SLASH-AND-BURN CULTIVATION AND DEFORESTATION IN THE MALAGASY RAIN FORESTS: REPRESENTATIONS AND REALITIES Volume 1
A Dissertation Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
by Jacques Pollini August 2007
© 2007 Jacques Pollini
SLASH-AND-BURN CULTIVATION AND DEFORESTATION IN THE MALAGASY RAIN FORESTS: REPRESENTATIONS AND REALITIES
Jacques Pollini, Ph. D. Cornell University 2007
Madagascar is a threatened biodiversity hotspot that has attracted the attention of conservation biologists and donor agencies for over two decades. The remaining primary forests are being cleared at a fast pace by farmers pushed toward the forest frontier and practicing slash-and-burn cultivation. A National Environmental Action Plan, supported by multilateral and bilateral donors, has been in place for 15 years but failed to decrease the overall deforestation rate. Results in terms of biodiversity conservation are local and have been achieved at the cost of negative impacts on local livelihoods. This dissertation analyzes the causes of this failure. The central hypothesis is that a gap exists between the realities of Malagasy farmers and the representation of these realities by actors committed to changing them. To test this hypothesis, I first analyzed the agrarian system of a community of slashand-burn farmers and the strategies of projects supporting the development of alternative land uses. Second, I deconstructed the environmental and development discourses that sustain conservation and development practices, using a method inspired from post-modernism and social constructivism. Special attention was paid to the alternatives to slash-and-burn paradigm, participatory ideology, and the demonization of fires. Third, I questioned these realities and discourses in the light of a conceptual framework aimed at avoiding the traps of positivist and relativist epistemologies.
The conclusion is that the politicization of scientific debates and the hegemony of technocratic controls led to a wide reality gap that no actor has been able to surmount. The real interactions between ecosystems and societies, the dynamics of agricultural intensification and the macroeconomic causes of deforestation are ignored. As a result, projects and policies target the wrong issues, have impacts contrary to their objectives and contribute to the marginalization of poor farmers whose dependence on natural resources persists. Only critical and creative analyses, inclusive ways to consider challenging representations and pragmatic and flexible intervention strategies could remedy this situation. I argue that in order to achieve this, more bridges and synergies must be created between research and decision making. The dissertation ends with a series of practical recommendations aimed at stopping deforestation in Madagascar without requiring the poorest farmers to pay the price for it. Among these propositions, subsidies aimed at compensating the negative impact of conservation on local livelihoods emerge as an inescapable utopia.
Keywords: Madagascar, Beforona, rain forest, deforestation, land degradation, slashand-burn cultivation, agrarian systems, environmental policies, political ecology, science and technology studies.
BIOGRAPHICAL SKETCH
Jacques Pollini (
[email protected]) holds a Master’s degree in biology (Maîtrise) from the Université des Science et Techniques du Languedoc, a Master of Sciences in Ecology (DEA) from the Université de Pau et des Pays de l’Adour and a Professional Master in Tropical Agronomy from the Ecole Supérieure d’Agronomie Tropicale de Montpellier (ESAT-CNEARC). He specialized in Comparative Agriculture at the Institut National Agronomique Paris-Grignon. He worked as a teacher in biology in France and as an agronomist and a forester in Burundi, Vietnam, Laos and Madagascar, from 1993 to 2006. He spent four years in Madagascar, between 2001 and 2006, including two years to conduct the fieldwork of this dissertation and two years as a technical advisor at the Ministry of Environment, Water and Forests for the Malagasy-French Cooperation.
iii
A mes parents, Joseph et Jacqueline
Aux paysans d’Ambodilaingo
iv
ACKNOWLEDGMENTS
I gratefully thank my committee members: Professor James P. Lassoie, my Chair Advisor, who has always provided me with the greatest care and encouragement and was just the right person to work with to address this comprehensive subject; Dr. Steven Wolf, for the intense discussions we had and for the guidance he provided me in entering the domain of science and technology studies; Professor Ronald Herring, who introduced me to political ecology and whose course provided me with a permanent feedback to my writing; and all of them, for the trust they had in me and the freedom they granted me for conducting my research. Thanks also to Norman Uphoff, Peter Marks, Charles Geisler, Davydd Greenwood, Paul Gellert and Louise Buck, for the great discussions we had and the new domains of knowledge they helped me to discover. Thanks to my amazing parents, Joseph and Jacqueline, my wonderful sisters Frédérique, Marie-Line and Cathy, and my great friends, François, Angèle, Guilhem, Emmanuelle, Eric, Cathy, Didier, Nicolas, Maria, Hubert, Karine, Pascal, Emmanuelle, Marième, Dean, John, Lorie, Guillermo, Vohangy, Livarilala, Dina, Misa, Narindra, Reni, Dada, Pascale, Olivier, Antonie, Virginie, Mickael, Isabelle, Narivo, Erica, Jutta, Simon, Farasoa, Valery, Myriam, Lovasoa, Minohery, Malala, Sarah, Federica, Pierre and Frank. They know they are in my heart and that this work could not exist without them. Thanks to all the people from Ambodilaingo who welcomed me in their villages, especially the Vavankaka and the Tangalamena, who are my second family. Thanks also to the people of Beforona, the staff of the NGOs Zanaky ny Ala, MITA and FFA and of the projects LDI, PTE and ERI, especially Mparany, Adrien and Faly, with whom I had great time in Madagascar. Thanks also to all the colleagues with
v
whom I worked in Madagascar, the staff from the Ministry of Environment, Water and Forest, and of the many organizations that collaborate with them. Thanks to all of them for the exchange of ideas and the discussions, sometimes intense, that we had together. Thanks, also, to the professors who accompanied me along my intellectual journey, especially my Professors from Cornell University already mentioned, Francis Hallé from the University of Montpellier and Marc Dufumier, Marcel Mazoyer and Hubert Cochet from the Institut National Agronomique Paris Grignon. I am also grateful to the Cornell International Institute for Food, Agriculture, and Development, the Mario Einaudi Foundation and the discretionary research funds of James P. Lassoie, for their financial support.
vi
TABLE OF CONTENTS VOLUME 1 Biographical sketch
iii
Acknowledgments
v
List of figures
xvi
List of tables
xvii
List of illustrations
xviii
List of abbreviations
xx
PART I: GENERAL INTRODUCTION
1
1. The Subject 2. Plan 3. Epistemological background 3.1. Complex objects 3.2. The limits of conventional science in the face of complexity 3.3. System analysis and modeling as a first answer 3.4. A second answer: beyond rationalization 3.5. How to check the validity of knowledge 4. Back to the subject 5. Conclusion
2 3 7 7 8 10 11 15 17 20
PART II: SETTING THE ISSUE: ENVIRONMENTAL DEGRADATION AND CONSERVATION IN MADAGASCAR 22 Introduction
23
Chapter 1. Environmental degradation in Madagascar 1. Introduction 2. Geographical features 3. Environmental and social History 3.1. First settlements and first environmental impacts 3.2. Chiefdoms and kingdoms 3.3. The first Europeans 3.4. From colonization to present 3.5. Deforestation in the twentieth century 4. The social causes of environmental degradation 4.1. Deforestation and the first frontier 4.2. Erosion and the second frontier 5. Conclusion
24 24 24 27 27 28 33 34 35 37 37 39 41
vii
Chapter 2. Environmental conservation policies and programs 1. Introduction 2. The first forest management and conservation efforts 3. The colonial period 4. The first republic 4.1. Economic context 4.2. The new legislation about vegetation clearing and fires 4.3. Forest control 4.4. The first environmental conferences 5. The transition to the second and third republic 6. Toward the National Environmental Action Plan (NEAP) 7. The Environmental Program, phase 1 (EP1) 7.1. Presentation of EP1 7.2. EP1’s impact 7.3. Conclusion 8. The Environmental Program, phase 2 (EP2) 8.1. Presentation of EP2 8.2. EP2’s impact 8.3. Conclusion 9. The Environmental Program, phase 3 (EP3) 9.1. Presentation of EP3 9.2. EP3’s implementation 9.2.1. The Durban Vision 10. Conclusion
43 43 43 45 50 50 51 53 54 55 58 62 62 65 71 72 72 79 81 82 82 90 90 107
PART III: THE REALITIES: SLASH-AND-BURN AGRICULTURE IN BEFORONA
109
Introduction
110
Chapter 3. Overview of the study area 1. Introduction 2. Biophysical environment 2.1. Location 2.2.Topography 2.3. Climate 2.4. Vegetation and deforestation 2.5. Soils 3. Population and society 3.2. Demography 3.1. ethnicity 3.3. Education 3.4. Healthcare 3.5. A brief political and socio-economic history 3.6. Social organization 3.7. Land tenure 3.8. Land-uses 3.8.1. Slash-and-burn cultivation (tavy)
113 113 113 113 113 115 119 126 130 130 130 133 133 134 138 140 140 141
viii
3.8.2. Paddy fields 3.8.3. Home gardens 3.8.4. Ginger cultivation 3.8.5. Vegetable 3.8.6. Improved fallow 3.8.7. Livestock husbandry 3.8.8. Forest and fallow exploitation 3.9. Farming-systems 4. Conclusion
151 151 153 157 157 157 160 162 167
Chapter 4. Methodology: comparative agriculture 1. Introduction 2. Key methodological principles 2.1. Participant observation 2.2. Qualitative methods, narratives and comprehensiveness 2.3. From enquiries to experience 2.4. Validating knowledge 3. A reading grid: comparative agriculture 4. Conclusion
169 169 169 169 170 171 174 176 182
Chapter 5. Ambodilaingo, a slash-and-burn farmers’ village 1. Introduction 2. Brief social history of the area 2.1. Settlement 2.2. Brief economic history 2.3. Social and cultural changes 2.3.1. Evolution of traditional ceremonies 2.3.2. The loss of solidarity 3. The present social organization 3.1. Traditional and legal authorities 3.2. Land tenure 4. Land uses 4.1. Cropping systems 4.1.1. Slash-and-burn systems (tavy) 4.1.2. Paddy fields 4.1.3. Home-gardens 4.1.4. Other cropping systems 4.2. Livestock systems 4.3. Forest products extraction 5. The strategies for intensification 6. Relations with Projects 7. An emotional portrait of the local economy 8. The influenza epidemic 9. The Agrarian system 10. Conclusion
184 184 191 191 195 199 199 202 206 206 207 210 210 210 218 221 222 224 226 229 231 232 239 241 247
Chapter 6. Rural development projects in Beforona 1. Introduction 2. The search for alternatives to slash-and-burn
249 249 251
ix
2.1. The BEMA/Terre-Tany research about alternatives 2.1.1. Agro-biological alternatives 2.1.2. Intensification of indigenous alternatives 2.1.3. Toward new farming systems 2.2. Nambena’s (2004) research on agroforestry alternatives 2.3. Styger’s (2004) research about fireless alternatives 2.4. Conclusion: future land uses in Beforona 3. The LDI and PTE projects 3.1. Main objectives 3.2. The CDIA 3.3. The Kolo Harena 3.4. The ban on tavy 3.5. Economic support 3.6. Technical support 3.7. LDI’s perception by beneficiaries 3.8. Conclusion 4. The Ankeniheny-Mantadia-Zahamena Corridor Restauration project 4.1. Context 4.2. Presentation of the AMZCR project 4.3. The transmission of fashionable alternatives 4.4. Overlooking the economy 4.5. The consequences 4.6. Some solutions? 4.7. Conclusion 5. Conclusion Chapter 7. Changing realities: a socio-economic experiment in Beforona 1. Introduction 2. Presentation of the non-governmental organization Zanaky ny Ala 3. Implemented activities 3.1. Irrigation of bottom land 3.1.1. The Marofody dam 3.1.2. The Hiaranana dam 3.1.3. The Ankorakabe dam 3.1.4. The other dams 3.1.5. Technical Supports 3.1.6. Socio-organization 3.1.7. Cost-efficiency 3.1.8. Conclusion 3.2. The agricultural contest 4. Activities planned but not implemented 4.1 Micro-credit 4.2. Construction of secondary roads 4.3. Land entitlement 4.4. Community granary 4.5. Community based natural resource management 4.6. Health and education 4.7. Others?
x
251 252 255 258 260 261 268 271 272 272 274 275 280 283 289 291 292 293 294 300 303 307 309 315 317 325 325 325 328 328 331 335 337 339 339 341 341 343 346 351 351 354 356 358 359 361 362
5. Activities not planned 6. Critical comments 7. Conclusion
363 364 366
Conclusion
368
VOLUME 2 PART IV: THE REPRESENTATION: DEVELOPMENT AND ENVIRONMENTAL DISCOURSES
372
Introduction
373
Chapter 8. Overview of Western development and environmental discourses 1. Introduction 2. Development discourses 2.1. The core meaning of “development” 2.2. The constructed meanings of development 2.3. High Modernism according to James Scott 3. Environmental discourses according to Dryzek (2005) 3.1. Dryzek’s analytical framework 3.2. Dryzek’s categories 3.2.1. Survivalism 3.2.2. Promethean 3.2.3. Environmental problem solving 3.2.3.1. Administrative Rationalism 3.2.3.2. Democratic Pragmatism 3.2.3.3. Economic Rationalism 3.2.4. Sustainability 3.2.4.1. Sustainable Development 3.2.4.2. Ecological Modernization 3.2.5. Green Radicalism 3.2.5.1. Green Consciousness 3.2.5.2. Green Politics 3.3. Reframing Dryzek’s categories 4. Environmental discourses in Madagascar 5. Conclusion
375 375 375 376 381 383 388 389 391 391 392 394 394 395 397 399 399 402 403 403 405 406 410 416
Chapter 9. Methodology: discourse analysis 1. Introduction 2. Michel Foucault’s epistemology 2.1. The first series of principle 2.2. The second series of principle 2.3. The third series of principle 2.4. The origin of discourses 2.5. Deconstructing discourses 2.6. Conclusion
417 417 417 418 420 422 422 423 424
xi
3. Bruno Latour’s epistemology 3.1. Latour’s key principles 3.2. Latour’s departure from Foucault 4. Synthesis 5. Avoiding the trap of relativism 6. A new model 7. Conclusion
425 425 428 429 432 434 439
Chapter 10: The Alternatives to Slash-and-Burn discourse (ASB) 1. Introduction 2. The birth of the ASB paradigm 3. Agroforestry at ICRAF 4. The received wisdoms about slash-and-burn cultivation 5. The first phase of the ASB Consortium 5.1. Main statements 5.2. Biophysical outcomes 5.3. Socio-economic outcomes 5.4. Outcomes in term of recommendations 5.5. Conclusion 6. The second phase of the ASB Consortium 6.1. A continuation of the first phase? 6.2. Toward a political economy of slash-and-burn cultivation? 7. Conclusion
442 442 442 444 454 457 457 458 459 461 464 464 465 468 476
Chapter 11: Community based natural resource management (CBNRM) 1. Introduction 2. The Participatory discourses 3. A few short case studies 3.1. Tsinjoarivo 3.2. The Zafimaniry area 3.3. Ikongo 3.4. Beforona 3.5. Two sites in western Madagascar 3.6. Conclusion 4. The GELOSE approach (Secured Local Management) 4.1. Background and origin 4.2. The GELOSE legislation and its potential effects 4.2.1. The contract 4.2.2. The COBA 4.2.3. The dina 4.2.4. Environmental mediation 4.2.5. The relative land tenure security (RLTS) 4.3. Conclusion 5. Pushing the GELOSE in the Malagasy environmental policies 5.1. The PCP evaluation 5.1.1. Methodology 5.1.2. Results 5.1.2.1. The integration into local, regional… policies
480 480 481 485 487 489 491 493 493 495 496 497 499 499 501 507 508 510 512 513 514 514 515 515
xii
5.1.2.2. Environmental impacts 5.1.2.3. Social impacts 5.1.2.3.1. Observed impacts 5.1.2.3.2. Potential impacts 5.1.2.4. Economic impacts 5.2. The IRD evaluation 5.2.1. Methodology 5.2.2. Results 5.2.3. The institutional driving of the IRD’s recommendations 6. Confrontation with other research outcomes 7. Conclusion
517 520 520 522 524 525 525 526 530 532 535
Chapter 12: The antifire discourse 1. Introduction 2. The antifire received wisdom according to Kull (2004) 3. Kull’s thesis about human impact on the environment 4. Kull’s conclusion 5. Conclusion
541 541 542 544 547 548
Conclusion
549
PART V: SYNTHESIS: TOWARD NEW DISCOURSES, EPISTEMOLOGIES AND PRACTICES Introduction
552 553
Chapter 13. Toward a Political Ecology of deforestation and slash-andburn cultivation 1. Introduction 2. A parallel with Land degradation and society 3. Brief overview of political ecology work 3.1. The foundation of the discipline 3.2. The key narratives 3.2.1. The marginalization thesis 3.2.2. The control thesis 3.2.3. The environmental conflict thesis 3.2.4. Environmental identity and social movements 4. The political ecology of deforestation in the Malagasy highlands 5. Toward a deconstruction of political ecology 5.1. From equilibrium to non equilibrium ecology? 5.1.1. The complementary character of equilibrium and nonequilibrium ecologies 5.1.2. Distinguishing vegetation change and vegetation development 5.1.3. Relativism and revisionism in social environmentalism 6. Conclusion
xiii
558 558 559 567 567 570 571 572 574 579 579 584 585 586 589 591 595
Chapter 14. Toward new epistemologies 1. Introduction 2. Confirming and strengthening an epistemological frame 2.1. Bruno Latour’s parliament of things 2.2. Actor-Network Theory 2.3. From Latour’s to Popper’s epistemology 2.4. Karl Popper’s epistemology 2.4.1. The falsification principle 2.4.2. The problem of induction 2.4.3. Dogmatism, critical attitude and uncertainty 2.4.4. The growth of knowledge 2.4.5. What is truth and can we know it? 2.4.6. Summary: three epistemologies 2.4.7. Application of Popper’s model to the subject of this dissertation 2.5. Paul Feyerabend: beyond Popper? 2.6. Edgard Morin’s science of complexities 3. Nature and culture 3.1. Introduction 3.2. What is nature? 3.3. The value of nature 3.4. The environment 3.5. Nature in political ecology 3.6. The dialogy of nature and culture 4. Conclusion
598 598 599 599 607 612 614 614 616 618 619 621 624
Chapter 15. Toward new practices 1. Introduction 2. Metis 3. Propositions to solve the issues of deforestation 3.1 Are tavy farmers the main cause of deforestation? 3.2. The reasons to stop deforestation 3.3. Solving the tradeoff between conservation and local livelihoods 3.4. Proposition of solutions 3.4.1. The role of legislation 3.4.2. Mitigation by supporting agriculture intensification 3.4.3. Subsidies 3.4.4. Conclusion 4. Institutional approach 4.1. The open society 4.2. Piecemeal social engineering 4.3. The open project 5. Conclusion
654 654 656 661 661 662 665 666 666 669 670 680 680 682 683 685 688
Conclusion
625 629 634 640 640 640 643 645 646 646 647
694
xiv
PART VI: GENERAL CONCLUSION
697
1. Introduction 2. Summary of outcomes 2.1. The conclusions from Part II 2.2. The conclusions from Part III 2.3. The conclusions from Part IV 2.4. The conclusions from Part V 3. Conclusion 4. Future research APPENDIX
698 698 698 699 703 705 712 720 726
EP3’s logical framework
726
WORK CITED
732
xv
LIST OF FIGURES Figure 1: Summarized table of contents Figure 2: Zoning and transect of the Betsimisaraka region Figure 3: Temperature and rainfall in the study area and surroundings Figure 4: Deforestation pace in Beforona from 1957 to 2000 Figure 5: Age pyramid in Marovoalavo, Beforona Figure 6: Distribution of work according to activities in Beforona Figure 7: Yield in tavy in relation to the type of fallow vegetation Figure 8: Relation between extension of home gardens distance from Road Figure 9: Main trees used for house construction in the Vohidrazana watershed Figure 10: Typical farming system in Beforona Figure 11: Income structure in Tanambao, Beforona Figure 12: Alternative farming system in Beforona Figure 13: Eco-design of an exemplary agroforestry hillside Figure 14: The chain of actors Figure 15: Match and mismatch between representations and realities Figure 16: The agroforestry pathways to change Figure 17: The evolution of agricultural landscape in Bukeye, Burundi Figure 18: Number of hits for several concepts in CAB Abstracts database Figure 19: Number of hits of several concepts in 3506 references Figure 20: Grid for analysis of farmer’s strategies Figure 21: Development pathways from slash-and-burn systems Figure 22: The ASB research framework Figure 23: Environmental degradation and recovery
xvi
6 116 118 121 132 144 148 154 164 166 168 259 262 323 435 445 448 451 453 460 462 469 473
LIST OF TABLES Table 1: Contributions to EP3 by donors and components, in million USD Table 2: Projects related to biodiversity conservation in Madagascar since 1997 Table 3: Categories of protected areas according to IUCN (2002) Table 4: Main differences between IUCN Categories V and VI Table 5: Malagasy protected areas in 2007 Table 6: Population in Beforona by ethnic group. Table 7: Cultivated species in Beforona Table 8: Plants cultivated in tavy in Ambinantsavolo, a Fokontany in Beforona Table 9: Dominant species in fallow in the region of Beforona and Moramanga Table 10: Main medicinal plants in the Vohidrazana watershed Table 11: The performance of cropping systems in Beforona Table 12: Comparison of tavy with alternatives Table 13: Effect of limited burning and no burning on rice productivity Table 14: Relative yields of fireless alternatives according to type of land Table 15: Relative monetary return of fireless alternatives in USD Table 16: Example of expenditures and investment in Ambodilaingo, in USD Table 17: How do organisms, societies… change, grow, develop and evolve? Table 18: Classification of environmental discourses Table 19: Comparison of Foucault’s (1970) and Latour’s (1989) concepts Table 20: ASB consortium criteria for comparing land uses Table 21: ASB Consortium criteria for comparing sites Table 22: Comparison of theories in relation with their verisimilitude
xvii
87 88 97 97 104 132 142 146 147 163 165 253 253 265 266 353 377 390 430 466 467 650
LIST OF ILLUSTRATIONS Map 1: Remaining primary vegetation in Madagascar Map 2: Existing protected areas and areas reserved for conservation sites Map 3: Localization of the study area Map 4: Regional map showing Beforona and Ambodilaingo. Map 5: Localization of sites investigated by BEMA and Terre-Tany Map 6: Forest clearing in the study area, from 1957 to 2000 Map 6 (continuation) Map 7: Land use change in comparison with prediction by model Map 8: Protected areas deforestation in the Mantadia-Zahamena corridor Map 9: Vegetation in Beforona surroundings Map 10: Population density in the communes of Beforona and Ranomafana Map 11: Land use in a few fokontany in Beforona Map 12: The 5 villages of the Ambodilaingo fokontany Map 13: Target area of the VMCRCC project Map 14: Location of management transfer sites investigated
26 93 111 112 114 123 124 125 127 128 131 143 185 296 486
Photograph 1: The Betsimisaraka escarpment Photograph 2: Bottom lands and hills in Beforona Photograph 3: A tavy in an ordinary fallow land, south from Beforona Photograph 4: Tavy in an eucalyptus woodland in Beforona Photograph 5: A home garden in Sahavolo (Beforona) Photograph 6: Banana transportation Photograph 7: Landscape transformation in Beforona Photograph 8: Indigenous techniques for erosion control Photograph 9: Manual plowing in a ginger field in Beforona Photograph 10: Drainage ditches planted with maize in Beforona Photograph 11: Wood exploitation in Beforona Photograph 12: Wood transportation Photograph 13: Satellite photography showing the 5 villages of Ambodilaingo Photograph 14: Tavy on the forest frontier in Ambodilaingo Photograph 15: Tavy inside the primary forest in Ambodilaingo Photograph 16: People of Ambodilaingo in their village Photograph 17: Children in Ambodilaingo Photograph 18: Ambodilaingo Photograph 19: Young villagers playing with a zebu before its sacrifice
117 117 145 145 152 155 156 158 159 159 161 161 186 187 187 188 188 192 200
xviii
Photograph 20: Meat distribution after a zebu sacrifice Photograph 21: The impact of wildfires in Ambodilaingo Photograph 22: The main street of Ambodilaingo Photograph 23: Visitors at the CDIA-LADIA in Beforona Photograph 24: The CDIA-LADIA in Beforona Photograph 25: Vegetation in Ambavaniasy Photograph 26: Land improvement in Ambavaniasy Photograph 27: Villagers from Ambodilaingo digging an irrigation canal Photograph 28: Irrigation canal and Rayamendrany in Ambodilaingo Photograph 29: A “tavy” farmer in Ambodilaingo, with her paddy field Photographs 30, 31 and 32: The Marofody bottom land in Ambodilaingo Photograph 33: Paddy fields irrigated by the Hiaranana dam Photograph 34: Paddy fields suffering from water shortage in Ambatoharanana Photograph 35: The Ankorakabe dam, in Beforona Photograph 36: The Ankorakabe canal, with irrigated paddy fields Photograph 37: Farmers’ improvement in the Andriamanavana site Photograph 38: Failed irrigation canal in Ambodilaingo Photograph 39: New paddy fields in construction in Ambodilaingo Photograph 40: Canal built by villagers to irrigate paddy fields, Ambatoharana Photograph 41: Masonry on a canal section in Ambodilaingo Photograph 42: Vegetable cultivation for the agricultural contest Photograph 43: Participants of the agricultural contest Note: all photographs except when indicated (# 3, 42 and 43) are by myself
xix
200 205 234 273 273 298 306 330 330 332 333 336 336 338 338 340 340 344 345 345 348 348
LIST OF ABBREVIATIONS ADRA AGERAS AGEXs AMZCR ANAE ANGAP ANGEF ANT ASB BEMA CAPE CBNRM CDIA CELCO CFSIGE
CGP CI CIIFAD CIRAD
CNCD COBA CSB I CTFT DCT DD
Adventist Development and Relief Agency Appui à la Gestion Régionalisée et à l’Approche Spatiale (Support to Regional Programming and Spatial Analysis) Agence d’Exécution (Line Agency) Ankeniheny-Mantadia-Zahamena Corridor Restoration Project Association Nationale pour les Actions Environnementales (National Association for Environmental Actions) Association Nationale pour la Gestion des Aires Protégées (National Association for the Management of Protected Areas) Agence Nationale pour la Gestion des Eaux et Forêts (National Agency for Water and Forest Management) Actor-Network Theory Alternatives to Slash-and-Burn Bilan Ecologique à Madagascar (Ecological Assessment in Madagascar) Composante Aires Protégées et Ecotourisme (Protected Areas and Ecotourism Component) Community Based Natural Resource Management Centre pour le Développement et l’Intensification Agricole (Development and Agriculture Intensification Center). Cellule de Coordination du PE3 (Coordination Unit of EP3) Centre de Formation aux Sciences de l’Information Géographique et de l’Environnement (environment and geographic information training center) Coordination Générale des Project (Project Coordination Service) Conservation International Cornell International Institute for Food, Agriculture and Development Centre International de Recherche Agronomique pour le Développement (International Centre for Agronomic and Development Research) Commission Nationale pour la Conservation et le Développement (National Commission for Conservation and Development) Communauté de Base (grassroot community) Centre de Santé de Base Niveau 1 (Level 1 Basic Health Center) Centre Technique Forestier Tropical (Tropical Forestry Technical Center) Development and Conservation Territory Direction des Domaines (Land Titling Directorate)
xx
DEF DGE DGEF DSRP DVG DWCT EMC EP1,2,3 ERI ESFUM ESSA EU FAGEC FAO FFEM FID FMg FOFIFA FORAGE FSP-GDRN
FSP-ADRA FTM GCF GDRN GEF GELOSE
Direction des Eaux et Forêts (Water and Forest Directorate) Direction Générale de l’Environnement (Environment General Directorate) Direction Générale de Eaux et Forêts (Water and Forests General Directorate) Document Stratégique de Réduction de la Pauvreté (Poverty Reduction Strategy Paper) Durban Vision Group Durrell Wildlife Conservation Trust Environnement Marin et Côtier (Marine and Coastal Environment) Environment Program Phase 1, 2, 3 Ecoregional Initiative Ecosystèmes Forestiers à Usage Multiple (Multiple-Use Forest Ecosystems Management) Ecole Supérieure des Sciences Agronomiques (school of agronomical sciences) European Union Fond d’Appui à la Gestion Environnementale Communautaire (Community Environmental Management Support Fund) Food and Agriculture Organization Fonds Français pour l’Environnement Mondial (French Global Environment Fund) Fonds d’Intervention pour le Développement (Social Fund Project) Franc Malgache (Malagasy Franc) Foibe Fikarohana momba ny Fambolena (Malagasy national research center for agriculture and rural revelopment) Fonds Régional d’Appui à la Gestion de l’Environnement (Regional Fund for Environment Management) Projet Gestion Décentralisée des Ressources Naturelles du Fonds de Solidarité Prioritaire (decentralized natural resources management project of the priority solidarity fund) Food Security Program of the Adventist Development and Relief Agency Foibe Taosaritanin’i Madagasikara (Malagasy National Geographic Institute Gestion Contractualisée des Forêts (Community Forest Management Contract) Gestion Décentralisée des Ressources Naturelles (Decentralized Natural Resources Management) Global Environment Facility Gestion Locale Sécurisée (Secure Local Management)
xxi
GNP GoM GTZ
Gross National Product Government of Madagascar Gessellschaft fyr Technische Zusammenarbeit (German cooperation agency) Ha Hectare ICDP Integrated Conservation and Development Project ICRAF International Centre for Research in Agroforestry (now the World Agroforestry Centre) IDA International Development Agency IEFN Inventaire Ecologique Forestier National (National Forestry Inventory) IMF International Monetary Fund IRD Institut de Recherche pour le Développement (Development Research Institute) IUCN International Union for Conservation of Nature (now the World Conservation Union) Km Kilometer LDI Landscape Development Interventions MECIE Mise en Compatibilité des Investissements avec l’Environnement (environmental impact assessment) MEF Ministère des Eaux et Forêts (Ministry of Water and Forest) MINENVEF Ministère de l’Environnement, des Eaux et Forêts (Ministry of Environment, Water and Forest) MT Management Transfer NEAP National Environmental Action Plan NGO Non-governmental organization NR2 National Road 2 NTFP Non Timber Forest Product OECD Organization for Economic Co-operation and Development ONE Office National pour l’Environnement (National Office for the Environment) OSIPD Office Statistique et Informatique pour Programmation du développement (office for data processing and statistic applied to development programming). PA Protected Area PAGE Projet d’Appui à la Gestion de l’Environnement (Environmental Management Support Project) PADR Plan d’Action pour le Développement Rural (Rural Development Action Plan) PSDR Projet de Soutien au Développement Rural (Rural Development Support Project) PTE Programme de Transition Ecoregional (Ecoregional Transition
xxii
SAGE SAPM SRA SRI UNCED UNDP USA USAID USD VMCRCC WCED WCS WHO WWF
Program) Service d’Appui à la Gestion de l’Environnement (Service for Support to Environmental Management) Système des Aires Protégées de Madagascar (Protected Area System of Madagascar) Système de Riziculture Améliorée (Improved Rice Cultivation System) Système de Riziculture Intensive (Intensive Rice Cultivation System) United Nations Conference on Environment and Development United Nations Development Program United States of America United States Agency for International Development United States Dollar Vohidrazana Mantadia Corridor Restoration and Conservation Carbon Project World Commission on Environment and Development World Conservation Society World Health Organization World Wildlife Fund
xxiii
PART I GENERAL INTRODUCTION
1
1. The subject Slash-and-burn cultivation is a land use that consists of slashing the vegetation on a piece of land, burning it and growing crops that will benefit from the nutrients stored in the biomass and made available in the form of ashes (Conklin 1957; Nye and Greenland 1963; Peters and Neuenschwander 1988; Ramakrishnan 1992). After one or a few years of cultivation, fertility is depleted and the land is temporarily abandoned to fallow. Vegetation develops again, which leads to the reconstitution of a nutrient stock that can be used through a new cycle of clearing, burning and cropping. In traditional systems still in equilibrium, the fallow period lasts 10 to 30 years. Farmers rotate the location of their crops across a wide landscape, sometimes moving their village, and this system is called shifting or swidden cultivation. Slash-and-burn systems are often located in remote areas with low population density, where primary forests still exist. As in any agricultural system, these forests are perceived as reservoir land. They are cleared at the rhythm of population increase, or when communities are pushed away from their ancestral land by groups that practice more intensive land use. For this reason, slash-and-burn systems are frequently involved in deforestation dynamics, whereas they are not necessarily the cause of deforestation. This dissertation investigates deforestation and slash-and-burn cultivation systems in the eastern rain forests of Madagascar. It considers these two issues as linked but distinct, and will examine how they are addressed by Malagasy and international organizations. More precisely, it will attempt to answer to the following questions: •
Why do Malagasy slash-and-burn farmers still clear the Malagasy eastern rain forest, despite the fact that international organizations have been supporting alternative land uses for several decades?
2
•
Why do the current policies and programs dealing with this issue fail, including projects that intervene in small areas and have elevated budget per beneficiary households?
•
What needs to be done to stop forest clearing and to decrease the poverty of slash-and-burn farmers, or to at least significantly slow deforestation without increasing poverty?
In response to these questions, I hypothesize that there is a mismatch between the reality of slash-and-burn farmers and deforestation dynamics on one hand, and the perception of these realities by those who act upon them on the other hand. In other words, the discourses produced by the organizations which address these issues are biased, incomplete, and describe realities that do not exist, or are significantly different from these descriptions. It is not necessarily the complexity of the reality that is causing this mismatch. It can merely be the subjectivity of those who produce representations. The result is the failure of what is implemented on the ground on behalf of discourses and representations. This hypothesis is tested by analyzing the practices of slash-and-burn farmers in a commune 1 of eastern Madagascar, the rural development projects that support these farmers and the development and environmental discourses within which these projects are embedded. The outcomes are presented according to the following plan.
2. Plan Part II sets the issues of slash-and-burn cultivation and deforestation in Madagascar. It proposes an overview of the social and environmental dynamics
1
A commune is an administrative unit. Madagascar is divided into six autonomous provinces (faritany), 22 regions (faritra), about 115 districts (departamenta) and about 1400 communes (kaominina) which can be rural or urban. Communes are divided into fokontany which are administrated by a fokontany President nominated by the commune.
3
associated with slash-and-burn cultivation (Chapter 1) and portrays the environmental policies that address deforestation in their historical context (Chapter 2). Part III addresses the same issues at local scale through the collection of empirical data. It analyzes the realities of slash-and-burn farmer communities in the commune of Beforona, located on the eastern verge of the Malagasy rain forest, and examines the rural development activities and environmental policies implemented in this area. The first chapter (Chapter 3) is an overview of the area using general information collected in the literature. Chapter 4 presents the methodology employed to collect original field data. Chapter 5 is a narrative drawn from first hand qualitative data collected in Ambodilaingo, a village where I resided six months, from July to December 2001. Chapter 6 analyzes the main projects that operate in the area and their impact. It includes a review of previous research work about alternatives to slash-andburn and an analysis of projects’ activities and their perception by beneficiaries. Chapter 7 presents the results of a rural development project implemented in Ambodilaingo and other villages by a small nongovernmental organization “Zanaky ny Ala” 2 that I created with my research assistant. This project was the translation into action of the preliminary conclusions of Chapters 5 and 6. It can be viewed as a social experiment aimed at testing these outcomes. Part III ends with a synthesis and conclusion that lists the main mismatches between discourses and realities, based on the outcomes of Chapters 5, 6 and 7. This led to the identification of three discourses that will be analyzed in Part IV. Part IV provides an analysis of development and environmental discourses at global scale and of their translation into policies in Madagascar. It is aimed at understanding the causes of the mismatches identified in Part III. It starts with an overview of Western development and environmental discourses, based on the 2
Children of the Forest in Malagasy language.
4
literature (Chapter 8) and introduces the methods (Chapter 9) I utilized for analyzing discourses. These discourses, which have been identified in Part III, are the alternatives to slash-and-burn discourse (Chapter 10), the participatory discourse, more precisely its translation into community based natural resources management policies (Chapter 11), and the antifire discourse (Chapter 12). In these three chapters, the literature is regarded as material providing me with empirical facts for the analysis of discourses, in the same way that the practices of farmers and projects provided me with empirical facts for the analysis of realities in Part III. Part V is a general synthesis and includes recommendations for future action. Chapter 13 confronts my results with the narrative developed by political ecology, a discipline that addresses similar issues to those I address in this dissertation and developed similar narratives and explanations, whereas some differences also exist. Chapter 14 proposes a new epistemological framework aimed at reducing the mismatch between representation and realities. It calls for a paradigmatic shift and synthesizes the outcomes of Parts II and III in the light of this shift. Chapter 15 proposes pragmatic modes of action aimed at addressing the issues of deforestation and slash-and-burn cultivation in this new paradigm. Eventually, the general conclusion (Part V) attempts to extract the main lines of this dissertation and to articulate its empirical and abstract content in a way that could be meaningful for field practitioners and policy makers. This plan is summarized in Figure 1, which visualizes the main articulation between chapters.
5
PART I: General introduction (Research question and epistemological framework)
Part II: Setting the issue Chapter 1: Environmental degradation in Madagascar Chapter 2: Environmental conservation in Madagascar
Part III: The realities
Part IV: The discourses
Chapter 3: Overview of the study area (Beforona)
Chapter 8: Overview of Western development and environmental discourses
Chapter 4: Methodology (comparative agriculture)
Chapter 9: Methodology (discourse analysis)
Chapter 5: Ambodilaingo, a slash-and-burn farmers’ village
Chapter 10: The alternatives to slash-and-burn discourse
Chapter 6: Rural development projects in Beforona
Chapter 11: Community based natural resources management
Chapter 7: Changing realities: a socio-economic experiment
Chapter 12: The antifire discourse
Part V: Synthesis Chapter 12: Toward a political ecology of deforestation and slash-and-burn cultivation Chapter 13: Toward new epistemologies Chapter 14: Toward new practices
PART VI: General conclusion Figure 1: Summarized Table of Content
6
3. Epistemological background The tools and methods used to collect empirical data will be presented in detail in Chapters 4 (for the analysis of realities) and 9 (for the analysis of discourses). In this section, I merely present the general epistemological framework of the research.
3.1. Complex objects The object addressed in this dissertation is complex. At a local scale, it is a network of biophysical and social interactions that links a group of human beings, their ecosystem and their land uses. At a larger scale, it includes a social, economic, political, cultural, and “scientific” context which determines ecosystems and land uses transformations. Biologists, sociologists, economists, policy makers and donors are all interested in understanding this object because they are committed to act upon it and determine its future. They each analyze it from their own perspective and draw detailed pictures of aspects of it. My research hypothesis is that no relevant comprehensive picture emerges from this, as revealed by the general failure of projects and policies. In order to avoid this pitfall, I chose to depart from conventional researches that “look at” aspects of this broad reality. Instead, I tried to understand why knowledge gaps always remain, committed to fill these gaps and adopted a comprehensive approach aimed at linking together different realms of knowledge. In this endeavor, I had to depart from conventional science because it does not provide appropriate tools to study broad and complex objects. This raised several questions about science, discourses, theories and knowledge. For instance, what is conventional science? How to differentiate scientific and nonscientific facts, ideas, discourses and theories? Where is the frontier between scientific and nonscientific knowledge? How to decide that some information is relevant? When we experience a reality, we permanently obtain nonscientific
7
information that relates to it. Do we have to use this information to build up our representations? In which manner? With what precautions? Shall we rather set up additional scientific protocols to collect this same information with more rigor? Is this always possible, technically and practically? If not, what to do? What is more scientific: to limit our enquiry to aspects of the reality that can be “scientifically” addressed or to fill knowledge gaps with imperfect material collected with “nonscientific” methods? Which of these two constructed representations will be closer to the reality, the incomplete or the imperfectly completed? In order to tackle these questions, it is necessary to examine how mainstream science is working, what are its limits to address complexity and how to go beyond these limits.
3.2. The limits of mainstream science in the face of complexity Classically, science is the "intellectual and practical activity encompassing the systematic study of the structure and behavior of the physical and natural world through observations and experiment" 3 . More recently, the social world has been added to the domains investigated by science. An essential characteristic of science, as it is practiced in Western civilization, is that it privileges quantitative methods. This leads to produce representations that are expressed in rational and explicit terms and, in some cases, by mathematical equations. Consequently, the accumulation of knowledge is regarded as a linear process of objective learning. It starts from an origin (a new paradigm or a series of hypothesis), then gathers empirical facts that validate or invalidate these hypotheses, create theories or representations, and finally conducts to a transcendental outcome that allows to prove that these representations match with the reality. We call positivism this conception of science. A more nuanced conception nevertheless 3
The New Oxford Dictionary of English 1998.
8
recognizes that truth cannot be known and considers that steps apparently “backward,” or the refutation of false theories, matter more than steps forward (the verification of existing theories). This conception, which can be called negativist, defines scientific knowledge, more modestly, as knowledge that can be refuted (Popper 1963). But still, an omnipresent “attractor” puts linear order into knowledge, with the risk of subjectively reshaping representations to make them fit with this linearity. The positivist conception of science makes sense for a large series of subjects, as has been shown by the fast technical progress that science enabled. But it also has some limitations that are revealed by the increasing doubts about the capacity of science to engage human civilizations in social and spiritual progress, and by the blurring of the frontier between sciences and humanities, or between physical, natural and social objects. Few processes are actually linear in the real world. Very simple ones are or can be regarded as being so, by approximation. But in most cases, social objects are too complex to be fully represented into a sum of quantifiable processes. The object addressed in this dissertation is clearly concerned by these difficulties. Comprehensiveness is bidirectional. It goes outward, but also inward, as has been shown by the theory of chaos. This theory led to paying more attention to the hidden domains of complexity that determine the behavior of seemingly simple objects (Gleick 1987). The system of forces that operates on such objects produces a complex behavior that can be represented in a geometrical space that is not determined by linear equations, but rather responds to fractal geometry. It is as if an excess of stochastic micro-variations of the system's state would spill over in a hidden nonEuclidean space, in a sort of microcosmic turbulent area where entropy that could not generate Euclidean order would plummet. Concerning this research, inward complexity appears quite abstract and I did not expect that addressing it would have significantly improved the outcomes. I thus rather oriented my investigation in the
9
direction of “outward” complexity, acknowledging that perfect bidirectional comprehensiveness is unattainable. Scientists address complexity by dividing objects into elements that they study separately (reductionism), but still have some difficulties in explaining the functioning of the whole system when they reconstitute it in the form of a model. Some parts are always missing, or remain in the state of black boxes. These missing parts relate to elements with less biophysical and more social content, resulting in a relative isolation of the studied object from the society it belongs to. The observer is also missing from the models, a sign that reflexivity has not made its entry in mainstream sciences yet. In order to address complex objects, it is thus necessary to go beyond reductionism and positivism, two paradigms that still largely dominate scientific practices.
3.3. System analysis and modeling as a first answer to complexities System analysis has been a first answer to complexity proposed over several decades in a large range of disciplines. Unfortunately, system analyses are often limited to the survey of objects and phenomenon that can be directly observed in the state they exist. They give little room to experimentation and have difficulties predicting what will happen in the future, which limits their role for action. They provide conclusions that are supposed to guide practitioners engaged in modifying the real world, but these field actors often fail in valorizing excessively complex sets of data. They prefer to ask system analysts to provide more precise guidelines for action, but these demands are rarely satisfied. System approaches also tend to excessively focus on what cannot be done when considering the whole. They end in viewing all relations between micro and macro levels as constraints. They analyze well the global state of a system, its connectivity with external factors, its general dynamics, but not enough its capacity to
10
absorb shocks and reorganize itself under perturbation. The comprehensive object that system analysts describe thus remains "frozen 4 " by the description of multiple interactions. Flexibility, that might be the counterpart of complexity, is not satisfactorily addressed. As a result, the outcomes of system analysis discourage field actors from most pathways they envision for changing realities. These actors then fall back into the reductionism paradigm that they eventually perceive as more appropriate for the scale and type of problems they address. In order to obtain a predictive character, system approaches have developed models that they call “expert systems.” Unfortunately, the reluctance or difficulties to go beyond the positivism and reductionism paradigms limits these models to a series of simple linear processes that can be computed. Key social issues are skipped in the process or regarded as black boxes. A significant part of what was gained by system analysts in terms of comprehensiveness is lost. Some dimensions of the system are discarded because they cannot be translated into clear algorithms, whereas human skills and intelligence, which go far beyond the capacity of such algorithms, could have addressed them. These models are certainly useful, if properly used, but are always strong simplifications. They cannot be regarded as being the last step in the acquisition of knowledge, whatever is the issue they deal with. In sum, it still remains to find what sort of knowledge formalization has to be built beyond systemic descriptions and predictive models.
3.4. A second answer: beyond rationalization The limits faced by science to investigate complex objects may reflect the limits of the capacity of humans to express in rational terms the experience of their 4
Thanks to Patrice Lamballe, a development expert who worked in Vietnam for the GRET (Groupe de Recherche et d’Echanges Technologiques; Solidarity and International Cooperation Association), a French NGO, for expressing and discussing this idea.
11
world. The assertion that “everything is rational,” for instance, is at the corner stone of many polemics in our time of chronic epistemological questioning and hesitation. I believe that these polemics result from a misunderstanding about the sense of the word "rational.” Does saying that "everything is rational" simply means that all phenomenon in our world are determined by the mechanistic laws of physics and chemistry and could theoretically be described using these laws? Or does it also mean, beyond that first assumption, that humans can in practice describe all phenomenon in rational terms, through expressions derived from these same mechanistic laws? The positivist paradigm that appeared in the enlightenment period produced a Western culture where the implicit answer given to both questions is yes. Kant (1787) criticized this radical positivism but his work, while influential, did not suffice to impede the mystification of science. Scientists framed in the positivist myth have the promethean belief that their rational methods can lead them to know the truth. They attribute to science the power of a god. I would agree that a perfect rationale spirit connected to all physical phenomenon and using science could have, through an unbiased perception of its environment, a complete understanding of all interactions that exist, from the microcosm to the macrocosm. This perfect spirit would be capable of putting the global world in a system of equations that would take account for its least components. He could also delineate the boundaries between the predictable Euclidean world and the fractal space that absorbs the excess of disorder on its microcosmic verges. But humans would be of an excessive arrogance if they would pretend to have such skills and achieve some day such an absolute objective comprehensiveness. The universe is rational but not rationalizable, or at least not rationalizable using human skills.
12
Hence the belief on which my methodology is based is that we should answer “yes” to the first question and “no” to the second one. Science and reason must not be rejected but we must acknowledge that scientific tools and methods have some limits, and that we ourselves have limitations in the use we can make of them. Science has failed to address social and complex objects that may be those that matter the most for our future. We thus have to go beyond the positivism and reductionism paradigms and the myth that rational thinking is the only way to produce scientific knowledge. Nature provided us with intellectual skills other than rationality. Intuition for example is a key intellectual faculty that helps to make decisions whose logic is not apparent, but whose sense can nevertheless be confirmed by the successes it helps us to achieve. Dreyfus and Dreyfus (1986) showed that if we consider a five-step scale in the process of building personal skills and knowledge (1: novice; 2: advanced beginner; 3: competent performer; 4: proficient performer; 5: expert), intuition is the key faculty for reaching the most advanced stages (4 and 5). Computed “expert systems,” because they are incapable of intuition, cannot go over stage 3 and should not be qualified as experts. Intuition is already recognized as an important quality for achieving scientific discoveries, as expressed by the symbolic importance, in the culture of all scientists, of Archimedes’ word “Eureka,” pronounced when he discovered the buoyant force. But intuition, unfortunately, is reserved to periods of deep questioning, of doubts about prevailing paradigms. It helps to make scientific revolutions and to open new realms of knowledge, as shown by Khun (1970), but as soon as revolutions are done, scientists go back to what Khun (1970) called normal science. They mostly apply inductive and hypothetical-deductive methods, which narrow their view and may delay future paradigm shifts. In short, intuition, if used cautiously and as a complement to rationalization, could help to bring humanity as far forward as it brings individuals when it is used in
13
their life’s experiences. It may be a more efficient way to summarize complex experiences than rational algorithms developed from sets of objective data. It may also justify taking some liberty with established methods, as Feyerabend (1975) argued. Intuition is sometimes rejected by scientists for being incompatible with rigor. It is viewed as a source of subjectivity due to its connectivity with emotion, a skill that is regarded as much linked to subjectivity as rationality is linked to objectivity. Emotions are felt, for example, when looking at a painting, having sympathy for a person or falling in love, events that matter much in our life but cannot be properly described in rational terms. But how can things that are necessary for peoples’ personal progress not be necessary for the progress of a group of people, i.e., of societies? If the purpose of science is to provide a comprehensive understanding of ourselves and our world in order to drive our destiny better, how can a mode of perception such as emotion be absent of scientific enquiry? Philosophy, ethics and arts already produced knowledge or created objects using nonpurely rational cognitive skills. Can the overall progress of human societies be possible with a separation of science from these other realm of knowledge? We can also seriously doubt about the absence of emotion in supposed objective and rational modes of cognition. It is now widely recognized that scientific work cannot be separated from cultural determinants. The personal background of researchers contributes to determining the direction they give to their work. Damasio (1994) showed that people who have been wounded in the area of their brain that relates to emotion are unable to act rationally. They can determine what they need to do, but they will not do it. This does not mean that emotion plays a role in the information treatment process itself, but shows that it is at least an incentive that helps to translate ideas into decisions and action. We can wonder what would be a scientist without emotion: what would be his/her capacity to go beyond the thinking dictated by
14
his/her culture. How could he/she engage in new hypothesis or theories and gather personal energy for achieving key challenges? Scientists, like any human, have emotional complexes, or affects, which have been built along their history. These affects can be either positive incentives or causes of biases and misrepresentations. Having an emotion is maybe not a scientific act and can go against science, but the scientific method has to find ways to use emotions positively, managing side effects that go against objectivity but also using them as incentives for more innovative, powerful, in depth, holistic and committed thinking. This would lead to establish more connections between science and philosophy. Even art, eventually, may not be separated from science (Feyerabend 2003). Scientists would then contribute to the embellishment of the world we live in, a world for which they already helped to build the material dimension.
3.5. How to check the validity of knowledge if rationality loses its exclusivity We saw that the use of intuition and emotion can increase efficiency in the advancement of knowledge, but may also increase the risks of subjectivity. This issue has to be taken into account seriously. How to do so? First, as contended above, emotion and intuition are already present in science because scientific research is carried out by people. Recognizing these skills as part of the cognitive processes, with the series of precaution it would imply, would not necessarily make science more subjective. To the contrary, it could help to delineate hidden areas of subjectivity and control their biases. Second, whatever the tools used to acquire knowledge are, there is only one way to check that a conclusion is sound: to make a prediction according to this conclusion and to check that this prediction actually occurs. Biophysical sciences already engaged in this logic and put experiments at the core of their methods. But
15
social sciences are not yet clearly engaged in it. I advocate for this engagement to occur, following the arguments of Bent Flyvjberg in his essay “Making social sciences matter” (2001). Thus, the matter is not whether a given knowledge is scientific or not, but whether it is useful for making a prediction and is validated or not by the occurring of this prediction. The frontiers between sciences, philosophy, humanities and arts could then be definitively blurred. Scientists could dare to go beyond the limits of their field and of what they supposedly could infer using only “objective,” rationale and “rigorous” arguments and methods. They could use any method and any of their skills, but they could also be systematically controlled in return by the results of their experiments. They would be accountable to the reality, rather than to their discipline and their peers. Does this mean that we would transform societies into guinea pigs submitted to experiments? It may be that societies are guinea pigs already, but that politicians, not the scientists, are those who conduct the experiments. What I propose hence is to make scientists (those who “know”) and politicians (those who “decide”) work together, making them deal with the same objects: our physical and natural world and the societies built upon them. There will be two ways to assess the results of their common experiments: through social feedback resulting from democratic control, and through biophysical feedback resulting from scientific monitoring. This would help to avoid knowledge traps such as fashions, unique thinking, received wisdoms, myths, ideologies and dogma. In the end, societies would win. With this perspective, researchers would have huge responsibilities as they would be responsible, together with politicians, for their discourse on behalf of the societies they are studying. Blurring the frontier between science and politics would lead to redefine not only science, but politics as well. Aristotle, two thousand years ago, developed the concept of phronesis that was aimed at bringing science into a
16
similar direction. Phronesis is a "true state, reasoned, and capable of actions with regard to things that are good or bad for humans" (Aristotle 1976, in Flyvbjerg 2001). "Phronesis goes beyond both analytical, scientific knowledge (episteme) and technical knowledge or know-how (techne) and involves judgments and decisions made in the manner of a virtuoso social and political actor" (Flyvbjerg 2001, 2). Phronesis is thus the only state of knowledge that implies the responsiveness of the researcher vis-a-vis society. It is the most mature state of knowledge due to this responsiveness. If science could evolve to satisfy the phronesis ideal, it could enter a new paradigm where the overall progress of human societies would be its only goal. It would deal with the real world, instead of constructing idealized societies that run perfectly in models (the communist society, the market society, and maybe, soon, the “sustainable” society) but would be, at best, of little spiritual, ethical or esthetic value, and at worst enslavement for human beings. In this new model, objective and rational thinking could be used with and not against our identity, which is that of beings that think but also feel and desire.
4. Back to the subject I must now go back to my dissertation topic to see how these quite abstract considerations determined my research questions, hypothesis and methods. Because I try not to do conventional science, the question I ask is neither “what are the effects of slash-and-burn cultivation on the Malagasy environment,” nor “what are the solutions to stop slash-and-burn cultivation in Madagascar.” Some answers to these questions have already been given by scientists belonging to the fields of biology, soil sciences, crop sciences, social sciences, and economics, using conventional tools such as field experimentation in controlled sites, enquiries through questionnaires, laboratory analysis, statistics, econometrics, system analysis and
17
modeling. Their results show that slash-and-burn cultivation leads to deforestation, soil degradation, and poverty. The solutions they propose to stop it are good governance, state control over deforestation and burning, community-based natural resource management, alternatives to slash-and-burn such as lowland rice intensification and agroforestry, income generating activities, access to market, and birth control. Nevertheless, and despite the fact that almost half a billion USD was spent to implement these solutions, the Malagasy forest continues to be slashed and burned, the soils to degrade, and the Malagasy farmers remain poor, including in the small areas where most of the technical and financial support provided by this aid has been concentrated. So, were the wrong questions asked, were the wrong answers given, or did the implementation of these answers fail? Consistently with the previous sections, I believe there is a mix of these three reasons. But more importantly, I am convinced that escaping from this situation of failure can be achieved by answering to the broader questions: “why is there deforestation in Madagascar, why do farmers still slash-and-burn the forests, and why did the Malagasy state and its partners fail in addressing these issues?” These questions do not look like usual scientific questions but are those that matter. They are in the mind of everybody involved in preserving the Malagasy environment and in developing the Malagasy economy: donors, practitioners, development actors, and environmentalists. But they have never been asked by scientists, who limit their work to certain aspects of the reality that they select according to their backgrounds, the disciplines they belong to and the methods they acquired. This dissertation is an attempt to fill this gap. It concerns a broad object which includes a biophysical environment; a local society that uses this environment and has its own history and culture, its own way of interacting with it, it own land use (slash-and-burn agriculture); and a global society that gets more and more concerned by this
18
environment and this land use and also has its own identity and culture. These local and global societies live on the same planet, exploit the same global environment, exchange things through a common economic system and share ideas through various communication means. In other words, they depend on each other for the construction of their destiny. Understanding deforestation dynamics and the land uses associated with them lead to a questioning of their identity, the discourses they generate, the activities they implement and the interactions between these two realms. Two cultural poles are involved, which for simplicity I will call the modern and the traditional. Two systems of values, two systems of knowledge, two systems of power and two senses of esthetic are derived from these two poles. We need to be more than “competent performers” (Dreyfus and Dreyfus 1986, see Section 3.3) to understand these two poles and the systems associated with them, and to merge these elements into a single representation. We have to experience the reality in depth, not only to measure it. As observers, we belong to one of these poles only. We must then be aware of our own culture and identity and question ourselves while doing our observations. We will need, as far as our methods and capacities will allow us, to deconstruct our representations and reconstruct more objective and comprehensive ones. How else than using all the skills that have been granted to us, including intuition, and by creating positive synergies between reason and emotion, can we get a slight chance to achieve such comprehensiveness and reflexivity? And how else than by testing empirically our outcomes can we get objective feedback on our insights? This objective is quite ambitious. But as the subject encompasses a large number of narrower issues that have already been addressed, including in the area where I conducted my field work, I will not hesitate to dedicate large sections to summarize and critically examine these works and insert them into my comprehensive picture. The same knowledge does not need to be produced twice, but different realms
19
of knowledge have to be articulated together, questioned by their confrontation, improved in the light of new facts, and completed by filling up knowledge gaps. All these, rather than the addition of new knowledge on a narrow issue, is be the endeavor of this dissertation.
5. Conclusion In conclusion, I contend that the limits of rational thinking have to be recognized in order to valorize in science, or with science, other faculties that can also play a positive role in the advancement of knowledge, for the benefit of human societies and their environments. The exclusion of knowledge domains, which is the corollary of reductionism and positivism, leads to the creation of artificial objects that are incomplete, mutilated and nonfunctional. These artificial objects are not the reality. They are just human constructions or representations. Uncontrolled emotions, cultural determinants and personal affects contribute to shape them. Hence they may not reflect properly the reality and in this sense be irrational. But as they are regarded as “scientific,” they are considered to be the only grid for understanding ourselves and our world. They are excessively trusted and are allowed to determine our future. In order to produce a more comprehensive and relevant knowledge, we have to go beyond this mode of thinking that denies an essential part of our identity and skills; that makes us blind to the spiritual, ethical and esthetical dimensions of our lives; that isolates us from people with different values, but engaged with us in determining the future of our world. We must not refuse the positive role that intuition and emotion can play for the advancement of knowledge. To recognize the presence of these skills will help us to better control the biases introduced in our representations by affects and other subjective determinants. A permanent dialectics between reason and emotion, between logic and intuition, between us and the others, between science and politics,
20
is necessary for not passing over key realities that can reveal the essence, the functioning and the becoming of complex objects. Slash-and-burn cultivation and deforestation in Madagascar strongly involve groups of people who do not know, or at least did not experience each other. They are issues that reveal high tensions between the local and the global, the social and the natural, and for these reasons crystallize significant attention from organizations engaged in the search for a better future. They are hence appropriate issues to produce a new sort of knowledge using nonconventional modes of scientific enquiry.
21
PART II SETTING THE ISSUE: ENVIRONMENTAL DEGRADATION AND CONSERVATION IN MADAGASCAR
22
INTRODUCTION
This second part will set the issues addressed in this dissertation (deforestation and slash-and-burn cultivation) in their historical, geographical, social and political context. Chapter 1 will draw a broad picture of environmental degradation in Madagascar based on the existing literature on the subject. Chapter 2 will describe the environmental policies designed to address degradation, with a special emphasis on the National Environmental Action Plan, started in 1991 and still ongoing. It will be mostly based on the grey literature produced by actors involved in rural development and conservation programs in Madagascar, and on my personal experience as a technical advisor at the Ministry of Environment, Water and Forest 5 .
5
I worked for the FSP-GDRN Project (Projet Gestion Décentralisée des Ressources Naturelles du Fonds de Solidarité Prioritaire; Decentralized Natural Resources Management Project of the Priority Aid Fund) of the Malagasy-French Cooperation, from February 2004 to February 2006.
23
CHAPTER 1 ENVIRONMENTAL DEGRADATION IN MADAGASCAR
1. Introduction In this chapter, I will set the issue of environmental degradation in Madagascar in a historical perspective. Section 2 will present the general geographical features of the island. Section 3 will describe the historical transformation of Malagasy ecosystems, from the first human settlement to present. It will show that environmental degradation has occurred for as long as people have lived on the island, but may have accelerated in recent times. Section 4 will formulate a preliminary explanation of the causes of environmental degradation. It will achieve this by linking social and environmental histories, with a particular emphasis on the social groups involved in the clearing of eastern rain forests. It will show that Madagascar is still in a process of conquest by its inhabitants and that unequal competition leads to the marginalization of the weakest groups, for whom primary ecosystems constitute a refuge and provide the main resource base.
2. General geographical features of Madagascar For most people, Madagascar evokes an island with a luxurious environment where astonishing animals, such as lemurs and chameleons, can be found, but which is threatened by humans who burn its land and clear its forests. Madagascar is indeed a land with striking ecological features. Located southeast from continental Africa, it is the fourth biggest island in the world with an area of 587,000 km2. It occupied a central position in the Gondwana continent in Cretaceous time but separated early from other continents (Wells 2003). This ancient isolation explains the presence of
24
numerous archaic groups of plants and animals, the high rate of endemism and the complex affinities with other bio-geographical zones. Now three hundred kilometers away from the African coast, the island is marked by a mountainous axis extending from north to south, the highest point being 2876 m. This escarpment is bordered by uplands on its west side, with altitude ranging from 800 to 1500 m. The substratum is principally made of a highly eroded Precambrian basement (gneiss, granite or quartzites). Sedimentary basins and volcanic intrusions are also encountered but do not cover vast areas. The climate is subequatorial to sub-tropical. The highest rainfall (3000 mm) is recorded on the east coast which is subjected to trade winds coming from the Indian Ocean, and the lowest (400 mm) in the semi-arid southwest. The dry season lasts from June to October but is absent in the east and more marked in the west and the south. It coincides with the lowest temperatures. Cyclones are frequent. They cause flooding and significant infrastructure and agriculture destruction in the eastern regions. These geographical and climatic features result in a large variety of ecosystems and a very high biodiversity (Humbert 1965; Koechlin et al. 1974; Jenkins 1987; Du Puy and Moat 1996; Lowry II et al. 1997; Gautier and Goodman 2003). Due to clearing for agricultural purposes and frequent burnings, most of the land is now covered by secondary vegetation. Primary vegetation covered 18% of the land in the 1970s (Map 1, from Du Puy and Moat 2003), still less at present. Secondary vegetation is dominated by grasses and sometimes invasive shrubs and trees. The primary ecosystems related to this dissertation are the mid altitude and lower montane evergreen, humid forests. They form a corridor that stretches from the north to the south of the island but which is fragmented in several locations (Map 1). Throughout this dissertation, I will employ the term rain forest for more convenience.
25
Map 1 : Remaining primary vegetation in Madagascar Source: Du Puy and Moat (2003, 52). Derived from Faramalala (1995) and based on aerial photographs taken from 1972 to 1979.
26
3. Environmental and social history 3.1. First settlements and first environmental impacts Madagascar was settled relatively recently, around 2000 years ago (Mac Phee et al. 1991). The pattern of settlement is quite complex and not well known, with groups of humans coming at different times from different continents, mainly Indonesia and Africa. According to Deschamps (1972), the first settlers shipped from Africa and were Indonesian navigators. They had partially integrated some elements of the African culture and could have been accompanied by populations from the EastAfrican coast. The first proven record of human presence shows that groups of hunters may have been present on the Malagasy coast as early as A.D. 100 or A.D. 300 (Wright and Rakotoarisoa 2003). The first known encampment site is dated around A.D. 450 and “archeological evidence of settlement on the coast increases significantly for the period after A.D. 800” (Wright and Rakotoarisoa 2003, 113). Traces of human settlement during the first millennium are nevertheless scarce and evidences of important settlement in several parts of the island are only attested for the tenth century A.D. (Wright et al. 1996). Anthropogenic fires occurred in the second century A.D. on the west coast, and in the seventh century in the central highland (Wright et al. 1996; Burney 1997b). They were probably ignited for hunting. At this time, the natural vegetation in the highlands was a complex mosaic of savannah, grassland and forest (Lowry II et al. 1997, Burney 1987). Fossils of now extinct hunted animals such as giant lemurs, a pigmy hippopotamus, a local elephant, giant birds (Aepyornis), turtles and other big animals have been discovered (Wright et al. 1996). It is possible that hunting and anthropogenic fires caused the extinction of megafauna but this thesis is still disputed (Dewar 2003).
27
A broad trade network across the Indian Ocean was later developed by Arabs (Radimilahy 1998), as attested by “a rapid emergence of diverse and often complex settlement systems” (Wright and Rakotoarisoa 2003, 114). These ports facilitated a continuous migratory flow, including several new waves of Indonesian settlements and the importation and exportation of slaves. The first known major port, Mahilaka, was established in A.D. 1150 in the Ampasindava bay and flourished until about 1400. This settlement significantly affected the environment of the surrounding area: The evidence of pollen and charcoal from dated sediment cores raised from lakes and bogs throughout the Ampasindava region shows increasing burning and a marked diminution in arboreal pollen beginning about A.D. 1100. By about 1300, most pollen was from grassy and brushy vegetation (Burney 1999). There seems to be no explanation for this deforestation other than human activity. (Wright and Rakotoarisoa 2003, 113) Around 1400, the city diminished in size, villages were abandoned and arboreal pollen became frequent again (Burney 1999, in Wright and Rakotoarisoa 2003), confirming the anthropogenic character of vegetation change. Concerning the highland, many villages were found as early as the thirteenth century (Dewar and Wright 1993, in Wright and Rakotoarisoa 2003). Settlement was preceded or accompanied by an increase of fire frequency and the extension of grasslands, as shown by a new increase of charcoal sediment and grass pollen during the twelfth century. Castor bean pollen (Ricinus communis L. 6 ), a commensal plant found in disturbed areas near villages, also appeared during this period (Burney 1987, in Wright and Rakotoarisoa 2003).
3.2. Chiefdoms and kingdoms In spite of a relative unity with respect to culture and language, distinct organized societies developed in several parts of the island during the second
6
Euphorbiaceae.
28
millennium (Wright et al. 1996). The most powerful group was probably the Merina, who developed a kingdom with a high level of political organization and conquered at least half of the island in the nineteenth century. It probably originated from new waves of Malaysian and Indonesian settlers who moved from the east coast to the highlands (Deschamps 1959). The Merina established around a cluster of lakes and marshes, close to the present capital Antananarivo (Wright et al. 1996). They assimilated the local population, which was composed of descendants of ancient African settlers (Wazimba). They developed irrigated rice cultivation and raised cattle (Wright et al. 1996). Their land uses allowed the production of a surplus which made possible the existence of an organized hierarchical society, with a specialized governmental apparatus typical of states (Wright et al. 1996). This productivity, together with the raiding of slaves from the surrounding territories, allowed a tripling of the highland population between the fifteenth and eighteenth centuries (Wright et al. 1996). Due to this prosperity, the whole central highlands have been almost completely covered by grassland since at least the mid-eighteenth century (Wright et al. 1996). Warfare may have played a role in this transformation, as denuded land facilitated the displacement of armies and control over territory. Large fortified bastions appeared in the eighteenth century and the frontier was repeatedly moved outward (Wright et al. 1996). The end of the eighteenth and the beginning of the nineteenth century was the time of famous Merina Kings such as Andrianampoinimerina (1787-1810) and Radama 1 (1810-1828), who conquered most neighboring polities and created the Malagasy Kingdom (1815-1895). As the Merina did not efficiently control this vast territory, the unity was relative. Nevertheless, the political organization allowed the development of larger irrigation systems and favored trade exchanges with coastal villages. Guns and powder were imported and New World crops such as manioc were
29
introduced. Close relationships were established with foreigners, first with British, and later with French, leading to the establishment of the French protectorate in 1896 (Section 3.3.). As the Merina were the main beneficiaries of education effort during colonial time, their descendants still dominate the administrative and economic life in independent Madagascar. Other important kingdoms were the Betsileo and the Sakalava. The history of the Betsileo is quite similar to that of the Merina, with whom they share the appellation of “people of the highlands,” distinguishing them from the “Coastal people,” a mosaic of groups living on the coast. The Betsileo history started with the settlement of groups coming from the east and mixing with Wazimba (Deschamps 1959). They created kingdoms south of the Merina region, where the conditions for rice cultivation were less favorable due to a hilly topography. The Betsileo never achieved complete unity and were conquered by the Merina in 1830 (Kottak 1980). However, they benefited from education efforts under colonial times and still have significant political and economic power. The Sakalava created the most powerful coastal kingdom. They once had a territory that covered extensive areas from the west coast to the central highlands in the seventeenth century, but were also conquered by the Merina, in the eighteenth century. Most other groups were village societies whose centers were located on the coast or closer to it. They benefited from links with Arab trade networks and had a quite sophisticated cultural life. Their military activities were limited to raids aimed at supplying slaves. Among these groups are the Betsimisaraka and the Tanala, who are the two main groups practicing slash-and-burn cultivation in the eastern rain forests. The Betsimisaraka (“the many who will not be sundered”) constitute most of the population in Beforona, the commune which will be studied in Part III. They result
30
from the unification of several tribes of the eastern coast by King Ratsimilaho in the early eighteen century. But this unification did not last due to internal conflicts and slaves raids by Sakalava and Sihanaka people. The Betsimisaraka were conquered by the Merina in 1823. Nevertheless, they have always been quite prosperous, due to the abundance of rice, tubers and fruits grown in slash-and-burn cultivation systems, and to trade exchange and piracy along its coast. They transformed the landscape of the eastern region, which early explorers described as a patchwork of dense forest, secondary vegetation and cultivated and grazed land (Flacourt 1658, Ellis 1858, in Agarwal et al. no date). The impact of colonization on the Betsimisaraka has been described by Althabe (1969) and Cole (2001). The eastern coast provided favorable conditions for growing cash crops such as coffee, which interested colonists. On the other hand, the dominant land use (slash-and-burn cultivation or tavy in Malagasy language) was stigmatized as a bad practice as we will see later. The Betsimisaraka suffered much from the imposition of a per capita tax, the obligation of growing cash crops to pay this tax, and the repression of slash-and-burn cultivation. They resisted by claiming their right to practice tavy, which was a way to express their culture and maintain their identity and unity (Althabe 1969). However, lowland rice cultivation was part of their farming system as well (Le Bourdiec 1974). The Tanala, or “People of the Forest,” is a group resulting from the aggregation of individuals fleeing from raids and slavery or removed from the coast and the central highlands by new settlers. The first Tanala were probably black people (Wazimba) coming from the coast and the highlands and fleeing other populations to find refuge in the forest. They practiced hunting, gathering, slash-and-burn agriculture and had a low level of social organization. In the twelfth and fifteenth centuries, they have been joined by noble Islamized groups who migrated to these areas. Migrations
31
of Tanala seem to have increased in the sixteenth century. This was probably due to the arrival of new settlers on the east coast and in the coastal valleys, to the development of the Merina kingdom in the highlands, and to the development of slavery and trade (Deschamps 1972). In the eighteenth century, the Tanala reached a higher level of social organization by associating with the Zafirambo chiefs, descendants of coastal Princes taking refuge in the forest (Beaujard 1983). The Zafirambo constituted the aristocracy, unified the various tribes, and created the Ikongo kingdom (Deschamps 1972). This process was a reciprocal assimilation rather than the subjection of one to the other, as attested by traditional tales (Beaujard 1985). It allowed the Tanala to resist to dominant groups such as the Merina. The current chiefs of the Tanala tribes (the Mpajanka) are still descendants of the Zafirambo. Changes in land use occurred at that time and wet rice cultivation was developed in the most favorable valleys (Beaujard 1985). The existence of both lowland and upland rice systems is attested by a traditional Tanala tale which explains that this combination is a strategy to reduce agricultural risks (Beaujard 1985): When Kotofamandrika, the bird hunter, and Zanahary, the god princess, founded the Tanala kingdom, they sowed rice in a cleared forest (tavy). But birds and wild pigs destroyed the field. Then they sowed rice in a paddy fields, but another kind of bird destroyed the harvest. Then they decided to sow both kind of fields, on slopes and in bottom wet lands, so that the animals will not eat the rice. The development of paddy fields has been shortened by the extension of the Merina kingdom in the eighteenth and nineteenth century. The Tanala were obliged to pursue their flee to the forest and to abandon their land, for conserving their freedom or not becoming slaves (Beaujard 1985). Nevertheless, the need to resist to the Merina kingdom encouraged them to organize into a stronger chiefdom (Beaujard 1985). The immigration and assimilation of Betsileo, who had more experience in irrigated rice
32
cultivation, allowed a second development of paddy fields in the second part of the nineteenth century (Beaujard 1985). But in the twentieth century, colonization exerted a new pressure on Tanala people, resulting again in the abandoning of their land and fleeing to the forest (Deschamps 1959). This history shows that the Tanala have various ethnic origins and are instead defined by their marginal social situation, their standard of living in the forest and their quest for free land. Migration of Betsileo people to the Tanala region is ongoing today, as attested by the existence of mixed Betsileo-Tanala villages where the social organization of the Tanala is adopted by both entities. This history also shows that Tanala people are not simply forest tribes practicing slash-and-burn agriculture. As most other groups in Madagascar, they have experience in both lowland and upland rice cultivation. But as they have been repeatedly marginalized and removed to remote areas, they never had the opportunity to definitively establish permanent settlements. Land use would thus be more a question of physical environment and economic resources than a matter of culture and traditions. Establishing sustainable paddy fields is a long process which can take generations, especially in mountainous areas where large plains are absent, and when political instability and conflicts impede farmers from increasing their social and economic capital.
3.3. The first Europeans 7 The first Europeans visiting Madagascar were the Portuguese around the year 1500. They established trade settlements and missions in the sixteenth and seventeenth centuries but later abandoned the island. The Dutch, the British and the French followed them. Until the end of the nineteenth century, the missions organized military expeditions into the central regions of the island but had no clear policy 7
The main sources used for this section are Brown (1995) and Deschamp (1959; 1972).
33
concerning the exploitation of resources and colonization. Colonization at that time was a failure despite the official annexation of Madagascar by France, in 1664. European settlers were successively massacred or abandoned the island and Madagascar became a shelter for pirates until the eighteenth century. When piracy declined, around 1825, the British and French created new establishments on the coast. The British developed an alliance with the dominant group, the Merina, on whom they had a large influence. The French colonized the coast but also developed links with the Merina Kingdom. In 1890, the British abandoned the island to the French in exchange for Zanzibar. France established a protectorate in 1895 and the colonization followed, opening a new period of social disturbances.
3.4. From colonization to present The twentieth century history will be narrated in more detail in Chapter 2 but needs to be briefly summarized here, in order to understand the dynamics of environmental degradation. The colonization by the French led to the development of infrastructure and organizational systems typical of a modern state, improving education and health services. But the goal was rather to serve the interests of the mother country. Taxes and forced labor (corvées) were instituted, land was taken to indigenous people and granted to colonists and cash-crops were introduced, often in a coercive manner. Dissatisfaction of the indigenous population increased until a popular strike occurred which was repressed by the military in 1947. Colonial pressure was loosened after these events and the country entered into a state of relative prosperity, leading a demographic boom in the second half of the century. Independence was gained in 1960. The political life of the independent state was dominated by the Merina, who were the main beneficiaries of educational efforts by the French. A socialist regime
34
was instituted in 1972 and enforced by a coup in 1975. But Madagascar later liberalized its economy and opened to foreign investors and international aid agencies, including conservation organizations and lobbies in the mid 1980s. An ambitious national environmental program was launched in 1990, with significant support from the international community. The current population is estimated to be 18.5 million, with an annual growth rate of 3% (CIA 2007) and an average density of 26 inhabitants/ km2. Most people live in rural areas and practice agriculture, which represents 26.9% of the gross domestic product (CIA 2007).
3.5. Deforestation in the twentieth century The pressure on ecosystems increased from colonization to the present time, provoking massive deforestation. Successive measurements of forest cover were done using different methods and the definition of “forest” varies according to the authors. For these reasons, comparison is difficult 8 and there is controversy about the exact rate of deforestation (Agarwal et al. no date). All estimations agree that it was very significant, Madagascar making no exception to what happened in the rest of the world, but exaggerated figures are sometimes produced: it is estimated that Madagascar lost about 12 million hectares of forest between 1960 and 2000, effectively reducing forest cover by 50% in just 40 years. Following the launch of the National Environment Action Plan in the late 1980s, deforestation rates have since declined from over 400,000 ha/year in 1975-1985 to around 100,000 - 200,000 ha/year during the 1990s. Based on satellite imagery, it is estimated that the total area of natural forest in Madagascar declined from 9.4 million hectares in 1993 to
8
According to Jenkins (1987), forest covered 25 to 30% of the land in Madagascar in 1987. According to Chauvet (1972), it covered 16.7 million hectares in 1972, including 4.3 million hectares of degraded forests. According to the World Bank (1988), 12 million hectares of forest remain in Madagascar, which represents 25% of the initial cover. Jolly (1989) estimated 7.7 to 10 million hectares of natural forest still remained in 1989, disappearing at a rate of two to 4 % per year. At this rate, the last patches of primary vegetation may disappear in a few decades.
35
8.5 million hectares in 2000, reflecting a national average rate of deforestation of about 0.86% per year. (World Bank 2004, 6) More precise data are available concerning the eastern rain forests, which are certainly the richest in terms of biodiversity. Sussman et al. (1994) and Green et al. (1990) estimated that rain forest probably covered 11.2 million hectares in eastern Madagascar before colonization by the French, of which 7.6 million remained in 1950 (calculated from the vegetation map of Humbert and Cours Darne 1965) and 3.8 million hectares in 1985 (from analysis of Landsat satellite images). According to these figures, the pace of rain forest clearing between 1950 and 1985 would have been 111,000 hectares per year. In the most densely populated areas, this pace would have decreased from 2.5% or 51,000 hectares per year, between 1950 and 1973, to 0.79% or 16,000 hectares per year between 1973 and 1984 9 . For Green and Sussman (1990), this decrease after 1973 may result “from a diminishing pool of accessible forest because of the elimination of forests on all but the steepest slopes.” These figures are, however, disputable. Dufils (2003) reviewed existing data sets and compared the estimation methods. He concluded that only three data sets can be compared concerning the eastern rain forests: the estimations by Humbert and Cours Darne (1965) (10.74 million hectares in 1953), by IEFN 10 (6.25 million hectares in 1993) and by JRC 11 (5.53 million hectares in 1999). According to these sources, 102,000 hectares would have been deforested every year between 1953 and 1993, and the same pace would have persisted between 1993 and 1999. The recent efforts of environmental conservation (Chapter 2) could have had no impact on the Malagasy rain forests. In order to answer this question, Hawkins and Horning (2001) compared deforestation rates between regions targeted or not targeted by environmental 9
No total coverage is available for the year 1973. Inventaire Ecologique Forestier National (National Forest Ecological Inventory). 11 Joint Research Institute of the Space Application Institute, Ispra, Italy. 10
36
programs during the 1990s. They measured a 0.3% loss per year in the Zahamena Mantadia corridor 12 , for the period 1993 to 2000, and a 0.5% loss in the RanomafanaAndringitra corridor, located in southeast Madagascar. These two sites are targeted by USAID 13 funded projects. These authors further measured a 1.1% loss in the Anosibe An’ala-Ranomafana corridor, which is not targeted by USAID projects and has been neglected by environmental programs (still for the period 1993-2000). The results of the comparison are exhibited in the form of posters in the office of many Malagasy state representatives, such as at the Water and Forest General Directorate. But in reality, the correlation between intervention and less deforestation does not mean there is a causal relationship. The situation may differ from one area to the other in terms of population migration, economic opportunities and ecological constraints. These factors, rather than the impact of projects, could explain the differences measured by Hawkins and Horning (2001). Project interventions can further lead to displaced pressures due to the mobility of Malagasy people. The higher rate in the control site, as well as the lower rate in the targeted sites, could both be the consequences of environmental programs.
4. The social causes of environmental degradation 4.1. Deforestation and the first agricultural frontier The main causes of deforestation during the twentieth century may have been logging, slash-and-burn agriculture, intensive extraction of forest products, and the establishment of plantations for export (Jarosz 1993). Construction of roads and railways induced the colonization of new areas, the burning of forest for producing charcoal, and the concession of forest land for extraction of precious woods, rubber,
12 13
The commune of Beforona, which will be studied in Part III, is located in this area. United States Agency for International Development, a major donor of NEAP.
37
raphia and other products (Olson 1984). Despite this diversity of causes, slash-andburn cultivation (tavy) was regarded as the main cause of deforestation and the French administration tried to regulate its practice (Chapter 2). There can indeed be little doubt that tavy was the most significant proximal cause. But if we adopt a more comprehensive frame, the role of slash-and-burn may not have been so significant. Slash-and-burn cultivation implies vegetation regrowth on the cleared land, after one or two years of cultivation. This secondary vegetation can be cleared again and the system can sustain itself, at least when the population density is low and stable. This could have been the case in Madagascar until 1950. According to Jarosz (1993), the population growth in Madagascar was very close to replacement level between 1900 and 1941, whereas huge areas of forest have been cleared during this period. Censuses between 1900 and 1950 expressed the anxiety of the French over population stagnation (Olson 1984), while millions of hectares of forest were cleared. The dramatic demographic explosion only occurred after 1950 (Olson 1984). Rather than the demographic increase or the unsustainability of tavy, the real “cause” of massive clearing could have been up-side dynamics obliging farmers to clear new land in forests. I already pointed out such up-side causes in the case of the Tanala people, in Section 3.2. These up-side causes may have been exacerbated and generalized to the whole country during the colonial period, leading to the disorganization of the local economy and land tenure systems. The institution of forced labor and taxes, the establishment of colonial concessions and forest plantations, and the granting of indigenous land to colonists disorganized farming systems and customary modes of regulation. They created monetary needs which encouraged migration of indigenous people in a search for new land and income opportunities (Jarosz 1993; Olson 1984). As forest is the only land that appeared to be in free access and with no owners,
38
farmers with few resources and little capital had no other choice than to exploit forest biomass when pushed away from their previous land by colonists’ settlements or by the pressure of forced labor and taxes. One could thus expect deforestation to decrease after independence. But we saw in Section 3.5 that this was not the case. That it did not decrease could be explained by poverty and the persistence of inequities in development driven by market forces. “Real income per capita has fallen by 40% since 1960 to about 240 USD per person in 1999” (Dorosh et al. 2003). During the 1980s, the Malagasy state was in bankruptcy 14 , forest controls stopped and poverty peaked, leading many communities to increasingly depend on natural resources, as we will see in the case of Ambodilaingo in Chapter 5. During the 1990s, market reforms engendered an overall GNP 15 growth but also created an increasingly risky economic environment, which affected the most vulnerable farmers and may have led to more deforestation (Barrett 1999). According to Paternostro et al. (2001, no pagination) “populations in rural areas witnessed persistent increases in poverty despite market reforms.” In response, small scale agricultural households “have been extending their land use by clearing and cultivating increasingly fragile lands” (Paternostro et al. 2001, no pagination). Today (in 2001), 69.6% of Malagasy people live under the poverty line, and 85% of the poor live in rural areas (Republikan’I Madagasikara 2003).
4.2. Erosion and the second frontier This massive deforestation does not mean that a large part of the country has been put under cultivation. According to Jolly (1989), only 5% of Madagascar is under cultivation and an additional 10% or so is regarded as cultivable. According to
14 15
The economical crisis of the 1980s will be analyzed in Chapter 2. Gross National Product.
39
Koechlin (1974), grassland covered an estimated 420,000 km2 in 1974, i.e., 72% of the surface of the island. This number may, however, include savannah, as aerial or satellite photographs do not clearly distinguish between the two. Grasslands have, however, some positive functions in the Malagasy agricultural landscape. They are used as pastures and are frequently burned, to provide early forage when the first rains arrive and to favor runoff and provide nutrients to paddy fields located in bottom valleys (Raison 1984; Kull 2000). But most fires are uncontrolled, cover extensive areas, and extreme erosion often occurs, creating huge gullies called Lavaka (Wells et al. 1997). The consequence is extensive flooding and silting, with significant negative impacts on crop production in irrigated plains downstream (Kramer et al. 1997). Due to their prevalence, grasslands constitute a second agricultural frontier to be conquered by farmers who have the capacity to develop more intensive land uses, to the detriment of those who currently occupy the land and practice extensive systems (mostly pastoralism). Conquest of the “grassland frontier” also needs more investment to be achieved than in the case of the forest frontier. Fewer natural resources are in place and investments have to be made in the form of work to complete improvements such as terracing, irrigating and reforesting. It is only when they already have some assets, or when they have no other choice, that farmers in the search for new land direct themselves to this second frontier. Those who establish more successfully and found the most prosperous communities are often Merina and Betsileo people, who have higher economic levels and have already developed intensive land uses and faced the problem of natural resource shortage for decades or centuries. Raison (1984) analyzed these dynamics of colonization in detail.
40
5. Conclusion In conclusion, there is a very acute environmental problem in Madagascar. Land degradation has been occurring ever since humans settled on the island, and continued shrinkage of the remaining primary ecosystems will lead to a very significant loss of biodiversity if it is not stopped in the next few decades. Madagascar, however, is not a particular case. Agricultural or pastoral activity cannot be implemented without transforming ecosystems and clearing primary forests. For this reason, humans have significantly transformed the ecosystems of the earth for as long as they have existed (Thomas 1956; Turner et al. 1999) and fires and forest clearing have played a central role in these transformations (Ponting 1991; Redman 1999; Pyne 2001; Mazoyer and Roudart 2001; Williams 2006). I can further assert that the main proximal cause of deforestation is the seeking of new land for cultivation, and that forests are mostly cut by farmers who practice slash-and-burn cultivation (tavy). But these farmers must not be blamed for that. Deforestation results from the mobility of the whole population, the extension of the most prosperous groups and the marginalization of the poorest ones. The expansion of the Merina kingdom and colonization by the French were probably the main forces that put the Malagasy forests and its resources under pressure. More recently, market reforms have not reduced, and may even have increased these pressures. Tavy farmers have been stigmatized as the main agents of forest clearing but they are just one aspect of a more global and complex dynamic that involves other groups. They are in the first line of the move to the forest, but they are not the main force that provokes this move. For this reason, they cannot be regarded as the main cause of deforestation. They are only the proximal cause, playing an instrumental role for the “benefit” of all actors that collect the advantages of human expansion. In short, “development,” in a context of limited regulation, may be the real cause of deforestation. Conservation discourses
41
need to build their representations inside this broad picture in order to avoid the construction of fragmented, incomplete and sometimes false “realities.” If we just focused on the last step in the chain of reactions that leads to deforestation, i.e., on what happens to the soil, the vegetation and the economy when a tavy farmer clears a piece of forest land to cultivate it, then we may fail in finding solutions to the problem.
42
CHAPTER 2 ENVIRONMENTAL CONSERVATION POLICIES AND PROGRAMS IN MADAGASCAR
1. Introduction In this Chapter, I will analyze the genesis of environmental policies and programs in Madagascar. Sections 2 and 3 will describe the first environmental and forest policies under the Merina kingdom and during the colonial period. Sections 4, 5 and 6 will continue this story from independence in 1960 (first republic), to the socialist revolution of 1972-75 (second republic), and the return to a liberal economy in the 1980s. Sections 7, 8 and 9 will analyze the three phases of the National Environmental Action Plan, with a special emphasis on its impact and on the socioeconomic and political context of its implementation. Section 10 will conclude and make the transition to Part III.
2. The first forest management and conservation efforts Before the arrival of Europeans and the organization of a modern state, the Malagasy society was feudal. The land was owned by kings or lineage leaders who allocated it to the population. The forest served as a reservoir of land for the installation of new households. The first known measures for environmental protection were promulgated by King Andrianampoinimerina (1710-1790), who transmitted them orally. These measures revealed awareness that the forest was the only resource available for the poor, and that it was not inexhaustible. Here is the forest,” said the King, “indivisible property where the orphans, the widows, and all the unfortunates can find their subsistence, because otherwise, they would not have any resource, nothing to sell. Nobody has
43
to worry them, because the small activities they will have will allow them to become, like the others, good subjects of their sovereign… Concerning my vassals and relatives, they will be free to take in the forest the materials they will need, for example to build a house, because nice houses are the beauties and the glory of my kingdom. It is however forbidden to burn the forest and to burn its wood, except to make charcoal for the forge. This interdiction is in your interest and is aimed at avoiding the complete and irremediable disappearance of the forest 16 . Radama (1861-1863) issued the first known legislation dealing with forest products extraction, the “Charte Lambert,” signed on June 28, 1855. This regulation authorized the free exploitation of mines and forests by a company owned by a foreign investor, Lambert. The Code of 70 articles, promulgated in 1862, included five articles of the “charte” and added two more for the regulation of mining and forest products extraction. On March 29, 1881, Queen Ranavalona II promulgated a code of 305 articles, which is regarded as the first significant move from a feudal society to a modern state. It included six articles (101 to 106) dedicated to forest protection and management (Lavauden 1934). Article 101 forbids forest burning and punishes it by “ten years in iron” (in prison). Article 102 forbids charcoal production in or close to forests, while article 103 forbids the use of big trees to make charcoal. Article 104 forbids settlement inside forests, a measure probably motivated by the difficulties of administrating populations living in remote places and fleeing from administrative pressures (Lavauden 1934). Article 105 forbids clearing and burning of forests for agricultural 16
From Julien (1909), in Lavauden (1934), translation from French to English by Pollini. Original citation in Lavauden (1934): “Voici la forêt, patrimoine non susceptible de répartition entre mes sujets. C’est là que les orphelins, les veuves, et tous les malheureux viendront chercher leur moyen d’existence, car sans cela, ils n’auraient aucune ressource, ne pouvant rien vendre. Que personne ne les inquiète, car les petites industries auxquelles ils se livreront leur permettront de devenir, comme les autres, des sujets utiles à leur souverain …. Quant aux grands, mes vassaux et parents, ils auront, eux aussi, toute liberté pour faire prendre dans la forêt les matériaux dont ils auront besoin, chaque fois, par exemple, qu’ils voudront faire bâtir une trano kotona (maison en bois), car les belles constructions sont l’agrément du pays et la gloire de mon royaume. Il est, néanmoins, interdit d’incendier la forêt et d’en brûler les bois, si ce n’est pour fabriquer le charbon qui sert aux travaux de forge. Cette interdiction, prise dans votre intérêt, a pour but d’éviter la disparition complète et irrémédiable de la forêt.”
44
purposes and limits the practice of slash-and-burn to secondary vegetation. Article 106, eventually, forbids cutting or damage of trees growing on the littoral. This legislation proposed very severe penalties but was hardly enforced.
3. The colonial period When Madagascar became a French Protectorate, a Forestry Mission was immediately sent by the French Ministry of Agriculture in 1896 (Lavauden 1934). Its objectives were to assess the forest resources, to organize their rational extraction, to control the permits that had been allocated (mostly to European that exported wood to Mauritius and La Reunion) and to prepare legislation that would organize forest product
extraction,
define
repressive
rules
and
create
a
Forest
Service
(Ramanantsoavina no date). According to Ramanantsoavina (no date), the key dates of the development of Malagasy modern forestry were: •
1897: Publication of the first legislative order (July 5), launching of the first reforestation program in the highlands, creation of nurseries in Antananarivo and introduction of exotic species,
•
1900: publication of the first forest decree (February 19),
•
1903: first “scientific” harvesting of nontimber forest products (rubber vines),
•
1905: establishment of the first forest map,
•
1908: creation of the first forestry research station in Analamazaotra,
•
1910 to 1914: important reforestation program along the railways in order to provide fire wood to the locomotives’ steam engines,
•
1911: introduction of pine trees,
•
1913: publication of the second forest decree (August 28),
•
1917: creation of the first body of Malagasy forest rangers,
45
•
1925: development of fisheries in the forestry research stations
•
1926: creation of five forest districts,
•
1923: creation of ten forest Natural Reserves (an eleventh will be added in 1929) covering 400 000 hectares,
•
1930: publication of the third decree (November 17), modified later but still valid at present,
•
1938: ratification of the International Convention for Fauna and Flora Protection in Africa, adopted at the London International Conference of October 8, 1933,
•
1946: creation of the first forestry school in Angavokely, and
•
1958: creation of the two first National Parks, opened to scientists and tourists only (Montagne d’Ambre and Isalo) and of 20 Special Reserves with less restrictive rules.
Until 1930, the enforcement of the forest legislation was almost ineffective. There were only two forestry officers for the whole country until 1927, assisted by no more than five forestry employees until 1922 (Lavauden 1934). Local administrators were flexible and continued to tolerate indigenous practices and forest clearing (Lavauden 1934). Facing the contradiction between conservation objectives and economic local realities, the legislation hesitated concerning the way to address fires. According to Kull (2004), Bertrand and Sourdat (1998), Ramanantsoavina (no date) and Lavauden (1934), pasture fires were forbidden in 1897, authorized in 1904, limited to indispensable ones in 1907, forbidden again by the second decree in 1913, and tolerated after 1913 because exceptions to the ban became the rule. The policy and the legislation about tavy had similar hesitations until the early 1920s. Madagascar was also visited by many biologists and naturalists during the nineteenth century, one of the most prominent figures being the French explorer and
46
naturalist Alfred Grandidier (Andriamialisoa and Langrand 2003). Under their influence, the French Governor Gallieni founded the Malagasy Academy in 1902. The first years of scientific surveys were accompanied by polemics concerning the original extension of forests and the type of ecosystems that existed in the highlands and the west before the arrival of humans. Perrier de la Bâthie (1921) estimated that 200,000 hectares of forest were cleared every year. He noticed that all species of the highlands were adapted to fires, with a capacity to reproduce vegetatively by their underground parts or sexually by producing seeds under a short cycle. Humbert (1927) renewed the arguments that tavy and pasture fires were the cause of this destruction. He drew the picture of an island that was completely covered by forest, woodland or bush-land before the arrivals of humans, except for a few swampy lands with bad drainage (Humbert 1949). Lavauden, Chief of the Forest Service appointed in 1928, went farther. He declared that Madagascar had only two botanical regions, the oriental region covered by evergreen forests and the occidental region covered by deciduous forests. He later contended that before the uplands were deforested, the climate was wetter and evergreen forests extended across the whole island, giving to it a uniform cover (Bertrand and Sourdat 1998). He created the myth of a lost evergreen forest paradise that has influenced environmental policies until now, as we will see. The third forest decree, adopted in November 1930, was strongly influenced by this myth and by the botanists’ call for more conservation efforts. It put a ban on any form of vegetation fire. According to article 36, “the fires, the destruction or clearing of forests and the bush fires for the preparation of crop fields or pastures are forbidden in the whole domain of the colony 17 .” When individuals were not identified, the 17
Translation by Pollini. Original text in French: “les incendies, destructions ou défrichements de forêts et les feux de brousse pour la préparation des cultures et pour les pâturages sont interdits dans tout le domaine de la colonie.”
47
communities were regarded as responsible and fined (article 58). Pasture fires were nevertheless authorized under certain conditions. The 1930 decree was followed by a series of strict instructions for its enforcement. According to Lavauden (1934), it was, however, not strictly enforced and exceptions supported by more or less legal administrative acts still proliferated. The forest services were, however, significantly strengthened (six forest officers and 39 employees working in six forests districts in 1931) and the population heard of these new authorities and of the regulations they enforced. This raised an awareness that something significant had changed: ancestral land now belonged to the state, an external and quite abstract entity, and accessing forest land through customary rights was rendered illegal (Bertrand and Sourdat 1998). This change was confirmed by the arrival of colonists and other outsiders who were granted land ownership and authorizations for forest resource extraction, with no respect for the traditional land tenure system. In consequence, local means of regulation started to erode, while the colonial administration was not strong enough to enforce the law, except along the main roads. Competition for access to natural resources increased, especially when colonists took part in it. Burning became a way to contest the taking of land by outsiders and to claim indigenous rights. This change, and the demographic explosion after 1950, are certainly the key explanation for the acceleration of environmental degradation in Madagascar during the twentieth century, as Jarosz (1993) and Kull (2004) argued. Parrot 18 (1924, in Bertrand and Sourdat 1998) 19 gives an example that illustrates this change. In the Antsirabe region,
18
Reference not in list of work cited because cited but not readable in the bibliography of my copy of Bertrand and Sourdat (1998). 19 Page of document not readable on my copy of Bertrand and Sourdat (1998).
48
the most exalted indigenous people declared that it was better to destroy the neighboring woods than to see them fall into the hands of others. They ignited fires secretly and thousands of hectares of beautiful forests disappeared in a few weeks. As soon as they belonged to the state, they were no longer protected. Parrot was the first to propose to implement community based natural resource management to address environmental degradation. The taking of land by the state and by colonists was not the only cause of tension. Farmers fled to escape forced labor as we saw in Chapter 1. They moved to remote areas, mostly into forests that they cleared to open new agricultural land. In some cases, the administration forced them to have a permanent settlement along the roads, to abandon tavy and to cultivate crops for export. Althabe (1969) showed how the Betsimisaraka people developed a dual system to resist to these pressures. They established permanent dwellings and coffee plantations but continued to practice tavy in the forest, far away from administrative controls. Submitted to these pressures, the rural population showed a discontentment that culminated in 1947. As in most colonized countries, the Second World War provoked an increasing demand for services from the population. Labor and products were requisitioned to support the war effort. But the war also showed that colonial power could lose a battle. People from the eastern region, led by strengthening nationalist movements, rebelled. Forest plantations were a prominent target (Kull 2004). The main forestry research station, in Analamazaotra, was destroyed and evacuated; the installations of four forest reserves were destroyed; in the eastern region many forest posts were attacked and the extent of tavy increased significantly (Ramanantsoavina no date). After the rebellion was repressed, and until the last years of the colonial period, repression decreased to a more realistic approach. Collective fines were seldom applied and the need to support farmers to develop alternatives to tavy, such as
49
irrigated rice cultivation, was more and more widely recognized. Pasture fires were again a subject of debate. Some administrators regarded them as necessary to reduce the accumulation of combustible materials. During the 1950s, economic opportunities increased, the infrastructure developed, the economy expanded rapidly and the population started to boom. But not until independence would a new law be issued to provide a legal framework for a softer approach to fire and tavy control.
4. The first republic 4.1. Economic context During the first decade after independence the first republic, under the presidency of Tsiranana, was a period of political stability and relative economic prosperity. There was a regular increase in industrial production and cash-crop export (coffee, vanilla and cloves), a self sufficiency in rice and a low inflation rate. The trade balance was in deficit, but this was never a threat due to membership in the Franc Zone and the substantial aid received from multilateral and bilateral donors (Brown 2000). Economic growth benefited the urban elite more, but the countryside also enjoyed its effects: In international statistics, Madagascar ranked as one of the twenty poorest countries, with a per capita income of only 80 USD in 1960, but it did not appear so. Many of the peasants outside the monetary economy lived relatively comfortably in well-constructed houses with their own ricepatch, poultry and cattle. (Brown 2000, 304) Nearly everyone “had access to the basic requirement of food, shelter and clothing and there was little sign of real hardship or deprivation.” (Brown 2000, 304). The road network was more developed than it has ever been since, providing to most regions an access to market and allowing state services to work more efficiently.
50
4.2. The new legislation about vegetation clearing and fires Not only the economy was more favorable to the population at this time. The repression of fires and tavy was more reasonable than it used to be under the colonial period. A new law was issued in 1960 to regulate vegetation clearing and fires (Gouvernement de Madagascar 1960). It is still valid today. Article 2, modified in 1962 (Gouvernement de Madagascar 1962) defined clearing as the series of operations aimed at allowing the cultivation of a piece of land that was previously covered by ligneous vegetation, and which consists in the slashing of all or part of this vegetation, followed or not by incineration and aimed at establishing crops or plantations. 20 The ordinance 60-127 is the reference to which Forest Services still refer. According to its articles 3 and 4, clearing is forbidden in the national forest domain, on slopes of more than 50%, on land subject to intensive erosion, on sand dunes of the littoral and on river banks. The Forest Service can grant clearing authorizations on other types of land. Concerning fires, three types are recognized by the ordinance 60-127: -
-
cleaning and cultivation fires, whose goal is to either incinerate the low density ligneous vegetation that covers a permanently cultivated piece of land in order to prepare new crops, or to clean the surroundings of perennial crops or of land dedicated to settlement for social or economic purpose, pasture fires, aimed at renewing the herbaceous vegetation on pasture land, wild fires, that propagate without control, without limit, across any type of vegetation and with no economic utility.21
20
Translation by Pollini. Original text in French: “on appelle “défrichement” la suite des opérations qui destinées à permettre la mise en culture d’un terrain préalablement recouvert d’une végétation ligneuse et qui consiste dans l’abattage de tout ou partie d'une végétation ligneuse suivi ou non d'incinération, dans le but de procéder à des plantations ou semis d'ordre agricole.” 21 Translation by Pollini. Original text in French: « Les feux “de culture” ou de “nettoiement” ont pour but “soit d’incinérer la végétation ligneuse peu dense qui recouvre un terrain cultivé de manière permanente en vue d'y préparer de nouvelles cultures, soit de nettoyer les abords de champs de cultures pérennes ou d'installation à des buts social et économique”; “Les feux de pâturage qui ont pour but le renouvellement de la végétation herbacée sur les pâturages”; “Les feux sauvages qui se propagent sans
51
Tavy practiced on dense ligneous vegetation matches with none of these categories, probably because the section about clearing indirectly addresses them. Tavy practiced on low density ligneous vegetation corresponds to the category “cleaning and cultivation fires.” According to article 8, “cleaning and cultivation fires can be ignited without authorization if they occur outside the national forest domain and the artificially reforested land.” 22 Tavy on low density ligneous vegetation is thus authorized by the law 60-127, without even needing authorization. The regulation, however, presents some ambiguity due to the difficult interpretation of the term “low density ligneous vegetation.” Most fallow land in eastern Madagascar is invaded by exotic weeds such as Rubus mollucana and Lantana camara, which are ligneous in the sense that their stems produce lignin, but which do not produce real wood, except on short stumps. These species are not even used as firewood and the term sub-ligneous may be more appropriate to qualify them. If they are mixed with herbaceous species, there is no ambiguity about the nondense character of the ligneous vegetation and tavy can be practiced without authorization. But if ligneous or sub-ligneous species dominate and form a dense vegetation, it is not clear whether this vegetation should be regarded as “low density ligneous” or “dense ligneous,” and whether a tavy practiced in it would imply authorization for “clearing.” Article 6 raises another ambiguity. It indicates that cleaning and cultivation fires only concern “permanently cultivated land.” If “permanently cultivated” means that crops are sown every year on the same land, then tavy fires on low density ligneous vegetation cannot be regarded as cleaning or cultivation fires. The problem is that they would not be addressed by the law in this case, because they could not be controle, sans limite, a travers n’importe quel type de végétation et sans utilité d’ordre économique.” (article 6). 22 Translation by Pollini. Original text in French: "Les feux de culture et de nettoiement peuvent être allumés sans autorisation, à condition que ce soit hors du domaine forestier national ou d'une parcelle artificiellement reboisée." (article 8).
52
regarded as a clearing in the sense of article 2 either. If we consider, to the contrary, that the fallow period is part of the cultivation cycle, then a tavy land can be regarded as a permanently cultivated land. The first interpretation makes more sense for agronomists, who usually use the term “permanent cultivation” to designate cropping systems with very short or no fallow periods. But only the second interpretation is possible if the legislation is to address all situations.
4.3. Forest control Due to a situation of relative stability and prosperity, the forest may have been under lower pressure during the first republic. The Forest Service had a level of efficiency they have not experienced since. Forestry staff remember this period as the golden age of their career. Rangers were respected and feared by the population. Their salary and field indemnities provided them a decent living and corruption was not widespread. They toured villages to grant collective tavy authorization, delineate cultivation land and control forests clearing and exploitation. They also identified sites suitable for building dams and creating irrigation schemes. This last responsibility obliged them to link rural development and conservation objectives. Unfortunately, they lost it later in favor of civil engineering services attached to the Ministry of Agriculture. The state efforts to develop communication infrastructure and the lower vulnerability of households also eased the rangers’ tasks. Certainly, land shortage and migrations still resulted in significant forest clearing and some communities may have suffered from repression campaigns. But we will see in the case of Beforona that farmers initiated a move to reduce their dependence on tavy and forest resources and remember this period as being most favorable to their livelihood.
53
4.4. The first environmental conferences Madagascar, after its independence, signed several international conventions and treaties and took the following steps for the implementation of environmental policies: •
Membership in the World Conservation Union (IUCN 23 ) in 1961.
•
Creation of the Superior Council for Nature Protection24 in 1962 to provide guidance to environmental policies. This council had a consultative role concerning fauna and flora management, the creation of protected areas and the application of international treaties.
•
Signature of the African Convention for Nature and Natural Resource Conservation in Alger (1968), ratified by the law 70-004 of June 23, 1970.
•
Participation in the “Biosphere Conference” in Paris in October 1969, organized by UNESCO 25 .
•
Organization of the Antananarivo International Conference on Rational Utilization and Conservation of Nature and its Resources 26 on October 711, 1970, sponsored by IUCN and UNESCO. This conference was followed by a costly antifire communication campaign. Antifire brigades and committees were created and communication and awareness campaigns were launched (Kull 2004).
•
Creation of the Environment and Nature Conservation Commission 27 in April 1971 to monitor the participation of Madagascar in the Man and the Biosphere Program and to prepare the United Nations Environment Conference to be held in Stockholm in 1972.
23
In French: Union Internationale pour la Conservation de la Nature. In French : Conseil Supérieur pour la Protection de la Nature. 25 United Nations Educational, Scientific and Cultural Organization. 26 In French : Conférence Internationale sur la Conservation de la Nature et ses Ressources. 27 In French : Commission Conservation de la Nature et Environnement. 24
54
The vice president of the Malagasy Republic, in his allocution at the Stockholm conference, insisted on the need to reconcile the satisfaction of human needs with the conservation of nature. He asserted that “the only realist attitude was a rational use of the resources, which includes necessarily the imperatives of conservation in order to insure the permanence of these resources” 28 (Repoblika Malagasy 1972, 40). The report addressed issues such as natural resource management and pollution, but also regional planning and rural and urban development. It contended that “in a world where economic rules do not allow increased prices and increased costs,” and where industrialized countries have a technological advantage, environmental management and conservation can only be achieved through international solidarity and aid between governments, people and enterprises (Repoblika Malagasy 1972, 37-38). But it also perpetuated the simplistic explanation of environmental degradation that developed during the colonial period. The Ministry of Agriculture and Rural Development, in his allocution, sent a “SOS” to gain support to fight against bush fires ignited by pastoralists. Several sections of the conference report put at the forefront the need to educate farmers to change their “mentality,” to repress their practices, to sensitize them to environmental issues and to propose them new techniques (Repoblika Malagasy 1972, 37-38). But after this conference, Madagascar entered into a period of political and economic turmoil and the new commitments would not be translated into programs and actions until the mid 1980s.
5. The transition to the second and third republic The relative prosperity of the first republic did not last and political protest arose in the late 1960s. 28
Translation by Pollini. Original text in French: “la seule attitude réaliste [est] celle d’une utilisation rationnelle des ressources naturelles, celle-ci englobant nécessairement les impératifs essentiels de la conservation afin d’assurer la permanence des ressources.”
55
First, there was a “growing perception that little had been changed by independence and that behind the facade of a Malagasy government, the French were still running the country” (Brown 2000, 307) – import-export trade was largely controlled by four French companies: the Lyonnaise, the Marseilleise, the Rochefortaise, and the Havraise. About 1000 technical assistants filled half the senior posts in the central administration. About 80% of the teaching staff at the university and state secondary schools were French. The consequence for the Malagasy youth and elite was fewer opportunities to get jobs and advance in their careers, and a feeling that the policies were not their own (Brown 2000). Second, the terms of trade started to degrade, which widened the economic gap between the rural mass and the urban elite. Impoverished farmers migrated to cities but were not completely absorbed by economic growth, which increased the number of unemployed urban poor and caused further tensions (Brown 2000). In 1971, student strikes started in Antananarivo and upheaval occurred in the countryside, mostly in the poorest southern regions. Internal divisions based on doctrinal differences, personalities, and ethnic or regional origins developed inside the regime. President Tsiranana contained them as long as he was in full vigor, but when his health started to decline, criticisms became increasingly open and were sharpened by the competition to be his successor (Brown 2000). The political turmoil peaked after President Tsiranana resigned, during the transition from the first to the second republic, between 1972 and 1975. Large-scale protests destabilized the economy, security forces killed protesters, a President was assassinated (Ratsimandrava, in 1975), and French economic interests and political and cultural influences were opposed. In 1973, Madagascar left the French Franc Zone. French technical advisors left the country and secondary education was required to be in Malagasy. Ratsiraka emerged from this turmoil as a new leader expected to
56
stabilize the political situation. But he launched a Socialist Revolution that led the whole country to bankruptcy. The impact of this turmoil on the environment was severe. The pace of deforestation by tavy picked up in the 1970s because power struggles in the capital and populist concerns led to a loosened regulation and enforcement, though the law did not change. In the context of breaking the links with the French neocolonial power…, the political message of the time was “do as you please.” (Kull 2004, 169) Macro-economic indicators were not threatening until the late 1970s. The balances of public and external finances were in good shape and remained so until 1977-78 (Jolly 2004) 29 . But there was no growth due to nationalizations that caused most companies to lose their competitiveness. The country started an all-out investment industrialization policy that created only white elephants and expanded bureaucracy and corruption (Jolly 2004; Brown 2000). Agricultural production stagnated due to the inefficiency of state farms and cooperatives that had replaced the foreign concessions, and to the low buying price of export crops by the state: between 20 and 30% of FOB price (Brown 2000). The worsening of the terms of trade and the second petrol shock aggravated the situation, leading to a total bankruptcy with a foreign debt of one billion USD in 1981. In the early 1980s, the debt service was not met despite a 65% debt relief. Madagascar went to World Bank and IMF and started its first structural adjustment programs in 1980. It came back, step by step, to a free market economy, lost its support from Moscow and resumed cooperation with France and other Western countries (Brown 2000; Jolly 2004). The social consequences of this economic collapse were huge. Rural bandit groups (Dahalo) proliferated. Crime, insecurity, and begging developed in the main cities, especially Antananarivo. The infrastructure was not maintained and degraded 29
This information is from an interview, reported in Jolly (2004), of Leon Rajaobelina, Director of the Central Bank at that time.
57
quickly. Many rural secondary roads disappeared, leading rural communities to revert to an autarky economy, as in the case of Beforona that will be studied in Part III. The living standard of the whole middle class declined and corruption proliferated. Not until the late 1980s did the economic situation start to improve. Growth was 4.9% of GNP in 1989 but the external debt amounted to 2.6 billion USD in 1986 (Brown 2000). Growth benefited urban people, but for most of the rural poor, impoverishment and degradation of the environment have continued until the present time. The crisis eventually forced the state to completely abandon the socialist ideology and to return to democracy, under tragic circumstances 30 . A new constitution (the third republic) was adopted in 1993, together with the election of a new President, Albert Zafy.
6. Toward the National Environmental Action Plan (NEAP) With the launching of the IMF and World Bank programs in 1981, Madagascar opened again to foreign investments and influences. The 1980s were a period of increasing international concern about environmental issues, while forest clearing and land burning still peaked due to the collapse of the economy. Among the first visitors were biologists who put lemurs and fires under the media lights in order to inform the public of Madagascar’s high biodiversity and rapid deforestation. The Malagasy government understood the opportunity it could get from this situation to capture the foreign aid that was so necessary for solving the debt crisis (Jolly 2004). In 1984, it created the National Commission of Natural Resource Conservation for Development, in charge of preparing the Malagasy Strategy of Natural Resource Conservation for Sustainable Development, called for by IUCN, UNEP and WWF (Kull 2004). The
30
An estimated 100 persons were killed by security forces during a peaceful protest march to the President’s palace in Iavoloha, on August 10, 1991.
58
strategy was adopted in 1984 and a second international environmental conference was held in 1985 to launch it. The Malagasy state engaged to improve its institutions, to adopt tools and methods for environmental management and to implement actions aimed at stopping degradation (World Bank 1988). The conference is also “remembered by many as the moment when Prince Philip, the international president of WWF, confronted President Ratisraka with the statement, “your nation is committing suicide” ” (Kull 2004, 239). Several programs were initiated after the conference. They were funded by the World Bank, bilateral donors (especially the U.S. and Switzerland), WWF and UNESCO. They concerned soil conservation, forest management and biodiversity conservation and included an environmental education program that reached all school districts (Kull 1996). They gave the opportunity to Malagasy biologists to be trained overseas and allowed increasing scientific exchanges and communication about environmental issues. Top staff of international nongovernmental organizations visited Madagascar, such as Russel Mittermeier of Conservation International and Tom Lovejoy of the World Wildlife Fund. In 1986, the De Heaulme Family won the J. Paul Getty Prize for Conservation for the private reserve they managed in Berenty, south Madagascar. This prize had been won by conservation figures such as Jane Goodall. It opened new opportunities to inform the international media of biodiversity and conservation efforts in Madagascar (Jolly 2004). In May 1987, three Malagasy decisions makers (the Ministry of Water and Forest, the Ministry of Higher Education and the Ministry of Applied Research) made their first trip to the United States to attend a meeting on Saint-Catherine Island, South-Carolina and to visit institutions working on environmental issues. They met representatives of the WWF, the New York Zoological Society (World Conservation Society), the Duke University Primate Center, the San Diego Zoo, the Jersey Wildlife Preservation Trust, the Missouri
59
Botanical Garden, and Yale University to prepare a collaborative plan (Wright 1996). The terrain to receive grants for environmental issues was very favorable at that time. The World Bank itself was in need of greening its activities. A few years before, crusading environmentalists scaled a building in Washington across the street from the World Bank. They hung out a huge banner whose dripping red letters proclaimed: THE BANK MURDERS RAINFOREST! The bank could not deny it. Bank-funded development in the Brazilian Amazon shouted its record of forest destruction to satellites in the sky. Madagascar was just what the World Bank – and all the donors – needed. Here was a virgin country where they had no project at all, so they had not yet been able to make any environmental mistake. Go for biodiversity aid to Madagascar! (Jolly 2004, 211) Hence, all the conditions were met to establish an ambitious National Environmental Action Plan (NEAP 31 ). When this plan was written in 1988, there were two National Parks, 11 Integral Natural Reserves, and 23 special reserves in Madagascar, representing 1,034,782 hectares or 1.76% of the country. These protected areas were poorly managed, some terrestrial ecosystems were not yet protected and there were no marine reserves (World Bank 1988). The main goal of NEAP was to set up a national network of about 50 protected areas that would cover the whole Malagasy biodiversity. The plan was written under the coordination of the Malagasy government, by a Permanent Technical Committee 32 working under the authority of a National Commission for Conservation and Development attached to the Planning General Directorate 33 . Financial and technical support was provided by the World Bank (leader of the donors’ group), USAID, the Swiss aid, UNDP, UNESCO and WWF. One hundred fifty national specialists from a wide range of disciplines were mobilized;
31
In French: PAE or PNAE (Pland d’Action Environnemental or Plan National d’Actions Environnementales) 32 In French : Comité Technique Permanent (CTP) 33 In French : Commission Nationale pour la Conservation et le Développement (CNCD)
60
eight working groups were set-up and trained and regional workshops were organized (World Bank 1990). The philosophy displayed by the plan was to favor dialog and communication instead of top-down recommendations, to insist on the benefits of the programs rather than on their constraints, and to set up mechanisms for funding a myriad of small projects (World Bank 1988). According to the World Bank (1990), the interdependence of conservation and development was the key tenet that guided the preparation of the plan, consistently with the recommendations from the United Nation World Commission for Environment and Development (WCED 1987) and of the UNESCO program Man and the Biosphere. This same World Bank report (1990) noted that Madagascar was one of the richest countries in terms of biodiversity, but one of the poorest with respect to its economy, and that “this contrast provoked, probably more than elsewhere, a tension between the partisans of conservation and those of rural development” 34 (World Bank 1990, 1). The Plan was legally recognized as being the translation into action of the national environmental policy by the adoption of the National Environment Charter (Gouvernement de Madagascar 1990). This law confirmed the willingness to link human development and environmental conservation. According to article 2, “the environment is constituted of the whole natural and artificial milieu, including human milieu and social and cultural factors that contribute to the national development 35 ” (Gouvernement de Madagascar 1990: article 2). According to article 6, its main objective is to “reconcile the population with its environment in order to achieve a
34
Translation by Pollini. Original citation in French: “Ce contraste a engendré, probablement plus qu’ailleurs, une tension entre les partisans de la conservation et ceux du développement.” 35 Translation by Pollini. Original text in French:“On entend par environnement l’ensemble des milieux naturels et artificiels y compris les milieux humains et les facteurs sociaux et culturels qui intéressent le développement national.”
61
sustainable development” (Gouvernement de Madagascar 1990: article 6). This would be achieved through the following objectives: -
to develop human resources, to promote sustainable development though a better management of natural resources, to restore, conserve and manage the Malagasy biodiversity heritage, to improve the livelihood of urban and rural population, to maintain an equilibrium between population growth and resource development, to improve environment management tools, and to support the resolution of land tenure issues. (Gouvernement de Madagascar 1990: article 6)
The National Environmental Action Plan was translated into action by a three phase program funded by the World Bank and other donors. I will now detail these three phases.
7. The Environmental Program, phase 1 (EP1, 1991-1997) 7.1. Presentation of EP1 When EP1 started, in 1991, the Malagasy government had no specific institution to address environmental issues (World Bank 1990). The only operational structures were the National Commission for Conservation and Development (CNCD) and its Permanent Technical Committee. They only had a consultative role. The Water and Forest Department of the Ministry of Animal Production, Water and Forest was technically responsible of forest and biodiversity management, reforestation and soil conservation. It benefited from World Bank support (Forest Management and Protection project 36 ) and from help of several nongovernmental organizations, universities and other international organizations, but this was insufficient (World Bank 1990). EP1’s main goal was to create an appropriate policy and institutional framework for environmental management (World Bank 1996). The program started 36
In French : Gestion et Protection des Forêts (GPF). Crédit N° 1878-MAG.
62
by creating the National Office for Environment (ONE), a small agency attached to the Ministry of Economy and Plan that would be in charge of EP1’s coordination. EP1 had a planned budget of 85.5 million USD for six years (World Bank 1990). Donors included the World Bank, (26 million USD granted by IDA 37 ), UNDP, UNESCO, USAID, the French, German, Swiss and Norwegian governments, WWF and CI (42.3 million USD in total) and the Malagasy government (17.2 million USD). The actual cost of EP1 was eventually 150 million USD (World Bank 1996). EP1 consisted of seven components (World Bank 1990): •
Component 1: protection and development of unique Malagasy ecosystems while developing peripheral zones (21.55 million USD). The creation of 14 new protected areas was planned, as well as the improvement of the management for the 36 already existing ones. Fifty operational protected areas should exist at the end of the program. A specific and independent organization, the ANGAP 38 , would be created to manage this network and set up Integrated Conservation and Development Projects (ICDPs) in the buffer zones of protected areas. Support would also be provided to the Water and Forest Directorate for the management of classified forests, despite the fact they would not be regarded as Protected Areas per se.
•
Component 2: promotion of soil conservation, agroforestry, reforestation and other rural development activities in the priority zones, including several watersheds (9.35 million USD). The target was to provide agriculture development to 100,000 rural households, through micro-
37
International Development Agency, a branch of the World Bank specifically dedicated to the poorest countries. 38 National Association for the Management of Protected Areas (in French : Association Nationale pour la Gestion des Aires Protégées).
63
projects for erosion control, agroforestry and watershed management. Seventy percent of these households would be located in the buffer zones of protected areas and 30% in areas threatened by erosion (World Bank 1990). A specific organization, the ANAE 39 , would be created to achieve this objective. •
Component 3: preparation and publication of maps and geographical information systems for the areas concerned by the program (10.14 million USD), through support to the Malagasy Institute of Geography (FTM 40 ),
•
Component 4: improvement of land security by issuing land titles (6.94 million USD), through support to the Direction of Domains of the Ministry of Agriculture,
•
Component 5: training of environment specialists, communication and education of the Malagasy population to environmental issues,
•
Component 6: research programs about marine, coastal and terrestrial ecosystems (4.24 million USD), and
•
Component 7: support activities including institution strengthening, the establishment of procedures for environmental impact assessment, the development of environmental databases and monitoring and evaluation activities (15.74 million USD).
These objectives reveal a concern about rural development issues, which were expected to be addressed by ICDPs (9.00 million USD) and by the soil conservation component (9.35 million USD). Hundreds of millions USD were further expected to arrive soon from other programs. In spite of this, EP1 was launched in an atmosphere of controversy about whether it was dealing with the right priorities in a country where 39
National Association for Environmental Actions (in French : Association Nationale d’Actions Environnementales). 40 In Malagasy: Foiben-Taosaritanin’i Madagasikara
64
poverty and misery were at a peak. Philippe Rajaobelina, Deputy Director of the Plan, “wrote to say that even in the biodiversity sector it was unacceptable to have more money allotted to the reserves than to peripheral development around the reserves” (in Jolly 2004, 215), because “there are more important primates in Madagascar than lemurs” (in Jolly 2004, 215). Newspapers made similar statements: A cartoon in Madagascar’s morning paper showed a fat fuzzy lemur, its ringed tail wrapped around hundred-dollars bills. Beside it two skeletal humans hold out clawlike hands. “Nothing for you!” sneers the lemur. “You’re not an endangered species.” (in Jolly 2004, 215) This controversy may reveal a mismatch between the intention of the program and its actual logic. This tension and the ideological debates associated with it will persist during the implementation of NEAP, as I will show later.
7.2. EP1’s impact After four years, the fulfillment of the program’s objectives was regarded as “generally satisfactory” by its main donor, although it “varied from one component to another” (World Bank 1996, 6). “Main strengths include[d] the effective development of institutions and partnerships, high visibility and a substantial demonstration effect, both domestically and internationally, and effective field results” (World Bank 1996, 8). Weaknesses were “a slow take-off, insufficient program integration, consolidated monitoring not yet operational, and an insufficient role given to environmental concerns when formulating policies that require hard choices, whether at the national or local level” (World Bank 1996, 8). The Beneficiaries’ Evaluation Reports provide some explanation for this lack of coordination and integration. According to Miara-Mita (1995), ONE was operational in 1993 only and its links with the Ministry were not clear. Communication between agencies was insufficient due to changes in personnel,
65
especially at the Water and Forest Directorate (DEF 41 ). Staff at the new line agencies (ONE, ANGAP and ANAE) had better salaries and working conditions and were better trained than those at the DEF, resulting in low motivation of the later. In its majority, staff from the DEF seemed to distance themselves from the donors, from the line agencies and from other partners such as consulting firms. DEF complained that it was set apart by ONE (lack of information, delay in payment). Similar problems also occurred at the FTM (in charge of mapping activities) and at the Direction of Domains (in charge of land entitlement) (MIARA-MITA 1995). In sum, the plan appeared to be like a graft that had some difficulty being accepted by the institutions in charge of its implementation. In spite of these difficulties, most activities were implemented. According to the Staff Appraisal Report of the Second phase, most of EP1’s objectives were in the process of being met in 1996: Some 42,000 families are benefiting from over 1100 mini projects for soil conservation by ANAE, twenty one national parks and other protected areas are being established and managed to the extent that ANGAP’s current funding allows; with nearly 840,000 hectares surveyed, land titling has begun to have impact in and around protected areas; FTM has significantly improved its mapping capacity and produced aerial photographs and other geographic instruments over 80,000 km2 of protected and surrounding areas, and ONE has developed activities in promotion, video production, training, policy formulation and regulatory framework improvement. New legislation has been established allowing the creation of foundations and trust funds, and a national environmental foundation (Tany Meva) has been established. A new forest policy has been formulated and is being translated into new legislation, providing for decentralization of forest management and rationalization of management plans and logging permits, and the reactivation of the existing National Forestry Fund. New legislation is also providing for the possibility of giving responsibility for natural resource management to local communities, under a negotiated contractual arrangement that will spell out the management system and the distribution of revenues. (World Bank 1996, 5-6) 41
Water and Forests Directorate. In French : Direction des Eaux et Forêts. It will be later changed into Water and Forest General Directorate (DGEF).
66
But the impacts of these activities on the ground are difficult to assess. According to the World Bank (1996), “inadequate attention has been given to the evaluation of costs and benefits associated with environmental protection activities” (World Bank 1996). Overall impact on the ground was regarded as limited “given that field activities ha[d] so far been limited in scope,” and because “the central objective was the creation and reinforcement of national institutions.” However, “ad hoc evaluations of field activities undertaken in soil conservation mini-projects reveal[ed] encouraging results” (World Bank 1996, Appendix 1 p.12). The beneficiaries’ evaluation reports were more pessimistic, especially with regard to the first component. They recognized that the creation of protected areas was effective but criticized the ICDPs for the excessive priority given to conservation objectives. The Ranomafana National Park ICDP, implemented through a partnership with ICTE 42 , provides a striking example of tension between conservation and development components. It is located east from Fianarantsoa, in a region inhabited by Tanala and Betsileo people (Chapter 1). This park is one of the three most visited in Madagascar and is sometimes regarded as a symbol of successful conservation programs (Wright and Andriamihaja 2002). But beyond the image presented by the promoters of the park (Wright and Andriamihaja 2002), the real situation in Ranomafana was difficult to assess. The OSIPD evaluators noticed certain constraints for accessing information. “The chief technical advisor of the ICDP required a National Park employee to accompany the evaluators and to be present during the interview” (OSIPD 1995, 67). Harper (2002) reports a similar control on her access to information when she implemented her research field work in the park. An anthropologist formerly working in the area would have been submitted to the same 42
Institute for Conservation of Terrestrial Ecosystems, Stony Brook University.
67
type of pressures. Nevertheless, the evaluation team managed to collect detailed information. Eighty percent of the interviewees were skeptical concerning the efficiency of conservation actions (OSIPD 1995, 70). Eighty three percent “expressed their participation in project activities in terms of a burden” (OSIPD 1995, 73). A traditional chief summarizes the situation in these words: “we are like trapped rats: we are here like in a jail and we cannot utilize all these things that are in our place 43 ” (in OSIPD 1995, 75). The fact that resources are “utilized” by foreign people (tourists and scientists) increases the frustrations still more. A frequent joke in Ranomafana would be that “Ranomafana is not a national park, but a family park,” 44 (OSIPD 1995, 77) meaning that only a few privileged people get benefits from it. Development activities were are also criticized. The evaluators reported that the project regarded them as an appendix to conservation objectives, or even as a lure for achieving conservation objectives. Forty-eight percent of the interviewees did not know who chose the project activities and asserted that they were not involved in the process. Fifty-seven percent complained that activities did not solve their problems despite the many promises. Twenty-eight percent considered that development activities were inefficient and that inequalities and thefts had increased in the area, because most aid was concentrated in a few villages. Community granaries were created but resulted in a division inside communities, because literate people had been put in charge of their management while traditional authorities had been marginalized. The evaluators concluded that activities were designed according to top-down logic. In the end, 25% of the interviewees (19 out of 77) considered that the project had 43
Translation by Pollini. Original text in French (translated from oral citation in Malagasy by OSIPD 1995): “Nous sommes comme des rats pris aux pièges: nous sommes emprisonnés ici, et nous ne pouvons pas utiliser toutes ces choses qui sont chez nous.” 44 Translation by Pollini. Original text in French (translated from oral citation in Malagasy by OSIPD 1995): “Le parc de Ranomafana? Ce n’est plus un Parc National, mais un Parc familial!”
68
positive socio-economic impact, while 75% considered that it had no or no significant positive impact (OSIPD 1995, 76). Concerning job opportunities, 20 out of the 24 conservation and development agents working for the project were recruited from other regions (OSIPD 1995). According to an elder of Ranomafana, the population would be like a wild oxen that one wants to tame: he is attached to a tree during two weeks and is not fed; after the two weeks, he is “broken,” and even kids can pet its nostrils and pull its ears without danger. 45 (OSIPD 1995, 78) In sum, after having been marginalized by slavery and colonization by the Merina and the French, the Tanala people are now squeezed in between the pressures of Betsileo immigration and conservation programs. The OSIPD evaluation team also investigated the case of Andringitra and Ivohibe National Parks 46 . Seven percent of the interviewees were forbidden to cultivate their paddy fields because they were located inside the park. Twenty hectares of paddy fields would be involved and the situation would be dramatic for the owners in terms of food security. In Ambatomboay, a village located in an area threatened by famine, the project proposed alley cropping technologies and beneficiary farmers complained about the distribution of seeds that were “not edible” and just served to control erosion (OSIPD 1995). Overall, 48% of the interviewees considered that the project had no socio-economic impact, 9% that it had a negative impact because of discrimination, and 21% that it had positive impacts. Most beneficiaries contended that in compensation for restricting access to land, the project should support the extension of paddy fields. For a program officer of the WWF, such a contract would be the wrong strategy because it would put the park in a position of blackmail, and 45
Translation by Pollini. Original text in French (translated from oral citation in Malagasy by OSIPD [1995]): “Cette population ressemble a un boeuf sauvage qu’on veut dompter: on l’attache a un arbre pendant deux semaines sans lui donner a manger; au bout de deux semaines, il est “cassé,” si bien que même un gosse peut sans danger lui caresser les narines et lui tirer les oreilles.” 46 Implemented by ANGAP and WWF.
69
because ICDPs should not be confused with rural development projects (OSIPD 1995). Two thirds of the beneficiaries, however, appreciated the project activities, despite the fact that conflicts emerged in the northern part of the intervention area (OSIPD 1995). Overall, the evaluators came to a similar conclusion as for Ranomafana: the project had limited or no positive impact due to the priority given to conservation over development activities. On the whole, they concluded that the main objective of the environmental charter (the reconciliation of humans with their environment) was not achieved. The same misleading excessive priority given to conservation over development objectives is found in the implementation of EP1’s second component (soil conservation). In Manakara, the nongovernmental organization TMM, which is in charge of implementing development activities with support from ANAE, complained about the objectives of this institution: “TMM appears to consider development first and soil conservation second, while ANAE seems to put soil conservation first and development second” (OSIPD 1995, 23). The evaluators recommended reversing the priority at the ANAE level, arguing that development activities could help to “open the door” to conservation activities (OSIPD 1995, 42). Other evaluation reports (CAST 1995; CCR 1995) provide less detailed analyses and are less pessimistic, but identified similar issues in several sites (activities not adapted to the situation of most farmers, discrimination in favor of the richer households, priority given to conservation objectives, top down approaches, protestation about restrictions on access to land and resources). Research publications provided similar conclusions to those of OSIPD (1995). Harper (2002), who implemented her research work in Ranomafana between 1993 and 1996, depicted a dramatic situation where residents struggle against malnutrition and diseases while access to the resources that could help them to survive is closed. She
70
concluded that conservation and development policies “were implemented in Madagascar with little or no regard for the rights of human subjects” (Harper 2002, 224). Ferraro (2002a) calculated that the cost for communities living in an area adjacent to Ranomafana National Park ranged from 19 to 60 USD per household, which is very substantial given the deep poverty, high vulnerability and low health status of people living around the park (Kightlinger et al. 1996; Kightlinger et al. 1998; Hardenbergh 1996; Hardenbergh 1997; Harper 2002). Still in Ranomafana National Park, Peters (1998; 1999) and Hanson (1996; 1997) describe a situation of conflict or tensions between local communities and park and ICDP’s managers. I myself visited Ranomafana and a few villages of the buffer zone in 2002 and witnessed that the park still had deleterious impacts on farmers’ livelihoods, due to very inadequate supports. A never-ending extraction of socio-economic data was still ongoing (I myself took part in it) but never translated into action with significant impact, provoking the anger of local communities. Concerning other sites, Shyamsundar and Kramer (1997) calculated that the creation of the Mantadia National Park (located close to the area that will be studied in Part III) cost 23 to 64 USD per year per household living around it. Moreover, some households were displaced when the park was created, provoking a deep economic crisis (McConnell 2002). Ghimire (1994) and Kull (1996) provide more case studies showing that the negative socio-economic impacts of NEAP and ICDPs were far from being the exception.
7.3. Conclusion In conclusion, it seems that the first phase of NEAP failed to satisfy the environmental charter’s ideal of reconciliation between humans and nature. Conservation objectives were raised above development activities. Supports to farmers
71
were envisioned mostly as a way to facilitate the acceptance of conservation policies. It is questionable whether communities received benefits from development activities, whereas the negative impacts of restriction on access to land and resources was clear. In this asymmetrical relationship between conservation and development objectives, conflicts rather than synergies emerged. In sum, the results of the implementation of NEAP’s first phase were unfortunately consistent with the early criticisms of the program.
8. The Environmental Program, phase 2 (EP2, 1997-2003) 8.1. Presentation of EP2 EP2 was launched in 1997. At that time, Madagascar possessed 44 protected areas (nine national parks, 11 natural reserves and 23 special reserves), covering 1.4 million hectares or 2.3% of the total land area. Over half of the highest priority research and conservation areas were still located outside parks and reserves (World Bank 1996). The management of this network encountered significant organizational and coordination problems. The objective of second phase was thus to extend and consolidate the results of the first phase. This would be done by creating new protected areas and by achieving a better integration of the whole of the components, through the consolidation of the programming, budgeting and monitoring of line agencies (World Bank 1996, 29). EP1 lacked legitimacy at community level, as we saw. The second phase would remedy this by using participatory approaches. As the staff appraisal report puts it, The overall development hypothesis of the program is that depletion of Madagascar’s natural resource base can be reduced by changing the enabling policies, institutions, incentives, and other conditions so that resource users have the authority to manage their own resources, and the
72
responsibility and the incentive to do so in a sustainable manner. (World Bank 1996, 11) The lessons learned from EP1 also gave more momentum to the will to integrate conservation and development objectives. According to the World Bank’s Staff Appraisal Report, the vast majority (70%) of Madagascar’s population (growing at over 3% per year) will continue to depend for its livelihood on low-productivity extensive subsistence agriculture – the main and most severe source of environmental degradation.” Hence “the battle to protect Madagascar’s biodiversity will be won or lost on agricultural land away from the forest, because the battle in which rural populations are engaged is about production and land use, not about the environment.” “In the absence of a land management and agricultural policy, there is no viable resource conservation policy, because how people manage land and production options determines what they do in the forest. Therefore, increased emphasis on rural development is required. (World Bank 1996, 11) EP2 had a planned budget of 155 million USD for five years, including 31 million provided by the Malagasy government. The main donor was still the World Bank (38.1 million USD). Other donors were UNDP, USAID, the French, German, Swiss, Japanese, Norwegian and Dutch governments, and WWF. The components of the program and their cost (consolidation of all donors) were as follows (World Bank 1996): •
Component 1: Field operations (123.1 million USD). It included the following activities: o Sustainable Soil and Water Management (GCES 47 , 43.5 million USD). This component is the continuation of the soil conservation and watershed management activities implemented under PE1 (Component 2) ANAE is still the main execution agency (30.5 million USD). The focus “would be on demand-driven, low cost and locally adapted intervention that can contribute to spontaneous adoption of improved
47
In French: Gestion Conservatoire des Eaux et des Sols.
73
practices by other farmers” (World Bank 1996, 17). In addition, bigger investments on bigger watersheds in specific and more vulnerable pilot sites would also be done (13.0 million USD). o Multiple-Use Forest Ecosystem Management (ESFUM 48 , 29.9 million USD). This component included the National Ecological Forestry Inventory, the identification of 300,000 hectares of new protected areas, the preparation of a pilot forest management scheme for 180,000 hectares, the formulation of participatory management plans involving local communities and decentralized Forest Services for 400,000 hectares, and the implementation of these plans on 150,000 hectares. These activities would be carried out by the Water and Forest General Directorate of the Ministry of Water and Forests. o National Parks and Ecotourism (CAPE 49 , 43.1 million USD). This key component, mostly carried out by ANGAP as a continuation of the first component of EP1, included the extension of the protected areas network; the purchase of equipment, infrastructure and staff for more effective
management;
the
promotion
of
ecotourism;
the
implementation of conservation research programs and environmental education. The continuation of some ICDPs were included in this component, but ANGAP would progressively take over direct management from ICDP operators. o Marine and Coastal Environment (EMC 50 , 6.6 million USD). This component did not exist in EP1. o Urban environment 51 . 48
In French : Ecosystèmes Forestiers à Usage Multiple. In French : Composante Aires Protégées et Ecotourisme. 50 In French : Environnement Marin et Côtier. 49
74
•
Component 2: regional planning and local management (14.5 million USD). It included the following activities: o Support to local natural resources management and land tenure security (GELOSE 52 , 6.9 million USD). The objectives of this transversal component were to transfer management rights to village communities while clarifying land tenure rights in these villages, to redefine land management policies and to propose a revised land law for the longer term. Transfer of management rights would be implemented by ONE, while land securing would be carried out by the General Direction of Domain and Land Security of the Ministry of Agriculture (DGDSF). o Support to Regional Programming and Spatial Analysis (AGERAS 53 , 4.3 million USD). This transversal component, implemented by ONE, created six Regional Programming Committees (RPC) aimed at providing guidance for the implementation of EP2. It illustrated the will to have a more decentralized, participatory and systemic approach that could address conservation issues at their root causes. After the recognized failure of ICDPs, it was expected that a larger scale decentralized approach would provide more development impact. o Creation of a Regional Fund for Environmental Management (FORAGE 54 , 3.3 million USD). This fund, administratively managed by ANAE, would serve to finance regional development programs identified by the Regional Programming Committees.
•
Component 3: Strategic activities (4.2 million USD). They consist of:
51
I do not detail these two last components because they are out of the scope of this dissertation’s subject. 52 Local Secure Management. In French : Gestion Locale Sécurisée. 53 In French : Appui à la Gestion Régionalisée et à l’Approche Spatiale. 54 In French : Fonds Régional d’Appui à la Gestion de l’Environnement.
75
o Upgrading the legal framework and the formulation of environmental policies (PSI 55 , implemented by ONE). o Assisting sector ministries in implementing policies and making environmental impact assessment operational (MECIE 56 , implemented by ONE). •
Component 4: Support activities. They are: o Finalized Environmental Research (REF 57 , implemented by ONE, 2.5 million USD). This activity was aimed at improving the coordination of research programs. o Environmental Education and Training (EFE 58 , implemented by CFSIGE 59 , 2.0 million USD). The objective was to improve the general content in environmental education. o Geographic Instrument (IG, 1.2 million USD). This activity consisted of support to the National Geographic Institute (FTM 60 ). o Environmental Information System (SIE 61 , implemented by ONE, 1.7 million USD). This activity was aimed at consolidating an environmental “dashboard” that would provide key indicators of the status of environmental resources and the impact of the program. o Communication, Monitoring, Evaluation, Program Coordination and Management (CGP 62 , 5.8 million USD). This component financed ONE for its coordinating role and supported other coordinating bodies
55
Policies, Strategies and Instruments. In French : Politique, Stratégies, Instruments. In French : Mise en Compatibilité des Investissements avec l’Environnement. 57 In French : Recherche Environnementale Finalisée. 58 In French : Education et Formation Environnementales. 59 In French : Centre de Formation aux Sciences de l’Information Géographique et de l’Environnement. 60 In Malagasy: Foiben-Taosaritanin’i Madagasikara. 61 In French : Système d’Informations Environnementales. 62 In French : Coordination Générale du Programme. 56
76
such as the National Environment Council, the Interministerial Committee and the Steering Committee. EP2’s high level policy guidance was entrusted in an Independent National Environmental Council (CNE), created by the Decree N° 97 822 of June 12, 1997. CNE played a consultative role, giving its opinion on new legislation before it was presented to the Ministries’ Council. CNE’s roles were to advise the government concerning environmental policies and to verify that the Environmental Program followed the legal framework of the environmental Charter. Policy level coordination was the responsibility of an Interministerial Environment Committee (CIME) attached to the Prime Minister. The roles of CIME were to provide advice to the Prime Minister for the implementation and monitoring of the program and to contribute to the integration of activities in the concerned sectors. CIME also played an arbitration role between line agencies in case of conflict. Its members were mostly the Secretary Generals of the concerned ministries. The creation and functions of CNE and CIME were confirmed by the Law N°97 012 of June 6, 1997. The operational level was carried out by state services (mostly the Water and Forest General Directorate of the Ministry of Water and Forest, the Domains Directorate of the Ministry of Agriculture and the FTM 63 ) and by the line agencies that were set up under EP1, mostly ANGAP and ANAE. General coordination was still the responsibility of ONE. Not surprisingly, this structure was quickly criticized for its excessive complexity and its dramatic inflation of institutions, components and activities. According to the supervision mission of October 2003,
63
National Geographic Institute.
77
EP2 was a highly complex operation to be implemented nation-wide, and its stated development objective of reversing environmental degradation was ambitious and unrealistic in the time framework allotted (in retrospect) as it involved: - The implementation of a wide range of activities (improved management, inventories, coordinated planning, production, introduction of new technologies, input supply, marketing, research, GIS, management transfer to local communities, and land tenure) on the local, regional, and national levels at the same time involving seven implementing agencies (some still new, the others in need of serious capacity building); - Extensive and ambitious policy formulation/reform, some politicallysensitive (Legal and Regulatory framework relating to the environment and environmental product exports; Mining, Energy, Industry, Tourism, Roads, Fishing, Aquaculture, Urban, Macro-economy, Biodiversity and Bioprospecting, Agriculture (e.g. use of agro-chemical inputs), and Mitigation of Unexpected Disasters); and - The challenging task of coordinating 10 multilateral and bilateral donors, four international nongovernmental organizations, and seven implementing agencies (AGEXs). (World Bank 2003, 7) Following the recommendations of supervision missions, EP2 has been simplified in 2001. The AGERAS, GELOSE, EMC and REF components have been grouped into a new operational component implemented by a new line agency called SAGE 64 , while the PSI, MECIE, SIE and CGP were grouped into a new component called PIIGE 65 . The FORAGE component has been stopped and activities relating to land security have been reduced. The number of line agencies has also been reduced, FTM, CFSIGE and DGDSF being regarded as consulting firms. The revised structure of EP2 is as follows: -
Protected Areas management: Institutionalize and sustainably manage a network of PAs 66 representative of Madagascar’s natural ecosystems; Forest eco-systems management: ensure a sustainable management of forest resources, and increase their economic, ecological, and social potential. The marine and coastal environment management would fit under this component…;
64
Environmental Support Service. In French : Service d’Appui à la Gestion de l’Environnement. In French : Politique, Instruments et Informations pour la Gestion de l’Environnement. 66 Protected Areas. 65
78
-
-
Sustainable soil and water management in priority zones: to increase the capacity of the rural population for a sustainable management of the productive natural resources (mainly soils and water); and Environmental policies and institutions: streamline environmental/biodiversity concerns; create an enabling environment for the sustainable management of the environment and biodiversity resources; capacity building of the related institutions; communication; training; etc. (World Bank 2003, 10)
In sum, EP2 is basically a continuation and consolidation of EP1’s key objectives: the conservation of Malagasy biodiversity by the creation of protected areas, and the extension of soil conservation practices. But it adopted a much more complex structure, at least during its early implementation. It attempted to solve the malfunction of EP1 by creating a more sophisticated techno-structure, but forgot that performances depended also on the root-stock upon which the graft was fixed. EP2 also put more emphasis on decentralization and community participation. It developed a new tool to achieve this: the GELOSE approach, which will be analyzed in detail in Chapter 10.
8.2. EP2’s impact The beneficiaries’ evaluation (OSIPD 2000) identified a series of institutional problems similar to those encountered during the first phase: lack of coordination and integration with other sector-based policies and inefficient communication between line agencies. This may be the consequence of the excessively complex structure that was adopted in the beginning. The report also revealed tensions between conservation and development objectives similar to those encountered during EP1. The rural communities targeted by the project considered that its main objectives were to protect forests and lemurs, and complained about the lack of support for development. According to the consultants, “it appears clearly that EP2 does not answer to the needs regarded as a priority by the
79
population. This is obvious in the regions of Antananarivo, Toamasina and Toliary, a little bit less in the other provinces” (OSIPD 2000, 50). Rural communities and regional and local authorities considered that just a few activities had a positive impact on development, poverty alleviation or food security (OSIPD 2000). The persistence of these problems put into question the legitimacy of the program. The analysis of this issue is central in the OSIPD evaluation but is done using a quite disputable methodology: the legitimacy is measured by the level of knowledge of the program and its components, by the level of integration of the logical framework in the beneficiaries’ environment, by the quality of the perception of the actions implemented by the program. 67 OSIPD (2000, 126) This measurement method is quite astonishing! Legitimacy and understanding are two aspects that do not necessarily mirror each other. A totally illegitimate program could be perfectly understood, and a perfectly legitimate one could be totally misunderstood. Assuming that these two aspects are positively correlated is equal to assuming, before starting the analysis, that the program is judicious, which renders the evaluation tautological! The consequence is that no recommendation can be made other than improving communication and justifying more funding to continue the same program: the advancement is insufficient because of an obvious lack of information and communication at all levels (ONE-OSIPD 2000, 126). It follows that a fundamental misunderstanding persists between EP2 and the beneficiaries concerning the meaning of “result,” despite the fact that both sides are in the pursuit of results (ONE-OSIPD 2000, 126). Experience has shown, first, that a certain lapse of time is necessary, in any case more than two years, for an activity to be legitimized, and second, that the signs, as mysterious as they are, can be understood by the population, 67
Translation by Pollini. Original text in French: “La légitimité se mesure par le niveau de connaissance du programme et de ses composantes, par le niveau d’intégration du cadre logique dans le milieu bénéficiaire, par la qualité de la perception des actions menées au niveau des composantes.”
80
so long as they are explained in a well conceived and well implemented act of communication. 68 (ONE-OSIPD 2000, 48) Five years before, the conclusions of EP1’s evaluation report (OSIPD 1995) showed that the program lacked impact in terms of development. They also made it clear that this failure created a crisis of legitimacy. Hence, when OSIPD conducted EP2’s evaluation in 2000, the supposed “lack of understanding” had lasted indeed for seven years already. Moreover, according to EP1’s evaluation (OSIPD 1995), the lack of success was not caused by a lack of communication but by inappropriate development supports and by the excessive priority given to conservation objectives. Research efforts, mostly the work of Peters (1998; 1999) in Ranomafana, confirmed this and provided guidelines for more appropriate strategies. Because they assumed that the lack of legitimacy could be just the result of a lack of understanding and appropriation, the EP2’s evaluators did not verify whether or not the issues identified by their colleagues under EP1 were still a concern. Without providing rigorous arguments, they implicitly put their colleagues’ conclusions aside
8.3. Conclusion EP2 faced similar difficulties as EP1 but its external evaluation concluded that communication was the key for successful implementation and for achieving impacts. Interestingly, communication is also a powerful tool to reshape representations of reality. NEAP, in its second phase, may have entered into a process of selfconstruction where instead of putting itself into question, correcting its mistakes and adapting to realities, it preferred to reshape realities, or at least representations of realities. The problem then would not only be that there would be a mismatch between
68
Translation by Pollini. Original text in French: “La preuve est faite, primo, qu’il faut un certain laps de temps, en tout cas plus de deux ans, pour qu’une activité puisse être légitimée, et, secundo, que les sigles, si mystérieux soient-ils, peuvent être compris par la population, pourvu qu’ils soient expliqués dans le cadre d’une action de communication bien pensée et réalisée.”
81
representations and realities. It would also be that this mismatch would be encouraged, leading to knowledge regression. The more sophisticated and abstract methods used in EP2’s evaluation (OSIPD 2000), where numbers and statistics replaced the narratives and direct citations of EP1’s evaluation (OSIPD 1995), may have contributed to this regression. The fact that researchers working in Ranomafana have been submitted to pressure for not publishing their results (Harper 2002) provides more evidence of this reluctance to face realities.
9. The Environmental Program, phase 3 (EP3, 2003-2008) 9.1. Presentation of EP3 EP3 started in 2003 and is still under implementation at present. Its overall objective is to manage eight million hectares of natural forests and 100,000 hectares of coastal zone and marine resources in a sustainable way. The target is to increase the area under protection status from 1.7 to six million hectares and to decrease the degradation rate of forest and wetland resources to less than half the 1993-2000 rate (World Bank 2004). The program reflects the Letter of Environmental Policy of the Malagasy government, adopted in September 2003. This letter asserts that the environmental policy is part of the poverty reduction strategy, as did the Environmental Charter in 1990. It reaffirms the commitment of the Malagasy government to achieve “sustainable development for the benefit of the Malagasy population” (World Bank 2004, 133). This is expected to be achieved “by contributing to broad-based economic growth, sustainable natural resource management and improving governance” (World Bank 2004, 4). This commitment is not new but it remains to assess if it is actually translated into fact during this third phase.
82
The program expressed an awareness of the negative economic impacts of conservation on local communities. The activities financed by the IDA/GEF grant are expected to induce a cost of 276 million USD and to generate a total benefit 292.7 million USD, resulting in a balance of +16.7 million USD (World Bank 2004, 82). But tavy farmers would be the “losers” of the program as the abandonment of this land use would represent a cost of 94 million USD (Carret and Loyer 2003). Restrictions upon charcoal production and the collection of nontimber forest products would also result in significant costs (28.9 and 31.5 million USD, respectively). These costs are expected to be mitigated by providing alternatives that would consist of: (i) stabilizing agriculture around natural forests through soil conservation techniques; (ii) promoting sustainable fuel-wood harvesting and nontimber forest products collection by promoting community based natural resource management (continuation of the GELOSE component of EP2); (iii) introducing more efficient carbonization techniques and diffusing more efficient wood stoves; and (iv) supporting reforestation programs. (World Bank 2004, 83) The main grants, provided by the International Development Agency of the World Bank (IDA, 40 million USD) and by the Global Environmental Facility (GEF, nine million USD), was ratified on August 19, 2004 69 , after a preparation period that started in December 2002 and was supervised by a “Task Force” created for this purpose. The first disbursement was made on November 7, 2004 (World Bank 2005). According to the Project Appraisal Document (World Bank 2004), the IDA-GEF program consisted of three main components: •
Component 1: Forest ecosystems management, implemented by the Water and Forest General Directorate (DGEF 70 ) of the Ministry of Environment,
69
Decree 2004-804 of August 19, 2004, ratifying the signature of the grant agreement from the International Development Agency of the World Bank, and Decree 2004-805 of August 19, 2004, ratifying the signature of the grant agreement from the Global Environment Facility. 70 In French : Direction Générale des Eaux et Forêts.
83
Water and Forests (MINENVEF 71 ) (18 million USD 72 ). This component includes: o The improvement of governance (6 million USD), by helping DGEF to carry out forest zoning and forest control, to set up an information system and to strengthen the concession rights allocation framework, and by supporting the Forest Sector Observatory (OSF 73 ). o The creation of Conservation Sites (4 million USD). This activity will be detailed later in this chapter, in the section dedicated to the Durban Vision Working Group. o The implementation of management transfer (4.5 million USD) in order to pursue the GELOSE component of EP2. This activity will be analyzed in depth in Chapter 11. o Reforestation (1 million USD), by supporting the creation of Land Reserves for Reforestation74. o Household energy (2.5 million USD), by improving the efficiency of energy production 75 (efficient stoves and efficient charcoal production). •
Component 2: Protected areas management, implemented by ANGAP 76 (22.5 million USD). When the program started, park entrance fees covered only 7% of ANGAP’s cost (World Bank 2004). This component supports:
71
In French: Ministère de l’Environnement, des Eaux et Forêts. Having 835 staff, it resulted from the fusion of the Ministry of Environment with the Ministry of Water and Forests, in January 2003. 72 When the program started, the annual budget of DGEF was about 400,000 USD (World Bank 2004a). 73 In French : Observatoire du Secteur Forestier. 74 In French : Réserves Foncières pour le Reboisement (RFRs). 75 It is estimated that 85% of domestic energy needs are covered by charcoal and firewood, which represents about 10 million tons of wood per year (World Bank 2004a, 6). 76 National Agency for the Management of Protected Areas. In French: Agence Nationale pour la Gestion des Aires Protégées. With a staff of 708 persons, it managed 1.7 million hectares of protected areas when the project started.
84
o The reduction of pressure around protected areas (1.5 million USD), by supporting micro-projects, and through capacity building and awareness raising, at the level of communes and involving the civil society. o The enhancement of the representativeness of the protected area system (1.5 million USD). o The consolidation of the protected area system (8 million USD), through support to zoning, control, and ecological monitoring. o The enhancement of tourism services (4 million USD). o The creation of a trust fund to insure the sustainable funding of the protected area system (7.5 million USD). This trust fund will be completed by other donors, with an objective of gathering 50 million USD by the end of the project. •
Component 3: Environmental mainstreaming, or the integration of the environmental dimension into sector policies, implemented by the Environment General Directorate (DGE) of the Ministry of Environment, Water and Forests, the Environment National Office (ONE) and the EP3 coordination office (CELCO) (8.5 million USD in total). This component includes: o Environmental information, education and communication (1.5 million USD), by supporting ONE to manage national and regional information systems. o Environmental legislation (2 million USD), by supporting DGE to carry out strategic environmental assessments aimed at improving the coherence of sector’s legislation with international environmental treaties and conventions.
85
o Environmental compliance (3 million USD), by supporting ONE to improve the application of the MECIE77 legislation. o Environmental management and coordination (2 million USD), by supporting MINENVEF and CELCO to put in place a financial management system and a monitoring and evaluation system The detailed objectives and activities of EP3 are summarized in a logical framework adopted by all the contributors of the program (Appendix). In addition to the 49 million USD granted by IDA and GEF, funding has been provided by other donors, mainly the USAID, the French Cooperation, the German aid, and Conservation International. The total amount granted to EP3 was eventually 171 million USD. About 75% of the MINENVEF budget was provided by donors (World Bank 2006). Table 1 summarizes the contributions by donor for each activity of the project. Table 2 lists the main projects related to biodiversity conservation initiated since 1997 (list established in 2003). The Malagasy State engaged to provide a contribution equal to 20% of the program’s global cost, i.e. 9.8 million USD, as a counterpart of the IDA grant, and 1.8 million as a counterpart of the GEF grant. Coordination is insured by a Project Implementation Support Unit, the CELCO 78 , established in the Ministry of Environment, Water and Forest (MINENVEF 79 ). The CELCO supports the line agencies for procurement, financial management, monitoring and evaluation, safeguards compliance, and reporting (World Bank 2004).
77
Environmental Impact Assessment for Investments. In French : Mise en compatibilité des investissements avec l’environnement. 78 This unit will eventually be named CELCO (Cellule de Coordination du PE3), or EP3 Coordination Cell. 79 Ministère de l’Environnement, des Eaux et Forêts, resulting from the fusion of the Ministry of Environment and the Ministry of Water and Forest.
86
87
2.0 7.2 8.7 0.07 0.73 4.2 1.38 0.33 0.88 0.06 1.2 0.12 0.5 0.08
10.2 5.6 2 4.25 0.15 4.2 0.99 0.19 0.8 1.4 0.12 0.8 0.05
0.97 28.4 17
31.7 18
Source: Moynot (2005, 8). Translation and adaption by Pollini.
2.1 Environmental education and sensitization
17 5.8 0.26 9 1.18 4.8 0.38 1.94 0.6 0.13 0.75 0.02 2.9 0.32 0.17 3.46 48.7 28
5.8 1.8 0.13 2.18 0.15
10 6
2 3.8 0.56 2.5 0.67 1.2 1.18 0.41 0.64 0.04 0.5 0.01 0.2 0.23 0.07 0.17 14 8
2.2
2.3
8.8 3.3 3.8 1.4 1 3 0.32
Environmental governance
1.4
Sustainable funding
1.3
Marine and coastal ecosystems
1.2 Forest ecosystems
CONTRIBUTORS The World Bank USAID French cooperation CI WWF German cooperation (GTZ/KFW) UNDP WCS Swiss inter-cooperation UNESCO Tany Meva Foundation Japanese cooperation (JICA) European Union Missouri Botanical Garden DWCT ICTE-Stony-Brook TOTAL %
1.1
Sustainable development
COMPONENTS
Protected areas and conservation sites
Table 1. Contributions to EP3 by donors in million USD
9 5.1 0.4 0.4 0.86 1.2 0.019 0.25 0.01 0.17
21.6 13
17.4 10
TOTAL 49 31 21.5 19.4 6.1 18.6 6.3 3.6 2.9 0.25 3.85 0.4 4.4 0.6 0.3 4.6 171
Table 2: Projects related to biodiversity conservation in Madagascar since 1997 Donor
Project
Beneficiary
Start
Finish
GEF GEF GEF GTZ
Water and forest management Protected area management Regional capacity building Efficient charcoal use to protect natural forest Integrated forest development Andringitra & Marojejy National Park Ankarafantsika Reserve Marovoay watershed management Zombitse Reserve management Sustainable use of natural resources Plateau Mahafaly ecosystem conservation Natural resource management transfer and training
MEF ANGAP SAGE/ONE GreenMad
06/97 06/97 06/97 04/97
06/01 06/01 06/01 03/06
Cost (millions) 0.9 USD 2.6 USD 4.6 USD 17.5 DEM
MEF WWF
03/98 06/98
02/06 06/03
25.5 DEM 11.0 DEM
CI Erosion Program WWF EP II
06/97 01/98
06/02 01/04
13.0 DEM 6.5 DEM
01/98 06/97
12/02 06/02
6.4 NOK 30.0 USD
WWF
10/01
10/05
6.0 FRF
EP II
09/97
03/02
12.0 FRF
Natural resource management and land tenure
EP II
06/02
06/05
5.5 FRF
Support to biodiversity and marine components Community forest Bemaraha National Park conservation & development Bemaraha phase 2 Support to ecoregional process, ANGAP, MWF Support to ecoregional planning, community forest management, and compatible economic development MWF MWF
EP II
01/98
12/02
9.6 USD
ANGAP
01/00 12/95
12/03 06/00
1.1 EUR 0.9 EUR
ANGAP Miray
06/00 07/98
12/05 06/02
5.0 EUR 10.0 USD
Landscape Development Initiative
07/98
06/03
19.0 USD
MEF MEF
06/90 06/97
06/02 06/04
5.5 USD 1.5 USD
GTZ KfW KfW KfW NORAD WB French GEF (AFD) French Cooperation (SCAC) French Cooperation (SCAC) UNDP EU EU EU USAID USAID
USAID DGIS
Source: World Bank (2004a, 164)
88
Overall guidance and strategic orientation is insured by a joint committee presided over by a delegate of the Minister of the Environment, Water and Forest (the Secretary General) and co-presided by a representative from the donors (the USAID environment task manager). Other members of the joint committee are the heads of ANGAP, ONE, DGEF, DGE, CGP, the CELCO coordinator, representatives from bilateral donors (France, Germany, Japan, Switzerland, USA), multilateral institutions (EU, IDA, UNDP), international and national nongovernmental organizations (SAGE, ANAE, CI, WCS, DWCT, WWF, Tany Meva, Fanamby), and other relevant stakeholders. The joint committee verifies that operations are consistent with the logical framework and informs the Ministry of the advancement of the program. The main differences between EP3 and the previous phase are the following: •
The abandonment of agricultural intensification activities, which were implemented by ICDPs under EP1 and by the AGERAS component and through ANAE’s activities under EP2. These activities are now transferred to projects that support the Malagasy Rural Development Action Plan, adopted in 1999. These projects are principally the Rural Development Support Project (PSDR) 80 of the World Bank (69.2 million USD grant for the period 2002-2007) and other projects funded by USAID, the French Cooperation, the Swiss aid, the European Union, etc.
•
The post-NEAP period has to be prepared because no fourth phase is planned. This is done by supporting the Malagasy government to streamline its environmental policies in all the concerned sectors (Component 3), and by the creation of the trust fund, which will finance conservation in the long-term.
80
In French : Projet de Soutien au Développement Rural.
89
•
The creation of a National Agency for Water and Forest Management (ANGEF 81 ), a new institution proposed but not created. ANGEF would have been in charge of all operational aspects concerning the management of forest resources outside the protected areas system. Its creation has been replaced by a support to the MINENVEF’s reform.
•
The creation of a Community Environmental Management Support Fund (FAGEC 82 ), a fund dedicated to supporting decentralized and community based environmental actions (reforestation, sustainable management of resources, etc.). The communes were expected to be the main beneficiaries of this fund. They would carry out activities formerly implemented by ANAE and SAGE, in partnership with these institutions and with other service providers.
9.2. EP3’s implementation EP3’s implementation is facilitated by a series of commissions and working groups, whose members are staff from the MINENVEF and its partners (donors, consultants, projects, private sector, civil society, nongovernmental organizations). I will present here the work of the Durban Vision Group (DVG), which coordinated the creation of Conservation Sites. This commission confronted with the key challenges of EP3 and its work is a good illustration of the overall logic of the program.
9.2.1. The Durban Vision During the World Congress of Protected Areas held in Durban, South-Africa, in September 2003, the President of Madagascar engaged his country to increase
81 82
In French : Agence National pour la Gestion des Eaux et Forêts. In French: Fond d’Appui a la Gestion Environnementale Communautaire.
90
protected area coverage from 1.7 million to six million hectares, in order to put 10% of the country under conservation and meet IUCN recommendations. The MINENVEF created the Durban Vision Group (DVG) to guide the achievement of this objective. The DVG’s Executive Steering Committee is composed of the Water and Forest General Director (DGEF), the Environment General Director (DGE), the Director of the Minister’s Cabinet, and the ANGAP General Director. A Technical Secretariat, which included representatives of CI, WCS and WWF, coordinates the work of five technical groups: management and categorization, biodiversity prioritization, communication, funding and legislation. At the beginning, the objective of the DVG was to facilitate the creation of Conservation Sites (Component 1). This new category of protected area was expected to be less costly than the existing ones managed by ANGAP. Conservation Sites would not be managed by this organization but would be under the MINENVEF’s authority. Conservation concessions could be established with conservations NGOs or other organizations for their management. Conservation Sites were also intended to provide more room for human activities and to be managed in a more decentralized way, with the strong involvement of communities and other local partners. Seventy five percent of the area would be allocated to the sustainable uses of natural resources by local communities, in order to contribute to poverty reduction and sustainable development. The main tasks of the DVG commissions were to identify priority conservation areas (marine and terrestrial), to define the minimal management rules for Conservation Sites, to set-up new legislation for giving them legal status, to support communications between actors and between regional and central authorities and to raise funds.
91
The categorization commission identified areas to be reserved for the creation of Conservation Sites (Map 2). The management commission worked on the definition of minimum management rules to be applied to Conservation Sites. Its first propositions, in 2004, were quite restrictive and at odds with what could be expected according to the initial intentions. According to the working document of DVG (Groupe Vision Durban 2004), ”no mining activity is accepted in Conservation Sites,” and “no commercial extraction of ligneous products, of any kind, is accepted in Conservation Sites.” Commercial harvest of nontimber forest products (NTFP) could be authorized, but would have to be localized inside Natural Resource Extraction Zones defined in the Management Plan. Harvesters would further be submitted to an Environmental Impact Assessment, following MECIE procedures, that would have to show that the population of the concerned species would not be threatened. A Sustainable Management Plan would have to be developed and extraction authorizations would be issued by the Water and Forest administration authorities. In sum, a ban would be put on the extraction of the most valuable products on six million hectares of forest, among the 7.5 million hectares remaining. Moreover, Map 2 shows that almost all forest patches of significant size were planned to be put under protection status. The remaining patches (the green area or dots on the map) are mostly fragmented forests submitted to intense pressure and it is unclear whether they still contain significant resources. Concerning nontimber forest products, local communities would be given priority for obtaining extraction permits. But due to the lack of knowledge about “modern” rules and regulations and the low literacy rates in remote villages, it is doubtful that these communities could master the procedure and be the actual beneficiaries of NTFP extraction. Instead, a local elite or a few outsiders could use the legislation to appropriate the resources. I will come back to this issue in
92
Provincial capital Existing protected areas Areas reserved for conservation sites Sensible areas (not reserved for conservation sites)
Map 2: Existing protected areas and areas reserved for conservation sites. Source: Durban Vision Group, September 24, 2004. Legend translated by Pollini.
93
the chapter dedicated to community based natural resource management approaches (Chapter 11). These restrictions led to intense discussions between the GVD members. Restrictions on mining activities were the subject of negotiation between the Ministry of Energy and Mines and the MINENVEF, through the mine-forest commission. Concerning the ban on ligneous products extraction, some organizations, such as the German cooperation (GTZ 83 ), the French cooperation, the SAGE, UNESCO and to a lesser extent UNDP and the WWF, advocated for a softer position where this extraction would not be excluded a priori from the whole protected area system, unless it would take an industrial form. The CIRAD 84 , with funding from the French GEF (FFEM 85 ), implemented a community based natural resource management project where communities were authorized to commercially extract ligneous resources. This project was implemented in the Ambohilero forest (Map 6, page 123), which is expected to be included in the future Mantadia Zahamena Corridor Conservation Site. It demonstrated that a sustainable extraction could be achieved by cutting 0.8 m3 of wood per hectare and per year, which was compatible with the creation of a protected area while providing significant income to local communities. On the other side it was argued, based on a similar experience implemented earlier by the MIRAY project in the Lakato region, that such management schemes would never be respected by communities and that the ban on extraction was the only realistic option. Concerning the Malagasy government, it expressed a preference for the second option in unofficial discussions, due to the influence of a “Malaysian” model of development where the exploitation of natural resources would allow a primary capital
83
In German: Gessellschaft fyr Technische Zusammenarbeit. French Agricultural Research Centre for International Development. In French: Centre International de Recherche Agronomique pour le Développement. 85 In French : Fond Français pour l’Environnement Mondial. 84
94
accumulation that would later serve to develop other sectors. However, the more restrictive conservation approach was officially adopted because it was the one favored by major donors. The lobby of conservation nongovernmental organizations has been particularly efficient maintaining this strategy. The discovery of illegal wood cutting in the Ambohilero forest, planned to be transformed into a Conservation Site, contributed to the justification of the conservation approach. This exploitation involved a joint venture between a Malaysian and a Malagasy firm. A road was opened across the forest and 17 kilometers had already been built when the operation was discovered. A sawmill was under construction on the site and heavy machines (bulldozers, trucks) were being used. This may have been one of the first industrial forest wood exploitations in Madagascar 86 . The Ministry stopped the extraction and issued a series of legal regulations to restrict forest product extraction. The Interministerial order 19560/2004, signed on October 18, 2004, suspended the granting of new mining and forest resources extraction authorizations in all areas reserved as Conservation Sites (the areas in red on Map 2, page 93) for a period of two years 87 . This ban was renewed in 2006 for an additional two years. The Ministerial order 21694/2004, signed on November 11, 2004, canceled all existing wood extraction authorizations, without compensation, and prohibited the extraction of ligneous resources in these same areas 88 . In spite of this, illicit wood extraction continued, as we will see in Chapters 3 and 5 in the case of Beforona. Containers of
86
In Madagascar, trees are usually cut up manually inside the forest and transported by people or oxen to the closest road (see the case of Beforona in Part II). 87 Translation by Pollini. Original text in French: “l’octroi de tout permis minier et de tout permis forestier est suspendu dans les zones réservées comme site de conservation, dont les limites sont annexées au présent arrêté” (Gouvernement de Madagascar 2004a, article 1). 88 Translation by Pollini. Original text in French: “Toute activité d’extraction de ressources ligneuses est prohibée dans les zones réservées comme sites de conservation tels qu’ils sont définis a l’article 2 de l’arrêté interministériel No 19560/2004 ” (Gouvernement de Madagascar 2004b, article 1). “ Les permis d’exploitation ou d’extraction de produits ligneux en cours dans ces zones réservées doivent être suspendus ” (Gouvernement de Madagascar 2004, article 2).
95
precious wood, including rosewood harvested in Masoala National Park, were regularly shipped for export in 2005 and 2006. In sum, it can be agreed that sustainable wood exploitation is not feasible in the context of Madagascar. But as the ban on exploitation is not enforceable, unsustainable wood exploitation will continue to occur. Moreover, the ban may create a situation of monopoly, by putting the forest sector into the hands of a few exploiters rendered untouchable by their strong political networks. The ban will also provide these exploiters with the “guarantee” that they will not pay taxes. Having these issues in mind, carefully monitored community wood harvesting by local communities may be worth trying again. Protected areas, which can afford to pay forest rangers and which may increasingly lack legitimacy in the future, may be appropriate locations to test this approach. In this context, the French aid had developed a new concept to challenge the conservation perspective: the Development and Conservation Territory (DCT). A DCT would consist of one or several core sites protected for their high biodiversity, and their surroundings, inhabited and already modified by human activities. A regional planning approach would be implemented to satisfy both conservation and rural development objectives. The approach is similar to that of the French Regional Natural Parks 89 and the DCT would match with the IUCN fifth category of protected areas (Table 3). At present, only IUCN categories Ia, Ib, II and IV are represented in the Protected Areas Network managed by ANGAP. The new protected areas are expected to fill the gap and to belong to categories V and VI. The Conservation Sites could match with the sixth category while the DCT would correspond to the fifth. 89
In French: Parcs Naturels Régionaux.
96
Table 3: Categories of protected areas according to IUCN (2002) Categories Ia Strict Nature Reserve
Main management objective Protected area managed mainly for science
Ib Wilderness Area
Protected area managed mainly for wilderness protection
II National Park
Protected area managed mainly for ecosystem protection and recreation
III Natural Monument
Protected area managed mainly for conservation of specific natural features
IV Habitat/Species Management Area Protected area managed mainly for conservation through management intervention V Protected Landscape/Seascape
Protected area managed mainly for landscape/ seascape conservation and recreation
VI Managed Resource Protected Area
Protected area managed mainly for the sustainable use of natural ecosystems
Source: IUCN (2002).
Table 4: Main differences between IUCN Categories V and VI Category V Protected Category VI Protected Area Area Core management Maintain harmonious Maintain predominantly philosophy interaction of nature and natural conditions as people basis for sustainable livelihoods Degree of modification of environment
Considerable: mainly a lived-in, working landscape
Predominantly natural (or near natural) condition
Typical dominant land uses
Agriculture, forestry, tourism
Hunting and gathering, grazing, management of natural resources
Source: IUCN (2002).
97
Category V “is unique among the six categories by making the core idea the maintenance of environmental and cultural values where there is a direct interaction between people and nature” (Phillips 2002, 10). It is defined as follows (IUCN 1994, in Phillips 2002, 10): Area of land, with coast and sea as appropriate, where the interaction of people and nature over time has produced an area of distinct character with significant aesthetic, ecological and/or cultural value, and often with high biological diversity. Safeguarding the integrity of this traditional interaction is vital to the protection, maintenance and evolution of such an area. The DCT could thus be a way to solve the asymmetry between conservation and development objectives that characterized NEAP. The DCT concept has been widely accepted by the Malagasy government and its partners as a tool for regional planning. But the idea that it would correspond to a protected area has been rejected. The main objection was that the focus of Category V “is not on nature conservation per se, but about guiding human processes so that the area and its resources are protected, managed and capable of evolving in a sustainable way” (Phillips 2002, 10). Emphasis is put on the cultural dimension of the landscape rather than on its natural dimension. Cultural landscapes exist in Madagascar, but their conservation is not regarded as a priority in comparison to less disturbed primary ecosystems that shelter a much higher biodiversity. As biodiversity conservation has to be the main objective of any protected areas system, even if other aspects are considered, the sixth category, which also authorizes multiple use of resources but concerns predominantly unmodified natural landscapes, was preferred to category V. Table 4 (page 97) summarizes the main differences between these two categories. In sum, the DCT concept may have been inappropriate in the context of Madagascar, if regarded as a protected area category, but relevant as a regional planning tool. It can be compared to the AGERAS approach of EP2, which was based
98
on the same statement that conservation programs failed due to a lack of integration with regional development schemes. In spite of this, advocacy for considering the DCT as a protected area category persisted 90 , which could be explained by the excessively political character of the discussions. The consequence was a lack of legibility of discourses elaborated by the Ministry’s technical advisors, creating confusion if not schizophrenia. At the political level, strict conservation policies were adopted due to the efficient lobby of conservation organizations. At the technical level, there was a clear willingness to open the system, though it was not clear whether the DCT was the appropriate way to do so. Eventually, IUCN experts were hired in March and July 2005 to calm the debate and provide more objective guidelines. The experts belonged to the World Commission on Protected Areas and the Commission on Environment, Economic and Social Policy of IUCN. They insisted on the new orientations of global conservation strategies that the conservation congresses in Durban 91 (2003), Kuala Lumpur 92 (2004) and Bangkok 93 (2004) defined. According to IUCN (2005a), some key recommendations from these congresses were the need to respect human rights; the minimization of negative impacts on the livelihoods of local communities, especially the poorest groups; the implementation of participatory approaches involving all stakeholders and answering to the aspirations of the whole society; the comanagement of protected areas involving local communities and the recognition of community protected areas. The consultants also pointed out that for the IUCN, the extraction of wood, of nonligneous forest products and even of mining products is not forbidden a priori in a protected area, in particular in the categories V and VI, but that this does not mean that these activities must
90
I took part in the advocacy effort. Protected Areas World Congress, September 2003, Durban, South-Africa. 92 Seventh Conference of the Convention on Biological Diversity, February 2004, Kuala Lumpur. 93 Third Conservation World Congress, November 2004, Bangkok. 91
99
be appropriated, and can be accepted. 94 (Borrini-Feyerabend and Dudley 2005a, 31) The consultant further reminded that management objectives have to be chosen with regard to biodiversity conservation objectives, which must remain the main goal of all protected areas 95 . The central question is to know if this extraction would result in biodiversity loss. The experts also contended that zones where biodiversity is not significant should not be included inside protected areas, but could constitute a peripheral zone where development activities could be implemented. In sum, they advocated for an open approach that would not be limited to protected areas with very limited use rights, but confirmed the arguments that led to rejection of DCT as a protected area category. In the end, the IUCN team proposed adopting an open, logical framework where all management objectives and governance modes could be combined (BorriniFeyerabend and Dudley 2005a). Their second mission, in July 2005, was aimed at supporting the creation of this new framework, through a participatory process that would start at the local level. The new framework would be formalized by a new legislation, constituting a Protected Area System (SAPM 96 ) that would render obsolete the notions of DCT and Conservation Sites. The proposition was accepted and from that moment, the SAPM became the new conceptual and legal framework within which conservation programs had to work. A ministerial order (Gouvernement de Madagascar 2005a) was issued in 2005 to give a preliminary status to the newly created protected areas. According to this
94
Translation by Pollini. Original text in French: “ Du point de vue de l’UICN, l’exploitation du bois ou des produits forestiers non ligneux et même des produits miniers n’est pas interdite a priori dans une aire protégée, en particulier dans les aires protégées de catégorie V et VI, mais cela ne veut pas dire pour autant que ces activités peuvent se révéler appropriées, et donc être acceptées.” (IUCN 2004a, 31) 95 The IUCN definition of a protected area is “An area of land and/or sea especially dedicated to the protection and maintenance of biological diversity, and of natural and associated cultural resources, and managed through legal or other effective means.” (IUCN 2002, 8) 96 Malagasy Protected Areas System. In French: “Système des Aires Protégées de Madagascar.”
100
order, the sites would be submitted to the “special regime,” a modality created by article 49 of the forest law (Gouvernement de Madagascar 1997). According to article 52 of this same law, “these perimeters cannot be object of any extraction, under any form. Clear cutting, clearing, burning and grazing are forbidden in them.” 97 These strict restrictions are, however, contradicted by the order, which authorizes nontimber forest product extraction under certain conditions. This order was followed by the Decree 2005/848, issued on December 13, 2005, which defined the new categories of protected areas to be created (Gouvernement de Madagascar 2005b). This decree addresses most of the issues that had been debated. For example, article 1 asserts that management objectives can include the conservation of cultural heritage. Article 2 creates four new categories which are the natural park, the natural monument, the protected landscape, and the natural resource reserve 98 . The commercial extraction of natural resources is forbidden from the first two categories, but not necessarily in the last two (protected landscape and the natural resource reserve), showing that the debate had eventually been fruitful. Articles 6 and 7 further insist on the necessity to consult local populations and decentralized territorial collectivities during the preparation of management plans, in order to consider local interests. However, the decree does not propose clear rules nor guarantees concerning the way to address the potential negative socio-economic impacts. First, the acceptance of protected areas by rural communes, districts and regions has to be formalized by official documents, but at the community level, the consultation is not followed by any formal acceptance. Second, communities “can” be compensated if 97
Translation by Pollini. Original text in French: “ces périmètres ne peuvent faire l'objet d'exploitation, sous quelque forme que ce soit. Les coupes rases, les défrichements et les mises à feu, ainsi que le pâturage y sont interdits.” 98 Translation by Pollini. Original text in French: “Parc naturel,” “monument naturel,” “paysage harmonieux protégé,” and “réserve des ressources naturelles.”
101
their access to resources is limited (Gouvernement de Madagascar 2005b), meaning that this compensation is just an option. Third, no socio-economic impact assessment is required, while conducting an environmental impact assessment is mandatory. This situation is quite astonishing as there can be little doubt about the environmental benefits of gazetting a protected area, while negative socio-economic impacts for local communities seem to be the rule, as shown in the evaluations of EP1 and EP2 (OSIPD 1995; 2000), by other consultant reports (Carret and Loyer 2003; Hockley and Razafindralambo 2006), and by research publications (Larson 1994; Ghimire 1994; Shyamsundar 1997; Peters 1998; Peters 1999; Harper 2002; McConnell 2002; Ferraro 2002). In sum, the decree reveals the same asymmetry and incapacity to address socio-economic aspects that led ICDPs to failure. Paradoxically, the only negative environmental impacts of protected areas that can be reasonably envisioned are indirect and will never be addressed by experts in charge of implementing the MECIE legislation. They are the leakages provoked by the volunteer displacement of farmers whose access to resources would be restricted, and only socio-economic impact assessments could anticipate these leakages! In spite of these concerns, the issue of this regulation allowed the creation of 919,000 hectares of protected areas in 2005 (four sites 99 ), followed by 1,102,566 hectares in 2006 (16 sites 100 ). In January 2007, the new protected areas were still
99
Sahamalaza (130,000 hectares), Ankeniheny-Zahamena (425,000 hectares), Makira (371,217 hectares), Anjozorobe-Angavo (52,200 hectares), and Loky-Manambato (70,837 hectares). 100 According to a presentation made at the EP3 Joint committee on December 31, 2006, the sites are the following: Nord-Ifotaky (22,256 hectares); Kodida (10,744 hectares); Ambatotsirongaronga (834 hectares); Mandena (230 hectares); Ambato Atsinanana (747 hectares); Central Menabe (125,000 hectares); Tampolo (675 hectares);Forest Corridor Bongolava (60,589 hectares); Montagne des Français (6,092 hectares); Multiple Use Forest Station Antrema (12,270 hectares); Forest Corridor FandrianaVondrozo (499,598 hectares); Analalava Forest Foulpointe (204 hectares) ; Andreba (31.7 hectares); Mahavavy-Kinkony (268,236 hectares); Amoron’I Onilahy (52,582 hectares); Alaotra Lake(42,478 hectares).
102
under temporary protection status 101 . More protected areas are expected to be created during 2007 and 2008, leading to a total of 4,539,000 hectares under protection by the end of 2008 102 . The objective to achieve 6,000,000 hectares has been postponed to 2012. Table 5 summarizes the state of advancement of protected areas gazetting in Madagascar. In this context of frantic gazetting, the debate about the potential negative impacts of protected areas on local communities emerged again at the end of 2006. The third EP3 supervision mission, implemented in September-October 2006, asserted that no sufficient funding was available to implement development activities around protected areas. The Bank explained these difficulties by the fact that, when the EP3 was conceived, it was envisioned to establish six new protected areas (three terrestrial and three marine) for a total area of 324,000 hectares, [while] complementary rural development projects such as the FID 103 and the PSDR 104 would finance support activities. However, none of the rural development projects had legal conditions that obliged them to finance the support activities of EP3 (although the PSDR and EP3 had established the protocols of the partnership agreement). 105 (World Bank 2006b) The PSDR (World Bank 2001) eventually did not contribute significantly to the financing of development activities around protected areas. The consequence was a financial gap that increased significantly when the Durban Vision planned the creation of 3,953,000 hectares (World Bank 2006b).
101
Special Regime status, defined by Gouvernement de Madagascar (2005a). According to a World Bank note (World Bank 2007), based on data from ANGAP and other partners. 103 Development Intervention Fund. In French: “Fond d’Intervention pour le Développement.” 104 Rural Development Support Project. In French: “Projet de Soutien au Développement Rural.” 105 Translation by Pollini. Original text in French: “Quand le PE3 a été conçu, il était envisage d’établir six nouvelles aires protégées (trois terrestres et trois marines) pour une superficie totale de 324,000 hectares. Les projets de développement rural complémentaires comme le FID et PSDR financeraient des mesures d’accompagnement. Néanmoins, aucun des projets de développement rural avait des conditionnalités légales qui les obligeait à financer les mesures d’accompagnement du PE3 (bien que le PSDR et PE3 ont établi des protocoles d’ accord).” 102
103
Table 5: Malagasy protected areas in 2007 Category Number Existing ANGAP protected areas
39
Total area in hectares 1,612,000
Extension of ANGAP protected areas
11
359,000
During EP3
Creation of ANGAP protected areas
5
272,000
During EP3
New protected areas, non-ANGAP
4
919,000
2005
New protected areas, non-ANGAP
15
1,080,000
2006
New protected areas, nonANGAP, with identified promoter
47
1,619,000
2007-2008
New protected areas, nonANGAP, without identified promoter
38
595,000
2007-2008
New marine protected areas
4
325,000
2007-2008
Total expected 2008
163
6,456,000
Source: World-Bank (2007). Translation by Pollini.
104
Date of creation Before EP3
The supervision mission further pointed out that the Malagasy government adopted, on October 3, 2003, a procedural framework for mitigating impacts provoked by the creation of protected areas, Conservation Sites and land reserves 106 (Association Nationale pour la Gestion des Aires Protégées and MINEVEF 2003, 4). According to this procedural framework, It is obvious that the protected area management component [of NEAP] can cause restrictions on access to the natural resources in the protected areas to be created. In the same way, the forest ecosystem management component can cause, to a certain extent, restrictions on access to natural resources. 107 The procedural framework recommends the implementation of participatory diagnosis to identify the more vulnerable groups, and the identification of compensation measures such as productive micro-projects (e.g., swine husbandry, chicken farming, fish farming, handicrafts and agricultural intensification), training, the implementation of the GELOSE approach, improved stoves, improved charcoal production and the construction of basic infrastructure. A technical note of the World Bank (2007) estimated that the financial gap to finance these mitigation was about 2.58 million USD for the 12 newly created protected areas, which cover a total of 1.8 million hectares. This gap extends to 12 million USD if we consider all of the protected areas newly created by EP3 or to be created through 2008. If this gap was not filled, the safeguard policy of the World Bank would be at risk of not being implemented, which would slow the creation of protected areas and put the Bank at risk of being criticized by human right lobbies.
106
“Cadre de procédure pour la mitigation des impacts de la création des aires protégées, des sites de conservation, et des réserves foncières.” Prepared by experts of the World Bank and adopted by the Malagasy government on October 3, 2003. 107 Translation by Pollini. Original text in French:“Il est évident que la composante gestion des aires protégées (AP) peut causer des restrictions d'accès aux ressources naturelles dans les AP à créer. De même, la composante gestion des écosystèmes forestiers peut causer, dans une certaine mesure, des restrictions d'accès aux ressources naturelles.”
105
This problem was intensively debated during the EP3 joint committee of June 12, 2006 108 . Two issues were raised. First, the term “compensation” created a certain fear on the side of partners involved in the creation of protected areas on the ground. It suggested the idea of financial compensation that they could not afford to provide, whose amount would be difficult to calculate and which could provide disadvantages to the communities that had already engaged in new activities with no negative impact on the environment. The term “mitigation” was preferred because it did not imply an exact match between the negative impacts and the alternatives, and because mitigations can be ordinary development activities such as those already implemented on the ground. Second, the partners rejected the idea of being subjected to a procedural framework defined by the World Bank. The Malagasy government had adopted this procedural framework, but it was not clear whether it applied to all new protected areas or only to those financed by the World Bank. Eventually, this last option was accepted, but the partners would engage in mitigating the negative impacts. Third, the partners pointed out that EP3 departed from the previous phase of NEAP 109 by focusing on environmental issues and that rural development aspects had to be addressed by the Rural Development Action Plan. They agreed, however, that the concerns were justified and accepted the idea that the government issued a regulation that would define the mitigation measures associated with the creation of protected areas. At the end of 2006, the target date for achieving the Durban Vision (i.e., six million hectares under protection status) was postponed to 2012, which would help to achieve the Durban Vision objectives at a more reasonable pace.
108
This paragraph synthesizes the information provided by 5 informants interviewed in February 2007. During EP1, rural development aspects were addressed, although unsuccessfully, by the ICDP. During the second phases, the AGERAS component enlarged this approach at a larger scale.
109
106
10. Conclusion In conclusion, a tension between poverty alleviation and conservation objectives persisted during the three phases of NEAP, despite the fact that the environmental charter (Gouvernement de Madagascar 1990) put at the forefront the necessity to reconcile people and their environment. The failure of development activities led to a rejection of EP1’s ICDP approach and to replacing it with regional planning (AGERAS component of EP2) and community based natural resource management (GELOSE component). Despite these new orientations, negative socioeconomic impacts persisted during the second phase of NEAP. The case study in Part III and Chapter 10 in Part IV will provide more insights for understanding the reasons and assessing the consequences of the failure of mitigation measures. The case study in Part III and Chapter 10 in Part IV will also confirm that communities of farmers that practice tavy are first in line to pay the price of conservation. NEAP was not able to adjust its strategy to face these difficulties. Instead of addressing realities, the project’s techno-structure favored representations that did not question its own strategy, as shown by the tautological arguments of EP2’s evaluation report (OSIPD 1995). Parts IV and V will explore more in depth the reasons for this process and its consequences. In its third phase, NEAP eventually chose to concentrate on conservation aspects. The PADR (World Bank 2001) was expected to achieve development objectives, but did not do so, probably because it adopted the same ill-designed approach to rural development, and because it preferred to target areas with more productive potential and to work with rural entrepreneurs rather than ordinary farmers, as we will see in Chapter 6. Due to the significant increase of areas under protection status (Durban vision), the tensions between protected area managers and local communities may increase significantly in the future. During a period that might last
107
one or two decades, these tensions may be limited because forest clearing will be “possible” in areas not under protection status. After these unprotected forests are cleared, the protected areas will be at peril if no solution is found to support viable alternatives and compensate for the restrictions on access to resources and land.
108
PART III THE REALITIES: SLASH-AND-BURN AGRICULTURE IN BEFORONA
109
INTRODUCTION
This third part is dedicated to the analysis of realities in a rural commune located on the eastern edge of the Malagasy rain forest corridor: Beforona, 160 kilometers east from Antananarivo (Maps 3 and 4). Beforona has already been investigated in depth due to its easy access: it is crossed by National Road 2 which is well maintained and links the capital Antananarivo to Toamasina, the first port in the country. More than 10 PhD dissertations and about 20 Masters theses addressing this region have been completed during the two last decades, mostly with support of the Swiss funded BEMA and Terre-Tany projects, during the 1990s, and of the USAID funded LDI and PTE projects, from 1999 to 2004. For this reason, I did not deem it necessary to engage in an in-depth analysis of the region. Instead, I have synthesized the existing knowledge (Chapter 3) and attempted to fill the most significant gaps. The main gap concerns the situation in villages having limited access to market, which justified conducting a fast agrarian system diagnosis (Chapter 4 describes the methodology) in a remote village, Ambodilaingo, located 16 kilometers south from the road (Chapter 5). A second gap concerns the reflexive analysis of development practices, which justified analyzing the work of rural development projects working in the area (Chapter 6). Chapter 7 eventually, will present a social experiment implemented in Ambodilaingo in order to put into practice the conclusions of Chapters 5 and 6.
110
Map 3: Localization of the study area Source: FTM (National Geographic Institute) and BEMA project, in Rakotoarijaona (2005, 3) Note: the left map indicates the six provinces of Madagascar. The right map shows the limits of the 13 Fokontany of Beforona.
111
Beforona Ambodilaingo
Map 4: Regional map showing Beforona and Ambodilaingo. Source: FTM (Malagasy National Geographic Institute). Notes: • National Road 2 is indicated in red. • The distance from Moramanga (Department capital) to Beforona is 49 kilometers. • The roads indicated in black and some roads indicated in white (including the road to from Vatomandry to Ambalabe) are impractical by car. • The forest cover (dark green) is based on aerial photographs taken in the 1950s or 1960s. • The light yellow lines indicate the locations I visited. • The blue square delineates the area visualized on map 6 and 7 (page 123 to
125).
112
CHAPTER 3 OVERVIEW OF THE STUDY AREA
1. Introduction In this chapter, I provide an overview of the study area using material available in the literature. I will successively present the biophysical environment (Section 2) and the human and social environment (Section 3). Most information, including the majority of figures, was extracted from research conducted by the Terre-Tany and BEMA projects from 1989 to 1998, in the sites indicated on Map 5.
2. Biophysical environment 2.1. Location Beforona is located in the Moramanga District, Mangoro-Alaotra Region, Toamasina Province (Map 3 page 111). It is 160 kilometers away from Antananarivo, on National Road 2 which links Antananarivo to Toamasina, the main port of the country (Map 4 page 112). A railway also links these two cities and crosses the commune. It was mostly abandoned during the past decades but has been rehabilitated recently.
2.2. Topography The topography shows a regular pattern of small valleys, at an altitude of about 500 meters, dominated by small hills with steep slopes, 150 meters higher in elevation. The western edge is closed by a north-south escarpment that culminates at 1300 meters. This mountain chain, called the Betsimisaraka escarpment, separates Beforona from the Mangoro depression where the city of Moramanga is located. It is still forested (Vohidrazana forest), whereas the rest of the land is covered by secondary
113
Primary forest Agricultural and fallow land Grassland
Studied villages National Road 2 River Watersheds investigated in details
Map 5: Localization of sites investigated by the BEMA and TerreTany projects. Source: Brand (1997b, 3). Notes: the Salampinga site has not been studied in this dissertation.
114
vegetation. This topography is summarized in Figure 2 (from Le Bourdiec 1974), and illustrated by Photographs 1 and 2. The forest cover in Figure 2 is overestimated. Forests did not cover the commune of Beforona in 1970.
2.3. Climate The area experiences a tropical wet climate (Figure 3). The average annual rainfall is 2757 mm at the Marolafo research station (Nambena 2004). It increases on the eastern slope of the escarpment and decreases on its western slope (Analamazaotra station). The strongest rains last from late December to March. Cyclones are quite frequent during this period, with winds of more than 100 kilometers per hour. Cyclones Hutelle, Daisy and Geralda in 1993 and 1994 were particularly destructive. Their impact has been described in detail and mapped by Brand et al. (1997). Home gardens and lowland crops were destroyed by inundations and landslides. Houses and banana trees were also destroyed by strong wind. Precipitation is lower during the rest of the year but takes the form of drizzle, which falls almost every day in the winter, from early July to late August. September, October and November are the driest months but rain can still occur. As some people say, we can only distinguish a rain season (with big rains) and a rainy season (with drizzle). Nevertheless, dry periods of about 10 days can occur with significant probability from September to November. At the Marolafa station, the probability of having 18 consecutive dry days is 0.1, meaning that such dryness occurs only 10% of the time during this period (Brand 1997a). In sum, the climate is quite favorable to agriculture overall, but risky due to rainfall irregularities during the “dry” period and to the destructive effects of cyclones.
115
Zoning
Dense forest Mixed fallow Grassland Swamps
Transect Panganalana canal Iaroka river Iaroka river
Vohidrazana mountain
Ranomainty swamp
Coast line
Ambodilaingo
Forested escarpment
Mixed fallow with Ravenala
Grassland
Figure 2: Zoning and transect of the Betsimisaraka region. Source: Le Bourdiec (1974, 258). Adapted and translated by Pollini.
116
Swamp area
Photograph 1: The Betsimisaraka escarpment. Altitude ranges from 550 meters (where the picture has been taken, 10 km south from Beforona) to 1220 meters (peak in the background). The vegetation on the escarpment is a primary forest (Vohidrazana forest) whose valuable wood has been exploited. The rest of the land is covered with fallow, a tavy, and a temporary dwelling visible in the middle ground (left).
Photograph 2: Bottom lands and hills in Beforona. This photograph is taken from the same location as photograph 1, but on the opposite side (looking west). This bottom land belongs to the Fokontany Ambatoharanana, where the NGO Zanaky ny Ala built a dam (see chapter 7). Altitude ranges from 500 to 650 meters.
117
Rainfall Temperature
Rain (mm)
Tº (Cº)
J. A. S. O. N. D. J. F. M. A. M. J. Rain (mm)
Tº (Cº)
Meteostation Road Railway Forest
J. A. S. O. N. D. J. F. M. A. M. J. Rain (mm)
Tº (Cº)
Rain (mm)
Tº (Cº)
J. A. S. O. N. D. J. F. M. A. M. J.
J. A. S. O. N. D. J. F. M. A. M. J.
Figure 3 : Temperature and rainfall in the study area and surroundings. Source: Nambena (2007, 14). Data from the National Meteorology and BEMA project Note: The study area corresponds to the Marolafa station. The Analamazaotra and Moramanga stations show the lower rainfall on the western side of the Betsimisaraka escarpment. Translation and adaptations by Pollini.
118
Air is wet all the year round, with a maximum of 100% humidity in the night and a minimum of about 70% at noon. The average humidity ranges from about 75% in November to 87% in July. This rainfall pattern results in a dense hydrographic network with water running all year round in all rivers and most streams. Flooding is frequent on the alluvial plains that border the rivers, making the cultivation of these fertile soils quite risky. When big rains occur, it becomes dangerous to cross the rivers, which constrains commercial exchanges and school attendance by children. The average temperature ranges from about 16C° in July to 26C° in January (Figure 3). The minimal temperature can reach 6C° at night in the winter and maximal 38C° in the summer (Brand 1997a).
2.4. Vegetation and deforestation The original vegetation is an evergreen rain forest of middle altitude dominated by trees from the genera Tambourissa 110 and Weinmania 111 (Fara Lala 1999). Palisander (Dalbergia sp. 112 ), a tree with a high market value, is abundant but has been widely exploited. The flora is similar to what is encountered in Andasibe, where it has been studied by Razakanirina (1986). Several species of lemur live in the Vohidrazana forest, including the Babakoto (Indri indri) and the Varika (Propithecus diadema diadema). The wet climate prevents frequent natural fires and rain forests may have covered the whole region, including the coast, initially. The costal area was probably inhabited from the tenth century and may have been covered by a mosaic of Savannah, 110
Monimiaceae. Cunoniaceae. 112 Fabaceae. The genus Dalbergia, in Madagascar, includes a few species of rosewood, which are harvested for export, and many species of Palisander, which are less valuable and are sold on the domestic market and are utilized for making furniture. Rosewood is not found in Beforona. 111
119
primary forest, secondary vegetation, and agricultural land at that time (Chapter 1). Brand and Zurbuchen (1997), based on bibliographical review, studied the dynamic of deforestation during the last century. When Europeans first arrived on the east coast of Madagascar, they noticed that grassland and savannah dominated the slopes and the top of the hills, whereas forests were mostly located on bottom land and along streams (Hastie 1817, in Brand and Zurbuchen 1997). Further west, a contiguous forest existed that had no or almost no inhabitants. During the first part of the nineteenth century, the savannah area progressed further west and in the second part, the forest between Ampasibe and Beforona was partially cleared, resulting in a mosaic pattern of forest and fallow land. This clearing may have been quite complete by the end of the nineteenth century (Ellis 1856, in Brand and Zurbuchen 1997; Julien 1909, in Brand and Zurbuchen 1997). It is believed that settlement in Beforona started in the eighteenth century. But population density was still low in the nineteenth century, with forests covering most land west of Beforona. Brand and Zurbuchen (1997) used aerial photographs from the years 1957, 1967, 1987 and 1994 to study deforestation dynamics during the last decades (Figure 4). The most intense clearing occurred after 1967, with a rate of 21.2 ha/year from 1957 to 1967, 39.3 from 1967 to 1987 and 46.5 from 1987 to 1994 (measurements were taken from a sample that did not cover the whole commune). This increase can be explained by the demographic explosion that started in the 1950s, by the low control that resulted from the political change in the 1970s and by the economic crisis in the 1980s. In the 1990s, most of the remaining forests were localized on the Betsimisaraka escarpment, at an altitude above 800 meters which is not very suitable for rice cultivation due to high rainfall. This factor, and/or a stronger enforcement of the ban on forest clearing, led to a decrease of the deforestation rate. Nambena (2007) studied the deforestation in the area between 1994 and 2000 and complemented Brand
120
89,7
88,0
85,8 51,4
0% 1957 1967 1987 Pourcentage non Non forested de land (%)Forêt (%) Pourcentage Forêt (%) Forested landde (%) Vites s e de défores tation Deforestation (Hectare perha/an year)
1994
1997
10,3
10 12,0
14,2
20,9
20
2000 Années
Figure 4: Deforestation pace in Beforona from 1957 to 2000. Source: Nambena (2007, 24). Data from: Brand and Zurbuchen (1997) and Nambena (2007). Translation by Pollini. Note: Based on aerial photographs of the area mapped page 143.
121
0
Défrichement Ha/an
79,1 46,5
63,0
26,6
30
37,0
25%
40
21,2
50% 41,4
% de surface
75%
60 50
39,3
100%
58,6
Ha/year
and Zurbuchen’s (1997) measurement (Figure 4 page 121). He showed that the edge of the forest main corridor did not greatly move, but that all the remaining forest patches east from the corridor were cleared during this period. McConnell et al. (2004) also studied deforestation in the area for the period 1957-2000 (Map 6). They created a model to predict the conversion of forest to agriculture using five variables: elevation, slope, distance from forest edge, distance from village, and an estimation of population density. They ran their model to predict land use change between 1957 and 2000 and compared the outcome with the actual deforestation. They showed that distance from a village, distance from forest edge and elevation have the highest explanatory value of observed forest clearing. Slope has the lower effect, which contradicts the outcomes of Green and Sussman (1990). For Mc Connell et al., farmers prefer to clear land with less slope but in the end, steep slopes are cleared as well when no more land is available. In a second step, McConnell et al. (2004) mapped the difference between the prediction of the model and the actual land use change (Map 7). They showed that the surroundings of the protected areas and the eastern side of the escarpment were less cleared than predicted by the model. The western side, on the other hand, was more cleared than predicted. This could be explained by an aspect effect that has not been integrated into the model: on the eastern side, the available land for agriculture is located at the altitudinal limit for rice cultivation, which seems to be at about 800-900 meters. Above this elevation, rice cultivation is more risky due to an excess of rain, as said earlier and as we will see in Chapter 5. On the western side, rainfall is lower and rice can be cultivated at higher altitude. McConnell et al. assert that facing this natural frontier, farmers “clean-up” the forest remnants (2004, 183), which is consistent with Nambena’s (2007) conclusion. On the western side, to the contrary, the pattern is that of recent agrarian colonization, with land being cleared inside large forest blocks. McConnell et al.
122
Beforona
Beforona
1957
1976
Ambodilaingo
Ambodilaingo
Map 6. Forest clearing in the study area, from 1957 to 2000. Sources: Maps from McConnel et al. (2004,173), using sources from “(a) IGN/FTM topographic maps S-47N and S-47S (PERINET and LAKATO), based on aerial photographs from 1957, scale: 1:50,000, projection: Conformal Laborde, publisher: L’Institut Geographique National (France)/Foiben-Taosarintanin’I Madagasikara (Madagascar); (b) Landsat MSS, 5 June 1976, Scene ID: LM2169073007615790; (c) Landsat TM, 21 November 1994, Scene ID: LT5158073009432510; (d) Landsat ETM+, 19 April 2000, Scene ID: LE7158073000011050.” (McConnel et al. 2004, 173) Names and arrows added by Pollini.
123
Beforona
1994
Beforona
2000
Ambodilaingo Map 6 (continued).
124
Ambodilaingo
Mantadia National Park
Analamazaotra Special Reserve
Ambavaniasy
Vohimana forest NR2
Beforona
Vohidrazana forest
Ambodilaingo
Legend:
The lower values (blue) are forests predicted to be cleared, but eventually not cleared. The higher values (red) are forests predicted not to be cleared, but eventually cleared.
Comments: Lower values are concentrated around the protected areas (the Vohimana forest is a concession managed by the NGO MATE). Higher values are concentrated to the west, which could be explained by an aspect effect not considered by the model (rice can be grown at higher altitude on the western side of the escarpment) or by a displacement of pressures. Map 7: Land use change compared to prediction by model. Source: McConnel et al. (2004, 178). Names, arrows and comments added by Pollini. 125 136
(2004, 171) conclude that “while overall land change patterns in the region are largely explained by elevation and village proximity, more specific, sub-regional, trajectories reflect the signatures of institutions governing access to land.” In other words, the creation of protected areas (Mantadia National Park) may have provoked a displacement of the pressures at regional scale. This conclusion, however, could have been biased by unforeseen aspect effects not integrated into the model. Map 8, elaborated by Conservation International Madagascar, shows the regional deforestation in the period 1993-2000. Today, most land is covered by secondary vegetation (Map 9). Deforested land is either cultivated or is dominated by different types of secondary vegetation that will be described in the section about land uses.
2.5. Soils The soils are ferralitic with strong desaturation, except on forest land and recently cleared land where they still have an elevated organic matter content, and in the bottom lands and lower parts of the slopes where they benefit from transfer of particles by erosion. The level of degradation is correlated with the number and frequency of clearings and burnings that have occurred and to the vulnerability of the soils, itself determined by the slope and organic content. These aspects will be addressed in more detail in the section about land uses. The main constraints for cultivation are a low phosphorus availability and aluminum toxicity. Concerning erosion, it does not lead to catastrophic events except landslides during cyclones (Brand and Rakotovao 1997; Brand et al. 1997). It is more intense on soils plowed for ginger cultivation than on slash-and-burn systems (Brand and Rakotovao 1997).
126
Zahamena National Park Betampona Strict Reserve Ambohilero forest
Mangerivola Special Reserve
Mantadia National Park
Analamazaotra Special Reserve
Beforona Ambodilaingo
Protected areas Classified forests Water Forest Cleared forest (1993-2000) Non forest Cloud on forest Cloud
Map 8: Protected areas and regional deforestation in Mantadia-Zahamena corridor. Source: Conservation International Madagascar, July 2003. Legend translated by myself. Blue square added by myself for showing area on map 7 and 8, page 134 to 136.
127
Map 9: vegetation in Beforona and in the surrounding. Source: Nambena (2007, 55). Translation by Pollini.
128
Based on Landsat image from September 1993
Area on maps 7 and 8 page 134 to 136
Railway National Road 2 River Lake
Provincial capital District capital Commune of Beforona Protected area
Low altitude savannah (0-100 m; rain > 2500mm) Higher altitude savannah (750-1000 m; rain < 1750mm) Low altitude fallow land (0-350 m; rain > 2500mm) Medium altitude fallow land (350-750 m; rain > 2000 mm) Medium altitude rain forest (750-1000 m; rain > 1750mm) Mountain rain forest (altitude > 1000 m; rain > 1750 mm)
129
3. Population and society 3.1. Demography As in the rest of Madagascar, a population explosion occurred in the second half of the twentieth century. Annual population growth is now 2.7%, which is close to the national average. Map 10 shows the demographic density in 1997 in Beforona and in Ranomafana, a commune located farther east. We can notice that the fokontany with lowest population densities are not necessarily those located close to the forest. This can be explained by a higher land degradation, which encourages emigration from these areas that have been inhabited for a longer period. Figure 5 shows the age pyramid in Marolafa, a fokonany 113 of Beforona. It shows that the population is very young, which is typical of most rural areas of Madagascar.
3.2. Ethnicity The population (Table 6) is mostly composed of Betsimisaraka people, who have a Bezanozano origin for part of them, as I will show in Chapter 5. The second most important group is the Sihanaka, who also practice agriculture. The other groups are the Antesaka, the Merina and the Betsileo, who established mostly to develop trading activities or take jobs, and a very small minority of Antandroy charcoal makers and traders of Chinese origin. The Merina and Chinese people dominate the economic, administrative and political life. Chinese and Merina traders use their earnings to buy land and are now the biggest landowners. They usually buy the most suitable land for intensive agriculture, especially the bottom land that can easily be irrigated. They also appropriate land by matrimonial alliance and by practicing usury: the debtors use their land asset as a guaranty to purchase necessity products on credit, frequently failing to
113
Beforona is composed of 13 fokontany (see note 1 page 3).
130
Map 10: Population density per Fokontany in the communes of Beforona and Ranomafana, in 1991. Source: Zurbuchen (no date), in Moor and Bark (1998, 12).
131
Age
Women
Men
Number of persons
Number of persons
Figure 5: Age pyramid in Marovoalavo, Beforona. Source: Zurbuchen (no date), in Barck and Moor (1998, 15). Translation by Pollini.
Table 6: Population in Beforona by ethnic group. Source: Barck and Moor (1998). Translation by Pollini. Ethnic group Betsimisaraka Merina Sihananaka Antesaka Betsileo Antandroy Chinese origin Total
Population 9331 981 928 316 58 5 2 11621
% of total population 80.3 8.4 8.0 2.7 0.5
100
132
Main occupation Agriculture Trade, administration, agriculture Agriculture Trade and agriculture Trade and labor force Charcoal production Trade
pay back these loans and eventually losing their land. Land appropriated in this way is mostly located along National Road 2.
3.3. Education According to Nambena (2004), the first schools in Beforona were built in the 1930s. All fokontany had an elementary school in 1975 and the junior high school opened in 1979. Children have to move to the district town of Moramanga when they want to attend high school. The education level decreased during the last decades. People educated in the 1950s are more literate than the new generation. In remote villages, state teachers often refuse to take their positions and parents replace them with nonprofessional teachers whose skills are not always sufficient. Parents recruit them locally and pay them themselves. Many schools destroyed by the cyclones have not been rebuilt. During the rainy season, the danger of crossing rivers can impede school attendance. Some parents are discouraged by the cost of school necessities or need their children to support them in domestic and agricultural activities, such as pest control.
3.4. Healthcare A local health center was built in 1983 (CSB I 114 ) and provided with more equipment and means in 1996 (CSB II 115 ). It is supposed to have a doctor and a nurse, but the appointed staff often refuses its assignment or manages to be moved to a bigger city after a short time. Staff with no or little medical training, but fortunately a long experience in the region, often have to provide healthcare. A private doctor also works in the commune. Consultation is cheap (a few USD for the most common
114 115
Basic Health Center Level 1. In French : Centre de Sante de Base de Niveau 1. Basic Health Center Level 2. In French : Centre de Sante de Base de Niveau 2.
133
illnesses, including the cost of pills) but still represents a sacrifice for poor households. The main constraint to health care is the distance to the health center. Often, only people living close to the road (less than five kilometers) consult a doctor. Others “wait,” use traditional medicine or practice auto-medication, because traveling to Beforona implies some costs they cannot bear (the time of the persons that accompany them, the need to find a place to stay and to buy food if the health care lasts several days). The influenza epidemic of 2002, presented in Chapter 5, will provide an illustration of these issues.
3.5. Brief political and socio-economic history 116 At the time of the first settlements, village life was controlled by the agricultural calendar and traditional ceremonies. As we saw in Chapter 1, Betsimisaraka people were organized into small kingdoms or chiefdoms and were peaceful. But those from the hinterlands suffered from raids for slavery from those living on the coast, which may explain their move westward and the population distribution in space. The economy has always been based on agriculture. There are a few mining activities in the commune (gold, corindon), but these do not provide significant income. Slash-and-burn cultivation has always been the main land use and forest clearing started to be repressed by the Merina administration in the nineteenth century. Merina people also introduced forced labor for the construction of the first road that linked the coast to the highland (Nambena 2004). Perturbation of the local economy increased during the colonial period. According to Nambena (2004), the first French colonists arrived at the end of the
116
The information is this section is mostly from Nambena (2004) and Styger (2004).
134
nineteenth century. They introduced the SMOTIG 117 (forced labor) for the improvement of the road, the construction of the railway and the planting of eucalyptus to feed the trains’ steam engines. The construction of these infrastructures attracted labor force and traders from other regions, resulting in a demographic increase that accelerated deforestation. A per capita tax also was introduced. A regulation issued on October 30, 1904, stated that those not able to pay this tax owed labor days to the administration. Farmers were asked to move to the Andasibe camp, located 25 kilometers west from Beforona, to plant eucalyptus or do other jobs, for the benefit of administrators or colonists. Farmers developed cash crops (mostly coffee and bananas) as a way to pay these taxes and to avoid forced labor. Cash crops also were aimed at propelling indigenous people into a market economy, leading to some changes in the traditional modes of consumption. Traders from other ethnic groups moved to the area to open small shops and collect agricultural products (Nambena 2004). During the colonial period, an agricultural service supported the development of irrigation schemes and the diffusion of new rice varieties and cultivation techniques (Nambena 2004). The first veterinaries also toured the region. Forest clearing was forbidden and the practice of tavy was submitted to increasing restrictions. Agents from the colonial administration were regularly visiting the area for the purpose of control. Wood extraction also started during this period, but rather in the neighboring commune of Ambatovolo. In Ambavaniasy (located on National Road 2, 10 kilometers west from Beforona), two colonial companies, the “Compagnie coloniale” and “grande Ile” initiated wood extraction in the 1930s, attracting immigrants that worked as loggers (Styger 2004).
117
Labor service for General Interest Work. In French: Service de Main d’Oeuvre pour les Travaux d’Intérêt élevé Général.
135
The colonial pressures provoked strong cultural and social resistances that have been analyzed in detail by Althabe (1969). The resistance peaked with the revolt of 1947, which resulted from an increase in the number of forced labor days in the early 1940s, due to the war effort required by the French government. After these events, economic and political pressure on the population decreased and a more prosperous era started. Forced labor was abolished by the end of colonization but the per capita tax lasted until the second republic. The period of the first republic, after independence, was relatively prosperous, as in the rest of the country (Chapter 2). The government provided support to develop irrigated rice cultivation. The road network provided more opportunities to sell agricultural products and veterinaries still visited the area. Education was more efficient than at present and cash crops, mostly coffee, were sold at a satisfying price. It seems that there was a relative regional specialization, Beforona producing coffee and buying rice grown in Ambatondrazaka. Staff from Forest Services were still patrolling the area to control forest clearing. They forbade tavy on the upper parts of the hills but authorized it on lower parts. They also provided support for creating irrigation schemes because the forest and hydraulic services depended on the same direction of the Ministry of Water and Forest at this time, as we saw in Chapter 2. Paddy fields were more extensive than at present, despite the fact that population density was significantly lower. Unfortunately, the situation started to degrade in the early 1970s. The price of coffee and all agricultural products plummeted. Hydraulic infrastructure started to degrade, due to the end of state supports and to the farmers’ lack of resources. The state entered into bankruptcy and state representatives stopped visiting the area. The new regime, after 1975, started a “do as you please” policy that encouraged forest
136
clearing. The ban on clearing of hill tops was not respected anymore (Styger 2004) and the population went back to an economy almost entirely based on tavy. The trend has been reversed in the late 1980s by the renovation of the road, which created new market opportunities, mostly to sell ginger and bananas, and by a return to a ban on forest clearing, which, however, had little effect due to the absence of Forest Services (Styger 2004). But the low price of agricultural products and the high cost of transportation limited these opportunities to villages located close to the road. Due to the permanent degradation of soil fertility, livelihoods have continued to decline to the present, at least in remote villages, as we will see in Chapter 5. A new economic shock occurred in 2002. In August 2002, the Malagasy government started a strong campaign against bush fires. The new president had just been elected and the donors supporting the National Environmental Action Plan needed a strong signal of his commitment to protecting the environment. On August 7, 2002, a decree (Gouvernement de Madagascar 2002) was issued that defined measures to encourage the prevention and eradication of bush fires. This decree required the creation of a bush fire committee in each Department. These committees, presided by the district chief, were to include representatives of decentralized territorial collectivities; of the environment, forestry, agriculture, and veterinary state services; of the meteorological services; of the army and of local environmental nongovernmental organizations (Gouvernement de Madagascar 2002, article 1). They were in charge of classifying the communes according to the following criterion: •
deserving communes, who made significant efforts to fight against fires, as shown by the fact that they have not been invaded by fires,
•
encouraged communes, who actively contributed to the extinguishing of wild fires in their territory and within which a diminution of burned land has been recorded,
137
•
failing communes, who did not make efforts and still need to become more responsible. (Gouvernement de Madagascar 2002, article 2).
Premiums and certificates were delivered to the deserving and encouraged communes, and these certificates were the condition for obtaining funding to implement the communal development plan (Article 3). This decree does not forbid tavy. It merely defines incentives and penalties for the communes that control or do not control fires. It is thus consistent with the spirit of the ordonnance 60-127 (Gouvernement de Madagascar 1960), still valid, which distinguishes different types of fires and does not forbid tavy per se (Chapter 2). But on September 28, 2002 the President of Madagascar made a speech on Malagasy radio. As in the decree, the word tavy was not mentioned in this discourse. But the speech was followed by a series of arrests of farmers who practiced tavy in Ambatovolo, the commune located west from Beforona. The information spread very quickly in the region. Most farmers were afraid of doing tavy, including on degraded secondary vegetation. The antifire campaign became an anti-tavy campaign. In 2003, the Toamasina Province confirmed this move by issuing a decree that forbids tavy. The repression was, however, less strong than in 2002 and the situation returned to normal in 2004. Chapter 5 will show the impact of the anti-tavy campaign in the fokontany of Ambodilaingo, while Chapter 6 will explain the relationships between these measures and the presence of rural development projects in the area.
3.6. Social organization The current social organization involves a state administration, whose mode of organization frequently changed along time, and traditional authorities, which always existed but are weakening.
138
The state is currently represented in each fokontany by a “Chef de quartier” (fokontany chief) appointed by the mayor and supported by a village chief in each hamlet inside the fokontany. These representatives are not paid but receive a small indemnity (about five USD/month). The fokontany chiefs sign administrative acts such as birth and death acts or residence permits. Their tasks are not burdensome. They have no office and usually live in the same way as other villagers. The traditional leader, at the scale of a community (fokonolona), is the Tangalamena. He is in charge of celebrating ceremonies, principally burials and land allocation for doing tavy. He serves as an intermediary between the population and the ancestors. He is supported by a spokesman, the Vavanjaka, who in certain case can have more influence than him on the population. Usually, all the inhabitants of a village constitute a fokonolona 118 and are represented by the same Tangalamena and Vavanjaka, but there can be exceptions. Important decisions are taken by a village elder council, which includes representatives of each lineage, plus sometimes representatives of the young, the women, and a few influential or more educated people such as the teacher or people who had the opportunity to travel far away outside the village. The village council, in collaboration with administrative representatives, can create community laws called dinas. The dinas are usually not written but are the basis on which the community life is ruled. The most important ones can nevertheless be written and registered at the commune, which gives them some legal power. Dinas define behavior rules, offences and punishment, concerning for example the ignition of uncontrolled fires, theft of agricultural products or violation of traditional taboos.
118
The fokonolona in the traditional sense thus does not correspond to the modern definition, where the fokonolona is the population of the fokontany.
139
Dinas can also be created at a lower level, by a group of farmers who want to organize collective activities such as maintaining an irrigation scheme. Ceremonies have an important role to play in every day life. They consist mostly in sacrificing oxen as a cult to the ancestors. The main reasons for these celebrations are funerals, the birth of a tenth child, circumcision, the obtaining of rights to cultivate certain land and the violation of taboos. I will describe these ceremonies in more detail in the Chapter 5 and will show that they changed over time. Reciprocal aid between households is used for certain agricultural works but tends to be replaced by paid labor.
3.7. Land Tenure The traditional authorities are in charge of allocating land to households, which allows a quite equitable repartition of resources. Most land is not legally owned and belongs to the state. Only traders and a few farmers with more education, more financial resources and more connection with state domain services manage to obtain land ownership. These privately owned lands are mostly irrigated rice fields located along the road. Migrant traders manage to appropriate the best lands by buying them, by matrimonial alliance or by usury. In this case, the land is used as a guarantee and taken from the borrower when he fails to pay. The land market concerns both domain land (not entitled but whose use rights are recognized by traditional authorities) and land privately owned (entitled). In the later case, the land can be sold at a higher price.
3.8. Land uses I will describe in more detail the farming, cropping and animal husbandry systems in Chapter 5, which focuses on the village of Ambodilaingo, and in Chapter 6, which
140
reviews the alternatives to slash-and-burn cultivation. As an introduction and for a better understanding of the sections that follow, I will provide a few elements her. Table 7 lists the main cultivated crops and Map 11 shows the repartition of land uses in the surroundings of Beforona. Fallow land clearly dominates the landscape. Figure 6 represents the allocation of work according to activities, showing that slash-and-burn cultivation employs most of the labor force.
3.8.1. Slash-and-burn cultivation Tavy (Photograph 3 and 4 page 145 and Photographs 14 and 15 in Chapter 5, page 187) consists in clearing and burning a piece of land in order to grow rice associated with a variety of other annuals crops the first year (see list and frequency in Table 8). Cassava, sometimes sweet potato, and rarely beans are planted the second year, on part of the plot only. Cassava can also be planted the first year after burning, if the land is too degraded to support rice cultivation. But in this case, the field is not called a tavy. Tavy gives better harvest if it is practiced on forest fertile soils but forest clearing is forbidden by the state, as we saw. Moreover, the remaining forests are located on the Betsimisaraka escarpment, which is not very suitable for rice cultivation due to excessive rainfall. For these reasons, tavy is mostly practiced on secondary vegetation. It is also sometimes practiced in eucalyptus woodlands (Photograph 4), which reveals the lack of space and the integrative approach to land management by farmers. Table 9 lists the main species encountered in fallow land. On the most fertile soils, which are mainly found in recently cleared areas, secondary vegetation is
141
Table 7: Cultivated species in Beforona Common name
Scientific name
Malagasy name
French name
Aspargus bean
Vigna unguiculata ssp sesquipedalis
Voanemba
Ambirique
Pineapple
Ananas comosus
Mananasy
Ananas
Egg plant
Solanum melongena
Angivy
Aubergine amère
Sugar cane
Saccharum officinarum
Fary
Canne à sucre
Carrot
Daucus carotta
Karôty
Carotte
Chayote
Sechium edule
Saosety
Chayotte
Chinese cabbage
Brassica chinensis
Petsay
Choux de chine
Cabbage
Brassica oleracea
Laisoa
Choux pommes
Cucumber
Cucumus sativus
Konkombra
Concombre
Squashes
Cucurbita pepo
Korozety
Courgette
Watercress
Nasturtium officinale
Anandrano
Cresson
Ginger
Zingiber officinale
Sakamalao
Gingembre
Common bean
Phaseolus vulgaris
Tsaramaso
Haricot
Yam
Dioscorea sp
Oviala
Igname
Maize
Zea mays
Katsaka
Maïs
Cassava
Manihot altissima
Mangahazo
Manioc
Watermelon
Citrulus lunatus
Voatango
Melon d’eau
Black nightshade
Solanum nigum
Anamamy
Morelle noire
Cowpea
Vigna unguiculata ssp unguiculata
Voantsoroka
Niébé
Onion
Allium cepa
Tongolo be
Oignon
Sweet potato
Ipomea batatas
Vomanga
Patate douce
Pea
Pisum sativum
Petipoa
Petits pois
Chilli pepper
Capsicum frutescens
Sakay
Piment
Bambara groundnut
Vigna substerranea
Voanjobory
Pois bambara
Lima beans
Phaseolus lunatus
Kabaro
Pois du cap
Taro
Colocasia antiquorum
Saonjo
Taro
Tomato
Lycopersicon esculentum
Voatabia
Tomate
Bambara groundnut
Voandzeia subterranea
Voanjobory
Voandzou
Source: Ravelomandeha (2003). Translation by Pollini (legend).
142
Villages National Road 2 Rivers and streams Home gardens Slash-and-burn rice (tavy) Irrigated rice fields Ginger Cassava Irrigated rice fields in fallow Natural forest Fallow land Reforestation
143
Map 11: Land use in a few Fokontany in Beforona. Source: Brand and Randriamboavonjy (1997, 92). Translation by Pollini. Note: Ambodilaingo is not mapped.
Mining (gold) Handicraft Mutual aid Building Ceremonies Other crops Ginger Home garden Paddy fields Slash-andburn cultivation Jul. Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May. June.
Figure 6: Distribution of work according to activities in Beforona. Source: Moor (1997, 38). Translation by Pollini.
144
Photograph 3: Tavy in ordinary fallow land, about 7 km south from Beforona. This tavy is typical of the region. The vegetation that was cleared is a mix of ligneous (Psiadia altissima), subligneous (Rubus moluccanus), and herbaceous species (Aframamum angustifolium, visible in the foreground). It is not clear whether these tavy are forbidden or not by the 60-127 law of 1960 (see chapter 2).
Photograph 4: Tavy in a eucalyptus woodland in Beforona. This picture shows that farmers manage space in an integrative way, and that space is insufficient despite the fact that fallow land covers vast areas. A few trees are not cut in order to provide construction wood in the future. Farmers contend that denser eucalyptus woodland grows after the rice is harvested.
145
Table 8: Plants cultivated in tavy in Ambinantsavolo, a fokontany in Beforona Malagasy name English name Scientific name Frequency (38 households) Tsaramaso Beans 92 % Phaseolus vulgaris Haricots Katsaka Maïs
Corn
Zea mays
83 %
Kokombra Concombre
Cucumber
Cucumis sativus
58 %
Kabaro or Tsidimy
Lima bean
Phaseolus lunatus
46 %
Tsiasisa or maley
Asparagus bean
Vigna unguiculata ssp sesquipedalis
38 %
Vonemba
Black eyed pea
Vigna unguiculata unguiculata
33 %
Voatavo or pongy
Squash
Cucurbita pepo
29 %
Anamamy
Black nightshade
Solanum nigrum, S. americanum,
25 %
Anamalao
Paracress
Spilanthes oleracea
Petsay, ramirebaka
Chinese cabbage
Brassica pekinensis, B. chinensis
25 %
Voatango
Watermelon
Cucumis sp.
8%
Voanjo
Groundnut
Arachis hypogaea
8%
Apemba
Sorghum
Sorghum bicolor
Source: Nambena (2004). Translation by Pollini.
146
Table 9: Dominant species in fallow in the region of Beforona and Moramanga.
147 Source: Styger (2004, 61).
Figure 7: Yield in tavy in relation to the type of fallow vegetation. Source: Styger (2004, 86).
148
dominated by woody species 119 which can form a closed canopy of about three to five meters. Trema orientalis and Solanum Mauritianum indicate the most fertile soils, usually on land that has been cultivated just one time after forest clearing. On the rest of the land, the vegetation is dominated by herbaceous perennials 120 or semi ligneous shrublets 121 when the soils are still in acceptable condition, or by a mix of shrubs and ferns 122 or grasses 123 when the soil has been degraded by more frequent or numerous fires. Aluminum toxicity and phosphorus depletion seem to be the main factors determining soil fertility (Pfund 2000). It is, however, still controversial whether soil fertility or weed invasion is the main cause of abandonment of degraded land (Pfund 2000). Styger (2003) studied the retrogressive succession of fallow species along successive slash-and-burn cultivation cycles and measured the rice yields obtained on each of these vegetation types. Her results are summarized in Figure 7 (page 148). The correlation between fertility and the number of previous cultivation cycles is, however, not as clear as it appears in this diagram. Topography is also an important factor as degradation is faster on steep slopes and slower on the lower part of the slopes, due to accumulation of nutrients and deeper soils (Pfund 2000). As Beforona has been inhabited for over a century (Section 2.4), most land has been cultivated more than six times, meaning that overall land degradation may be slower than what is shown in Figure 7 (page 148). We will see in Chapter 5 that uncontrolled fires, rather than tavy, may explain the later stages (ferns and grasses) of this retrogressive succession.
119
Psiadia altissima (DC.) Drake (Asteraceae), Harungana madagascariensis Lam. ex Poir. (Clusiaceae), Solanum mauritianum Scop.(Solanaceae) and Trema orientalis L. Blume (Ulmaceae). 120 Aframomum angustifolium [Sonn.] K. Schum (Zingiberaceae). 121 Lantana camara L (Verbenaceae), Rubus moluccanus L (Rosaceae) and Clidemia hirta [L.] D. Don (Melastomaceae). 122 Pteridium acquilinum [L.] Kuhn and Sticherus flagellaris [Bory ex Willd.] Ching. 123 Imperata cylindrica [L.] Raeusch (Poaceae).
149
In Beforona, tavy is often practiced on land more or less densely covered by Rubus, Lantana and Aframomum, meaning that rice yield ranges from 0.5 to 1 ton per hectare. Whether the legislation about vegetation clearing and burning (Gouvernement de Madagascar 1960) applies to this type of tavy is unclear, as we saw in Chapter 2 (Aframomum is not ligneous and while Rubus and Lantana stems are ligneous, they can hardly be regarded as woody). Farmers living close to the forest have access to less degraded land and enjoy higher yields (1 to 2 tons per hectare). But in this case, tavy implies the clearing of dense ligneous vegetation (Trema, Harungana and Psiadia, small trees or shrubs used as firewood) and is clearly forbidden by the law (although the law has seldom been enforced). Vegetation degradation follows an east to west gradient, the less degraded land being encountered in the west, where clearing has been more recent. One could conclude that the region has been populated by successive waves of migrants coming from the east and presents a pattern typical of most agricultural frontiers. I will show in Chapter 5 that despite the fact that this tendency is real, things are actually not that simple. Vegetation type indicates how much yield can be expected on average from a tavy, but the variation around this average is very elevated, due to climatic risks and frequent pest attacks, mostly rats, a small bird called Fody (Foudia madagascariensis) and Behatoka, a small black coleopterea (Heteronychus plebejus Klug, 1832) that appeared in the area around 1965. The practice of tavy just requires a machete, an axe if it is done in the forest and a stick for sowing. It is accessible for all households. Tavy also resists cyclones better than most other crops. According to Moor (1998b), tavy would be the less profitable system, with a labor productivity ranging from 2000 to 3000 FMg/ day of labor (about 0.25 to 0.35
150
USD). The average production per household would vary from about 600 to 1200 kilograms of rice. We will see in Chapter 5 that the risk factor, the limited investment capacity of farmers and the higher opportunity cost of labor during certain periods of the year can explain why this land use is still practiced, in spite of its low profitability. Normally, firebreaks are created around tavy to avoid wildfires. These efforts vary according to the presence of state control. They were systematically made during the anti-tavy campaign. Overall, wildfires are encountered but less frequently than in the highland or western Madagascar. They result from the absence of firebreaks, or from conflicts between villages or families, usually over land tenure issues.
3.8.2. Paddy fields Beside tavy, irrigated rice cultivation is the second most important cropping system if we consider the labor force allocated to it (Figure 6 page 144). It is however not practiced on large areas, due to the narrowness of bottom lands, to the limited investment capacity of farmers and to the damage caused by cyclones to irrigation schemes. Rice can be harvested twice a year on the same land (in January and in June) if the plot is irrigated. The government provided some support for the development of irrigation schemes in the past, mostly during the first republic, but most dams have been destroyed by cyclones and never rebuilt. Irrigated rice cultivation, and its profitability in comparison with tavy, will be studied in more detail in Chapter 5 and 6.
3.8.3. Home gardens Home gardens (Photograph 5) are important in the local economy. Finding an appropriate spot to create a home garden is a concern for most new households. This system is usually located in the lower part of the slopes or along small streams, due to more fertile soils and wetter conditions. Home gardens are dominated by a mix
151
Photograph 5: A home garden in Sahavolo (Beforona). Photograph by Nambena (2004, 202). The big tree on the left is an Albizzia. Banana trees are visible in the foreground and coffee trees grow behind the farmer.
152
bananas and coffee trees, both introduced around 1940 (Nambena 2004). Banana became an important cash crop in the area after the renovation of the road in 1987. It has the advantage of requiring little care once established and of providing a low but constant income which helps to buy first necessity products. The size of home gardens and the number of banana trees decreases when distance from the road increases (Figure 8), because the cost of transportation renders banana cultivation unprofitable (Photograph 6). Coffee used to provide significant income after its introduction during the colonial period, but its price declined in the late 1990s. Sugarcane, which provides sugar for local consumption and serves for the fabrication of local rum 124 , plus a few fruit trees, mainly avocado, oranges, grapefruits and jackfruits, are associated with banana and coffee trees. Their commercialization is rendered difficult by the lack of market opportunities and by transportation issues (Photograph 6). Fruits are often left rotting on the ground.
3.8.4. Ginger cultivation After the decline of coffee cultivation, ginger became the main cash crop. But as with banana, its cultivation is profitable only along the road (less than ten kilometers) due to transportation costs. Its price also tends to decline. Ginger can cover significant areas close to the road. It leads to the appearance of new land uses because it implies plowing the land, which allows exploiting better soil fertility. Rice and sometimes other crops are planted on the same plot after ginger is harvested, which opens the way to an evolution toward new cropping systems where crop rotation is applied while the fallow period is shortened (Photograph 7). The problem is that plowing can cause considerable erosion if it is practiced on steep slopes (Brand and Rakotovao 1997). Farmers attempt to reduce soil loss by creating drainage ditches 124
In Malagasy language: tokagasy.
153
Number of trees
Nom bre de pieds plantés
80
Bananas Banane Coffee Café Fruit trees Fruitiers CSugar annecane à sucre
60
40
20
0 5
10
15
20
Distance from the road in Km
Distance (km) à la RN2
Figure 8: Relation between the extension of home gardens and the distance from National Road 2. Source: Nambena (2007, 27) with data from Randrianarisoa (1998).
154
Photograph 6: Banana transportation. These young men are in good health and live close to the road (5 kilometers). They carry about 50 kilograms of bananas, which they will sell for about 300 FMg/Kg (15000 FMg or 1.5 USD in total). By doing this several times a week, they can buy rice and first necessity products for their family during the rice shortage period and can invest in intensification (see chapter 7). The photograph was taken 5 kilometers south from Beforona.
155
Photograph 7: Landscape transformation in Beforona. This hillside announces a transition from Boserup’s stages 3 (short fallow) to 4 (annual cropping), as we will see in Chapters 5 and 6. Plowing is the operation which makes this transition technically possible. Markets (for ginger) provide the opportunities and incentives that render it profitable to invest significant labor in the new system. Crop rotations are not yet fixed but farmers collectively learn by a series of trials and errors.
156
around the plot (Photographs 8, 9 and 10). In spite of these problems, the future land uses of Beforona may be derived from this system (Photograph 7 page 156), as we will see in Chapter 6. Displacing ginger fields to the bottom part of the slopes or inside home gardens could be a way to conciliate farmer strategies with better soil management practices. Chapter 6 will explore this possibility.
3.8.5. Vegetables Vegetable cultivation is not much developed. A few vegetables are mixed with rice on tavy (Table 8 page 146) and some families create small vegetable gardens close to their house. Chinese cabbage (Brassica chinensis) is the most frequently encountered vegetable. Only a few families specialize in market oriented vegetables, usually with support from projects (Chapter 7). Market opportunities are presently too limited for a generalization of their experience. Vegetable cultivation is also rendered difficult by climatic uncertainties and by the limited investment capacity of most farmers.
3.8.6. Improved fallow Fallow land and forests also contribute to the household’s economy. Farmers improve fallow land by planting pineapples that can grow well with almost no care. This system is highly profitable despite the fact that pineapples are sold at a quite low price.
3.8.7. Livestock husbandry Livestock husbandry is very little developed but was much more important in the past. According to Messerli (2002, in Nambena 2004), the decline started in 1947, when the colonial power killed many animals to repress the nationalist insurrection. In
157
Photograph 8: Indigenous techniques for erosion control. Confronted by elevated soil losses on plowed hillsides, farmers create vertical and oblique drains to evacuate excessive water. These drains are sometimes consolidated with rocks or planted with maize or cassava. Techniques proposed by projects (contour lines), to the contrary, provoke water infiltration, which can favor landslides according to (Brand and Rakotovao 1997).
158
Photograph 9: Manual plowing in a ginger field in Beforona. This plot is on a steep slope. It is narrow in order to facilitate the evacuation of water by the vertical drains. The weeds are thrown out of the plot because otherwise, they would regrow after the first rain. This still increases soil losses.
Photograph 10: Drainage ditches planted with maize in a ginger field in Beforona. These improvements are certainly insufficient but their effect would deserve to be measured.
159
the late 1950s, the invasion of unpalatable perennial plants (Lantana camara and Rubus mollucanus) resulted in a reduction of available forage, favoring the development of diseases. The end of veterinary services in 1975, the economic crisis of the 1980s, and the conflicts between herders and agriculturalists would have led to the current situation where less than 10% of households raise cattle (Messerli 2002; Nambena 2004). Pigs are not often raised because of traditional taboos that put a ban on husbandry and consumption. These taboos tend to disappear and more and more households raise pigs (Nambena 2004), which they feed mostly with banana stems and cassava. Sheep and goats are absent. Chicken are raised by almost every household in an extensive way. They usually eat what they can find around the dwellings. A very few families, usually those that have a rice surplus, raise fowls in a more intensive manner. Rabbits recently started to be raised but are stricken by diseases. Manure is usually not used, but starts to be so in villages with higher population density such as Ambinanisahavolo (Nambena 2004).
3.8.8. Forest product extraction and land exploitation Forest extraction in Beforona is practiced in a way similar to what occurs in Madagascar in general. Operators hire the service of intermediaries that live along the road and collect boards and beams transported by hand up to the road (Photograph 11). This provides jobs to farmers living close to the forest (Photograph 12). Wood is cutup directly in the forest, by loggers hired by the operators or by farmers who specialize in this activity. Extraction occurs at a distance that varies from 20 to 50 kilometers off the road, in total disregard of permit boundaries. Beyond this distance, extraction is not profitable due to transportation costs. Palisander (Dalbergia sp.), a first category tree, has been widely extracted from the primary forest over the last ten
160
Photograph 11: Wood exploitation in Beforona. Wood is harvested mostly south from the commune of Beforona, in an area that has been reserved as a potential conservation site and which includes the Vohidrazana forest and the remaining primary forests of Ambodilaingo. (Ankeniheny site, see map 2). Exploitation permits in potential conservation sites were suspended in 2004 (see chapter 2), as well as the delivery of new permits. This exploitation is claimed to be legal because forest services provide authorization for stock evacuation, but the trees collected now were actually cut after the 2004 suspension.
Photograph 12: Wood transportation. Salary for this work is about 15000 FMg (1.5 USD) for transporting a board weighing 30 kilograms a distance of about 20-25 Km.. The income is higher than for Banana transportation for a given weight transported. Farmers from remote villages, such as Ambodilaingo, were much involved in this activity over the ten last years.
161
years. Its stock is almost depleted but other types of wood started to be harvested recently. I will describe this extraction in more detail in Chapter 5. Wild vegetables, material for the construction of houses, medicinal plants and firewood are collected in fallows and in the forest (see list of medicinal plants in Table 10 and useful woods in Figure 9). Wild plants collected in fallows represent an important part, probably the main part, of the vegetables that are consumed. Hunting is also important and contributes significantly to the diet, at least for the households that live close to the forest. Fishing is practiced in most rivers and stream. All these activities contribute to the economy of most households, but no quantification is available.
3.9. Farming systems Table 11 summarizes the performances of cropping systems in Beforona. It shows that the most profitable system, in terms of labor productivity, is the home garden, which can be explained by its perennial character and by the low labor required for its maintenance, once established. Tavy seems to be the less profitable system, certainly due to land degradation. But secondary crops associated with rice, which represent a significant value, should be considered for a more accurate measurement. Concerning irrigated rice fields, Moor (1998b) does not provide data in Beforona, but calculated that in Salampinga, located 15 kilometers east from Beforona (Map 5 page 114), 118 days of labor are necessary to produce 666 kilograms of paddy rice, resulting in a productivity of 4231 FMg (about 0.65 USD) per day of labor. The combination of these cropping and livestock systems at the household scale constitutes a farming system. Figure 10 represents a typical farming system in Beforona and Figure 6 (page 144) shows the distribution of work among the different activities. These two figures show that farmers in Beforona are far from being
162
Table 10: Main medicinal plants in the Vohidrazana watershed (See Map 5 page 114 for localization) # 1 2 3
Scientific name Panicum brevifolium Musa spp Cabucala spp
Part used Entire Leaves Leaves
Treated disease Wound Stomach ache Stomach ache
4 5
Local name Ahipody Akondro Andriambavifohy (F) Angadoha Antafanala (F)
Elephantopus scaber Terminalia sp
Leaves Bark
6 7
Befelatanana (F) Dingadingana
Vitex hirsutissima Psiadia altissima
Leaves Leaves
Stomach ache Stomach ache, Sexually transmitted disease Protection Stomach ache
8
Dipaty (F)
Bosqueia obovata
Leaves
Tiredness
9
Fatraina
Bark
Measles
10
Goavintsinahy
Erythroxylum corymbosum and/or Evodia spp Psidium catleanum
Leaves
Stomach ache
11
Goavy
Syzygium spp
Leaves
Stomach ache
12 13
Harongana (F) Malaomanitra
Harungana madagascariensis
Leaves Leaves
Wound Protection
14
Mandresy
Phyllanthus ninuroïdes
Leaves
Stomach ache
15
Menahihy(F)
Erythroxyllum corymbosum
Bark
Cold, Measles
16
Nonoka (F)
Ficus pyrifolia
Leaves
Vertigo, Tiredness
17 18
Ramitsiry Rarà (F)
Ethulia conyzoïdes Brochoneura vourii
Leaves Leaves
Wound Fever
19
Ravinala
Ravenala madagascariensis
Leaves
Bleeding
20
Ravintsompatra
Clidemia hirta
Leaves
21
Rotra (F)
Eugenia spp
Leaves
Stomach ache, Wound Stomach ache
22
Sompatra
Clidemia hirta
Leaves
23
Takoaka
Rubus mollucanus
Leaves
Stomach ache, Wound Stomach ache
24
Tanamasoandro
Tithonia diversifolia
Leaves
Stomach ache
25
Taolampotsy
Ravenala madagascariensis
Powder
Wound
26
Tsiandrova (F)
Pauridiantha lialyi
Leaves
Tiredness, Nausea
27 28
Tsilavondrivotra Tsitrotroka (F)
Phyllantus sp Dichaetanthera spp
Leaves Leaves
Stomach ache Wound
29
Vahimainty (F)
Agelaea pentagyna
Stem
Tiredness
30
Voatrotroka
Tristema verusanum
Leaves
Wound
31
Volo (bambou)
Dendrocalamus spp
Leaves
Tiredness
Source: Razafy (1999). Translation by Pollini.
163
11,72
13,28
10,16
24,22
73,44
% use
24,22 50,78
40,63 46,88
50,00
Harungana madagascariensis
Weinmannia spp
Oliganthes spp
Eucalyptus spp
Dichaetanthera spp
Bosqueis spp
Faucherea spp
Tambourissa spp
Albizzia spp
Eugenia spp
Figure 9: Main trees used for house construction in the Vohidrazana watershed, with frequency of utilization. Source: Razafy (1999, 124). Sample of 128 households.
164
Table 11: Performance of cropping systems in Beforona, on less degraded land Cropping system Tavy (rice only) Home-garden Ginger Other crops
Number of labor days per year per household 280 63 76 53
Production Production Labor value productivity (kg)
1140 424
(FMg/day of labor)
855,000 480,925 373,286 223,082
3050 7685 4890 4222
Source: Moor (1998b). One USD = about 6000 Fmg Note: In 2006, 1 USD = 10000 FMg. This rate probably doubled from the time these inquiries were done. As a gross estimate, we can consider 1 USD = 6000 FMg. But the price of most crops increased less that the USD exchange rate.
165
Ecological cycles
Economic cycles
Figure 10: Typical farming system in Beforona (in fokontany located close to the road). Source: Messerli (2003, 279). Translation by Pollini.
166
specialized in the practice of tavy, despite the fact that this land use dominates and shapes the landscape. Figure 11 shows the level and distribution of income provided by this farming system in Tanambao, a village investigated by BEMA and Terre-Tany (see Map 5 for village location). This diagram applies to most villages located close to the road (less than five kilometers) but the income provided by coffee, and maybe ginger, seem to have significantly decreased after these data were collected. Moreover, the situation is significantly different farther from the road because the main cashcrops (ginger and bananas) are no longer profitable, due to the cost of transportation. This issue will be investigated in Chapter 5.
4. Conclusion Beforona presents a diversity of cropping systems that seems to evolve in response to increasing pressure on land. Tavy is the main land use but its future is put into question by land degradation and by the limited access to forest land. Farmers living close to the road have several strategies to reduce their dependence upon tavy. They improve bottom land by creating irrigation schemes, extend paddy fields and home gardens (mostly by planting banana trees), or specialize in cash crops such as ginger. These alternatives necessitate some investment, at least in the form of labor, and the most profitable ones (ginger and banana cultivation) are dependant on access to market. Given these constraints, it remains questionable whether these intensification strategies can be applied in villages located far away from National Road 2. Chapter 5 will show that they may not be applied, meaning that the outcomes of the farming system research conducted in Beforona may not be representative of the situation of ordinary tavy farmers.
167
Total annual income: 1,138,832 FMg (about 175 USD)
Figure 11: Income structure in Tanambao, Beforona (See map 5 page 114 for location of village) Source: Moor and Rasolofomanana (1998, 45). Translation by Pollini.
168
CHAPTER 4 METHODOLOGY: COMPARATIVE AGRICULTURE
1. Introduction In this chapter, I present the methodology I used to collect data in the intervention area. This methodology has mostly been applied to the village of Ambodilaingo, presented in Chapter 5, during the six months I spent in this village. It was also used to collect the field data of Chapters 6 and 7, in a less formal manner, during short visits I made in the commune from 2002 to 2007. Section 2 will present the key methodological principles that sustain the method for data collection. Section 3 will present the key concepts of comparative agriculture, a discipline which provided me with a grid for reading realities. I will conclude (Section 4) with a brief statement about the difficulties of satisfying both the need of efficiency and the nondisciplinary ideal which this dissertation adopted.
2. Key methodological principles 2.1. Participant observation Initially, I planned to spend two years in Ambodilaingo, from July 2001 to July 2003, and to engage a participant observation approach. Local language would be learnt and everyday activities would be shared with villagers. As the initial subject was centered on the analysis of vegetation dynamics in fallow, technicians would be hired to conduct quantitative ecological measurements. Eventually, staying two years in the village was not possible due to budget and schedule constraints and the subject changed to the analysis of environmental policies. I only spent six months in Ambodilaingo, which I dedicated to data collection using the services of a translator. The first months were dedicated to ecological
169
measurements to satisfy the initial subject. In order to compensate for this limited time, it was proposed to five Malagasy students to conduct their Masters research in Ambodilaingo. They all spent several months in the village and four of them completed their degree. Their theses provided relevant information that will be used in this chapter. As I eventually stayed four years in Madagascar 125 , I continued to make short but frequent visits to the village, from 2001 to 2007. The creation of the nongovernmental organization Zanaky ny Ala (Children of the Forest), whose President was my translator, also allowed for complementary data collection as I will explain later.
2.2. Qualitative methods, narratives and comprehensiveness Qualitative methods have been privileged over quantitative ones because they were regarded as more efficient ways to achieve comprehensiveness. Quantitative methods imply making replicate measurements of the same variable in order to apply statistical procedures and obtain a satisfying level of accuracy. There is a price to pay for replication, in terms of time and budget, which leads to elimination of certain variables deemed secondary. Quantitative methods are further associated with disciplines that determine which variable to collect and which method to use. Qualitative methods, conversely, are less time consuming and can more easily let the reality drive the investigation. The “qualitative researcher” can make observations about all that is presented to his senses. He will open all black boxes because he prefers uncertainty to ignorance. He will produce a narrative that will attempt to tell
125
This includes six months in Ambodilaingo from July to December 2001; 13 months based in Antananarivo, but doing frequent visits to Ambodilaingo, from June 2002 to July 2003; two years working as a technical advisor at the Ministry of Environment, Water and Forest, and supporting the work of the NGO Zanaky ny Ala, from February 2004 to February 2006; five months based in Antananarivo and making frequent visits to Beforona and Ambodilaingo, from March to July 2006; and one month interviewing people and doing field work in Beforona and Andasibe in February 2007.
170
true stories about what is actually going on in the real world. He will extent his research object without limit, along a chain of actors and processes that goes from the local (a farmer who slashes and burns a piece of forest) to the global (international organizations concerned about the behavior of this farmer). Researchers often avoid comprehensiveness because they are afraid to lack time to address large objects. But for any research work, there is a point of departure which is the studied object, and a point of arrival which is the level of detail the researcher wants to address. The time spent to implement the research depends on the distance between these two points. Whatever is the point of departure, the point of arrival can be at an infinite distance (the structure of molecules, atoms, quarks, etc.) or can be very close (a simple look at the elements that constitute a large object). This dissertation chose to take a large object as a point of departure (a community of slashand-burn farmers in a global political, social, economic and ecological environment) and to stop at the level of detail that can be addressed given the time and resources available. This choice resulted from the statement that most previous research work dealing with deforestation and tavy in Madagascar looked at some aspects of the issue, not to the whole issue, as I explained in the introduction.
2.3. From enquiries to experience Only a few formalized enquiries have been implemented during this field work. One reason is that a large number of enquiries have already been conducted in the area. Their results provided the knowledge basis that was presented in Chapter 3 and will feed the discussion about alternative land uses that will be presented in Chapter 6. These investigations mostly concern villages located close to the road, where the situation is quite different from Ambodilaingo, but they provide references from which the situation in Ambodilaingo can be derived.
171
The second reason for few formalized enquiries is that farmers living in remote villages are reluctant to respond to enquirers who “do not bring anything good in return.” 126 They are hostile to projects and see foreigners, as well as urban educated Malagasy people, as intruders who come to “take their land” and disturb their livelihood. This fear of land-taking by foreigners is a frequent assertion. It is explained by the vicinity of the Mantadia National Park, where land had actually been taken from people as a consequence of the creation of the park (McConnell 2002). Development project staff are not welcomed because projects are perceived as associated with anti-tavy policies, which they actually are, as we will see in Chapter 6. Students who do enquiries are regarded as the first signs of future intrusion and livelihood disturbance. In this context, making quick visits to fill questionnaires would not have been a reliable way to access information. Concerning inquiries conducted in less remote places 127 , another type of problem arises. Most students that did their field work in the area collected data in a few fokontany located close to the road (less than ten kilometers). Farmers developed a discourse which they expect would fit the expectation of these enquirers. They minimize the environmental impact of their land use because they know projects are associated with antifire and anti-slash-and-burn policies; they exaggerate their poverty and hide certain resources because they expect projects will target the poorest communities first; they hesitate to criticize government policies and project activities. These biases can be mitigated, mainly by developing a closer relationship with 126
This expression has been formulated in a similar way by many farmers during enquiries. In this chapter, expressions between quotation marks are citations of farmers’ words, as I recorded them during my enquiries. These citations result from a double translation, from Malagasy to French by my translator, and from French to English by myself. I did not use a tape recorder and translation may sometimes have been imperfect. Only expressions that relate to clear and explicit facts have been quoted in this chapter, in order to minimize the risks of mistranslation. When long citations are presented, they associate answers to several questions, separated by comas. 127 Despite the fact that I centered my analysis on Ambodilaingo, I also conducted interviews in other villages, mostly to understand how rural development projects are perceived in the region.
172
villagers and gaining their confidence. But they cannot be completely avoided and are particularly important when dealing with politically sensitive subjects such as deforestation and tavy. The consequence is that drinking a cup of coffee with farmers was a more efficient data collection methodology than filling out questionnaires. This coffee drinking metaphor illustrates a methodological approach which I call “experiencing the reality.” It consists of a long immersion in the village’s life, in order to participate in villagers’ activities, instead of asking them to participate in ours (our enquiries, our focus groups, our MARP). In contrast to ethnologists’ participant observation, “experiencing the reality” implies to go beyond observation and to be more actively involved. This involvement does not mean to go hunting, to plow the land with farmers or to participate in their ceremonies. These things can be done if they happen naturally; if they are done selfishly for “us” and not artificially to please “them.” Experiencing reality simply means to play our own role, which in my case was that of an external person that wants, as a student, to understand the way communities live and, as a development professional, to support rural development. In other words, it means to be genuinely curious of who the people are, what they do and what they want, while still doing our own job, remaining aware of who we are and of what society we belong to. Experiencing realities optimizes, if not maximizes, authenticity, transparency and confidence, which may improve information reliability and relevance. The type of data I collected partly depended on the events that occurred everyday. When a series of people visited me to ask support to buy rice seedlings for their tavy, I paid attention to investment and credit issues. When an eight year old child walked far away to purchase firewood while eucalyptus grew close to the village, labor force availability and land tenure became my concern. When another person left the village with a gun and returned after only a few days, I had a starting
173
point to investigate lemur hunting. Some facts remained hidden, but at least what was apparent had more chance to be significant, whereas things which people would lie about remained unapparent. But the events of each day were not the only determinants of my enquiries. I also employed a reading grid inherited from my education in comparative agriculture, which I will present in the next section. My position with regard to questionnaires is, however, not radical. I did pass a closed questionnaire to about 400 households in Beforona and Ambodilaingo. The objective was to compare the information obtained in this way with the information obtained by “experiencing the reality,” to assess the biases of each approach and to measure a few quantitative variables such as income level and investment capacity in remote villages. The results of these inquiries have not been processed yet, due to schedule constraints, and will not be presented in this dissertation.
2.4. Validating knowledge Due to the informal methods employed for data collection and to the ideal of comprehensiveness, it was impossible to test every single data. It was, however, possible to test the overall validity of knowledge by confronting data together and verifying their consistency. Contradictions between informants occurred but the accumulation of knowledge progressively allowed me to point out mismatches between information. When such mismatches were encountered, information was exposed to more scrutiny. A consistent representation progressively built up, until a point was reached where answers to most questions could be anticipated from the informants. The attainment of this point was a first validation of the representation. It was however not sufficient. Observing and experiencing realities have an emotional dimension which can result in subjectivity and be source of significant bias, as we saw in the general
174
introduction. Quantitative methods can help to solve this problem, but the choice of what variable to measure, how to do the measurements and what statistical procedure to use can be subjective as well. The psychology of the surveyors and of the “surveyed” can introduce strong biases, especially when dealing with sensitive issues such as tavy and deforestation. Emotions are always present when two people interact in order to share knowledge and transmit information and no method can pretend to be a guarantee against subjectivity. At least qualitative methods claim less to be objective, which helps to anticipate and correct their biases. Considering these issues, and in order to provide better validation of my analysis, I decided to confront my outcomes with the reality by conducting a social experiment. I thus applied the experimental method to social sciences, as recommended by Flyvjberg (2001) and Greenwood (1998). In other words, I attempted to change social realities in the village of Ambodilaingo, based on the outcomes of my analysis. If change occurred according to my predictions, my analysis and the representation derived from it would be corroborated. They would not be proved but they would become less uncertain, which already represents progress. The nongovernmental organization Zanaky ny Ala conducted this social experiment. There is nothing new in this approach in the sense that data collection is always connected with the implementation of activities, more or less directly. In Beforona, there have always been research programs and rural development projects working hand in hand. But I contend that a procedure that consists in successively collecting data, processing them and implementing activities derived from the conclusions, each step being implemented by different persons and organizations, does not result in appropriate action. Some knowledge is transmitted at each step but not the experience of the researchers. Action is severed from observation and conversely. In contrast, the method I employed had the advantage of employing the same persons at
175
all steps of the process. In case of failure, researchers cannot be excused by a misunderstanding of their work, and development actors cannot complain about the irrelevance of research work. The responsibilities of both, who are the same person, are strongly engaged, by doing what Aristotle (1976) called phronesis (Part I).
3. A reading grid: comparative agriculture The concepts and methods that most significantly influenced my data collection have been developed by the School of Comparative Agriculture of the National Institute of Agronomy, Paris-Grignon, elaborated by Mazoyer and Roudart (2001), Dufumier (1996) and Cochet (2005). Based on this method, I used the following hypotheses to carry out my work: a. Farmers act rationally from an economic standpoint. They make the choices that answer better to their interests, given the resources that are available to them. Their interest can be the maximization of yield, the maximization of labor productivity, or the minimization of risks. If land is the limiting resource, the maximization of yield will be the main objective. If labor force is limiting, then the main objective will be the maximization of labor productivity. Risk reduction is a strategy that prevails in situations of vulnerability. b. In the case of strategies based on risk minimization, neither high yield nor high labor productivity may be the main goal. Farmers prefer to diversify their system to decrease their vulnerability, instead of specializing in the most profitable crops. They choose the systems that are less subject to market instabilities and less vulnerable to climatic catastrophes or fluctuation. The “best” farming system would be the one that produces more in the worst years, not necessarily the one that produces more in the best years or on average.
176
c. The asset that is perceived as a sign of wealth (as capital) differs according to societies. In modern societies, it is money, and production is dependent on the capacity to mobilize financial resources. In traditional societies, capital can take the form of herds (pastoral societies), fertile land (societies where access to this resource is unequal and competitive), water (if irrigation is vital and needs investments) or valuable trees. The “rich” households are those with the largest herd, the most fertile land, the largest irrigated fields or the largest tree plantations. d. The capacity of households to accumulate capital is usually linked to the capacity to reproduce fertility over years and generations. An unsustainable system is a system whose mode of fertility reconstitution is inefficient, or in crisis. The most frequent modes are: •
Vertical transfer of nutrients from subsoil to surface, by perennial plants in fallow. This mode characterizes slash-and-burn systems. The nutrients are pumped from the subsoil by the ligneous vegetation and returned to the topsoil in the form of ashes. The longer the fallow, the more biomass there is and the more available nutrients there are after burning. The succession can, however, be arrested at an early stage, which sets a limit to the amount of nutrients that can be stored (as in the case of Imperata grassland).
•
Horizontal transfer by animals. Cattle can graze in pasture land and be brought onto cultivated land or into cowsheds during the night. Manure is the medium through which nutrients are transferred from pastures to cultivated land. If the cattle graze on the same cultivated land where manure is deposited, there is no net transfer. However, this can still
177
have a positive impact by improving biomass and nutrient management. •
Horizontal transfer by flooding. A famous case is the fertilization of cultivated land in Egypt by the flooding of the Nile River, but all irrigated systems benefit from this transfer to a certain extent.
•
Application of chemical fertilizer. This is the dominant mode of fertility reproduction in Western agriculture.
e. Agrarian systems are developed by the transformation of ecosystems, and similar biophysical conditions determine similar agrarian systems. This is why we can find slash-and-burn cultivation systems in most regions of the world that are characterized by a wet climate, hilly topography, unfertile soils, low population density and abundant biomass, even if no contact or historical link exists among these systems. f. Agrarian systems are dependent on population density. Population growth provokes increasing pressure on resources. In the end, population increase leads to reaching the limits of the system, which forces the development of a new system with higher carrying capacity. This change is dependent on the adoption of new technologies and on the capacity of farmers to take risks and invest labor or financial resources to acquire these technologies. If these conditions are not filled, the system enters into a Malthusian crisis: the resources are excessively tapped to satisfy short-term needs, and the land degrades. The asset that represents capital is tapped to fill the productive gap: fallows are not allowed sufficient time to produce significant biomass, animals that produce manure are sold, or hydraulic infrastructures are not maintained. The society enters into a phase of decapitalization, until food shortage, diseases and famine occur. If investment in new technology is possible, then intensification occurs and a new asset replaces the
178
one that has been lost. Soils can be rebuilt, for example using animal manure after they have been degraded by shortened fallows. Some authors call this process environmental recovery (Tiffen and Mortimore 1994). g. The successive stages of intensification are quite similar in many places around the world. Different pathways of change are encountered according to the type of environment (great flooding plains, mountain areas, hilly landscapes), but similar ecological and economic environments determine similar pathways. For example, hunting and gathering is usually followed by shifting cultivation, followed by manual polyculture associated with livestock husbandry, followed by polyculture using draft animal traction, and then by specialized agriculture based on the use of chemical inputs, machines and fossil energy. The two first stages are referred to as extensive systems because much land is necessary to sustain a limited population (the carrying capacity is low). Extensive systems are characterized by high labor productivity and low land productivity. Less work is necessary to feed a household and more free time is available for social events. This situation is reversed when intensification occurs. However, animal traction and draft force later increase both land and labor productivity. This succession is only a general trend that should not lead to adoption of deterministic views. Variations of this general pattern are always possible. h. Social differentiation is dependent on the capacity of farming systems to produce surpluses. The more surpluses there are, the more possible it is to sustain nonfarmers such as administrators, militaries, religious leaders, etc. In the case of societies with fewer surpluses, nonagriculturalist groups can be absent but there is always a principle of authority in charge, at least, of land allocation, traditional ceremonies and conflict mediation.
179
i. Agrarian systems are constituted of farming systems usually managed at the household level. Several types of farming systems are encountered due to social differentiation and environmental heterogeneity. Farming systems are themselves constituted of cropping systems, livestock systems, plus sometimes a forestry component and hunting and gathering in primary ecosystems. j. The social relations between households are determined by their level of assets. Those unable to produce surpluses, or having a deficit, typically hire out their labor to those who can produce surpluses but lack a labor force to do so. Lending land is another typical relation between households. k. Farming systems can be characterized by a productive trajectory: young households in good health with not many children are able to produce more surplus. More children, or the necessity to feed elders not able to work results in less surplus. Aging households with a low labor force can fall into poverty unless they are supported by their children. In consequence, to have many children is a way to guarantee the future. Along their lifespan, households concentrate their effort on the accumulation of the asset that is a sign of wealth (capital) in their society (large herd, larger home garden, improved land, etc.). l. The agricultural calendar is a key determinant of the combination of cropping and livestock systems. The opportunity cost of labor is a more important criteria than daily cost (salary) to understand these combinations. All working days are not equal. A farmer can be reluctant to do a well paid job if this opportunity occurs when he is very busy, whereas he can accept a job with a low wage if he has nothing else to do. This remark is quite trivial but we will see that this aspect of realities is almost never addressed in agricultural profitability calculations. m. Cropping systems are characterized by rotations and associations of crops, the varieties that are used, a succession of technical operations, the calendar
180
of operations, the labor cost of each operation, and the tools that are used. In this chapter, I will not analyze the details of cropping systems because they have already been extensively studied in the region. n. In order to characterize a cropping system in economics terms, it is necessary to measure: •
The investments necessary to implement the system, independently of the area under cultivation. These investments are mostly the buying of tools and the building of infrastructure.
•
The cost of the inputs that have to be used, proportionally to the surface under cultivation. Labor force is not regarded as an input when it is provided by the family and is better assessed by measuring labor productivity. Labor can, however, be regarded as a service to be paid in case it is used for a specific task during a limited period.
•
The labor productivity of the system, per day of work or per laborer, and the yield. The first criteria matters more when labor force is scarce, the second when land is scarce, as we already have seen.
•
The maximum area that can be cultivated per laborer. It depends on the task that is most constraining in the agricultural calendar. For example, in some cases, sowing determines the maximum cultivable area because only a very short period is available to sow, between the arrival of the first rain and a limit date that depends on climatic factors.
•
The maximum production of the system per laborer. It is equal to the yield times the maximum area that can be cultivated by one worker. It is an indicator of labor productivity but integrates the constraints of the agricultural calendar.
181
o. Livestock Husbandry Systems are characterized by a forage system, a pasture system, the type of animal, and veterinary cares. Most of the information necessary to conduct this analysis was provided by the researchers of the BEMA,Terre Tany, and LDI projects (Chapter 3). For this reason, I did not conduct detailed investigations. For example, I did not calculate the variables listed in point n. But I attempted to test the outcomes of previous research studies, to compare them with the situation in Ambodilaingo and to fill knowledge gaps, on the basis of direct observation and discussion with farmers. I assumed that the criteria enumerated in this section are intuitively integrated by farmers in their decision making system and that qualitative approaches can even render them more accurately and finely, as soon as the reality matters more than the model that can be created out of it.
4. Conclusion I chose to focus on the collection of qualitative data and employed informal methods which I called “experiencing the reality,” and which can be regarded as a form of active participant observation. I did not reject quantitative methods but considered that the priority, given the large amount of data already available for this region, is now to produce a comprehensive representation of realities. The reality under investigation is a community of farmers who practice slash-and-burn cultivation and clear primary forest in a given economic, social, cultural and political context. It is a broad and complex object that links local and global realities. Due to this complexity, the representation I will produce will have to be tested in real world, by conducting a social experiment, as we will see in Chapter 7. In spite of this ideal of being driven by the reality itself, I also chose to use the concepts of comparative agriculture as a reading grid aimed at increasing my
182
efficiency. Farmers know their reality because they experience it from their birth. But in my case, the time available to acquire knowledge was severely limited. There is unfortunately a trade-off between objectivity and efficiency, between method and the absence of method, which, again, justifies the translation of my outcomes into a social experiment.
183
CHAPTER 5: EXPERIENCING THE REALITIES IN AMBODILAINGO
1. Introduction The fokontany of Ambodilaingo is located 16 kilometers walking south from the National Road 2 (Map 3 and 4 page 111 and 112 and Map 12). It consists of five villages (Ambodilaingo, Ambotanefitra, Ambohimiadana, Apasintsiry and Sarotriva), covers an area of 32 square kilometers and has a population of about 288 households or 1500 inhabitants. Four of these five villages (Ampasintsiry is the exception) still have primary forest on their land (Photograph 13). This forest is cleared for practicing tavy (Photographs 14 and 15) and many families live on its verge to find more fertile soils to cultivate. The density is about 38 inhabitants per square kilometer (Map 10 page 131). The people (Photographs 16 and 17) belong to the Betsimisaraka group but have some Bezanozano origin, as we will see. They live scattered in the whole territory in order to remain close to their fields. The information in this chapter is mostly the result of semi-formal inquiries conducted in Ambodilaingo (the village) between late July and late September 2001. Interviews often last several hours and sometimes included a visit to farmers’ fields. There were no sampling strategies nor interview guidelines. Each interviewee created a specific relationship, expressed particular concerns and interests and was as determinant as myself of the direction taken by the interview. The reading grid (Chapter 4) was utilized to classify the quotes and to organize the results, not to orientate the discussion. Questions were formulated in the most imprecise and simple way and answers were never suggested 128 . To give just a few examples, it was asked
128
To make no suggestion about answers is an obvious necessity, but I experienced that enquirers rarely respect this essential rule.
184
Map 12: The 5 villages of the Ambodilaingo Fokontany. Source: Institut Geographique National (1963). Based on aerial photographs taken in 1957. Additions by Pollini. The blue circles show the five villages of the fokontany.
185
Ambotanefitra
Sarotriva
Ambodilaingo
Ampasintsiriry
Ambohimiadana
Photograph 13: Satellite photography showing the 5 villages of Ambodilaingo and the forest frontier. Source: Google Earth. Year of satellite image unknown.
186
Photograph 14: Tavy on the forest frontier in Ambodilaingo. The effect of fires on the forest edge is clearly visible. Farmers purposely let fires burn the edge in order to increase their land without explicitly “clearing” the forest.
Photograph 15: Tavy inside the primary forest in Ambodilaingo. This situation, typical in areas recently colonized, is uncommon in Ambodilaingo. We can distinguish a field that has been cleared and cultivated once (right part of the photograph) and a field that has been cleared and cultivated a second time (left part). The altitude is about 800 m.
187
Photograph 16: People of Ambodilaingo in their village. Many houses have been destroyed by the Geralda cyclone in 1994 and not rebuilt since then.
Photograph 17: Children in Ambodilaingo. The picture is taken on the main square of the village, which serves for zebu sacrifice, as shown by the ritual pole with zebu skulls. Houses are built with bamboo (walls) and Imperata cylindrica thatch (roof).
188
•
Where do the people from this village come from? Why did they come here?
•
What changed in this landscape from the time of your parents?
•
What is the difference between this village and those located close to the road?
•
Why do some farmers grow rice in paddy fields and others not?
•
Did the people have more zebu in ancient times? Why?
•
What are the main causes of poverty?
•
What are the first objectives of a young man and his wife after they marry?
•
How do the people here get some money?
•
How do you see the future for your children?
•
What did you cultivate on this plot? What will you cultivate next year? Why?
After this two month period of intensive interviews, exchanges were more brief and still less informal, but villagers continued to provide relevant information. The five Malagasy Masters students who did their final training practice in Zanaky ny Ala, between 2004 and 2006, provided complementary information utilized in this chapter, when indicated. I took notes only during the two months of intensive interviews. For this reason, most quotes in this chapter are from the period between July and September 2001 or are retrieved from the thesis of the Malagasy students. Quotes may not always be an exact translation because interviews were translated from Malagasy to French by my assistant before I could take any notes. Information that was not corroborated by direct observation or by other interviews was discarded. The names of the interviewees were not recorded, in order to preserve their anonymity and to favor confidence.
189
I conducted more than 100 interviews in total in Ambodilaingo but interviewed a few households in Ambotanefitra and Ampasintsiriry in addition (Map 12 page 185 and Photograph 13 page 186). The exact number of interviews is not known because many inquiries were informal and notes were not always taken. But a group of about 25 informants, usually elders or farmers with more education than the average, provided the most detailed information and most of the long quotes reported in this chapter. This may not have caused significant bias because the community of Ambodilaingo is relatively unified. There were limited inconsistencies between the ideas expressed by the informants. Basically, all the information they provided and that has been written in my notebook is reproduced in this chapter. Only redundancies have been discarded. The results will be presented in eight sections. Section 2 will present a brief history of the fokontany, with a particular emphasis on early settlement (Section 2.1.), and on the economic and social history (Sections 2.2. and 2.3.). I will then describe the present social organization (Section 3) and land uses (Section 4). Section 5 will present the farmer strategies for intensification, while Section 6 will describe their relations with projects. Section 7 will show the high vulnerability of these communities by drawing an emotional portrait of poverty in Ambodilaingo. Section 8 will show the consequences of this vulnerability, by telling the story of the influenza epidemic that struck the village in 2002. Section 9 will synthesize the outcomes by describing the agrarian system, in the light of Boserup’s (1965) theory of agricultural development. It will allow envisioning future land uses, but more insights will be provided on this matter in Chapters 6 and 7.
190
2. Brief social history of the area The analyses of this section are limited to the epoch and events still vivid in the memory of villagers. They are presented according to three topics: settlement, economic life, and cultural changes.
2.1. Settlement According to the field investigations of Rakotoarijaona (2005), the ancestors of the villagers settled in the area several centuries ago. The original location of Ambodilaingo was on the other side of the Iaroka river, a little bit downstream. It was chosen for the abundance of fertile soils along the river. It was moved to the present location (Photograph 18) by the French administration, in 1910 or 1920, due to a change in administrative divisions. At that time, the main crops were rice cultivated in tavy, cassava, sweet potato, and bananas: In the first half or in the beginning of the century, there was no coffee and almost no paddy fields. Banana trees were cultivated around the village but were not traded. Sweet potatoes and cassava were cultivated for selfconsumption and tavy for rice cultivation was the main cropping system (an elder of Ambodilaingo). The historical leader of the community was Andrianambo, also called Ranambo, a member of a Bezanozano lineage called Zafindrianambo. Discussions with elders provide contradictory information, showing that the settlement was complex and involved both Betsimisaraka and Bezanozano lineages. Rakotoarijaona (2005) conducted bibliographical research in the National Archives. She confirmed that the Zafindrianambo was a noble Bezanozano lineage who migrated from the Ankay Region 129 in the seventeenth century, after a war between tribes. 129
Ankay is an area located southwest from Moramanga, on the eastern side of the western branch of the forest corridor (eastern side of the Angavo mountains). A mix of Merina, Betsimisaraka and Bezanozano people live there. Aubert and Bertrand (2003), Razafiarison (2003) and Aubert (2003) analyzed the agrarian systems of this region and found social dynamics similar to those that I will describe in this chapter. They showed that the Merina people, mostly recent migrants, are dominant
191
Photograph 18: Ambodilaingo. The forest escarpment is visible in the background. The large bottom land, on the left, is frequently flooded, which explains the reluctance of farmers to create paddy fields. But it could be more intensively cultivated if the land was plowed. Fallow land frequently burnt by wildfire and invaded by ferns (Pteridium aquilinum) is visible on the foreground.
192
According to Mayeur (1785, in Razafiarison 2003), Bezanozano people are the descendants of escaped Merina slaves. They became a powerful tribe in the eighteenth century and served as intermediaries for trading between Europeans and people from the highlands. They refused to submit to the Merina and King Radama I rejected them eastward, to the forests of the Angavo escarpment. They now live mostly in the Mangoro Region, from Moramanga to Ambatondrazaka, where they practice both irrigated and upland rice cultivation. Ampasintsiry was founded by a Bezanozano from another lineage who married Ranambo’s daughter. Despite the fact that it is located west from Ambodilaingo, on more degraded land, Ampasintsiry was founded more recently. This eastward move shows that forest was not the only attractive resource that determined settlement. Bottom land suitable for more intensive systems may have been coveted as well, although not for irrigated rice cultivation as this land use seems to have been nonexistent at the beginning of the twentieth century. Sarotriva, on the other hand, would have been created by families interested in wood extraction and in need of new land. According to Rakotoarijaona (2005), the site of Ambotanefitra was settled by people from Ambodilaingo who found suitable land to establish paddy fields. But the village itself was created in 1947, by a French collector of precious stones (Corindon). Mining created job opportunities which attracted people from the entire region. These migrants settled definitively after the end of mining activities. Ambohimiadana is the only village that has no common history with Ambodilaingo. It is clearly a different economically and appropriate the land most suitable for agricultural intensification. The Betsimisaraka, on the other hand, are often the poorest and their livelihood is dependent on tavy. The Bezanozano would be in an intermediate position. They enjoy a dominant position with respect to the Betsimisaraka but fear the Merina people. I visited, in 2005, the Tsiazompaniry region, located on the western side of the forest corridor (western side of the Angavo mountain), about two days walking from the Ankay. I observed that migration of Merina people eastward still occurred and that this move had been accelerated by the reforestation of the Tsiazompaniry watershed, with new villages created east from the lake. My informant described the Bezanozano as “shy people” that flee to more remote areas for fear of being in trouble when Merina people settle in their region. This provides a supplementary example of how reforestation can accelerate social dynamics that favor deforestation.
193
fokonolona. Its population is not of Bezanozano origin and is feared for its sorcery skills. It would have arrived in the region from the east or southeast. Marriage, however, occurred between members of the two villages, contributing to the blending of Bezanozano and Betsimisaraka people. This complex settlement shows that the region cannot be regarded as a simple agricultural frontier where slash-and-burn farmers (Betsimisaraka people) would migrate westward, abandoning degraded grassland and clearing forest to appropriate new land with fertile soils. The two agricultural frontiers described in Chapter 1 (the forest frontier and the “grassland” or “fallow land” frontier) may indeed be intertwined in Ambodilaingo. The forest frontier’s main assets are biomass and fertile soils, while the fallow land frontier’s assets are the streams that can be used for irrigation and the deep soils in bottom land. Bezanozano people are familiar with both irrigated rice and slash-and-burn agriculture, due to their mixed origin. As the region was less favorable to cultivation in bottom land, probably because of frequent cyclones and flooding, they ended in adopting the Betsimisaraka land use (tavy), culture, and way of life. This is why no Bezanozano people are recorded in Beforona’s census data now (Chapter 3). These outcomes challenge the received wisdom that culture is one of the most significant factors that impedes agricultural innovation and land use changes. But they are consistent with the works of Kottak (1971), Astuti (1995) and Bloch (1995), which showed, in the case of the Tanala, Vezo and Zafimaniry people, that ethnic identity can result more from adaptation to a physical environment than from traditions and cultural heritage.
194
2.2. Brief economic history Precise information about the early colonial period was difficult to collect. Most people who were adults in the first half of the twentieth century are now dead. However, the forced labor in the Andasibe camp, the difficulties in finding sufficient money to pay taxes, the touring of colonial administrators and the interdiction to do tavy are still present in memories: We had to work 30 days per year for the colonial power, and also had to work for creating improvements and to transport people of the administration. We also had to pay taxes and to grow coffee in order to get the money, or to sell other products including rice. Usually we paid in December-January. (an elder from Ambodilaingo 2001) Those who could not pay the taxes were arrested. This was told me by my parents. We had to pay a tax for each zebu head. Some people fled. (a farmer from Ambodilaingo 2001) Before the arrival of the French, people were doing tavy inside the forest and had their dwelling there. About 50 persons were living in this way. But when the French arrived, they did not dare to do that anymore and got out of the forest to cultivate the land that had already been cultivated. (an elder from Ambodilaingo 2001) The most remembered event is the 1947 upheaval and the food shortage it provoked. We did not ignite fires in order to avoid being seen by the French, because they flew over the sky with planes. 130 (an elder from Ambodilaingo 2001) My father has had the bad luck of walking outside during the day and shots have been fired on him. 131 (an elder from Ambodilaingo 2001) There was a strong food shortage in 1948. This was because of the upheaval of 1947, due to the colonial pressure. People had to flee to the forest and abandon their crops, which were damaged by wild boars. People had to harvest wild plants in order to eat. (an elder from Ambodilaingo 2001)
130
Translated from a citation by Rakotoarijaona (2005): “au risque d’être repérer par les vazaha pendant la nuit on n’allumait pas le feu, parce qu’ils survolaient le ciel, en avion au dessus de nos têtes.” Vazaha is the Malagasy term used to designated foreigners, or the “white man.” 131 Translated from a citation by Rakotoarijaona (2005): “mon père a eu le malheur de se promener en plein jour, et on lui tiré dessus.”
195
Concerning more recent times, farmers frequently recall the vivid contrast between the first and second republic. At the end of the colonial period, rice production was relatively good and the economy quite prosperous. Cash crops, mostly coffee, were sold at a satisfying price and first necessity products were cheap. Cattle seemed to have a central role in the economy, as shown by the following testimony: About half of the families had big cattle in 1959. Those who had animals had many: 15 to 30, sometimes more. Those who did not have animals loaned them for free to the others. They did not refuse because people had good relations at that time. Those who had large herds hired people to watch over animals and paid them by giving calves. Usually the herds were all put together to avoid being lost…. There was much area available for grazing….The animals were put in corrals during the night, but only during the cultivation season, in order to avoid damage to the crops….Animals had good care and were visited by veterinarians….They were used for earning money and for sacrifices. There were more sacrificed animals than sold animals. (an elder from Ambodilaingo 2001) Cattle manure was not used, but cattle nevertheless played an essential role in agricultural intensification, because zebus were used to tread paddy fields before sowing: We did not use manure. There was no taboo about that, but the land was fertile. (an elder in Ambodilaingo 2001) The first paddy fields were created in 1914-1918. This crop did not conflict with livestock husbandry. There was a time when paddy fields covered more areas than now, because people had many more cattle. The animals were used for treading of land before sowing. If we do not have cattle, we can only cultivate small areas of paddy fields. (an elder from Ambodilaingo 2001) Cattle were used to tread paddy fields before sowing. My father had 50 head. He had a paddy field because treading was easy. Paddy rice was a big activity in ancient times because people had many cattle. (an elder from Ambodilaingo 2001) The decline of cattle husbandry started in the seventies, maybe earlier, for a variety of reasons. It provoked a decline of paddy fields, because treading was no longer possible or not efficient with a small number of animals. The multiplication of conflicts between farmers more involved in livestock husbandry and those focusing on
196
cultivation, due to population increase, may have played an essential role in this decline: There has been a big decrease of cattle in 1970. The main reason is that the people started to be bad and asked penalties when cattle damaged their crops….These damages mainly concerned cassava, sweet potatoes and coffee, in the bottom land and in the lower part of the slopes. As a consequence, cattle started to lack access to pastures. People did not build fences because agriculture officers told the herders to care for their herds. In ancient times, there was not frequent damage because there were fewer crops in bottom lands and people were not so bad. (an elder from Ambodilaingo 2001) Livestock was also the main form of saving to withstand times of economic difficulties and was sacrificed during traditional ceremonies, which contributed to herd reduction. Eventually, cattle suffered much from diseases: Other reasons for the decrease in cattle are the sale because of poverty and the sacrifices….Cattle also died because of diseases: some people bought cattle from other regions, and this brought the disease. The animals get quickly thin and die [when they get this disease]. The liver is full of small animals and cannot be eaten. This disease developed around 1983 132 and still exists. (an elder from Ambodilaingo 2001) We saw in Chapter 2 that the Malagasy state entered into bankruptcy in the early eighties. In Ambodilaingo, veterinary services and support for the development of irrigation schemes stopped, which accelerated the spreading of diseases and the degradation of infrastructure by cyclones: Another reason of the decrease of paddy fields is the inundations that destroyed the dams and improvements. The ancient paddy fields became land for tavy or have been planted into banana trees. (an elder from Ambodilaingo 2001) The crisis was also aggravated by the degradation of the terms of trade. Coffee was sold at a very low price due to the nationalization and was progressively abandoned:
132
Around 1978 according to another informant.
197
Coffee cultivation developed around 1950, under the authorities of the contremaitre 133 . Coffee generated higher income at that time. One kilogram was sold for 150 F, while one kilogram of rice cost 30 F. Today, one kilogram of coffee is sold for 2500 F, and one kilogram of rice costs about 2000 F. (an elder from Ambodilaingo 2001) The population felt abandoned by the state and went back to autarky, by practicing tavy. Fear of repression had disappeared after independence: The main period concerning interdiction of clearing was after the visit of Mr. Abraham and the forest guards in 1954…. Interdiction mainly concerned the upper part of the slopes. People clearing forest at that time were sent to jail, but this did not occur anymore after the independency. (an elder from Ambodilaingo 2001). During the colonial period, it was forbidden to clear the forest. There were forest guards. It was strictly forbidden to clear the primary forest and the crests. This lasted until independence. (an elder from Ambodilaingo 2001) Chapter 2 showed that the Malagasy economy started to recover during the nineties, due to the opening to foreign investments and the financial support provided by the international community. But for the villagers of Ambodilaingo, this recovery never occurred. The period from the seventies to present is one of continuous degradation of their livelihood. As they put it, “we are always becoming poorer.”134 The renovation of the road and the opening to foreign investments and international aid led to some changes in Beforona but did not reach their village, except during a few years when the ginger price was high. Poverty is now dramatic, as I will show later.
133
Staff from the colonial agriculture services. This statement was frequently made during enquiries and contrast with the situation in the villages close to the road, where some villagers assert that their situation improved during the last decades, due to increasing opportunities to market products, mostly ginger and bananas.
134
198
2.3. Social and cultural changes The main social transformations in Ambodilaingo concern traditional ceremonies and solidarities between households. They are consequences of economic changes.
2.3.1. Evolution of traditional ceremonies Zebu sacrifice for burials (Photograph 19 and 20) was practiced from as far back as people can remember. In earlier times, the nuclear family had to provide the zebu. The reduction of herds and the economic crisis forced this rule to be changed. In 1974, the Vavanzaka and the Tangalamena of Ambodilaingo decided that the cost of sacrifices would now be born by the whole fokonolona (Rivolalao 2006). With this new system, sacrifices turned from being a burden to being an opportunity for a collective buying of meat. Due to the economic crisis, this collective buying is now an excessive burden as well. Several months can be necessary to gather the money. In some cases, several deaths are celebrated through one single sacrifice. In Ampasintsiry, the community is divided concerning the continuation or abandonment of zebu sacrifice (Rivolalao 2006). As a village elder puts it, the difficulties of the living people are already profound, due to the economic situation of the region, so it is unbearable to purchase one or several oxen for one deceased person! (an elder from Ampasintsiriry, in Rivolalao 2006) 135 Some villagers propose to organize a special ceremony to ask the ancestors to forgive them the abandonment of the cult. Others are afraid of a possible wrath. They refuse the change and do not want to donate for the ceremony, which leads to social discord (Rivolalao 2006).
135
An elder in Ampasintsiry, In Rivolalao (2006). Original citation in French (translated from Malagasy language by Rivolalao): "les malheurs des vivants sont déjà assez, vue la situation économique de la région, et c'est insupportable de fournir un ou plusieurs bœufs pour un mort.”
199
Photograph 19: Young villagers playing with a zebu before its sacrifice in Ambodilaingo. Accidents in these games are frequent but usually result in minor wounds.
Photograph 20: Meat distribution after a zebu sacrifice. Each household of the community contributes to pay for the zebu and receives a share that also depends on social rank.
200
With respect to the other celebrations (birth of the tenth child, circumcision, new harvests and access to new land), zebu sacrifices have been abandoned except for the later. In the past, the birth of the tenth child and circumcision were opportunities to call people from the neighbor villages. They are now celebrated in a more modest and intimate way in order to reduce the costs (Rivolalao 2006). A long time ago, it seems that a sacrifice was necessary to authorize any vegetation clearing (Ratsimijanona 2006). In more recent times, fewer and fewer places, usually those affected by a taboo or located close to graves, are subject to this mode of regulation. Only the richest households can contribute to the sacrifices, meaning that they became a way to reserve the most fertile land for the richest members of the community: We have to sacrifice a zebu every two years in order to do tavy on this land. There is such a rule because there were much cattle and good harvests in ancient times. But the rule is not so well respected now. We had to sacrifice every year in ancient times. Another change is that we can grow other crops than rice without doing a sacrifice while in ancient times, we had to do a sacrifice for any crop…. This rule exists in all hamlet of Beforona. All people from Ambodilaingo can do tavy on this land if they contribute to the sacrifice. Usually, five to ten families associate together for buying a zebu. It is also possible to pay the participation for another family. We were 10 people associated last year, and we also sacrificed a zebu three or four years ago. It is interesting to do so because the land is less cultivated here and the fertility is higher…. There are about five zones affected by this rule in Ambodilaingo and they are subject to regular sacrifices. They are called "Vazimba" and are close to the ancestor's graves. (an elder from Ambodilaingo 2001) These examples show that far from being static, cultural features change over time according to economic constraints. As we saw in the case of ethnic identity, which switched from Bezanozano to Betsimisaraka in certain lineages, the physical environment and the economy it can support determine culture, rather than the contrary. There can, however, be a transition period during which culture impedes adaptation to the environment, as shown in the case of Ampasintsiriry where some farmers refuse to change the rules. As long as it can be preserved, cultural identity
201
may help to decrease anxiety in the face of changes. The practices of the ancestors sustained livelihood until the present whereas new practices represent the unknown. But when ecological or economic realities render ancient practices obsolete, farmers pragmatically change their practices. In sum, culture can slow change but does not necessarily block it. As the fokontany president of Ambodilaingo put it: I am in this village since 1956. I never left. People do not respect cultural rules like they did in the past. People did not raise pigs because there was a taboo on this animal. [three households raise pigs now in Ambodilaingo]. Working was forbidden certain days, but this is usually not respected now. This lower respect of taboos is because the rules do not fit with the current realities of our life. For example, people did not go to the hospital. Now, they do. They thus have to find some money. In the past, the state helped the people buy medicine. People have to pay themselves now. (fokontany President 2001)
2.3.2. The loss of solidarity The social disruptions that resulted from the economic crisis are not only expressed in ritual changes. Several citations in the previous section evoked the idea that people “were not bad” in ancient times and helped each other more. Rivolalao (2006) showed how poverty favored individualism and the decomposition of social fabric in Ambodilaingo. Solidarity between households is expressed by a mutual support system that facilitates agricultural work. In the past, this system operated at the scale of large families and was used for most agricultural work. All participants brought some food and shared their meal together. Now, farmers expect a salary for any work they do outside their own family (Rivolalao 2006). Another sign of social fabric degradation is the increasing frequency of thieves: There are now people stealing cassava in the fields. This problem did not exist in the past. This is because some people are lazy and do not work, but
202
still have to eat. You should force the people to plant cassava. 136 (an elder from Ambodilaingo 2001) In response to this problem, farmers build a secondary house close to their field and stay there all the year round to watch over their crops. They sometimes harvest banana regimes before maturity to avoid theft. The dispersal of households is both a cause and a consequence of solidarity loss. In ancient times, farmers had their main residence in the village and often stayed there to participate in ceremonies or visit relatives. Secondary houses were built with cheap material and rebuilt every year in the location of new tavy. Now, secondary residences tend to be the main ones, resulting in decreasing socialization. The commune threatens the villagers with a five to eight USD fee 137 if they do not stay in the village at least a few months per year. When asked why they stay away, farmers first give the conventional answer: they need to watch over their fields in order to avoid pests and theft. But when conversation goes more in depth, they often respond, in a more passionate way, that “life in the village is not for the poor,” and that they would “be ashamed to show their poor condition” (a farmer from Ambodilaingo 2004, in Rivolalao 2006). Here, [close to the forest], we live alone and nobody comes to disturb us and criticize our condition. We choose to live like this because anyway, nobody in the village would come to help us to solve our problems. Now, it is every person for himself because everyone has his own problems and we do not support each other anymore. 138 (a farmer from Ambodilaingo 2004, in Rivolalao 2006)
136
This answer certainly suffers from a “postcolonial” bias. But the informant expressed this idea as a joke, and after a good glass of local rum. 137 50,000 FMg. 138 Translation by Pollini. Original citation in French: "le village n'est pas vraiment fait pour les pauvres.” "Ici (près de la forêt), nous sommes seuls et personne ne vient nous déranger et critiquer la situation économique de notre famille. Nous choisissons de vivre comme ça parce que même au village, personne ne viendra vous aider pour résoudre nos problèmes. Maintenant c'est chacun pour soi car chacun a ses propres problèmes et on ne peut plus s'entraider.”
203
To the contrary, those less in need prefer to remain isolate in order to not be under the pressure of their poorer neighbors. Farmers also evoke the loss of solidarity to explain why they do not control wild fires, whereas the lack of control is also an explanation as we will see (Photograph 21): In ancient times, when there was a fire, we called our neighbors and everybody came to stop the fire. Now, people just think of themselves and do not come. (a farmer from Ambodilaingo 2001) Farmers also evoke individualism to explain the difficulties to gather sufficient money for zebu sacrifices, to mobilize the community to repair the school and to organize the transportation of diseased people to the commune’s hospital. It is, however, mostly the young that refuse these solidarities, which leads them to criticize the elders and sometimes to enter into conflict with them. Rivolalao (2006) gives the example of a group of young people who were asked by the elders to transport a foreign visitor back to the road after she twisted her ankle. The transporters were not supposed to be paid for this service and just received a minimal indemnity. They all accepted the duty but complained during the trip: Only the dina 139 is important for them [the elders]. It is not surprising that they are so poor. They do not realize that everybody needs some money. 140 (a young farmer in Ambodilaingo 2004, in Rivolalao 2006) This last example shows that, beyond poverty, new consumption needs may also explain the development of individualism.
139
Community rule issued by the traditional authorities. Dina can also be given a legal power if they are registered at the commune. 140 Translation by Pollini. Original citation in French: “ seul le dina reste important pour les raiamandreny [elders]. Ce n'est pas étonnant qu'ils soient si pauvres. Ils ne se rendent pas compte des besoins importants en argent.”
204
Photograph 21: The impact of wildfires in Ambodilaingo. These fires can be criminal or result from negligence and the absence of firebreaks. As they are not associated with management rules, they are much more destructive than tavy. Most households agree that they should be forbidden. They almost disappeared after the repression campaign of 2002-2003.
205
3. The present social organization 3.1. Traditional and legal authorities The distribution of power in Ambodilaingo is similar to the case of Beforona (Chapter 3). The representative of the legal power (fokontany President) collaborates with traditional representatives (Tangalamena and Vavanjaka). Conflicts between traditional and legal power occur but do not seem to be significant. The fokontany president 141 summarizes community ruling in the following way: There has always been a juxtaposition of authority. There are exchanges between legal and traditional representatives. We try to eliminate the traditional rules that do not match with the legal rules…. The rules of the fokontany are elaborated by the advisors of the commune. They are the same for all 13 fokontany of the commune. They are typed on a paper and we have a copy here. These documents indicate several rules, with the price to pay. There is a stamp of the commune…. The documents do not include rules concerning tavy. We do not have other papers coming from the province or district…. In the case of violation of the law, the judgment is done by the commune. But in the case of small conflicts, we try to arrange that with the village elders first, in the presence of the whole community. The mobile guard and the fokontany president have to execute the decision…. There are also traditional rules. They concern, for example, when someone damages the house of another person. He is punished and has to buy a zebu for sacrifice. All the elders of the region are called to apply pressure on the person. This happens when people have bad relations. This rule is not written anywhere. It is in the memory of the people…. There are also national laws, such as not making tavy in the primary forest and not raising unvaccinated animals. (fokontany President 2001) This description shows the pragmatic attitude of local administrators. National laws are adapted to local realities. The ban on tavy, for example, concerns only primary forests whereas the national law authorizes tavy on nonligneous and nondense ligneous vegetation only (Chapter 2).
141
Representative of the administration, appointed by the mayor.
206
3.2. Land tenure Land in Ambodilaingo is divided into several sections occupied by different fokonolona. Each of the five villages is a distinct fokonolona and has its own land. Four of them still have primary forest on their land, Ampasintsiriry being the exception. Forest land is cleared under the authority of the Tangalamena, who decides which area to clear and allocates tracks of land to each lineage. It seems, however, that the Tangalamena’s authority is decreasing and that these rules apply only for newly cleared forest land. Concerning fallow lands, they theoretically belong to the lineage that first cleared the forest but every household can take the liberty to choose an area to crop, though this sometimes provokes conflicts. The following testimonies show that the rules are not always clear. We only work the land of our ancestors. (an elder from Ambodilaingo 2001) When someone clears a piece of land, he continues to cultivate the same land along generations. People know the limits of their land because of the messages left by their parents. (a farmer from Ambodilaingo 2001) We do not need to ask to the owner for doing tavy because we are all owners, except if we want to do it in another hamlet, but it is usually accepted. (a farmer from Ambodilaingo 2001) We cannot remember the year of clearing and cultivation for all land. We cannot locate all the land of our ancestors either, but there is no conflict on this point. (a young farmer from Ambodilaingo 2001) Now, we do not really know who cleared the land the first time, except for the land surrounding the village and for the bottom land. So, if we find a land far from the village, we usually can cultivate it. (a young farmer from Ambodilaingo 2001) A general tendency can however be drawn. Land tenure seems to evolve from a situation where the owner was the first lineage who cleared the land, to a situation where the owner is the household who actually cultivates the land. The sense of ownership hence shifts from collective (fokonolona or lineage) to individual
207
(household) and from permanent (maintained during the fallow period) to temporary (lost if the land is fallowed). This confusion of the land tenure system may be a constraint to intensification, because extensive systems are easier to implement and are privileged by farmers who want to quickly secure their land tenure: If we use a land, and if we have a problem to continue cultivation, then the other people insert into the land. This concerns the paddy fields, which are difficult to improve, but not the banana trees because they are perennial. For example, someone makes a paddy field and abandons it, and another one takes the land to make a home garden. This happens many times because we do not always remember which ancestors cleared the land. Such problems have to be solved by the elders. The story of the ancestors is fuzzy from the second or third generation. Orphans are the main victims. (a farmer from Ambodilaingo 2001) The following testimonies show that land suitable for the most profitable systems (paddy fields or home gardens) is already quite difficult to purchase, which could also increase the risks of conflicts: Only about 20 families own land in the large bottom land. Some families own land in the small bottom land. Less than 50% of the families own bottom land. (an elder from Ambodilaingo 2001) Some families have no bottom land because it was already cleared when their ancestors arrived. They have to request some land if they want to cultivate. This is usually accepted, but only for annual crops. It is impossible to make paddy fields or home garden in this way, except after buying the land. (a farmer from Ambodilaingo 2001) Some people do their home garden far from the village and have their house out of the village, for looking after the home garden. There is a risk of exodus out from the village, because people go far to find land for the home garden. (an elder from Ambodilaingo 2001) We prefer to build our house in the village because there is more communication between the people. But if we want a big home garden, we have to find the land quite far from the village. If we accept to move, it is quite easy to find such a land. (a young farmer from Ambodilaingo 2001) It is hard to find a piece of land for the home garden because the soil is hard. There is available land but it already belongs to someone. If we want to borrow, the owner refuses. Much land is in this case because people start to improve the bottom land and progressively cultivate the upper part of the
208
slope. There is no case of land renting. (a young farmer from Ambodilaingo 2001) This land shortage, however, discourages the development of systems that would maximize yields rather than labor productivity (see the hypothesis of comparative agriculture in Chapter 4). Once suitable land is found, farmers attempt to secure their ownership by extending their home garden or their paddy field over the largest possible surface, which is not compatible with intensive management. This securing of the land is regarded as essential due to the fear of land-taking by foreigners, especially if a road would be constructed: One big worry for the people of the village concerns the taking of land by foreigners in the future. The problem is that they have no money to pay for titling their land. (a farmer from Ambodilaingo 2001) If they build a road, people will come to get the best land. New people are strong and we are weak. We do not know about land titling. We have to do that if it is possible. (a farmer from Ambodilaingo 2001) No entitlement has been done until now to protect the land from outsiders. A few farmers tried but failed due to corruption of state services and to a lack of resources: In order to title the land, we have to meet the topographer, check that the land is used, go to the domain services, and check that the land has no title deed already. The titling itself is cheap but the traveling of the officers is very expensive (two persons), about 400,000 FMg 142 for two days. We can lose all this money because we do not know if it will work. People have to organize together in order to title their land. The titling of several families together is possible and these families will share their land. The titling of all Ambodilaingo is also possible. But all demanders have to be on the title. This is complicated. This is what the state wants to do, rather than letting the land to the domain. But people are not motivated to do that. They do not have the habit of doing so. They are afraid of offices. Most are not intellectuals. (a farmer from Ambodilaingo 2001) I tried to get the title for five hectares of land (1 hectare bottom land and adjacent slope). I paid 150,000 FMg 143 but I failed. I had a discount because I was a military, and that was a long time ago. It is still more expensive now. I failed because I had to find some money to make the land 142 143
About 60 USD at the time of the enquiry. About 23 USD.
209
officer come here, but I did not have this money, so I lost everything. There is no way to know how much we would have to give. (a farmer from Ambodilaingo 2001) In sum, land suitable for tavy is abundant, but land suitable for intensification (paddy fields and home gardens) is insufficient and unequally distributed. The first is usually collectively owned, while the second tends to be individually appropriated. The boundary between what is collective and what is private is not clear, resulting in conflicts that are quite rare but may increase if intensification is to occur. Households either have no land suitable for intensification, or have more land than what they can improve. If land tenure securing programs were implemented in the area in the future, they would have to consider this distinction and give priority to titles for the most coveted and individually appropriated land. Such programs could also favor intensification because tenure appropriation, which implies cultivating the largest possible tracts of land, currently tends to favor extensive systems.
4. Land uses 4.1. Cropping systems I made a brief presentation of the land uses encountered in Beforona in Chapter 3. More details are provided here in the case of Ambodilaingo.
4.1.1. Slash-and-burn systems (tavy) A farmer of Ambodilaingo provides the following description of slash-andburn systems: When we clear the primary forest, we first grow rice, and sweet potato on part of the plot when the rice is 30 cm high. Clearing the forest is the basis of our way of living. We cultivate the land a second time after 3-4 years, with rice and sweet potato, in the same way, with cassava in addition. We do not wait more because we lack area to cultivate, because of population increase. After that, we wait six years before cultivating again, because the
210
soil is not so fertile. The fallow period after this third cultivation cycle will have to be still longer…. The harvest is 20 vat 144 [400 kilograms of white rice] for one vat of seedlings [20 kilograms] at the first cultivation cycle, if the harvest is good, and five vat [100 kilograms] if it is bad. It is often bad because of rain excess. For the second cultivation cycle, we harvest 10 vat [200 kilograms] from five kopy 145 [10 kilograms] if the harvest is good, or five vat if it is bad. We have to use more seedlings: five kopy instead of four for the same area, because the soil is less fertile. The worst is if the rain falls in November because the seeds do not germinate. For the third cultivation cycle, we have to sow seven to eight kopy where we sowed four or five, because there are no trunks on the soil and because the soil is less fertile. We get eight to 15 vat…. 146 The distance between seed holes is about 25-30 cm for the first and second cycles and much closer for the third cycle. The number of seeds per hole does not change. We associate beans, maize and tsidimy 147 with the rice. The rice is sowed first. Maize grows better in the first cycle. We do not stop cultivating until the land is invaded by Tenina [Imperata cylindrica] and Rangotra [Pteridium aquilinum 148 ]. This usually occurs at the fifth cultivation cycle, but these plants can disappear if we wait about 20 years. But the Tenina can also invade from the third cultivation cycle. For the fourth cycle of cultivation, we usually wait 8-10 years because the soil is not fertile. We sow one vat and harvest eight to 20 vat…. We often move, when the soil is not fertile anymore. We also have some land in the bottom land. We have banana trees and coffee there but the cyclones made much destruction. We also lack water for making paddy fields. In regards to the aspect, it is better to grow rice on east and northeast slopes. We can also cultivate on other exposures if the season is good…. Forest guards do not come anymore from many years. But clearing of forest is not so fast because this depends on the families. It is a very hard
144
One vat is about 20 kilograms of hulled rice. One kopy is about two kilograms of hulled rice. 146 According to a questionnaire passed to the whole population of the village (quantitative survey, whose results have not been processed yet except for what follows), during the 2002-2003 season, farmers in Ambodilaingo produced 525 kilograms of paddy or 350 kilograms of white rice on average in their tavy. That year, however, was a bad year due to the antifire campaign. The price of the rice varies from about 0.25 to 0.6 USD per kilogram. It is lower during the harvest period, which is when most farmers sell it, so I will use the value of 0.3 USD per kilogram. Considering that 600 kilograms is a good harvest, I hypothesize, considering also the secondary products in tavy (beans, maize and vegetables planted in association with rice), that tavy produces no more than about 250 USD worth of agricultural products per year. This number is just a gross estimate and more precise calculation remains to be done. 147 Phaseolus lunatus. 148 See Table 9 page 147 for more information about these species. 145
211
work and people are also afraid of the law. Forest edges are also more favorable for cultivating. (a farmer from Ambodilaingo, 2001) This description is representative of the situation in the region and matches with the results of other interviews, except concerning the quantity of seed that is sown. According to other enquiries, the amount sowed ranges from about two to 15 kopy and the harvests from five to 50 vats (100 to 1000 kilograms). Harvests beyond one ton per household are exceptional. These results match with the fact that the rice shortage period lasts three to nine months for most households. The variation in the amount sowed is explained by the variety of strategies in terms of combination of cropping systems: I clear the land for cassava in August but I only can do a small area. Then I leave the vegetation drying, and I clear another land for rice, and go back to the cassava land for burning and planting. Then I burn the tavy and plant the rice. Weeding starts after one month. We do it twice and also weed the cassava. Weeding is continuous. We do not stop until rice fructification. The worst weeds are Bemimbo [Ageratum conizoides] and Fatakana [Pennisetum sp?]. We have to throw the weeds far from the plot in order to eliminate them. (a farmer from Ambodilaingo, 2001) We sow tavy first, in October-November and we sow rice in paddy fields later, in November-December-January. We can sow in paddy fields at different seasons because there is no problem of water.” “Plowing the paddy is in January, at the same time as weeding in tavy. We always lack time. We also lack time for watching over paddy fields and tavy in the same time. (a farmer from Ambodilaingo, 2001) Clearing in primary forest (Photographs 14 and 15), it is not frequent due to fear of the law, except in forest patches. This tendency was put in evidence by Nambena (2007) and is confirmed by farmers from Ambodilaingo: In ancient times, there were forest patches outside the forest. We kept forest patches to have medicinal plants because we do not use modern medicine here. These forest patches belonged to everybody. We also harvested fruits and construction material in them…. But as the people were later not allowed to clear primary forest, they cleared all the remaining forest patches. (a farmer from Ambodilaingo 2001)
212
Only one farmer in the whole community still tries to maintain a forest patch on his land. He is a traditional healer and needs to conserve this patch for collecting forest products: I do not want anybody to touch anything in this forest patch. I use it for collecting firewood and other products. The latex of this tree 149 is used for trapping small birds. I planted this Pandanus for making the roof of small houses [with the leaves]. I did not plant other trees…. This forest patch will not last a long time. My sons will fight each other to know who will burn it. This is why I asked my girls to plant eucalyptus. (a farmer from Ambodilaingo 2001) Concerning the definitive abandonment of degraded land, it seems to result from the frequent passage of uncontrolled fires (Photograph 21 page 205), rather than from tavy itself. The invasion by Imperata cylindrica and Pteridium aquilinum is the main cause of abandonment. But most farmers contend that if no fire occurs, any land can become cultivable again. The initial subject of this dissertation consisted in an analysis of vegetation dynamics in fallow. Some vegetation measurements have been done on 98 fallow plots. The data have never been processed, but I observed that almost every time I made measurements on plots dominated by Imperata cylindrica, Pteridium aquilinum, or other species typical of advanced degradation, the land had actually not been submitted to tavy itself in recent times but was subject to the frequent passage of wild fires. Farmer’s testimonies are consistent with this fact: We do not stop to cultivate until the invasion by the tenina [Imperata cylindrica] and rangotra [Pteridium aquilinum]. This usually occurs at the fifth cultivation cycle, but these plants can disappear if we wait about 20 years. But the tenina can also invade from the third cultivation cycle. It is the case in this land that we cleared nine years ago and cultivated three times. (a farmer from Ambodilaingo 2001) If we do not control fires, then the vegetation burns outside the cultivated land. This helps much the development of the tenina (a farmer from Ambodilaingo 2001).
149
Maybe Voacanga thouarsii.
213
Some plants come into the fallow from outside, after 4-5 years. They can eliminate the tenina if there is no fire. The best fallow plants are dingadingana [Psiadia altissima 150 ] and harongana [Harungana madagascariensis]. Thakok [Rubus mollucannus] and longoza [Aframomum angustifolium] are quite good also. All these plants can come from outside and invade the plot. Longoza comes by its rooting system. (a farmer from Ambodilaingo 2001) The invasion of Imperata cylindrica causes them much concern: The production of rice decreased because of soil degradation. Tenina [Imperata cylindrica] started to develop around 1967. It is necessary to wait 10 years to see it disappearing. Only six years were necessary some decades ago. Thakok [Rubus mollucannus] existed for a long time, but expanded in about the same period. (a farmer from Ambodilaingo 2001) But they have management rules aimed at controlling Imperata (tenina): There is some land where the dingadingana is mixed with tenina. This is not good for doing tavy. The dingadingana makes the tenina disappear, but not completely because the roots are still there for 3-4 years. If we put the fire, then we will have much tenina in the next fallow period. (a farmer from Ambodilaingo 2001) The tenina comes where the soil is hard. It is mixed up with dingadingana.… The dingadingana eliminates the tenina if we wait six to 10 years. (a farmer from Ambodilaingo 2001) The tenina develops after clearing and burning. It also develops better in altitude, and in dry land. It can be completely suppressed by banana trees and by sugarcane but these plants do not grow well in areas where there is much tenina. There is no solution against this weed except in the bottom lands, where we can plant bananas and sugarcane. (a farmer from Ambodilaingo 2001) They also recognize that controlling wild fires is necessary to fight against this invasion but refer to poverty, the lack of solidarity (see also Section 2.3.2.), and the lack of state control to explain the frequency of wildfire: We usually burn the land that we want to cultivate but the fire can propagate beyond the field. We have to call friends in order to control fires better. Sometimes, we also prefer to burn in the night in order to avoid accidents. The propagation depends on the day we burn rather than on the period we burn. (a farmer from Ambodilaingo 2001)
150
See Table 9 page 147 for more information about these species.
214
If we do not control fires, then the vegetation burns outside the cultivated area. This helps much the development of tenina. People do not take care and let the fire extend everywhere. This is frequent but some people take care. We respect the land because we came back to the village four years ago and we do not have our own land. We have to take care of the land that we borrow. People were more careful in the past because there was the police and the rules were strict. They were calling friends for controlling fires together. Now, everybody just takes care of himself. There has never been community rule on this point. (a farmer from Ambodilaingo 2001) Concerning the tools used for slash-and-burn cultivation, farmers use a machete, plus sometimes an axe, to clear the vegetation. Seed holes are made with a stick and weeding is done with the machete again or an old small angady. Rice is harvested with a simple knife. In other words, almost no resources other than labor and vegetation biomass are necessary to practice tavy. Anybody that has access to land (granted by traditional authorities) and owns a machete and some seedlings can practice it. In terms of labor productivity, the system is perceived as efficient if it is practiced in the forest. Most households consider that it is more profitable than lowland rice cultivation, unless a good spot is available in bottom land and can be irrigated. Weeding is not an issue on land cleared in the forest but can cause abandonment of a field on degraded land. It seems that weeds, more than low fertility, are the cause of low rice harvest on tavy. Fighting against pests, mostly Behatoka (Heteronychus plebejus Klug), rats and red fody, a small bird of the Passeridae family (Foudia madagascariensis Linnaeus) can be a significant work burden. These pests are a frequent cause of bad harvests: Rats are a big problem, as well as parrots. They are more numerous on the crests. (a farmer from Ambodilaingo 2001) Behatoka is a big problem on rice, in tavy, and in paddy fields as well. There are also many rats. (a farmer from Ambodilaingo 2001) Another essential constraint to tavy is the repression of this practice by the government. Tevy ala (primary forest clearing) is perceived as being forbidden by the
215
government but was still practiced in 2001, due to the absence of control (Section 2.2.). The legislation actually puts some restrictions on the practice of tavy in fallow land (Chapter 2) but we saw that these restrictions are usually not enforced. In sum, until 2001, farmers in Ambodilaingo did not worry about state control when they did tavy on fallow land, but they expressed a certain fear of state control in case of forest clearing: It is forbidden to clear the forest. The state is very strict. It seems that there is something in the forest that is useful for the state. We do not really know what would be the punishment but most people are afraid. Maybe we could go to jail. People cultivate bad soils despite the fact that they have bad harvest. They move into fallow land and put their production in the hands of god. When we have a bad harvest uphill, we cultivate the bottom lands. (a farmer from Ambodilaingo 2001) This situation changed in 2002, when the government launched the antifire campaign (Chapter 3). The Presidential speech was heard on the radio and information about farmers that had been arrested for practicing tavy quickly reached the village. Some farmers did not dare to burn the vegetation they had already cleared and did not sow rice. Others moved deep into the forest to hide from state control or came back to the village to do tavy on degraded lands (Ratsimijanona 2006). Others waited to see what would happen and delayed sowing, resulting in a low harvest. Those who had no or a bad harvest employed the same strategies they use in case of cyclonic damage. They switched their diet earlier to cassava, harvested food in the forest, and searched for income opportunities (Ratsimijanona 2006). In Ambodilaingo, a decrease in school attendance was observed because children were asked to contribute more to support their family (Ratsimijanona 2006). Parents send them to Beforona to sell bananas. The 2002-2003 season may indeed have been the less appropriate for developing more intensive land uses. It is remembered as a period of deep crisis by most people. Forest activities, mostly the transportation of wood from harvesting sites to the road, were the main source of income that helped people to survive. Fortunately, the campaign was
216
not repeated the next year and the situation came back to normal, despite the fact that provincial authorities issued a decree that forbid agricultural fires (Chapter 3). Certainly this campaign had a positive impact on the environment (it clearly resulted in less primary forest clearing and better fire control), but it was unethical due to the deep poverty of most households in this region. Moreover, the same results could have been obtained targeting only tevy ala (primary forest clearing) and wildfires. As a consequence of this history and these practices, tavy in Ambodilaingo encroaches relatively slowly on the forest. The limit between agricultural land and forest land follows a straight line. Not many tavy are encountered inside the primary forest, almost no forest patch remains in the agricultural land and secondary forests are nonexistent. Long fallows are never practiced except in case of arrested succession or land invaded by Imperata cylindrica and Pteridium aquilinum. But state control does not seem to be the only explanation. It seems that rice reaches it ecological limit on the forest edge, at an altitude of about 800 meters. Big rains result in high risk and cultivation on degraded land located at lower altitude is sometimes preferred for this reason. In other regions of the world, farmers move over a vast territory and allow a complete forest recovery when they abandon their village for a better location. Such systems are qualified as shifting or swidden agriculture. They cannot be compared to the situation in Ambodilaingo. Here, villages never move once they are established. When pressure on the land is too elevated and fertility degrades, some households can leave the village and create a new dwelling closer to the forest. Those who move are usually young households in the search for new land to establish themselves. In other cases, villages split due to internal conflicts and a complete lineage leaves.
217
But at present, even this move toward forest land does not seem to occur. Young households that leave the village rather search for locations suitable for establishing a home garden, as we saw in the section about land tenure. Tavy is not the only land use and may even be on the verge of not being the main land use.
4.1.2. Paddy fields We saw that irrigated rice cultivation has been in decline since the seventies due to the quasi-disappearance of cattle. This land use is, however, still practiced, usually by the few families that own zebu. In general, paddy fields are created where water can be derived from a stream with a minimal effort, for the benefit of a single or a few families. The fields are usually in the form of small terraces as the land is rarely completely flat. Larger collective improvements are rare, unless support is provided by the government or by projects for the construction of cement dams (Chapter 7). Areas suitable for irrigated rice cultivation are limited and are unequally divided between households. Rice can be cultivated twice a year if there is efficient irrigation. The counterseason cycle provides lower yield due to lower temperatures and a higher susceptibility to pests. The techniques are usually traditional, but direct sowing is quite rare. Rice is sown in nurseries and transplanted after four to six weeks. The plot is prepared with an angady or a hoe, or by treading using zebu for the few families that still raise cattle. The yield usually ranges from one to two tons per hectare. After a few years, it can decrease significantly, depending on the location of the plot. When yield reaches a bottom line, the plot is abandoned into fallow for a few years. Families that have sufficiently large plots cultivate just half or less and switch from one part to the other over the years. Manure is not used, even by households that own cattle. Techniques recently popularized by projects, such as sowing in line, transplanting at
218
an earlier stage or using fewer seedlings per seed hole are encountered in some locations. But the use of the whole SRA 151 package is quite rare, and SRI 152 is totally absent 153 . Here is a farmer’s description of a typical paddy field in Ambodilaingo. I have four zebus for treading but this is not enough. I have to work the land with the angady. If we do not tread, we do not have a good harvest. We prepare the land for the nursery using an angady. To prepare the land with an angady takes much time. After we sow, we have to watch over the field for 15 days because of the Fody birds…. We cultivate this paddy field from 13 years. We were doing tavy before, on a land that we now want to sell. All this bottom land is ours and we cultivate it every year, on the same plots. We usually sow one vat [20 kilograms]. There are variations in the yield because of the cyclones. The knowledge for doing this comes from our ancestors…. People do not use the plow because there are not many zebus in this area…. We do not have to weed if we sow in October, because there is much water. We do not extend our paddy field because part of the land lacks water and cannot be cultivated properly. (a farmer from Ambodilaingo 2001) This description shows that the main issue for the extension of paddy fields is the availability of animals to facilitate labor, for treading and plowing the land. More and more farmers attempt to work the land with an angady and to cultivate paddy fields without cattle treading. But many are discouraged by the fact that bottom land is covered by vegetation that is dominated by fast growing perennials, mostly Rubus mollucanus, Lantana camara and Aframomum angustifolium 154 , which are very difficult to eliminate with a simple angady: Not everybody has land but everybody can ask for land for cultivating rice in the bottom land. This is for free. We do not ask for this land because it is much work to eliminate the weeds. One needs to have money and food to invest in cultivation in bottom land. Rice in bottom land is cultivated like in tavy if we cannot irrigate, but there are many weeds and if we do not do a good weeding, then we lose. Nobody has worked this land with a plow. We just use the angady and the machete. (a farmer from Ambodilaingo 2001) 151
Improved Rice Cultivation System. In French: Système de Riziculture Améliorée. See Chapter 6. Intensive Rice Cultivation System. In French: Système de Riziculture Intensive. See Chapter 6. 153 I will explain these systems in more detail in the paragraph dedicated to the work of rural development projects. 154 See Table 9 page 147 for more information about these species. 152
219
Once the land is cleared, it is also necessary to dig irrigation canals, which represents still more work and is constrained by a shortage of food: People tried to dig a canal. But they lacked food and did not finish in time. There is a problem of lack of zebu, but irrigation is very important also. (a farmer from Ambodilaingo 2001) In spite of this, there seems to have been an increase in land improvements in recent years. Those who cultivate rice in paddy fields still practice tavy and the two systems complement each other. I grow paddy rice in the bottom land. I plow by hand, but I have enough water in this plot. Last year, I sowed 50 kapoks [10 kilograms] in my paddy field and 100-200 [20-40 kilograms] in my tavy. Paddy necessitates more work, mainly for plowing because we do not have a plow. I have been doing that for 10 years but I can only grow a small area and I can see that it is very hard and complex to do that. The yield is decreasing because we cannot provide all the care. We cultivate the same land every year. Weeding on paddy is more difficult because we have to remove the weeds to avoid letting them regrow in the water. (a farmer from Ambodilaingo 2001) This farmer’s statement further reveals that paddy fields are usually regarded as more constraining in terms of labor. Buying a plow, which costs about 40 USD, is often mentioned as a way to increase labor productivity, giving more evidence that the lack of labor force is a more severe constraint than the lack of land to develop irrigated rice cultivation: There has never been any plow here in the past, and now it is the same. I have a project to buy one for my kids but life is difficult. (a farmer from Ambodilaingo 2001) I know that you just do research, but I insist that people here need some means. There is much good land in the bottom land and the water does not come until here. But we need a plow in order to cultivate this land. People here lack means and tools for working. (an elder from Ambodilaingo 2001) We already thought about working with the plow but it is impossible for one household to do the investment. We could try to organize to buy one with four or five households together. If one person had this equipment, maybe everybody would use it. Those cultivating rice in paddy fields, at least, could use it. But people have to be sure that they can successfully use it. (a farmer from Ambodilaingo 2001)
220
In this last testimony, the interviewee lists what could be the three main constraints to intensification: the lack of capacity to invest, the lack of solidarity, and the risk factor.
4.1.3. Home gardens Besides tavy and irrigated rice fields, farmers in Ambodilaingo have home gardens where they grow cash crops, mostly bananas and coffee. These crops were marginal until the mid twentieth century. Coffee was an essential cash crop during the 1950s and 1960s, but has declined since the 1970s due to low price. Bananas mostly developed after 1990 because the rehabilitation of the road in the late eighties increased market opportunities. Creating a new home garden is the first objective of a newly established household. Farmers sometimes go far away in the search for suitable places for creating their home garden, as we saw in the paragraph about land tenure. Other constraints are the important work necessary to establishing home gardens and transportation cost, which confirms Nambena’s (2007) outcomes: It is easy to plant 1000 banana trees in one year if we want and if we have many friends. We get the seedlings from friends. But we do not plant that much because we have many other activities. For instance, now is the time for weeding the home garden, but we also have to clear the tavy. Most people just plant 10 to 20 banana trees. But if we specialize in this crop, we can plant more. If there would be a road, sure, we would plant more banana trees because we could sell. (a young farmer from Ambodilaingo 2001) However, home garden maintenance has a low cost. It consists of one or two weedings per year and can be done at any time, whereas some periods are more appropriate. There is no special time to weed banana trees. We do that when we can, little by little. (a farmer from Ambodilaingo 2001)
221
Bananas also have the advantage of being harvested all year round, providing a limited but permanent income. Unfortunately the sale price is low: about 0.5 USD per kilogram. Collectors know that farmers will not return to their village with a 30 kilogram load. They impose a very low price and it happens that farmers prefer to dispose of bananas rather than accept this price. Bananas are also consumed, mainly during the rice shortage period. Coffee played a significant role in the economy a few decades ago, as we saw. At present, coffee plantations are almost abandoned due to low and unstable prices and coffee is harvested mostly for self-consumption. The other products found in home gardens (e.g., avocado, jackfruits, grapefruits, oranges, limes, and sugar cane) are locally consumed and often go to waste.
4.1.4. Other cropping systems Ginger used to be cultivated in Ambodilaingo as a cash crop until a few years ago, using techniques similar to those employed in Beforona. It has now almost disappeared because “farmers along the road produce big quantities and the price went down” (a farmer from Ambodilaingo 2001). Due to transportation costs, ginger is not profitable anymore in Ambodilaingo 155 . It has a comparative disadvantage due to the remoteness of this fokontany. The elevated cost of seedlings is also a significant constraint. Beans and vegetables are grown on plowed land, on flat land, on river banks or down the slopes on the most fertile soils. They are not frequent due to the difficulty in purchasing seeds, the problem of transportation (in the case of vegetables) and climatic risks. The profitability of cropping systems based on plowing and crop 155
In 2006, Ginger was sold as low as 750 FMg/kilogram in Beforona (0.07 USD), while transportation cost 300 FMg/kilogram (0.03 USD). Ginger can however be sold a more elevated price if its harvest can be delayed until the price gets higher. But poverty leads to selling it as soon as it is mature.
222
rotation, in comparison with tavy, is disputable. The labor cost of plowing discourages most households due to the difficulties of eradicating the root systems of invasive perennial herbs and shrubs. Plowing is, however, practiced on small plots and may become more frequent in the future, as we saw in Chapter 3. Chapters 6 and 7 will assess this option in depth. The LDI project 156 attempted to develop crop cultivation in collaboration with village teachers: I cultivate beans in the bottom land, on land previously cultivated two years in rice. Last year, I sowed 10 kapoks and harvested 100 kilograms. This year, I sowed the 10 kapoks again and harvested 120 kilograms. The harvest increased because the roots decompose in the soil…. The land has been slashed and burnt for rice before growing the beans. It was a ThakokLongoza-radriaka 157 fallow. There was no tenina…. The rice was sowed like in tavy. But for the beans, I plowed the land to a 20 cm depth, using an angady…. I am the only one doing that. People did not try because plowing is difficult. The beans are for self-consumption…. This has been done with the support of LDI. The seedlings come from the agriculture center of Moramanga. The difference with the traditional system is that the land is plowed, instead of just taking out the upper part of land with all superficial roots. This asks four times more work but plowing gets easier every year. I had a training in Andasibe for making compost and I will start soon to do that. (the teacher of Ambodilaingo 2001) The main constraint for this system seems to be the labor cost of plowing, as in the case of paddy fields. But plowing gets less difficult when it is repeated successive years, because the root systems of perennials are progressively eliminated. The extension of permanently cultivated plowed land can be seen as a long-term investment that will need years to occur, unless it is encouraged by outside supports. The main constraint may be this initial investment in terms of labor, rather than the profitability of the system. Chapter 7 will show how this constraint can be lifted.
156
Landscape Development Interventions, financed by USAID (Chapter 6). Rubus mollucannus, Aframomum angustifolium, and Lantana camara. See Table 9 page 147 for more information about these species.
157
223
Pineapple is cultivated by a few farmers in Ambodilaingo, on fallow land like in Beforona and with minimal care. This system can be seen as an improved fallow having high labor productivity (Chapter 3). Only one household grows pineapple at a significant scale, however, due to the lack of opportunities to sell the fruits and the elevated transportation cost. The fruits are used for self-consumption or to feed the pigs.
4.2. Livestock systems Livestock husbandry is not much developed. Pigs are not consumed because of taboos but are raised by a few households. Fowls are raised by almost all families in an extensive manner. For most farmers, the lack of agricultural surplus is the main constraint to raising these animals: We do not have means to raise pigs. We must have food to give to them, like cassava or maize. (a farmer from Ambodilaingo 2001) Pig raising, however, seems to be paid increasing attention by farmers: People started to raise pigs eight years ago. There was a taboo on this animal before. Four people raise pigs in the village now. (the fokontany President of Ambodilaingo 2001) Cattle were abundant until the 1960s but declined for the reasons I explained earlier. Farmers who have cattle either managed to find sufficient cash to buy the first animal, or inherited them from their ancestors. Here are two descriptions of livestock husbandry by two of the few farmers who own cattle. Both are better-off farmers. The first one receives income from a small trade while the second has one of the largest irrigated rice fields in the village: We put the zebu on the paddy field after the harvest. We also use them for treading. We have a corral for the night. During the cropping period, we bring the animals in the forest. The main problem is that we have to look after the animals all the time, because there are crops everywhere. Our small boy can do that…. We had zebu from our ancestors but they died. We bought these ones…. Dingta is not the only disease. There is another
224
disease: when you put the flesh of the dead animal in a pan, it gives some water. We can eat the meat but it needs much preparation…. We bought the animals in 1994 because they were cheap, due to the Geralda cyclone. We get some money from our small trade…. We do not use manure but there is no taboo for that. We never heard that we could use manure. The corral is always in the same place…. We do not give forage to the cattle in the corral…. Our male is dead last year from the disease we described. We have to ask other herders in order to fertilize our female. We now have the mother and two young. We do not sell animals except in case of absolute necessity. (A farmer from Ambodilaingo 2001) I have these four zebus because I bought them six years ago, from the money of rice. I was trading zebu before. I bought them because the price was very low after the Geralda cyclone. I had two births. One is dead of disease and one fell into a hole in the forest…. I feed my animals in the bottom land and also bring them inside the forest. There is no special period for this but if there is no grass in the bottom land, I bring them inside the forest…. We always have to look after the animals. I join with my son for doing so…. The difficult period for feeding the animals is when there is much rain or a cyclone, because it is difficult to look after them, or when we have health problems…. We do not join forces with other families because not many people have zebu…. The animals spend the night in a corral. We have one in the forest and one in the bottom land…. There is no taboo concerning the use of manure, but we do not use it because we are not used to it. We prefer to rotate the corral and cultivate on it, but this is not always done…. Nobody complained about our animals because we always look after them, me or my sons. This is difficult but we have no choice. We tie them up when we leave for lunch…. My zebu were vaccinated. The veterinary came and we just had to pay the vaccine. But the veterinary did not come this year. (A farmer from Ambodilaingo 2001) These descriptions show that cattle husbandry necessitates investment and is risky due to the frequency of disease and the difficulties in finding veterinarians, and to “accidents” of cattle that fall in the steep terrain of the forest. They also show that potential damage to crops results in an important labor burden to look after the animals, and that the use of manure is only embryonic. In sum, small animal husbandry, especially pig husbandry, may be more suitable to local conditions than cattle husbandry in the short term. But the situation could evolve if the problem of conflict with agriculture was solved, for instance, by enclosing permanently cultivated land, and if the vulnerability of livestock was
225
decreased, by increasing veterinary care. The lack of capacity to invest, however, will remain an essential constraint to the development of cattle husbandry.
4.3. Forest products extraction The forests of Ambodilaingo have not yet been depleted of their wood resources. Palisander has been harvested for about ten years. The resource may be on the verge of depletion and second category woods are starting to be cut. Harvesting is controlled by a few operators. Some received a legal permit, but cut beyond the authorized period and area. They employ intermediaries to collect the wood on the road (Chapter 3). Wood is cut inside the forest and carried to the road in the form of boards. This provides significant job opportunities to villagers. Some villagers also search and cut trees themselves in order to receive a larger part of the benefit. It seems that between ten and twenty households specialized in this business in Ambodilaingo. In 2003, a road was constructed by an operator to transport wood from Ambodilaingo to Beforona. Trucks transported a few loads until the extraction stopped, certainly because of the permit suspension that occurred in 2004 (Chapter 2). The road is now impassible but exploitation recently started again. Staff from the forest service offices sometimes comes to visit. They deliver documents that allow wood to circulate in apparently legal conditions. For example, they authorize the circulation of ancient stocks but the documents are used for transporting newly cut trees. According to informants from Beforona, people with high political connections protect this business. An operator recently attempted to utilize the GELOSE legislation to allow legal extraction by local communities. He prepared all the legal documents and made a few promises to convince villagers to collaborate with him. The villagers of Ambodilaingo were going to sign the documents to form a COBA (Chapter 11) that
226
could be granted extraction permits. But the mayor, aware of the manipulation, opposed it. In Ambodilaingo, most men of working age transport boards from the forest to Beforona. They earn the equivalent of about 5 USD a week transporting two or three boards that weigh about 30 kilograms each. This represents three to six days of work but most workers need to rest one day after each transport. In spite of this, the job is very attractive because it provides immediate cash that serves to buy basic products like rice, salt, kerosene for lamps and rum as a reward after the effort. Some young people manage to do important saving they use as a dowry to marry. People over 55 years old have been seen engaged in transporting boards. Some men specialized in this activity and almost abandoned agricultural work. They are criticized by other farmers who remind them that forest product extraction can stop any day and that people “should better plant cassava if they do not want to starve some day” (an elder from Ambodilaingo 2001). A rigorous calculation remains to be done but it may be that agriculture production actually pays less than working as a board transporter. In sum, forest exploitation, and especially board transportation, plays a very significant role in Ambodilaingo’s economy. But this activity is perceived in a very different manner by the elders and the young. Elders are often against it: Young people rush into this activity, but it has no future. People should plant banana trees, coffee, and grow rice. (an elder from Ambodilaingo 2001) It is very hard to do this job. People do that because of poverty. Young people should better invest in agriculture. The money they earn is only used for eating. There is nothing remaining, except for drinking rum. They always think that they will earn money again the following day…. They get so tired when coming back that they cannot work the land. They usually have a family and a young child, and some are still older. Some families are dependant on this income and forget to grow rice. (an elder from Ambodilaingo 2001)
227
Young men are more enthusiastic because board transportation represents the fastest way to get cash: To get a wife, one needs about 500 000 FMg [about 75 USD]. One has to cultivate and earn money two to three years to get this money. Ginger is the crop that allows earning more money, but it necessitates much work. Sometimes, 10 years [of saving] is still not enough. Some people carrying boards can get the money in a few months. Most young men collect the money in this manner. We can pay part of the sum to ask for the woman but if we do not finish paying, we can lose the wife and the money, and the kids if we already have kids. This money is for the wife’s parents. (a young farmer from Ambodilaingo 2001) A large number of other products are harvested in the forest. They are mostly materials for house construction, wood to produce furniture or other items (such as mortar and pestles), medicinal plants, orchids to be sold in Andasibe (only a few households do this), and food such as meat from hunted animals and forest yams. The later seems to play a significant role in the diet during the food shortage period. Besides the harvesting of forest products, reforestation exists but is limited to a few small eucalyptus plantations. The first eucalyptus were planted in 1957. We prefer forest patches but there were not anymore, so we planted the eucalyptus. Some plantations belong to the village and others belong to individual owners. In this case, we must ask their authorization in order to harvest firewood or construction wood. But we prefer to harvest firewood in fallow land…. In this village, we do not produce charcoal. (a farmer from Ambodilaingo 2001) In 1964 and 1972, planting eucalyptus was the condition for getting authorization for doing tavy. Eucalyptus were planted on land not suitable for rice because they grew well on this land anyway. (a farmer from Ambodilaingo 2001) Eucalyptus tends to be invasive. Farmers sometimes cut them to do tavy. They argue that they lack land to do tavy and that this clear-cutting allows dense eucalyptus woodland to grow after the crops has been harvested.
228
5. The strategies for intensification We saw that agriculture in Ambodilaingo was relatively prosperous until the seventies. Livestock husbandry was more developed and paddy fields and coffee cultivation complemented tavy to sustain livelihood. The situation degraded after the seventies due to the decapitalization of livestock, the abandonment of paddy fields, the degradation of terms of trade, the degradation of land, and the ban on forest clearing. At present, slash-and-burn systems are in a situation of crisis because they face their ecological and political limits. For these reasons, farmers in Ambodilaingo perceive that tavy is not the crop for the future. In contrast to what their ancestors did, they search for new economic opportunities: In ancient times, rich people did not make cash crops but extended their tavy cultivation. (a farmer from Ambodilaingo 2001) Now, rich people invest their money to pay hired people to extend their home garden, their paddy fields, or for coffee or cassava cultivation. (a farmer from Ambodilaingo 2001) If I had money, I would invest in agriculture and in raising pigs. A 15 kilograms female costs about 200,000 FMg [about 30 USD]. We cannot buy a younger one because there is a big risk that it will die, because there are many diseases. We can sell the pig for 400,000 in April, after it gives babies. (a farmer from Ambodilaingo 2001) If we had money, we would invest in ginger and rice. (a group of farmers from Ambodilaingo 2001) But they lack investment capacity and market opportunities to adopt these alternatives: There is much land to cultivate but the problem is to find money. The area is favorable to ginger and pineapple but the problem is to sell the products. There are big price variations. (a farmer from Ambodilaingo 2001) The cultivation and market for ginger started to develop in 1987. Ginger was sold for 3000 FMg/kilograms five years ago [about 0.45 USD]. Now, it is sold for 750 FMg [about 0.12 USD]. (a farmer from Ambodilaingo 2001)
229
We grow pineapples that are very easy to sell in Beforona because they grow big here. But the transport cost is 300 FMg per kilograms 158 [0.45 USD]. We also have problems with rats but we can eat pineapple. It is good for the children. (a farmer from Ambodilaingo 2001) The construction of a road is often evoked as a way to solve these problems: The main problem is the transport to Beforona. We could grow coffee, banana, rice, cassava, sugarcane but it is difficult to sell the products. People would be very happy if there was a road, because they could sell easily. Now we go to Beforona once a week to sell our products and buy something. (a farmer from Ambodilaingo 2001) I am here for one year, and I think that the number one problem is getting products out. Most people would agree to the creation of a road. Only a few would be against…. The priest had a project to create a road, but the commune did not agree. I do not know the reason why they disagreed. I also heard that a forest operator had a project to build a road for carrying the wood. 159 (the catechist of Ambodilaingo, recent migrant 2001) But the road is also perceived as a danger because it could favor the appropriation of land by immigrants: The problem is also that people coming from outside can take the land. The population living along the road does not encourage its construction because they may have to move. They nevertheless need a road to get their products out. The solution? We have to take the risk and to be smart. We have to title the land. (the catechist of Ambodilaingo, recent migrant 2001) The problem is that the road also brings new people who seek for new land. We will have to cultivate land in the forest to have something when there will be land tenure problems. We do not title the land because we do not know about that. (a farmer from Ambodilaingo 2001) Beyond the lack of investment capacity and market opportunities, cash crops and other alternatives to tavy are also constrained by climatic risks: The main problem in Ambodilaingo is the climate. Harvest can be good but there are cyclones and hail storms. The hail gives cold to the rice’s roots. This happens almost all years. There are two kinds of hail storms. The small ones that do not do much damage and the big ones that cool the rice roots. (a farmer from Ambodilaingo 2001)
158
In 2006, pineapple were sold for about 500FMg per piece in Beforona (0.075 USD), and the transportation cost was about the same. 159 This road was eventually constructed in 2003-2004, but was impassible after a few months, due to big rains and the stopping of forest exploitation.
230
When there is a lack of rain, the production is bad. But it is the same when the rain is too big. We put our production in the hands of god. (a farmer from Ambodilaingo 2001) There is also a big risk of inundation in the bottom land when rice is flowering. These inundations destroy the improvements in the bottom land, so it is dangerous to invest by paying people to do these improvements. The slopes are not affected by the cyclones. (a farmer from Ambodilaingo 2001) Fields and houses have been destroyed [by the Geralda cyclone, in 1994]. Banana trees, cassava and almost all crops have been destroyed by inundations on bottom land and wind on slopes. People left the village and did not come back because they had to find food. (a farmer from Ambodilaingo 2001) In the end, even energy purchase is insufficient to work and improve the land, as we already saw: People now have fewer paddy fields, fewer zebu and less food. They thus have less strength for working their land. (a farmer from Ambodilaingo 2001) In sum, several pathways for intensification already exist (paddy fields, home gardens, cash crop cultivation such as ginger and livestock husbandry) but are constrained by the lack of investment capacity, by vulnerability and by insufficient access to markets.
6. Relations with Projects Facing such difficulties, one could expect that farmers would ask for support from projects. We saw that projects operate in Beforona from many years. People from Ambodilaingo know about that. Market days, every Thursday in Beforona, provide opportunities to exchange information with people from other villages. But in the end, projects are perceived as being associated with anti-tavy policies and are not welcomed in the area, as shown by the following declarations: I know the people from here. People do not dare to tell you that but I want to be sincere. People in Ambodilaingo are afraid of LDI because they have little means to cultivate the bottom land and only do tavy. People are afraid
231
of LDI because LDI impedes doing tavy and cutting the forest. People here also observe that LDI does not really have a solution to replace tavy. They have some solutions, but it is impossible to do what they propose. LDI also tries to convince the foreign people that they have some solutions. (a farmer from Ambodilaingo, among the most educated 2001) Chapter 6 will show that this fear is rational. My own presence also created fear in the village, at least at the beginning: Maybe you did not follow that in the meeting, but there was a big discussion when you first came. The elders did not want you to come but the young people insisted for the village to receive you. People are not confident with LDI because LDI does not support them well for the new techniques. (a farmer from Ambodilaingo, same as for the previous citation 2001) When you came for the first time with [the representative of the commune], I proposed that we wait for the agreement of the elders from the five hamlets of the fokontany, before giving approval. I have some difficulties feeding my 11 kids and I resisted at the time of your arrival, but the person who was accompanying you insisted on getting my immediate acceptance, without agreement from the whole fokontany. (an elder of Ambodilaingo 2001) The main concern is, of course, the attitude of visitors with regard to tavy: When one forbids tavy, one kills people. (a farmer of Ambodilaingo 2001) If you do not come for forbidding tavy, then we are safe. (a farmer of Ambodilaingo 2001)
7. An emotional portrait of the local economy In this section, I will describe the poverty situation in Ambodilaingo in the way it has been felt during the six months I spent in the village. I believe this is necessary because the most talkative farmers, those who provided me with the longest and more informative citations in the previous section, are often the better off. The poorest often remain shy. They hesitate to receive visitors because they are ashamed of having nothing to offer. There is also much information that is collected by direct observation and that does not require asking questions, such as the belongings people have and the materials they use to build houses. This section will thus complement the previous
232
one. It concerns a significant part of the population of Ambodilaingo, certainly a large majority; however, no precise measurement as been done of poverty level and distribution. The first sign of poverty in Ambodilaingo is that almost all new houses are built with the same local materials collected in forest and fallow lands. Roofs are made of thatch and walls and floor are made of bamboo or other cheap materials. Only ancient houses are made of boards and have a corrugated iron roof. The Geralda cyclone destroyed most of these houses in 1994 and they have not been rebuilt since that time (Photograph 22). Nobody now seems to have enough money to buy corrugated iron. Every day’s life is about the same for everybody. It is marked by a food shortage period (indeed, a rice shortage period) which lasts three to nine months. During this period, households attempt by all means possible, mostly by transporting boards or bananas for sale, to find money to buy rice. If they do not manage to purchase rice, they eat mostly cassava, which has a lower nutritious value, but also bananas, yams and sweet potatoes. For the households that have no revenue (for example, widowed women or elders who cannot transport palisander boards or banana racemes), the temptation is great to eat the seeds kept for the next agricultural season. Meat is eaten during a few weeks of the months following oxen sacrifice, which occurs about twice a year. Households living close to the forest also hunt and eat game meat. During the rice shortage period people are quite depressed. They complain and dispute more with their neighbors and almost beg support from foreigners (myself and the Malagasy students I worked with). During the harvest period the atmosphere changes. Foreigners are more frequently invited for lunch or dinners (it can happen that we had five or more meals a day). Farmers are not ashamed of their condition
233
Photograph 22: The main street of Ambodilaingo. The village was almost abandoned after the Geralda cyclone in 1994, but farmers started to move back and rebuild their houses in 2004. This photograph also reveals that the economy was more prosperous in the past: households are unable to buy corrugated iron at present.
234
because they still have food for the next day. Children are more alive, move more and speak more. The bodies of children as well as of adults change: they recover their normal weight and muscles. The basic goods that almost all households have are a few cooking pans, some blankets, a bag containing a few nice clothes used for ceremonies and other social events, some plates and spoons, a set of woven items made by the wives, and sometimes a small backpack and a tape player, used on party days only due to the high cost of batteries. Even these basic objects are missing in many households. Blankets are not sufficient in number for the whole family or are very old. The missing items, especially new clothes, are bought before the national day (June 26). This period of the year is a time of party and joy and people do everything possible to eat and dress decently. As rice harvest occurs in April and May, some rice is sold to purchase clothes and to donate for the collective buying of zebu for sacrifice. This rice is sold at very low price (no more than 0.2 USD per kilogram), despite the fact that rice will have to be bought at twice this price during the food shortage period. Children are educated in the same public school despite the fact that their parents know that the education quality is poor. Only a handful of households send their children to a private boarding school, managed by the catholic church of Andasibe, 27 kilometers from Ambodilaingo. People usually stay in their village and do self-medication or use traditional medicine when they are ill, because going to Beforona results in costs they cannot bear (Chapter 3). When they are seriously ill, they cannot afford to move to the district town and pay the hospital. It would cost one to a few hundreds USD to cure a disease such as a tuberculosis in its final stage, or to do basic surgery. This sum would need the contribution of the whole community to be gathered.
235
In this context of high poverty, the capital of farmers, except land, is reduced to almost nothing. Agricultural tools are only the machete, a stick for sowing rice in tavy and a small angady, usually very old and used for weeding. Only one person in the whole fokontany would have a plow. For many households, increasing this capital by buying a new and bigger angady that costs about two USD cannot be done without some sacrifice. Even labor force is insufficient in this weak economy. Very few documents about the region raise this issue 160 but it was one of the facts I most deeply experienced during my stay in the village. When asked to explain why they do not improve a larger part of their bottom land, farmers often answer that they lack food to work hard. Things are that simple. This issue is not reported in the previous section except in one citation, because I had no idea to ask about it and people are ashamed to say that they lack food. But in the end, I had more opportunities to discuss this issue and it was clear that lack of food was a major constraint. Others farmers asserted that health issues or the lack of cash are their main concern, but this still reflects the same problem of bad diet, low calorie intake, and high vulnerability. A comparison with the situation in other regions of Madagascar helps one to understand that this “energy” constraint is not a new myth. When the Betsileo farmers of the highland, who are regarded as hard workers, spend their whole day plowing, they need a huge plate of rice to recover their strength, which makes them famous for the quantity of rice they are able to eat in one meal! If they do not have such a plate, they will obviously not be able to do the same work the next day. If 2000 calories per
160
Styger (2004) collected testimonies that express a similar situation in Ambavaniasy, a fokontany located farther north on the forest edge, on the National Road 2. Food shortage creates income needs that result in the negligence of agricultural production: “When I plant rice, I can only harvest after six months. If I am in need of food or money, I have to go cut wood, which does not allow me to do a good weeding. The rice harvest is therefore low and I have to concentrate more and more on the wood” (a farmer in Ambavaniasy, in Styger 2004, 39).
236
day is the minimal necessary intake to have ordinary activities, the need can reach 8000 calories per day in the case of intense efforts 161 . Plowing the whole day with an angady, creating terraces on slopes, digging canals for irrigation with a hoe, transporting 50 kilograms cement bags for 17 kilometers to build a dam may necessitate much energy and be problematic when meat is rare and calorie intake insufficient. If farmers prefer to use their strength to carry palisander boards, it may merely be because they immediately earn sufficient cash to buy rice and recover their strength. Concerning income, even the one USD per day economy is out of reach for most farmers in Ambodilaingo. Cash is absent in most households at anytime of the year. When it is present (after the transportation of wood, bananas, or rice, or more rarely after the sale of honey or local alcohol) it never represents more than a few USD and is generally spent to buy rice and basic products, sometimes a small backpack, a radio, a watch or some clothes. When foreigners want to buy something using a big note (50,000 FMg, equivalent to 5 USD in 2006), farmers never have the change. In sum, farmers are much poorer in Ambodilaingo than in the villages located close to the road. Due to transportation cost and limited access to market, people almost live in autarky and the economy still depends mostly on tavy. The repression on this land use, the fact that it reaches its ecological limits and the reduction of the fallow period result in insufficient production to sustain livelihoods. Palisander transportation provides significant income but mostly serves to compensate for the low rice production and to purchase basic products. As the villages located close to the road provided most of the knowledge basis about the region (Chapter 3), there could
161
Cycling, for example, can require up to 8000 calories a day (http://www.ultracycling.com/nutrition/calories.html).
237
be a significant bias in the representation of tavy farmer communities by aid organizations. It may be, however, that more remote villages located farther south are not in the same situation of deep crisis, due to the absence of forest control. If this was true, it would also mean that the negative economic impact of an enforced ban on forest clearing would be very significant. A few households, nevertheless, are the exception to this depressing picture. They are also at a quite low economic level in comparison with urban standards, but they can be asked for change on a 5 USD note. Some send their children to the private school of Andasibe; some have a more powerful tape player that they hire out during village celebrations; others have some unusual items in their house, such as a gun, a radio or a tape player that looks quite new, and batteries to make it work all day. In most cases, the relative prosperity of these atypical households has a nonagricultural origin. The head of household may be a representative of the state administration (fokontany President); a person who traveled and worked outside the village, found good job opportunities, saved money and came back; a traditional musician that gets paid for playing at ceremonies; a man who specialized in wood extraction and logs trees himself; a merchant who opened a small shop in the village. I would call these atypical people the local elite. In the few cases where they owe their success to agriculture, this results from a combination of advantages: they inherited cattle from their ancestors, have a good spot for irrigated rice cultivation and a large track of fertile land to grow banana trees. We will see in Chapter 6 that this local elite is first in line to capture the aid provided by projects.
238
8. The influenza epidemic An influenza epidemic occurred in July-August 2002. This epidemic has been recorded by the World Health Organization in the Tanala region, located several hundred kilometers farther south. It moved northward to Beforona, where it has not been recorded officially but was witnessed by myself, the students I worked with and by medical staff working in the commune. The disease came to the village like a wave. The first day, a few people were ill and had to rest. Their family provided them with care. The second day, there were ill people in every family. The third day, almost everybody was ill in the village and lay on a mat, rolled in a blanket. Diseased people were visited by the few relatives still in good shape. Many households had no food in their house, no cash available and insufficient strength to collect cassava from their fields. So they just waited for the disease to go. The third day, people started to die, mostly young children and old people. Food was purchased by the World Food Program through the Food Security Program of the nongovernmental organization ADRA 162 , financed by USAID, and healthcare was provided by private donors. The head of Beforona’s hospital was the first to come on the ground and to provide healthcare. He made a report to the district health center in Moramanga. The Director of ADRA-PSA and a representative of the Red-Cross office in Moramanga also came to state the situation. The private local doctor was hired (he actually worked almost for free) and was given a stock of drugs. He provided healthcare to about 800 persons in the three fokontany he visited. People actually did not die from the influenza directly. The virus, in synergy with malnutrition, weakened their immune system, resulting in a malaria crisis and in the development of other latent pathologies, mostly respiratory diseases. Both the private doctor and Beforona’s health center chief of staff agreed on
162
Adventist Development and Relief Agency.
239
this mechanism. Their statements are consistent with the WHO 163 report about the epidemic that occurred in Ikongo. WHO and a nongovernmental organization were contacted in Antananarivo. They were unfortunately unable to act unless they were requested to do so by the Ministry of Health, whose decision depended on reports received from provincial and district offices. The district health office sent a medical team which refused to recognize the existence of the epidemic and provided no healthcare. After one night in Ambodilaingo, the team traveled back to Moramanga. Some messages were broadcast on the radio contending that the epidemic was a rumor. Eventually, facing the evidence, the district health services provided some drugs, one week after the epidemic had passed. The disease may have killed 36 persons in Ambodilaingo out of a population of about 1500. The mayor recorded 27 deaths. The difference can be explained by the fact that the mayor did not have time to visit all the scattered households and by the fact that many people, mostly young children, are legally nonexistent because births are often declared only after a few years, or are not even declared. I decided to investigate the situation in villages farther south, to see if the epidemic was still striking and to understand where it came from. I visited the commune of Ambalabe (Map 4 page 112), located 55 kilometers southeast from Beforona, guided by the person who would later become the mayor of Beforona. In all villages we crossed, local authorities described the same epidemic and established a list of deceased people. The death toll was always the same: about 2% of the people, mostly babies and old persons, during a one month period. Ambalabe had been stricken two months earlier and people explained that the epidemic had come from farther south, from Ifasina II (Map 4 page 112). If one would walk until Ikongo, located 300 kilometers 163
World Health Organization.
240
farther south, he would cross just two roads, in Marolambo and Ifanadiana, the first one being closed almost all year round. But he would cross a village about every two hours. It may be that people died silently in the villages located along this itinerary. The disease did not strike all villages in the same way. There were reported cases in many villages of the commune but people died only in three fokontany: Ampasimazava, Ambodiara, and Ambodilaingo (about 100 persons in total). In the other fokontany, people just had a “cold,” managed to cure themselves by a better diet and had access to healthcare, at the communal hospital or at the private doctor’s office. These differences reflect the lower or higher vulnerability of communities. The three fokontany that suffered from the outbreak are the most remote ones. Cash economy is almost absent, the agrarian system is in a situation of crisis and access to healthcare is problematic. The healthy period only lasts a few months after the rice harvest. After that, the diet is based on cassava and people lack the strength and courage to resist diseases. In sum, the influenza outbreak confirmed the high vulnerability of tavy farmers living in remote areas and having no access to market, forest land or other resources. To put a ban on tavy, the main land use that sustains the livelihoods of these vulnerable communities, would be a crime that only ignorance could explain. Fortunately, the Malagasy government is not ignorant of this vulnerability and enforced the ban during a short period, in 2002, just enough time to please the donors and favor the launching of EP3.
9. The Agrarian system We saw that the region is in a situation of agricultural crisis due to the reduction of the fallow period. The reproduction of fertility over time is not allowed. Fallow periods usually last from three to ten years in Ambodilaingo. The consequence is yield decrease that forces the development of new land uses. If there had been no
241
repression of forest clearing, the same situation would have occurred sooner or later, after the clearing of all forest land or when the remaining forest land became unsuitable for the existing land uses. But the fact that the crisis might have happened anyway does not imply that it could not be avoided. Agrarian system analysis can help us to envision what pathways could be used to avoid this crisis and to anticipate land use change, before the systems reaches its political or ecological limits. The challenge is to develop a new system which would have a higher carrying capacity, i.e. which could develop and sustain the life of more people on the same land. We saw that the elements of this new system already exist. They are mostly paddy fields, home gardens, ginger cultivation systems and livestock systems – mostly pig farming. A more general framework is necessary, however, to understand more comprehensively the changes to come and the challenges associated with them. This leads me to introduce the work of Esther Boserup (1965) about the conditions for agricultural growth. We will see that the dynamics analyzed in this chapter, to which more insights will be given in Chapter 6, can be better understood in the light of Boserup’s framework. Boserup (1965) criticizes the Malthusian perspective, which limits its enquiry to the understanding of “how changes in agricultural conditions affect the demographic situation” (Boserup 1965, 11). Her own concern is to understand “the effects of population change upon agriculture” (Boserup 1965, 11). According to the Malthusian theory, population growth is the dependant variable, determined by changes in agricultural productivity which, in their turn, are explained as the result of extraneous factors, such as the fortuitous factor of technical invention and imitation. (Boserup 1965, 11) Boserup’s assumption, to the contrary, is that
242
the main line of causation is in the opposite direction: population growth is… the independent variable which in its turn is a major factor determining agricultural development. (Boserup 1965, 11) Boserup’s main argument is that population growth increases the labor force available and renders the development of new land uses necessary and possible. More labor input facilitates the adoption of new technologies, which can lead to a total landscape transformation. This phenomenon would have been overlooked by classical economists because they regarded soil fertility “as a gift of nature, bestowed upon certain lands once and for all” (Boserup 1965, 13). They believed that the only pathways to develop agriculture were “the expansion of production at the… margin, by the creation of new fields, and the expansion of production by more intensive cultivation on existing fields” (Boserup 1965, 12). Moreover, they did not explicitly consider fallow land and pastures as agricultural land. For example, they saw North America as a virgin land to be conquered, whereas the landscape was already anthropogenic when the first colonists settled in New England (Cronon 1983). Challenging this classical view, Boserup proposed the concept of frequency of cropping. She showed that early stages of agriculture are characterized by systems with long fallow and that fallow periods are progressively reduced in later stages, until the land becomes permanently cultivated. The reduction of the fallow period leads to lower soil fertility, but this fertility can be rebuilt in the later stages. This is permitted by population increase, which provides a labor force with the power to create new agro-ecosystems and to rebuild the soils that have been lost during the period of extensive cultivation. The consequence is that soil fertility, instead of being treated as an exogenous or even unchangeable “initial condition” of the analysis, takes its place as a variable, closely associated with changes in population density and related changes in agricultural methods. (Boserup 1965, 13)
243
Using the concept of frequency of cropping, Boserup proposes a classification of land uses with respect to their degree of intensity. Stage 1 is “forest-fallow cultivation.” Forest land is cleared and cultivated one or two years and then left in fallow for a long time span that allows secondary forest to regrow. Shifting or swidden cultivation systems usually belong to this category. Stage 2 is “bush-fallow cultivation.” The fallow period is shorter (six to eight years) and just allows the development of a mix of shrubs and herbaceous plants. This stage is the one that predominates in Ambodilaingo. Stage 3 is “short fallow cultivation.” The fallow period (a few years) only allows herbaceous vegetation to grow. Some villages in the commune of Beforona are already at this stage, especially the most densely populated ones located close to the road (Photographs 7 through 10 page 156 to 159). Stage 4 is “annual cropping.” Here, the fallow period lasts just a few months and crops are rotated in order to avoid soil depletion. The agriculture of the Malagasy highlands is at this stage, at least in the bottom land, and this system is progressively being exported to the small plains and large bottom lands of coastal regions, including in Beforona. Native farmers of Beforona also attempt to reach stage 4 on their own, by struggling to create paddy fields and by inventing new crop rotations where ginger and plowing play a prominent role (Chapter 3). Stage 5 is “multiple cropping.” Several crops are grown on the same plot the same year, in association and/or in succession, allowing several harvests per year. These systems are encountered in densely populated areas such as the highlands of Burundi (Pollini 1993), as we will see in Chapter 10. This classification is extremely simple, even trivial. It is based on the simple assumption that soil fertility is a function of land use as much as land use is a function of soil fertility. Soil fertility is depleted from stage 1 to 3, which we can call extensive stages, but is increased during stages 4 and 5, which we can call intensive stages. Fire is usually the agricultural tool that is used in stages 1 to 3 because it allows the
244
concentration of fertility in space (on the top soil) and in time (on the cultivated year). Farmers’ work replaces fire during stages 4 and 5. But as farmers are less efficient tools than fires, this replacement can only be progressive. Farmers build new soils and new agro-ecosystems over the years, but they need a transition period where fire continues to play its role and where extensive and intensive systems are combined, until the new systems achieve their full productive capacity. In this perspective, soil degradation is just a transitory stage to be reframed into a longer historical process. This conception of soil fertility brings into question the vicious circle of land degradation and poverty, a concept which shapes the strategies and practices of development projects, as I will show in Chapter 6. There could be, indeed, a population threshold above which the available labor force allows the rebuilding of soils, and a soil fertility threshold below which farmers are motivated to change their land use and rebuild its fertility. This second threshold explains why the shift does not occur at low population density, by concentrating the available labor force on part of the land. In the light of the previous sections, we can guess that labor productivity is the criterion to consider for determining when these thresholds are reached. Farmers allow soil degradation to occur during the early phase of agriculture development because a decrease in yields can be compensated for by cultivating larger tracks of land, or because labor productivity is still sufficient to satisfy household needs or is superior to that of other systems. When the labor productivity of extensive systems reaches the level of more intensive ones, then the switch can occur. However, things are not that simple because implementing new systems is also a move to the unknown, which implies technological innovation and the taking of risks. Intensification further needs investments, at least in the form of work as we saw. These constraints explain why the switch to new systems does not always occur at the most appropriate time, and
245
sometimes does not occur, especially if the fertility threshold is attained brutally, as appears to be the case in Beforona (Styger et al. 2007). The validity of the Boserupian model, hence, does not imply that Malthusian crises are always avoided. Such crises actually occur, as shown by the history of European agriculture (Le Roy Ladurie 1969; Mazoyer and Roudart 2001) and by the situation in Ambodilaingo. They just cannot be explained by a simple linear relationship between a level of resources supposedly bestowed once and for all and a population having a certain density and technological knowledge. It is also important to notice that primary ecosystems are sacrificed while passing from stage 1 to stage 2, because their clearing extends the land available for cultivation, allowing long fallows to be maintained, and thus delays reaching these thresholds. In contrast to soil chemical properties, biodiversity cannot recover when population density increases, unless primary ecosystems have been maintained in some places and are allowed to recolonize some land once the switch to more intensive systems has taken place 164 . The challenge, thus, is to allow the switch to occur before the complete clearing of primary ecosystems. A ban on forest clearing, logically, is a condition for this to occur, unless forests are located on marginal land with no potential for agriculture. One could wonder why such agricultural development dynamics, which do not appear to be particularly complex and have occurred for many decades in Beforona, have been overlooked, as we can already guess from the previous chapters and as we will see with more scrutiny in Chapter 6. The reason may be that the models of the classical economists that Boserup put into question are still excessively influential, including in development programs in Madagascar. To understand this, we must get
164
This situation is encountered in a few tropical countries which managed to modernize their agriculture before all forests were cleared (for example: Costa-Rica and Puerto-Rico).
246
back to the logic of the agricultural conquests that formed the background of these classical economists. When European settlers arrived in the new world, they did not exactly “develop” their land use from the beginning. Rather, they transferred it from Europe to America. They had the advantage that earlier stages of resource depletion had not yet occurred on the new continent, because native agricultural systems were only at the first or second stage except in a few locations. They greatly benefited from the conjunction of several resources being available at the same time: a capital of knowledge and some finance that they brought with them, and a natural capital that was still in place. Interestingly, the situation in Madagascar may be comparable. Intensive agriculture systems have already developed in the highlands and are now being exported by migrants to less densely populated areas where extensive systems are still practiced by local people with less knowledge and less economic capital. The classic model of agricultural development led development actors to confuse these two agricultural dynamics: colonization at the margin by migrants and development by autochthonous groups. The consequence is that development actors could support the logic of the first process to the detriment of the second, whereas only the second leads to an increase in the carrying capacity and can reduce the pressure on natural resources. We will see in the next chapter that this is what actually occurs in Beforona.
10. Conclusion In conclusion, the challenge in Ambodilaingo is to accelerate the switch to Boserup’s stages 3 and 4 and to make this switch happen before the remaining primary forests are cleared. A ban on forest clearing is a condition to achieve this. Otherwise, the forest will be used to avoid reaching the thresholds that determine land use change. The forest, however, has to remain open to the harvesting of its products, so it can
247
contribute to sustaining livelihood, rendering the ban on clearing more economically and socially acceptable. Closing the forest frontier, however, must be mitigated 165 by support for agricultural intensification, for a quicker passage to stages 3 and 4. This can be done by supporting the development of more intensive cropping systems, such as irrigated rice cultivation, crop rotation and perennial crops. For all these systems, some investment will be necessary, at least in the form of labor, and some risks will have to be taken until the new systems are adjusted to local conditions. Villagers from Ambodilaingo have an almost null investment capacity and their high vulnerability renders them risk averse. Support from outside is thus necessary if this intensification is to happen quickly. This support could take the form of subsidies to energy, in order to favor investment in the form of labor, while the risk factors could be alleviated by taking in charge agricultural innovation. I will finish this chapter by giving two reasons for hope. Communities of farmers in developing countries have two advantages for avoiding a Malthusian crisis. First, the new technologies that are necessary to change the system have long been in existence (developed countries switched from slash-and-burn cultivation to more intensive systems hundreds or thousands of years ago). Second, investment capacity could be huge if the international community was committed to provide it. Resources are indeed available, but what remains to be questioned is rather the form this support could take. Chapters 6 and 7 will attempt to answer these questions.
165
I will eventually conclude, after the analysis of development discourses in Part V, that mitigations will not suffice because they offer no guarantee that the ban on clearing will be actually compensated. In the last chapter of this dissertation, I will argue that a more direct compensation, in the form of subsidies, will be necessary. But for now, based on the analysis of the agrarian system in Beforona, I can only argue about the necessity to support agricultural intensification.
248
CHAPTER 6 RURAL DEVELOPMENT PROJECTS IN BEFORONA
1. Introduction We saw that Beforona is easily accessible from Antananarivo (three hours driving on a good road) and is still covered by rain forest on its western limit (the Betsimisaraka escarpment). For these reasons, several major rural development projects aimed at stabilizing or eradicating slash-and-burn cultivation have worked in the commune for decades. From 1969 to 1972, the CTFT 166 supported 12 households to intensify their land use in a 150 hectare watershed in Marolafa (Map 5 page 114). It proposed various rice intensification techniques, crop rotation on slopes, organic fertilization, fireless alternatives to tavy and anti-erosive techniques, introduced draft force for plowing as well as several forage crops and improved irrigation by the creation of small dams (De Coignac et al. 1973). Some farmers remember CTFT as being “a good project” (a farmer in Beforona, 2002) 167 because some dams were built and “agricultural tools were provided for free” (a farmer in Beforona, 2002). It is not clear whether the project had actually an impact but it probably contributed to lowland rice intensification. In 1974, after the end of the first republic and the departure of most foreign experts, the Marolafa watershed became a research station managed by FOFIFA, the Malagasy National Research Center for Agriculture and Rural Development. Applied research on lowland rice intensification continued. A few technicians having their 166
French Technical Center for Tropical Forestry. In French: Centre Technique Forestier Tropical. This institute is now the Forest Department of CIRAD. 167 Farmers frequently say that this project was the best project they remember. This information can be biased by the fact that the project was run by a French organization while I am French myself. Moreover, the fact that the project was well perceived does not imply that it had a positive impact.
249
office in Moramanga visited the site and supported rice and coffee intensification, but this support did not last due to the bankruptcy of the state. In 1989, in the context of opening to foreign intervention, new aid programs started with support of the Swiss government, which funded two projects: Terre-Tany from 1989 to 1998 and BEMA 168 from 1994 to 2001, whose research outcomes have been utilized in Chapter 3. The main objective of these projects was to provide a comprehensive understanding of the agrarian system and its dynamics. Beyond this research effort, these projects collaborated with the Malagasy nongovernmental organization SAF-FJKM, 169 from 1994 to 1998 (Nambena 2004). The main activities implemented through this collaboration were the construction of community granaries, reforestation and intensification. Farmer groups were supported in nine villages (Nambena 2004). The USAID funded project LDI 170 started in 1999 and resumed the management of the Marolafa station. LDI lasted until 2002 and was followed by a transition project called PTE, 171 until 2004, and by the ERI 172 project which is still ongoing today. These projects have other working sites in the Toamasina and Fianarantsoa Provinces. They work hand in hand with other USAID programs implemented in the area and presently form the Toamasina eco-regional alliance, a platform for coordinating and monitoring conservation and development activities in the Mantadia-Zahamena corridor. In this chapter, I will analyze the rural development activities implemented by the LDI and PTE projects in Beforona. Section 2 will critically examine the
168
Ecological Assessment in Madagascar. In French: Bilan Ecologique à Madagascar. An NGO tied to the Malagasy protestant church. 170 Landscape Development Interventions, operated by the consulting firm Chemonics International. 171 Ecoregional Transition Program. In French: Programme de Transition Ecorégionale. This program is operated by Chemonics International and by the NGO PACT. 172 Eco-Regional Initiatives, operated by Development Alternative International. 169
250
knowledge base that has been produced by the BEMA/Terre-Tany and LDI/PTE projects for developing alternatives to tavy. Three PhD dissertations have been the hallmark in this endeavor: the works of Messerli (2003), Nambena (2004) and Styger (2004). 173 Many other relevant works exist but will not be reviewed here because these three authors already synthesized their outcomes. In Sections 3 and 4, I will analyze the activities implemented by the LDI and PTE project in Beforona and will study the case of the VMCRCC project (Vohidrazana Mantadia Corridor Restoration and Conservation Carbon Project), implemented in the neighboring commune of Ambatovolo. The VMCRCC project is still in its very early implementation but illustrates a new turn that could be taken soon by conservation and rural development policies in Madagascar.
2. The search for alternatives to slash-and-burn 2.1. The BEMA/Terre-Tany research about alternatives Messerli (2003) reviewed most of the existing and potential alternatives to slash-and-burn cultivation applicable in Beforona. He synthesized most of the research work done on this subject by the Terre/Tany and BEMA programs. This section is mostly based on his work. I will start by presenting his review of alternatives based on fire suppression and the maximization of biomass production, which are usually those paid more attention by projects. I will call these alternatives agrobiological technologies for convenience. These technologies are, for instance, mulching, improved fallow, and alley cropping. In a second section, I will present alternatives already practiced by farmers (some of them being practiced in Ambodilaingo as we saw in Chapter 5) and the pathways envisioned by Messerli (2003) to improve them. 173
My own research work was partly financed by CIIFAD (Cornell International Institute for Food, Agriculture and Development), an institute based in Cornell University and partner of the LDI project for the implementation of its activities in the Toamasina Province.
251
2.1.1. Agrobiological alternatives Messerli showed that mulching provokes a very significant yield decrease in rice due to very low germination rates and insufficient nutrient availability, and necessitates a very significant additional labor, in comparison with slash-and-burn systems, as shown in Table 12. These negative effects can, however, be corrected by applying the mulch six months in advance, in order to favor its decomposition, and by implementing the system on flat land (Table 13). In this case, a positive effect of no burning on rice yield is observed the first year, but this effect is not sufficient to compensate for the increased labor requirement. It must be noted that the experiment, whose results are presented in Table 13, has no control: T1 is not a traditional tavy as biomass is only partially burned and is cut six months prior to cultivation as for T2. In order to remedy this, Messerli (2003) compares the results obtained in his trials with the yield of traditional tavy, which averages 1670 kilograms/ha or 4 kilograms/day of labor on less degraded land, according to Moor (1998b). But this comparison can be misleading because Moor’s (1998b) data concern ordinary tavy usually implemented on slopes, whereas Messerli conducted his experiment on flat land. An experiment with a control plot which would be a tavy practiced on the same flat land should be done to check the advantage of fireless practices. Other options, such as plowing or the application of manure on land cleared using fire, would also be worthwhile to test and compare with mulching. Concerning improved fallow, Messerli (2003) obtains a significant biomass increase on bottom land by planting Crotalaria, but identified no plant that can produce more biomass than native vegetation on slopes. However, he did observe a 20% yield increase, with no additional labor cost, when he associated Tephrosia with cassava (Messerli 2003, 179).
252
Table 12: Comparison of tavy with alternatives Tavy (control)
Treatment Biomass mulching and composting
Biomass mulching and cutting
Germination rate (after 5 weeks)
10 stems/m2
42% of control
47% of control
Yield
100% (834 kg/ha)
41% of control
31% of control
Invested labor
100% (4.5j/120 m2)
580% of control
320% of control
Criteria
Source: Messerli (2003, 151). Translation by Pollini.
Table 13: Effect of limited burning and no burning on rice yield and labor productivity Treatment First year T1 (limited burning the first year)
Second year
T2 (no burning)
T1 (limited burning the first year)
T2 (no burning)
Criteria Required days of labor
190
292
185
200
Rice yield (kg/ha)
2430
3350
1540
1520
Labor productivity (kg/day)
12.8
11.5
8.3
7.6
Source: Messerli (2003, 162). Translation by Pollini.
253
Concerning alley cropping, ginger plots protected by Tephrosia vogelii Hook. f. planted in contour strips resulted in a maximum 40% decrease in erosion, in comparison to traditional systems (Messerli 2003, 197). This allowed a 29% yield increase (Messerli 2003, 192) if the hedgerow was appropriately managed. However, competition between hedgerow and crops eliminates the advantage if pruning is delayed, which can easily happen due to constraints in the farmer’s work schedule. On the other hand, no supplementary work is required in alley cropping because the better control of weeds by mulching compensates the time spent establishing and managing the hedgerow. In sum, mulching appears to be irrelevant in Beforona due to the excessive labor it requires, while other agroforestry technologies may allow a 30% productivity increase 174 in the best cases. This gain is significant but appears to be insufficient to expect an agrarian revolution to occur through these technologies. Agroforestry may be of interest for broadening the scope of potential future land uses but should not be regarded as a panacea for intensifying agriculture in the short-term. Even if the 30% productivity increase was confirmed, the fact that alley cropping and improved fallow have not been adopted, despite important extension efforts, should be considered seriously. The risk factor, the opportunity cost of labor, the trade-off between higher yield obtained in the long-term and lower yield obtained in the short-term, in comparison to tavy, have never been analyzed in depth, whereas these factors may be the main concerns at the farmer level. The integration of fast growing agroforestry shrubs in farming systems must also be investigated in the long-term. It can be
174
In the context of Beforona, where land is not the limiting factor, labor productivity is a more important criterion than yield to assess the profitability of cropping systems. Labor productivity can be increased without converting more land to agriculture, by increasing the frequency of cropping (Chapter 5), i.e. by decreasing the fallow period.
254
expected that the positive impact on yield will increase with time, but the labor requirement may increase as well. Shrubs planted in hedgerows may grow fast and require increasingly frequent pruning so not to compete with crops. Those planted in fallow could constrain the development of more intensive systems with crop rotation and plowing, because their root systems would render plowing excessively burdensome. Nambena’s (2004) outcomes will confirm this view. Moreover, if research efforts in agroforestry are to be continued, it will also be necessary to demonstrate that these techniques are more profitable and more suitable to farmers than more conventional technologies such as plowing, manure application or irrigation. It will also be necessary to demonstrate that erosion is indeed a key problem in Beforona and that these technologies are efficient for limiting soil and nutrient losses. Based on existing knowledge, even this point is doubtful. According to Brand et al. (1997, 32): the traditional technique is largely better adapted, ecologically, to the torrential rain than most proposed or already practiced technical innovations.… An improvement of the physical quality of soils (for example by mulching or agroforestry) on steep slopes could favor crumbling and landslides. The anti-erosive improvements should be adapted to extreme events rather than to daily rains because most soils are lost during these rare events. 175
2.1.2. Intensification of indigenous alternatives This second series of alternatives concerns the integration of agriculture with livestock husbandry and the development of permanent cultivation on the most fertile land, usually on bottom land. These techniques are already practiced by the 175
Translation by Pollini. Original text in French: “ la technique agricole traditionnelle est écologiquement et largement mieux adaptée aux pluies diluviennes que la plupart des innovations pratiquées ou proposées.… Une amélioration de la qualité physique des sols (par exemple par mulching ou agroforesterie) sur pentes raides pourrait favoriser le déclenchement des éboulements et des glissements de terrain. Les dispositifs anti-érosifs devraient s’adapter plutôt aux événements extrêmes qu’aux pluies quotidiennes parce que c’est pendant ces moments rares que la plupart des terres sont perdues.”
255
population, in a rudimentary form and at a limited scale, as we saw in Chapters 3 and 5. First is ginger cultivation on plowed land. This crop has existed in Beforona for several decades but boomed in the late 1980s, when the renovation of the road opened access to a wider market (Chapter 3). Beforona is now the leading area for ginger production in Madagascar. Ginger is usually cultivated on slopes (Photographs 7 to 10, page 156 to 159). The land is cleared, burned and plowed, which provokes elevated soil losses 176 (Brand and Rakotovao 1997). These losses are the main motivation of projects to improve the system. Messerli tested an improved technique consisting in growing ginger on bottom land and applying 25 tons of cattle manure per hectare. Yields increased by 288% and labor productivity increased by 203% in comparison with a control with no manure (Messerli 2003, 198). Clearly, the system is much more profitable than tavy and agrobiological technologies, but the constraint is the availability of manure as most farmers do not own cattle. The quantity of manure applied to the plot was very high and more modest amounts should be tested. These results, however, show that integration of agriculture with livestock husbandry could be a wiser pathway than agrobiological technologies for future intensification. Messerli (2003) showed that swine husbandry could be developed to favor this intensification. The initial investment is much lower than in case of cattle husbandry. 177 Beyond the production of manure, pig farming offers multiple advantages to farmers. Rice bran and by-products of cassava and bananas can be used to feed pigs. The sale of animals can provide a significant amount of cash that helps to 176
These losses are much more elevated than in the case of tavy: 50 ton/hectare/year on average in ginger cropping systems, if we consider the whole rotation, including the fallow period; five ton/hectare/year in the case of tavy, also for the whole rotation (Brand 1997). 177 In 2002, a young pig cost 100,000 FMg (about 15 USD) while a young zebu cost about 800,000 FMg (about 120 USD).
256
satisfy social obligations or to resist natural calamities. Farmers are always very motivated to develop this activity, as Nambena (2004) showed in the case of Ambinanisahavolo. 178 The main constraints, on the other hand, are the insufficient investment capacity to buy the first animals and the risks of disease. A second pathway for intensification is the improvement of home gardens. Messerli (2003) confirmed that the home garden is the most profitable agricultural system in the area in terms of labor productivity (Chapter 3) but also needs important labor investment for its establishment (Chapter 5). He proposed to associate annual crops such as ginger, beans and other plants traditionally grown in tavy, as a way to provide short-term high returns and amortize these investments. He demonstrated that these crops increase the profitability of the system and allow a shorter-term return, but it remains unclear whether the association is actually an advantage. Farmers do their empirical economic calculations at the farming system scale, not at plot scale, so growing ginger in or out home gardens will not make any difference for them unless bananas and ginger have positive interactions. Messerli (2003) did not provide insight on this point. In a context where land is not rare and considering that ginger accommodates poor soils, the advantage of the association remains doubtful, which Nambena (2004) will confirm. Lastly, Messerli investigated the potential of irrigated rice cultivation on bottom land. His outcomes are similar to those of Chapter 5. The lack of investment capacity and the damage caused by cyclones would be the main constraints to the creation of new paddy fields. Land tenure and socio-organizational aspects would also be of prime importance. In this context, the introduction of new techniques and varieties would not be the priority (Messerli 2003). Ultimately, Messerli (2003) recognizes that paddy fields cannot substitute totally for tavy, but considers that the potential for its 178
A village in the commune of Beforona.
257
development is, however, significant. He also showed that permanent rice cultivation can stimulate the establishment of other permanent systems, such as home gardens and crop rotation in and around the bottom land.
2.1.3. Toward new farming systems Based on these results, Messerli (2003) assisted three households to implement a combination of alternative systems. All three significantly increased their production. Labor productivity ranged from 0.8 to 3.5 USD per day of labor 179 (the maximum was obtained with animal husbandry). Home gardens, irrigated rice cultivation, livestock husbandry and the use of manure were the main elements of this success (Figure 12). Agrobiological technologies were also used but did not appear to be essential. Only one household, which was living close to the forest, continued to do tavy but on a significantly reduced area. Messerli (2003) concluded that the key factors for this successful intensification were (1) the integration of agriculture with livestock husbandry; (2) the existence of opportunities to develop irrigated rice cultivation (access to bottom land and efficient social organization to manage the water); (3) the possibility of selling products from home gardens, which allowed the purchase of cash crop seedlings and (4) the abandonment of tavy, or at least of large tavy practiced in remote places. This abandonment was essential to allocate more labor force to more profitable activities. Off-season cultivation of beans and vegetables was particularly successful because it did not compete with rice cultivation (in tavy or in paddy fields), still regarded as essential to assure food security. Most revenue was provided by the home garden (mostly planted in bananas), by the sale of animals and by ginger cultivation on
179
About 6000 to 27,000 FMg. Gross estimation due to the variations of the exchange rate and to the uncertainty about when the data were collected.
258
Ecological cycles
Economic cycles
Figure 12: Alternative farming system in Beforona (to be compared with figure 10 page 166) Source: Messerli (2003, 279). Translation by Pollini.
259
bottom land fertilized with manure. One household also speculated on the price of products (rice price at least doubles between harvest and food shortage periods), which almost doubled his revenue.
2.2. Nambena’s (2004) research on agroforestry alternatives Nambena (2004) strengthened Messerli’s (2003) conclusions. Her research was explicitly aimed at developing alternatives to tavy based on agroforestry technologies such as agroforests, improved fallow and alley-cropping. She tested a large number of agroforestry species chosen for their potential to produce green manure, improve nutrient balances and provide by-products. Concerning improved fallow, she concluded that fast growing agroforestry species cannot easily compete with the invasive species that already grow in fallows. She proposed to move straight to systems that are more intensive than improved fallow. She observed that the elimination of invasive shrubs, by uprooting, is “a necessary condition for the abandonment of burning and the adoption of permanent agriculture”180 (Nambena 2004, 218). As ginger cultivation necessitates plowing, she proposed to use this crop to eliminate perennials fallow plants and amortize the important labor required for this elimination. She observed that some farmers already do this and plant beans or cassava on plowed land after ginger’s harvest. Eventually, she proposed manure, compost or mulching application as a way to maintain sufficient fertility in these new systems. In short, Nambena (2004) proposes the transition to Boserupian stage 4 that I described at the end of Chapter 5 and observes that this transition is already on the verge of happening. It is interesting to notice that her working hypothesis was to favor the plantation of perennials (agroforestry species), including in fallow land, while her 180
Translation by Pollini. Original text in French ”le nettoyage a fond par déracinement de ces arbustes rudéraux [est une]… condition indispensable pour le renoncement au brûlis et pour une mise en culture prolongée”
260
outcomes lead her to propose the uprooting of perennial species in fallow. To use the jargon of the working hypothesis of this dissertation, the reality has been stronger than the representations that framed her research (the belief that agroforestry technologies were the solution). Chapter 10 will show that the demystification of agroforestry alternatives did not occur only in Beforona. Nambena, however, still advocates for the use of agroforestry species planted in contour lines to produce mulch and green manure and protect the soils from erosion, as shown by the exemplary hillside design she proposes (Figure 13). She is aware that some factors could constrain the adoption of these techniques (labor requirement, risk factor) but she did not fully assess these constraints. In sum, Nambena (2004) may have found the clue for the future development of the region and provided some arguments to demystify agroforestry technologies. But she partially remained framed in this agroforestry myth and may have not apprehended all the consequences of her outcomes.
2.3. Styger’s (2004) research about fireless alternatives Styger (2004) conducted research in two villages located close to National Road 2: Ambinanisahavolo, where Nambena (2004) did part of her research, and Ambavaniasy (commune of Ambatovolo, west from Beforona). Her research focused on the development of fireless techniques for upland rice cultivation, using technologies such as manure, green manure, mulching, improved fallow and the application of guano-phosphate. Styger (2004) analyzed land degradation in detail. She showed that thresholds that render slash-and-burn unprofitable can be quickly reached at the scale of a whole landscape and are not anticipated by farmers, who are then projected into a situation of crisis they cannot escape, due to a lack of suitable alternatives (Styger et al. 2007).
261
Forest trees and cover crops Hedgerow Agroforestry trees and cover crops Hedgerows Forest trees and ginger Hedgerow Cassava Hedgerow Ginger in young home-garden Hedgerow Home-garden with cash crops Hedgerow Paddy field Hedgerow
Figure 13: Eco-design of an exemplary agroforestry hillside. Source: Nambena (2004, 221). Translation by Pollini.
262
She quickly reviewed existing alternatives practiced by farmers or paid attention by former projects. She noticed that improved fallow technologies developed by BEMA had quickly been abandoned by farmers after the end of trials, because “the techniques were developed only in preliminary form and were not fine-tuned to the current system” (Styger 2004, 36). Farmers were “unable to solve emerging problems… [and] it was easier to return to their traditional practices” (Styger 2004, 36). She stated that irrigated rice cultivation could not produce enough rice to substitute for tavy, due to insufficient availability of bottom land, and derived from this statement a less optimistic conclusion than Messerli’s about the potential of this land use. Concerning home gardens, she pointed out the insufficient quality of their products (except for bananas) and low market price as the main constraints for their development. But she missed the main advantage of this system (the high labor productivity) and the main constraint for its development (the need of investment for its establishment, at least in the form of labor). Concerning off-season crops, mostly beans and vegetables, she considered them to be the most successful alternative. Finally, Styger (2004) deemed that composting is not applicable at a large scale. After this review, Styger (2004) concludes that “people will continue tavy as long as possible, because it is the only technique they know. The result is that all uplands degrade and there is no reversal of this trend in sight” (Styger 2004, 50). This conclusion contrasts with the conclusions of Messerli (2003) and Nambena (2004), as well as with my outcomes in Chapter 5, which showed that irrigated rice cultivation, home gardens, and ginger cultivation on plowed land are alternative systems that most farmers already know, but that a lack of investment capacity constrains the development of these systems at a larger scale. The difference between the conclusions of these authors may, however, be explained by the fact that Styger (2004)
263
conducted part of her research in Ambatovolo, where the topography is particularly hilly and where bottom land suitable for intensification cover limited areas. Based on these analyses, Styger (2004) proposed to develop alternatives that would minimize nutrient losses and improve nutrient cycling. She proposed the application of guano-phosphate (which provides an important level of the most limited nutrients P and Ca) in combination with mulching and fallow improvement with leguminous species. Her treatments were: •
T1: slash-and-burn (control)
•
T2: slash-and-mulch without guano-phosphate
•
T3: slash-and-mulch with 600 kilograms/hectare of guano-phosphate
•
T4: slash-and-mulch with 1200 kilograms/hectare of guano-phosphate
She applied the same crop rotation on all three treatments: •
Year 1: upland-rice in the main season, followed by off-season beans
•
Year 2: ginger
•
Year 3: improved fallow (Crotalaria grahamiana Wight and Arn.)
Table 14 summarizes her results. These results are consistent with Messerli’s (2003) outcomes. They show that slash-and-mulch (T2) leads to lower yields than slash-and-burn (T1) on the early rotation (rice and beans) and moderate yield increases later on (for ginger). When guano-phosphate is applied in addition (T3 and T4), significant yield increases are obtained but only for beans and ginger. The more degraded land (Rubus and Imperata fallow) are those where the yield gains are the most significant. Styger (2004) also compared the treatments in terms of monetary return (Table 15). She showed that more production and added value is obtained using guanophosphate but not with slash-and-mulch alone. But the use of monetary return as the main criteria for economic assessment is disputable, because it does not consider labor
264
Table 14: Relative yields of fireless alternatives according to type of land. (% of yield obtained in slash-and-burn system = % of T1) Dominant species in fallow Trema Psiadia Rubus
Imperata
Average for all fallow types
Crop
Treatment
Rice
T1 (SaB)
100
100
100
100
100
T2 (SaM)
53
21
53
34
31
T3 (Gp600)
87
47
63
81
56
74
33
62
265
T4 (Gp1200) Beans
T1 (SaB)
100
100
100
100
100
T2 (SaM)
128
56
86
91
96
T3 (Gp600)
177
196
265
275
221
349
299
324
T4 (Gp1200) Ginger
T1 (SaB)
100
100
100
100
100
T2 (SaM)
155
108
161
117
126
T3 (Gp600)
119
127
226
152
156
254
159
195
T4 (Gp1200)
Source: Styger (2004: p.192). Adapted by Pollini.
Table 15: Relative monetary return of fireless alternatives in USD Treatments
Crops
T1 (tavy)
Rice
T2
T3
T4
Guanophosphate cost 0
Return from harvest 351
Return input cost 351
Relative monetary return 100
Beans
0
107
107
100
Ginger
0
1312
1312
100
Total
0
1770
1770
100
Rice
0
110
110
31
Beans
0
103
103
96
Ginger Total
0 0
1647 1859
1647 1859
126 105
Rice
62
198
136
39
Beans
62
236
175
163
Ginger
62
2042
1980
151
Total
185
2475
2291
129
Rice
123
218
94
27
Beans
123
346
223
209
Ginger
123
2562
2439
186
Total
369
3126
2757
156
Source: data from Styger (2004, 239). Adapted by Pollini.
266
costs. We saw that labor availability is limiting for most households and that labor productivity is a criterion that matters more than yield for farmers in Beforona, because much land is still available. Chapter 5 provided arguments for this hypothesis and the Section 3 will confirm it again. Styger (2004) did not calculate the labor cost of the techniques she proposed and the description of mulching operations she provided leads one to expect that these costs are elevated. 181 The systems proposed by Styger (2004) should also be compared with other options. From an agronomic perspective, its makes sense to use tavy as the control treatment. But from an economic perspective, this is not the comparison that matters. The application of guano-phosphate necessitates significant investments (185 to 369 USD for the complete rotation). If a farmer would be proposed to make such investments, he would compare this option with other possible investments, such as buying bottom land, paying labor force to create paddy fields, planting more banana trees or buying animals to raise. This last option would appear more appealing to him because it would allow the production of manure which could play the same role as guano-phosphate. The investment would not need to be renewed in the future unless the animals died from disease. Animals could even multiply and provide income and draft force, in addition to manure. One hundred eighty-five to 369 USD (the cost of guano-phosphate in T3 and T4) is the price of two to four young zebus or nine to 18 young pigs. There is little doubt that this option would be preferred by farmers and that this choice would be a wise choice. In conclusion, Styger’s (2004) statement that no alternatives are suitable to farmers is certainly right in the sense that existing alternatives are not easily adopted
181
“For the mulch treatments in the trema, Psiadia and Rubus fallows, the woody stems and thick branches were removed from the mulch plots just before planting. The leafy biomass, small twigs, and branches remained on the plot as part of the mulch. The mulch was cut with the machete into pieces of 50cm to 1 m in length, and was evenly distributed on the plot.” (Styger 2004, 179)
267
and fail to substitute for tavy at a significant scale, whereas this land use is clearly in a situation of crisis. But the alternatives she proposes face the same constraints as the existing ones (paddy fields in bottom land and terrace; development of livestock husbandry and use of manure; extension of home gardens) for their adoption. They are probably even more constrained due to the elevated investment they require, in the form of cash to buy guano-phosphate and in the form of labor to prepare the mulch.
2.4. Conclusion: future land uses in Beforona In conclusion, the excessive focus on agrobiological pathways to intensification may be misleading. Mulching clearly appeared to be unprofitable at least in the short and mid-term. Agroforestry technologies did not appear to be the most promising alternatives due to their longer term return, the limited productivity increase they allow, and uncertainties about management constraints that could arise in the future. These limitations have been underestimated because the strong desires of development and conservation organizations to suppress fires and maximize biomass production narrowed the range of envisioned options, and because alternatives were often compared with tavy, instead of being compared to one another. Messerli’s work (2003) was a benchmark in the production of knowledge about land uses in Beforona because it made this transversal comparison. Unfortunately, fire-suppression policies and technological fashions have been stronger than Messerli’s outcomes and researchers continued to produce biophysically perfect techniques while overlooking economic constraints to their adoption. A new series of agrobiological technologies appeared recently, based on the use of cover crops and the absence of tillage. A few projects started to develop them in Beforona in collaboration with ANAE. Not many conclusive results are available yet but it seems, according to the few discussions I had with project staff, that the main
268
difficulty faced by these techniques is the need to eliminate the perennials plants that dominate fallow land. These plants grow faster than the cover crops and ANAE proposed to eliminate them using chemical herbicides. 182 A second problem often evoked is that the advantage of cover crops, in comparison with burning, occurs only the third or fourth year. In sum, these new alternatives may be submitted to the same economic constraints for their adoption. 183 In spite of this, zero tillage technologies and cover crops are already planned to be the new technological package to be developed and proposed to farmers in Beforona by the Marolafa research station, as we will see in the next section. As they offer new agronomical challenges, they may play the same role as agroforestry did before in maintaining the belief that finding solutions to land degradation implies to tackle biophysical issues, whereas the main challenge may rather be economical. In parallel to these research efforts, farmers will continue to develop their own land uses. Based on the outcomes of Chapter 3 and 5 and on the work of Messerli (2003) and Nambena (2004), we can guess that future farming-systems in Beforona will be a combination of irrigated rice fields, home gardens, permanently cultivated plots fertilized with manure, cassava plots on degraded land, livestock systems, and tavy, which would persist but decrease in extent during a transition period. Support to investments, the opening of new roads to access markets, labor productivity maximization and the minimization of risks are certainly key criteria for the successful development of these new systems. Significant investments will be necessary to create irrigation schemes and terraces, to acquire animals in order to produce manure, to pay labor force for the extension of home gardens and to buy cash crop seedlings. 182
According to discussion I had with ANAE experts who visited Beforona. I am discussing here the suitability of these techniques in places like Beforona. Zero tillage technologies, on the other hand, proved to be profitable and are widely adopted in the Vakinankaratra and Alaotra Regions (Groupement Semis Direct Madagascar 2006), where agriculture is more advanced and prosperous.
183
269
These new farming systems may develop in clusters around new dwellings located close to bottom land (Messerli 2003). This new organization of space is already visible in the villages with more degraded land and easy access to markets, such as Ambinanisahavolo, (Nambena 2004). As farmers put it, “when the soil is unfertile, we move to the bottom land.” 184 From this perspective, erosion is not as much an issue as it was expected to be. It can occur but does not necessarily lead to a dead end because part of the nutrients and soil lost on slopes accumulates in bottom lands. Once these bottom lands have been totally cultivated, farmers will be interested in reconquering land on hill sides. Then they will be interested in agrobiological technologies, which can explain the success of zero tillage technologies and cover crops in the more densely populated highlands (Groupement Semis Direct Madagascar 2006). In sum, the works of Messerli (2003) and Nambena (2003) provided more evidence of the Boserupian dynamics envisioned in Chapter 5. They showed that the seeds of future land uses are already present in the landscape and revealed that projectdriven technological innovations have a different logic to that of farmers. Farmers focus on bottom land, whereas projects develop sustainable systems on slopes. Farmers utilize the capital of fertility present in deep soils, whereas projects develop technologies aimed at maximizing biomass production and at restoring this biomass to the top soil. Farmers attempt to maximize the return of their efforts, while projects are mostly concerned by yield maximization or monetary return. But the main mismatch between project’s representations and realities may concern tavy itself. Clearly, tavy is less profitable than most existing alternatives in terms of yield, as well as in terms of labor productivity when it is practiced on degraded land. For this reason, it has often been considered as uneconomical (Styger 2004). But tavy also has three key economic 184
From field enquiries in Beforona in 2002.
270
advantages: it is more secure, necessitates less investment 185 and provides the main staple food. In a context of frequent cyclones, high poverty and limited access to market, these advantages may be determinant. In other words, tavy is not uneconomical but is regarded so because economic analyses are too often limited to a few criteria (yield, monetary return, labor productivity) which are not sufficient to account for the complexity of farmers’ economic strategies. The consequence is that unsuitable or unprofitable alternatives continue to be proposed to farmers even though much research has been implemented in Beforona on this subject. The next section and Chapter 7 will confirm these discrepancies between representation and realities and Chapter 10, in Part IV, will show that this excessive focus on agroforestry and agrobiological technologies is a general pattern, not specific to the case of Beforona.
3. The LDI and PTE projects I stayed in Beforona (in Ambodilaingo) during six months in the year 2001 (Chapter 5) and made short but frequent visits in the commune until 2007. This allowed me to conduct informal interviews with farmers (beneficiary or not) and project staff and to make my own direct observations about the strategy and the impact of the project, who actually benefited from it, and how it was perceived. I will present here a brief summary of the knowledge I accumulated during these six years and from the reading of LDI’s and PTE’s reports. I also hired two students to pass a questionnaire to Kolo Harena associations in 2002. As said earlier, the quantitative results of these enquiries have not been processed yet and will not be presented here,
185
Moor (1998) showed that the average household revenue in Tanambao (a village in the commune of Beforona) is about 140 USD per year. Only 13% of this income is invested (about 15 USD per year), the rest being spent to buy food and first necessity products and to satisfy social obligations.
271
but these students also conducted informal interviews and took some notes which confirmed my own experience and are exploited to draw this section.
3.1. Main objectives According to their annual reports, the primary objective of LDI (2003, 60) and PTE (2004, 19) is: to reduce environmental pressures on the Mantadia-Zahamena forest corridor through an eco-regional approach to agricultural intensification, community based natural resource management, environmental education, conservation enterprise development and farmer capacity building. (LDI 2003, 60)
3.2. The CDIA In Beforona, these objectives are achieved by supporting the research and extension station of Marolafa, now called CDIA-LADIA. 186 CDIA owns 10 hectares of land where trials and demonstration plots are implemented (Photographs 23 and 24). It is equipped with two training rooms, dormitories that can welcome 30 persons, an electricity generator, computers, audiovisual equipment, and a radio for communication with Antananarivo, as there is no electricity or telephone in Beforona. 187 The CDIA offers farmers, nongovernmental organizations, and other development projects the opportunity to see first hand an integrated farming system that provides sustainable alternatives to tavy that improve agricultural 186
CDIA is the name that was used under the LDI and PTE project. It means “ Centre de Diffusion pour l’Intensification Agricole ” (Extension Center for Agriculture Intensification). LADIA is the new name that has been coined in June 2006. The ladia is a plant that serves to make strong ties, and the acronym means “Lapa Ara-Drafitra Ivoaran’ny Ambanivolo” (community house for rural development in Malagasy language). In addition to the LDI, PTE and ERI projects, the CDIA-LADIA had collaborations with the Roger Haus Foundation, Cornell University, the Heidelberg University, the Faculty of Sciences of the University of Antananarivo, the HEPHO (Haute Ecole Provinciale du Hainaut Occidental in Belgium), the LEGTA St-Paul (Lycée d’Enseignement General des Techniques Agricoles de Saint-Paul, La Reunion), the EPSA Bevalala (Ecole Professionnelle Supérieure Agricole de Bevalala), the Peace Corps, and the Madagascar Green Healthy Communities Project (MGHCP). 187 According to the mayor, Beforona should receive electricity in 2007.
272
Photograph 23: Visitors at the CDIA-LADIA in Beforona. A group of farmers and officials visit the center. We can distinguish fish ponds (foreground, left), contour lines (middle ground, left), and buildings for raising pigs and chickens (behind the visitors).
Photograph 24: The CDIA-LADIA in Beforona. Most land is in fallow (picture taken in February 2007, during the agricultural season). We can distinguish trees planted on the upper part of the hill (left) and fallow land.
273
production. 188 In the last year [2002-2003], the CDIA has provided more than 60 technical training sessions that have reached over 800 farmers.” (LDI 2003, 63). In 2006, “83 farmers’ associations, 3209 farmers, 1094 technician, 51 researchers and 160 private operators benefited form the activities of the center. 189 In sum, the CDIA is a training, research, and extension center similar to what is encountered in most rural development projects worldwide
3..3. The Kolo Harena LDI’s approach is also based on support to farmers’ associations, the Kolo Harena, 190 who are now organized into a federation. These associations are expected to be the privileged partner for the implementation of bottom-up approaches: Kolo Harena Federation farmer members meet annually with the University of Antananarivo to define research objectives and assure that the research carried out at the center [the CDIA] addresses the real needs of the farmers. Farmer defined research objectives are oriented towards increased productivity with the focus on promoting initiatives for farming without burning, and protection of the forest corridor. (LDI 2003, 61) We will see later that Kolo Harena are not perceived as legitimate representatives by most farmers. The citation above also reveals that despite the fact that alternatives have to be defined by farmers, they have to be fireless. Farmers have indeed to abandon tavy to be accepted as Kolo Harena members, a rule that was, however, applied with certain flexibility. Support for social organization by the creation of Kolo Harena associations is indeed central in LDI’s strategy. Kolo Harena were supported to create cooperatives that facilitate access to inputs and commercialization, and benefited from a large number of trainings. But LDI’s support also took other forms. Farmer extension agents
188
According to a flyer given to visitors (in 2006), these trainings mostly concern intensive rice cultivation, upland rice with no burning, ecological ginger cultivation (no burning), agroforestry, fish farming and chicken farming. 189 According to a flyer given to visitors. 190 In 2003, 26 Kolo-Harena federations grouped 20000 members, at national level (LDI 2003, 6).
274
(“paysans vulgarisateurs”) and farmer organizers (“paysans animateurs”) have been trained to favor a direct farmer-to-farmer extension of new techniques. In 2003, the LDI and PTE projects had trained, at national level, more than 600 paysans vulgarisateurs and animateurs. Among them, around 400 received a certificate from the Ministry of Environment, Water and Forest, and from the Ministry of Agriculture, Livestock and Fisheries (LDI 2003).
3.4. The ban on tavy Another important element in LDI’s strategy is the collaboration with the Malagasy government to enforce the ban on slash-and-burn cultivation. The 20022003 antifire campaign (Chapters 3 and 5) has been implemented with technical and financial support from USAID funded projects MIRAY, LDI and PTE. These projects provided communication materials, organized workshops, identified green communes and provided support to these communes (MIRAY 2004; PTE 2004): With USAID support, the MIRAY team was mandated to support the campaign in collaboration with the Ministry of Environment, Water and Forest (MinEnvEF). (MIRAY 2004, 19) By the 2002 year’s end, 11,973 posters (format A2) and 24,995 (format A3), 79 banners, 28,890 stickers, 15,130 animation manuals, 9,700 legislation texts, 3,893 campaign fact sheets, 3,543 President speeches, two radio spots, and two TV spots were produced and disseminated at the national level. (MIRAY 2004, 20) The presidential decree (Gouvernement de Madagascar 2002) and the presidential message broadcast on radio did not employ the term tavy. They spoke of the necessity to control fires, not to eradicate them, and planned to offer a premium to “green” communes that would not be invaded by fires. The PTE and MIRAY projects, as revealed by their annual reports, switched the focus from bush fire control to tavy eradication. Several passages of the LDI and PTE reports reveal this strategy.
275
For example, the arrest of farmers that practiced tavy, in Beforona Ambatovolo, was perceived as giving opportunities to achieve more results and was encouraged: Since August 2002, the new administration has taken unprecedented steps to strictly enforce laws and regulations against bushfires, which considerably facilitate our task, broaden the impact of our actions and encourage rural communities to actively seek alternatives to tavy. In that context, we teamed up with the Ministry of Water and Forest, local communities, and ten communes in our zones of intervention, to promote a campaign against tavy and launch a small-scale reforestation program. Those communes will soon be certified as “green communes.” This initiative deserves special attention and should be replicated at a larger scale, because it demonstrates how a drastic change in governmental policy can induce a much more conducive framework and lead to productive partnerships that can effectively reduce bushfires and tavy. (LDI 2003, 6) Another passage reveals that repression is viewed as a way to alleviate the effect of traditions, which seem to be regarded as an important constraint to change: Since July 2002 when the government arrested six farmers practicing tavy, the pressure has diminished considerably, and in spite of their tradition, the farmers are now beginning to more intensively exploit fallow lands. (LDI 2002, 73) Economic constraints to change are however also recognized: The disadvantaged parts of the population are by far the most affected by this situation [by the repression]. Their lack of financial and land resources has led them to practice tavy because this is the only way they can make ends meet. The gap between the advantaged and the disadvantaged has widened in the last year because of the ban. (LDI 2003, 73) This statement could have led to questioning the validity of an approach that uses repression as a mean to facilitate adoption and may conduct the poorest farmers to more poverty. Instead, the report continues with a note of optimism: The farm families that join the Kolo Harena associations have not been adversely affected by the ban; their production improves (ginger, bananas, coffee, irrigated rice), they have access to agricultural inputs and equipment, and their investment capacity is increasing (resulting in the purchase or rent of new land, or hiring of labor), their influence is dominant in the federation, and they tend to dominate the local market. In light of the tavy ban, there has been an increase in the number of requests
276
to join and to set up new Kolo Harena associations, where farmers are enthusiastic about learning methods to reclaim abandoned upland areas. (LDI 2003, 73) But this optimism fails to hide the risk of favoring a minority of competitive farmers that can afford to adopt alternatives while the majority would just suffer from the ban. The minority could in the end be empowered by controlling the local economy, as the excerpt itself states. Remembering the problematic of marginalization evoked in Chapter 1 and the situation in remote villages (Chapter 5), one can wonder if this dynamic could contribute to alleviate poverty and reduce pressures on primary forests. The repressive campaign provoked significant protest in the region, leading to the organization of a provincial workshop in the district capital Moramanga, on August 8, 2003. Representatives of all communes were invited to talk. The Ministry of Environment, the Senate and the National Assembly vice presidents, the Toamasina Province President, the deputies and the senator of Moramanga attended the workshop. There were more than one hundred participants and it was asked that one farmer of each commune speak. The large majority of them (about nine out of ten) asked for an authorization to practice tavy in fallow land. Their interventions were always followed by strong applauses. They argued that they did not have sufficient bottom land and that the cropping systems with no burning were not profitable. They also expressed distrust about the technicians working for projects. They asked for the building of infrastructures: dams and canals for irrigation and the creation of new roads to market their products. A representative of the Kolo Harena association also spoke. He encouraged farmers to stop doing tavy and to apply the techniques proposed by technicians. His intervention was disturbed by hecklers and he was forced stop his talk before finishing, demonstrating that the Kolo Harena were not legitimate representatives of tavy farmers.
277
Eventually, the decisions taken during the workshop were the following: •
Communities should ask authorizations to do tavy in fallow land and the Ministry would send staff for approbation. But in all cases, clearing on steep slopes and on riversides was forbidden.
•
The ministry would support land titling and would send agriculture technicians.
•
Irrigated rice cultivation on terraces should be encouraged.
•
Some local committees should be constituted to control fires.
•
In the long-term, the forest corridor should be restored from the fallow vegetation.
Being present at the workshop, I had the sense that the requests by the farmers had been heard and that an opening to the practice of tavy on fallow land was about to occur. Many participants had the same optimism. This did not happen, however. A provincial order outlawing fires in the Toamasina Province was issued after the workshop. 191 This contradiction is typical of what frequently happens in workshops. The participatory ideal justifies the organization of workshops and meetings with local and regional stakeholders but these workshops mostly serve to legitimize decisions that depend on centrally designed policies. The issuance of the Provincial order can more easily be explained by the lobby of conservation organizations and development projects than by the workshop’s outcomes. The fact that the PTE project “Assisted the 191
Direction du Département chargé de la sécurité civile et de la conservation de l'environnement (2003). According to article 2 of this order, the use of fire is strictly forbidden on the whole Toamasina Autonomous Province, on the littoral plains zone, the hill zone, the escarpment zone, and along streams and rivers. The burning of detritus (herbs and branches) without protection and watch over, and without taking account of the wind direction, is also forbidden. Translation by Pollini. Original text in French: “Sur l’étendue de la Province autonome de Toamasina, l’usage du feu, sur la zone des plaines littorales, les zones de colline, zone des hauts massifs, ainsi que la zone située aux abords des rivières et fleuves, est strictement interdit. L’incinération de déchets (herbes, branchages) ou écobuage sans protection, sans surveillance et sans tenir compte du vent est également interdit.” Some contradictions appear in the text but the main message is a strict ban on all types of tavy, whereas the national legislation authorizes the practice of tavy on herbaceous vegetation (Gouvernement de Madagascar 1960).
278
provincial government in drafting and eventually finalizing a provincial decree outlawing forest fires” (PTE 2004, 27) provides arguments in this sense. The order was, however, never enforced in Beforona, showing that realities were eventually stronger than decisions taken from misrepresentations. One decision, however, was eventually translated into action: the restoration of the forest corridor (Section 4). This decision is also the only one that is at odds with the outcomes of the discussions that animated the workshop. In sum, the LDI, PTE, and MIRAY projects used two strategies expected to act in synergy. On one side, repression was viewed as a way to discourage farmers from practicing tavy. On the other, training and communication campaigns were designed to convince farmers of the interests of the new systems. A behavioral change was expected from the combination of these two elements. This approach is not consistent with the outcomes of Chapter 5 and the conclusions of Messerli (2003). There is much doubt about the profitability of fireless alternatives and investments are necessary to adopt and develop more intensive systems. We further saw in Chapter 5 that the antifire campaign was suffered like a calamity by vulnerable people. The supporters of tavy repression forgot that projects operate mostly along roads, whereas the presidential message and information about the arrest of farmers have been heard in all villages, regardless of their distance from roads and markets. Due to this asymmetry, it is wrong to pretend that development activities and anti-tavy campaigns act in synergy. The repression campaign may have had a high price indeed, both politically and economically. I did not meet one single person of the Malagasy administration that considered it realistic and wise. Only the lobby of conservation organizations, if not of the whole aid system, can explain such a radical action. One could wonder why such importance is given to fire suppression in the area. It may be that biodiversity conservation and erosion control are not the only explanations. A new “actor” may be
279
on the verge of powerfully shaping development discourses and practices around the forest corridor: carbon. The last section of this Chapter, about the VMCRCC carbon sequestration project, will provide arguments to this thesis. Economic aspects and the necessity to support farmers in making investments required for intensification have, however, also been tackled by the LDI and PTE projects, as I will now show.
3.5. Economic support Economic aspects have mainly been addressed in three ways: support in accessing credit and other sources of funding, support in accessing markets, and support for private entrepreneurship (eco-enterprises) and ecotourism. Beforona has mostly been affected by the two first options. LDI helped the Kolo Harena cooperatives to establish commercial contracts, mostly for the sale of ginger. It also helped farmers to obtain funding through the OTIV 192 system and the PSDR 193 project, mostly for developing ecological ginger cultivation, livestock husbandry and vegetable cultivation in a few cases. I did not collect specific information about these grants in the case of Beforona. But for the overall LDI project (activities mostly implemented in the Toamasina and Fianarantsoa provinces), 143 farmers have been helped during the year 1998-1999, mobilizing 16,500 USD at OTIV. This allowed the cultivation of irrigated rice using intensified and sustainable techniques on 242 hectares of land. A second request of about 2,700 USD “helped 68 farmers produce upland beans on 48.3 hectares using anti-erosive and soil fertility improvement technologies that will stabilize the need for continued slash burn agriculture on the hillsides” (LDI 2000, 4). 192
In Malagasy: Ombona Tahiry Ifampisamborana Vola. A Malagasy NGO that provides rural credit. Rural Development Support Project. In French: Projet de Soutien au Développement Rural. Project financed by The World Bank for supporting the Malagasy Rural Development Action Plan (PADR).
193
280
About 90,000 USD have further been mobilized, from OTIV and from the Bank of Africa, to fund 45 community granaries. During the year 2002-2003, OTIV loans “totaling 158,313,350 FMg 194 were disbursed to 218 Kolo Harena farmers beneficiaries [110 USD per beneficiary] with 60% of these loans funded by the revolving fund, and the remainder as new loans.” (LDI 2003, 73) Concerning the PSDR, this project accepted 18 grant requests during the year 2001-2002 (LDI 2002, 27). In the year 2002-2003, with assistance from LDI staff, the Kolo Harena Federations begun submitting requests to the World Bank-funded PSDR project for grants involving as much as 60 million FMG [about 9200 USD]. These grants were destined to help the farmers develop new, or expand existing production activities (e.g., ducks, peanuts, fish farming) which are implemented with LDI and farmer extension agent support. Grants approved thus far to Kolo Harena total 20 billion FMG [about 3 million USD]. (LDI 2003, 11) Considering this general information and the few examples given in the reports (not specifically from Beforona), it seems that a limited number of beneficiaries benefited from most of the aid, and that the amount of PSDR grants represented a large sum in comparison with the economic level of slash-and-burn farmers. Moor (1998b) calculated that households invest about nine to 13% of their revenue, which represents 22 USD for an annual revenue of about 175 USD. 195 The farmers supported by Messerli (2003) had an annual benefit 196 ranging from 700,000 to 900,000 FMg (100 to 140 USD) and invested 13 to 41% of this benefit into agriculture, which represented about 10 to 40 USD per year. These amounts appear to be insignificants at first glance but actually allowed farmers to significantly improve their livelihood. This even sometimes resulted in jealousy from the rest of the community (Messerli 2003).
194
About 24,000 USD. 1138,000 FMg. 196 I do not consider speculation on rice price, which allowed one farmer to obtain a much higher benefit. 195
281
In comparison, the amount of OTIV grants may be adequate only in the case of richer or less poor farmers. For others, they may represent amounts they could not reimburse, especially if a cyclone, a flood or any other natural calamity were to destroy their harvest. This would be particularly true in the case of remote villages having no access to market, such as Ambodilaingo. The PSDR grant represented a still more elevated amount. Several thousand USD were usually granted to Kolo Harena associations 197 which grouped less than ten households (in the case of Beforona). However, loans were nonreimbursable, meaning that PSDR grants represented no risk and could even be used for nonprofitable activities. PSDR loans may have indeed been the main incentive for the creation of Kolo Harena associations. But farmers with more education and economically better off, often migrants that never practiced tavy or were never dependant on this land use, were the first to benefit from the loans. Even nonfarmers (teachers, employees from projects, etc…) received nonreimbursable grants from PSDR. Certainly these people deserve support because not being poor does not mean being rich in Beforona. But if the objective of the projects was to reduce tavy and poverty, their economic empowerment may be counter productive, as we will see later. Concerning the smaller loans made by OTIV, they may, however, be appropriate if properly managed. But they remained inefficient because they were 197
To give a few examples viewed in the LDI and PTE reports (not specifically from Beforona): the Hery Miaradia association in Mahanoro received 43 million FMg from PSDR (about 6,600 USD) for developing duck farming; the Vonona Tsarahonenana in Fenerive Est received 25 million FMg (about 3,800 USD) from IPPTE (Heavily Indebted Poor Countries Initiative of the World Bank, a fund managed by the Malagasy government) for reforestation); The TLA association in Amboditonononana received 25 million FMg (about 3,800 USD) from CEQUIP to buy a huller; five cooperatives and a COBA received 135 million FMg (about 20,000 USD) from IPPTE for reforestation (PTE 2004a, 20); two associations (Vorontsaradia-Ambonivato and Miavotena–Mahanoro) received nonreimbursable funding totaling nearly 65 million FMG (about 10,000 USD) for the acquisition of an improved still and a huller (from various LDI and PTE reports). In Beforona, according to local informants, the first association that benefited from a PSDR grant received 55 million FMg (about 8,400 USD) to grow “ecological ginger” (ginger using fireless alternatives).
282
conditioned to the adoption of fireless techniques, mostly for ginger cultivation, which did not prove to be profitable as we saw in the previous section and as we will see again later. Many farmers had difficulty repaying the loans and complained about the low profitability and high risk of these “supports.”
3.6. Technical support In this section, I will review the techniques proposed to farmers by the LDI project. I will use as a starting point the CDIA center, which contributes to the development of these techniques and offers demonstration plots. The small watershed where CDIA is located shows a zoning from its upper to its lower part. Up the hill are trees that have been planted by the CTFT project in the early 1970s. They mostly belong to the genera Pinus, Araucaria and Agathis. Under this woody crown, banana trees have been planted in association with coffee and a few fruit trees. These home gardens show some differences as compared with those of local farmers: they are not invaded by weeds or invasive species such as Rubus mollucanus, whatever is the period of the year. As a consequence, their yield is certainly higher than in traditional systems. But we saw that high yield is not the priority in the area. The goal of farmers is to maximize their production at a farming systems scale and labor productivity is the main criterion to achieve this. We saw in the case of Ambodilaingo that home garden maintenance can be done at any time of the year. Due to this flexibility, the opportunity cost of labor in home gardens is very low. If cleaning and weeding would be done several times a year, this advantage would be lost and the system would become less profitable. Looking down the home garden, hedgerows planted on contour lines delineate plots either cultivated or in a state of herbaceous fallow. These improvements may
283
represent a significant investment in terms of labor for their maintenance. A permanent cultivation of ginger, beans, rice, and other crops in the alleys between the hedgerows may be necessary to amortize these investments. Ginger is the main cash crop promoted by LDI to achieve this and is sometimes seen growing at CDIA. The project proposes fireless techniques (“ecological” ginger), which are supposed to be more profitable than traditional systems due to higher yields: Organic ginger and improved rice cultivation have been the centerpiece of agricultural intensification activities [in Beforona]. (LDI 2001, 63) [Ecological ginger yields] from 7.5 [to] 14 T/ha (traditional ginger yields ~ 5 tons/hectares), and even though it requires increased labor and financial investment (the use of compost, limited chemical fertilizer and hillside stabilization techniques such as contour vetiver planting), it is very popular along the eastern side of the forest corridor. (LDI 2003, 74) Farmers contest these yields, but more importantly, they explain that higher yield does not necessarily result in more production because the new techniques also demand more labor and they can “only apply the technique on a small plot.” As a farmer extension agent of Beforona explained, Fire is doing much work for us. It clears the land and kills the insects and the weeds. So if we do not use fire, we have to apply many supplementary techniques to clear and fertilize the land and to control the pests. Each of these techniques takes much time and then we cannot cultivate a large area. So I apply the new techniques on a small piece of land for demonstration and I continue to practice traditional systems on the rest of my land. Sometimes I also use 80% or 90% of the technique. But what really takes much time is the preparation of compost. This I really cannot do. 198 This assertion is consistent with Messerli’s (2003) review of agrobiological alternatives to tavy. The supposed high profitability of ecological ginger is also contradicted by the fact that this crop is grown only sporadically and on limited
198
A farmer in Beforona, 2007. Other testimonies from farmers have been received during enquiries in Beforona (2001 and 2002): “those that are not members and continue to do tavy actually live better than us”; “the technique of LDI can produce much, but they need more work and we can implement them on a small piece of land only.” “We do not have the means to implement these techniques. We have been trained but we do not get any support for implementing the techniques after the training.”
284
surface in the CDIA itself. This absence could, however, be explained by a reason other than a lack of profitability. According to CDIA staff, organizational, and management issues would be the cause of this situation. Management decisions are taken along a chain of actors that goes from the center to the project’s regional office, and from there to the head national offices, which itself has to answer to expectations from the donor and from its headquarters in Washington, D.C. This chain of actors reduces the array of agricultural technologies that the center can implement and slows the decision process, resulting in difficulty getting cash at the right time and following the agricultural calendar. In other plots of the hill sides, cinnamon trees are planted but are not yet in production. Small paddy fields are encountered in the bottom land, which is quite narrow. A small cement dam and a canal divert water from a small stream for irrigation. Two types of rice cultivation systems are implemented: SRI 199 and SRA. 200 SRI is an innovative rice cultivation system that can achieve high yields without requiring chemical inputs (Stoop et al. 2002). It is based on a simple principle: whereas conventional systems imply a permanent inundation, SRI necessitates alternately inundating and drying the field, with a weekly frequency or more. This provokes an oxygenation of the soil that strongly benefits the root system, allowing rice to express its total biological potential if other factors are not limiting. Many thallus can be obtained from a single stem and more spikelets are born by each stem. The rice can be sowed at larger spacing and with just one seed per seed hole, which allows saving many seedlings. Yields of 10 tons per hectare and more can be achieved. Soil oxygenation is higher, which stimulates biological activity and makes
199 200
Intensive Rice Cultivation System. In French : Système de Riziculture Intensive. Improved Rice Cultivation System. In French : Système de Riziculture Améliorée.
285
nutrients more available and better preserved in the long-term. SRI works with all varieties of rice and does not require “modern” inputs such as fertilizers or machines. The diffusion of SRI in Beforona, however, faces several problems: farmers criticize it for the high labor input it requires. The reason is simple: the oxygenation of the soil also benefits the weeds. The main function of the water table in rice cultivation systems is actually not to feed rice in water, but to eliminate weeds. It is true that water stress is detrimental to rice and that the natural environment of wild rice is wetland, but a permanent inundation is not necessary to cultivate rice, including in the case of bottom land varieties. But in the absence of inundation, SRI plots have to be weeded three or four times (some technicians recommend weekly weeding). The consequence is a lower labor productivity that causes concern to farmers, as for ecological ginger. I did not find data to verify this because most evaluations focus on yields, but farmers directly experience this constraint. As they put it, “we do SRI on a small plot and traditional rice on the rest of the land.” 201 They try it rather than adopt it, and abandon it as soon as projects leave, as shown by Moser and Barrett (2003a; 2003b). SRI may also be riskier because transplanting from the nursery is done at an earlier stage (8 to 15 days after sowing), which renders the plot more vulnerable to flooding and cyclones. A perfect control of the water table is necessary to achieve success. This risk factor may indeed be as important a constraint for adoption as low labor productivity. If everything goes right, SRI may also be profitable in terms of labor productivity because high yield could cover the supplementary work. But cyclones, flooding, pests, and diseases can annihilate the efforts a farmer invests in
201
All farmers practicing SRI that I have visited until now (about 6, in Beforona and Ikongo) made exactly this same assertion. Discussion with project staff shows that they experience the same statements.
286
SRI. As I showed in Chapter 5, the objective of vulnerable farmers is to produce more in the worst years, not to produce more in the best years. SRI’s rational is to express the complete yield potential of rice. This implies that there are no other limiting factors, which may hardly be the case in Beforona. For example, it is doubtful whether the soils in Beforona are sufficiently fertile to allow high yields and amortize the high labor input invested in SRI. Eventually, SRI’s high yields have to be verified in the long-term. A significant organic fertilization may be necessary to compensate for the exportation of nutrients when very elevated yields are obtained. The LDI and PTE projects were aware of these limitations and proposed another pathway for irrigated rice intensification: the SRA. SRA is based on more conventional techniques such as transplanting in rows, using varieties with higher yield potential and doing an earlier transplantation. Its diffusion started during the first republic (and maybe earlier), through support by the CTFT project. Some ingredients of the SRA package are sometimes adopted by farmers. The main advantage they notice is that new techniques (both SRA and SRI) allow saving seedlings due to a larger spacing and a lower number of seedling holes. SRA extension thus certainly contributed to the progress of rice cultivation in the area. However, the labor costs of the new techniques are also criticized, although to a lesser extent than in the case of SRI. The main work constraint results from transplanting in rows (this also applied to SRI). Rows allow mechanical weeding, which saves some time, but on the other hand, more time is needed for transplanting. Furthermore, weeding machines are not always available, are expensive with regard to the investment capacity of most households, and the labor gain for weeding does not necessarily compensate for the labor cost of transplanting in a row. In order to compare these two tasks, labor has to be compared considering its opportunity cost. This has not been done yet.
287
The CDIA further developed livestock husbandry activities. It raised improved breeds of pigs and chickens but these activities were not profitable due to the high cost of food supplements. They were abandoned in 2002 but restarted later. The CDIA also created fish ponds but their production was low or uncertain for reasons that are not yet clear, and they necessitated labor investments that farmers cannot easily afford. Here again, a calculation remains to be done to know if the profit compensates for the investments, and to compare the profitability of fish ponds with other activities. Farmers have done their empirical “calculation” already and are reluctant to create fish ponds on their land. Fish ponds are even the subject of jokes and those who adopted this activity complain about the low production. All these techniques are supported by training. CDIA proposes training programs about agroforestry (contour strips, alley cropping, and improved fallow), agrobiological agriculture (mulching, zero tillage), home garden intensification, “ecological” ginger cultivation (fireless techniques), horticulture (litchis and vegetables), SRA and SRI, compost production, and swine and chicken husbandry. When beneficiary farmers are asked about their interest for the training they receive, they almost unanimously answer that the topics are interesting but that they usually cannot apply the proposed techniques because they do not have sufficient time or financial resources. The situation I described is indeed common in agricultural research and extension centers. The limits of this type of approach have been already analyzed and criticized in depth. The CDIA, similar to other centers of the same kind, is isolated from the farming systems it is supposed to support. In spite of this, the CDIA is the subject of many newspapers articles that speak in praise of its achievements. 202 It may 202
Articles in Midi Madagascar, July 28, 2003; Madagascar Tribune No 4414, July 28, 2003; L’express de Madagascar, July 29, 2003; Le Quotidien de Madagascar, July 20, 2004; L’express de Madagascar, September 24, 2004; Midi Madagascar, May 30, 2006 ; Les Nouvelles, May 31, 2006 ; La Gazette, May
288
be possible that its implicit function is the demonstration of the project’s achievements to journalists, rather than to farmers. Representatives of the Malagasy government, of the donors, and even of citizens from donor countries (American congressmen) come to visit the center and leave with an enthusiasm that contrasts strongly with farmer’s criticisms and complaints, as we will now see.
3.7. LDI as perceived by the beneficiaries When asked to say what they think of the LDI and PTE projects, people in Beforona speak of excessively frequent training that makes them lose time while providing them with techniques they cannot apply, and complain about the fact that the project requires the abandonment of tavy. As they, the “beneficiaries” of LDI, often put it, “those who still do tavy are less poor than us, because the alternatives we adopt cannot be applied on large surfaces.”203 LDI is accused of lying to farmers or of “just talking about alternatives without even trying to implement them” (a farmer in Antandrokomby in 2002). Fireless alternatives are criticized for the excessive work they require and the increased risk of pest damage they lead to. Beside these issues, “beneficiary” farmers recalled a few other problems: they had been promised certain prices for ginger sales but the price had eventually been lower; seedlings arrived late with regard to the agriculture calendar; the management rules of community granaries had unexpectedly changed and the profit farmers hoped to receive were eventually directed to Kolo Harena federations; some farmers complained that they were obliged to buy coffee seedlings. These problems have
31, 2006 ; Madagascar Tribune, June 1, 2006 ; Le Quotidien, June 1, 2006 ; La Gazette, June 3, 2006 ; Les Nouvelles, June 3, 2006 ; Midi Madagascar, June 3, 2006 ; Madagascar Tribune, June 3, 2006. 203 Many testimonies, coming from the enquirers I hired, from my own informal enquiries, from discussion with project staff, include this statement. I faced exactly the same statements during visits of the LDI project in the Fianarantsoa Province (Ikongo and Ranomafana) and of the PDFIV project in Tsinjoarivo (Ambatolampy district).
289
sometimes been the cause of anger, but can be regarded as minor issues because they do not put into question the strategy of the project itself. They just illustrate the logistical and communication difficulties encountered by any rural development project. But in a context where the local elite receives the main benefits, these small mistakes have an elevated cost in term of discontentment. The concentration of the aid in the hands of a local elite may indeed be the main pitfall of the project. As we saw, the Kolo Harena associations that received PSDR grants are often headed by traders and product collectors who are not dependant on tavy for subsistence or who never even practiced tavy. Key positions such as the presidency of the local federation are held by the most prosperous traders. This category is indeed the only fraction of the beneficiaries that expressed satisfaction over the program with no ambiguity. It benefits from access to significant nonreimbursable loans (PSDR), from a better commercialization, and maybe from a better social and political status in the commune. The rest of the population usually sees these people as opportunists, not as their representatives. This support of local entrepreneurship is a laudable objective but is not consistent with the objective of supporting tavy farmers and alleviating poverty. The project argues that if rich farmers adopt new techniques, the poorer will follow, but this logic did not prove to be true. The empowerment of the most competitive farmers lead them to fill market niches and to lower the prices of cash crops to the detriment of farmers living in remote areas, as we saw in Chapter 5 in the case of ginger. As rich farmers appropriate the land most suitable for intensification, their empowerment may contribute to the marginalization of the poorest ones, with both environmental and social negative consequences. The problem of inadequate target has been recognized by the project itself. The ERI program is now committed to avoiding excessive focus on what he calls “false
290
farmers” (Staff from the ERI project, 2007). It seems that only the neighboring commune, Andasibe, is still concerned by this problem. Kolo Harena leaders, however, recognize themselves that targeting the poorest is still a challenge. Eventually, people in Beforona often put forth support to irrigation as a key strategy to develop the commune and complained about the absence of support at this level. Even project staff complained that LDI neglected this activity (only one dam has been built with LDI’s support, irrigating about 10 hectares). The reasons behind this neglect are the lack of budget for this activity, the fact that the construction of irrigation schemes is funded by the World Bank Project FID, 204 the elevated cost of cement dams and the limited potential of the commune for irrigated rice cultivation. Concerning the elevated cost, Chapter 7 will show that dams are actually quite cheap. Concerning the area suitable for irrigation, the Mayor established, in 2002, a preliminary list of 60 sites that could be equipped or repaired, for a total of about 260 hectares. It must be verified whether the improvement of these sites is actually feasible, but Chapter 7 will show that this number is not unrealistic if the case of Ambodilaingo could be extrapolated to the whole commune. If rice fields in terrace could be created in addition, the impact could be significant.
3.8. Conclusion (LDI and PTE projects) In conclusion to this section, the LDI and PTE projects focused excessively on training and on techniques, whereas farmers expect mostly economic support to implement the new techniques. The proposed alternatives imply more risks and necessitate investments that most farmers cannot provide, while the more realistic alternatives, those already known by farmers, are overlooked. Modalities of funding lacked flexibility, consisted of excessively large loans that farmers had difficulty 204
Social Fund Project. In French : Fonds d’Investissements pour le Développement.
291
repaying and concentrated nonreimbursable grants in the hands of a better off minority. Excessive importance was given to repressive policies whose overall effect on livelihood was destructive, especially in remote areas. As a result, LDI and PTE may have ended up empowering a rural elite which has never been dependent on tavy to subsist, and which may be in an unequal competition with tavy farmers to seize market opportunities and to access to the most fertile lands, those suitable for alternatives if the required investments can be done. These dynamics, together with the negative impact of the anti-tavy campaign, may have increased the vulnerability of poor farmers. I did not prove these facts but a large bundle of evidence renders this interpretation of the reality much more defensible than the optimistic picture produced in the LDI’s and PTE’s reports, such as the belief that fireless alternatives and PSDR grants will allow the development of the area. The consequence is that most farmers in Beforona complain that the LDI project “just helps the rich” (several farmers in Beforona 2002) and “does not propose viable alternatives” to tavy (several farmers in Beforona 2002). It is true, however, that agriculture has developed in the commune over the last decade, but this may result more from the increasing pressure on land, which forces farmers to develop more intensive systems (usually other than those proposed by LDI). The development of cash crops, such as ginger and bananas, also played a role by providing opportunities to increase incomes.
4.
The
Ankeniheny-Mantadia-Zahamena
Corridor
Restoration
project
(AMZCR) In this section, I will present the Ankeniheny-Mantadia-Zahamena Corridor Restoration project (AMZCR), currently in its early implementation in the neighboring
292
communes of Ambatovolo and Andasibe, west from Beforona. This project is aimed at reestablishing connectivity between corridor fragments located north and south from National Road 2. It will be partly funded by the sale of carbon newly stored by reforestation. It is the first attempt to implement the Clean Development Mechanism in Madagascar and reveals the rural development and conservation strategies that could prevail in the corridor in the future.
4.1. Context With the signature of the Kyoto Protocol and the implementation of the Clean Development Mechanism, carbon has become a commodity that can be bought and sold on the world global market. Land use change and forestry, if they lead to an increase of carbon stocks, create opportunities for transactions on this market (Brown, in press; World Bank 2006). It seems that the idea of using the carbon market to finance conservation and development activities in Madagascar was first expressed in the mid 1990s, concomitantly with an increasing will to reestablish connectivity between fragments of the eastern rain forest corridor. A workshop about the Clean Development Mechanism was organized in Antananarivo in October 2001 (IRG 2001). Surveys were launched to assess the potential of Malagasy forests to store carbon. Rarivoarivelomanana (2001, 15) showed that the lowland rain forests store 120 to 450 tons of carbon 205 per hectare, whereas the middle altitude forests store 80 to 310 tons/ hectares. The Wildlife Conservation Society launched the Makira Forest Project (Meyers 2001; Meyers and Berner 2001), partially funded by the sale of emission offsets on the voluntary carbon market. But the AMZCR project is the first attempt to use the Clean Development Mechanism. It was prepared with support from the nongovernmental organizations Conservation International (CI) and Tree for 205
One ton of carbon is equivalent to 3.664 tons of CO2.
293
Life. 206 It would last 30 years and would be worth eight million USD, including 3.2 million USD from the BioCarbon Fund of the World Bank. 207 The main other donors are IDA, 208 GEF, 209 USAID, CI through its Global Conservation Fund, the Climate Trust, 210 and Dynatec (a private mining company) (UCFB 2006). Conservation International has been working in the area for over a decade. It collaborated with ANGAP 211 in the early 1990s to implement the ICDP212 approach in the Zahamena National Park, during the first phase of the NEAP. It further implemented the ESFUM 213 component of the second phase of NEAP, with the USAID funded MIRAY project (MIRAY 2004), from 1998 to 2004, and collaborated with the projects LDI, PTE, ERI, and FSP-ADRA, which, together with their partners, constituted the Toamasina USAID eco-regional alliance. These USAID funded projects are expected to provide the main knowledge base and capital of experience for the implementation of the AMZCR project, meaning that a continuity can be expected between the activities proposed by LDI and PTE and those that will be proposed by the AMZCR.
206
This NGO “promotes the restoration of the world's degraded ecosystems as the most important task facing humanity in the coming decades” (http://www.treesforlife.org.uk/about/aboutus.html). 207 The Biocarbon Fund, managed by the World Bank, serves as an intermediary between carbon buyers and sellers. See World Bank (no date) for more details about this fund. 208 International Development Agency, a division of the World Bank dedicated to the poorest countries. 209 Global Environmental Facility. 210 “The Climate Trust is a leading nonprofit organization dedicated to providing solutions to stabilize our rapidly changing climate.” (http://www.climatetrust.org/) 211 National Agency for the Management of Protected Areas. In French : Agence Nationale pour la Gestion des Aires Protégées. 212 Integrated Conservation and Development Project. 213 Multiple Use Forest Ecosystems. In French: Ecosystèmes Forestiers a Usages Multiple.
294
4.2. Presentation of the AMZCR project The AMZCR project has three objectives: to reestablish connectivity between fragments of the corridor, to create carbon sinks and to develop sustainable cultivation systems as alternatives to slash-and-burn. It would achieve these objectives through three components (UCFB 2006): •
Component 1: Restoration of the forest corridor (3020 hectares of reforestation with native species) along National Road 2, between the Mantadia and the Vohidrazana forests (Map 13).
•
Component 2: Protection of 405 000 hectares of natural forest, by avoiding the deforestation of 80 000 hectares over the next 30 years, between and around the Mantadia and the Zahamena National Parks.
•
Component 3: establishment of forest or sustainable gardens (928 hectares), fruit gardens (334 hectares) and firewood plantations (661 hectares) (Map 13).
Component 1 is eligible for the clean development mechanism and will be partly funded by the Biocarbon Fund. Component 2 (avoiding deforestation) is not eligible at present, 214 but may be supported by the voluntary carbon market and be eligible for a second “window” of the Biocarbon Fund. Component 3 is a support activity designed to avoid leakage: as farmers live in the reforestation area, they may move and clear forest in other places to compensate for the loss of agricultural land in favor of reforestation. Sustainable gardens, fruit gardens, and firewood plantations are aimed at providing food and income to these farmers, in order to allow them to stay in the same location while improving their livelihood.
214
There are currently discussions to render avoided deforestation eligible to the Clean Development Mechanism in a series of pilot countries, and Madagascar could be chosen as one of these pilot countries.
295
Andasibe Mantadia National Park
Target area for components Ambavaniasy 1 and 3
Communes Villages Footpaths National road Fruit gardens Firewood plantation Reforestation Sustainable gardens
Analamazoatra reserve
Beforona
Km
Vohidrazana forest
Ambodilaingo
Map 13. Target area of the VMCRCC project. Source: Conservation International Madagascar. Additions by Pollini (Blue arrows).
296
The Biocarbon Fund of the World Bank accepted the project identification document (Project Idea Note) in November 2004, and agreed to implement the project with the government of Madagascar as a host party (Intention Letter signed in March 2005). A Project Design Document has been prepared but addresses components 1 and 3 only. The area of intervention is limited to an area surrounding National Road 2, close to the Mantadia National Park 215 and the Perinet-Analamazoatra reserve,216 between the communes of Beforona and Andasibe. Its main objective is the reconnection of the south and north parts of the forest corridor along National Road 2 (Photograph 25). This subproject has been named the Vohidrazana-Mantadia Corridor Restoration and Conservation Carbon Project (VMCRCC) and the analysis that follows will be limited to it. The VMCRCC’s Project Design Document has not been validated yet, but carbon sales have already been negotiated by the Biocarbon Fund and are ready to be signed. The Malagasy government has put in place a coordination unit (in April 2005), which includes staff from its central and decentralized services. It also promised to engage 1.5 million USD from its GEF-IDA funding (third phase of NEAP). But until now, only Conservation International’s financial contribution has been disbursed, enabling the start of field activities such as the establishment of nurseries, the identification of plots for reforestation and the training of field partners in new agricultural techniques. Seven local partners are involved (6 nongovernmental organizations 217 plus the Mantadia National Park), but a project manager remains to
215
The most visited national park in Madagascar, managed by ANGAP. 10 000 hectares. A small reserve (810 hectares) that shelters a population of Indri indri, the largest lemur in Madagascar and a threatened species. 217 Mitsinjo (an NGO based in Andasibe), Ecophi (an association of experts and consultants of the Ambohitsaina Faculty of Sciences), MATE (Man and the Environment, an NGO that runs a project in Ambavaniasy, between Andasibe and Beforona), SAF-FJKM (the development and relief agency of the Church of Jesus-Christ in Madagascar), NAT (a German NGO), AGA (Association of the Andasibe Guides). 216
297
Photograph 25: Vegetation in Ambavaniasy. This area is targeted by the reforestation component of the VMCRCC project. The dominant land use is tavy, which results in a landscape dominated by young fallows. Cassava is visible in the foreground and forest patches subsist in the background. The landscape is expected to be transformed into a mosaic of agroforestry gardens, sustainable gardens and restored ecosystems.
298
be recruited to insure the link between the coordination unit and these seven partners. Decentralized collectivities and state services are also involved. Contracts with farmers have not been signed yet but negotiations to prepare these contracts are ongoing. It has been proposed that farmers lend part of their land for reforestation, on a voluntary basis, for the thirty year period of the project. In some cases, they own this land legally, in other cases, their “ownership” is a use-right granted by traditional authorities but recognized as legitimate by the state and the project. In the first case, it may be possible that farmers would be financially compensated for lending their land, but the matter is not clear and still in discussion. In the second case, they would not be directly compensated but would benefit from three advantages: •
The legal recognition of their land ownership through the granting of land certificates by the communes, 218
•
Priority access to job opportunities provided by the project (mostly for the plantation of trees)
•
Support for the development of new farming systems through the implementation of component 3.
Concerning the benefit from carbon sales, it will be captured by the Malagasy government and will serve to finance reforestation operations (including the salaries of beneficiary farmers hired by the project to plant the trees). The key question that arises from this model is whether the advantages proposed by the project will compensate for the disadvantage of losing agricultural land through reforestation. As the main new resource (the profit generated by carbon sale) is captured by the Malagasy government, the fate of farmers will depend on the
218
Land certificates are not exactly land title, but according to the new land law passed in 2006 (Gouvernement de Madagascar 2006), they provide the same rights.
299
capacity of new land use to compensate the loss of usufruct on reforested land. If the alternatives compensate for this loss, there will be a win-win situation and both rural development and conservation objectives will be achieved. If they do not, the VMCRCC project could impoverish already vulnerable communities, leading to outmigration and a significant risk of leakage. In other words, sustainable forestry is in the hand of successful agricultural development. Unfortunately, the capital of knowledge and experience that will be used to design agricultural support may be the one accumulated by the LDI and PTE projects, which does not encourage optimism. In the next section, I will analyze the available documents about this project and will synthesize the results of discussions I had in February 2007 with people working in several organizations involved in it. 219 Section 4.3. will analyze the transmission of technical knowledge, while Section 4.4 will address economic aspects. These analyses are quite speculative as the VMCRCC project did not start many activities, but so are the hypotheses that served for its conception. In the case of projects adopting new strategies, there is no other choice than adopting uncertain hypotheses. Uncertainty can be significantly lowered only once these activities are completed.
4.3. The transmission of fashionable alternatives Several observations indicate that the alternatives proposed to farmers by the LDI and PTE projects will be transmitted to the VMCRCC project.
219
I met with representatives from the Ministry of Environment, Water and Forests, Conservation International, Man and the Environment (a partner NGO working in Ambavaniasy, about 10 kilometers from Beforona), and Mitsinjo (a partner NGO working in Andasibe).
300
First, the preliminary PDD 220 explicitly refers to technologies developed in Beforona as sound alternatives that farmers could implement. 221 Interviews with staff from the MINENVEF and from CI confirmed that this perspective prevails. Much hope is put on agroforestry technologies and on cover crops and zero tillage, despite the fact that these techniques have yet to be proven to be adapted to the region. Field partners, however, did not share this optimism. Staff from the NGO Mitsinjo contended that zero tillage would result in lower yields the first year, equal yields the second year and hopefully higher yields the third or fourth year, in comparison with tavy. Moreover, more inputs would be necessary, such as cover crop seeds and chemical herbicides to eliminate the weeds that compete with the cover crops. Fruit gardens, fallow gardens and sustainable gardens are other systems expected to play a significant role. Unfortunately, no information is available concerning these techniques except in a few training documents of the CTHT, 222 which insist on the necessity of using compost and applying mulch, and on the web site of the “Restore the Earth Project” (no date) of the nongovernmental organization “Trees for Life.” The information provided on this site may be out of date but is the only one available. The site presents the project “Catalysing Rainforest Restoration in Madagascar: Andasibe-Mantadia Corridor project,” which supported the preparation of the VMCRCC project. According to this site, sustainable development would be achieved mainly by the creation of “savoka gardens” (fallow gardens), which would consist of 220
Project design document form for afforestattion and reforestation project activities (CDM-AR-PDD) version 02 (no date). This document is a working draft and its content may change significantly in the future. However, I exploit it in this article because despite the fact that it is a draft version, the project’s implementation has already started on the ground, and because this draft PDD is the only institutional document available for analyzing the strategies that are driving this implementation. 221 See Section F1, social risks, of the PDD. This document is a draft with no pagination. 222 Centre Technique Horticole de Toamasina. In French: Horticulture Technical Center of Toamasina. This center is a public organization which has been supported by aid from the French in the past and is dedicated to the development of fruit and vegetable cultivation.
301
a mixture of local native forest plants and exotic fruit and vegetable plants. The aim… [would be] to transform land that would otherwise be colonized (due to abandonment after rice harvest) by alien invasive plants of no use to people or the rest of native biodiversity, into productive, soil-restoring land for a five year fallow period. (Restore the Earth Project no date, no pagination) The NGO also proposes to establish “permanent community gardens,” which would associate “plants which yield produce over the longer-term (e.g., fuel wood, timber, fruit)” (Restore the Earth Project, no date). Both types of gardens would contain a high proportion of native forest trees and shrubs, acting as a buffer between the corridors and areas of mono-cultures, as well as providing resources which would otherwise be obtained from intact forest.”(Restore the Earth Project no date, no pagination) We can suppose that “fallow gardens” and “permanent community gardens” will constitute the 928 hectares of “sustainable garden” that are visualized on Map 13 (page 296), whereas the fruit gardens would constitute the 334 hectares area. The alternatives developed in Beforona may rather be dedicated to intensification on the part of the land that will not be directly targeted by the project, and the financing for these alternatives remains unclear. If this pattern was confirmed, the target area would consist of a mosaic of reforested land, fruit tree gardens, fallow gardens and sustainable gardens, all having a significant ligneous component. This perspective matches with the importance given to agroforestry by most actors involved in the VMCRCC project and is consistent with the fact that, initially, the carbon stored in sustainable gardens and fallow gardens was expected to be sold on the global market as well. The project eventually abandoned this idea in order to give more flexibility to farmers, which appears to be a wise choice. However, to create a mostly forested landscape linking the northern and southern fragments of the corridor remains one of its main objectives. Farmers will, however, not be obliged to lend their land to the project. But as they say, “we have no choice, because tavy is forbidden now, so we must at least try to
302
get these advantages” (a farmer in Ambavaniasy, 2007). These words may reveal a supplementary explanation for the unexpected outcomes of the workshop held in Moramanga during the fire repression campaign of 2002/2003 and the issuing of the provincial order that forbids tavy in 2003 (Section 3).
4.4. Overlooking the economy There is no doubt that the technologies proposed by the project will achieve great ecological performances, but their economic performances (their capacity to sustain livelihoods) are doubtful. No data seems to be available concerning the profitability of fruit gardens, fallow gardens and community or sustainability gardens, nor about the marketability of their products. Bananas and ginger are the main cash crops in the area but little is said about the importance that will be given to them in the gardens. The technologies proposed by the VMCRCC can only be assessed on the basis of previous experience in Beforona, which does not allow optimism as we saw in Section 2. In sum, the VMCRCC inherited the belief that fires and erosion are the main enemies, which led to the implicit assumption that maximizing biomass production is the main criterion to select appropriate alternatives. In consequence, it developed a blind faith in agroforestry and agrobiological technologies that led to overlook more essential criteria such as profitability. This blind faith is confirmed by the draft version of the Project Design Document, which indicates that “no socio-economic negative impacts are expected from this project” (Draft Project Design Document no date, no pagination). The PDD’s section about cost analysis further asserts that “as the project activity is reforestation without any generation of financial or economic benefits other than CDM related income, simple cost analysis… should be applied” (Draft Project Design Document no date, no pagination). Considering that the project plans to convert about
303
5000 hectares of agricultural land into forest or agroforestry systems whose profitability is doubtful, these assertions are highly disputable. The draft PDD reveals other received wisdoms that can explain the overlooking of economic aspects: There are potential economic risks to members of communities within the project area, especially in the form of “the unknown.” A number of the project activities will be entirely the responsibility of the people in terms of undertaking the development of savoka [fallow] gardens, sustainable gardens, etc. The amount they do, what they do and their time input will be their choice, but it will involve some input of time to activities which are relatively new to them and the benefits of which they will only know for certain through experience. (Draft Project Design Document no date, no pagination). Economic uncertainty is evoked in this paragraph but the problem of “the unknown” is explained by the potential wrong choices made by farmers, while the possibility that the project would make the wrong choice itself is not even mentioned. Another section, about social risks, makes explicit that the project considers that tradition is the main potential cause of wrong choices: The primary social risk to the project was considered to be the possibility that local people will refuse to change their historical use of tavy [slashand-burn cultivation] as their main form of agriculture. Because tavy has been used for generations as the main form of rice production for a large number of the local population, major efforts needed to be made to convince people that it is no longer feasible. (Draft Project Design Document no date, no pagination) It is certainly true that a threshold can be reached where tavy is not anymore feasible. But farmers are already aware of this and what should be cause of concern now is the feasibility of the proposed alternatives. Other sections of the document reveal an insufficient understanding of agrarian dynamics, the excessive optimism about economic impact and the influence of received wisdoms. The authors contend, referring to reforestation, that the income to be generated will be higher than actual of rice cultivation since rice net benefit is decreasing each year due to shorter tavy-savoka
304
cycles, therefore, less productivity of the land and the consequently need to move further into the forest looking for clearing new productive lands. (Draft Project Design Document no date, no pagination). Several aspects are overlooked here. First, production will not necessarily decrease forever because local farmers will not necessarily practice slash-and-burn cultivation forever, as we saw in the previous chapters. Second, the benefits of reforestation will be perceived at national levels (by the Malagasy state that will receive payments from carbon sale) and at the global level (by a reduction of global warming) whereas the costs will be born by local farmers (by restrictions to access to land). In these conditions, it is doubtful whether the “income generated” will actually be higher. It is right that farmers will indirectly receive part of the profits generated by carbon sequestration in the form of salary for planting trees. But this will just last a few years and will be a legitimate payment for their work. It is right also that farmers will be authorized to harvest products in the reforested area. But the main usufruct of the trees (the revenues of carbon sale) will not be directed to them. Leakages (forest clearing by impoverished farmers) would be the consequence of this unequal distribution of profit, not of the balance between the overall costs and benefits of the project. In sum, analyses of available documentation and interviews with staff revealed that the VMCRCC project inherited the received wisdom that dominate the conservation and rural development projects that operated in the area. This may lead to the same result: the proposition of unprofitable or unsuitable alternatives to slashand-burn cultivation while the alternatives implemented by farmers are not supported (Photograph 26). As a result, the project will probably fail to alleviate poverty and to reduce pressure on ecosystems. But in this case, and to the contrary of what happened in Beforona, farmers will lose part of their agricultural land, due to the reforestation component of the project.
305
Photograph 26: Land improvement in Ambavaniasy. Despite the absence of bottom land, paddy fields are a possible alternative to tavy. But significant labor investment is necessary if this option is to play a significant role.
306
4.5. The consequences The consequences of such inefficient actions would be the impoverishment of farmers and carbon leakage. Fires would be ignited as a means of resistance (farmers would sabotage or burn the plantations established by the project). Migration to remote forests not under conservation status would be the last alternative for the most vulnerable households. The reforestation program of the Fanalamanga, in the Mangoro region (60,000 hectares, mostly pine trees) allows these effects to be anticipated. 223 When the plantation was established, the local farmers lost part of their land, mostly pasture land where they raised cattle. Some households benefited from job opportunities from the Fanalamanga and may have improved their livelihood. But when the Fanalamanga entered into bankruptcy, they unsuccessfully claimed their land and ignited fires in pine plantations to support their claim. Other households did not get job opportunities and had to decrease their herds, with negative consequences on their economy. Some moved to the forest frontier where more land was “available” because it was still covered with primary forest. We can meet some of these people farther east, in the commune of Fierenana for example, where they use forest to provide forage to cattle. According to informants in Moramanga, other households would have moved south, to Beparasy. These migrations may have contributed to the advancement of the agricultural frontier to the detriment of the primary forest, whereas the importance of this contribution remains to be estimated. It is true, however, that the VMCRCC reforestation covers a smaller area (3020 hectares) and concerns “only” a few hundred or thousand households. 224 But the project is regarded as a pilot for larger scale operations in the future. 223
The following information was gathered during a consulting work done in 2002 for the Food Security Program of the ADRA NGO. 224 No precise account is available, but it seems that the seven local partners work with a few hundred households each, some of them having all their land in the project area, and some of them having only part of their land in it.
307
A political consequence can also be anticipated: the reinforcement of wrong beliefs that lead to inappropriate strategies, such as the rejection of fires and the taste for high yields. We will see in Chapter 7 that farmers are empiric and not reluctant to change, as soon as they can see that new techniques are profitable and suitable to them. But the beliefs of conservation and rural development organizations are more solid. Projects are under less pressure than farmers to achieve results because it is not their own survival that matters. Moreover, projects depend on donors whose staff cannot invest significant time to assess realities on the ground. The projects’ annual reports are supposed to be the medium that allow donors to be in contact with realities but are often biased by their second function, which is to justify more funding. Moreover, lack of impacts rarely puts projects at risk of being stopped. So, if projects are usually not blamed for their failure, farmers often are. The consequence is that false beliefs are reinforced. Efforts will be increased to convince farmers that they need to adopt new technologies with higher yields, that they have to stop using fires, and that they must adopt zero tillage and mulching technologies, produce compost, plant fruit trees and develop other cash crops for their own good. Coercive conservation policies or “participatory top down” will escalate in the name of biodiversity conservation and sustainable development. Strong misleading beliefs will engender the closure to new forms of knowledge. Farmers will pay the price not only of conservation efforts, but also of our reluctance to face our own ignorance of realities. But all this may have already happened indeed, and the tavy repression campaigns of 2002 and 2003 may be the expression of it.
308
4.6. Some solutions? Critical analyses alone are useless and I must now propose strategies to reduce the risks of negative socio-economic impacts and leakage of the VMCRCC project, in the light of previous chapters. First, obviously, the agrarian system needs to be understood. Future evolution can be anticipated by comparing the region with other areas of the world where similar systems are encountered. As we saw already, the work of Ester Boserup (1965) provides a powerful guideline that helps to understand agrarian system dynamics. It shows that annual cropping, followed by multiple cropping, perennial crops, crop rotation, the integration of livestock husbandry with agriculture and the use of manure are the natural evolution of extensive systems when the frequency of cultivation increases. Green manure, mulching and the association of trees with annual crops occur only at later stages of intensification, when the pressure on land reduces the area available for grazing animals and for planting trees. The Outcomes of Messerli (2003) and Nambena (2004), as well as Chapter 5, confirmed this dynamic. Second, support must not focus on technical aspects. Intensification needs investment to purchase labor force, tools, inputs, and to avert risks. If farmers do not receive economic support, which can be in the form of credit or subsidies, only the most competitive will intensify, yet very slowly, and the others will move to the forest in the search for new land, provoking leakage. Third, farmers have to be supported in the development of their own alternatives. Other technologies can also be proposed but it would be highly risky to invest all efforts in the “unknown.” Endogenous alternatives (irrigated rice, home gardens, rotation on plowed land burned at least the first year) have the advantage to be there already and do not necessitate high investment. Acquiring a pig to raise and better agricultural tools (a hoe, not even a plow), which only represent about 15 USD,
309
can already make a significant difference. Small cement dams can also have a very significant and short-term impact, and energy can be subsidized in the form of food for work operations. These strategies will be presented in Chapter 7. Some have been tested already in the Ambodilaingo fokontany. But in the specific case of the VMCRCC project, the central point is the need to implement a rigorous cost-benefit analysis at farmers’ level, in order to compare the profitability of the new systems with the ancient ones. This analysis is essential because as soon as trees are planted on their land, farmers will have no other choice than to successfully intensify, whereas in ordinary projects such as LDI or PTE, they can reject unprofitable alternatives and continue their usual land use. In order to achieve this analysis, the number of targeted households, the area of land they own, the part of the land that will be reforested, the part that will be dedicated to community garden, fallow gardens and other alternatives, and the labor productivity of each of these systems must be calculated. The level of investments that are required and the availability of such investments, either at farmer level or from external sources must be accounted for too. However, these analyses may not be sufficient because no reality can be exhaustively measured. The complexity of farming systems may have no limit when quantification is undertaken. For example, labor productivity cannot be evaluated just in terms of added value produced per day of work because some farmers cannot pay labor force and only use family labor. In this case, days of work are not all equal, which obliges assessment of the opportunity cost of labor. The effects of the risk factor, of the investment capacity, of credit opportunities, of remoteness, of increasing pressure on forest land, of social differentiation, of competition and of culture and habits, which are by far less
310
significant than commonly believed but nevertheless do exist, are also difficult to quantify. In the face of this complexity, another option would be to limit the analysis to qualitative aspects and to adopt an adaptive management approach (Lee 1993; Lee 1999; Johnson 1999) where agroforestry, mulching, and fire suppression would not be proposed a priori. The VMCRCC project would become an experiment, which it is already indeed as it is the first attempt to implement this approach in Madagascar. But it would become a cautiously run experiment. Several qualitative comprehensive analyses have been made already: in this dissertation and in many other works about the area, such as Messerli’s (2003) and Nambena’s (2004) dissertations. They could serve as a basis for designing this adaptive management approach. Fifth, part of the carbon profit will have to be directed straight to beneficiary farmers to compensate for the loss of agricultural land. Only by doing so can the project be legitimate in the long-term because farmers will see the reforested land as their own, which it will actually be as they will be delivered land title. If the trees do not provide them with significant income, they will see this land as an area for future extension of their farmland, for their children if not for themselves. Rather than encouraging farmers “to do nothing,” 225 the subsidies would be consistent with the support for intensification because it would increase the investment capacity of farmers. The implementation of this approach would confront the belief that poor farmers are not able to manage cash. It is true that households have obligations of social redistribution that can divert their income from productive investment. But Messerli (2003) showed that farmers do invest part of their earnings in productive activities. Chapter 7 will confirm this outcome. Moreover, social redistribution can be 225
This is one of the arguments of the detractors of payment for environmental services.
311
viewed as a sort of social security system. In a context of vulnerability, social redistribution may be a rational factor of economic efficiency because it allows sharing risks. When a farmer receives financial support from his family, this can help him to avoid a decapitalization of assets (for example the sale of land), to maintain productive capacity (by buying rice to have more strength to work on the land). Modern states catch and redistribute up to 45% of Gross National Product. When the state is weak and taxes absent, it may be rational that individuals take the initiative of implementing this redistribution themselves. Observations in other areas further show that farmers invest their money in agricultural intensification when they have a significant source of cash. The Tanala farmers from the Ikongo region invested the earnings from coffee sales in the creation of terraces for irrigated rice cultivation, when the price of coffee was high in the early nineties. 226 In the same manner, the Zafimaniry people from the Ambositra district travel to western Madagascar to work as loggers during the off-season and invest their earnings to pay labor to improve their land (Rabetsimamanga 2005). Concerning the farmers of the Mananara region, they invest part of the profits generated by clover cultivation in the creation of irrigated terraces (Locatelli 2000). In order to assess how much profit farmers could obtain from carbon “grown” on their land, I propose a hypothesis of five hectares of reforested land per household. According to the draft PDD, the reforestation of one hectare would remove 390 tons of CO2 from the atmosphere after 30 years.227 If CO2 is sold five USD per ton228 and the money (390 x 5 x 5 = 9,750 USD) deposited in a bank account with a 6% interest rate, 226
Results from enquiries I made in the Ikongo region in 2002. Draft PDD, Section D4. No pagination. A total of one 179 998 tons of CO2 would be removed from the atmosphere after 30 years, by the reforestation of 3020 hectares. 228 According to IETA (2006), the price of certified emission reductions issued under the clean development mechanism ranged from about 2.5 to 24 USD / Ton CO2equivalent. Five USD per ton appears to be a reasonable hypothesis for the future if Clean Development Mechanism projects become less risky, which could be achieved by a more appropriate benefit sharing. 227
312
reforestation would generate 585 USD per year per household. If farmers received half of this money 229 (292 USD), it would compensate for the abandonment of slash-andburn agriculture on these five hectares. 230 A quick estimation, based on the hypothesis of an average rice yield of about 800 kilograms/hectare on tavy with fallow period, shows that the value of crops grown on these five hectares (one hectare being cultivated every year, with a five year fallow period) would be about 300 USD per year. 231 But as farmers would continue to practice agriculture on the rest of their land and dedicate the soils with less potential to reforestation, the real opportunity cost of not cultivating the five reforested hectares would actually be lower. It would be equal to the difference of yield that would result from the reduction of agriculture land, and this difference could be null or even negative if the 292 USD perceived annually was invested for intensification on the remaining land (for digging irrigation canals, buying cattle for the production of manure, or buying agricultural tools). Minten (2003) estimated that a majority of farmers would abandon tavy if they received 85USD per year per household as a compensation (in the form of rice). This survey was conducted in the Maroantsetra region, which is less poor than Beforona due to the presence of more profitable cash crops, such as vanilla. The abandonment concerned all types of tavy, not only tavy in forest. This unexpected result (that growing carbon can generate more value than growing rice) has to be confirmed by a more precise measurement using rigorous
229
The rest could be directed to the government, could serve to finance monitoring, or could stay on the bank account to increase the capital over time. Farmers would plant the trees for free because in this case the usufruct would be for them. 230 We saw in Chapter 2 that the opportunity cost of protected areas ranges from about 10 to 60. USD/year for households living around protected areas (Ferraro 2002; Shyamsundar 1997). But in this case, agricultural land is reforested which results in higher opportunity cost. 231 800 kilograms x 0.3 USD = 240 USD. To this must be added the value of maize, beans and other minor crops associated with rice on tavy. 300 USD may be a gross estimation of the total value of these crops and more precise calculation must be done.
313
economic evaluation tools, but it reminds us of the huge economic and productivity gap that exists between the agricultures of developed and developing countries. 232 There is however an important potential pitfall in this approach. The payment would not compensate for the nondevelopment of future agricultural land use, unless carbon prices increased in the future. For the part of their land affected by reforestation, farmers would be dependent on other sources of income, such as profits they could receive from other forest products. Bio-prospecting contracts and the sustainable extraction of forest products could provide these supplementary incomes. Moreover, the carbon contract will be terminated after the 30 year period of the project and the carbon could then be sold again, allowing for increase in the capital and the annual income generated by it. In sum, payment for planting trees and maintaining them in place could serve as a basis for a new form of development based on the sustainable extraction of forest products, rather than on agriculture, and carbon would just be one product among others. Local communities would “develop” in the sense that they would receive increasing revenue from reforested land. If, in spite of these opportunities, the revenue generated by forest products was still inferior to what would be obtained by converting land to agriculture, it would mean that agriculture would still have an externality (the loss of the last primary forests) that would have to be suppressed by asking the international community to continue to pay for the difference. Or, from the local community perspective, the compensation (the subsidy) will have to be permanently adjusted to the opportunity cost of not clearing the reforested land. This aspect of the bargain could be perceived as a danger by organizations committed to the 232
Mazoyer and Roudart (2001) showed that this productivity gap is now about one for one thousand, meaning that a well equipped farmer in the developed world produces a thousand fold what a tavy farmer produces in Madagascar (he can yield ten tons per hectare on a 100 hectare farm, whereas the Malagasy tavy farmer can manually cultivate about one hectare only, and yields about one ton per hectare).
314
conservation of ecosystems. But this fear may reveal that through compensation communities would be genuinely empowered, which could facilitate their genuine participation in resource management. Chapter 11 and 15 will provide more arguments in favor of this strategy.
4.7. Conclusion In conclusion, the VMCRCC project reveals a global political economy issue and illustrates a pattern that may prevail in most land use-change and reforestation carbon sequestration projects. Planting trees has always been a way to appropriate land and it is well known that reforestation projects often result in land tenure conflicts. The supplementary value of trees provided by carbon may still increase this tension, unless part of the profit generated by carbon is directed to the traditional users of the land. It is not only the ethic of conservation that is in play here, it is also its potential success or failure. In the global competition for the appropriation of resources, states have the power to rule and international organizations have the power to shape knowledge. But the main power is in the hands of farmers: they will decide the fate of the forest, by clearing it or not, by burning it or not, in the remote areas to which they will flee if not in the reforestation sites themselves. In this political economy framework, carbon, in addition to biodiversity, may actually have become a key “actor.” Carbon may have been the main cause of the importance given to fireless alternatives and fire repression policies adopted by the LDI and PTE projects. Even erosion is not as important as assumed in the region to justify fireless techniques, as shown by Brand et al. (1997). But the logic of conservation policies and development projects may now be on the verge of passing to a second stage, initiated by the VMCRCC project, where avoiding carbon release to the atmosphere is no longer regarded as sufficient. Carbon must now also be
315
sequestered. Being given this new momentum, carbon policies may become still more detrimental to farmers because in addition to “justifying” fire repression, carbon will now occupy their land. There is a paradox in this situation. Carbon sequestration is actually a profitable alternative to slash-and-burn cultivation. But because of this profitability, it favors the appropriation of land usufruct by actors that are not the historical beneficiaries of this usufruct. The lesson from this paradox is that markets provide potential solutions, through new opportunities for revenue, but also contribute to marginalization of those who are less prepared to seize its opportunities. On the verge of markets, primary ecosystems are the last pieces of land not integrated into the global economy, and slash-and-burn farmers living on this forest frontier are among the less prepared people to seize market opportunities. Forest clearing and the poverty of slash-and-burn farmers may be two interdependent “externalities” that need to be internalized simultaneously, through appropriate market regulation mechanisms, if we are to solve the problem of deforestation. A way to do so is to channel the profits from carbon sales straight to those having traditional use rights on the forests. If not, we may lose both the last primary forests and their people. My demonstration may not be complete yet due to its speculative character, as I explained in the introduction of this section. But there is unfortunately only one way to be more demonstrative: to implement and experiment on behalf of assumptions. This is actually what the Malagasy government, Conservation International and its partners are doing. The experiment is the VMCRCC project itself. The problem is that it is not recognized as being an experiment, although it is based on a series of weak assumptions. It is now necessary to question these weak assumptions in order to address uncertainties, adopt an adaptive management approach and remain open to new practices and debate. The alternative is that the project continues to its end as if
316
based on established certainties. Both the Malagasy forests and its people would lose if the institutions that carry the project played that game.
5. Conclusion This analysis of the activities conducted by projects in Beforona appears to verify the hypothesis of this dissertation: there may be a mismatch between realities and the representation of these realities by the projects that address them. This mismatch results from a series of wrong beliefs, which we could also call received wisdoms, myths or dogmas. The LDI, PTE and VMCRCC projects did not necessarily create these received wisdoms. They rather strengthened them, meaning that it is not the organizations that run projects that need to be put into question. It is rather a larger framework within which these organizations operate. This larger framework will be studied in Part IV, but as a transition to this fourth part and in order to synthesize the outcomes of Part III, I will review here these received wisdoms. They will be put inbetween quotation marks in order to show that they have an “author,” although this author is a quite abstract entity, as we will see in Chapter 9. “Farmers will implement slash-and-burn cultivation forever.” As a result, “there is a Malthusian vicious circle of land degradation and poverty: land productivity will decrease forever, leading to catastrophic soil degradation and to a deep agrarian crisis.” This can happen but is far from being the rule. Slash-and-burn is not the only land use in the area and most households are committed to giving increasing importance to other systems, which are irrigated rice, rotation on plowed land and home gardens, as shown by Chapter 5 and by the works of Messerli (2003) and Nambena (2004). Lack of investment capacity, risks, and difficult access to market are the main constraints to the development of these new land uses. Culture and tradition can slow change but farmers are eager to adopt new techniques as soon
317
as they experience that these techniques are more profitable and economically suitable. Chapter 7 will confirm this point. Boserup (1965) showed how pressure on land can be an incentive to develop new land uses (Chapter 5). According to her model, population increase does not lead to a dead end but is rather a precondition for a new step of agricultural development. More population means more labor force available to farm the land, reshape the landscape, plant trees, and improve soils. What matters is to implement wise policies that will help farmers to engage in this dynamic and not be trapped into poverty. “Yield is the main determinant of agricultural production.” I showed that yield it is the main criteria used by the LDI and PTE projects to assess their results. But farmers’ economic strategies are rather the maximization of production in relation to the rarest resource. When much land is available and agriculture in a phase of conquest (in a transition from extensive to intensive systems), labor, and not land, is the limiting resource and labor productivity, and not yield, is the main criterion to assess profitability. In case of vulnerability, risk minimization is also an essential strategy. The case of Ambodilaingo (Chapters 5), the perception of LDI’s alternatives by farmers (Chapter 6) and Messerli’s (2003) outcomes verified this hypothesis and showed that farmers in Beforona are committed to increasing labor productivity, not yield, because this will determine how much they will produce and eat in the end. “Technology is the key to agricultural development.” Most projects invest much effort to the development of new technologies, such as high yielding rice, agroforestry, fireless technologies, improved fallow, zero tillage and cover crops. They allocate much of their resources to the technical training of farmers. LDI and PTE were no exception. But one of the most frequent complaints by farmers, as we saw in Chapter 6, is that they cannot apply the technologies they have been trained for, because they lack the necessary resources (investment issue) or do not have time
318
(labor productivity issue). Support should thus be centered more on economic aspects (credit and subsidies). “Integration to market will help the poor.” Again, this is often not the case due to a strong competition for access to market niches that can be quickly filled, and due to low prices that allow only the most competitive (those with better land, easier access to market, more investment capacity, more capacity to support risks and greater opportunities to get external support) to make a profit. The degradation of the terms of trade is also a concern. Coffee used to be an important cash crop in the area but is almost abandoned now due to its low price (Chapter 3). Profit from ginger cultivation is also decreasing. Farmers from remote villages abandoned this crop because “big producers along the road lowered the price” (a farmer in Ambodilaingo 2002) (Chapter 5). Markets are also uncertain, especially with regard to minor crops such as fruits and vegetables which are also the most profitable. In Andasibe for example, avocados, “can provide 20 USD of revenue per tree and per year… if we find someone to buy the fruits!” (a staff member of the NGO Mitsinjo 2007). These risks result in a reluctance of the poorest to invest in marketable crops. In a remote place like Ambodilaingo, support to food crops, especially rice cultivation, may be the priority. “Fires are demonic.” This is certainly the most radical dogma adopted by projects that operate in the area (Chapter 6). The implicit assumption that fires are demonic led to considering all fires equal, which they are not despite the fact they obviously all have a destructive effect on vegetation and soils. The main difference between fires is that some are associated with management practices (pasture fires and slash-and-burn agriculture fires) while others are not (accidental fires, political fires, fires associated with land tenure conflicts). The first are less destructive because the management objectives determine a bottom line under which degradation has no reason to occur (Photograph 21 page 205). It is not worth it, for a farmer, to burn a
319
piece of land if it will not yield sufficient rice to compensate the work invested. Thus slash-and-burn fires never degrade the land more than what is compatible with rice cultivation. Farmers do not burn land if doing so would favor invasion by Imperata cylindrica and they do not cultivate land invaded by this species, unless they have no choice. For this reason, Imperata invasion would be reversible if fires were ignited only for tavy. In the same way, pasture fires (which are not frequent in the region) have to be compatible with the maintenance of forages. What should have been targeted by antifire policies are the uncontrolled fires that continue land degradation after the bottom-line determined by tavy fires is attained, and fires in primary forest, which threaten the last significant biodiversity reservoir. If this strategy was adopted, most households would be ready to collaborate with state authorities, at least in the case of wild fires which they also perceive as a plague. On the other hand, the repression of all types of fires, aside from its unethical character, leads to hostility and resistance. “Erosion is demonic.” The logic behind this assertion is similar to the one concerning fires. It is true that erosion degrade soils in Beforona and has led to catastrophic events in many areas of Madagascar. But demonization impedes addressing the diversity and complexity of the situations in which erosion occurs. In the case of Beforona, landslides can occur here and there and measures can be taken to prevent them (Brand et al. 1997), but lavaka 233 is not a significant feature of the landscape. In spite of this, projects in Beforona continue to invest significant research efforts in the development of agroforestry technologies and to promote mulching, whose effect against landslide is more than doubtful (Brand et al. 1997). Besides, farmers develop their own anti-erosion systems, which are based on the assumption that water has to be evacuated through drainage ditches (Photographs 7 to 10, page 233
A particularly destructive form of erosion very frequent in Madagascar (Chapter 1).
320
156 to 159), rather than stopped and infiltrated by hedgerows planted along contour lines. Their practices are certainly imperfect but match better with the outcomes of Brand et al. (1997) (Chapter 3). “Farmers are backward and reluctant to change.” This statement is also widespread in the literature about rural development. But certain forms of reluctance exist in all societies. It is true that for many Malagasy farmers, it would have been more rational to develop alternative land uses earlier, before they got trapped into poverty and had their investment capacity reduced to zero. But in the same manner, it would have been more rational for Western countries to develop alternative energies and reduce green house gases emissions earlier. Human induced global warming was first discovered in 1957 (Revelle 1957) and widely recognized by the scientific community in the late eighties. In spite of this, significant decisions are still far from being taken. For Malagasy farmers, as for Western societies, there are factors of inertia which have to be addressed properly and levered rather than just used as an argument for blame. Focusing on blame, to the contrary, leads to oversimplified explanations and to the overlooking of realistic solutions. “Culture and traditions explain land uses.” Again, empirical observations show that this is not true. The main Malagasy ethnic groups classically regarded as slashand-burn cultivators (the Tanala and Betsimisaraka) have developed a specific culture associated with this land use (for example, the ceremonies associated with the clearing of new forest land). But Rakotoarijaona (2005) showed that the ancestors of “tavy” farmers in Ambodilaingo are a mix of Betsimisaraka and Bezanozano people (Chapter 5). The Bezanozano group is close to the Merina of the highland, for whom irrigated rice cultivation is the main land use. But Bezanozano farmers in Ambodilaingo adopted slash-and-burn because this land use is more suitable to the area. Kottak (1971), Astuti (1995) and Bloch (1995) described similar dynamics of cultural change
321
and adaptation in the case of Tanala, Vezo and Zafimaniry people. In short, culture may be a consequence of land use rather than the contrary. In most regions of the world where demographic density is low, rain is important, soils low in fertility, access to markets problematic and slopes steep, people practice slash-and-burn agriculture. When one of these criteria changes (for example population density) land use also changes and with it culture and traditions. Rather than changing culture, the challenge may be to accelerate this existing change, and economic support may be the key to this. As I said earlier, received wisdoms are not the consequence of choices made by people or institutions. They operate from a higher level. They are present in the very substance of the systems of ideas and practices within which people and projects operate. This substance, which I will call discourse, will be studied in Part IV. There is, however, an institutional mechanism that favors the persistence of received wisdoms or that impedes the reality to “jump” into the face of actors and debunk misinterpretations. This mechanism was described, empirically, by LDI’s and PTE’s staff. I called it the chain of actors (Figure 14). It is significant that almost all LDI, PTE and VMCRCC staff I interviewed shared with me the opinion that the techniques proposed to farmers were not adapted to their socio-economic situation. LDI and PTE’s staff explained that when they had opportunities to work closely with farmers (some were living in villages and spent most of their time there), they came to understand clearly farmers’ expectations and needs and that these expectations conflicted with those of the project. Through discussions with staff at a higher level (regional office), I realized that exactly the same frustration existed at their level. And while working later in Antananarivo, I suspected (not specifically in the case of LDI and PTE) that this frustration existed there also, due to the dependence of Antananarivo’s offices vis-à-vis decisions taken in Washington, D.C. The closer the
322
+
Donors in their home country
_
Donors in host country Experts in donors’ country Decision power
Project coordinator
Knowledge of local realities
Experts in host country Local coordinator
_
Local staff
+
Farmers
Figure 14. The chain of actors.
323
actors are located to beneficiaries, the less authority they have to take decisions, as illustrated in Figure 14. In other words, two types of knowledge confront with each other: one that is experienced on the ground and one that uses very few inputs from the ground. The next chapters will provide more insights for understanding these different forms of knowledge. But I raise this issue here because it means that received wisdoms, although they are hegemonic and shape the discourses and practices of rural development organizations, could be weakened by appropriate institutional arrangement. The solution may be to break the chain of actors, to bring the centers of decisions back on the ground, inside the reality. The social experiment conducted by Zanaky ny Ala, which I will present now, was aimed at testing this possibility.
324
CHAPTER 7 CHANGING REALITIES: A SOCIO-ECONOMIC EXPERIMENT IN BEFORONA
1. Introduction This chapter presents an experiment that has been conducted in Beforona in the form of a pilot rural development project, on the basis of previous chapters’ outcomes. Some activities have been designed but not implemented due to the lack of time and funding. They will however be presented as well. From a more practical perspective, this chapter can be read as a series of recommendations for implementing more efficient rural development activities aimed at addressing the issues of deforestation and slash-and-burn cultivation. Section 2 will present the NGO in charge of implementing this experiment. Section 3 will describe the activities that were carried out by this NGO, while Section 4 will present activities that were planned but eventually not implemented. Section 5 will describe what the NGO did not intend to do, which may better reveal its philosophy than the list of its activities. Section 6 will review the mistakes made by the NGO, before I will conclude (Section 7).
2. Presentation of the nongovernmental organization Zanaky ny Ala Zanaky ny Ala 234 was created in 2003, by my initiative and with strong involvement of my assistant and translator, who became the coordinator of the project that will be presented here. Zanaky ny Ala further recruited nine field agents in 2004. Five were master’s program students from the University of Antananarivo and four were recruited in Beforona. The students belonged to the department of Geography, History and Economy. Their role was to manage the logistics of the project and to 234
In English: Children of the Forest.
325
support community organization. They also dedicated time to collecting data for their master’s thesis. The staff recruited in Beforona assisted them. I was not myself a member of Zanaky ny Ala but was involved as a benevolent technical adviser, helping to design strategies and activities, to prepare technical documents and to raise funding. The first funding came from private donations.235 They were complemented in 2003 by a grant from the Food and Agriculture Organization236 and, in 2004 and 2005, by food donations by the World Food Program, which allowed implementation of a “watershed management” pilot project. 237 Most activities were implemented in the fokontany of Ambodilaingo, but other fokontany (Sahanonoka, Ambatoharanana and Beforona) also received support. A partnership with the Food Security Program of the nongovernmental organization ADRA, 238 funded by USAID, was established but lasted only the first year. 239 Zanaky ny Ala was an NGO in its infancy, which had advantages and disadvantages. It was open to nonconventional approaches. The staff had no previous experience but was highly motivated and the functioning costs were limited. Two thirds of the budget was directly invested into the activities of communities, in the form of food distribution (food for work), equipment (construction of dams), and
235
In total, the NGO received about 10,000 USD from private donators for the implementation of this project. 236 An 8,560 USD grant from the Telefood Fund (contract TFD-1/MAG/003) was received on October 8, 2003. In theory, Telefood funds can be granted only to farmers’ organizations. In practice, NGOs often serve as an intermediary to manage the funds (as did Zanaky ny Ala), due to the illiteracy and low formal education level of most beneficiaries. 237 The name of the TELEFOOD FAO project (8,500 USD) is “ Projet Pilote d’aménagement de bassins versant en alternative à la pratique de l’agriculture sur brûlis ” (Watershed management pilot project for the development of alternatives to slash-and-burn cultivation). To the contrary of what this name could lead one to expect, the project did not employ an integrated watershed management approach, for reasons I will explain later. Concerning World Food Program support, Zanaky ny Ala received a first grant of about 60 tons in 2004, and a second grant of about 40 tons in 2005, to implement food for work activities. A third grant has been received in 2006 (amount unknown) to implement a new project in a neighboring commune (Ampasimbe). I was not anymore involved with Zanaky ny Ala at that time. 238 Adventist Development and Relief Agency. 239 The differences in the approaches of Zanaky ny Ala and ADRA resulted in difficulties for the establishment of a partnership protocol.
326
inputs (agricultural tools and seedlings). Zanaky ny Ala had no office in Antananarivo and no vehicle. The staff used public transportation and had its office in Beforona. 240 It spent most of its time in Ambodilaingo and in other fokontany, where it shared roof and meals with the villagers. Expertise was provided by me and some operations failed while others succeeded. The work was sometimes done with amateurism and the logistics were complicated due to the absence of communication means. The main objectives were to achieve agricultural intensification and to reduce the dependency on natural resources by putting into practice the conclusions of my PhD research, at the scale of the Ambodilaingo village. Zanaky ny Ala adopted an Action-Research philosophy, stressing the reciprocal benefits that research and social action can get from each other (Greenwood, 1998). Two main activities were implemented: the creation or rehabilitation of irrigation schemes and the support to technical innovation by the organization or agricultural contests. The first activity was an answer to the lack of investment capacity of slash-and-burn farmers while the second addressed the issue of risks, by taking in charge farmers’ technical innovation. Natural resources management objectives were not addressed directly, but indirectly, by supporting agriculture intensification. The challenge was to increase agricultural production in order to reduce the dependence on tavy and the pressure on primary ecosystems. Consistent with this, we never made any judgment about the fact that farmers practiced tavy and cleared forests. We believed that by answering to their immediate problems, we would encourage them to later go to a second step and respond to longer term issues. Short-term impact on agricultural production was also viewed as a necessary step to create mutual confidence, before initiating more challenging activities.
240
For a cost of five USD per month.
327
Reforestation and support for perennial crops were included in the project proposal in order to answer better to the expectations of potential donors. These activities have not been implemented in the end, due to the difficulty of involving farmers in activities whose benefits occur in the medium or long-term, and due to the short duration of the project (one year, eventually extended to 18 months). Moreover, the challenge was to reduce the pressure on primary ecosystems and reforestation would have increased this pressure. Others activities were deemed essential by Zanaky ny Ala but had not been included in the proposal for several reasons. First, the Telefood fund that has been granted by FAO had to be exclusively dedicated to finance equipment and inputs. Second, a 12 or 18 month period and the 8,500 USD Telefood grant were not sufficient to implement these activities. These other activities included building a road for opening the southern part of the commune to markets the entitlement of bottom lands, the construction of community granaries, the sustainable extraction of forest products and support to education and healthcare.
3. Implemented activities 3.1. Irrigation of bottom land Seven cement dams were constructed between 2003 and 2005: four in the fokontany of Ambodilaingo (Marofody, Ambia, Andriamanavana and Savia), one in the fokontany of Sahanonoka, one in the fokontany of Ambatoharanana (Hiaranana) and one in the fokontany of Beforona (Ankorakabe). The Marofody, Amby and Andriamanavana dams were constructed in partnership with the ADRA-FSP project, financed by USAID. ADRA provided rice and beans for food for work activities and engineering expertise to design the dams. Aside from the construction of the dam and canals, no other support has been provided to the beneficiaries. Support for social
328
organization and extension of new techniques were not implemented because they were deemed too costly in relation to their potential impact, because we preferred to let farmers find their own modes of organization and because several projects already provided technical support in the area (LDI, PTE and SAF-FJKM 241 ). A mason team was hired for constructing the dams under the supervision of ADRA-FSP engineers or of technicians directly hired by Zanaky ny Ala. Stone workers were hired to prepare the rubble stones and gravel, directly on the site where the dam was built or on the closest place where appropriate rocks were available. The rest of the work (transportation of cement from the road, excavation of sand in the rivers, digging of irrigation canals) was done by beneficiary farmers using a food for work approach (Photograph 27). This subsidy in the form of food was justified by the strong food insecurity of most households (Chapter 5). About 70 tons of food was provided in total by the World Food Program, through the ADRA-FSP project in the beginning and by direct contracts with Zanaky ny Ala later on. The Zanaky ny Ala staff ensured the general coordination and organization of the work. The traditional leaders (Photograph 28) were strongly involved in the design of irrigation canals and in organizing and controlling the work. The food for work approach raised several problems. The amount of food distributed per day of work was too high: the value of the food distributed was equal to a salary, and sometimes even higher, which is not consistent with the economic logic of a subsidy system. The risk was that households could do work having no interest for them beyond the salary they received. Many nonbeneficiary farmers indeed came to work with Zanaky ny Ala just to have a well paid job. On the other hand, several canals ended in not being functional. If farmers had not been “paid” to
241
A large Malagasy NGO linked to the protestant church.
329 330
Photograph 27: Villagers from Ambodilaingo digging an irrigation canal. This activity received support from Zanaky ny Ala (food for work and engineering).
Photograph 28: Irrigation canal and Rayamendreny in Ambodilaingo. This canal (Savia site) was dug with support from Zanaky ny Ala (food for work and engineering). In order to be functional, it still needs about 100 meters of masonry work. Revegetation has not been achieved yet. The three persons are, from foreground to background, the Tangalamena (traditional chief), the former Vavanjaka (his spokesperson) and a rayamandreny (a member of the village council).
330
dig them, they would have born the mistakes of the NGO unless they anticipated the failure and refused to do the job. Eventually, the seven dams were built and irrigations canals were dug. Three sites are currently functional (Marofody, Hiaranana and Ankorakabe), two sites are partly functional (Ambia and Sahanonoka) and two are a failure (Savia and Andriamanavana). The three functional sites were assessed in 2007 and the results are presented here.
3.1.1. The Marofody dam The Marofody dam, built in partnership with the NGO ADRA, 242 can irrigate about seven hectares of paddy fields. Paddy fields had been created on this site in the past but later abandoned because of cyclone damage. The land was totally in fallow and dedicated to tavy before the creation of the dam. One household created his paddy field the first year after the construction of the dam (Photograph 29). There were six paddy fields the second year and 14 the third year, among a total of 15 beneficiaries. The Photographs 30, 31 and 32 show the evolution of the bottom land from 2005 to 2007. Rice production averages 500 kilograms 243 per household in the main season plus 150 kilograms in the off-season. All households continued to cultivate rice on tavy (producing about 800 kilograms in average). They plan to increase the paddy field area over the years. By joining their tavy and paddy field harvests, they hope to produce sufficient rice for the whole year. Some envision abandoning cassava slash-and-burn cultivation. In sum, the impact of the dam is not sufficient for the abandonment of tavy but is sufficient to decrease the pressure on primary forests. The increasing mobilization 242
Food Security Program of the Adventist Development Relief Agency (FSP-ADRA), financed by USAID. 243 White rice (after hulling).
331
Photograph 29: A “tavy” farmer in Ambodilaingo, with her paddy field in the background. This young woman and her husband were the first household, among 15, to create paddy fields in the Marofody bottom land, where Zanaky ny Ala and ADRA built a dam. The second year, they were followed by 6 more households. The third year, all households but one had started to create paddy fields. This household used to rely on tavy and on income from wood transformation before the creation of this paddy field.
332
Photographs 30, 31 and 32: The Marofody bottom land in Ambodilaingo, in July 2005 (upper), April 2006 and February 2007 (lower). A dam has been built by Zanaky ny Ala, but no support was provided after completion of the work. This confirmed that farmers invest in improvements as soon as irrigation schemes are functional. In 2007, 14 households out of 15 beneficiaries had created a paddy field in the site. The canal is properly maintained and has been repaired after big rains, but no cyclone has stricken the site yet. The next cyclone will be the next test to the validity of the approach. The beneficiaries continue to practice tavy because they need both systems to produce sufficient rice. Some households expect to be self-sufficient in rice once improvements are completed. It seems however that tavy surface is diminishing and that the farmers’ concern is now to concentrate efforts on the site, which verifies Messerli’s thesis (chapter 6). The dam cost about 2500 USD including the payment of labor force using a food for work approach. This cost could be significantly lowered by scaling up the project.
333
334
of labor in paddy fields may in the end decrease tavy areas. The perspective to abandon cassava slash-and-burn systems is a first sign of this evolution. Farmers could also be more and more interested in the development of intensive systems around the bottom land, following the logic analyzed by Messerli (2003) and reviewed in Chapter 6. Several households established their principal dwelling and built a new granary close to their paddy fields, which is a first sign announcing this process. The main problems encountered are the damage caused by rats which discouraged most households from growing off-season rice, and the lack of food, tools and money. Only one household hired labor force (he spent five USD in total). Organization for managing the water and maintaining the canal is not perfect but seems to improve over time. The older farmer has been appointed “President” and canal damages have been repaired. Each household is in charge of the part of the canal adjacent to his field. Some land tenure conflicts occurred but were solved. All families belong to the same lineage and the distribution of land is relatively egalitarian. As one explained, “we try to provide a little piece of land for each family” (a farmer in Ambodilaingo, 2006). More households than what was initially estimated eventually got a piece of land in the perimeter.
3.1.2. The Hiaranana dam On this second site (Photograph 33), rice was cultivated in the perimeter before the construction of the dam. Precise information about the situation at that time is not available but all farmers interviewed explained that the main impact is the possibility to grow off-season rice, which was not possible before. The dam further decreases the risk of water shortage (Photograph 34). At present, among 16 beneficiary households, four do not cultivate irrigated rice regularly and 12 obtain an average of 450 kilograms of rice in the main season and 375 kilograms off-season. Five households still practice
335
Photograph 33: Paddy fields irrigated by the Hiaranana dam, built with support from Zanaky ny Ala. The cost of this dam was about 3500 USD, more than half to pay the farmer’s labor (food for work). About 15 households can utilize the irrigation scheme.
Photograph 34: Paddy fields suffering from water shortage in Ambatoharanana. This picture was not taken at the Hiaranana site, but illustrates the consequences of poor water control in the area.
336
tavy (7 before the construction of the dam), obtaining about 500 kilograms on average. One householder clearly said that he abandoned tavy because he produces off-season rice now. The dam would, however, not be the only reason for this abandonment. Other causes act in synergy. Ambatoharanana is located five kilometers from the road and benefits from significant market opportunities, in contrast to Ambodilaingo. These revenues are an advantage for land improvements. They are partly invested to pay labor force: nine of the 16 households spent 17 USD in average to hire laborers. This confirms that when farmers have the capacity to invest, they do so. From this statement, one could expect that farmers could also afford to buy cement (about 10 USD / 50 kilograms bag) and to pay a mason in case of damage by cyclone or big rain. Concerning the maintenance of the canal, no major problem seems to happen. Crumbling occurred after the 2007 big rains but the canal has been repaired collectively. The main constraints to increased production are the lack of equipment, the damage caused by rats and the lack of water. The area that can be irrigated would be too large with respect to the size of the dam and canal. Water management could certainly be improved to remedy this problem.
3.1.3. The Ankorakabe dam Concerning the Ankorakabe dam (Photograph 35 and 36), located in Beforona, less than one kilometer from the road, the situation is similar to Hiaranana. All 17 beneficiaries cultivate irrigated rice, with production averaging 450 kilograms in the main season and 250 kilograms in the off-season. Only six households still practice tavy. Unfortunately, no data is available for the period before the construction of the dam. The problems are also similar to the case of Hiaranana: lack of water, lack of tools and damage caused by rats. The lack of collaboration between beneficiaries for
337
Photograph 35: The Ankorakabe dam, in Beforona, built with support from Zanaky ny Ala. This is the biggest dam built by the NGO. The capacity of the beneficiaries (17 households) to maintain it in the long term is still questionable.
Photograph 36: The Ankorakabe canal, with irrigated paddy fields in the background. Beneficiary farmers proved to be capable of maintaining this canal during big rains. The scheme has not been stricken by cyclones yet.
338
the maintenance is also mentioned but the main damages, after the big rain of 2007, have been repaired.
3.1.4. The other dams Concerning the four other dams, three of them (Ambia, Andriamanavana and Sahanonoka) led to the creation of new irrigated rice fields in spite of low water flow (Photograph 37 ). Improvements on the canal should be made to increase the irrigation potential of these dams. Some canal sections crumble frequently (Photograph 38) and can create landslide in case of cyclone. The last site (Savia) is not functional at all and needs about 100 meters of masonry canal (Photograph 28, page 330). These dams are regarded as partial or total failures and no enquiry to the beneficiaries has been done yet. The problems result from a bad choice concerning the site for the dams and the pathway for the canal. However, the dams may be saved if there is more investment in masonry on certain sections of the canals. These failures do not put into question the approach itself but point out the insufficient experience of the NGO to implement the approach and the difficulty of finding skilled hydrologists.
3.1.5. Technical Supports Concerning rice cultivation techniques, no support has been provided as said earlier. Consistent with Messerli’s (2003) outcomes, we considered that the main obstacle to the development of irrigated rice cultivation was water availability and management. Water is still not sufficient for many farmers of the Hiaranana and Ankorakabe dams, as we saw. To remedy this point is a prerequisite before the extension of more intensive techniques. Moreover, the traditional techniques implemented at present already result in significant economic improvement and extension efforts have already taken place in the region for decades. Extension efforts
339
Photograph 37: Farmers’ improvements in the Andriamanavana site. The light green vegetation is rice. These fields are not irrigated and terraced yet. Farmers first do tavy during one or two seasons in order to eliminate most stumps and perennial root systems. Then they create terraces and plow the land. This site is not in a bottom land and shows that the potential for irrigated rice cultivation is higher than expected if considering bottom-lands only. The fact that farmers concentrated their tavy on the site reveals that they planned to make improvements, despite the fact that the Andriamanavana dam was not functional (and is still not).
Photograph 38: Failed irrigation canal in Ambodilaingo. The lower canal was poorly designed and abandoned due to landslides. The second is now functional, except in its upper section where rocks bar it. It catches the water from a small secondary stream (see picture 41).
340
are still ongoing, mostly through the ERI and SAF-FJKM projects. If more profitable varieties and techniques were available, farmers would end in adopting them. We actually observed that farmers tried new varieties in their fields, on their own initiative.
3.1.6. Socio-organization Concerning social organization, no support was provided beyond the construction of the dam and the digging of the canal as mentioned earlier. Beneficiaries organized according to modalities they defined themselves. Basically, the work of Zanaky ny Ala consisted in creating the infrastructure. Farmers were abandoned after this, due to the termination of the project. The main initiatives they took were to forbid the passage of zebu on the canal, to ban tavy on the watershed above the canal (this has been done only on one site) and to mobilize collectively to repair important damage (after big rains). The termination of the project may indeed have been an advantage because the farmers understood they should expect no more support. They decided to continue the work by themselves.
3.1.7. Cost-efficiency Concerning the profitability of the investments, each dam cost from 1000 to 2000 USD for the material, plus the food provided by the World Food Program, which represents about 2500 USD per dam. As a gross estimate, we can consider that each dam cost about 4000 USD for 15 beneficiary households, which represents an investment of 266 USD per household. This represents about five years of the subsidies that could be raised by storing carbon on one hectare (Chapter 6). Moreover,
341
the price could indeed have been much lower if canals were dug in appropriate locations 244 and if beneficiaries had provided part of the work for free. This cost can be amortized in about three years 245 by the additional rice production. This is just a gross estimation as there can be many sources of error in data collection. A real cost-benefit analysis, with precise measurements of production before and after the dam and assessment of direct and indirect negative and positive impacts, at a farming system scale, should be done to assess the real profitability of the operation. However, the fact that the amortization would take two or six years does not make much difference. The dams have been designed to last several decades or more if appropriate maintenance is done. My brief calculations just serve to show that the profitability of the operation is obvious if farmers actually do the maintenance, but is not if no maintenance is done. It was clear from the beginning that the challenge, beyond the construction, was to ensure long-term maintenance. This experiment, conducted on these seven sites, is not finished yet and the key test will be the passage of the next cyclone. To achieve long-term sustainability, other measures that go beyond what could be implemented on an 18 months project are necessary. For example, a communal tax on irrigated land could be raised to pay a communal mason that would maintain the sites. Zanaky ny Ala showed that dams are cheap and that farmers quickly make optimal use of them, but a larger infrastructure policy that addresses long term maintenance issues must also be put in place to validate the approach.
244
In a certain site, a first canal was dug, then abandoned and replaced by a second canal, doubling the cost. 245 The additional rice production is about 500 kilograms in the case of Marofody, and about 300 kilograms in the case of Hiaranana and Ankorakabe, these 300 kilograms are of similar value to the 500 kilograms of Marofody farmers, due to higher rice price in the off-season. 500 kilograms of rice x 0.3 USD /kilograms x 10 households in average = 1500 USD/year worth of additional rice production (4500 USD after three years, assuming a constant production).
342
For a complete assessment of the operation, indirect impact should be also measured. It seems that households that are not beneficiaries from the dam, observing the creation of new paddy fields by their neighbors, decided to create irrigation schemes on their own (Photograph 39). But the repression of forest clearing is also an incentive and the effect of both factors cannot be easily separated. Due to the relatively high cost of the dams and the failure of certain sites, Zanaky ny Ala started a new strategy after the construction of these seven dams. Smaller dams were constructed in locations where the more ambitious initial project had failed (Photograph 40), and food for work activities were implemented to support the extension of irrigated rice fields independently of dam construction. Contracts were establish with groups of farmers. A certain amount of food was given for digging canals, building a traditional dam (with wood, rock and mud) and creating terraces. Data are not yet available concerning these achievements but informal interviews in Beforona indicate that their impact was significant with regard to the cost. In some cases, however, food was used to pay the labor force of big landowners, revealing a risk of clientelist drift in favor of the rural elite.
3.1.8. Conclusion In conclusion, the construction of dams had a significant impact on household economy, but this impact has to be consolidated by achieving the long-term maintenance of irrigated schemes. In the case of Marofody, tavy is still practiced and paddy fields provide additional production (about 500 kilograms) that helps to achieve food security. In the case of Hiaranana and Ankorakabe, the off-season cultivation allows an additional production of about 300 kilograms. Off-season harvest is lower than first season harvest but has the advantage of occurring during the food shortage period, when rice can be twice as expensive. In sum, irrigated rice cultivation has a
343
Photograph 39: New paddy fields in construction in Ambodilaingo. The owner of this land received no support from Zanaky ny Ala. According to villagers, the success of the Marofody dam encouraged non-beneficiary farmers to create paddy fields on their own. Plowing and terracing new fields requires significant labor investment. In the background, we can see a paddy field newly created by a beneficiary of the Andriamanavana dam.
344
Photograph 40: Canal built by villagers to irrigate paddy fields, in Ambatoharana. This picture was taken in 2006. In 2007, nothing remained from this canal due to big rains. Hundreds similar sites may exist in the municipality of Beforona which could be improved by masonry for a very low cost.
Photograph 41: Masonry on a canal section in Ambodilaingo, built with support from Zanaky ny Ala. Such masonry is very cheap to build (less than 50 USD) and suffices to provide water to one or a few households. Hundreds of sites are probably suitable for such improvements in Beforona. In the future, this type of improvement, rather than bigger dams such as in Ankorakabe (photograph 35 page 361) should be privileged. A team of communal masons could be constituted to build them.
345
significant impact on food security, but access to irrigated rice alone is not sufficient to lead to the abandonment of tavy. Furthermore, uncertainty remains concerning the long-term maintenance of the schemes, which encourages a focus on smaller schemes in the future (Photograph 40 and 41 page 345). It would also be risky to focus excessively on this one activity, which leads us to the second activity implemented by Zanaky ny Ala: the agricultural contest.
3.2. The agricultural contest This second activity has been designed to potentially involve a larger number of farmers. The objective was still to encourage the development of alternative systems to slash-and-burn cultivation. But in this case, neither big investment nor collective management is necessary, the limited extent of bottom land is not a constraint and everybody can implement the new practices, down the slopes if not in bottom lands. The techniques proposed were land plowing (tillage), crop rotation, and organic fertilization. The choice results from the statement that in most regions of the world where slash-and-burn practices are abandoned, they are replaced by polyculture and crop rotation on plowed land and integration of agriculture with livestock husbandry (Boserup 1965, see Chapter 5). Beforona actually is not an exception and these techniques already exist in their infancy, mostly in the more densely populated villages located close to the road. We saw, for example, that crop rotation and plowing are starting to be applied for ginger cultivation (Chapter 3 and 6). Even rice is starting to be planted in rotation with ginger on plowed land. The agricultural contest was aimed at putting more households in contact with these new techniques and at allowing vulnerable farmers to implement them without taking risks. Participants were encouraged to plow and to use manure but were free to use the technique of their
346
choice, including burning the land. 246 In this way, there would be as many tested techniques as participants. The role of the contest was just to accelerate an innovation process carried out by farmers, by increasing communication and information exchange between farmers and by taking risks in charge. The long-term objective was to develop a crop rotation system, but this was rendered impossible by the lack of funding after the first harvest. About 220 persons from four villages (Ambodilaingo, Ambotanefitra, Ampasintsiriry, and Ambohimiadana) participated in the contest. All persons old enough to cultivate were authorized to participate. Sometimes, two persons from the same household registered. The crops proposed to farmers were beans and vegetables (zuchini and chinese cabbage). The support consisted in providing tools and seeds. In a context of high vulnerability and poverty, farmers do not take the risk to invest time and money in uncertain new practices (Chapter 5). So we gave a free hoe, a free angady and about 100 grams of bean seeds and vegetable seeds to each participant. This represented a 6 USD nonreimbursable grant per participant. Each farmer was asked to cultivate these crops on two 12 square meter plots so that the investment in time was negligible and he could continue his usual activities. In order to favor comparison of results, the participants were asked to grow their crops together on the same field, at the scale of one village (Photograph 42). At first they were concerned about land tenure issues (the owner of the land was afraid that the participants would appropriate it), but eventually found a flat piece of land on a river bank. In some villages, farmers separated into a few groups.
246
To avoid fire the first year appeared to be difficult or impossible, due to the abundant, spiny, ligneous and fast growing vegetation (mostly Rubus Mollucanus, Lantana camara and Aframomum angustifolium) that covered the fallow land. But after a first plowing had occurred, it seemed that fire could be avoided because the vegetation was dominated by annual herbs, due to the elimination of most rhizomes and strong root systems by the tillage. Unfortunately, the contest did not last enough time to verify this point.
347
Photograph 42: Vegetable cultivation for the agricultural contest. The fields are grouped together on fertile bottom land, on a small river bank. The land belongs to one or a few households who lend it to the participants. Photograph by Herimamonjy Lovasoa Ratsimijanona.
Photograph 43: Participants in the agricultural contest. Beans grew well, but production was highly variable from one participant to another. Farmers with low harvest did not complain, but enjoyed receiving an angady and having an opportunity to learn. Photograph by Herimamonjy Lovasoa Ratsimijanona.
348
The crops were to be sowed in July-August, which is the off-season. At that time, rain falls in the form of drizzle and there is no significant risk of erosion on plowed land. Harvest occurred in November, which is already the food shortage period for many households. Beans could significantly help to provide income because they can be sold at a good price (about 0.8 USD per kilogram) and amortize transportation costs, in contrast to bananas, the main cash crop at present. 247 The main risk was the irregularity of rain fall during this period. Beans need a relatively dry period for maturing and also for flowering. Vegetables suffer as well from excessive rain, which favors diseases. Farmers were aware of these risks but already had some experience with bean cultivation. They considered it worthwhile to try the system. All the farmers who received the tools and seeds actually participated in the contest. They all plowed the land and some of them applied manure. There was a wide diversity of sowing density, with densely sowed beans that grew well but did not produce any pods. Nobody complained about that, due to the small size of the plots. Failures were rather the subject of jokes and everybody could see the results of the most successful participants, learning about what to do the next time (Photograph 43 page 348). It was really unfortunate that crop rotation was not implemented because this was actually the main challenge. Plowing fallow land represents a significant labor investment that cannot be justified for just one year’s cultivation. When crops are put in rotation, plowing becomes easier every year. If two harvests are done on the plot the same year (for example, off-season beans or vegetable followed by rice or maize associated with beans during the main season), plowing could be avoided or be
247
A standard charge (30 kilograms) is worth 30 USD. Transportation costs about 0.03 USD/kilograms (about one USD in total). In comparison, 30 kilograms of banana is worth about 1.5 USD and its transportation costs one USD. This shows that the value per kilogram is a much more important criterion than the cropping system’s profitability per se for choosing cash crops in remote areas.
349
superficial during the main season. This would result in lower labor cost and lower erosion. Beans, or other crops such as sweet potato or groundnut, could indeed end in playing the same role as cover crops used in zero tillage technologies, with the difference that they provide edible products. They may be less efficient than cover crops such as Mucuna but this difference may not be problematic due to the still relatively fertile soils encountered in Beforona, at least in the bottom land and on the lower part of slopes which are, as we saw in Chapter 5 and 6, those paid attention by farmers who want to intensify. Rotation over several years could further favor vegetation with more forage value than the perennial herbaceous plants typical of most fallows. Actually, this effect was felt even after one cultivation cycle. A villager that owns cattle brought his zebu to graze on the contest plot after the harvest. Land submitted to crop rotation with short herbaceous fallow could thus be permanently used, either for cultivation or for grazing. This would motivate farmers to enclose it, either to protect it from cattle or to maintain cattle inside, during the fallow period. This would facilitate the integration of agriculture with livestock husbandry, which is essential for sustaining the fertility of this new land use in the long-term. The participants of each village defined criteria to designate the winner and created a committee of villagers for this designation. This allowed local knowledge to be integrated in the appreciation of the best techniques. A large series of criteria, not only the yield, could be taken into account, empirically if not arithmetically. The only message sent to the farmers was that they had to designate the cropping system they deemed more useful for their community. An inter-village committee also designated the “best” village. I did not purchase the data about who won but the village that won is Ambotanefitra. Unfortunately, no premium has been distributed so far due to lack of funding. Most participants did not complain because they enjoyed the experience
350
anyway. Some kept part of their seed and renewed the experience the following year, on their own land. We noticed an increased number of farmers sowing beans on plowed land after the contest, but the relationship is not clear as this practice already existed, though at a very limited scale (Chapter 5). In sum, this experience allowed the testing of a new approach for agricultural innovation. It showed that farmers are curious and want to try new techniques. Unfortunately, the contest did not result in significant technical innovation because the duration was too short to achieve the main objective: to test crop rotations that include rice over several years of permanent cultivation. Furthermore, taking in charge the risk may not be sufficient for future development of bean cultivation. According to most farmers, the main constraint of this crop is the lack of money to buy seeds. Even rice cultivation on tavy is constrained by this problem. Poverty leads some families to eat all their stock and to borrow seeds to sow the following year. A micro-credit system may be necessary to remedy this problem. The next section will present the system we designed (but did not implement) to provide this credit.
4. Activities planned but not implemented In this section, I will describe activities that were designed by Zanaky ny Ala but not implemented due to insufficient resources.
4.1. Micro-credit We imagined a system that was supposed to replace the free distribution of seeds after the agricultural contest, allowing farmers to cultivate beans at a larger scale. The existing micro-credit systems are not adapted to the situation of Ambodilaingo’s farmers as we saw in Chapter 6. Credit purchased at OTIV or granted
351
by the PSDR range from about 50 to 1000 USD per household. Most farmers in Ambodilaingo are not used to managing such large amounts of money, even 50 USD, and would have difficulty repaying. As food shortage lasts about six months, they would be tempted to use part of the credit to buy rice. Some OTIV beneficiaries in Beforona did this. In other words, the existing micro-credit systems are in fact macrocredits if we consider the economic level of most slash-and-burn farmers living in remote villages. Table 16 below gives a few examples of expenditures and investment that illustrate the economy of Beforona. Another problem with regard to access to credit is that it implies preparing a sort of business plan with the support of project field technicians. In this way, projects orientate the use of credits to certain techniques, mostly fireless technologies whose profitability is doubtful (Chapter 6). In order to remedy these difficulties, we imagined a system where anybody could receive a loan for any purpose, just by showing his ID. Maximum loans would amount to a few tens USD, which is insignificant with regard to our standard but significant in the perception of farmers. Credit would be linked to productive activities such as the purchase of seed but would be flexible. Rather than by the activity, the amount of the loans would be determined according to two criteria: the number of persons that offer their guarantee and the number of former times the same person borrowed and paid back. The persons giving guarantee would also have to present their ID and could get credit as well. One person alone borrowing for the first time would receive very little (no more than one to two USD) but ten persons mutually guaranteeing each other and having borrowed and reimbursed several times could receive a significant amount (a few tens USD each). The system would educate people to manage credit, not by training but by incentive. This system would match better with the empirical approach of farmers with a low formal education level. If one
352
Table 16: Example of expenditures and investment in Ambodilaingo, in USD (based on the 2007 exchange rate: 1 USD = 10,000 FMg) Objective Amount in Comment USD Investment to buy a young zebu
100, less after cyclones
Investment done on very rare occasion, when income has been earned by doing a well paid job, or when zebu are particularly cheap (after cyclones)
Investment to buy a pig
20
Investment possible for households that sell product at market (ginger). Limited by a traditional taboo but expanding, especially in villages located close to the road. Only a few households involved in Ambodilaingo
Investment in vegetable or bean seed
5
Rarely done, probably because of the risks associated with bean cultivation and the lack of cash.
Salary for wood transportation (for 2 days of work)
1.5
Some farmers specialized in this activity and transport 2 or 3 boards per week. Seems to be more profitable than agriculture on degraded land if job is permanent.
Renewing basic items (clothes, etc…) before the national day
5-20
Rice is sold to purchase these products. The price is very low and rice is re-bought at twice this price during the rice shortage period.
25
Rarely done. The profitability of this tool is unclear, except for the few people with large paddy fields, or unless the plow would be shared among several households.
Buying a plow
Potential profit from carbon sale
60 / ha / year
See Chapter 6
353
person would not pay back, access to credit would be cut to the whole group. In this way, the risk would be shared and solidarity would be encouraged. Reimbursement could be done in kind (part of the harvest) or in money. In the case of credits granted for productive activities with a short-term return, a high interest rate could be practiced (15% per year). A farmer borrowing 10 USD to cultivate beans, for example, would have to pay back 10.5 USD at harvest time (four months later). The credit office could be opened in Beforona, one day per week in order to minimize its cost (for example, on market day). It would be administrated by staff from the NGO. A register (on paper and on an Excel spread sheet) could suffice for its management. The risk of no reimbursement would be elevated, but the small size of the first loans would minimize the cost of this risk. In the end, those who would reimburse would have access to increasingly larger loans, which would encourage the others to eventually reimburse the few USD they would owe. For this system to work, monitoring would have to be minimal. Otherwise, the management costs would be far above the amounts of the loans. The only mode of regulation would be the sanction (no more loans) of the groups that do not reimburse. Maybe all this just would not work, but the only way to get an answer is to try the system.
4.2. Construction of secondary roads We saw in the previous chapters that National Road 2 has a significant economic impact. Farmers who live less than ten kilometers away from it can market their products (mostly ginger and bananas) and are clearly better off than those living in remote areas. Building secondary roads for a better access to market is thus essential. However, the demand in the market may be limited and increasing offers of marketable products could result in lowering the prices.
354
Secondary roads also present some danger. They favor immigration of people who could appropriate the most profitable lands. This phenomenon already exists in Beforona, but not yet in Ambodilaingo. Farmers are aware of it and sometimes are opposed to the idea of a road that would come to their village (Chapter 5). The forest frontier would certainly be the destination they would privilege if their land would be taken (Chapters 1 and 5). The answer to this pitfall resides in land entitlement, as I will show later. Roads further lead to increased pressure on primary ecosystems if they facilitate access to them. National Road 2 resulted in cutting the corridor into two pieces. On the other hand, roads facilitate control by state authorities. In a context of efficient control, they may be beneficial. Otherwise, they may be detrimental. The impact of most roads is however quite limited in space. Most migrants settle close to the road, usually less than five kilometers away. But this attraction can be an advantage. If the road is constructed at a reasonable distance from the forest, it can motivate people to leave the forest edge. In Ambodilaingo, some farmers attempt to move to villages closer to the road, or at least to get land there from a branch of their family. Their concern is to be able to grow and sell cash crops. As we saw in Chapter 5, ginger and bananas are not profitable anymore beyond a distance of 15 kilometers (Figure 8). Based on these outcomes, I conclude that roads should be built parallel to the forest corridor, about 20-25 kilometers east from it. This would provide access to market for a significant territory, while discouraging farmers from moving farther west and encouraging an opposite movement from the forest to the east. I showed in Chapter 5 that water and biomass are the two main resources that farmers look for to sustain their livelihood. Roads and the market opportunities they provide may be a third option. But the condition to achieve this is to avoid the appropriation of the most
355
fertile land and the filling of market opportunities by migrants to the detriment of local farmers, who would then move toward the forest instead. A road not too distant from the forest would also facilitate forest control. Tavy on the forest edge (on the escarpment) can be easily observed with binoculars at a distance of 15 kilometers. In term of forest products extraction, a distant road could encourage traditional harvesting (neither bulldozers nor other mechanical means) which has the advantage of being less destructive and of distributing a larger part of the added value to local communities.
4.3. Land entitlement Land entitlement should be implemented to secure improved land from appropriation by outsiders but also to encourage intensification. As land entitlement is costly, it has to be limited to areas more susceptible to appropriation and intensification: the bottom land that can be irrigated, the home gardens and the plowed land. Still more reasonably, a titling of paddy fields alone could be a wise policy to encourage improvements. We imagined a process involving state domain services (in charge of land entitlement) and leaving significant responsibilities to villagers. As a prerequisite to the operation, farmers would identify land that they deem privately owned (for example paddy fields), at the scale of households, not lineages. They would physically mark the boundaries of the fields in these perimeters. They would establish a list of plots with the owner’s name. The whole list would be validated (agreed and signed) by both traditional and legal authorities of the fokontany. Potential conflicts would be solved at this stage, by the community itself, with the incentive of being granted a title for free if the process gets completed. After completion, the domain services staff would be invited for the reconnaissance and for establishing a map or sketch of the
356
area, based on the delineation made by the community. Zanaky ny Ala would finance this visit. The cost of the operation would be relatively low because much would have been done by the community and because permanently cultivated fields are easily accessible and have clear boundaries. The next step would be the granting of titles. At this level, the intervention of the NGO would be minimal. A financial incentive would be given through a contract with the domain services, in the form of a negotiated sum of money per title effectively granted. An advance could be given to facilitate the procedures for the first set of titles and the rest would be paid after delivery of the titles. Land not permanently cultivated could also be titled despite the fact that this does not appear to be a priority. In this case, titles would be established in the name of lineage elders. Such an entitlement would be cheaper and would be sufficient to avoid the appropriation of land by outsiders. Under the authority of lineage elders, access of each household to land would still be managed following traditional rules, allowing more flexibility. As Boserup (1965) showed, the privatization of land is less marked in less intensive systems as increase in the frequency of cropping is associated with a move from collective to individual ownership. This logic is verified in the case of Ambodilaingo (Chapter 5). When intensive and extensive systems coexist, private ownership and community management coexist as well. The system we propose is aimed at taking account of this diversity and at allowing a match between legal and legitimate land tenure. A new approach to land entitlement is currently under development in Madagascar. According to the new land tenure law (Gouvernement de Madagascar 2006), land certificates can be delivered by communes. These certificates are not exactly land titles but open to the same rights. Donor agencies support the creation of communal land tenure offices in charge of delivering these certificates. The objective
357
is to reduce the cost of titling and to reform procedures, in accordance with the decentralization policy. This approach is at present in an experimental phase at the scale of pilot projects. Its impact will have to be carefully assessed before considering it to be the appropriate answer to the inefficiency of the current system. Like the VMCRCC project, it must be regarded as a large scale social experiment. Potential pitfalls could be the power it would provide to mayors to satisfy their clientele, the lack of capacities of remote communes to manage the system (skillful people are mostly urban and are often reluctant to work in remote areas), the difficulty of controlling abuses due to the large number of communes, and the temptation to condition the granting of title to the implementation of land use planning policies that would provoke new disturbances on local farming systems.
4.4. Community granary Zanaky ny Ala envisioned community granaries as a way to reduce the rice shortage period while taking advantage of the elevated price difference between harvest and food shortage periods. Granaries would be built at the level of each lineage or village. Farmers would provide the work and the NGO would provide materials to build them. The concerned communities would appoint a manager. A management system would be proposed but the community could modify, accept, or refuse it. The management system we elaborated was as follows: the villagers would sell rice to the granary for 25% above the market price at harvest time and could buy it for 25% off during the food shortage period. The buying and selling periods would be bounded. As the rice price usually doubles between the harvest and the food shortage period, the difference between buying and selling prices would still allow a significant benefit for the granary, while representing a virtual benefit for the farmers. In
358
comparison with the present situation, they would save money two times: by selling rice at a higher price and by buying it six months later below market price. Overall, farmers would still lose, but they would lose less than they currently do, which represents significant progress. They would also save the cost of going to Beforona to sell and to buy rice. A certain cynicism is necessary to accept the approach but farmers should be allowed to pragmatically judge the system. Moreover, the profit gained by the granary would remain inside the community. The key issue is how this money will be used, a point that farmers have to define for themselves as well. For simplification and to avoid conflicts, the maximum quantity of rice a person could buy should be equal to what he sold the same year. Rice bags should be tagged so farmers would actually re-buy the same rice they sold. But after a limited storage duration that still has to be defined (for example, until the next agricultural season), rice from the granary could be re-bought by anybody.
4.5. Community Based Natural Resource Management Favoring watershed protection, especially where dams are built, is an activity that could be started in the short-term. Farmers themselves took the initiative, by creating a community rule that forbids tavy on the Ambia dam watershed. But we had in mind a larger scale and longer term program: the creation of a forest reserve that would be co-managed by an NGO and the community. We believed that in the context of Ambodilaingo, where the education level is very low and where a few opportunist persons could take advantage of this, 248 a passage though comanagement
is
necessary
before
legal
transfer
of
resource
management
responsibilities to local communities. Chapter 11, which analyses community based
248
Chapter 11 will address this issue in details.
359
natural resource management in Madagascar, will provide the justification for this choice. The legal framework for this co-management approach would be the application of article 24 of the Malagasy forest law (Gouvernement de Madagascar 1997). An application decree of this article still needs to be issued to clarify the modalities of delegation, but contracts have already been issued by the ministry of environment on behalf of this article. Most of these contracts are between the state and private operators aimed at developing ecotourism, but a few concern NGOs that have been granted rights to manage forest resources with multiple purpose. Recently, the Malagasy state started to develop a legal framework for creating conservation concessions that may provide a legal frame for co-managed reserves (Chapter 2). In order to respect the expectation of the Ambodilaingo community, the reserve would be opened to logging. Some minimal rules, such as a minimum diameter for extraction and annual quotas, would be defined by the co-managers. Under these minimal rules, the community would organize as it wants, resolve conflicts by itself and distribute the benefits in the way it would consider the wisest. Conflicts would certainly occur but the community itself would have to solve them. It could establish dinas, written by hand on paper or even orally transmitted. It would just have to be made very clear that in case forests would be cleared and basic management rules not respected, the sub-contracts would be broken. Both rights and duties would be defined, for the state, the NGO and the community, allowing an escape from coercive state ruling and radical conservation approaches, as well as from the idealization of supposed harmonious relationships between communities and natural resources. As we saw in Chapter 5, the forest is traditionally managed as a reservoir land for the extension of agricultural land and a ban of forest clearing is necessary if we are to achieve a transition to more intensive land uses before the whole
360
forest gets cut. Only a top-down ruling, designed in the name of global interests, can lead to such a ban. This ban would have an important and social cost and would have a low legitimacy. For this reason, the ban would need to be compensated and support to agriculture intensification may not be sufficient for this compensation. I will come back to this issue in Chapter 11 and 15. If the ban and this compensation were put into place, the community would have to take its responsibility but would also be free to find its own way to put this responsibility into practice. In other words, beyond the delegation of legal forms of authority upon the resources, the community would be granted confidence to achieve its duties. With the support of the NGO, the biomass and the biodiversity could contribute to development through the carbon market (Chapter 6) and bio-prospecting contracts, in addition to the conventional uses (logging and nontimber forest products extraction). In other words, the co-managed reserve would be a way to avoid the appropriation, by the state or by other actors, of natural resources (products, biodiversity, and carbon). The value of these products may significantly increase in the future and may serve as a basis for a new form of development where biomass and biodiversity would play a significant and increasing role, beside agriculture. The main pitfall, however, is that the NGO that has been granted the contract could work for its own good instead of genuinely helping the community to capture these benefits. A law needs to be passed to guarantee that the benefits will actually be directed to communities.
4.6. Health and education Needs at this level are obviously huge but Zanaky ny Ala did not elaborate health and education programs. Actions were limited to emergency support during the influenza outbreak in 2002, just before the legal creation of the NGO. In case of
361
supplementary funding, it was also planned to equip a few schools with didactic items (posters, books) and to grant scholarships to children from remote villages that would complete primary education but could not afford to go to Beforona’s junior high school.
4.7. Others? The list of activities could never stop, but the risk would be to dissolve key actions in a mass of activities with low impact. In a context of huge needs and limited intervention means, the essential has to be implemented, not the necessary. Activities must also be designed in order to be replicable at regional or national scale in case of success, which implies that they have to be cheap. Our approach included an infrastructure policy (dams and roads), an intensification policy (the promotion of tillage, crop rotation and organic fertilization through agricultural contests), a subsidy policy (food for work activities), and a natural resources policy (co-management of natural resources). If implementation costs were excessive, these activities would be condemned to remain at the stage of pilot projects, whatever their impact. It can further be noted that the activities implemented or planned by Zanaky ny Ala are not new. In the vast area of development practices, it may be that everything has been tried already. The novelty of Zanaky ny Ala was rather in the approach and in what it did not do. First, the NGO employed inexperienced but highly motivated young students instead of confirmed “development” specialists, in order to render possible a cleansing of the misleading beliefs that dominate rural development projects. Second, Zanaky ny Ala did not pay much attention to training, socioorganization, monitoring and paperwork, and allocated about 70% of its budget to direct support to farmers, in the form of food for work and materials purchase (cement and other construction material, seeds and tools). Third, it selected just a few activities
362
judged as being the clue for successful development, instead of designing the perfect project that would address all aspects of the reality. To complete the portrayal, I must then say a few words about activities that were discarded and explain the reasons for this.
5. Activities not planned Bee keeping was discarded because the market cannot absorb much honey and because it does not fit well with the day to day economy that prevails in Ambodilaingo. Significant investments are necessary to intensify bee keeping and returns are uncertain and not immediate. The most interested farmers are often those belonging to the rural elite and they could quickly fill the limited market opportunities if they received support from projects. Fisheries and arboriculture were discarded for about the same reasons. Fowl and swine husbandry were discarded for three reasons. First, agricultural surpluses (rice for fowl, cassava for pigs) are necessary to feed animals, whereas farmers in Ambodilaingo do not produce such surpluses. Second, the risk factor is also significant (diseases). Third, soils in bottom land are still quite fertile in Ambodilaingo. Swine husbandry, however, would deserve to be supported if it helped to achieve better food security. It already deserves support in the fokontany located close to the road, which are economically better off. As we saw in Chapters 5 and 6, the production of manure is essential for the future and is a much less costly way to produce organic fertilizer than composting. An initial investment is necessary (which can be provided by credit) but labor inputs are much lower and benefits are multiple. Pig farming was actually intended to be supported later on, through the creation of the micro-credit system.
363
Agroforestry and agrobiological technologies were discarded because farmers never adopted them despite the fact they have been subject of extension efforts for decades, because other projects are already committed to their diffusion and because their profitability is disputable (Chapter 6). SRI and SRA, as well as the diffusion of other intensive techniques that consist in increasing yield, were discarded for similar reasons, because work productivity and low risk are more important criteria than yield for most poor farmers, because a paddy field that produces 1.5 tons per hectare is already a success and is already more profitable than tavy on degraded land, and because innovation has to be created by farmers themselves, which they will do if we provide them incentives and information about the external world. The agricultural contest could be applied to rice intensification and new rice varieties and techniques could then be proposed, but the first step was to create new paddy fields. Training was discarded because many farmers in the commune complain about the too many trainings they received, which reveals the difficulty of conducting training with the right content and led us to prefer social learning through shared experience (through the agricultural contest). Socio-organization was discarded because it was assumed that organization would be more efficient if it was designed by the beneficiaries themselves, and that changing the material basis of the production (by building dams) would create new needs that would encourage social organization (for managing water and canal maintenance).
6. Critical comments Some mistakes have been made by Zanaky ny Ala that served as lessons, allowing the following recommendations for future action:
364
•
The design of irrigation schemes necessitates hiring confirmed hydrologists or masons with strong empirical experience.
•
Communities should be supported for the construction of traditional dams and the digging of canals before the construction of cement dams, in order to verify the feasibility of irrigation schemes.
•
The smaller the dams, the easier their maintenance. Very small dams that bar a small stream and irrigate a few hectares are worth building, even if they benefit only one or a few households. They are very cheap (less than 100 USD), can be constructed without engineering input and a very large number of sites are suitable in the commune for such dams.
•
It would be worth training a team of communal masons to build and repair dams. An irrigation tax could be created to finance this team in the long-term.
•
Concerning food for work activities, no more than the equivalent of two thirds of a daily salary should be provided to workers. This represents a little bit more than one kilogram of rice per day. The rest would be the voluntary contribution of the beneficiaries. Contracts could, however, be established for payment of the remaining third in case the work does not lead to functional schemes.
•
The agricultural contest should be replicated over a period of at least three years and extended to the whole commune. The approach could also be applied to irrigated rice cultivation.
•
New tools should be developed for facilitating intensification in bottom land. Hoes and angady do not allow high labor productivity. Plows necessitate much draft force, are expensive and are not adapted to small plots. In between, harrows could be a suitable option. Zanaky ny Ala
365
attempted to develop a harrow, in collaboration with an engineering firm from Antananarivo, but the results were not conclusive. During the trials, however, the farmer who provided the zebus noticed that his animals did not get tired like they do with a plow, and that it was easier to maneuver and to work on small plots with this new tool than with a plow. The development of this harrow would be worthwhile to continue.
7. Conclusion The social experiment conducted by Zanaky ny Ala showed that farmers, even in communities regarded as being more difficult to work with such as Ambodilaingo, 249 can engage quickly in new activities if they are provided with appropriate supports. Abandoning environmental friendliness as the main criteria for selecting alternatives to tavy allowed to work more efficiently. Zanaky ny Ala chose to support endogenous dynamics (descending to the bottom land) rather than to force farmers to engage in a completely new pathway (developing sustainable systems on slopes using agrobiological technologies). Concerning the first activity, farmers in Ambodilaingo created paddy fields as soon as they had water arriving to their land. They had a 300 to 500 kilogram net increase in rice production and constructed new granaries to welcome it. This relative success, however, needs to be confirmed in the longer term as the impact of cyclones and the capacity of beneficiaries to maintain irrigation schemes are still questionable. Beyond Zanaky ny Ala’s operations, which were limited in space and time, a
249
When I arrived in Beforona in 2001, several persons tried to discourage me from working in Ambodilaingo, asserting that “people there are backward.”
366
communal strategy, if not a regional or national policy, may need to be designed to achieve larger scale and longer term impact. The agricultural contest was also a relative success. It confirmed that farmers are not reluctant to try new techniques. But as it just lasted one season and was not followed by support to purchase seeds, its impact was limited. These results, although limited, are sufficiently encouraging for justifying a longer term and larger scale project where economic support, rather than training, would be central, and were field experience, rather than technical expertise, would be required from staff.
367
CONCLUSION
This third part showed that two worlds actually coexist in Beforona. First, there is the world of ordinary farmers. It can be divided between those who live close to the road and have access to market, who were investigated by the Terre-Tany and BEMA (Chapters 3 and 6) and received “support” from development projects (Chapter 6); and those who live in remote villages, such as Ambodilaingo (Chapter 5) and are out of reach of markets and project interventions. Farmers living in such remote places may be more representative of typical Malagasy slash-and-burn farmers due to the very limited extension of the road network in eastern Madagascar. 250 In the case of Ambodilaingo, households survive in a day by day and almost autarkist economy based on the practice of tavy. Mining the natural resources (clearing forest and harvesting wood) helps them to survive but is increasingly compromised by stricter law enforcement. Improving bottom lands to create paddy fields is the most obvious alternative to tavy and the technology is already available (Chapter 7). It will not suffice, however, and necessitates significant investments. Home gardens could also play a role but are constrained by difficult access to market. Crop rotation on the most fertile soils is another option but trials over long periods must be implemented to accelerate the development of new systems. Integration with livestock husbandry will be essential to create a new mode of fertility reproduction that will replace the role played by fallow. If these alternatives are encouraged, tavy will continue to be practiced, but its extension will decrease over time, while the new systems will cover larger and larger areas. Getting back to farmers living close to the road, market offers them more opportunities to receive income that can be invested in 250
If one walks from Beforona to Ifanadiana, about 500 kilometers south, he will pass a village about every one or two hours, and cross just one road that even four wheel drive vehicles cannot use most of the year, in Marolambo.
368
intensification. But due to land degradation, yields are low and investment capacity remains limited for the majority of households. The second world is encountered mostly along the road. It consists of households that I called the rural elite (Chapter 5). It includes an entrepreneurial group of merchants that buy some of the best land, pay labor force and specialize in the most profitable crops, mostly rice and ginger. These entrepreneurs have strong links with influential local stakeholders such as civil servants, association leaders and project staff, who constitute the rest of the rural elite. The two groups are culturally, socially and economically less distant from each other than from the rest of the population. This gives to the merchants and entrepreneurs an advantage to channel most of the aid provided by NGOs and other organizations (Chapter 6). Zanaky ny Ala did not make the exception and ended in privileging collaboration with the local elite. Due to this wrong target, projects may eventually contribute to the marginalization of the poorest farmers, those who would sell their land or would move away to the forest frontier, due to fear of that unequal competition (Chapter 1). Unequal competition cannot be avoided. Social differentiation is inherent to any society. It is also normal that people who are culturally closer to each other collaborate more closely. But in Madagascar, those who get marginalized do not move to cities in the search for a job, as European farmers did during the enclosure and the industrial revolution. Instead, they move to the forest. If the objective of rural development projects that work in the area is to fight against poverty and deforestation, these entrepreneurs are the wrong target, a fact that the ERI project also recognized. This is not to deny the necessity of having a class of entrepreneurs and traders. But this class will benefit from the development of the area anyway, through increased opportunities for business.
369
This third part further showed that projects perpetuate received wisdoms (conclusion of Chapter 6), and that this maintains them is a state of ignorance about local realities. One reason for this may be that there is no physical contact between those who take decisions and those who are subject to these decisions. They never experience each other. Actors located in Washington D.C. meet their partners from Antananarivo, who meet their partners from Moramanga, who meet their partners from Beforona, who meet their partners from villages located close to the road. In a few cases, the two ends of the chain are briefly connected. But this occurs in a virtual world recreated purposely for this occasion, in the form of a workshop, such as those organized at the CDIA center, where PowerPoint “re”-presentations replace realities. Moreover, nobody meets the population from villages located ten kilometers away from the road whereas this population is practicing tavy, suffers from deep poverty and is subject to the same global economic and political system. This situation may worsen in the future because the centers of decisions will move farther and farther away from Malagasy realities, due to the globalization of environmental issues and the increasing legitimacy of global actors to enforce global regulations. Already now, carbon released by Western citizens may be the main determinant of development and conservation discourses and practices in Madagascar (Chapter 6), whereas carbon is still an abstract entity for Malagasy farmers. A second reason may be that the methods used to acquire knowledge focus on spare parts of the reality rather than on the whole, and on measurable aspects rather than on nonmeasurable ones. In consequence, biophysical realities are paid much more attention than social ones, economic realities being in an intermediate position. This explains why projects developed alternatives to slash-and-burn that maximize positive soil-plant biophysical interaction, while the main problems are located elsewhere, in
370
the lack of investment capacity of farmers, in the vulnerability of marginal groups to market forces, and in the unequal distribution of power between stakeholders. We are now getting to the turning point of this dissertation. This will lead us from the local to the global, from projects to policies, from facts to discourses, from the analysis of realities to the analysis of representations, and, eventually, to a more thorough analysis of the mismatch between representations and realities. In order to achieve this, three main discourses need to be extracted from the outcomes of this second part: the alternatives to slash-and-burn discourse, whose translation into project realities has been studied in Chapter 6 (agrobiological and agroforestry alternatives); the participatory discourse, present in Beforona in the form of supports to Kolo Harena associations (Chapter 6) and also evoked in Chapter 7 (community based natural resource management approaches); and the antifire discourse, which motivated the excessive focus on fireless alternatives and the tavy repression campaigns described in Chapter 6. These discourses will be analyzed in the fourth part of this dissertation (Chapter 10, 11 and 12). Once these analyses will be completed, the outcomes will be joined to those of Part III in a synthesis that will constitute the fifth part of this dissertation.
371
SLASH-AND-BURN CULTIVATION AND DEFORESTATION IN THE MALAGASY RAIN FORESTS: REPRESENTATIONS AND REALITIES Volume 2
A Dissertation Presented to the Faculty of the Graduate School of Cornell University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
by Jacques Pollini August 2007
PART IV THE REPRESENTATIONS: DEVELOPMENT AND ENVIRONMENTAL DISCOURSES
372
INTRODUCTION
I highlighted in the introduction that my research was driven by concerns about a real object: a community of farmers who practice slash-and-burn cultivation and clear primary forest in a given biological, social, economic, cultural, political, local and global context. I dedicated Part III to describing this object in detail, avoiding as far as possible to be driven by disciplines. This led me to the following conclusion: Slash-and-burn farmers in Ambodilaingo are poor, lack food during half of the year, have almost no capacity to invest, have low willingness to take risks, suffer from environmental policies and do not receive support from projects. They struggle to develop more intensive land uses and most often fail to do so. In villages located closer to the road, markets offer some opportunities, but poor farmers are in competition with a rural elite, usually migrants and traders, who buy land, hire labor force and benefit from foreign aid. Rural development projects fail to consider these social dynamics and may eventually contribute to the marginalization of the poorest farmers. Their failure can be explained by the existence of wrong beliefs, myths or received wisdoms that orientate projects and policies to unwise strategies. In other words, there is a mismatch between the realities and the representations of these realities by the actors who act upon them. In this third part, I will analyze the genesis of these myths or received wisdoms, in order to deconstruct the discourses within which they are embedded, as far as this is possible. This exercise will provide more evidence of the existence of a gap between representation and realities and may help to reduce this gap. I will start, based on the existing literature, with an overview of Western development and environmental discourses within which these received wisdoms emerged (Chapter 8). I will then introduce the concepts and methods which I will
373
subsequently use to deconstruct more specific discourses (Chapter 9). This deconstruction will focus on three discourses which articulate several received wisdoms together and that I considered particularly relevant for understanding the situation in Beforona. First is the alternatives to slash-and-burn discourse (Chapter 10) which helps to explain the excessive focus on agrobiological techniques and fireless alternatives analyzed in Chapter 6. Second is the participatory discourse (Chapter 11), which motivated the creation of Kolo Harena associations and the implementation of a farmer to farmer approach by the LDI and PTE projects (Chapter 6). The participatory discourses will be analyzed through the implementation of the GELOSE legislation, presented in Chapter 2. This approach was not implemented in Beforona but represents one of the major tools of Malagasy environmental policies. It is utilized by several operators, including LDI (in places others than Beforona) to address deforestation and slash-and-burn cultivation. The third discourse (Chapter 12) is the demonization of fires, or antifire discourse. We saw that it is central in development and conservation practices and policies. It may sustain, indeed, the other discourses, especially the alternatives to slash-and-burn discourse. As it has been extensively studied by Kull (2004), Chapter 12 will be mostly based on the outcomes of this author.
374
CHAPTER 8 OVERVIEW OF WESTERN DEVELOPMENT AND ENVIRONMENTAL DISCOURSES
1. Introduction In this chapter, I will present the Western development and environmental discourses that constitute the background of development and environmental policies, programs and projects implemented in Madagascar. Section 2 will be dedicated to development discourses. It will be based on personal insights developed along my career as a development professional. Section 3 will be dedicated to Western environmental discourses. It will be based on a recent synthesis on the subject by Dryzek (2005). In Section 4, I will use the categories of discourses presented by this author to clarify the environmental debates analyzed in Chapter 2 and to understand better the tension between two conflicting ideologies. This will enable the establishment of a link between Part II and Part IV, jumping over Part III.
2. Development discourses The concept of “development” engenders an abundance of discourses, ideologies, policies and passions. I will not review these discourses and the abundant literature (Sachs 1992; Rist 1997; Roberts and Hite 2000; Todaro and Smith 2003) that relates to this concept. Instead, I will attempt to cleanse the concept from its ideological layers, in order to bring back to the surface its core meaning, as far as this can be done. I will use two tools to achieve this: a dictionary and an analytical frame that will compare meanings of this buzz word in different domains.
375
2.1. The core meaning of “development” From the Oxford English Dictionary On Line, 251 development is: 1. A gradual unfolding, a bringing into fuller view; a fuller disclosure or working out of the details of anything, as a plan, a scheme, the plot of a novel. Also … that in which the fuller unfolding is embodied or realized. 2. Evolution or bringing out from a latent or elementary condition; the production of a natural force, energy, or new form of matter. 3. The growth and unfolding of what is in the germ; the condition of that which is developed: a. Of organs and organisms. b. Of races of plants and animals: The same as evolution; the evolutionary process and its result. development theory or hypothesis (Biol.): the doctrine of Evolution; applied especially to that form of the doctrine taught by Lamarck (died 1829). c. The bringing out of the latent capabilities (of anything); the fuller expansion (of any principle or activity). d. The act or process of developing … a mine, site, estate, property, or the like; also, a developed tract of land.… e. The economic advancement of a region or people, esp. one currently under-developed (sense b). 4. Gradual advancement through progressive stages, growth from within. 5. A developed or well-grown condition; a state in which anything is in vigorous life or action. 6. The developed result or product; a developed form of some earlier and more rudimentary organism, structure, or system. We can notice from these definitions that confusion exists between the concepts of development, evolution and growth. However, one feature appears repeatedly and seems to characterize “development”: the unfolding of new potentialities. When something develops, it does not remain equal along time, neither quantitatively nor qualitatively. New characteristics appear which are determined by previous stages. In order to illustrate this specificity, I propose to compare the concepts of change, growth, development and evolution in relation to five objects: a living organism, an ecosystem, an economic system, a society and our planet. Table 17 summarizes my propositions. It can be read as an attempt to answer the following question: How do
251
http://dictionary.oed.com.
376
Organisms
Ecosystems
Economies
Societies
The Earth
Change?
Through polymorphism, which results in changes along generations; these changes can be random (nondirectional)
Non-equilibrium ecology reveals ecosystems change, from one state or subclimax to another
Through external changes such as unstable markets and prices, which provoke adaptive changes, such as restructuring, delocalization, elimination or creation of corporations
Through external changes such as cultural influences, conflicts, and environmental changes, which result in adaptive changes such as migration, new technologies, new legislation
Due to movements in the solar system, which provoke climate change, displacement of ecosystems; due to tectonic movements, which provoke climate change on continents
A small organism becomes a big one over its lifetime (from a small fish to a big fish)
Through the colonization of new spaces, usually by spatial extension on the edges
Through GNP increase and the colonization of new spaces (globalization)
377
How do…
Through It does not! demographic increase and the colonization of new spaces (globalization) Table 17: How do organisms, ecosystems, economies, societies, and the earth change, grow, develop and evolve? Grow?
Table 17 (continued) How Organisms do…
Ecosystems
Economies
Societies
The Earth
From fecundation to death (from egg to tadpole and frog)
Through Secondary Succession (from cleared land to a subclimax)
By unfolding new sectors of activity (primary, secondary, tertiary); By improving well being and equity; By becoming more complex and powerful
Through the production of surplus and the division of labor; From traditional to industrialized societies (unless this is evolution?); From small polities to large empires; From a feudal society to communism (according to Marx); By becoming more complex and powerful
From the geosphere to the biosphere and the noosphere (Vernadsky 2002); from creation to annihilation, in relation with the life and death of the sun; Through god’s plan
Evolve ?
Through Natural Selection (Darwinism)
Through Natural Selection of all organisms belonging to an ecosystem; through co-evolution of these organisms
Through “Market Selection”: the most efficient people and firms win and eliminate the others
Through “Social Selection” (social Darwinism): the best fit groups win and marginalize or eliminate the others
It can’t because there is just one earth and it just has one life!
378
Develop?
organisms, ecosystems, economies, societies and the earth change, grow, develop and evolve? The propositions of Table 17 are questionable conjectures but help to draw general features of the four concepts in the left column. Change can be viewed as the most elementary transformation. A sum of changes of different natures determines growth, development and evolution. Growth, at least when it concerns physical entities, is a quantitative change of one thing made at the expense of other things. It is one dimension of more complex transformations such as development and evolution. It cannot occur indefinitely in a finite system, such as the earth. Development is the unfolding of new potentialities determined by a program. This program is the genetic code in the case of organisms, or a sum of genetic codes in the case of ecosystems. In the case of the earth, development is linked to the life cycle of the sun, from birth to death, and the elementary laws of the physical world set its program. Concerning social and economic systems, the program is more flexible but still exists. It can be a collective project determined by the collective willing of the individuals that constitute a society, or of a group of individuals that dominate this society. This program shapes the knowledge produced by this society and orients changes in a given direction. The Enlightenment’s system of values is an example of such pervasive programs. Communism and neoliberalism, on the other hand, are subprograms defended by subgroups of individuals but become hegemonic when these individuals obtain a dominant position in the formal structures of power. Environmental discourses may now announce the emergence of a new paradigm which we could call the “sustainable program.” Evolution implies the replacement of ancient systems by new ones that are more efficient. To proceed to this replacement, programs have to be improved, made
379 377
more complex and even created to restart development from zero. In the case of living organisms, this is done by mutation and natural selection. The entities that receive the improved programs have an advantage and eliminate the others, unless these modifications are flawed. With respect to the earth, I contend that it cannot evolve because it is unique, cannot be put into competition with other systems and cannot change its program, which is determined outside, by the sun. Earth is just the perimeter within which complex objects unfold (develop), inside boundaries delineated by universal laws. In the case of societies, a change of social program can be seen as paradigmatic change; the passage from a civilization to another one; a total change in the system of values. When civilizations disappear, others usually replace them that are more powerful and more efficient, in the sense that they are able to satisfy more needs for a greater number of people, still exploiting the same ecosystem. In the future, there will be a point where our own society will face the same fate. It will be unable to develop following the same program and conserving the same system of values. It will have to evolve in order to restart its development following a new program and a crisis may occur before this happens. A social entropy may exist which can be paralleled to the chemical entropy of living organisms and which determines an end to the unfolding of potentialities, i.e., an end to development. In sum, evolution works by the creation of new programs aimed at starting new societal projects when development cannot be sustained anymore. Over long time spans, it may be that societies have to evolve rather than “develop.” Evolution is thus mandatory precisely because development, by definition, cannot be sustained. I conclude from this overview that we cannot and should not get rid of development, whatever the type of system we deal with. But this concept must not be confused with change, growth, and evolution. All systems are dynamic, are determined by laws, have a birth and a death. Discarding a concept such as
380
development would result in an ontological vacuum because there would be no word to name this process of unfolding potentialities. If development is the unfolding of a program, this program has to be deconstructed and analyzed, not the fact that there is a program. The key questions to ask are what develops and for what achievements. I will now ask these questions, in order to identify the constructed meanings that Western culture has built upon the concept of “development.”
2.2. The constructed meanings of development From the Encyclopedia Britannica, Development or Economic Development is “the process whereby simple, low income national economies are transformed into modern industrial economies” (Encyclopaedia Britannica 2002). This definition goes far beyond the core concept as delineated in the previous section. It renders explicit the content of the program associated with development. This program is the march to an industrialized society, and modernization is regarded as a synonym of industrialization. Moreover, the Encyclopedia states that the term development is sometimes used as synonym for economic growth (Encyclopaedia Britannica 2002). It refers to the entry “Economic Growth and Planning” for “full treatment” of the concept and has no entry for social development. In sum, the development of human societies is assimilated to economic development, is framed inside industrialization, is universalized by making industrialization a synonym of modernization, and is nearly reduced to a quantitative process: growth. The consequences of this reduction are many and I will not address them here. But to say just a few words, this conflation of concepts may be the cause of the failure of our development models to solve issues such as deep poverty, food scarcity, social violence, deforestation, pollution, global warming and other social and environmental havocs.
381
Many theories have been proposed to support, criticize or challenge this model of development. I will not discuss them here because it would lead us far away from the empirical subject of this dissertation. But to summarize, we can distinguish three general trends according to the way the dominant development paradigm is regarded. First, classical development theorists put the emphasis on a positive succession of stages that all societies would have to follow to successfully achieve development (Rostow 1960). The fact that the program must lead to industrialization is not questioned. Second, dependency theorists insist on the submission of a periphery to a center (Galeano 1974; Walleirstein 1974; Amin 1977). The development of this center (the developed countries) depends on the maintenance of the periphery (the third world) in a state of underdevelopment. Here again, industrialization is not questioned but capitalism is not regarded as the appropriate pathway to achieve it. Third, postmodernists reject the notions of development, industrialization and modernization all together or are ambiguous in the way they consider these concepts (Sachs 1992). They usually fail to provide an alternate paradigm, probably because they reject the ontological necessity of development, instead of just rejecting the social constructions (the program) built upon this concept. In spite of the postmodernist critique, modernization in the sense of a march to industrialization is still, and by far, the dominant discourse when dealing with “developing” countries. I will show, following Dryzek’s arguments (2005), that industrialism is also the key concept upon which environmental discourses articulate. There is little doubt that this hegemony has been harmful and that the postmodern critique (Sachs 1992) was necessary and relevant. I would contend, however, that the concept of modernization should not be rejected for not falling, like in the case of development, into an ontological vacuum. According to the Oxford English Dictionary
382
On Line, 252 modernization is “the action or an act of modernizing something,” and modern means “being in existence at this time; current, present,” or “relating to the present and recent times, as opposed to the remote past.” There can thus be as many modernization processes as there are systems of values. But due to the framing of this concept in Western culture and the hegemony of industrialism as a development program, “industrial societies” now replaces “the present” as the benchmark to refer to when speaking of modernization. It is that benchmark that has to be criticized and questioned. There can be no world that does not try to adjust to its time, but there can be a world that does not consider industrialization as the benchmark for this adjustment. The more radical forms of Western modernization have been studied in detail by James Scott (1998), who called them High Modernism. Due to the huge economic, social, political and cultural gap that exists between “developed” and “developing” countries, the sometimes extreme realities Scott describes are particularly relevant to understanding how the dominant development paradigm operates in countries such as Madagascar. I will hence dedicate the following section to Scott’s work.
2.3. High Modernism according to James Scott For Scott (1998, 2), modern states need to make societies legible, to arrange the population in ways that [simplify] the classic state functions of taxation, conscription, and prevention of rebellion. The premodern state, conversely, was, in many crucial respects, partially blind; it knew precious little about its subjects, their wealth, their landholding and yields, their location, their very identity. It lacked anything like a detailed “map” of its terrain and its people. It lacked, for the most part, a measure, a metric, that would allow it to “translate” what it knew into a common standard necessary for a 252
http://www.oed.com/.
383
synoptic view. As a result, its interventions were often crude and selfdefeating. (Scott 1998, 2) In this perspective, modernization is the process by which states create legibility for the administration of their subjects. It includes the creation of permanent last names, the standardization of weight and measures, the establishment of cadastral surveys and population registers, the invention of freehold tenure, the standardization of language and legal discourses, the design of cities, and the organization of transportation. (Scott 1998, 2) This administrative engineering is not just aimed at creating legibility. States “enable much of the reality they depict to be remade” (Scott 1998, 3). Nature and society are reordered according to the canon of modernity. The society becomes “a reified object” (Scott 1998, 91) that can be scientifically described. It is reshaped in order to match with this description and be assigned more productive functions. “In place of the plasticity and autonomy of social life, [administrative engineering creates] a fixed social order in which positions [are] designated,” leading to various forms of “social taxidermy” (Scott 1998, 93). This administrative engineering project is what Scott calls high modernism. It is “a vision of how the benefits of technical and scientific progress might be applied – usually through the state – in every field of human activity” (Scott 1998, 10). Its ideology is a strong … version of the self confidence about scientific and technical progress, the expansion of production, the growing satisfaction of human needs, the mastery of nature (including human nature), and, above all, the rational design of social order commensurate with the scientific understanding of natural laws. (Scott 1998, 4) Some of the most extreme high modernist episodes are the great leap forward in China, collectivization in Russia, villagization in Tanzania and other ideologies “driven by utopian plans and an authoritarian disregard for the values, desires, and objections of their subjects” (Scott 1998, 7). These cases involved “an authoritarian state … willing and able to use the full weight of its coercive power to bring [its]
384
designs into being” (Scott 1998, 7). However, states are not the only promoters of high modernism: large scale capitalism is just as much an agency of homogenization, uniformity, grids, and heroic simplification as the state is, with the difference being that, for capitalists, simplification must pay. (Scott 1998, 8) High-modernism is challenged by approaches that put the emphasis on “practical knowledge, informal processes, and improvisation in the face of unpredictability” (Scott 1998, 6). Scott uses the Greek concept “metis” to depict this alternative. I will come back to this concept in Chapter 14. Scott proposes three examples of high modernism that are particularly relevant for the subject of this dissertation. A first example is scientific forestry. Its invention dates from the eighteenth century and is located in Prussia and Saxony. It influenced tropical forestry as early as the late eighteen century, in Indonesia (Peluso 1992), India (Guha 1990) and East Africa (Neuman 1998) for example. It led to ignoring the vast, complex, and negotiated social uses of the forest for hunting and gathering, pasturage, fishing, charcoal making, trapping, and collecting food and valuable minerals as well as the forest’s significance for magic, worship, refuge and so on. (Scott 1998, 12) The forest as a habitat disappeared and was replaced by the forest as an economic resource to be managed efficiently and profitably.…. The actual tree with its vast number of possible uses was replaced by an abstract tree representing a volume of lumber or firewood. (Scott 1998, 12) A geometrization and uniformization of the forest occurred, which was “intended to facilitate management and extraction” but “quickly became a powerful aesthetic as well” (Scott 1998, 18). Natural resources replaced nature and lines of trees replaced the disorder of natural ecosystems.
385
A second example is the cadastral survey and other mapping devices aimed at controlling land tenure and land uses. The modern state increasingly aspired to “take in charge” the physical and human resources of the nation and make them more productive.… An inventory of land, people, incomes, occupation, resources and deviances was the logical place to begin. (Scott 1998, 51) Mapping reduced the autonomy and the power of local communities, who resisted by acts such as land-invasions, squatting, and poaching. The third example concerns agriculture. State-sponsored high modernist agriculture has recourse to abstractions [such as] the simple “production and profit” model of agricultural extension, [whose] guiding idea was the maximization of the crop yield or profit. (Scott 1998, 263) High modernist agriculture considers the maximization of yield as its “holy grail” (Scott 1998, 293). It also developed a “visual aesthetic” (Scott 1998, 281) that gave more value to regular geometric pattern symbolized by monocultures in row. This geometry is aimed at facilitating the quantification and control of natural processes. Seedling densities and yields can be easily measured and factors that determined production can be separated from each other. High modernism also encouraged agronomists to work on a few variables, doing simple experiments where all other factors would be controlled. Artificial conditions were created that had little significance with regard to reality but could be described in simple terms. In the end, agronomy had “a weak peripheral vision” (Scott 1998, 292), especially in the context of complex and heterogeneous traditional systems. But what it needed to see was the reality it intended to create, not the existing one. The alternatives to tavy developed in Beforona, with their emphasis on yield and perfect biophysical design, match with this description, whereas they can be viewed as a soft version of high modernism.
386
In sum, high modernism failed in important ways to represent the complex, supple, negotiated objectives of real farmers and their communities, … the space in which farmers plant crops – its microclimates, its moisture and water movement, its microrelief, and its local biotic history. (Scott 1998, 262) This failure has been particularly visible in the case of shifting cultivation. From the perspective of a sovereign state and its field extension agents, [this land use was] an exceptionally complex and hence illegible form of agriculture [that] violated in almost every respect [the] understanding of what modern agriculture had to look like. (Scott 1998, 282) The fields were “fugitive, going in and out of cultivation at irregular intervals – hardly promising for a cadastral map” (Scott 1998, 282). The cultivators themselves were “moving periodically to be near their new clearings. Registering or monitoring such populations, let alone turning them into easily assessable taxpayers, [was] a Sisyphean task” (Scott 1998, 282). However, high modernism did not necessarily impede the rise of productivities. It even allowed it in many countries and regions of the world. It then must be asked why it also led to so many failures. For Scott, the main reasons were the inheritance of “unexamined (Western) assumptions about cropping and field preparation that turned out to work badly in other contexts” (Scott 1998, 264); the fact that “the actual schemes were continually bent to serve the power and status of officials and of the state organs they controlled” (Scott 1998, 264); a shortsightedness that led to focus on short-term productive goals and neglect of long-term outcomes and externalities; the incapacity to deal with complexity. It was not science and modern agriculture that had to be put into question. The culprit was rather the incapacity to integrate local experience, knowledge and culture into innovation and extension strategies.
387
3. Environmental discourses The content of this section is mostly derived from the book The Politics of the Earth, by John S. Dryzek (2005), with a few additional insights when indicated and when giving examples that relate to Madagascar. Dryzek’s essay matches exactly with the intent of this chapter. It proposes categories of discourses consistent with those I would have derived from my own representations and experiences. However, I will propose a different analytical frame to articulate in my own way these categories. According to Dryzek, a discourse is a shared way of apprehending the world.… Embedded in language, it enables those who subscribe to it to interpret bits of information and put them together into coherent stories or accounts. Discourses construct meaning and relationships, helping to define common sense and legitimate knowledge. Each discourse rests on assumptions, judgments, and contentions that provide the basic terms for analysis, debates, agreements, and disagreements. (Dryzek 2005, 9) Discourses are [further] bound-up with political power. Sometimes it is a sign of power that actors can get the discourse to which they subscribe accepted by others. Discourses can themselves embody power in the way they condition the perception and values of those subject to them, such as some interests are advanced, other suppressed (Foucault 1980).… Discourses are also intertwined with some material political realities. (Dryzek 2005, 9) This way to consider discourses is clearly influenced by the work of Michel Foucault, upon whom more will be said in the next chapter, but some distinctions are also explicitly made. For Foucauldians, “individuals are for the most part subject to the discourses in which they move, and are seldom able to step back and make comparative assessments and choices across different discourses” (Dryzek 2005, 22). In Foucault’s writing, discourses are portrayed “in hegemonic terms, meaning that one single discourse is typically dominant in any time and place, conditioning not just agreement but the terms of the dispute” (Dryzek 2005, 22). Dryzek disagrees with that. For him, “discourses are powerful, but they are not impenetrable” (Dryzek 2005,
388
22). Dryzek will thus attempt to describe “the basic structure of the discourses that have dominated recent environmental politics” (Dryzek 2005, 10). He will “present their history, conflicts and transformations” (Dryzek 2005, 10) and delineate their core determinants. Whether Dryzek is right or wrong does not really matter in this chapter, which is merely aimed at reviewing Western environmental discourses in order to provide a background to Chapter 10 and 11.
3.1. Dryzek’s analytical framework Dryzek starts his analysis by referring to a distinction, made in most histories of
U.S.
environmentalism, between
“the anthropocentric rational resource
management [perspectives], advocated by the U.S. Forest Service’s first chief forester Gifford Pinchot, and the deeper respect for nature propounded by Sierra Club founder John Muir” (Dryzek 2005, 11). Pinchot emphasized the utilitarian value of natural resources. Muir, on the other hand, put at the forefront an esthetic and spiritual dimension of nature that would be worth conserving for its intrinsic value and that Americans could enjoy through outdoor activities. These anthropocentric and biocentric ways of thinking remained in conflict until present time. Muir’s perspective has received increasing attention in recent decades. What most Americans used to consider as an agricultural frontier to be conquered is now regarded as a wilderness to be preserved. Dryzek further analyzes the wider frame within which environmentalism developed. First, he reminds us that environmental discourses “have to be positioned in the context of the long dominant discourse of industrial society” (Dryzek 2005, 13), which he calls Industrialism. In Dryzek’s words, “Industrialism may be characterized in terms of its overarching commitment to growth in the quantity of goods and
389
services produced and to the material well-being that growth brings” (Dryzek 2005, 13). However, this framing “eventually started to disintegrate, yielding the range of environmental discourses now observable” (Dryzek 2005, 22). Two criteria determined the pathways to depart from industrialism. First is the distinction between “reformist” and “radical” discourses (Dryzek 2005, 14). This distinction refers to the pace and scale of change. Reformists support a step by step change along a series of pragmatic reforms, whereas radicals consider that a wholesale transformation of the society is necessary to cope with environmental issues. The second criterion is a distinction between “prosaic” and “imaginative” discourses (Dryzek 2005, 14). It relates to the way environmentalism articulates with larger frames, and especially with industrialism. “Prosaic departures take the politicaleconomic chess-board set by industrial society as pretty much given” (Dryzek 2005, 14). “Environmental problems are seen mainly in terms of troubles encountered by the established industrial political economy” (Dryzek 2005, 14). Hence action is required but does not imply moving to a different kind of society. “In contrast, imaginative departures seek to redefine the chessboard” (Dryzek 2005, 14). “The environment is brought into the heart of the society and its cultural, moral and economic systems” (Dryzek 2005, 14). It provides an opportunity to put industrialism into question. By cross-checking these two criteria, Dryzek distinguishes four main environmental discourses, as shown in the table below: Table 18: Classification of environmental discourses. Source: Dryzek (2005, 15). Reformist Radical Prosaic Environmental problem Survivalism solving Imaginative Sustainability Green radicalism I will present now the environmental discourses and their sub-categories, as identified by Dryzek.
390
3.2. Dryzek’s categories 3.2.1. Survivalism “Survivalism is the discourse popularized in the early 1970s by the Club of Rome” (Dryzek 2005, 15), through the publication of The Limits to Growth (Meadows et al. 1972). “The basic idea is that continued economic and population growth will eventually hit limits set by the Earth’s stock of natural resources and the capacity of its ecosystems to support human agricultural and industrial activity” (Dryzek 2005, 15). Survivalism is fundamentally Malthusian. Key concepts that laid the foundation of survivalism are Garrett Hardin’s (1968) Tragedy of the Commons and Ehrlich and Ehrlich’s (1970) Population Bomb. Ehrlich and Ehrlich (1970) put into question the right to give birth and Hardin (1974) went still farther by proposing his famous “lifeboat” metaphor, which pictures developed nations as committed to their own survival while developing countries are allowed to drown in misery, unless they accept controlling their population. This quite extreme perspective reveals a tendency of survivalist discourse to defend technocratic and authoritarian policies that could open the doors to some forms of totalitarianism, as also argued by Rossi (2003). William Ophuls, for example, recommended the “establishment of a governing class of ecological mandarins” (Ophuls 1977, in Dryzek 2005). In such a scheme, the absence of democratic feed-back could impede local knowledge from being taken into account. In the best cases, short term results in term of environmental conservation could be achieved but at the cost of ethical violations that would render long term results questionable. In the worst cases, received wisdom would spread without obstacle and result in irrational policies whose impact may be both unethical and the reverse of their objectives. The lobby of international organizations to forbid agricultural fires in
391
the Toamasina Province, in Madagascar, provided an illustration of this logic (Chapter 6). In sum, survivalism is radical because it questions the permanence of growth, which is a key feature of all industrial societies, and prosaic because “it can see solutions only in terms of the options set by industrialism, notably, greater control of existing systems by administrators, scientists, and other responsible elites” (Dryzek 2005, 15). Most of its prominent early figures have a background in biology but they have been joined in 1995 by a few leading economists (Arrow et al. 1995, in Dryzek 2005, 35), who defended less radical positions. Authoritarianism was softened and more emphasis was put on the “creation and enforcement of environmental quality standards” (Dryzek 2005, 37), opening the way to the “problem solving” discourses which I will present later.
3.2.2. Promethean As a dominant environmental discourse during several decades, survivalism received counter-attack from defenders of the established industrial economy. These defenders argued that humans are characterized by unlimited ingenuity, symbolized in Greek mythology by the progress made possible by the theft of fire from the gods by Prometheus. (Dryzek 2005, 25) Dryzek calls these counter-attackers “prometheans.” They increased their influence in the 1980s. They asserted that the resources of the earth were wholly unlimited and that “as soon as one resource threatened to run out, ingenious people would develop a substitute” (Dryzek 2005, 26). Key figures of the promethean discourse are Barnett and Morse (1963), who contended that decreases in the price of natural resources since the beginning of the century proved there was no scarcity. Simon (1981) continued these arguments and contended that there was no scientific
392
proof of global environmental degradation. Lomborg (2001), who became famous through his book The Skeptical Environmentalist, defended a similar thesis. The U.S. administration, under the presidencies of the Bushes and Reagan, has been strongly influenced by promethean discourses. In spite of this revisionist character that may be detrimental to knowledge advancement, promethean discourse can be meaningful if envisioned in a more moderate perspective. The key argument, that “nature is, indeed, just brute matter…infinitely transformable” given enough energy is available (Dryzek 2005, 57), is theoretically acceptable. The flows of energy and matter used by human societies are still low in comparison with the total flows of the world we live in. Energy consumption could greatly increase if we would switch to renewable sources. But to expect that governments could develop appropriate technologies in time to remedy environmental problems, and that markets would provide the right incentives to adopt these technologies is certainly just a dream. With respect to the impacts of our industrial activities, outside of science-fiction literature, the efficiency of new technologies to solve environmental problems will remain only anecdotal for the near future. Nevertheless, abandoning science-fiction and getting back to more trivial realities, the promethean belief in science provides useful criticisms to the excessively Malthusian perspective of survivalism. For Julian Simon, “the ultimate resource is people.… Skilled, spirited, and hopeful people will exert their wills and imagination for their own benefit, and [for] the rest of us as well” (Simon 1996, 589, in Dryzek 2005, 59). So, “the more people, the better” (Dryzek 2005, 59). In the case of changes in agricultural development, this argument is consistent with Boserup’s (1965) hypothesis presented in Chapter 5. The promethean perspective helps in understanding
393
how communities of farmers in a situation of crisis eventually develop more intensive systems, allowing the population to grow and livelihoods to improve. In sum, promethean discourse can be viewed as the econocentric response to the biocentric character of survivalism, but an answer that became too radical. This bipolarity presents some similarities with the opposition of the biocentric and anthropocentric perspectives of Muir and Pinchot, but is not superimposable since survivalism has its foundation in both Muir’s and Pinchot’s ideals. Two bipolarities exist and both have to be utilized if we are to adopt a comprehensive analytical framework. Muir’s and Pinchot’s perspectives, as well as survivalist and promethean discourses, can all contribute to a better understanding of our world.
3.2.3. Environmental problem solving Environmental problem solving is defined by taking the political-economic status quo [of industrial societies] as given but in need of adjustment to cope with environmental problems, especially via public policy. (Dryzek 2005, 73) Both state regulations and market mechanisms can be used to achieve these goals. The approach is prosaic because it is implemented using the tools of the industrial society, and reformist because changes are implemented by successive reforms. The three main pathways for doing so are “administrative rationalism, democratic pragmatism, and economic rationalism” (Dryzek 2005, 73).
3.2.3.1. Administrative Rationalism Administrative rationalism emphasizes the role of administrations to solve problems. Experts are called to assess environmental issues and propose solutions that are translated into policies. State organizations such as national park services and pollution control agencies are created and new regulations are issued.
394
A first limit to this approach is that it solves problems only when they become political issues. Environmental degradations that have gradual impact are insufficiently addressed because people get used to them and tolerate them more than catastrophic events. Administrative rationalism also tends to displace problems: administrations act by sectors, at geographic scales determined by administrative boundaries, and lack coordination, which limits their capacities to address global and comprehensive issues (Dryzek 2005). Eventually, administrative rationalism is faced with a conundrum when problems are global but solutions have to adapt to local realities: “the more an organization is disciplined, the less it can be expected to learn” from the local, and “the more it learns, the less easy will it be to maintain discipline by administrative means” (Dryzek 2005, 96). This problem was also raised by Scott (1998) in his analysis of “high modernism” and is particularly relevant in the case of weak states. As an answer to this limitation, the most recent methods developed by government to manage their resources put the emphasis on local knowledge, flexibility (adaptive management) and negotiation. By moving from a culture of government to a culture of governance, the administration rationale could escape the conundrum: “a central agency [would] set standards, while compliance [would be] negotiated locally with activists and corporation” (Dryzek 2005, 96). But the distinction with the next sub-discourse (democratic pragmatism) may then not be clear.
3.2.3.2. Democratic pragmatism Democratic pragmatism is influenced by the philosophy of pragmatist philosophers such as John Dewey (Ratner 1939). For these philosophers, the world is full of uncertainties that science cannot remove, but that we need to consider to
395
determine our acts. In consequence, experimenting, in the wide sense of being confronted directly with reality, is essential. As Dryzek puts it, The most rational approach to problem solving, in life as in science, involves learning through experimentation. For problems of any degree of complexity, the relevant knowledge cannot be centralized in the hands of any individual or administrative state structure. [It must involve] many voices, and cooperation across a plurality of perspectives. (Dryzek 2005, 100) All stakeholders are involved in negotiations prior to decision and neutral mediators can be called to solve disputes and help to find the best consensus. Impact assessment, public enquiries, decentralization, participatory methods, right-to-know legislation, and alternative dispute resolutions are concepts and tools that took their impetus from this new framework (Dryzek 2005). The experiment conducted by the NGO Zanaky ny Ala in Beforona (Chapter 7) illustrates this approach to problem solving through experimentation. Chapter 11, about the GELOSE legislation, will analyze another attempt to implement this approach in Madagascar. Democratic pragmatism thus confirmed the shift from “government” to “governance.” Government has a top-down imagery in which administrative rationalism rules once goals and principles have been set.… Governance, in contrast, dispenses with a central locus of authority, revealing in informal interactions.… The image of a network replaces that of the hierarchy. (Dryzek 2005, 109) In this network interactions occur such as “committee meetings, legislative debate, hearings, public addresses, legal disputes, rule-making, project development, media investigation, and policy implementation and enforcement,” which involve “lobbying, arguing, advising, strategizing, bargaining, informing, publishing, exposing, deceiving, image building, insulting, and questioning” (Dryzek 2005, 108). The working groups of NEAP (Chapter 2) illustrate the increasing momentum given to democratic pragmatism and governance in policy making.
396
Critics of democratic pragmatism argue that its chaotic character puts it at odds with a scientific and rational administration. But Karl Popper showed that science itself advances in such a chaotic schemes. He advocated for an open and democratic problem solving approach in politics, which he called “piecemeal social engineering” (Popper 1966, in Dryzek 2005). In this perspective, as in the case of pragmatic philosophy, the implementation of a policy becomes comparable to a scientific experiment. Uncertainty is both accepted and addressed. Chapter 15 will come back on this approach. Democratic pragmatism is exposed to some limitations. First “is the simple existence of political power” that can have “large financial resources at their disposal” and “try to skew the outcomes of policy debates and decision making processes in their direction (Gonzales 2001)” (Dryzek 2005, 117). Chapter 11 will provide a good illustration of this problem. Second, stakeholders can think as citizens but still act as consumers. Instead of changing their behavior, they expect that the state will use its authority and expertise to enforce regulations that will oblige them to behave as good citizens (Dryzek 2005). Global warning provides a good example: the public awareness of this event increases constantly, but results more in expectation of state and international regulations than in changes in behavior. The solution may be an appropriate balance between administrative rationalism and democratic pragmatism. In sum, democratic pragmatism and administrative rationalism both allowed some achievement. As in the case of survivalism and prometheans, the two discourses have to be regarded as complementary rather than contradictory. But they may also have to articulate with the third sub-discourse: economic rationalism.
3.2.3.3. Economic Rationalism Economic rationalism may be defined by its commitment to the intelligent deployment of market mechanisms to achieve public ends. It differs from
397
administrative rationalism in its hostility to environmental management on the part of government administrators - except, of course, in establishing the basic parameters of designed markets. (Dryzek 2005, 121) Economic rationalism considers that market and private properties are the solutions to the Tragedies of the Commons. When dealing with resources that cannot be privatized, such as air and water, it proposes the internalization in market mechanisms of pollution and other environmental externalities, through regulation and legislation. Green taxes can be created to oblige individual consumers to pay a price that better reflects the real cost of what they buy. The trading of emission rights makes it possible to reduce the cost of pollution control. This mechanism helped to achieve targets in the reduction of industrial sulfur pollutants in the United States and is included in the Kyoto protocol as a mechanism to reduce green house gases (Clean Development Mechanism). The VMCRCC project analyzed in Chapter 6 provided an illustration of this approach. In these internalization mechanisms, the market is not a consequence of the behavior of industries and consumers. It is artificially created by states or international organizations in order to answer to market failure (externalities). According to Dryzek, nature, in the perspective of economic rationalism, “exists only to provide inputs to the socio-economic machine, to satisfy human wants and needs” (Dryzek 2005, 134). “Missing from economic rationalism is any notion of active citizenship” (Dryzek 2005, 135). Humans are reduced to their Homo economicus dimension and nature to its marketable value. This argument is, however, disputable because there is no reason to raise a green tax on behalf of an intrinsic value of nature if this value is recognized. It is true, however, that the utilitarian perspective that prevails in our world does not encourage using economic rationalism to defend ethical values.
398
3.2.4. Sustainability Problem solving approaches do not state clearly where they posit themselves between anthropocentrism and biocentrism. They do not address explicitly the question of limits. Their concern is empirical and local, rather than theoretical and global. They focus “on the work to be done in the here and now” (Dryzek 2005, 142). Sustainability, on the other hand, attempts to solve this conflict between anthropocentrism and biocentrism and to combine environmental conservation and economic growth. I will show that, unfortunately, the sustainability discourse has been a victim of its ambitions. Sustainability appeared first under the label sustainable development. A second sub-discourse developed later, which Dryzek called ecological modernization. Dryzek addressed these two sub-discourses separately.
3.2.4.1. Sustainable development For Dryzek, sustainable development is an imaginative attempt to dissolve the conflicts between environmental and economic values that energize the discourses of problem solving and limits. The concepts of growth and development are redefined in ways which render obsolete the simple projections of the limit discourse. (Dryzek 2005, 16) The term appeared in the 1980s and was coined by the World Commission on Environment and Development of the United Nations (WCED), which defined it as follows: sustainable development is a process of change in which the exploitation of resources, the direction of investments, the orientation of technological development, and institutional change are all in harmony and enhance both current and future potential to meet human needs and aspirations. (WCED 1987) In other words, humanity “meets the needs of the present without compromising the ability of future generations to meet their own needs” (WCED
399
1987, in Dryzek 2005). This definition is one of the most frequent citations in the sustainable development literature. If we look closer, the WCED Sustainability discourse appears to balance more on the anthropocentric side. “The environment does not exist as a sphere separate from human ambitions, action and needs…the environment is where we all live” (WCED 1987, in Drykek 2005, 153). The intrinsic value of nature, which may exist independently of our “ambitions, action and needs” (WCED 1987, xi, in Dryzek 2005, 153), is not explicitly recognized. What has to be conserved is the environment that surrounds us, not necessarily a nature “out there.” The sustainable development discourse also departs from survivalism and balances toward a moderate form of promethean discourse. According to Dryzek, it contends that “growth has no set limits in terms of population or resource use beyond which lies ecological disaster,” in part because “accumulation of knowledge and the development of technology can enhance the carrying capacity of the resource base” (Dryzek 2005, 154). In spite of this, both anthropocentric and biocentric perspectives appropriated the sustainable development discourse in their own way. Sustainable development is now a paramount concept in development, business and environmental circles alike, where it reflects quite different objectives. Brundtland’s definition has been repeatedly quoted and other definitions proliferated, questioning “what is to be sustained, for how long, for whom, and in what terms” (Dryzek 2005, 146). In the end, sustainable development may have become “an empty vessel that can be filled with whatever one likes” (Dryzek 2005, 148), a point with which Dryzek disagrees. My personal understanding is that sustainable development is indeed an oxymoron. Table 17 showed that development, the unfolding of new potentialities according to a program, has a beginning and an end. When the program finishes, i.e.
400
when an organism or a society achieves the complete unfolding of its potentialities, chemical or social entropy may lead to crisis, decay and death. Societies may need to be reinvented to remain dynamic. They have to evolve, a process which should not be confused with development as we saw. The concept of sustainable development created this confusion. Those who believe in it keep on hoping that the same program, the same system of values can be indefinitely conserved and support population and GNP increase, with just a few corrections. Those who reject it refuse the idea that human population could continue their expansion, because they are detoured from the concept of evolution and from the possibility of adopting new and more appropriate “programs.” In sum, the Sustainability discourse can be seen as an attempt to reconcile the anthropocentric and biocentric perspectives, or to create synergies between economic growth and environmental conservation, but an attempt that has failed due to a confusion between development and evolution. Things can evolve forever but development has a limit which is set by its program and the environment within which this program runs. This misleading confusion between two concepts is revealed by Dryzek’s assertion that in a world dominated by market liberalism, sustainable development’s prospects are poor unless it can be demonstrated that environmental conservation is obviously good for business profitability and economic growth everywhere, not just that these competing values can be reconciled. (Dryzek 2005, 160-161) In other words, liberalism is the program that has been adopted by our society and a change in this program may need to occur in order to allow future evolution, a process
which
was
wrongly
called
sustainable
development.
Ecological
modernization, which is more imaginative than sustainable development, may reveal the first attempts to switch to new programs.
401
3.2.4.2. Ecological modernization Ecological modernization refers to a restructuring of the capitalist political economy along more environmentally sound lines…. Environmental criteria must be built into the re-design of the system. (Dryzek 2005, 167) Industry is invited to cooperate to this redesign. It receives incentives to “embrace rather than resist ecological modernization, provided only that business is sufficiently far-sighted rather than interested only in quick profits” (Dryzek 2005, 167). Key assumptions are that pollution are “a sign of waste” (Dryzek 2005, 167), that solving them in the future will be “vastly more expensive for both business and government” (Dryzek 2005, 168), and that “there is money to be made in selling green goods and services” (Dryzek 2005, 168). The challenge is to guide industrial societies to an “environmentally enlightened era” and to “re-establish the commonality” (Dryzek 2005, 171) of economics and ecology, remembering that both concepts are derived from the same Greek word oikos (household). Ecological modernization could be the arena where the sustainability concept achieves its maturation, through a commitment to practical objectives and comprehensive societal changes. However, being in continuity with sustainable development, ecological modernization “does not easily admit the idea that nature might have intrinsic value beyond its material uses” (Dryzek 2005, 179). It also maintains the conflation between the concepts of modernization and industrialization. It can thus be questioned whether it is a significant departure from Western societies driven by utilitarian values and whether it significantly opens the way for imaginative evolutions. Moreover, “ecological modernization was long silent about what might be the appropriate developmental path for Third World societies, and attempts by theorists to
402
apply ecological modernization concepts to these societies often misfire” (Dryzek 2005, 172).
3.2.5. Green radicalism Green radicalism may be the most imaginative discourse; the only one that could be located on the other side of the great partition that determined the march to industrial societies. Its adherents reject the basic structure of industrial society and the way the environment is conceptualized therein in favor of a variety of quite different alternative interpretations of humans, their society, and their place in the world.… It is both radical and imaginative. (Dryzek 2005, 16) Its project is to provoke an in depth change of human and societal values and behaviors, which determine two sub discourses: green consciousness and green politics.
3.2.5.1. Green consciousness One of the most popular forms of green consciousness is deep ecology, which advocates for a minimal human footprint on the planet and, in its more radical forms, for the reduction or extinction of human population. As green consciousness is anecdotal or absent from Madagascar, I will not present in length the movement and its variations such as ecofeminism, bio-regionalism or eco-theology. Nevertheless, this discourse is important because it can help to delineate the frontier between discourses embedded in industrialism and their alternatives. Romanticism can be seen as the alternate pathway that Western societies could have chosen instead of industrialization. Historically, the term “was attached to an intellectual movement in the eighteenth and nineteenth centuries that opposed the rise of modern science and liberal politics” (Dryzek 2005, 191-192) that characterized the
403
Enlightenment. Green consciousness posits itself on the romantic side. It considers that “environmental destruction [is] caused by modern science and technology wielded in human arrogance” (Dryzek 2005, 192). Green consciousness is idealistic, in the sense that it considers that “ideas, not material forces, … move history” (Dryzek 2005, 194). Supporters of green consciousness appeal to new wisdoms, as is expressed by their taste for spiritual experiences, contemplation and harmony with unknown forces. If nature is what is set out of the control of human societies, green consciousness replaces it at the center of their life. It considers that nature cannot coexist with rationale industrial societies and see it as a source of knowledge to be used for the reconstruction of new systems of values. The adept of this movement take the risk of advancing into new realms of knowledge in spite of the prejudice this creates with respect to the way they are perceived. In sum, green consciousness could help to recover a dimension of our world that was lost after the great partition of the enlightenment. This dimension could be a harmony with the unknown and the uncontrolled. It could help us to re-discover things that were not given paramount importance in the representation of our world that we created and in the societies and the environment that we built from these representations. Green consciousness movements, however, “cannot tell us how to get from our current severe disequilibrium to this harmonious state” (Dryzek 2005, 201). They have often taken radical positions that made their movement unpopular and as exclusive as the ideologies they criticize. From their statement that there is a conflict between nature and culture, they often end in rejecting human societies and cultural productions altogether. In short, because it was the most radical and imaginative departure from industrialism, green consciousness also engendered some of the most
404
exclusive and intolerant environmental discourses. Instead of opening up society, it closed it in a different way. Green politics may be the answer to these excesses.
3.2.5.2. Green politics Green politics is an attempt of Green Radicalism to enter the arena of political life and decision making, through the creation of political parties aimed at challenging the dominant development pathways of modern societies. This was done at the cost of internal divisions, the more radical members preferring to confront the political system rather than working within it. Green politics also favored the establishment of links with other activist networks, on themes such as environmental justice, development of the poorest nations, globalization, and global justice. These connections were manifest during the global protest that accompanied the global conference of the World Trade Organization in Seattle, in 1999 (Dryzek 2005). Green politics adopted more moderate positions concerning the way it regards human activities and culture. One of its sub-movements, Social Ecology, treats humans as part of nature, but “the only bit of nature that has yet achieved selfconsciousness” (Dryzek 2005, 206). Culture would be a nature “aware of itself” (Dryzek 2005, 206). Culture is viewed as being part of nature, or as Bookchin (1986, in Dryzek 2005, 206) puts it, a “second nature.” The intrinsic value of nature is still recognized but extended to culture, for the same ethical reasons. In sum, green politics is an imaginative attempt, embedded in radical movements, to create a new world where the relationship with nature will be modified and oriented to a balanced stewardship instead of a submission of nature to humans. It “appeals to ideals of progress beyond an irrational industrial order, rather than promises return to some primal eden” (Dryzek 2005, 218). Green politics was influential in Germany and in Scandinavian states and contributed significantly to
405
integrate green discourses into political debates. But it lacked an elaborate blueprint for defining a new societal project. In consequence, ecological modernization often absorbed the ideas of green politics and made them lose their imaginative capacity. Between a political pragmatism and a creative radicalism, green movements face a dilemma. Green politics, in the end, could become “just a minor set of irritants to [a] monolithic, global, transnational, capitalist political economy” that is “more secure and more entrenched than ever in history” (Dryzek 2005, 226).
3.3. Reframing Dryzek’s categories I clearly adopt the categories of discourses that Dryzek identifies. However, I propose an analytical framework that is different from the one proposed by Dryzek (Table 18 page 390). Environmental discourses can be organized along an axis whose extremes are the most radical version of the limit discourse on one side (survivalism), and the promethean negation of limits on the other side. Different positions on this axis reflect the way interactions between human beings and their physical and natural environment are seen. On one side, we find a broken harmony that has to be recovered if we are to survive, while on the other side, the environment is permanently reshaped thanks to an infinite supply of energy. A second axis can be drawn to map John Muir’s and Gifford Pinchot’s philosophies. It would reflect the value that is given to nature: an immanent realm with intrinsic value on one extreme, and a reservoir of resources whose value is proportional to their utility on the other extreme. The two axes do not superimpose. Promethean discourse clearly rejects the limit discourse and adopts Pinchot’s utilitarian anthropocentric perspective. Adepts of survivalism, to the contrary, adopt the limit discourse but can be either romantic or
406
utilitarian; biocentric or anthropocentric. Sustainable development is an attempt to solve the conflict between promethean and survivalism perspectives, or between economy and ecology, but failed to achieve this synthesis. Sustainable development eventually adopted a rather utilitarian perspective and challenged the limit discourses. Concerning green consciousness, it is radically biocentric and survivalist. These four discourses (survivalism, promethean, sustainable development and green consciousness) make it possible to map the different ways human beings think about their environment, at least in Western societies and cultures. They help us to understand how we conceive our environment philosophically. The other discourses (problem solving [administrative rationalism, democratic pragmatism and economic rationalism], ecological modernization, and green politics), to the contrary, relate to the strategies we can employ to manage the environment in practice, independently of the way the environment is conceived. It would not make sense to map them according to the two axes I defined. Problem solving, for example, proposes tools and methods to implement policies according to three sub-strategies that focus on state regulation, public participation, or market mechanisms. I agree that problem solving is reformist, in the sense that it proposes incremental changes on specific issues, and prosaic, in the sense that it does not see beyond the framework of industrial societies. This makes problem solving distinct from ecological modernization and green politics. But problem solving can be applied by adherents of promethean, survivalism or sustainable development discourses. Ecological modernization may be the translation of sustainable development discourses into policies. It is more imaginative because this comprehensiveness obliges it to question the way industrial societies work. It can be adopted by the adherents of promethean, survivalism, sustainable development, and even green
407
consciousness. As it addresses large scale issues, it goes beyond problem solving but can integrate its tools and methods, which determines several sub-currents. For example, “weak ecological modernization” emphasizes “a technocratic corporatist style of policy making monopolized by scientific, economic, and political elites” (Dryzek 2005, 173) (as does administrative rationalism) whereas “strong ecological modernization” features “open, democratic decision making” (Dryzek 2005, 173) maximizing “participatory opportunities to citizen” (Dryzek 2005, 173) and communication (as does democratic pragmatism). Green politics is the entry in the political arena of green radicalism. It is both radical and imaginative because its role is to question and provoke from the verge. But when its proponents become actors that implement policies (when they get elected), pragmatism obliges them to revert to ecological modernization, which actually “developed in the European heartland of green party politics” (Dryzek 2005, 219). The lack of clear distinction between ecological modernization and green radicalism may reflect the dilemma of choosing between changing systems from the inside, with a range of actions determined by subordination to a structure, or changing them from the outside, as a free thinker involved in radical activist movements. In sum, survivalism, promethean, sustainable development and green consciousness perspectives provide the key hallmarks to understanding environmental discourses. The three first analyze the industrial society from different perspectives but do not put it into question. Achieving material wealth is still accepted as being the main goal of human development. This has to be done with respect for the environment, in the case of survivalism and sustainable development, and is not regarded as irreversibly harmful to the environment in the case of prometheans. Conventional science and rational thinking are still viewed as appropriate and sufficient tools to achieve human progress. Green consciousness, conversely, attempts
408
to analyze Western society from outside and to put into question the dominant paradigms within which Western people are all embedded. Unfortunately, this leads this discourse to a disconnection from the society, to an incapacity to propose pragmatic pathways to change, and sometimes to misanthropy. The other “discourses” (problem solving, ecological modernization and green politics) are translations into policies of these four perspectives. Their combination depends more on the actors’ position (more or less embedded in the system, more or less pragmatic, committed to local or global issues) than on the way they actually see their environment and its relations with society. This framework enables one not to choose a camp among the discourses identified by Dryzek (2005). Discourses are representations of the reality; not the reality itself. All contribute to a comprehensive understanding and can be seen as ingredients to pick up and combine in different ways according to what category of issue is being dealt with (local or global, practical or theoretical, political or technical). The same person could adopt survivalism when analyzing a Malthusian agroecosystem crisis, be promethean (Boserupian) when theorizing about the long-term future of this same agro-ecosystem, promote sustainable development if located in a favorable position to influence this future, and exhibit green consciousness when recognizing in what remains of a natural ecosystem an intrinsic value that has to be preserved. If involved in the implementation of policies or programs, this same person could adopt problem solving and pragmatically combine administrative rationalism, democratic pragmatism, and economic rationalism, according to the resources available and the current institutional, social and economic environment. With more power and more financial resources, he could envision ecological modernization. As an activist in a lobby group or member of a green party, he could be more radical and
409
imaginative and adopt green politics. But he may have to move back toward ecological modernization if elected.
4. Environmental discourses in Madagascar It is now time to go back to the Malagasy realities, in the light the analytical frameworks proposed in this chapter. This will allow a new reading of the Malagasy environmental policies presented in Chapter 2. I had the chance to directly experience the reality of environmental discourses and policies in Madagascar, by working during two years as a technical advisor for the Water and Forest General Directorate, at the Malagasy Ministry of Environment, Water and Forest. 253 As I mentioned earlier, I was a member of several working groups that advised the Malagasy government with respect to the implementation of the third phase of the National Environmental Action Plan (Chapter 2). My colleagues were representatives of Malagasy state services and of their partners (public organizations, bilateral and multilateral donors, NGOs, civil societies, and the private sector). I became a part of my study object and promoted some of the discourses and strategies analyzed in this dissertation, including some of those I criticized. The section that follows is mostly derived from this experience. Due to this particular situation, I did not use any particular methodology to carry out this part of the “research.” During the two years I spent in the decision makers’ arena, I had no research plan, no questionnaire and no schedule. My job led me to question myself about what was going on in these various working group meetings, and out of these meetings, as they were not the main location where
253
Decentralized Natural Resources Management Project of the Priority Aid Fund of the MalagasyFrench Cooperation. In French: “Projet Gestion Décentralisée des Ressources Naturelles du Fonds de Solidarite Prioritaire de la Coopération Franco-Malgache”, financed by the French Ministry of Foreign Affair. I worked on this project from February 2004 to February 2006.
410
decisions were taken. I had not yet read Politics of the Earth by Dryzeck (2005). I knew about sustainable development, deep ecology, green politics, and the works of Muir, Pinchot, Hardin, and Ehrlich, but I did not articulate these elements in a clear system of representations. However, a picture came eventually, through a rather intuitive process. I will now describe this picture, using Dryzek’s words and linking it with the conclusions of Chapter 2. The adherents of Muir and Pinchot clearly confronted each other in the Malagasy political arena at the central level. But the confrontation was not between Malagasy decision makers. It was rather between their partners. On one side was a group of actors whose main goal was to enlarge the system of protected areas from 1.7 to six million hectares, in accordance with the commitment made by the Malagasy President at the National Park Congress of Durban, in September 2003. These protected areas would only allow the strict minimum of human activities, as we saw in Chapter 2. Under this vision, the Malagasy state was encouraged to relocate forest exploitation from the harvest of valuable trees in natural ecosystems to plantation forestry. This group was led by conservation NGOs such as Conservation International, the Wildlife Conservation Society and the Durrell Wildlife Trust, and by the main donors to the Environmental Action Plan (the World Bank and USAID), with sometimes some hesitation for the later. Muir’s vision clearly dominated the discourses of this group, while administrative rationalism was the privileged approach for implementation. On the other side were the bilateral agencies, some research projects and some conservation and rural development agencies or NGOs who held that natural resources, including timber, could be harvested, under strict regulation, in the new protected areas that would be created. No activity would be excluded a priori but sustainability had to be proven and was the condition to allow forest product
411
extraction under the status of protected area (Chapter 2). Industrial exploitation involving powerful companies was rejected. The goal was to allow local communities to sustain their livelihood from the extraction of natural resources, including wood. This extraction could be market oriented but would remain traditional (no roads, no machines). The GELOSE legislation (Chapter 11) would provide the tools to implement this system. The state would get its share of the benefit through taxation. It was assumed that management by local communities, the supplementary means provided to authorities and the presence of a partner committed to conservation, due to the classification into protected area, would make possible successful control and tax collection. This group was led by the German and French cooperation, SAGE 254 and the WWF. The heritage of Pinchot was clearly present and democratic pragmatism characterized the approaches that were promoted. This contrast was, however, not explicitly formalized and democratic pragmatism was officially accepted by both groups. The GELOSE legislation was envisioned as a frame to empower local communities to achieve conservation as well as development. The consequences will be analyzed in Chapter 11. The sustainable development discourse was also accepted on both sides but different meanings were implicitly given to the term, “thanks” to the oxymoronic character of this term. Sustainable development provided the illusion that a consensus existed, which provided more confusion than clarification to the debate. Malagasy officials usually let their partners have their polemics during the meetings. But when they were asked their opinion, as occurred sometimes when debates became too animated, they clearly chose the second group. Forests included great resources, with high value on the international market, and most Malagasy
254
Environmental Management Service. In French: Service d’Appui a la Gestion de l’Environnement. See Chapter 2.
412
officials believed that these resources had to be harvested. Management had to be improved to make this extraction sustainable and all recognized the many risks and difficulties. But to abandon the objective of sustainable extraction did not raise any enthusiasm due to the loss of potential income for the country and the discontentment of the private sector. Nevertheless, when decisions were made by the sovereign state, they mostly followed the recommendation from the first group, which was that of the main donors. Hence, Muir won the debate and the area of forest dedicated to his vision was much wider than that dedicated to Pinchot’s vision, as we saw in Chapter 2. This was, however, only virtual. On the ground, forests continued to be exploited. Many Malagasy officials tolerated it, due to important regional, private and political lobbies. Muir’s vision just forced the extraction of forest products to remain illegal. Concerning the second axis of analysis (the opposition between survivalist and promethean discourses), survivalism prevailed in the first group, as illustrated by the strategies of projects operating in Beforona (Chapter 6). Madagascar was regarded as overpopulated, leading to a Malthusian vicious circle of environmental degradation and poverty that was transforming the green island into a red one. Population had to be controlled. In the second group, survivalism was not the rule. Deforestation and the existence of limits were accepted and efforts were made to de-dramatize the situation. Boserupian dynamics were more often evoked. There was also something promethean in both groups. The same technical or legislative solutions, such as agroforestry, SRI, zero tillage, the GELOSE, or zoning were promoted on both sides. A soft, high modernist project pervaded in all representations. The sciences, whether biophysical or social, were expected to find solutions to the existence of limits. Experts and engineers were called upon to propose the most appropriate pathways to achieve sustainable development. In sum,
413
survivalism prevailed when assessing the state of the environment while a promethean dream dominated when envisioning the remedies. Concerning green consciousness, there was no such movement in Madagascar. Nevertheless, we can say that the first group gave more intrinsic value to the forest than the second. This argument was, however, marginal in the debates. Both groups preferred to refer to the utilitarian and economic value of ecosystems, whether this value would result from forest products extraction, tourism, or environmental services. We arrive now at the second set of discourses, which relate to the way environmental policies are implemented. Ecological modernization and green politics were totally absent or were not expressed. Clearly, the privileged approach was problem solving, the problems to solve in priority being deforestation, slash-and-burn cultivation, erosion and fires. Problem solving employed a mixture of administrative rationalism, democratic pragmatism, and economic rationalism. Administrative rationalism, however, was more tangible than the two other approaches. It was expressed by the ban on fires and tavy, by the conditionality of the World Bank grants and by the importance given to expertise. Any subject of debate could result in the launching of a new study and funding was most often available for this. But science and expertise were called on to provide legitimacy rather than to produce knowledge. They were part of an administrative process. Democratic pragmatism was expressed by the implementation of the GELOSE legislation, by regionalization and decentralization efforts, by the organization of workshops and by the multiplication of working groups in the Ministry. Democratic pragmatism was, however, an illusion because these groups, like the associations created by the GELOSE legislation (Chapter 11), rather served to legitimate decisions taken at the upper level. Lobbies from conservation organizations on one side, and
414
from private interests and regional political powers on the other side, were more influential on the decisions taken. At the local level, the policies decided in Antananarivo were often perceived as imposed and not legitimate. Communication tools were set up in the name of democratic pragmatism but mostly served to obtain legitimacy through manipulative “public awareness” and education programs. The Moramanga workshop about the practice of tavy gives an example (Chapter 6). The Durban Vision Group (Chapter 2) also followed this logic. Economic rationalism was expressed by attempts to develop a carbon market and to put in place payments for environmental services. The VMCRCC project, analyzed in Chapter 6, provides an illustration. Economic rationalism is not yet a dominant paradigm in Madagascar but is certainly promised to have more weight in the future. In sum, a few actors had more power and more influence: the state’s higher authorities, due to their sovereignty to take decisions; the main donors, due to some explicit and implicit conditions associated with the grants; and a few conservation organizations, such as Conservation International, due to their strong lobby power on the donors and on the Malagasy state. Beneath this, staff from Malagasy services, minor donors, and field partners had less influence. They had different opinions but had to follow the general move in favor of Muir’s vision, renamed the Durban Vision (Chapter 2). An environment family was formed which built up a common culture and committed to achieving this vision. This vision became a dream that widened the gap between representations and realities. It ended in being self-enforcing and selfjustifying, showing that little may have changed from the tautological evaluation of EP2 by OSIPD (2000) (Chapter 2). The reason may be that the most essential actors, those who shape the reality on the ground, were not members of this family. These most important actors are, of course, the Malagasy farmers. Their presence in NEAP’s
415
working groups was just in the form of numbers, as in the OSIPD’s (2000) evaluation report of EP2 (Chapter 2).
5. Conclusion This Chapter showed that development, and now sustainable development, are central concepts in our representations of our world. This central position is justified by the fact that development has an essential ontological signification. Nothing can be static and no change can occur without being under the influence of forces that give it a direction, although these forces are so complexes that the direction cannot be predicted in most cases. But development, beyond this core meaning, received supplementary layers of signification that transformed it into a buzzword in the best case, an ideology or a dogma in the worst case. The same occurred to sustainable development, with maybe more deleterious consequences due to the oxymoronic character of the concept. Reconciling the environment with human development thus remains a crucial challenge, a conclusion which is consistent with NEAP’s difficulties in achieving success and with the ideological battle that occurred in this program (Chapter 2). I will now analyze in more detail the three sub-discourses that I identified in Part III. But first, I have to explain the methodology I will use, which I derived from the works of Michel Foucault (1970) and Bruno Latour (1987).
416
CHAPTER 9 METHODOLOGY (DISCOURSE ANALYSIS)
1.Introduction In this chapter, I will present a series of concepts which I will use to analyze discourses in the subsequent chapters. These concepts have been elaborated by Michel Foucault (1970) and Bruno Latour (1987) and had a strong influence on poststructuralism and postmodernism for the first, on social constructionism and actor network theory (ANT) for the second. More will be said about social constructionism and ANT in Chapter 14. Section 2 will be dedicated to Michel Foucault’s epistemology, and Section 3 to Bruno Latour’s. I will then synthesize the outcomes of these authors (Section 4), and propose a way to avoid the relativist trap toward which a radical interpretation of their epistemology could lead (Section 5). This will lead me to propose a personal model which can be regarded as a form of critical realism (Section 6) and which I will use for my investigations in Chapters 10, 11 and 12.
2. Michel Foucault’s epistemology This section is mostly derived from a concise introduction to Michel Foucault’s (1970) work: his first course of at the College de France, in 1970. A central assumption made by Foucault is that in every society, the production of discourse is controlled, selected, organized and distributed according to procedures aimed at controlling its powers and dangers, its random events, and at avoiding its heavy and fearsome materiality.255 (Foucault 1970, 10) 255
Translation by Pollini. Original text in French: “dans toutes les sociétés, la production de discours est a la fois contrôlée, sélectionnée, organisée et redistribuée par un certain nombre de procédures qui ont pour rôle d’en conjurer les pouvoirs et les dangers, d’en maîtriser l’événement aléatoire, d’en esquiver la lourde, la redoutable matérialité.”
417
In other words, knowledge is not built through a neutral collection of empirical facts. It is produced and organized by a society that needs it in order to justify and maintain its rules, its organization and its culture. Subjective forces emanate from the social and pervade into discourses, reshaping certain facts and discarding or rearticulating others. Discourses are thus both the consequence and the cause of power. They can be hegemonic and pervade into the mind and body of people, controlling their behavior and shaping their common culture. Rather than mirroring the world, knowledge, in this perspective, would be the incarnation of a social and cultural identity. It would be unquestionable from inside. We can say, in the light of the outcomes of Chapter 8, that discourses are intimately linked to the program run by a society. They may hence be the cause of the difficulties for societies to evolve, while providing guarantees of social cohesion. To maintain the hegemony of their dominant discourse, societies develop a set of tools to control the emergence and evolution of new forms of knowledge. They create what Foucault called domains of exclusion. They fix the boundaries or conditions of possibility of discourses. Foucault proposed a series of tools and principles to analyze these forces of exclusion.
2.1. The first series of principles This first series of principles concern modes of exclusion that operate from outside discourses. First are the statements and the taboos that reflect them. Statements are discursive elements whose presence is hegemonic even if they are not visible. They spread and unfold in the whole discursive field, fill interstices between discursive elements which they link and reshape. They are the sources and the manifestation of power. They are taken for granted by most subjects. They determine what we can
418
think and say, what we dare to say and what the persons we talk with can hear from what we say. A second mode of exclusion is the divide that all societies establish between reason and madness. Foucault (1961) showed that the location of this divide varies according to time and place. In Europe, there was a period when madness was regarded as capable of revealing important truths. It was thought that phenomenon not explained by reason could be better perceived and understood by people who lost their reason. This belief disappeared with the emergence of the enlightenment philosophy and the raise of positivist epistemologies. The third mode of exclusion is the partition between true and false, which Foucault calls the “great partition.” It occurs at the early stages of the formation of societies and cultures. It determines the dominant statements that will become hegemonic (the truth) and those that will be rejected (the false). Its determinants have to be sought at the origin of discourses and culture and on their outlets, through a process of deconstruction that is aimed at re-questioning true and false, revealing their relative character. Foucault (1969) calls Archeology of knowledge the exercise that consists in searching the domains of exclusion of discourses, by identifying statements, taboos and partitions. This exercise enables him to delineate the condition of possibility of discourses, or to reveal their will for truth. His approach presents some similarities with Nietzsche’s (2003 256 ) Genealogy of Morals, which may have been an earlier attempt to deconstruct Western civilization, culture and discourses. Foucault did the archeology of knowledge to study reason and madness (1961), medical discourses (1963), and prisons and education (1975). These exercises were challenging because none can totally escape from the will for truth. “Thus appears to our eyes only a truth 256
First published in 1887.
419
that would be wealth, fecundity, a sweet and insidiously universal force. And we ignore on the other hand the will for truth, as prodigious machinery destined to exclude” 257 (Foucault 1970, 22).
2.2. The second series of principles Foucault identified another series of exclusion modes that, in contrast to the first ones, are directed toward what is inside discourses. These forces of exclusion determine the internal organization of ideas. They delineate boundaries between subsets of discourses and help to control the emergence of new ideas and paradigms, in the interstices of knowledge. The first of these internal modes is the commentary. By linking discursive elements to previous texts, it allows the re-articulation and refinement of discourses and obliges new knowledge to remain embedded in existing discourses. The commentary impedes the random occurrence of subversive new ideas and thinking. It leads to a consolidation of discourses through redundancy: along the subsequent exegeses of a text, some elements crystallize. They become automatically accepted by the commentator because they gain increasing momentum from the authority of successive commentators. The commentary is widely used in scientific work where it takes the form of the citation, as I will show later. The second internal mode of exclusion is the author; not the physical author, but the author as a point of origin of discourses. In the same manner as “the commentary limited random events by the play of an identity that would have the form of the repetition, … the principle of the author limits chance by putting at play an
257
Translation by Pollini. Original text in French : “ainsi n’apparaît a nos yeux qu’une vérité qui serait richesse, fécondité, force douce et insidieusement universelle. Et nous ignorons en revanche la volonté de vérité, comme prodigieuse machinerie destinée à exclure.”
420
identity that has the form of the individuality and the self” 258 (Foucault 1970, 31). The author is a sign of the unity of discourse. It provides an identity to a set of ideas and facts, guaranteeing their consistency. The third internal mode of exclusion relates to a concern I already expressed in the introduction of this dissertation. It determined my willingness to be driven by the studied object rather than by a discipline. This third mode is actually the discipline. It is significant that the same term refers to “a control or order exercised over people or animals,” “the system of rules used to maintain this control,” and “a branch of instruction or learning.” 259 According to Foucault, “a discipline is not the sum of all the truth that can be said about something” 260 (Foucault 1970, 32). Rather, “inside its limits, each discipline recognizes true and false propositions; but it pushes away, on its margins, a teratology of knowledge” 261 (Foucault 1970, 35). It does this in several manners. First, in order to be accepted inside a discipline, facts have to concern a certain category of objects. Second, facts must to be collected using certain tools and concepts. Third, they must lead to propositions that match with a given theoretical framework. It can nevertheless happen that some facts challenge this framework and lead to a scientific revolution, or a paradigm change in the sense of Kuhn (1970). But disciplines try to retard this as long as they can. They may be the key machinery that controls and shapes discourses, fixing their boundaries.
258
Translation by Pollini. Original citation in French: “ le commentaire limitait le hasard par le jeu d’une identité qui aurait la forme de la répétition et du même. Le principe de l’auteur limite ce même hasard par le jeu d’une identité qui a la forme de l’individualité et du moi.” 259 The Illustrated Oxford Dictionary, 1998. 260 Translation by Pollini. Original citation in French: une discipline, ce n’est pas la somme de tout ce qui peut être dit de vrai a propos de quelque chose.” 261 Translation by Pollini. Original citation in French: “ A l’intérieur de ses limites, chaque discipline reconnaît des propositions varies et fausses; mais elle repousse, de l’autre cote de ses marges, toute une tératologie du savoir.”
421
2.3. The third series of principles Eventually, Foucault presents a third set of principles. They are systems of control exerted upon the persons that enunciate discourses. First are the rituals that accompany speeches, such as the behaviors, the body movements and other signs that speakers express under the control of their peers. Second are the societies of discourses, which serve to create affinities and connections by which discourses are recognized, shared, adopted, spread and given more momentum. Third are the doctrines, which determine more narrowly than disciplines the areas around which discourses revolve. Fourth is education: “every educational system is a political way to maintain and modify the appropriation of discourses, with the knowledge and power they carry with them” 262 (Foucault 1970, 46). Education may be the key to control societies because it trains the future leaders. Through it, institutions get embedded in the dominant discourses and are provided with appropriate knowledge, authority and power.
2.4. The origin of discourses After reviewing these principles, Foucault (1970) analyzes the mechanisms that explain the creation of discourses and their hegemony. He asserts that the need for rationality leads to considering the will for truth as the truth itself. As a result, the specificity of discourses, as entities distinct from the reality itself, is denied by three myths that are indeed the roots of discourses. First is the myth of the original subject. It is the belief that there is an objective point of departure at the origin of commentary chains. This point is regarded as untouchable because it would be real, anchored in the reality. Second is the myth of the original experience. According to it, meanings 262
Translation by Pollini. Original citation in French: Tout système d’éducation est une manière politique de maintenir ou de modifier l’appropriation des discours, avec les savoirs et les pouvoirs qu’ils portent avec eux.”
422
existed in the world before discourses. They would just have to be deciphered and articulated together. “An early complicity with the world would be the foundation of our possibility to speak about it and in it, to designate and name it, to judge it and to know it in the form of a truth” 263 (Foucault 1970, 50). Foucault rejects this conception and contends that “we should not expect the world to show us a readable image that we need only to decipher. It is not accomplice of our knowledge” 264 (Foucault 1970, 55). The third myth is the “universal mediation.” It consists in substituting discourses to real objects. Through it, discourses have their own life and exist for themselves, along subsequent exegesis. They would become “nothing more than the reflection of a truth that is revealed under its own eyes” 265 (Foucault 1970, 51).
2.5. Deconstructing discourses Foucault committed his life to disputing the will for truth, to restore to discourses their incidental nature and to lift the sovereignty of signs (the myth of the original experience) in order to liberate human knowledge. In order to do so, he proposes four methodological principles: First is the principle of inversion. It implies to distrust discourses and to question the will for truth: In the proper location where, according to the tradition, we would expect to recognize the source of discourses, the principle of their diversity and continuity, in these figures that seem to play a positive role, such as the figure of the author, of the discipline, of the will for truth, we should rather
263
Translation by Pollini. Original citation in French: “Une complicité première avec le monde fonderait pour nous la possibilité de parler de lui, en lui, de le designer et de le nommer, de le juger et de le connaître finalement dans la forme de la vérité.” 264 Translation by Pollini. Original citation: Il ne faut pas “s’imaginer que le monde tourne vers nous une image lisible que nous n’aurions plus qu’à déchiffrer. Il n’est pas complice de notre connaissance.” 265 Translation by Pollini. Original citation in French : “Le discours n’est guère plus que le miroir d’une vérité en train de naître a ses propres yeux.”
423
recognize the negative game of the division and rarification of discourses. 266 (Foucault 1970, 53) (emphasis added) Second is the principle of discontinuity. It requests us to recognize the accidental existence of discourses, the absence of linearity in their construction. Third is the principle of specificity. It implies that discourses cannot be reduced to a series of signs that would reveal the truth. Rather, they would have their own truth. Fourth is the principle of exteriority: discourses have to be analyzed for what determines them outside rather than what they are inside. We have to understand “the condition of their possibilities.” Having these principles in mind, the architecture of discourses can be deconstructed using the three series of principles listed in the previous section. An attempt can then be made to go beyond discourses, although a total emancipation may be impossible.
2.5. Conclusion In sum, Foucault proposes two series of concepts to analyze discourses. The first serves to do the genealogy or archeology of discourses: to understand how they got formed, under which system of constraint, and what are the conditions of their apparition. The second series serves to delineate their current boundaries, to identify domains of exclusion, to analyze the ways exclusion operates. For both series of concepts, the objective is to understand the location of the determinants that shape discourse. Only the perspective changes, from a temporal one for the first series of principles to a spatial one for the second series. These concepts contributed to challenging the positivist epistemology and to initiating an intellectual revolution that led to postmodernism and social
266
Translation by Pollini. Original citation in French: “La ou, selon la tradition, on croit reconnaître la source des discours, le principe de leur foisonnement et de leur continuité, dans ces figures qui semblent jouer un rôle positif, comme celle de l’auteur, de la discipline, de la volonté de vérité, il faut plutôt reconnaître le jeu négative d’une découpe et d’une raréfaction du discours.”
424
constructionism, two movements characterized by the assumption that truth is relative to the observer. The question that remains to answer is until what point this revolution must go, between a strict positivism that already showed its limits and a radical relativism that could lead to ontological vacuums, as occurred with the postmodern deconstruction of concepts such as development and modernism (Chapter 8). The analysis of Latour’s epistemology will help us to answer this question.
3. Bruno Latour’s epistemology 3.1. Latour’s key principles Bruno Latour is an influential figure of social constructionism. But he later departed from this movement, which he criticized for the relativist dead-end it led to. Latour (1986) analyzed more specifically the production of scientific knowledge. He started his career by observing a group of researchers in an endocrinology laboratory. He employed a participant observation approach as classical anthropologists did when they studied exotic tribes. His main conclusion was that scientific facts are socially constructed: cultural, social and economic factors, not the actual realities of the biophysical world, determine the direction in which knowledge advances. The similarity with Foucault’s work is clear. In his book Science in Action, Latour (1987) developed a methodology for studying these social processes. He proposed to create a new sociology of science which, contrary to mainstream approaches, would neither be a social history of sciences, nor the search for socially or culturally determined scientific mistakes. The focus would shift from the ideas themselves to their construction; from the analysis of the “ready made science” (Latour 1987) and its relation with society, to the analysis of the social process of “science in the making” (Latour 1987). This shift determined a first methodological principle: “[the] entry into science and technology will be
425
through the back door of science in the making, not through the more grandiose entrance of ready made science” (Latour 1987, 4). This principle being stated, Latour identified the type of knowledge that needs to be collected in order to do this new travel into science and technology. He observes that “context and content fuse together” (Latour 1987, 6) in the making of science: Suspense, game, tone, delay of publication, awe, six week delay, are not common words for describing a molecule [sic] structure. This is the case at least once the structure is known and learned by every student. However, as long as the structure is submitted to a competitor’s probing, these queer words are part and parcel of the very chemical structure under investigation. (Latour 1987, 6) In other words, the social relations between researchers and institutions, in a context of competition, lead to discarding certain facts and retaining others. At the end of this social process, a truth is accepted and a controversy can be closed. A new scientific fact is created and enters the realm of ready made science. The concept of controversy does not have its equivalent in Foucault’s epistemology, where the emphasis is put on the hegemonic character of discourses. Controversies play a central role in Latour’s epistemology because they justify the mobilization of resources, by the creation of laboratories. Laboratories, with their careful experiments, provide the equipment that scientists need to create facts and translate these facts into texts (articles), which strengthens their position in controversies. Laboratories are supported by networks that can involve media, politicians, and research clients far beyond their technical boundaries. The aggregate of these technical, human and social resources produces what Latour calls the machines. The stronger machines are those that will mobilize more resources, create more facts and win a controversy in the end. The machines involved in controversies have several weapons: rhetoric (the way to articulate arguments in the literature and in talks); the recruitment of more
426
“friends”; the creation of new laboratories in order to mobilize more resources; the argument of authority (an alliance with scientists that have more recognition or social status in their field); stylization, through pictures, figures and diagrams that provide more visual impact to the text; stratification, that operates by relating a new statement to others that are out of controversy; and citation. This last weapon is essential because “the strength of the original statement does not lie in itself, but is derived from any of the papers that incorporate it” (Latour 1987, 42). But the most powerful of these strategies may be stratification. It is “the transformation of linear prose into, so to speak, a folded array of successive defense lines” (Latour 1987, 48). In consequence “the dissenter will be unable to expose the text to the real world out there, since the text claims to bring within it the real world “in there” (Latour 1987, 49). In sum, a constructed world replaces the real one. As this new world is clearer and, apparently, better organized and easier to understand than the real one, the reader will think it is an accurate reflection of the reality. Stratification is hence “the surest sign that a text has become scientific” (Latour 1987, 48). Latour also mentions several tactics used for stratification: stacking facts in logical order, which leads the reader to pay more attention to inductive arguments; staging, which makes explicit what type of reader is targeted by the text and convinces opponents to consider it seriously; framing, which impedes the reader from wandering around the arguments and to being out of control; captation, “the subtle control of the objector’s moves” (Latour 1987, 57), achieved by all tactics working together. If captation is successfully achieved, the text will be laid out in such a manner that “wherever the reader is, there is only one way to go” (Latour 1987, 57). This canalization may eventually lead opponents to become supporters of the key statements (to be recruited). They will still feel they are free but will be convinced that the text is logical and reflects reality.
427
3.2. Latour’s departure from Foucault In sum, for Latour as for Foucault, social forces of exclusion and inclusion are the determinants of knowledge. Latour, however, focuses more specifically on a specific mode of knowledge that we call science and on the way it produces facts, while Foucault analyses discourses at a broader scale and searches their determinants outside. But the main distinction between Latour and Foucault is the way they consider social constructions. For Foucault, they are hegemonic representations that shape our mind and our body. We should attempt to depart from them in order to be free and fulfil ourselves. For Latour (1987), they are the outcomes of a social process but these outcomes have to be accepted. Latour (1987), like Foucault, believes that the real world is not an accomplice of our knowledge. Nature does not determine how controversies are closed because no researcher can have it on its side. Latour (1987, 97) refuses the idea that “the goliath of rhetoric with his laboratory set-up and all his attendant Philistines [would] be put to flight by one David alone using simple truths about nature and his slingshot.” But in order to avoid the relativist trap this could lead to, he declares that the constructed nature is the one we have to deal with; that it is the real nature. As he puts it, the settlement of a controversy is the cause 267 of Nature’s representation not the consequence. (Latour 1987, 99) More than that, he considers that we can never use the outcome - Nature – to explain how and why a controversy has been settled. 268 (Latour 1987, 99) This is probably Latour’s main departure from Foucault’s thinking and the epistemological consequences are important. If we take Latour’s assertion above to the 267 268
Italicized by Latour. Italicized by Latour.
428
letter, the real natural world is assimilated to its representation. No realities can exist other than the representations that have been fixed by closing controversies. This very radical and provocative assertion has been perceived by many as a declaration of war against science and empiricism. Gross and Levitt (1994, 58), for example, parody Latour with the following example: imagine that a few of us are cooped up in a windowless office, wondering whether or not it’s raining. Opinions vary. We decide to settle the issue by stepping outside…. We retreat into the office and say to each other: Wow, it’s really coming down! We all now agree that it’s raining. Insofar as we are disciples of Latour, we can never explain our agreement on this point that it is raining. Rain, remember, is the outcome of our “settlement,” not its cause! We can however doubt that this parody reflects what Latour wanted to express. In several passages of his books, Latour recognizes the existence of a world out there, distinct from its representations. There is indeed another way to understand Latour. Saying that nature is the cause of the settlement of a controversy could simply mean that the representations we create contribute to shape our world in return, and that this created world is the one that matters. The epistemological consequences of this perspective are important as well, and can be criticized, but I will not address this issue here. I will come back on Latour’s work in Chapter 14. For now, I am just interested in the tools proposed by Latour to analyze science in action, assuming that these tools will help to analyze discourses in the next chapters.
4. Synthesis Foucault and Latour proposed a series of tools and concepts that will be very useful to analyze discourse. Table 19 summarizes and compares these concepts: The second and third set of principles, which concern the internal driving forces of discourses, have their equivalent in Foucault’s and Latour’s epistemology. But Foucault’s first series of principles, which deals with what is located at the outlet
429
Table 19: Comparison of Foucault’s (1970) and Latour’s (1989) concepts. First series of principles (operating from outside discourses) Foucault Latour Strong statements and Taboo ? Reason-Madness ? Great partition ? Second series of principles (operating inside discourses) Foucault Latour Commentary Citation Author Principle of authority Discipline Machine Third series of principles (shaping the behavior of those who pronounce discourses) Foucault Latour Rituals Staging Societies of discourse Recruitment Doctrines Framing Education Captation
430
and at the origin of discourse, does not have equivalent in Latour’s epistemology. This absence may explain why Latour needs to find a truth inside the discourses themselves, and adopts the idea that nature is the outcome of closed controversies. If the truth cannot be found outside, it must be located inside. The consequence is that, if we follow Latour’s conception, the world out there is internalized through the social construction of discourses. For Foucault, the world out there takes a form that we do not know and cannot pretend to know. It is located outside discourses. There may be indeed a symmetry between Foucault’s original subject and experiences and Latour’s ready–made science. Foucault’s original experience may be the starting point or the cause of the great partition. It determines which facts participate in the production of knowledge and which facts do not. Latour’s readymade science may be the knowledge that has been produced in the end. It could thus reflect Foucault’s original experience, closing a cycle that started with this original experience. By accepting the outcome of controversies as being the real world, Latour may have refused to question the original experience, the great partition, the conditions of apparition of discourses, which may explain why his epistemology does not propose some equivalent to these concepts. Getting back to the subject of this dissertation, I will use the concepts of both authors. Foucault’s vocabulary will be better adapted to analyze broad systems of knowledge and to search for hidden truth outside discourses and controversies. Latour’s concepts may be better adapted to analyze scientific discourses and the process of their creation. But before starting my analysis, I have to point out an essential distinction between the stance I will take and the one proposed by Foucault and Latour.
431
5. Avoiding the trap of relativism As I already quoted, Foucault (1970, 55) contended that the world “is not accomplice of our knowledge.” The danger of this conception is that it could lead to radical forms of relativism. If knowledge is not linked to reality in one way or another, all representations become equivalent and the world cannot be understood. In a pragmatic perspective, this would be nonsense. Radical forms of deconstruction could lead to an ontological vacuum once a concept was completely deconstructed. In order to avoid this trap, I propose to add just one word in Foucault’s assertion, and to say that
The real world is not the only accomplice of our knowledge.
Theoretically, this changes everything because linearity and a positive axis in the advancement of knowledge are possible again. The unfolding of signs from nature can tell us again something and give a positive dimension to knowledge. But from a pragmatic perspective, as has been adopted for this dissertation, the difference may be marginal. The forces of exclusion and subjectivity that Foucault identified are still recognized. They may even overwhelm the positive forces that emanate from the natural signs. They just do not act alone. This model is not a return to positivism because the real world cannot act alone: it is just one determinant of knowledge among others. It can be the major determinant in some cases, and be marginal in others. What matters will be to identify the factors that render this “natural” positivity marginal. In the case of complex objects embedded in the cultural, social construction may be more determinant than the unfolding of signs from the real world. But in the physical act of jumping from the Tower of Pisa, we should listen to the advice of Galileo, a conventional positivist
432
scientist, instead of risking our life after deciding not to trust the signs that unfold from nature. We could even listen to Aristotle’s advice, despite Galileo proving he was wrong. This trivial example shows that from a pragmatic perspective, positivities cannot be denied and even false or imperfect positivities can be useful. The position I take with regard to Latour’s work is similar to the one I took with Foucault’s: the social construction cannot explain everything. I disagree with Latour’s assertion that nature is the consequence of closing controversies, as I disagreed with Foucault that the real world is not an accomplice of our knowledge. I prefer to consider that nature is indeed behind the laboratory, despite the fact that its role is often marginal. This explains why Lysenko’s biology was eventually rejected. As for Foucault, I would modify Latour’s assertion and contend that
The settlement of a controversy is the cause of nature’s representation, whereas the real nature is one of the causes of the settlement of controversies.
In short, the interest of Foucault’s and Latour’s epistemologies is that they provide us with powerful tools for debunking the social and cultural forces that bias representation and cause mismatches between representation and realities. We have to use these tools to make all possible attempts to get closer to reality, to control the forces that bias knowledge, to read more directly the “words” that nature says to us. If the real world or nature were not accomplices to our knowledge, we would fall into a relativist trap where these tools would become useless. This dissertation would then be a fun exercise but would have no practical use. The controversies I am dealing with will close some day and the reality would just be the consequence of this closing. So why bother with it? This is the paradox within which postmodernism and social constructionism have been trapped.
433
6. A new model These comments lead me to propose a different epistemological position that I call “relative relativism.” In this stance, we can use the real world to explain how and why a controversy is settled. But we cannot use it alone to provide the explanation. The other causes of settlement are preexisting false representations or false statements embedded in the social, the political and the cultural. The construction of representation thus results from the influence of the real world on one side (nature being part of it) and of preexisting false statements on the other side. Figure 15 helps explain this model. The asymptotes represent the cases of extreme match or mismatch between representations and realities. Representations are closer to the 100% mismatch asymptote in cases where they are more influenced by false statements, and closer to the 100% match in the opposite case. The influence of false statements on one side, and of the real world on the other hand, are represented by the two magnets behind the asymptotes. These influences are the sum of factors of subjectivity in the first case (left magnet), and the sum of factors of objectivity in the second case (right magnet). A represents a situation in which no reality is yet represented. It is absolute ignorance. B represents a situation in which the whole world is represented and in which all elements of this representation (all statements) are true. Such a representation could be produced by a perfect spirit (a unique god) whose perceptions and skills would have no limit. The starting point would be ignorance but no false statements would appear along the construction of representations. Knowledge would be elaborated following exactly the right asymptote, from A to B.
434
Part of the world that is integrated into representations 100%
Paradigms
C
D
100% mismatch asymptote
100% match asymptote
F
False statements 0%
B
E
A
Figure 15: Match and mismatch between representations and realities.
435
Real world
C represents a situation in which the whole world is represented, but everything in this representation is wrong. It is the negative of B, made of all its inverse statements. The line from A to D relates to representations in which there is a balanced mix of true and false statements. On the right side of this line, representations are more true than false. On the left side, they are more false than true. Paradigms are located at different levels of balance between true and false. For example, Lysenko’s biology could be mapped on the left side of the diagram. E is the “area of representations” of an object that relates to a small fraction of reality and about which there can be little doubt. E can be, for example, the statement that “fire volatilizes nitrogen and burns organic matter.” F relates, to the contrary, to a broader object whose representation can be largely put into doubt, such as “Fire leads to poverty.” One could contend that there are cases where statements are proven and are located on the right asymptote, for example: 2+2=4. I contend that this could be true in practice, but must not be accepted in theory. The probability that a statement is false can tend to zero, but cannot be equal to zero. This precaution allows knowledge to always remain open. It is a recognition of the limits we face in the perception of our world. The following statement illustrates the necessity to allow knowledge to be open: “two persons having the same age will necessarily have the same age all their life.” One century ago, no doubt was possible about this statement. We know now that it is false. The following assumptions can further be made about how this diagram works. First, there is no way to know for sure where a given representation is located inside the diagram, in between the false and the true. The magnets act but we cannot know how much they act, and which one acts more. This is what differentiates this
436
model from certain forms of critical realism according to which certainty would be difficult to achieve, with many traps along the way, but could be achieved in the end. In the present model, a doubt always remains. The consequence is that controversies are never closed and representations always remain open to modifications. This model may indeed illustrate a form of Popperian critical realism where knowledge advances by falsification and where scientific facts are uncertain by definition (they are facts that can be falsified). In spite of this uncertainty, positive progress is still possible. It just remains uncertain and this uncertainty is precisely what renders progress possible. The advancement of knowledge is dependent on the acceptance of the “Cartesian anxiety” (Bernstein 1983), not on its rejection. The relation between knowledge and reality is probabilistic: there is no way to know for sure whether a representation matches with realities, but the chances for a better match can be increased by diminishing the influence of the left magnet (by identifying factors of subjectivity and distancing oneself from them) and by increasing the influence of the right magnet (by having a more direct experience of the real world). What matters then is to identify the factors that decrease these chances. Certainty is one of these factors because it freezes knowledge or fixes once and for all the direction of its advancement. This is why uncertainty is a condition of progress. Second, representations are never located on the asymptotes of the diagram, nor do they reach its bottom and upper lines. They navigate inside these boundaries. The consequences are that there is no absolute ignorance, no absolute certainty, no absolute doubt, and that representations are never complete. Representations that match with the points A, B, C, and D do not exist.
437
Third, in case of controversy, the observers whose representation match better with realities will be given a supplementary chance to win the controversy, as a result of the action of the right magnet. But the observers whose representations match better with preexisting false statements will also have an advantage, as a result of the action of the left magnet. Reality does not determine who wins the controversy, but gives a greater chance to the representations that match better with it. In the same manner, false statements give more chance to the representations that match with them. Fourth, the representations of simple objects mostly embedded in the biophysical world are less prone to the effects of false statements and are more prone to the effects of realities, compared with complex and broader objects more embedded in the social world. In sum, this “relative relativism” model departs from positivism as well as from relativism. The term relativism relates to the departure from positivism, while the term relative relates to the departure from relativism. The model could be regarded as a form of critical realism but critical realism is not always clear whether it accepts the possibility of certainty. In the model I propose, the key feature is explicitly the acceptance of uncertainty, which is even recognized as the conditions for the advancement of knowledge. This uncertainty is the consequence of the presence of two positivities which interact. I will call them the natural positivity and the cultural positivity. The natural positivity drives representations in a direction that makes them match better with realities. It is the line that was rendered hegemonic by positivist epistemologies. The cultural positivity, on the other hand, drives representations in a direction that makes them match better with themselves. It is the line that received much attention from Foucault (1970, 51), who described it as “nothing more than the reflection of a truth that is revealed under its own eyes.”
438
The need for certainty may be a characteristic of our Western civilization and culture. By giving exclusivity to the cultural positivity (by asserting that nature would be the consequence of closing controversies), Latour may have indeed refused uncertainty in the same way as positivist scientists did before. He just replaced one positivity by another one. This may have been necessary to demystify a form of science that gave a hegemonic power to the natural positivity. But now that the shift toward relativism has been digested, the natural positivity should be considered again in order to avoid other forms of abuse. By recognizing a dialectic relationship between the cultural positivity and the natural positivity, we can impede scientific positivism from becoming hegemonic again, while avoiding a fall into the trap of relativism. I will come back to these issues in Chapter 14.
7. Conclusion This chapter identified a series of powerful concepts which I will summarize here. First, discourses have a point of origin and to identify this point (to uncover their genealogy) makes it possible to identify a great partition between discursive elements that were set apart (domains of exclusion) and elements that remained accepted. For example, the Enlightenment set apart Romanticism, leading to our positivist rationale civilization and to the conflation of development with the concepts of modernization and industrialization (Chapter 8). Second, discourses have implicit or hidden statements that are the consequence of this great partition. For example, the belief that nature has no intrinsic value, that farmers are irrational, that regular geometry is good, that yield increase results in production increase and that fires are evil, are widespread hidden statements among development and conservation actors in Madagascar, as we saw in Part II.
439
Third, the agents of discourses are committed to recruiting more people and producing more knowledge, in order to gain more authority or power. As a consequence, discourses shape knowledge and power cannot be separated from knowledge. Science, which is the privileged way to produce knowledge in Western civilization, is thus intimately linked to power and politics. Fourth, discourses that get consideration from rulers and institutions vested with authority spread in the rest of the society through education, determine or reshape disciplines and produce doctrines and received wisdoms. In this way, they increase their control upon the subsequent production of knowledge and impede the emergence of alternative discourses. By examining these mechanisms, it will be possible to identify who the discourses serve, what are their objectives and who benefits or suffers from the actions implemented in their name. Chapters 10, 11 and 12 will now show that the concepts elaborated by Foucault (1970) and Latour (1987) are powerful tools for understanding the condition of apparition of wrong beliefs or received wisdoms, such as those listed in the conclusion of Chapter 6. But I departed from the epistemologies of these authors by recognizing that the real world is an accomplice of our knowledge, though often a powerless accomplice, and by considering that controversies should never be closed. In this way, I avoided falling into the trap of relativism. I recognized that uncertainty exists in all representations, but also that some things are less uncertain than others. The facts and theories that are less uncertain, in a given state of knowledge, have to be accepted to drive our decisions. They can sometimes be regarded as certain in practice despite the fact that they are not certain ontologically. If they were, this would create a risk of producing received wisdom that would be impossible to debunk. Ignorance, indeed, is not uncertain representation. It should rather be defined as a representation that does
440
not match with the facts but that is regarded as certain. This is why Latour’s epistemological model, which accepts the closing of controversies, can lead to the creation of realms of ignorance just as positivist science did before.
441
CHAPTER 10 THE ALTERNATIVE TO SLASH-AND-BURN DISCOURSE (ASB)
1. Introduction In this chapter, I will present a research and extension program, the ASB Consortium, whose objective was to develop alternatives to slash-and-burn cultivation (ASB). This program promotes agroforestry technologies and the outcomes of its work have been influential on rural development projects addressing slash-and-burn cultivation worldwide. This analysis will help to understand the faith in agroforestry and other agrobiological alternatives of the LDI, PTE and VMCRCC projects in Beforona (Chapter 6). I will start by relating the birth of this program (Section 2) and the early development of agroforestry research, from which the ASB concept emerged (Section 3). Section 4 will present a series of received wisdoms about slash-and-burn cultivation that existed when the program was launched. Sections 5 and 6 will analyze the work of the Consortium and the conclusions of its two phases. In the conclusion (Section 7), I will show that the ASB Consortium eventually proposed a significant paradigm shift which is consistent with the outcomes of the case study presented in Part III.
2. The birth of the ASB paradigm In 1987, the Brundtland Report (WCED 1987) called for more attention to tropical deforestation and its relation with poverty. It pointed out “the serious lack of funding “for conservation projects and strategies that improve the resource base for development” ” (Brady 1996, 7). As a consequence, the UNDP 268 “commissioned the 268
United Nations Development Program.
442
World Resources Institute (WRI) to study the problem” (Brady 1996, 7). This institute recommended the establishment of an International Environment Facility (World Resources Institute 1989), leading to the creation, in 1991, of the Global Environment Facility (GEF), co-sponsored by the World Bank, UNDP and UNEP. 269 One of the first GEF operations was the funding of development and conservation organizations committed to favor the abandonment of slash-and-burn cultivation. Several international and national research centers and NGOs, led by the International Centre for Research in Agroforestry (ICRAF), 270 were involved (about 40 organizations at present). A workshop was held in February 1992 in Brazil to launch the program. It was “attended by 26 environmental policy makers and research leaders, five NGOs, six international research centers, three regional research organizations and six donor agencies” (Sanchez et al. 2005, 10). The ASB Consortium was born. It became a CGIAR 271 program in 1994 and its legitimacy increased after the adoption of the Convention on Biological Diversity and of the Agenda 21 of the 1992 United Nations Conference on Environment and Development. According to Chapter 11 of Agenda 21 (section “combating deforestation”), governments, with the participation of the private sector, nongovernmental organizations, local community groups, indigenous people, women, local government units and the public at large, should act to maintain and expand the existing vegetative cover wherever ecologically, socially and economically feasible, through technical cooperation and other forms of support. (UNCED 1992, no pagination) Among the activities to be considered is Limiting and aiming to halt destructive shifting cultivation by addressing the underlying social and ecological causes. (UNCED 1992, no pagination) The first outcomes of the Consortium were presented at the XVth World Congress of Soil Science, held in 1994 in Mexico and published in Sanchez and Hailu 269
United Nations Environment Program. Now the World Agroforestry Centre. 271 Consultative Group on International Agricultural Research. 270
443
(1996). A second research phase followed, whose outcomes are synthesized in Palm et al. (2005a). Before examining these two syntheses, I will briefly review the agroforestry research conducted at ICRAF and analyze the discourses about slash-andburn cultivation that prevailed at that time.
3. Agroforestry at ICRAF ICRAF’s work started in the early 1980s with an inventory and description of agroforestry systems around the world. Most outcomes of this early agroforestry research were published in the first issues of the journal Agroforestry Systems, in the early 1980s. As shifting agriculture systems associate forests and agricultural components, ICRAF was concerned by the results of this inventory and the lack of sustainability of shifting agriculture in the contemporary economic and demographic context was pointed out (Raintree and Warner 1986). At that time, ICRAF was developing sophisticated agroforestry technologies such as alley-cropping, hedgerows planted on contours lines and improved fallows, which were quickly viewed as the potential alternatives to slash-and-burn systems (Figure 16). ICRAF also carried out research on indigenous agroforestry land uses, such as the multistory agroforests described by Michon (1985). But the long cycle of these systems, due to the presence of forest trees, did not facilitate research funding and extension programs. Alley-cropping and contour lines, to the contrary, were perfect candidates as research objects. They were appealing to donors because they privileged a Western conception of productivity where geometry, high yield and sustainability, a third criteria introduced by the Brundtland report (WCED 1987), are put at the forefront. ICRAF and its partners received significant funding to develop these technologies, allowing the selection of more productive grasses and legume trees and the design of optimal management practices.
444
Figure 16: The agroforestry pathways to change. Source: Raintree and Varner (1986, 44).
445
Alley-cropping and improved fallows clearly showed their superior efficiency in creating favorable biophysical soil-plant interactions, improving soil properties and supporting high yields. Their performance in terms of sustainability gave the illusion of a departure from high input agriculture, but this departure was only partial. In contour strips and alley-cropping systems, simple geometry and regularity are still regarded as essential (Pollini 1994), as they are in the high modernist scheme of Western agriculture (Chapter 8). Hedgerow management implies defining a multitude of norms about plantation spacing, pruning schedule and frequency, which have to be rigorously respected, as shown by Messerli (2003) in the case of Beforona (Chapter 6). Optimum values are defined at cropping system scale and farming system constraints are often overlooked (Pollini 1994). Complexity and heterogeneity, on the other hand, are not addressed, leading to systems that lack flexibility. In Burundi, for example, rural development projects complained about the low adoption rate of alley-cropping and contour strips technologies, while farmers planted perennial grasses as live fences around their fields. 272 As the landscape is a patchwork of contiguous fields with no regular geometric forms, these fences rarely follow contour lines. Their efficiency in terms of erosion control may be lower than hedgerows planted on contour lines, but the competition with crops and the management costs are lower. By conditioning the promotion of these grasses to their being planted along contour lines, projects limited their adoption and hence their overall impact in terms of erosion control. In sum, the high modernist ideal of a perfectly managed landscape was confronted with the logic of heterogeneous and complex social landscapes, where larger spatial and temporal scales matter more and where optimum is not only biophysical.
272
Statement based on my experience as a research associate in Agroforestry at the Burundi Agronomy Institute, from January 1993 to January 1994.
446
Beyond this lack of flexibility, the excessive attention paid to agroforestry technologies led to the neglect of pathways to intensification which emerged spontaneously from shifting agriculture systems in a situation of crisis, as described in Chapters 5 and 6. Figure 17 shows the case of the Burundi highlands, where slashand-burn systems coexisted with more intensive land uses in the early twentieth century (Pollini 1993). Fallow and livestock husbandry both contributed to maintain soil fertility until the lack of space forced farmers to practice very short fallows. Today, several crops are associated and two or three harvests per year are obtained on the same land. This history provides supplementary evidence in favor of the Boserupian model of agricultural intensification (Chapter 5). Raintree and Warner (1986), however, did not overlook Boserupian dynamics. In their diagram showing the pathways to intensification (Figure 16 page 435), they use Boserup’s (1965) terms “annual cropping” and “multiple cropping.” But they propose a departure from this “natural” pathway, by a shift from techniques aimed at increasing labor intensity to techniques that increase land use intensity. In other words, they propose to maximize yield rather than labor productivity. We saw in Chapter 6 the consequences of this excessive focus on yield in a context where land is not the limiting resource. In sum, ICRAF’s agroforestry technologies, beyond the fact that they have real biophysical qualities, 273 became fashionable and diverted attention from other pathways to intensification and erosion control. The concept of agroforestry ended up in being synonymous with contour lines, enriched fallows or alley-cropping. Saying that shifting cultivation is an agroforestry system is now perceived as nonsensical by
273
Agroforestry also includes a variety of low input land uses which recently developed in temperate countries (Lassoie and Buck 1999). Their advantages in this different context, notably to improve the quality of agricultural products and provide alternative pathways to agro-chemical intensification, are indisputable.
447
Permanent agriculture
Household Herd
Shifting cultivation
Primary forest (degraded)
Browsing area (savannah) Hunter Pasture
1910
Manure
Swamps (papyrus) Food crops (sorghum, maize, beans, peas, tubers)
Banana trees Hedgerows Pasture
Coffee
Eucalyptus Papyrus
1950
Dry season crops
Figure 17: The evolution of agricultural landscape in Bukeye, Burundi. Source: Pollini (1993). Note: hedgerows are usually located on slight slopes.
448
Agro-arboriculture (Grevillea, Cedrella, fruit trees) Banana trees Houses Hedgerow (Setaria, Tripsacum, Pennisetum)
Eucalyptus
1992
Coffee
Food crops (cassava, sweet potatoes, maize, beans, wheat, etc…)
Figure 17 (continued).
449
Dry season crops (Maize/beans/ potatoes)
many rural development actors, whereas shifting cultivation is indeed the only agricultural land use that integrates natural ecosystems (secondary vegetation if not primary forests). Ironically, the term forest, which initially designated natural areas not under control — an “outside” territory as we are reminded by the Italian word “fuori” (outside), derived from the same Latin root — ended up being synonymous with perfectly managed rows of shrubs and grasses cut three times a year. Model Burundian farmers who received agricultural medals for their perfect appropriation of agroforestry technologies, in the early 1990s, would not have been at odds in the Cartesian world of a French garden. But in the hills of Burundi, they were just temporary artifacts that provided donors with some illusions that projects achieved results. 274 For the bureaucracies of donor agencies and for the states they support, “agroforestry” became a unique and bright pathway to convert slash-and-burn systems. This pathway gave the illusion that a fast and radical change was possible and provided a justification for the implementation of repressive policies aimed at "eradicating" fires and slash-and-burn cultivation. In other words, ICRAF’s agroforestry technologies were central in the building of what Latour (1987) would call a strong rhetoric. This rhetoric contributed to strengthening many of the received wisdoms identified in Part III. To use the concepts introduced in Chapter 9, the “author” was ICRAF, which was provided with “authority” by the “discipline” of soil sciences and built a “machine” by creating a worldwide “network” of “laboratories” having strong ties with donors and “recruiting” partners among the elite of developing countries. In the end, agroforestry became a “doctrine” and the indigenous pathways to change were “domains of exclusion.” To illustrate the building of this strong rhetoric, Figure 18 shows the number of hits of certain concepts in CAB abstracts, one of the main databases used by experts 274
Based on the visit of rural development projects I did in Burundi, in 1993.
450
4201 3946
Indigenous know ledge
1889 1781
Gliricidia
671 631
calliandra
6414 6038
Leucaena
Improved fallow
255 179
Contour cultivation
473 427
In all fields In descriptor fields
1300 1127
Alley cropping
843 316
Marginalization
C o n c e p ts
Poverty
29753
8530
Market
10203
Erosion
68218
24282
Agroforestry
46287
7662
172
Political ecology
2325
Political economy
Geography
24349
2965
Socio-econom*
6175
Anthropology
11001
32096
655
0
10000
20000
30000
Number of hits 40000
50000
60000
70000
80000
Figure 18: Number of hits for several concepts in the entire CAB Abstracts database (2003).
451
working in international rural development programs. The scores are indicated for searches done in all fields or in descriptive fields only. Erosion receives the highest score, followed by market and agroforestry. Leucaena, a species used in alleycropping and contour strip technologies, has a higher score than indigenous knowledge. Figure 19 repeats the same exercise but restricts the search to references that present the keywords slash-and-burn, swidden or shifting cultivation in any of their fields (3506 references in total). 275 Soil has the highest score, which could be explained by the fact that it is a common word. But the words “farmer” and “people,” also very common, have much lower scores, whereas “nutrient” and “erosion,” which are more specific, still have a relatively high scores (blue dots). In the case of more specialized concepts, those that relate to biophysical elements (mulch, legume, organic matter, alley-cropping, runoff, leucaena; see black dots) have higher scores than those that relate to social aspects (social capital, marginalization, local knowledge; see red dots). Economic concepts are in an intermediate situation (credit, poverty, investment; green dots). The fact that agroforestry has a score similar to polic*/polit* and a higher score than social and anthropol* is also compelling These figures show that agroforestry, which is just a type of agricultural technology, turned out to weigh as much as classical academic disciplines among development actors. This new discipline provided a justification for the monitoring of rural development programs by agronomists and soil scientists. Social sciences were not absent, however. Low adoption rates of contour strips and alley-cropping technologies raised awareness that something was going wrong. In order to understand the reasons, biophysical scientists called for the support of social scientists and economists for applying participatory methods and for considering indigenous knowledge. Figures 18 and 19 (pages 451 and 453) confirm the important attention 275
These searches were done in 2003.
452 451
Paup erizat io n
1
Po lit ical eco lo g y
3
So cial cap ital
3
Po litical eco no my
6
M arg inaliz*
8
Number of hits
10
Lo cal kno wled g e Green manure
31
Fo o d s ecurity
33
leucaena
44
hed g ero w
47
Po verty
52
Cred it
59
To o l
67
Runo ff
68
Inves tment
69
Co nflict
71
Taung ya
74
Tillag e o r p lo ug hing
74
Anthro p o l*
75
alley cro p p ing
80
manure
90
Refo res tat io n
97
Leg ume
103
M ulch
103
M ig rat io n
107
Land tenure
109
Bio d ivers ity
111
Lives to ck
161
Co mmunity
167 195
Org anic matt er
218
Ind ig eno us Peo p le
271
M arket
273
Fire
335
So cial
341
Defo res tat io n
351
Ero s io n
397
Nut rient
402
Ag ro fo res t ry
408 421
Farmer Co ns ervatio n
447
Po lit* o r p o lic*
451 831
Eco no m*
1435
So il
0
200
400
600
800
1000
1200
1400
1600
Figure 19: Number of hits for several concepts in 3506 references retrieved from CAB Abstracts and having the word slash-and-burn, swidden or shifting cultivation in any of their fields (2003).
453
given to indigenous knowledge. But social scientists were called on to find a way to make adoption occur, rather than to help define how erosion could be controlled and slash-and-burn systems could evolve. They were at the service of biophysical scientists who had already determined the pathways to privilege. Concerning the "participatory methods," we can still question whether they were genuinely translated into action, for reasons that can be guessed from the outcomes of Chapter 6 and which I will investigate more thoroughly in Chapter 11. Concerning "local" or "indigenous knowledge," quotes of this concept are very frequent but mostly relate to work that concerns the collection of ethno-botanical data. What was needed from the social sciences was a criticism of the approach in the light of social realities. Unfortunately this was seldom done, except by a few researchers as I will now show.
4. The received wisdoms about slash-and-burn cultivation I explained how agroforestry research favored certain pathways to change to the detriment of others. But misconceptions about the dynamics of slash-and-burn systems preceded the development of this discipline. I will not analyze these misconceptions in length here, as my whole dissertation is dedicated to this endeavor. I will merely present, based on the work of Dove (1983), the main received wisdoms or “myths” that prevailed when ICRAF started to develop its technologies in the early 1980s. Dove (1983) argued that much of the debate “deals not with the empirical facts of swidden agriculture, 276 … but rather with widely accepted myths, and that this explains the failures of developmental schemes involving swidden agriculturalists.” He examines three of these myths and proposes alternate conceptions with a quite provocative and passionate style which is worth quoting at length: 276
Synonym for shifting cultivation.
454 452
One myth is that swidden agriculturalists own their land communally (or not at all), work it communally, and consume its yields communally. The truth is that their land (including land under secondary forest fallow) is typically owned by individual households. It is worked by individual labor forces and/or by reciprocal but not communal work groups, and its yields are owned and consumed privately and individually by each household. A second myth is that swidden cultivation of forested land is destructive and wasteful, and in the worst cases results in barren, useless grassland successions. The truth is that swidden cultivation is a productive use of the forests, indeed more productive than commercial logging in terms of the size of the population supported, and forest-grassland successions are typically a function not of rapaciousness but of increasing population/land pressure and agricultural intensification – the grasses, including Imperata cylindrica, having value both as a fallow period soil-rebuilder and as cattle fodder. A third myth is that “swidden agriculturalists have a totally subsistence economy, completely cut-off from the rest of the world. The truth is that swidden agriculturalists, in addition to planting their subsistence food crops, typically plant market-oriented cash crops as well, and as a result they are actually more integrated into the world economy than many of the practitioners of more intensive forms of agriculture. (Dove 1983, 85) The reality Dove describes in this paragraph, despite sometimes being a slight caricature, matches with the situation in Madagascar. In Beforona, slash-and-burn farmers individually appropriate their land, under the authority of their traditional chief, and consume their harvest at household level (Chapter 5). The fallow period is regarded as part of the cultivation cycle and fallow land is not in free access, as long as no external factor disrupts traditional authorities. Only when the land is fallowed for a duration that exceeds what is regarded as necessary for a new cultivation cycle can it return to the collective and be appropriated by another household. But even so, this can provoke land tenure conflicts between households, confirming that use-rights and land ownership are associated with individuals. Concerning the second point (the intensive character of shifting agriculture), it is more disputable, as shifting cultivation systems have lower carrying capacity than most other agricultural systems. It is nevertheless clear that the productive capacity of slash-and-burn cultivation, when it is not in a situation of crisis, is much higher than
455
what is usually thought. Hunting and gathering of forest and fallow products complement the diet of slash-and-burn farmers. The products are often marketed, can be highly valuable but are usually overlooked in economic assessments. Concerning the third point, we clearly saw, in the case of Beforona (Chapter 3 and 5), that shifting cultivators market their products and develop cash crops (coffee, ginger, bananas) just as other farmers do, as soon as they have investment capacity, access to market and equal advantages with their competitors. When they do not have “comparative advantage,” because of remoteness and lack of capital, they remain out of the market, but this has nothing to do with their culture or traditions. Dove concludes that these myths “have generally facilitated the extension of external administration and exploitation into territories of … swidden agriculturalists, and hence can perhaps best be explained as a reflection of the political economy of the greater societies in which they dwell” (Dove 1983, 85). They support the domination of an "urban and governing elite" (Dove 1983, 96) over a "politically vulnerable peasantry." They help to justify "self-serving administration, expropriation, and exploitation of rural lands, especially forest lands" (Dove 1983, 95), for the development of plantations and logging activities. These conclusions are consistent with the logic of the VMCRCC project analyzed in Chapter 6. We will see in Chapter 13 their importance to understand the situation of slash-and-burn farmers worldwide. Dove’s (1983) outcomes may have received little attention at that time, as shown by the low scores of “political economy” and “marginalization” in Figures 18 and 19 (pages 451 and 453). It was important, however, to mention Dove’s work because it anticipated some of the conclusions that the ASB Consortium will derive twenty years later, when its strong rhetoric will eventually collapse. This institutional and scientific background being set, I can now go back to the work of the ASB Consortium and analyze the building of the strong ASB rhetoric.
456
5. The first phase of the ASB Consortium In this section, I will summarize the main outcomes of the first phase of the ASB Consortium. I will present first the main statements of the Consortium, then present
its
biophysical
outcomes,
its
socio-economic
outcomes,
and
its
recommendations.
5.1. Main statements The first synthesis of the ASB Consortium (Sanchez and Hailu 1996, 1) starts with a statement of the situation concerning Slash-and-Burn cultivation: The continuing pace of tropical deforestation and its negative effects on global climate, biodiversity and watershed stability is one of the major worldwide environmental concerns of today. About 60% of the current tropical deforestation is carried out by small-scale farmers for agriculture settlement (World Bank 1991). The bulk of this rain forest land being deforested, which totals about 10 million hectares per year, is cleared by slash-and-burn methods. Most of it is eventually abandoned in some form of shifting cultivation…. Many farmers practicing slash-and-burn are migrants to the humid tropics, displaced from other parts of their country because of a series of socioeconomic constraints. Others are indigenous peoples faced with increasing land pressures. Together they number between 300 and 500 million (FAO 1985), are among the poorest people in the rural tropics, and face formidable challenges. The problem, therefore, is not only one of environment but one of human equity as well. (Sanchez 1996, 1) This introduction reveals a strong awareness of the links between poverty and deforestation and states explicitly that conserving tropical forest is both an environmental and a social challenge. A first paper, by Brady (1996), establishes more precise assumptions, by making a distinction between (1) traditional slash-and-burn systems implemented in areas with low population density by communities that acquired a deep knowledge of their ecosystem, (2) slash-and-burn systems practiced under higher densities and in a
457
situation of crisis, due to reduced fallow period, and (3) slash-and-burn systems practiced by migrant farmers who lack local knowledge and adopt irrelevant techniques. In the first case, the system is regarded as sustainable. In the second and the third, it is not. Demography and migration, by determining the shift from equilibrium to nonequilibrium states, would be the main causes of deforestation. Slash-and-burn systems would meet their limits sooner or later due to demographic expansion. However, technological changes are expected to allow increasing carrying capacity and avoidance of this crisis, which leads the ASB Consortium to eventually adopt the sustainable development discourse. This recalls the survivalism discourse presented in Chapter 8 and the Malthusian perspective where technological innovation would be the cause, not the consequence, of population growth.
5.2. Biophysical outcomes Juo and Manu (1996) and Palm et al. (1996) analyze the chemical and biological dynamics of soil under slash-and-burn systems. They reinforce the nutrient cycling model of Nye and Greenland (1960), proposed 35 years earlier, but “have not added much to understanding the controls or improving and manipulating certain components in the system” (Palm et al. 1996, 61). Alegre and Cassel (1996) compare soil physical properties under unsustainable slash-and-burn systems and their alternatives. They show that slash-and-burn degrades the soil less than tillage and that the best results, in terms of erosion control, are obtained with agroforestry systems that include cover crops and trees. But the alternatives they compare are mechanical clearing and tillage using a bulldozer, enriched fallow using exotic species and alleycropping. They say nothing about the systems based on crop rotation and plowing implemented by communities of farmers on the basis of their local knowledge. If these
458
techniques, rather than working the land with a bulldozer, had been used as a control treatment, the outcomes would have certainly been quite different.
5.3. Socio-economic outcomes Vosti and Witcover (1996) explored the “household perspective” of slash-andburn agriculture. These authors propose a simplistic linear decision making model where households set their objectives, assess their resources, consider external constraints, choose their activities, provoke environmental impacts, reassess their resources and then repeat this cycle (Figure 20). They recognize that their model is a simplification and that things may be less sequential and more “simultaneous, or iterative” in the reality (Vosti and Witcover 1996, 26). They also assert that there can be a trade-off between long-term and short-term objectives that their model fails to account for. They cite the work of Rausser (1980), who showed that “an extremely limited natural resource base, or severely constrained opportunities to transform resources at hand, … can limit goals to survival goals, and the action time horizon to a short one” (Vosti and Witcover 1996, 26). Using the terms of their model, they assert that farmers may place the highest value on maintaining or enhancing the human resource base, perhaps at the expense of the natural resource base. Under such survival conditions, a short-term action horizon is not myopic but rational.… Natural resources mining may be the only way, not the shortsighted way, to meet short-term goals. (Vosti and Witcover 1996, 26) The case study in Beforona (Part III) confirms these limitations of Vosti and Witcover’s (1996) model. I showed that the trade-off between long-term and short term objectives is of paramount importance in systems in a situation of crisis, because the environmental aspect of the crisis calls for long term solutions while the economic aspect calls for short term ones.
459
Figure 20: Grid for analysis of farmer’s strategies. Source: Vosti and Witcover (1996, 25).
460
Another limitation of Vosti and Witcover’s model is that despite the fact that it assumes that farmers make rationale choices, it says nothing about the strategies that illustrate this rationality. We saw in Chapters 5 and 6 that these strategies are the minimization of risks and the maximization of labor productivity, at least in the case of vulnerable communities practicing extensive systems. Vosti and Witcover’s (1996) model did not consider these strategies in a satisfactory manner.
5.4. Outcomes in term of recommendation The last paper of the ASB’s first phase synthesis (Hardwood 1996, 78) proposes a diagram that visualizes “development pathways toward sustainable systems following slash-and-burn” (Figure 21). These pathways depend on geophysical, technological, social, economic and political variables such as land degradation, available and affordable technologies, market opportunities, population pressures and public support in the form of subsidies and incentives. Both the endogenous and externally designed intensification pathways described in Chapter 6 are represented in the diagram. Home gardens (a form of “mixed tree systems”) and paddy fields (“intensive lowland rice” cultivation) were among the most promising alternatives to tavy in Ambodilaingo (Chapter 5). “Plantation crops” was also a viable alternative when coffee cultivation was profitable, until the early seventies. The development of “low intensity permanent cultivation,” followed in the long-term by more intensive “mixed arable systems with animals,” was the strategy of the NGO Children of the Forest (Chapter 7). It was translated into action by the organization of agricultural contests. There is, however, a significant difference between what is represented in this diagram and the realities I presented in Part III. New land uses in Beforona usually do not result from the transformation of slash-and-burn systems. They rather develop along side them. Bottom land and the lower parts of slopes are improved by
461
Figure 21: Development pathways from slash-and-burn systems. Source: Hardwood (1996, 78).
462
building traditional dams, digging irrigation canals and planting fruit trees and banana trees. Beside this, slash-and-burn continues to be practiced on steep slopes and degraded land. In the farmers’ perspective, the two systems need to coexist and complement each other until the new system achieves its full productive capacity. The case of Bukeye in Burundi (Figure 17 pages 448 and 449) showed the same logic. Hardwood’s (1996, 78) diagram, and the expression “development pathway toward sustainable systems following slash-and-burn” may lead to overlooking this point. Moreover,
Hardwood’s
analysis
did
not
lead
to
precise
policy
recommendations. The author calls for imperative technological changes but is not explicit about the direction for this change. He advocates for policies that address the variability of ecological and socio-ecological contexts and calls for long-term investments, opening to markets and economic development in general. But he does not derive recommendations from the pathways he presents in his diagram. This shortcoming may reveal an implicit assumption that agroforestry technologies (alleycropping, contour strips and improved fallows) are the alternatives that should be supported, whatever can be concluded from the diagram. Hardwood’s (1996) article may be aimed at pushing an agenda, not at questioning it. Coming back to Vosti and Vitcover’s paper (1996), it remains similarly vague in terms of recommendations. Tools for policy implementation are proposed (market development, infrastructure, credit, land tenure policies, development of new technologies) but no pattern emerges clearly defining their specific objectives. The authors contend that “the ultimate objectives of policy-makers and development programs should be to enhance the compatibility among sustainability, growth and poverty alleviation” (Vosti and Witcover 1996, 35), because an effect on “any one of these three development goals “will likely have ripple effects on the other two” (Vosti and Witcover 1996, 35). The authors propose to apply these objectives “at the level of
463
technologies and institutional arrangements” and call for more interdisciplinary research in a “more comprehensive conceptual framework” where “biophysical and social scientists must share a common three-part goal”: sustainability, growth, and poverty alleviation (Vosti and Witcover 1996, 36). In sum, they put a strong emphasis on the need to carry out more research and provide a framework to guide this research, but they do not propose and question strategies. As for Hardwood (1996), they push an already defined agenda rather than question it in the light of their outcomes. The controversy about the best pathway to change may have already been settled.
5.5. Conclusion In conclusion, the excessive focus of the ASB Consortium on biophysical sciences led it to exclude domains of knowledge and impeded it from questioning its research agenda. A first domain of knowledge exclusion is the political economy framework of slash-and-burn cultivation, which Dove (1983) identified as being paramount. The consequence is that the real causes of deforestation and slash-andburn cultivation are not analyzed while solutions are proposed on the basis of weak assumptions about these causes. A second domain of knowledge exclusion concerns the indigenous pathways for change, such as those represented in Hardwood’s (1996) diagram. The consequence is that the Consortium developed strategies that were at odd with the economic constraints and the strategies of slash-and-burn farmers, as we saw in Chapter 6. We will see, however, that these forces of exclusion will be levered after the second phase of the Consortium’s work.
6. The second phase of the ASB Consortium In this section, I will present the outcomes of the second phase of research conducted by the ASB Consortium. I will show that, despite the fact that this research
464
was based on the same initial statements and used similar methodologies, it led to a significant paradigmatic shift and displaced deforestation from being a technological challenge (the finding of alternatives to slash-and-burn) to being a political economy issue (the destruction of forests by market forces).
6.1. A continuation of the first phase? Palm et al. (2005a) propose a synthesis of the ASB Consortium’s outcomes at the end of the second phase. A first series of articles concerns biophysical aspects such as greenhouse gas fluxes (Murdiyaros et al. 2005), carbon losses and sequestration in relation with land use change (Palm et al. 2005b) or below ground biodiversity (Bignell et al. 2005). A second series of articles concerns the assessment, at a farming systems’ level, of alternative land uses such as rubber agroforestry systems (Wibawa et al. 2005), coffee and pasture systems (Carpentier et al. 2005) or improved Imperata grassland (Purnomosidhi et al. 2005). Macro-economic aspects are addressed by Kurniatur et al. (2005) (forest conversion dynamics), Cattanao and San (2005) (macroeconomic driving forces of deforestation), Carpentier et al. (2005) (effect of coffee price in the case of pasture and coffee plantations) and to a lesser extent by researchers who worked at national scales (Valentin and Vosti 2005; Partohardjono et al. 2005; Gockowski et al. 2005; White et al. 2005; Suraswadi et al. 2005). Overall, the researchers concentrated their attention on biophysical aspects and local socioeconomic dynamics. Land uses and sites were compared using mostly quantitative criteria (Tables 20 and 21) which were not instructive about social dynamics that occurr at larger temporal and spatial scale, such as population migrations, environmental and forest policies, conflicts over resources, the accumulation of assets, market dynamics or the development of infrastructure. Policy recommendations were made, but rather as a way to provide an appropriate regulatory framework for
465
Table 20: ASB consortium criteria for comparing land uses
466 Source: Palm & al (2005, 26)
Table 21: ASB Consortium criteria for comparing sites
467 Source: Palm & al (2005, 17)
implementing best bet alternatives defined according to micro-economic and biophysical criteria (see research framework in Figure 22). In short, political economy issues are present but are not at the core of the research agenda of this second phase. As in the first phase, the criteria used to assess farming systems and alternatives are disputable. For example, Vosti et al. (2005) do not raise the question of labor opportunity cost. The consequence is that cropping systems are assessed in isolation, not in combination. This can be misleading because poor farmers do not specialize in the most profitable system according to a simple criterion. They prefer to combine cropping systems, which allows them to fill their working calendar, to avoid dependence on a single crop and to minimize risks, as I showed in the case of Beforona (Chapters 5 and 6). In other words, the concept of “best bet alternative” may not be appropriate, especially in the case of vulnerable communities. Vosti et al. (2005), for example, contend that swidden agriculture will disappear when it will become unprofitable in terms of return to labor, and when markets provide more profitable opportunities. But we saw, in the case of Beforona (Part III), that vulnerable farmers who have access to markets often continue to grow food crops in low profit slash-and-burn systems in order to guarantee food security, especially in case of market instability (Barrett 1999).
6.2. Toward a political economy of slash-and-burn cultivation? In the end, the consortium nevertheless produced relevant outcomes at political economy level. Its researchers were involved in collaboration with policy makers worldwide, which led them to address issues beyond their research framework. For instance, “work with the Indonesian Ministry of Forestry resulted in a presidential decree that recognized the property rights of the people managing the complex agroforests on government lands in Sumatra (Fay et al. 1998).” (Palm et al. 2005a, 28)
468
Figure 22: The ASB research framework. Source: Palm et al. (2005, 12).
469
“At the regional level ASB scientists have promoted enabling policies to support community-based forest management plots with the government of the state of Acre in Brazil” (Palm et al. 2005a, 28). These involvements led the Consortium to propose explanations of deforestation dynamics that went beyond the simplistic Malthusian model it formerly adopted and the distinction between equilibrium and nonequilibrium systems that it suggested during the first phase. Slash-and-burn farmers were now regarded as being connected to a larger political and economic context. It was acknowledged that solutions for transforming this land use may not be only technical. A foreword by Jeffrey Sachs reveals the Consortium’s commitment to push these new conclusions in the international agenda: The traditional interpretation of tropical deforestation has usually proceeded in something like the following way. A rising population of smallholder farmers at the forest margin – the boundary between farm and forest – leads to deforestation as forests are cut to make room for new farms. At the same time, existing farmland is abandonned because of land degradation, soil erosion, and soil nutrient depletion. The loss of existing farmland is exacerbated by the pressure of shortened fallow, caused by the rise of population densities. In this traditional view, the best way to slow or stop deforestation would be to raise productivity on existing farms in a sustainable manner – for example, through the systematic replenishment of soil nutrients, so that pressures to expand into new lands can be ceased. There are of course important aspects of truth in this conventional view, but as the studies in this volume make clear, the situation is far more complex. Natural population growth on the forest margin is not the only, or even the key, driver of deforestation. Population growth often results from in-migration of settlers, rather than from the natural population increase among existing residents. Ironically, in such circumstances, intensification of agriculture techniques, even in a sustainable manner, can increase rather than decrease the rate of deforestation, by raising the profitability of farming and thereby inducing the in-migration of settlers to the forest margin. There may be a strong case for improving the productivity of agricultural practices, but that step alone may not solve the problem of deforestation.” “Policies will be needed that explicitly aim to tilt the incentives toward forest conservation. It may be advisable, for example, to compensate landowners for the conservation of nonmarketed ecosystems services such as conserved habitats and sequestered carbon. (Sachs 2005, ix)
470
This new interpretation clearly announces a paradigmatic change. The profitability of land use conversion replaces demography as the main explanation for deforestation and payment for environmental services replace intensification as a privileged strategy to halt forest clearing. Several articles in the book confirm this move. Sanchez et al. (2005) provide more arguments in favor of non-Malthusian explanations of deforestation. Based on the work of other scholars, they assert that 72% of the original cover of tropical forests was converted to other land uses (Myers 1991, FAO 1997, in Sanchez et al. 2005); that 96% of these forest losses result from the activities of small-scale farmers (Amelung and Diehl 1992, in Sanchez et al. 2005); that an estimated 37 million of these farmers practice some form of shifting cultivation (Dixon et al. 2001, in Sanchez et al. 2005); but that in spite of this, “no direct relationship between deforestation and population growth or economic growth has been found” (Sanchez et al. 2005, 4). Several studies would converge to this conclusion: Myers (1991) noted that whereas the population of forested tropical countries increased by 15 to 35% in the 1980s, deforestation expanded by 90% during the same period. The recent analysis by Geist and Lambin (2002) shows that in-migration to the forest margins is a much larger factor of deforestation than high internal population growth. (Sanchez et al. 2005, 8) In this new explanatory framework, Sanchez et al. (2005, 9) evoke the marginalization of poor farmers as a key explanation to deforestation: Many of these migrants are unfamiliar with the humid tropics, are largely unaware of the knowledge-intensive practices of indigenous shifting cultivators, and attempt to establish cropping systems that work where they come from (Moran 1981). People in these situations usually lack alternative employment opportunities; have limited access to markets, credit, and information; and often are politically marginalized (Sanchez et al. 2005, 9) In many cases, … [these migrants] are marginalized from society and government support programs. (Sanchez et al. 2005, 4)
471
These outcomes are consistent with the conclusion of Dove (1983), made twenty years earlier (Section 4) and with the analysis of Jarosz (1993), who showed that huge deforestation occurred in Madagascar in the first half of the twentieth century although the population did not increase (Chapter 1). They are also consistent with the situation in Beforona, where a rural elite supported by projects seizes market opportunities and appropriate the land most suitable for intensification (Chapter 6). The outcomes of the ASB Consortium, indeed, converged with narratives developed over more than a decade by political ecology, a sub-discipline of political economy which will be presented more in length in Chapter 13. This Chapter will confirm that marginalization is indeed a key narrative with regard to deforestation dynamics and shifting cultivation systems worldwide. Sanchez et al. (2005) also resituate slash-and-burn cultivation systems in a longer term perspective, leading them to de-dramatize the significance of negative environmental impacts. They adopt a Boserupian model which confirms the departure of the Consortium from simplistic Malthusian explanations. They show that soils, if properly managed, can recover fertility when population increases (Figure 23). Based on the works of Tiffen et al. (1994) and Sanchez et al. (1998), they assert that more people and population growth can lead to less erosion and more trees. This means that there can be a “natural” dynamic that leads societies out of the Malthusian trap, a perspective which matches with the Boserupian model and with the outcomes of Chapter 5. Rather than a series of practices threatened by a Malthusian crisis and in need of technological inputs provided by agronomists, slash-and-burn cultivation is now presented as a stage in land use history that just needs wise policies to evolve: “in some cases, the policy environment does not provide incentives to rehabilitate … degrading lands [line b in Figure 23].” Hence “the challenge is to find policy tools that will provide those incentives” (Sanchez et al. 2005, 7). In sum, the second phase of the
472
Figure 23: Environmental degradation and recovery. Source: Sanchez et al. (2005, 7).
473
ASB work eventually allowed a shift from an approach centered on techniques (developing alternatives to slash-and-burn) to an approach centered on economics and policies (providing the right incentives for the adoption of existing alternatives). Vosti et al. (2005) also propose conclusions that are typical of political economy analyses. First, they contend that “traditional swidden agriculture systems have or will soon disappear because of population pressure and low rates of return to labor” (Vosti et al. 2005, 410). Second, and more importantly, they assert that as “returns to forest conversion are high at all benchmark sites” (Vosti et al. 2005, 411), the rain forests will continue to disappear. They recognize, so speaking, that macroeconomic factors, rather than efforts to develop new technologies, will determine the fate of slash-and-burn systems. They conclude that “most of the fundamental economic factors driving land use system adoption [are] beyond the scope of local, regional, and sometimes even national policy makers” (Vosti et al. 2005, 408). The last paper of the book, by Tomich et al. (2005), confirms again the shift to a political economy of deforestation. It reflects the strong frustration that ASB researchers had during their work when they realized that the key issues were at another level than what they studied. The article expresses this by a desperate and passionate appeal that deserves in length quote: Indeed, everyone in the world seems to want something from the tropical forests. Forest dwellers want to continue aspects of their traditional way of life based on hunting and gathering while improving the welfare of themselves and their families. They are losing their land to migrant smallholders, who clear small amounts of forests to earn a living by raising crops and livestock. Both these groups tend to lose out to larger, more powerful interests – ranchers, plantation owners, large scale farmers, or logging concerns – whose aim is to convert large areas of forest into big money. Outside the forest is the international community, who wants to see the forest preserved for the carbon they store, which would otherwise contribute to global warming, for the wealth of biological diversity they harbor, and for the many other ecosystem services they provide.
474
Deforestation continues because converting forest to other land uses is almost always profitable for the individual, household, or firm that engages in it. However, the society as a whole bears the cost of lost biodiversity, global warming, smoke pollution and the degradation of water resources. (Tomich et al. 2005, 416) Clearly, the authors put into question the global economic system, which leads them to reverse another received wisdom: markets may be the problem rather than the solution, because they drive the deployment of capital and labor in ways that give no consideration to social and ecological issues. Market opportunities may indeed just be a source of hope, a temporary glimmer, as during the boom of oil, coffee and cocoa in Cameroon in the late 1970s and early 1980s (Tomich et al. 2005). Markets also create new needs and provide incentives to tap natural resources and convert forests into commodities. Karl Polanyi (1944) proposed a prophetic vision of the disruptive consequences of the commodization of societies and nature which is worth comparing to the Tomich et al.’s (2005) citation above. Polanyi identified what we call now externalities and expressed in a passionate and relevant way their unforeseen consequences: [As] labor, land, and money are essential elements of industry, they also must be organized in markets.… But labor, land and money are obviously not commodities.… Labor is only another name for a human activity which goes with life itself…. Land is another name for nature, which is not produced by man.…The commodity description of labor, land and money is entirely fictious…. [But it supplies] a vital organizing principle in regard to the whole society, affecting almost all its institutions in the most varied ways…. To allow the market mechanism to be sole director of the fate of the human beings and their natural environment indeed … would result in the demolition of the society…. In disposing of a man’s labor power the system would, incidentally, dispose of the physical, psychological, and moral entity “man” attached to that tag. Robbed of the protective covering of cultural institutions, human beings would perish from the effect of social exposure; they would die as the victims of acute social dislocation through vice, perversion, crime, and starvation. Nature would be reduced to its elements, neighborhoods and landscape defiled, rivers polluted, military safety jeopardized, the power to produce food and raw material destroyed.… No society could stand the effects of such a system of crude fictions even for the shortest stretch of time unless its human and natural
475
substance as well as its business organization was protected against the ravage of this satanic mill. (Polanyi 1944, 76-77) Polanyi wrote his book during the second world war and interpreted the collapse of Western civilisation at that time as a consequence of the development of market driven industrial societies. Sixty years later, the tropical forests and its inhabitants may be the last frontier that market economies are destroying. Areas affected by slash-and-burn cultivation may be the ultimate locations where the commoditization of land (nature) and labor (people) operates its destructive effects. In other word, slash-and-burn farmers and tropical forests may pay together the same market externalities, for the benefit of developed nations that can purchase products harvested or grown on the forest frontier. This phenomenon is not new but as long as the agricultural frontier was open, the forest played the role of absorbing the less competitive people on the market’s verges. Now that the consequences of global environmental impacts are felt, conservation programs close the frontier and deny access to these last resources without providing alternatives, while the market continues its exclusion process. This will lead to increasing starvation, poverty, migration, urban violence and crime, to use the words of Karl Polanyi. Something different from repression has to be done to stop this tradeoff between people and biodiversity. If not, impoverished farmers will reopen the frontier to sustain their life, until no forest will remain, and both nature and its people will be lost.
7. Conclusion In conclusion, the work of the ASB Consortium was trapped by the fact that it defined the solution (the technical alternatives) prior to comprehensively addressing the problem (the regional dynamics of deforestation). The Consortium was victim of a strong rhetoric built up by 20 years of agroforestry research centered at ICRAF. It implicitly assumed that creating cropping systems that maximize positive soil-plant
476
interactions would lead to wise alternatives to slash-and-burn and solve the problem of deforestation. This assumption was not a working hypothesis because the research program was not designed to test it. It was rather the consequence of scientists’ preference for “hard” quantifiable facts. It was a sign of the excessive trust in technologies and of the preference for biophysical sciences that characterize the positivist reductionist paradigm. Agroforestry, which does not focus on the use of machines and chemical inputs, gave the illusion of a departure from Western modes of modernization. But the attention paid to yield and the design of perfect biophysical systems under controlled conditions showed that the “high modernist” schemes (Chapter 8) were still highly influential. This trust in “technologies” may have provided an excuse for decision makers to not address the problems politically. At the level of researchers, however, there was a genuine commitment to address the real causes of deforestation. Therefore, the Consortium was able to end the second phase of its work with strong advocacy for a political economy of deforestation, despite the fact that this was not an explicit objective of its mandate. It is true that the Consortium’s ultimate objective, according to its research framework (Figure 22 page 469), was to propose policy instruments and institutional mechanisms. But as we saw, the outcomes of the cross site comparison mostly concerned simple biophysical and measurable parameters at farming system scale. These outcomes were not sufficient to elaborate macro-economic policy recommendations beyond advice about how to favor the most profitable land uses. If the Consortium eventually provided macro-economic recommendations, it was because the realities “jumped” into its face. The researchers fully experienced the reality while they did their work. They faced social complexities and saw the limits of the quantitative tools and methods they had designed, and of the questions they were asking. They expressed their outcomes on the margin of their work, putting in their
477
words the same emotion they felt during their experience. They experienced a reality which became accomplice of their knowledge and helped them to reject the constructed
misrepresentations
that
had
determined
their
hypotheses
and
methodological choices. This story justifies the methodology I employed to analyze the agrarian system of Beforona (experiencing rather than measuring realities). It also validates the epistemological model I adopted in Chapter 9: the real world has a word to say on the advancement of knowledge. It is one of the accomplices of our knowledge and can help us to distance ourselves from our misconceptions. We just have to give it more opportunities to unfold its signs under our eyes. To accomplish this, we sometimes have to take some distance from our methods and to listen to our emotions, as artists would do (Part I). Unfortunately, discourses also unfold their own signs under their own eyes (Chapter 9), which is a strong factor of inertia. The ASB rhetoric will pervade through the representations of development and conservation actors for a long time still, retarding the translation into action of the last Consortium’s outcomes. Most operators in Madagascar are not yet engaged in political economy analysis and will continue to believe that “best-bet” alternatives elaborated according to a minimum set of criteria and proposed to farmers — with no regard to regional dynamics and international context — will suffice to achieve results (Chapter 6). We must hope that development and conservation actors will soon follow the work of the Consortium and shift their focus to political economy issues. The VMCRCC project (Chapter 6) could have been an opportunity to implement this shift but failed to seize this opportunity. It is true that the agroforestry alternatives developed by ICRAF are no longer as fashionable as they used to be in Beforona. But unfortunately, zero tillage and cover crop technologies are now replacing alley-cropping and improved fallows for providing the illusion that
478
deforestation and poverty can be fixed by technical solutions. 277 Based on these concepts, a strong new rhetoric is appearing and could send development actors back to ignorance.
277
This argument is not a point against zero tillage and cover crops. It is a point against the fashion that surrounds these techniques, and the risk of hegemony this fashion could lead to.
479
CHAPTER 11 COMMUNITY BASED NATURAL RESOURCE MANAGEMENT (CBNRM)
1. Introduction In this chapter, I analyze the community based natural resource management policies implemented in Madagascar during the last ten years. Central to these policies are the GELOSE 278 and GCF 279 approaches, supported by the GELOSE law (Gouvernement de Madagascar 1996) and several application decrees (Gouvernement de Madagascar 1998; Gouvernement de Madagascar 2000a; Gouvernement de Madagascar 2000b). These approaches have not been implemented in Beforona, but are a central component of the National Environmental Action Plan (Chapter 2). They have been utilized by the LDI and PTE projects on other sites as a tool to control slash-and-burn cultivation, as we will see. But this chapter will also contribute to understanding the underlying logic that has lead the rural elite to capture most of the aid provided by projects in Beforona (Chapter 6). This GELOSE legislation enables the transfer of management rights to local communities and has been implemented in about 500 sites across the country. The GCF approach relates specifically to forest resources and is framed within both the GELOSE law and the Forest law (Gouvernement de Madagascar 1997). For more convenience, I will use the terms Management Transfers (MT) to designate both the GELOSE and GCF approaches. The chapter starts with a critical analysis of the participatory discourses within which community based natural resource management approaches are embedded (Section 2). This analysis, mostly based on the scholarly work of Rahnema (1992),
278 279
Local Secure Management. In French: Gestion Locale Sécurisée. Forest Management Contract. In French: Gestion Contractualisée des Forêts.
480
enables the formulation of the hypotheses that are utilized in the ensuing sections. Section 3 describes the implementation of community based natural resource management approaches in a few sites I had the opportunity to visit during my stay in Madagascar. It compensates for the fact that I did not present cases of management transfer in Part III. 280 Section 4 presents the GELOSE legislation itself. It summarizes the historical process that led to its creation and critically examines the GELOSE law and its application decrees. Section 5 analyzes the way the GELOSE approach is promoted in policy maker circles. It critically analyzes the outcomes of a national evaluation that was conducted in 2004. It uses the reports of this evaluation as empirical material for analyzing the discourses associated with the GELOSE. In Section 6, I confront my outcomes to those of other work conducted on the subject. In the conclusion (Section 7), I propose solutions for a move toward a more genuine community based natural resource management approach.
2. The participatory discourse I will not review here the abundant literature about participation. Such reviews have already been conducted and I consider that it is not worth repeating this exercise. Instead, I focus on an article by Majid Rahnema (1992) and a review of the subject by Cooke and Khotari (2001). These works critically analyze what is actually going on in the real world in the name of participation, as opposed to a large number of essays which focus on the theoretical framework of the concept and on the tools and methods created on behalf of it. Rahnema (1992) starts by reminding us of the basic definition of the word “participation”:
280
No GELOSE or GCF was implemented in Beforona when I conducted my inquiries, in 2001.
481
for the Oxford English Dictionary, participation is “the action or act of partaking, having or forming a part of.” In that sense, participation could be either transitive or intransitive; either moral, amoral or immoral; either forced or free; either manipulative or spontaneous. (Rahnema 1992, 116) From this definition, Rahnema (1992) argues that a subject can participate in a project’s activities to defend his interests (transitive meaning), or just to be there and show his presence (intransitive meaning). Participation can also be decided by the participant himself (free participation) or result from his incorporation by a mechanism out of his control (forced or manipulative participation). In sum, participation means nothing if we do not indicate participation in what, for what (if for anything) and in which manner. It is also important to set the historical background that determined the appearance of the concept. Still according to Rahnema, the words “participation” and “participatory” appeared for the first time in the development jargon during the late 1950s, [due to the influence of social activists who] attributed most of the failures of development projects to the fact that the populations concerned were kept out of all the processes related to their design, formulation and implementation. In their great majority, they started to advocate the end of “top-down” strategies of action and the inclusion of participation and participatory methods of interaction as an essential dimension of development. (Rahnema 1992, 116) On another side, the Development Establishment was obliged, some years later, to recognize a structural crisis. Donors and recipient national governments were witnessing the fact that the billions spent on development projects had failed to produce the expected results, often even adding new problems to the old. (Rahnema 1992, 116) This story explains how a concept hitherto regarded as subversive became accepted and wished by all parties. During the 1980s, participation quickly became the norm in development projects. The work of Robert Chambers (1985) was particularly influential in its application in the field of rural development. Rapid Rural Appraisal and Participatory Rural Appraisal methodologies were designed under the
482
participatory ideal. The “ostensible aim” was “to make “people” central to development by encouraging beneficiary involvement in interventions that affect them and over which they previously had limited control or influence” (Cooke and Khotari 2001, 5). Using Dryzek’s (2005) terminology, participation was expected to favor a shift from administrative rationalism to democratic pragmatism (Chapter 8). In this context, the concept was “no longer perceived as a threat” (Rahnema 1992, 117) by state power but became politically and economically attractive. It displayed the image of a democracy at work and was expected to lower the cost of development policies. Unfortunately, “participation has come to be “disembedded” from the sociocultural roots which had always kept it alive” (Rahnema 1992, 120). It became “perceived as one of the many “resources” needed to “keep the economy alive” ” (Rahnema 1992, 120). Early criticisms tried to reframe the approach in its early ideal and called for a “popular participation” (Rahnema 1992, 120) committed to a true empowering of local people. “Popular knowledge” (Rahnema 1992, 121) was expected to help destroy “the pernicious monopoly of the dominant paradigm” (Rahnema 1992, 122). But for Rahnema, there is not enough evidence to indicate that a new kind of knowledge did emerge…. Instead, there are indications that the way many activists interpreted their mission contributed to dis-valuing the traditional and vernacular forms of power. (Rahnema 1992, 122-123) In the end, participation could have acted like a Trojan horse which may end up in substituting a subtle kind of teleguided and masterly organized participation [substituted] for the old types of intransitive or culturally defined participation, proper to vernacular societies. (Rahnema 1992, 125) In sum, participation may have favored the destruction of local forms of knowledge and power instead of the empowerment of local people. It ended in becoming a manipulative tool spreading a global culture and favoring the control of local communities and minority groups by states and other powerful actors. The
483
tension between the activist root of the concept and the form it eventually took led to a situation of double bind: participants were called to participate in the design of projects and activities that had already been designed for them. Chapter 6 provided an example of this mechanism. Farmer associations (the Kolo Harena) were created to favor a bottom-up process of agricultural innovation but the rules (no fire, yield maximization, agrobiological technologies) had already been fixed by actors located up the chain of decision. The “participatory top-down,” to use an expression coined by a Malagasy colleague, was at work while the fashionable and idealistic character of participation, in the virtual and idealistic world of discourses, prevented the concept from being criticized. This helped schizophrenic practices to become hegemonic and widened the gap between realities and representations. Cooke and Kothari’s (2001) conclusions are of the same vein. These authors reviewed the criticisms formulated against the participatory discourse. They distinguished two types of critics. The first “focus on the technical limitations of the approach and stress the need for a re-examination of the methodological tools used, for example in PRA” (Cooke and Kothari 2001, 5). They are criticisms “within the orthodoxy” (Cooke and Kothari 2001, 5). The second “pay more attention to the theoretical, political and conceptual limitation of participation” (Cooke and Kothari 2001, 5). They are a critique of “the orthodoxy per se” (Cooke and Kothari 2001, 7). Cooke and Kothari (2001), as does Rahnema (1992), enter in the second category. These authors get to the same conclusion that participation can be used as a subtle and manipulative way to assert control over communities. They use the term “tyranny” to shake the received wisdom that participation is democratic and identify three particular sets of tyrannies. First is “the tyranny of decision making and control” (Cooke and Kothari 2001, 7): participatory facilitators can “override existing legitimate decision making processes” (Cooke and Kothari 2001, 7). Second is the
484
“tyranny of the group” (Cooke and Kothari 2001, 7): already powerful actors can strengthen themselves still more by participating in decision making processes. Third is the “tyranny of the method” (Cooke and Kothari 2001, 8), which favors the exclusion of groups who have advantages they cannot use in participatory methods they did not design. In sum, participation discourses and methods were appropriated by mainstream development actors and now serve as a tool to spread the dominant modernization paradigm. They are used “like a Trojan horse” (Rahnema 1992, 125), sent to reshape local knowledge and representation in order to make realities match with this paradigm. They attempt to fill the interstices of local discourses with representations elaborated outside, which can lead to robbing entire populations “of their possibilities of relating to each other, in their own best interest” (Rahnema 1992, 126). They are, to use the words of Foucault, a technology of power which acts by reshaping local knowledge and practices. But fortunately, participation approaches can also fail to serve these objectives, as we will see in the analysis of the GELOSE legislation.
3. A few short case studies This section presents the outcomes of short field visits to management transfer sites I had the opportunity to make between 2002 and 2006. Map 14 shows the location of these sites. I will first present the four sites located around the eastern rain forest, and then present the two sites located in western Madagascar.
485
Miandrivazo
Ambondromamy
Beforona
Tsinjoarivo
Ambositra (Zafimaniry)
Ikongo
Map 14: Location of management transfer sites investigated. Map from FTM (National Geographic Institute), in Rakotoarijaona (2005, 3). Additions by Pollini (sites indicated with blue arrows)
486
3.1. Tsinjoarivo Tsinjoarivo is located in the Ambatolampy district, on the western edge of the rain forest corridor, about 80 kilometers southeast from Antananarivo. The PDFIV 281 project, implemented by the GTZ 282 agency and funded by the German government, works in the area 283 and supported the creation of forest management associations. I was invited to visit the PDFIV project in 2002 and spent one week in Tsinjoarivo, accompanied by a field technician. During this stay, I conducted informal qualitative interviews with a total of about 15 households. 284 Two ethnic groups live in this commune: the Betsimisaraka, who come from the east and typically sustain their livelihood by doing slash-and-burn cultivation (tavy), and the Merina, who come from the west (from the more populated highlands) and already have a long experience of irrigated rice cultivation (Chapter 1). The Betsimisaraka were the first settlers and for this reason, their traditional leaders managed customary rights, controlling access to land and deciding which patch of forest could be cleared for slash-and-burn cultivation. The Merina were more recent migrants attracted by trading opportunities. They mostly came to buy sugar cane alcohol produced by the Betsimisaraka, which they sold in the highlands. Tsinjoarivo is an important center for sugar cane alcohol trading. But during their travel across the region, the Merina had the opportunity to find land suitable for developing irrigated rice cultivation. They recognized the legitimacy of Betsimisaraka traditional leaders
281
Integrated Forest and Village Development Project. In French: Projet de Développement Forestier Intégré Villageois. 282 German Agency for Technical Coopération. In German: Gessellschaft fyr Technische Zusammenarbeit. 283 The PDFIV project works in several communes of the Ambatolampy district. The information in this section only concerns the Tsinjoarivo commune. 284 Aubert (1999) also visited this site and expressed concerns similar to those evoked in this section. My outcomes are not based on her work as I read it after I wrote this section.
487
and negotiated their installation through matrimonial alliances or by economic transactions. When the project started to work in the area, it created forest management associations which were not ruled according to the GELOSE law but were intended to be later adapted to this legislation. The Merina, due to their higher education level and better knowledge of urban and foreign culture, were in a better position to take leadership in the associations. As these “modern” associations were granted some legal power to control vegetation clearing, the control of access to land shifted from traditional Betsimisaraka authorities Merina people supported by the PDFIV project. As the Merina do not depend on slash-and-burn cultivation, they accepted management rules that matched with the conservation objectives of state and project representatives. This provoked a significant reduction of the land available to practice slash-and-burn cultivation, leading to social conflicts between Merina and Betsimisaraka people. Many Betsimisaraka left the region and moved eastward, in the search for other land to do slash-and-burn cultivation. Some just went across the river, where the project does not intervene, and “waited for the foreign people to leave so they could take their land back” (a farmer in Tsinjoarivo, 2001). Those who stayed in the region often hire their labor force to the Merina people. Between the bottom land occupied by migrants and the top hills protected by new management rules, they have little room remaining to sustain their livelihood. The project proposes alternative cropping systems (mostly agroforestry technologies such as alley-cropping and contour strips) supposed to compensate for this lack of land. But most interviewed beneficiaries, including the Merina people, consider that these systems are not applicable because they require excessive investment for their establishment and excessive labor input for their maintenance. The views they expressed were similar to those expressed by farmers in Beforona (Chapter 6). In sum, due to the creation of
488
forest management associations, pressure on land and forest was displaced in time and space, social conflicts increased, management authority was transferred from legitimate community leaders to a minority that accepted to promote centrally designed conservation policies, and inefficient alternatives to slash-and-burn served as an excuse for enforcing these coercive policies through forest management associations. In the end, however, all this could just not work. The project will leave, the Betsimisaraka people could come back and the traditional mode of regulation could be put back in place.
3.2. The Zafimaniry area The second site is the Zafimaniry area, located farther south in the Ambositra district, still on the western edge of the rain forest corridor. I visited the area twice in 2005 to supervise the field work of a Malagasy student (Rabetsimamanga 2005) of the Antananarivo forestry school. 285 I spent a total of six days in the area, visiting about five villages with the student, her professor (during three days) and the president of the COBA 286 federation. In this area, migrants from the highland are not many. The people who occupy the region belong to the Zafimaniry group, which has cultural affinities with both the Betsileo, who live in the highlands, and the Tanala, who live farther southeast and practice slash-and-burn cultivation (Chapter 1). Management Transfers ruled according to the GELOSE legislation were implemented in the area with support from SAGE, 287 a line agency in charge of implementing the GELOSE component of NEAP (Chapter 2). The main objective was to improve the management of valuable woods 285
Département des Forêts de l’Ecole Supérieure des Sciences Agronomiques. Lending support to this student was part of my job as a technical advisor at the Ministry of Environment, Water and Forest, from February 2004 to February 2006, for the FSP-GDRN project (note 5 page 23). 286 The type of association created by the GELOSE legislation. See Section 4.2.2. 287 Environmental Management Service. In French : Service d’Appui a la Gestion de l’Environnement.
489
utilized to fabricate sculptures sold in the district town of Ambositra, and sometimes exported. This activity is a significant source of income in certain villages and rendered the region famous and attractive for tourists (Rabetsimamanga 2005). The associations in charge of managing the resources (the COBA) are administrated by the Zafimaniry people. A federation of all COBAs was also created. It is headed by a local leader who has both a high level of education (he has been a teacher and speaks excellent French), a strong commitment to the well being of his community and a strong legitimacy (he seems to belong to a royal lineage). To the contrary of the preceding case, the conditions are very favorable for the implementation of a genuine community based natural resource management approach. However, most farmers consider the management rules enforced by the COBA as something coming from outside. Their main concerns are the restrictions on the practice of slash-and-burn cultivation and on access to new land by forest clearing. They accept the rules but explain that they will not be able to support this sacrifice much longer if no alternatives are provided to compensate the restrictions. The leader of the federation himself expresses views that appear sometimes contradictory. He is proud to show foreign visitors that he assumes his responsibilities, that no forest is cleared, but he also agrees with the criticisms made by ordinary farmers. In sum, the COBA basically works but is an instrument for the implementation of an environmental agenda designed outside. The main point in this agenda is that no more forest must be cleared, whereas according to community management logic, forests are a reservoir land must remain available for future agriculture expansion, except for sacred places that shelter graves or are protected by taboos. The GELOSE approach creates the illusion that the new rules are designed by the communities, as shown by the pride of COBA leaders when GELOSE objectives are fulfilled. This creates a second illusion: that restriction to forest clearing is socially accepted and does not need to be compensated. The
490
schizophrenic pattern of the GELOSE approach was sometimes very explicit. Once mutual confidence was established between interviewers and interviewees, farmers asked us to do something to compensate the restrictions if the ban had to be maintained. Even the COBA leader agreed with these requests. It seemed that the community considered that we, foreigner, were responsible for the ban on forest clearing, which we are indeed.
3.3. Ikongo The third site I visited is located in the Ikongo district, on the eastern side of the forest corridor (Map 14 page 486). I stayed one week in the area, in 2002, in the company of staff from the LDI project 288 which invited me for this visit. I only conducted a few interviews but had long, intense and very interesting discussions with the project’s team, which was at that time conducting an evaluation of the impact of its activities on the practice of tavy. The people living in this region belong mostly to the Tanala and Betsileo groups and practice slash-and-burn cultivation (mostly the Tanala). The project approach consists of implementing community based natural resource management and in providing alternatives to tavy. Restrictions to the practice of slash-and-burn cultivation and the enforcement of the ban upon forest clearing were clearly the main objectives that justified the creation of Management Transfers. As in Beforona, the condition to receive support from LDI was to stop slash-and-burn cultivation. In order to mitigate these restrictions, the project proposed alternatives to this land use similar to those proposed in Beforona. As in Beforona, both beneficiaries and project staff agreed that these alternatives did not compensate for the productivity loss that resulted
288
LDI intervened in the Toamasina and Fianarantsoa Provinces and had a regional office in each of these provinces.
491
from the abandonment of slash-and-burn cultivation. The preliminary version of an evaluation report even calculated that the loss in terms of agricultural production was between 25% and 50%. 289 The situation is thus similar to that encountered in Zafimaniry region: management transfers are used as a tool to enforce a centrally defined political agenda whose main objective is to stop forest clearing and slash-andburn cultivation. In order to avoid any misreading of these outcomes, I must outline that I am not contending that local communities should be allowed to manage their land according to their own customary rules and continue to clear primary forests. My position on this issue, which is more political than technical and raises ethical concerns, will be developed in Chapter 15. What I am saying here is that the expectation that community-based natural resource management will lead to stop forest clearing is an illusion, and that in cases where this objective is achieved, it is because the COBA serves to channel repressive policies designed at the central level. We will see later that empowering communities can help to slow deforestation, by controlling the intrusion of outsiders who would otherwise contribute significantly to clearing. But the fact that forests are a reservoir land for future agricultural expansion will not be changed by this empowerment. Community can be committed to slow down deforestation, not to stop it. It is rather the productive base of the society that has to be changed if we want to conserve primary forests, as expressed by communities in their repeated appeals for alternatives.
289
I saw this document when I was in Ikongo but did not manage to purchase a copy and lost the reference.
492
3.4. Beforona The last site I visited in eastern Madagascar is the commune of Beforona, to which several Chapters have already been dedicated (Part III). The implementation of Management Transfers was envisioned by the LDI project but was not yet realized when I did my field enquiries, in 2001. Recently (in 2006), a local wood collector tried to collaborate with a few communities, including villagers from Ambodilaingo (Chapter 5), to set up Management Transfers. This collector works for forest operators who have difficulties buying wood legally, due to the ban that was put on the delivery of forest permits until a new procedure would be applied. 290 Besides the few permits that are still valid, Management Transfer is now the only way to harvest wood legally in Madagascar, until the new regulation is operational. 291 Clearly, these new actors of Management Transfers are not concerned by community empowerment. Community leaders were nevertheless ready to sign the Management Transfer documents prepared by the wood collector, because wood extraction provides them with jobs and because their low level of education makes them easy to manipulate (they were promised to be supported for the creation of new paddy fields). Fortunately, the mayor was aware of the manipulative character of the operation, did not share interests with the operator, and blocked the process.
3.5. Two sites in western Madagascar The last two sites I visited are located in Western Madagascar. As for the sites in the Zafimanity region, I visited them when I worked for the FSP-GDRN project, which contributed to the financing of management transfers.
290
The adoption of a competitive and transparent procedure for the delivery of forest exploitation permits was one of the conditions for the implementation of the National Environmental Action Plan. 291 In 2007, the new regulation was applied for the first time in a few sites.
493
One site is located in the Ambodronmamy (Mahajanga Province). Management transfers were set up in this area with technical support from CIRAD, 292 for achieving sustainable charcoal production. I spent three days on the site and interviewed traditional leaders and representatives of management transfer associations (COBAs) in three villages. The COBAs had allowed charcoal makers to legalize activities that were previously illegal and the operation was positively perceived by the beneficiaries. The functioning of the COBA, however, was not clearly understood by most people I interviewed. From their perspective, the main change that occurred concerning charcoal production, after the creation of the COBA, were the legalization of this activity, taxation, the attribution of production quota and a more efficient control. It can be questioned whether the GELOSE approach was necessary to implement these elements, which are essential in any resource management policy. Moreover, private charcoal producers were still delivered authorization by Forest Services while they did not have to pay taxes. This created an unfair concurrence which was cause of discontentment. A workshop aimed at synthesizing the lessons learned from this experience was organized in Mahajanga, in 2004 (Anonymous 2004). 293 One of the main concerns expressed by participants was the unequal concurrence between COBAs and private producers. In the end, I concluded that the operation could lead to improved management if it was generalized to the whole Mahajanga region but that it was questionable whether it was actually a form of community based natural resource management. The next sections will provide more evidence showing that these doubts were justified.
292
International Center for Agronomic and Development Research. In French : Centre International de Recherche Agronomique pour le Développement. 293 This was an activity financed for the project within which I worked. I visited the field site as part of my monitoring activities and attended the workshop.
494
The last site I visited in western Madagascar is located in the Miandrivazo District, where several COBAs were set up to improve fishing management in order to protect endemic fish species encountered in a series of small lakes. The visit I made was typical of the kind of field work donors set up when they want to identify projects to fund. We were many, moving in four wheel drive vehicles, attending meetings and never staying more than a few hours at a site. It was almost impossible to collect relevant and trustable information in these conditions. However, an “anecdotal” history deserves to be related. We visited a community whose members apologized for having cut the grasses that grew around the lake, in the vicinity of their village. They explained that traditionally, they had always cut the grasses along the part of the lake adjacent to the village, but that the COBA changed the rule and no longer allowed them to do that. They violated the COBA rule because the grasses sheltered a crocodile that ate a child and his mother just a few weeks earlier. One can hardly believe that this COBA rule was a genuine community rule.
3.6. Conclusion In conclusion to this section, it can be questioned whether community based natural resource management can be an efficient answer to key environmental issues such as deforestation. Madagascar still has a forest frontier and it cannot be expected that communities will renounce clearing it, which is legitimate in the context of the high poverty that prevails in this country. If we are to improve natural resource management, the main challenge may be to transform the economic base of agricultural production. As the efforts of the international community to conserve the last reservoir of biodiversity is also legitimate, the solution may reside in a negotiation between restriction over access to resources on one hand, and support to the development of new land uses that would reduce the dependence on resources on the
495
other hand. In other words, it would be a matter of compensation for restrictions that are no longer negotiable (restriction of forest clearing), rather than a matter of improving governance. Management Transfers, by maintaining the illusion that improved governance alone can be the solution, lead to overlook the necessity to provide these compensations. The reason for this situation may be the oxymoronic character of the sustainable development concept. Communities need to sustain their development, which implies having access to resources and extending their agricultural land. From their perspective, forests can be depleted of the various products they shelter because they will be converted into agricultural land some day. For the international communities, sustainability implies the conservation of biodiversity. Sustainable development provides the illusion of a common agenda whereas local and global objectives are incompatible, resulting in schizophrenic policies. Only in cases where there is no conflict between these two forms of sustainability, i.e., when primary ecosystems have already been cleared, could this tension be avoided.
4. The GELOSE approach (Secured Local Management) In this section, I will analyze the GELOSE legislation itself and the discourses associated with it, as they are revealed by the GELOSE law and decrees and by the story that led to issue them. I will start with a brief overview of the context within which the approach originated. I will then present the law and its application decrees. I will pay particular attention to the GELOSE contract itself, to the institution created by the law to manage the resources (the COBA), to the community rules created by this institution (the dina), and to two devices aimed at facilitating the negotiation and the securing of management contracts: environmental mediation and relative land tenure security.
496
4.1. Background and origin Community based natural resource management was the main mode of governance during the whole history of Madagascar, due to the absence or weakness of the state and to the autonomy of local polities that created their own customary systems. The colonial administration itself was quite decentralized until the publication of the 1930 forest decree and the strengthening of the forest administration that followed (Lavauden 1934). From 1930 to the publication of the GELOSE law in 1996, forest management policies did not change significantly from the centralization ideal of Lavauden and the governance of natural resources mostly depended on the political and economic situation (Chapter 2). The move from coercive state policies to a large scale implementation of community based natural resource management approaches started in the early 1990s. The first initiative, labeled Participatory Forest Management, 294 was implemented by the Swiss aid in the Menabe Region (Western Madagascar). The debate that led to the creation of the GELOSE law was initiated a few years later and matured during a series of workshops (Maldidier 2001), in Mantasoa 295 (September 1994), Mahajunga 296 (November 1994), Antsirabe 297 (May 1995) and Antananarivo (June 1996 298 and October 1998 299 ). Weber (1994), during a workshop organized in Mahajanga by ONE 300 and ANGAP, 301 proposed solutions to the illicit occupation of protected areas. Weber’s 294
In French: Gestion Participative des Forêts. This workshop was the conclusion of a study about customary rights and the local modes of governance, financed by the KEPEM project of USAID. 296 International workshop about human occupation in protected areas. 297 National workshop about community based natural resource management, aimed at formulating general recommendations for a decentralization policy. 298 International workshop about secure local management, which contributed to the design of the GELOSE legislation. 299 See the proceedings in Programme POLFOR and Equipe MIRAY (1998). 300 National Office for Environment. See Chapter 2. 301 National Agency for Protected Areas Management. See Chapter 2. 295
497
strategy consisted in “granting exclusive use rights to local communities living around the protected areas” (Weber 1994, 14). These use rights would be integrated into the negotiated management plans of buffer zones and would put an end to a de facto free access assumed to be the main cause of degradation. Granting management rights was, however, not sufficient. To make this work, local communities needed to find a material interest beyond the recognition of their authority to manage the resources. CBNRM approaches would thus go hand in hand with economic development and natural resource extraction (Weber 1994). By paying attention to these economic aspects, Weber (1994) anticipated the tradeoff between sustainable resource use and agricultural development objectives revealed by the case studies presented in Section 2. This new way to envision conservation was a significant departure from the still widespread received wisdom that population growth and resource scarcity were the main causes of degradation. Against these Malthusian perspectives, Weber defended the Boserupian argument (Chapter 5) that population increase is an incentive to agricultural intensification and better resource management. Human and social capital would be the solution rather than the problem, which justified the granting of management rights to local communities. Also influential on the approach was the French school of heritage management 302 (Barthod and Ollagnon 1991), which assumes that facilitating rational and free communication among stakeholders allows finding of a consensus that optimizes the satisfaction of each group. This approach is considered to be a departure from participatory methods and was intended to avoid the pitfalls analyzed by Rahnema (1992) (Section 1). We will see later that this notion of rational and transparent communication may indeed be quite idealistic, even naïve. 302
In French: “gestion patrimoniale.”
498
A consulting team was later created, with support from the French cooperation, to create a national policy aimed at decentralizing natural resource management (Montagne and Rakotondrainibe 2006). This led to the preparation of the GELOSE law and its application decree, which provided a legal framework for community based natural resource management.
4.2. The GELOSE legislation 4.2.1. The contract The GELOSE law (Gouvernement de Madagascar 1996) 303 was voted on and passed by the Malagasy National Assembly and promulgated on September 30, 1996. It defines the conditions and the procedures for the signature of management transfer contracts 304 that involve the community, the commune it belongs to and the decentralized state services it depends on. 305 Local communities are granted rights and responsibilities for the conservation or extraction and commercialization of natural resources in their territory (Gouvernement de Madagascar 1996 : article 42). An NGO or a project usually intervenes as a facilitator to negotiate the contract and prepare official documents, i.e. the contracts and its appendixes (community rules or dinas, management plans, a book of responsibilities, and sometimes resource maps). An environmental mediator is involved to favor a consensus between stakeholders (Section 4.2.4), and relative land tenure security is normally implemented to better guarantee resource use rights (Section 4.2.5). Contracts are valid for three years and can be renewed for ten more years after evaluation.
303
Secure Local Management. In French: Gestion Locale Sécurisée. The term Secure relates to the fact that the law implies a relative land tenure security, which I will explain later. 304 In French: Contrat de Transfert de Gestion. 305 This service can be a decentralized office of the ministry of environment, water and forest, of the ministry of agriculture and livestock husbandry, or of the ministry of fisheries, according to the type of resource.
499
Each management contract relates to one or several resources in a given territory, but not to the territory itself, although the legislation is sometimes ambiguous on this point. Marine or terrestrial resources, natural and plantation forests can be the object of contract. Specific decrees were issued or are planned to be so to define more precisely the modalities of transfer for each type of resource. The GCF decree (Gouvernement de Madagascar 2001) relates specifically to forest management. A second decree, which relates to pastoral resources, has been issued recently and a third decree is in preparation to treat the case of marine resources. Most contracts that have already been signed concern forest resources. In the case of GCF, contracts are signed between the communities and the Forest Services only. The environmental mediator is replaced by a facilitator provided by the NGO and there is no relative land tenure security. These differences result from the ambiguous status of the GCF decree, which, according to its first article, is an application decree of article 24 of the forest law (Gouvernement de Madagascar, 1997) rather than an application decree of the GELOSE. According to article 24 of the forest law, “the state can delegate forest management to other persons, public or private.” 306 The difference of meaning between management delegation and management transfer is unclear. The GELOSE law could indeed have been an application decree of this article 24 to the case of communities but was issued before the forest law. Two pathways that relate to two legal frameworks thus exist for implementing community based natural resource management in Madagascar: management delegation according to the forest law and management transfer according to the GELOSE law. The GCF decree used this ambiguity to take some liberty from the GELOSE law. On the
306
Translation by Pollini. Original text in French: “L’Etat peut déléguer la gestion de ses forêts a d’autres personnes publiques ou privées.” (Gouvernement de Madagascar 1997 : article 24).
500
ground, however, both GCF and GELOSE contracts are regarded as management transfers.
4.2.2. The COBA We saw that the GELOSE legislation was envisioned as a way to transfer management rights to local communities. I will show in this section that this is not what occurred eventually because the definition of the COBA matches with the concept of association, not of community. According to the Compact Oxford English Dictionary On Line, 307 a community is (1) a group of people living together in one place; (2) the people of an area or country considered collectively; (3) a group of people with a common religion, race, or profession: the scientific community. (4) the holding of certain attitudes and interests in common; (5) a group of interdependent plants or animals growing or living together or occupying a specified habitat. The concept of association presents some similarities with that of community. An associations is (1) a group of people organized for a joint purpose; (2) a connection or link between people or organizations; (3) a mental connection between ideas. The third definition of community is quite similar to the first definition of association but a distinction remains: people in a community are linked by their identity, whereas members of an association are linked by common objectives or purposes. In Madagascar, the fokonolona is the concept that matches better with that of community. Members of the same fokonolona share a common history and culture and are dependent on the same territorial resources, at least in the case of rural fokonolona. They usually come from the same ancestor and are buried in the same location. They 307
http://www.askoxford.com/?view=uk.
501
have social and economic relationships and owe reciprocal duties to each other (the fihavanana). They have a common leader or traditional chief and follow common rules they formalize by issuing dinas. The rights of all members concerning access to resources are not equal due to social stratification, but when they are not disrupted by new stakeholders coming from outside (migrants, state representatives, private operators, conservation and development actors), the inequalities remain moderated and are perceived as legitimate. Fokonolona are also in constant transformation. Due to migratory movements, several fokonolona from different ethnic groups can share the same territory until they fuse together. A fokonolona can also split when population increases or when internal conflicts provoke the departure of a group of people (Chapter 3). The term fokonolona is linked with the term fokontany, which designates the land occupied and utilized by the fokonolona. But for the state administration, the fokontany is an administrative unit (each commune is divided into fokontany). There can thus be several fokonolona in the traditional sense, usually living in different hamlets, inside one single fokontany in the administrative sense, despite the fact that in an administrative sense, all of the inhabitants of the fokontany constitute a single fokonolona. The status of the fokonolona has changed several times during the history of Madagascar. Under the first republic, the smallest decentralized territorial collectivity was the commune. The fokonolona was not legally recognized as a legal body and was not granted legal rights. But the state defined duties that had to be ensured at the fokonolona level, in order to utilize this customary organization for giving responsibilities to the population (Aubert 1999). During the second republic, the fokonolona was became a legal body (Aubert 1999). The general assembly of the fokonolona was in charge of managing its territory (the fokontany) and enforcing
502
regulations. According to Aubert (1999), this allowed legalization of legitimate ways of ruling that were already implemented by traditional authorities. The situation changed again in the early 1990s, after President Zafy was elected. The commune became again the basic administrative structure and the fokonolona lost its autonomy (Aubert 1999). This situation still prevails today, the fokontany being administrated by fokontany Presidents appointed by the mayor (Chapter 3). The constitution of 1997, however, asserted that the fokonolona could take measures aimed at improving resource management. But it remained unclear about what form these measures can take and what legal value they have. To be consistent with the initial objectives proposed by Weber (1994), the GELOSE should have been designed in a way to grant legal natural resource management rights to the fokonolona. But in the political context of the mid-1990s, this proved to be impossible. The fokonolona had just lost its autonomy and the members of the National Assembly opposed the idea that management rights would be transferred explicitly to the fokonolona (Aubert 1999). This led the GELOSE legislation to create a new concept, the basic community (communauté de base), abbreviated by the term COBA, whose legal statute would be defined by an application decree. According to article 2 of the GELOSE law, the COBA “is composed of any voluntary group of individuals united with the same interests and obeying the rules of a common way of life. It groups, according to the cases, the 308 inhabitants of a hamlet, a village or a group of villages” 309 (Gouvernement de Madagascar 1996: article 2). This article reveals that the initial ideal of genuine community management was still 308
I highlighted. Translation by Pollini. Original text in French: “La communauté de base est constituée par tout groupement volontaire d’individus unis par les mêmes intérêts et obéissant a des règles de vie commune. Elle regroupe selon les cas, les habitants d’un hameau, d’un village ou d’un groupe de villages.”
309
503
present when experts designed the GELOSE legislation. Unfortunately, in the end, it was not the inhabitants who would constitute the COBA but some inhabitants, which completely changed its nature. As the GELOSE law mentions in later articles, “any inhabitant residing within the limits of the territory of the basic community can be accepted as a member [of the COBA].” “The application to become a member is submitted to the general assembly, which deliberates within the conditions fixed by the statute” 310 (Gouvernement de Madagascar 1996: article 5). The possibility to belong to the basic “community” (the COBA) is hence defined by an administrative act and by individual choices, rather than by history, kinship or geography. Rather than a community of people who share a common territory and identity, the COBA was an association of people having a common objective and making the choice to group together. A small fraction of the community can create a COBA and the legitimacy of this group of people to defend community interests is not a formal criterion for this creation. Theoretically, any group of people with common objectives can constitute a COBA, decide who will be in and out and elaborate the rules that best matches its interests. Clearly, the new institution so created is an association, not a community. This opens the way to inequities and division at the community level if the COBA gains significant power. It is not excluded, however, that communities use the COBA statute to modernize themselves, but several factors will impede this from occurring, as I will show. The decree about the COBA statute (Gouvernement de Madagascar 2000b) confirms the fact that the COBA is an association. It states that the statute have to indicate the purpose of the COBA, the territory it manages, its management organs, its functioning, the list of its members and have to be deposited at the commune the 310
Translation by Pollini. Original text in French: “Peut être accepté comme membre tout habitant résident dans les limites du terroir de la communauté de base.” “La candidature pour devenir membre est soumise à l’assemblée générale, qui délibère dans les conditions fixées par le statut.”
504
community depends on. The commune issues a certificate that is the legal validation of the COBA’s creation. The decree also states that “all the decisions taken during meetings have to be written and classified in a book. The President or one of the members of the management structure is in charge of all writings” (Gouvernement de Madagascar 2000b, articles 25 and 26). 311 In these conditions the rural elite, which usually has better literacy than traditional leaders, has a significant advantage to take control of the COBA. This elite can use the COBA to channel to itself most of the profit that can be generated from natural resource extraction, in the same manner as the Kolo Harena associations channeled to themselves most of the support and funding granted by development projects (Chapter 6). The case of Tsinjoarivo (Section 2) where Merina people with better formal education took control over access to land after they took control of forest management associations, illustrates this risk. The associations created by the PDFIV project were not COBAs but one can wonder why things would be different if they were. Urban people having interest in rural areas, such as forest operators, can also learn very fast how to use the GELOSE to their own advantage and to manipulate local communities and appropriate their resources. This could have happened in Beforona as we saw in Section 2. The operator failed in this case, but according to the syndicate of forest exploiters, Beforona would not be the only case in which forest collectors have attempted to use the COBA statute to take control over resources. GCF contracts, which do not imply environmental mediation, may pave the way to the worst abuses. Andriambololona et al. (2004) give the example of a GCF contract involving 7% of the population (35 persons), but allowing the exploitation of a 677 hectare forest in Ankodondona and Mahafinaritra
311
Translation by Pollini. Original text in French: “Toutes les décisions prises lors des réunions doivent être rédigées par écrit et classées dans un livre réserve a cet effet.” “Le Président ou l’un des membres de la structure de gestion se charge de toutes les rédactions écrites.”
505
(Anjojorobe district). One family controls the COBA and the rest of the community is excluded from resource extraction (Andriambololona et al. 2004). The GELOSE law designed two devices to reduce this risk. First, the commune, in partnership with state decentralized services, must implement an enquiry before validating the COBA creation. This enquiry must check the social adhesion of the demand, verify the legality of member designation in the COBA’s managing structure, and assess the natural resources and the COBA’s capacity to manage them (Gouvernement de Madagascar 1996, article 12). But these assessments can easily be subjective. No objective criteria, such as the relations of COBA leaders with traditional authorities or their mode of designation, were defined to guarantee the rigor of this survey. Local political leaders often share common interests with the opportunistic local elite, resulting in a significant risk of subjective assessments. The second device is the environmental mediator, about whom more will be said later (Section 4.2.4.). In order to avoid misunderstanding about my arguments, I need to indicate, again, that I do not reject the concept of COBA in itself, nor of course of community based natural resource management. A COBA according to its statutes is a sort of association and associations are useful to improve the social fabric and strengthen the civil society. But problems arise when their roles are confused with those of communities. If associations are granted legal rights that were hitherto legitimately held by communities, the result can be an abusive appropriation of resources which can lead to conflicts between COBA members and nonmembers (between the association and the rest of the community). Furthermore, the confusion created between communities and private associations can result in unfair regulations. With respect to fiscal aspects for instance, collectivities usually raise taxes whereas private operators that extract resources pay these taxes. In the case of COBAs, it is not always
506
clear whether it should raise taxes (like any territorial collectivity) or pay taxes (like any private operators).
4.2.3. The DINA Besides the mismatch between the concepts of COBA and local community, the dinas provide another example of the confusion that can be provoked by the GELOSE. The dina is a community rule designed at fokonolona level but requiring approval by the commune to have a legal value. It existed prior to the GELOSE, but was incorporated into it. According to the GELOSE law, “the relationships between COBA members are regulated through the creation of dinas, [which are] approved by the COBA members following the customary rules of the community” (Gouvernement de Madagascar 1996: article 49). 312 Dinas are further validated by the commune (Gouvernement de Madagascar1996: article 51), which gives them the status of a law (Gouvernement de Madagascar 1996: article 52). The consequence of this innovation is that two entities (the fokonolona and the COBA) are now able to create legal rules at the community level. The first one can do that in the name of its historical legitimacy while the second one receives its legitimacy from the GELOSE legislation. It is then questionable whether a dina created by the COBA applies to the whole fokonolona or only to the COBA members. Important problems arise whatever answer is given. If the dina applies to the whole fokonolona, it will lack legitimacy because the institution that established it (the COBA) is not necessarily controlled by the legitimate authorities at fokonolona level. If it applies to the COBA members only, the members of the fokonolona will have the possibility to chose between being submitted
312
Translation by Pollini. Original text in French: “Les rapports entre les membres de la communauté de base sont réglés par voie de “Dina.” Les “Dina” sont approuvés par les membres de la communauté de base selon les règles coutumières régissant la communauté.”
507
to it or not (by being member or not of the COBA). Such a dina could hardly be called a law. It would just be an internal rule as there are in any association.
4.2.4. Environmental mediation The designers of the GELOSE legislation anticipated the pitfalls described in the previous sections and incorporated an actor aimed at mitigating them: the environmental mediator. A specific decree (Gouvernement de Madagascar 2000a) was issued to define the role and responsibilities of environmental mediators and the procedure for their training. According to this decree, the goal of environmental mediation is to facilitate discussion and negotiation between the various partners involved in local resource management, by contributing, through the establishment of an information flow between parties, to bring points of views closer to each other. [This will] facilitate the emergence of a common vision and a common strategy of long-term resource management and the consensual definition of procedures that allow effective management. (Gouvernement de Madagascar 2000a: article 2) 313 Mediation concerns the relation between community members as well as between the community, the state, the commune and the other actors involved. In addition, the mediator is involved in practical and strategic tasks such as the establishment of rules for access to resources, the determination of modalities concerning natural resource extraction and benefit sharing, the determination of how the income will be used and the determination of fees and penalties in case of fraud (Gouvernement de Madagascar 1996: article 17). He/she is recruited among people trained specifically for this task and who have received an agreement, delivered by a committee that includes members of seven different ministries. The intervention of 313
Translation by Pollini. Original text in French: La médiation environnementale a pour but de faciliter les discussions et les négociations entre partenaires impliques dans la gestion locale de ces ressources, en contribuant, par l’établissement d’un courant d’information entre les parties, a rapprocher les points de vue et objectifs en présence et a faciliter ainsi l’émergence d’une vision commune et d’une stratégie commune de la gestion a long terme de ces ressources et la définition consensuelle des procédures permettant leur gestion effective.”
508
mediators is mandatory in the case of the first management transfer demands in a given commune and when two or more communes or communities are concerned with the same resources. It is facultative in other cases. The mediator is required to be neutral (Gouvernement de Madagascar 2000a: article 31). The training he receives is intended to prepare him for this neutrality. However, mediators are in most cases urban people foreign to the regions where they work and are financially and technically supported by the NGO or project that implements the GELOSE process. This support consists of the sensitization of stakeholders, the organization of meetings, the training of community leaders, the implementation of socio-economic studies and diagnosis, the purchasing of technical tools and equipments, the assessment and mapping of resources, the design of “modern” or “scientific” management plans, the search for market opportunities and the communication of results. It can hardly be imagined how mediation work could remain independent of these multiple tasks and of the actor’s representations associated to them. Aubert’s (1999) analysis of environmental mediation reveals the ambiguity of the mediator’s role. According to this author, the implementation of the GELOSE could imply a “social restructuring" (Aubert 1999, no pagination), 314 and one of the mediator’s role would be to legitimate this new social organization. This reveals the existence of a social engineering project that implies much more that just favoring transparent communication and the emergence of a common vision. In practice, the work of the mediator may serve to guide communities to “participate” in this social engineering project, the architect of the whole process being the NGO or project that supports the GELOSE. In which case, the mediator would mostly serve to support social organization, as does any development agent specialized in this kind of task. This confusion of roles may explain why the GCF decree eliminated the 314
No pagination because Html file.
509
necessity to utilize environmental mediators (as we will see later), allowing his duties to be assumed by NGO’s staff.
4.2.4. Relative land tenure security (RLTS) Relative land tenure security is another key element of the GELOSE legislation. It consists of delineating the boundaries of the territory put under COBA’s authority; in defining zones inside this territory according to the resources which are encountered and according to the rules applied to these resources; and in some cases (not mandatory), in the establishment of a cadastral map aimed at solving land tenure conflicts locally (Gouvernement de Madagascar 1998). RLTS does not lead to land title granting (which explains the term “relative”) but can serve as a base for title granting in the future. In most management transfers, RLTS was not implemented because it is costly, can slow the overall process by revealing land tenure conflicts and is not perceived as being essential by the projects or NGOs that support the implementation of the GELOSE. Management transfers typically involve a single resource (but can be extended to several resources), which poses a practical problem for the implementation of RLTS. When several resources are encountered on the same land, it would not be legitimate to secure land tenure in favor of an association of people extracting one of these resources. The zoning that accompanied management transfers for charcoal production in Ambondromamy (Section 2) can help in anticipating the problems that could arise. In Ambodromamy, Zyziphus savannahs were zoned in order to optimize charcoal production by COBA members. Five zones were defined and charcoal producers have to follow a five year rotation in these five zones. When more pressure occurs on the land, due to demographic increase, one may wonder what will be the legitimacy of this zoning for farmers who do not produce charcoal, are not COBA
510
members but need to convert Zyzyphus savannahs into agricultural land to sustain their livelihood. Muttenzer (2006) investigated the functioning of charcoal COBAs in other sites of the region. He showed that trees cut to make charcoal are mostly cut in areas dedicated to be later converted into agricultural land. Five year rotations would not make sense in this case: once the trees are transformed into charcoal, the land is cultivated and no tree is allowed to grow again. From the farmer’s perspective, this logic is rational because agricultural land is more profitable than charcoal production, the later being just a transition activity for recent migrants (Muttenzer 2006). If the management plans and the five year rotations were successfully implemented, they would compete with the customary modes of regulation which support this multiple use logic. The land use would become fixed according to a pattern determined by the harvesting of a minor resource (Ziziphus) to the benefit of a subset of the population (the COBA members). This would reduce the overall profit that can be obtained from the land and concentrate these profits in the hands of a minority. Moreover, the logic that leads to favoring a particular mode of resource utilization (charcoal production) is strongly influenced by outside actors. Charcoal producers have more chance to be granted management rights because sustainable charcoal production allows maintaining trees in place, which matches with the expectations of donors that fund development or conservation programs. On the other hand, farmers who need to clear land to extend their fields are disadvantaged, despite the fact that their activity may produce more value per unit of land, allowing a higher population density and hence lower pressure on remaining primary ecosystems. This scenario will not necessarily occur as communities will certainly find their own mode of regulation when they will be confronted to these problems. They will probably initiate negotiations between the COBA and the rest of the fokonolona. According to Muttenzer (2006), this is indeed the situation that prevails.
511
In sum, the GELOSE legislation and the management plan, zoning, and tenure security associated with it may favor the introduction of exogenous development schemes that do not address the complexity of the reality, to the detriment of community regulations based on more comprehensive, empirical and contextualized knowledge designed to manage multifunctional landscapes. This recalls the high modernist project presented in Chapter 8 where the realities are reshaped according to simplified management schemes. The GELOSE and its zoning and management plans may favor a “social taxidermy” (Scott 1998) aimed at rendering communities more legible
for
administrative
and
control
purpose.
In
this
perspective,
the
nonimplementation of RLTS in GELOSE contracts may then be a chance because RLTS would render the zoning less easily reversible, would strengthen the control over access to land by resource users whose activities are more compatible with shortsighted conservation agendas, and could impede future land use changes that would be profitable to the overall community.
4.3. Conclusion This analysis showed that the GELOSE legislation can easily be used by a group of stakeholders for taking control over access to natural resources. This situation results from the confusion created between the concepts of association (which the COBA is) and of community (which it is not). The consequence can be a marginalization of traditional and legitimate modes of control over resources in favor of new modes implemented by legal associations whose objectives match better with the international conservation agenda. Paradoxically, instead of transferring management rights to communities, the implementation of the GELOSE may actually lead to transfer these rights from a community that was the de facto manager of the resources, due to the absence or the weakness of the state, to a restricted group of
512
individuals which is willing to adopt management rules designed by the state and its partners. The interests of the five parties involved (the state, the commune, the COBA, and the NGOs and the donors that support the process) can easily match together. The management plans and zoning provide the state and the commune with an increased legibility of the population and of the resource uses, which facilitates the implementation of “modernization.” The NGOs’ and donors’ objectives in terms of improved natural resource management are better achieved, at least in the short-term. The rural elite, which is better prepared to manage modern associations such as COBAs, increases its control over access to land and its local prestige. But it is doubtful whether these interests match with those of the community (the fokonolona). At the community level, resources are managed according to a much larger array of activities which management transfer contracts fail to account for.
5. Pushing the GELOSE in Malagasy environmental policies In this section, I analyze the institutional dynamics that surround the implementation of the GELOSE legislation eight years after its vote by the National Assembly. I use as empirical material a comprehensive evaluation of management transfers carried out in 2004, the RESOLVE 315 evaluation (RESOLVE-PCP-IRD 2004b; RESOLVE-PCP-IRD 2005). This evaluation consisted of four phases. The first phase (RESOLVE-PCP-IRD 2004a) included a comprehensive bibliographic review, an inventory of existing contracts, the creation of a database about these contracts, the elaboration of the evaluation methodology and the test of this methodology on eight sites. The second phase (RESOLVE-PCP-IRD 2004b) consisted of evaluating 26 supplementary sites. 315
RESOLVE is the name of the consulting firm that implemented the work.
513
The third phase concerns the set up of the monitoring system, a matter which I will not address here. The fourth phase (RESOLVE-PCP-IRD 2005) is a synthesis of the whole work. Two teams were involved in the evaluation: the PCP 316 team and the IRD 317 team. They worked separately and their results are presented in different sections of the phase II report (RESOLVE-PCP-IRD 2004b). I will concentrate my analysis of this second phase report, which provides the most complete presentation of the evaluation’s outcomes. The results of the two teams will be presented in two different sub sections in order to show the difference in institutional logic that accompanied their work.
5.1. The PCP evaluation 5.1.1. Methodology The PCP team consisted of a group of researchers from the CIRAD, 318 the FOFIFA 319 and the ESSA. 320 It employed a semi qualitative rapid assessment method and used a survey guide passed to stakeholders by small teams of two or three researchers. It surveyed eight sites during the first phase (test of the methodology) plus 17 sites during the second phase. It focused on governance aspects and on the environmental and economic impacts of management transfers (RESOLVE-PCP-IRD 2004b). Five survey guidelines were elaborated which targeted COBA members, representatives of state services, representatives of territorial collectivities, economic 316
A research consortium comprised of CIRAD, FOFIFA and the University of Antananarivo. In French : Pole de Compétence en Partenariat. 317 Development Research Institute. In French : Institut de Recherche pour le Développement. This institute was formerly named ORSTOM. 318 French International Center for Agronomic and Development Research. In French: Centre International de Recherche Agronomique pour le Développement. 319 The Malagasy national institute for agronomic research and rural development. 320 The main Agronomy School in Madagascar. In French : Ecole Supérieure des Sciences Agronomiques.
514
operators, and representatives of NGOs or projects which provided technical support (RESOLVE-PCP-IRD
2004a).
Interviews
of
legitimate
community
leaders
(fokonolona representatives) was not planned but could have occurred in cases where these leaders were COBA members as well. Interviews of non-COBA members was not planned either. This fact was pointed out in the conclusions of phase 1, and the phase 2 report (RESOLVE-PCP-IRD 2004b, 13 and 48) mentions that non-COBA members should be interviewed. However, the lists of actors actually encountered by the PCP team does not mention non-COBA members or fokonolona representatives (RESOLVE-PCP-IRD 2004b). The time spent on each site was only one or two days on average. Given the sampling strategy and this limited time spent on the ground, it is doubtful whether this methodology allowed assessment of social dynamics at the community (fokonolona) level.
5.1.2. Results I will present the outcomes of the PCP team according to four topics: (1) the integration of management transfers into local, regional and national policies; (2) the environmental impacts; (3) the social impacts; and (4) the economic impacts.
5.1.2.1. The integration of management transfers into local, regional and national policies The authors state that management transfers are far from being integrated into state supported rural development policies at the communal level: The Communal Development Plans insist on basic population needs (health, roads and education) but fails to consider environmental risks as major risks. They also forget that this sector can certainly be a source of
515
income for the communes by taxing the flow of ligneous and nonligneous products. 321 (RESOLVE-PCP-IRD 2004b, 8) The mayors are rarely aware of their role and duties concerning the GELOSE procedures. Hence there cannot be any communal strategy able to structure and plan management transfers 322 (RESOLVE-PCP-IRD 2004b, 8) and it appears difficult at this stage to envision a vertical integration of management transfer into rural development policies at a larger scale.323 (RESOLVE-PCP-IRD 2004b, 8) These results show that the management transfer is like a graft that does not take at the communal level. But instead of questioning the reasons and meaning of this lack of integration (could it be a rejection?), the authors insist on the “good” reasons (environmental risks, tax collection) to consider management transfers (MTs) 324 as a priority. It could, however, be questioned whether the significance of the revenue that communes could gain from MTs,
the difficulties of collecting such taxes, in
comparison with taxes raised on trade, agriculture, land and natural resource extraction in general, 325 would justify this priority. Given the limited resources of communes in Madagascar, the many challenges they face for providing basic infrastructure and services, 326 it is still questionable whether MTs are indeed a priority for them. It is 321
Translation by Pollini. Original text in French: “pêchent de ne pas considérer les risques environnementaux comme des risques majeurs. Ils oublient également que ce secteur peut être certainement une source de revenus des communes au travers de prélèvements sur les flux de produits ligneux et non ligneux.” 322 Translation by Pollini. Original text in French: “le maire n’ayant que rarement conscience de son rôle et de son devoir en particulier pour les procédures GELOSE. Il ne peut y avoir dès lors de stratégies communales à même de structurer et de planifier les transferts de gestion et il paraît difficile à ce stade d'imaginer des intégrations verticales du transfert de gestion dans des politiques de développement à plus vaste échelle.” 323 Translation by Pollini. Original text in French: “il paraît difficile à ce stade d'imaginer des intégrations verticales du transfert de gestion dans des politiques de développement à plus vaste échelle.” 324 For more convenience, I employ the abbreviation MT for Management Transfer. 325 Management transfer is just one modality among others for natural resources extraction. 326 According to Republikan'I Madagasikara (2003), 33% of the Malagasy communes have no access to a national road, 30% have no access to a provincial road and the average time required to access a bus station is 10 hours. Security condition is a second important issue. It is perceived as bad or very bad in 39% of Malagasy communes.
516
hence not the importance of management transfers that should be questioned, but their relative importance and the synergies and trade-off between sectors in a context of limited funding and low institutional capacity. The main objective of the evaluation may have been, even if this was not explicit, to push MTs in the local development agenda, rather than to question their relative importance from a local standpoint. The authors further propose to use the local zoning component of MTs as a starting point to establish communal and regional zoning. But communities and communes may have other forms of knowledge which allow them to identify infrastructure and development priorities. Zoning and mapping, which would certainly be done using external expertise, would not necessarily identify these needs any better and could be as costly as creating some basic infrastructures. Local actors may be aware of this, which may explain their lack of interest in mapping and formal management tools. If the institutions which implement management transfers contend that zoning and mapping are a priority, it may mean that it is a priority for them, mostly because they need this to be done in order to be delivered data that allows them to monitor management transfers better. Consistent with this, the authors propose to integrate zoning in a national monitoring system of management transfers and to use the number of communal zonings as an indicator of the GELOSE implementation. When basic transportation, health, and education infrastructures are absent or nonfunctional, mapping and zoning centered on resource management issues may be luxury devices.
5.1.2.2. Environmental impacts Concerning this aspect, the PCP analysis concludes that setting aside management transfers that have failed and transfers that are not yet functional, we can assert that natural resource management transfers contribute in a significant way to improving the environment in
517
the sites that have been transferred: reduction of bush fires, effective protection of the environment against excessive pressure on the surroundings, organization of product collection that can be supported by the environment. 327 (RESOLVE-PCP-IRD 2004b, 8-9) The reason for this apparent success may be that MTs provide a framework that encourages all members of the community to unite against intruders, and that it gives opportunities to the state and other organizations to provide support for this resistance (RESOLVE-PCP-IRD 2004b). But this conclusion is based on the observation of a correlation (fewer fires after the implementation of MTs) that can be spurious: it does not necessarily imply a relation of causality. The MT evaluation was implemented in 2004, a short time after the repressive antifire campaigns initiated in 2002 (Chapter 6). It is not possible to distinguish the effects of this campaign from the effects of MTs because sites with no MT were not studied. Furthermore, in a political context where fires in all forms are seen as an evil, coercive logic can be present in MTs as well, as I showed in the short case studies (Section 2). The frequent passage of foreigners, NGO staff and state representatives is often a source of fear, including when their presence is not justified by control. If there would be a causal relationship between management transfers and a better control of resource degradation, it could actually be the consequence of a coercive logic channeled by MTs (intentionally or unintentionally). The report does not provide arguments that would allow rejection of this hypothesis. The next section, about social and economic impacts, will provide insights that corroborate this hypothesis and help to envision the long term impacts of MTs on the environment.
327
Translation by Pollini. Original text in French: “Mis à part les transferts de gestion qui ont échoué et les transferts qui ne sont pas encore fonctionnels, on peut retenir que le TGRNR participe de façon significative à l'amélioration de l'environnement sur les site transférés: réduction des feux de brousse, protection effective contre des pressions excessives sur le milieu, organisation des prélèvements jugés supportables par le milieu.”
518
Another observation showing that we cannot be conclusive yet about the positive environmental impacts of community management is that in some cases, the pressures on ecosystems were displaced to other areas by management transfers, especially when economic alternatives were not provided (RESOLVE-PCP-IRD 2004b). When these outside resources are depleted, the pressures may come back to the transferred area. Such “leakages” may be more frequent than observed by the RESOLVE team, as the evaluators just spent a few days on the ground and did not have time to tour the whole territory (RESOLVE-PCP-IRD 2004b, 30). The report also says very little about the motivation of communities to improve resource management. When COBAs successfully protect forest resources from clearing by outsiders, their motivation could be to conserve a reservoir of land for future agriculture expansion. If this was the case, management transfers would reduce the pace of deforestation but would not stop it. Concerning the monitoring of the environmental impact of MTs, the authors insist on the necessity to produce baseline data, to establish a detailed diagnosis of the situation before transfer, to define quantified objectives and indicators in relation to them and to produce periodical reports (RESOLVE-PCP-IRD 2004b). This reveals a temptation for a bureaucratization and formalization of the process which may be unrealistic in the context of the low education and literacy of most communities. It could also favor the empowerment of a rural elite with higher literacy, as we saw in Section 3. These propositions match better, again, with the logic of Scott’s (1998) high modernism than with the ideal of local empowerment and community based management advocated by the first advocates of the GELOSE approach. In short, the assumption that MTs have had a positive impact on the environment is based on very little evidence. The “impact” observed on the ground cannot be contested but can be explained by other factors (coercive policies), can hide
519
a displacement of the problem and may be short-term if no alternatives are provided to compensate the restriction on access to land and resources. Rather than questioning these issues, the authors advocate for a strong monitoring of management transfers and propose tools that could in the end be used to control local communities, rather than favoring their empowerment and giving them more rights and responsibilities.
5.1.2.3. Social impacts 5.1.2.3.1. Observed impacts The main issue addressed by the PCP report with respect to social aspects is the impact of MTs on social cohesion and the situation in terms of conflicts. The authors explain that conflicts can occur inside the community (in the sense of the fokonolona), sometimes inside the COBA (especially in cases where land tenure is not clear), and between the COBA, the commune and the state services involved in the transfer (RESOLVE-PCP-IRD 2004b). The environmental mediator is expected to reduce these conflicts. Consistent with the observations I made in the Section 3, the evaluators observed that on the ground, the role of the mediator was often confused with social organization and capacity building, which are tasks under the responsibility of the NGO or project that provides support. They contend that this confusion results from the fact that MTs are regarded as a way to transfer administrative tasks and duties from state services to COBAs, which minimizes their political dimension (RESOLVEPCP-IRD 2004b). This argument is consistent with the analysis of Rahnema (Section 1), who asserted that participatory approaches can be used as a way to discharge states services from costly responsibilities they are unable to assume. In the context of weak states which prevails in Madagascar, it is logical that the mediator gets diverted from mediation work and redirected to these administrative tasks.
520
In spite of this issue, the PCP report depicts a situation where social conflicts are not very significant. They exist in some cases, 328 especially between members and nonmembers, but communities find ways to solve them. For example, nonmembers create alternative associations to challenge COBAs. The limited extent of conflicts can also be explained by the fact that many COBAs are nonfunctional. According to the authors, the reasons for this inefficiency can be a lack of interest, the refusal of new rules or the lack of administrative capacity (RESOLVE-PCP-IRD 2004b). In sum, conflicts do not seem to be prominent. But as we saw earlier, leaders of the fokonolona were not targeted by the interviewers and the time spent on each site was quite short. The interviewees were mostly actors directly involved for whom pointing out conflicts would not be the right strategy for guaranteeing future funding and support. The overall picture can thus be biased. We will see in the next section that the conclusions of the IRD consultants, who spent more time on the ground (but in fewer sites), are quite different. The fact that social conflicts would be limited is, however, consistent with the conclusions of Muttenzer (2006), who contends that local communities re-appropriate MTs in a way consistent with their interests. But even if this dynamic was the rule, it can still be asked what would be the impacts of MTs in the case of more efficient implementation. Understanding these potential impacts is equal to understanding the real intention of the GELOSE, which may exist in an implicit form hidden behind explicit statements.
328
For example in Sadjaovato (RESOLVE-PCP-IRD 2004b, 49) and in Marosely (RESOLVE-PCPIRD 2004b, 50).
521
5.1.2.3.2. Potential impacts We saw in Section 3 that the confusion between the concepts of association and community, the question of whether or not the COBA represents the community, the skills required for managing a COBA and the way the COBA articulates with traditional leaders are central issues that have to be addressed. They reveal the implicit intentions of the GELOSE and determine the potential impact it could have if COBAs became more powerful. I must now see how the PCP team addressed these issues. According to the PCP report, the question of the representativeness of the COBA must not be a dogma. We cannot ask the COBA to represent all the fokonolona, as we expect from elected institutions that they represent all citizens. Instead, it must be representative of the different categories of natural resources users. 329 (RESOLVE-PCP-IRD 2004b, 9) I would agree with this remark if the COBAs were created with the intent to grant resource use rights to economic operators grouped into associations. But if the objective was to implement community based natural resource management, then the COBA should be the fokonolona as we saw in Section 4, because the Malagasy entity that matches better with the concept of community is indeed the fokonolona. Under the authority of the fokonolona, extraction rights could be granted to groups of users organized into associations but these two aspects (community management and private extraction) should not be confused. There can be associations of charcoal makers, loggers, medicinal plant collectors, honey collectors, herders, fishers or hunters, but these groups are not communities. Several types of institutions and actors such as communities, associations and entrepreneurs are necessary to constitute a functional social fabric, but have to play different roles that must not be
329
Translation by Pollini. Original text in French: “question de la représentativité des COBA ne doit pas être un dogme. On ne peut pas demander à la COBA de représenter tout le fokonolona, comme on attend des institutions élues qu’elles représentent les citoyens. Par contre, elle doit être représentative des différentes catégories d’usagers des RN.
522
confused. Natural resources can be managed by small private groups (associations) and this can even be preferable to community management, but the significance would be different. For example, if only a few farmers are interested in harvesting trees in a forest, there is no reason to grant extraction rights to the whole community. This small group of local loggers, organized into an enterprise (or a “lucrative association” having a COBA statute), could receive a permit or any sort of convention that would define its rights and responsibilities on behalf of the state. Beside this, the community could be involved in the creation of new rules which would be locally designed, could raise local taxes paid by the logger association and could implement control. The COBA (the group of loggers) could then render accounts to the community instead of the state. But in order to achieve this, the community needs to be recognized as a legal body able to establish contracts with the group of loggers. The GELOSE legislation does not allow this to occur because the COBA is not the fokonolona, by definition. Trapped by this contradiction, the PCP team ended in defending similar arguments. It asserted that “the legitimacy, and hence the recognition of the COBA by the fokonolona, should be translated into explicit obligations, in the contract and the book of responsibilities, of the COBA toward the fokonolona” 330 (RESOLVE-PCPIRD 2004b, 9). By saying so it recognized that the COBA is not the fokonolona. However, these conclusions did not solve the problem because the fokonolona is not a legal body which can be legally accountable. The COBA was indeed the institution aimed at giving the fokonolona this legal recognition but failed to achieve this. We thus end in a complex situation where a virtual and externally designed community (the COBA) should theoretically render accounts to the historical but still real community (the fokonolona) whereas in practice the COBA, due to its recognition by 330
Translation by Pollini. Original text in French: ” La légitimité, donc la reconnaissance de la COBA par le fokonolona, devrait se traduire par des obligations explicites du contrat et du cahier des charges, de la COBA envers le fokonolona.”
523
the state as a legal body, often considers itself to be the community, which reverses the relationship just evoked. As the COBA renders account to the commune, to the decentralized state services and to the NGO that provides technical and financial support, it could marginalize the role of the fokonolona and substitute for it if it became sufficiently powerful. The real intention of the GELOSE legislation, expressed by the reluctance of the members of the National Assembly to give legal status to the fokonolona, may indeed have been to disempower the fokonolona while empowering a rural elite which incarnates better the canon of “modernity.” But all this just does not work. Local communities could just reshape the system in their own way and continue their activities as usual, the COBA being just a supplementary postcolonial creation in a complex social network determined by production and power relations and by the diversity of resource use strategies. According to Muttenzer (2006), this would be the situation that prevails. The GELOSE legislation would then be mostly hot air blowing across the Malagasy countryside.
5.1.2.4. Economic impacts As I mentioned earlier, no precise data was collected concerning the profitability of MTs. Some outcomes are nevertheless provided by the PCP report. First, it appears that MTs lead to increased profits when a valuable resource is harvested (wood to make charcoal, timber trees, or fish) (RESOLVE-PCP-IRD 2004b). However, it can be questioned whether the GELOSE was necessary to achieve these results, as they could just be the consequence of the regulation and legalization of illegal activities (see the case of Ambodromamy in Section 3). Second, MTs dedicated to conservation, where only noncommercial use rights are authorized, are profitable if they help to control illegal immigration (RESOLVE-PCP-IRD 2004b).
524
Land is itself an asset and its protection from appropriation by outsiders can be regarded as a profit. But from the perspective of farmers, this profitability may mostly rely on unsustainable exploitation of resources followed by the conversion of ecosystems into agricultural land.
5.2. The IRD evaluation In this sub section, I will present the evaluation conducted by the IRD team. I will start by presenting the methodology, followed by the results, before analyzing the institutional forces that drove the evaluation, as they are revealed by the recommendations provided by the evaluators.
5.2.1. Methodology The IRD team involved researchers from the IRD and students and professors from the University of Antananarivo. The students spent several weeks or months at their sites and were visited by their professors and by IRD researchers. The IRD investigated six sites in depth. It had already worked in these sites for about ten years, which provided it with knowledge of the situation before MT implementation. The team further investigated three sites more superficially. The method was derived from the participant observation methods of anthropologists, in order to achieve a more in-depth analysis of social organization at the community scale. The evaluators conducted a cartography of community land, a history of village settlement and an in-depth analysis of social organization (traditional authorities, existence of associations, and other forms of power). MT contracts were analyzed in detail, with a special emphasis on COBA’s composition, and management structure (RESOLVE-PCP-IRD 2004b). We will see that the conclusions are quite
525
different from those of the PCP, which can be explained by the different methodological choices.
5.2.2. Results The main conclusions, according to the executive summary, are that the type of resource, the social organization and the history of the community are the main determinants of the type of problems encountered. Ethnicity and the differences between regions would be less determining (RESOLVE-PCP-IRD 2004b). The consequences are that when a MT is implemented without a prior analysis of social organization and structures, it is from the beginning bound to fail, but even more grave still, has every chance of destroying the traditional natural resource management system that existed before its implementation by provoking, inside the community, conflicts, disruptions, power reversals and structural modifications, often irreversible, that sometimes result in the exacerbation of behavior and uses that are totally in opposition with management transfer objectives. 331 (RESOLVE-PCP-IRD 2004b, 108) This conclusion is consistent with the short case studies presented in Section 3 and summarizes most concerns brought up during the analysis of the legislation and of the PCP’s outcomes (Sections 4 and 5.1). The IRD team further provides guidelines about how to describe the forms of powers and proposes a list of criteria to consider when establishing a COBA. It asserts that the main feature of traditional power structures, valid for all ethnic groups in Madagascar, is that legitimate powers typically belong to the first settlers in a given location. When migrants are not many, they do not challenge this power and prefer to 331
Translation by Pollini. Original text in French: “est dès le départ voué à l’échec, mais bien plus grave encore, il a toutes les chances de mettre à bas le système de gestion “ traditionnel ” des ressources naturelles préexistant au contrat de transfert en occasionnant, au sein de la communauté, des conflits, des fractures, des renversements de pouvoirs et des modifications structurelles souvent irréversibles qui parfois entraînent l’exacerbation d’usages et de comportements en opposition totale avec les objectifs d’un transfert de gestion”.
526
establish connections through matrimonial alliances, which provides them with legitimate access to land and a certain political influence, especially if they marry the daughters of traditional chiefs. Things are different when migrants are more numerous, develop their own economic activities and social organizations, and designate their own leaders. If they develop profitable activities and utilize resources that were not paid attention by the first settlers, their economic power and political influence can increase and become dominant (RESOLVE-PCP-IRD 2004b). This occurs for example when migrants improve bottom lands while historical settlers practice more extensive land uses such as livestock husbandry or slash-and-burn cultivation. The situation in Tsinjoarivo, described in Section 3, provides an example. From this analysis of social dynamics, the authors propose to distinguish two types of communities: those where the historical power is not yet challenged and where migrants use marriage to access land and gain some influence (type 1), and those where a redistribution of power and in-depth changes in the social organization occur (type 2). In the first case, the main feature of social organization is a vertical hierarchy of power determined by history and matrimonial alliances. In the second case, social organization reflects a variety of economic activities and the alliances (still often matrimonial) passed between the groups that implement these activities (RESOLVE-PCP-IRD 2004b). Once this distinction is made, the authors raise the question of COBA’s representativeness, proposing a different answer to that of the PCP team. They assert that despite initial plans to make the COBA match with the fokonolona, a drift in the GELOSE approach led to creation of an entity that has little sociological value and can be regarded as a “colonial invention” (Goedefroid 332 2002, in Cormier-Salem et al. 2002, in RESOLVE-PCP-IRD 2004b). This view matches with the arguments I 332
Reference not found. Not in the list of work cited.
527
developed in Section 4. But the authors go further by questioning what it means to be representative. Do the COBAs have to follow the canon of Western democratic culture or do they rather have to represent the main groups that have authority over resource management in the traditional system? They opt for the second version and, from there, propose two options in order to address the two types of communities. In the case of type 1 communities, they contend that the COBA should include legitimate leaders of the foundation lineage in the key positions. The absence of representatives of migrant groups and of women can be regarded as normal and an excessive representation of families that belong to the foundation lineage should not be regarded as a problem. In the case of type 2 communities, by contrast, the management structure should include a wider range of actors belonging to different lineages having different economic activities and different origins, while the excessive representation of certain lineages, including the foundation one, could be a danger (RESOLVE-PCPIRD 2004b). The authors found that the actual composition of COBAs rarely follows this logic. Leaders are members of marginal groups in most cases. Typically, they are people under the age of 40 who spent a few years outside the village, where they learned to read and had contacts with foreign people. They want the traditional system of power and the land use practices to change, which leads them to criticize traditional leaders and elders. Their skills and areas of interest allow them to seduce the agents of foreign aid. They can assume bureaucratic tasks, participate actively in meetings, maintain accounts, understand the procedures to create associations, assume the presidency of these associations (not only COBAs), and channel financial support to the village. Their interest is more or less individualistic according to the cases but they often see involvement with aid actors as a way to elevate themselves in the power structure (RESOLVE-PCP-IRD 2004b). The risk is that tension could appear between
528
them and traditional leaders, who still have a strong legitimacy and could block their projects. In other cases, however, traditional leaders perceive the GELOSE as a “foreign thing” and “do not see the interest to enter into a project that would allow them to acquire a management and control power they already have” 333 (RESOLVEPCP-IRD 2004b, 114). They leave recent migrants, or other people that are more “modern” or have a better understanding of this foreign process, the task to manage them (RESOLVE-PCP-IRD 2004b). These outcomes encourage a slightly more optimistic vision than the one I proposed up until now. Societies and cultures are not frozen and are open to change. Internal and contextual constraints make evolution necessary and the deployment of new forms of power, in relation to the arrival of new actors, is a natural process that can favor wise adaptations to a changing context. The COBA, being a colonial invention, could in the end favor the creation of links between the modern world inherited from colonization and the traditional world that Malagasy communities inherited from their history. If this were true, it could still be questioned, however, in what way the GELOSE would differ from other socio-organizational supports provided by projects, such as the creation of Kolo Harena (Chapter 6) or of any farmers’ association. The same optimism could indeed be applied to coercive approaches, which can also lead to local empowerment by giving incentives to communities to resist and better organize themselves. But whether we adopt an optimistic or a pessimistic stance, only the future will tell us if local communities, in the end, will be empowered by the creation of COBA and other social devices created by projects.
333
Translation by Pollini. Original text in French: “le contrat de transfert est souvent perçu par eux comme une “ chose étrangère ” et ils ne voient pas très bien l’intérêt de rentrer dans un projet qui leur permettrait d’acquérir un pouvoir de gestion et de contrôle qu’ils ont déjà.”
529
5.2.3. The institutional driving of the IRD’s recommendations The anthropological work of the IRD team helped in understanding the social organization of local communities, their dynamics and the impact of external interventions. I will show, in this section that, beyond these insights, the recommendations of the IRD team reveal, again, the pervasiveness of Scott’s (1998) high modernist project in the institutional framework within which the consultants conducted their work. One of the main goals of the RESOLVE evaluation, defined by the donors, was to propose monitoring indicators that would help to better implement management transfers. This objective reveals two implicit assumptions. The first one is that MT is regarded as a wise approach that just needs to be better implemented. The second is that better monitoring is the solution for this improved implementation. In order to satisfy this expectation, the IRD team proposed rules aimed at defining the COBA’s composition and indicators relating to these rules. Consistent with their outcomes, the consultants proposed indicators such as the type of community, the presence of associations prior to the creation of the COBA, the presence and position of migrants in the COBA, the representation of minority or marginal groups and external influences (RESOLVE-PCP-IRD 2004b). Doing so, they provided more precise rules aimed at better designing social organizations, pushing farther the social engineering project of the GELOSE legislation. The consultants apprehended, however, the contradiction within which they were trapped. As they put it, how to implement management transfers adapted to the specific cases of each community, while simplifying and diminishing the number of
530
monitoring indicators? Isn’t it contradictory to call for a finer and simpler process all together? 334 (RESOLVE-PCP-IRD 2004b, 108) In other words, the main question may be who has to address complexity, rather than how to address it. Scott (1998) showed that externally designed social schemes usually fail because they cannot totally account for the complexity of realities. They are always strong simplifications whatever the sophistication of the models they use. External designs that operate on realities may need to be simple, precisely because the reality is complex. Only in this way can the local actors, who carry with them contextualized knowledge, address complexity effectively. The concept of metis, which will be introduced in Chapter 15, will help to understand this better. I will just give here an example to illustrate this argument. If a traditional leader with low literacy led a COBA in a type 1 community (in order to create a COBA that satisfies the proposed indicators), he could lose his prestige by being vested in responsibilities he could not assume in the end. If he abandoned the leadership to migrants, he would take the risk of being challenged by the COBA. Facing this conundrum he would, however, be in better position than outside actors to make this choice, because he knows exactly his skills, his position in the society and the skills and position of his challengers. He owns a contextualized knowledge that can hardly be translated into simple rules aimed at transforming his community. In other words, communities should not be socially engineered from outside, an endeavor that started with the creation of the COBA and could be pushed farther if the IRD team recommendations were to be translated into new formal rules. Rather, communities should find their own way to change and learn from their own mistakes.
334
Translation by Pollini. Original citation in French: “comment mettre en place un contrat de gestion “ sur mesure ” pour chaque communauté, tout en simplifiant et en diminuant le nombre d’indicateurs ? N’est-il pas contradictoire de vouloir affiner le processus tout en le simplifiant ?”
531
Even if ill designed, the GELOSE should be left to the actors for whom it was designed. These actors will decide how to use it and what to do with it. The stronger the external designs will be, the less communities will be able to make their own choice. The problem of the GELOSE may be in the approach itself (in the fact that it is a social engineering project), rather than in the way the approach is implemented (in the fact that it would be a badly implemented social engineering project). The solution, rather than to define a finer and more precise procedure, may be to open the framework and make the monitoring and implementation pathways more diversified, more flexible and more fuzzy, a concept that unfortunately does not match the high modernist project that most states and international organization want to implement.
6. Confrontation with other research outcomes Before concluding, I have to evoke the main studies of management transfers that were conducted in Madagascar. Referring to these works now is a way to test my own outcomes. If the conclusions converged, then my arguments would be strengthened. Randrianasolo (2000), conducted a national evaluation of GCF 335 contracts. He showed that the GCF approach, which is simpler than the GELOSE, is insufficiently understood at all levels, is appropriated by influential persons concerned with their own private interests, is accepted by a fraction of the community (30% is regarded as a good rate for COBA membership), sometimes leads to conflicts between members and nonmembers, but can work well to protect forests from fast destruction by outsiders. Maldidier (2001) conducted a national evaluation of the GELOSE approach. He showed that management transfers are associated with a ban on forest clearing and
335
Randrianasolo’s (2000) evaluation only concerns the GCF approach, a modality of management transfer which applies to the case of forest resources only, as we saw.
532
often with a ban on forest product commercialization. They effectively lead to less forest clearing but villagers tend to consider them as a new form of control and ask for alternatives to the restrictions on their land use. Maldidier (2001) also recognized the ambiguous character of the COBA (community or association) and refers to controversies that accompanied the early implementation of the GELOSE approach. According to him, the GELOSE was criticized for being “unrealistic,” “inapplicable,” “complex,” costly, rigid, formal and “intellectual,” which led to the creation of the GCF approach. In overall, these evaluations identified similar pitfalls to those I evoked in this chapter. In the end, however, these authors drew conclusions more optimistic than the one I will propose. This could be explained by an excessive pessimism on my part, by the institutional framework within which these author worked (Randrianasolo worked as a consultant, while Maldidier was technical advisor at ONE, the agency in charge of implementing the GELOSE component of EP2), or by the fact that the GELOSE was still in its early implementation stage when they conducted their work. Kull 336 (2004) proposes less optimistic conclusions despite the empirical facts he reports are still quite similar. For this author, the GELOSE may transfer responsibilities to communities, not rights, in “a top-down attempt to expand control” (Kull 2004, 262). It uses a “potentially troubling” new institution, the COBA, which ignores “the potential for fractious intracommunity politics” (Kull 2004, 257) and can have difficulty establishing its legitimacy. The consequences of ignoring the dynamics of community institutions could be “noncompliance, elite capture, or reshuffled local power relations” (Kull 2004, 262). Kull argues that “for GELOSE to seriously change
336
The work of this author does not concern specifically management transfers and will be presented in more details in Chapter 12.
533
fire policy and empower communities, the entire suite of fire and forestry laws must be reformed as well” (Kull 2004, 249). Eventually, studies conducted in other parts of the world also present some similar outcomes. Ribot (2002) synthesized the experiences of a large number of community based natural resource management approaches conducted worldwide. He contends that despite CBNRM is
a
promising
approach,
“most
current
“decentralization” reforms are characterized by insufficient transfer of powers to local institutions, under tight central-government oversight” (Ribot 2002, 1). In many cases, “local institutions do not represent and are not accountable to local communities” (Ribot 2002, 1). Ribot (2002) argues that representation by institutions that are both accountable and empowered is essential. “Transferring power without accountable representation is dangerous” (Ribot 2002, 2). It “favors the most organized and powerful groups, … [leading to] elite capture” (Ribot 2002, 1). Establishing accountable institutions with no power, on the other hand, “is empty” (Ribot 2002, 2). For Ribot (2002, 2) “most resistance to decentralization comes from within government” and mediation could play a significant role to achieve better results. Ribot (2002) concludes that decentralization, in spite of these difficulties, is too young for a definitive conclusion to be given and that efforts need be pursued. Other analyses put into question the real principle of negotiating management rules with arguments that are consistent with my criticism of environmental mediation. For Leach et al. (1999, 241), It would clearly be naive… to assume that negotiation processes take place on a level playing field. Indeed, the very idea of negotiation conjures up an image of parties equally able to voice their positions and argue for them, which is very far from reality in most of the situations confronted by CBNRM. Just as power relations pervade the institutional dynamics of everyday resource use, so they would pervade any negotiation process.
534
The use of participatory methods as control devices for implementing the international conservation agenda and their disruptive effect on social organization at community level have also been pointed out by Brosius et al. (1998), Belsky (1999), Kellert et al. (2000) and Harrison (2002). In sum, many investigations about community based natural resource management produced similar narratives, but conclusions varied in their level of optimism or pessimism. Only the future will say whether optimism or pessimism is justified.
7. Conclusion Management transfers can be regarded as a tool for injecting modern administrative logic in the organization of local communities. They are instruments in the service of what Rahnema (1992) called “manipulative participation,” despite the fact that they were initially not intended to be so. This manipulative character results from a fundamental lack of congruence between state and community objectives. On one hand, the institutional framework within which the GELOSE and GCF approaches were designed is the National Environmental Action Plan, whose main objective is to stop forest clearing and fires. On the other hand, communities need to clear forests and to ignite fires to sustain their livelihood. According to Antona et al. (2004), congruence between stakeholders’ objectives is the main factor that determines the success or failure of management transfers. It seems that this congruence does not exist and we may face, here, the fundamental problem that the GELOSE failed to address. The heritage management approach pretends to depart from participatory methods that already proved their manipulative character, but is embedded in an ideal of rational and transparent communication that leads to overlooking power issues,
535
such as the domination of nonnegotiable objectives defined by the global environmental agenda. In the relationships between NGOs, state services and local communities, the three forms of participatory tyranny identified by Cooke and Khotari (2001) have more effect than this rational communication. The tyranny of decision making and control is incarnated by the COBA, a new institution which favors decisions that match with the global environmental agenda. Customary management by fokonolona leaders are the overridden “existing legitimate decision making process” (Cooke and Kothari 2001, 7). The tyranny of the group is incarnated by an opportunistic elite of more educated people who has a comparative advantage for taking control over the COBA. The tyranny of the method, eventually, is the exclusion from COBA’s leadership of those who cannot manage paperwork and other “modern” administrative tasks. It is also the fact that all COBA must adopt the same structure, implement the same RLTS, work with the same mediators, establish the same tripartite contracts, implement the same mapping and zoning and provide the same monitoring indicators. In other words, and referring to the concepts introduced in Chapter 9, the GELOSE may favor the hegemony of global conservation discourses. The ideal of rational and transparent communication has led to overlooking the issue of power. The COBA may actually serve to channel to communities a form of knowledge which is in reality the expression of coercive powers. This was not intended but is a result of the naïve hypotheses of the heritage management approach and of the political context of the mid-1990s in Madagascar. Designed to facilitate the shift from administrative rationalism to democratic pragmatism (Chapter 8), the GELOSE eventually became a tool for more thorough administration of the land. Using devices such as mapping and management plans elaborated with input from experts, it unsuccessfully attempted to address the complexity of multifunctional landscapes and of community organization. Moreover, this lack of success may be a chance for the targeted communities.
536
Creating COBAs can thus be seen as a radical modernization project at work, but a modernization that concerns the social and organizational sphere rather than the technical one. The COBA, the DINA, the Mediator and the RLTS are the instruments of a sophisticated social engineering project. They help to reshape local communities and in this sense can be regarded as a manifestation of what James Scott (1998) called high modernism (Chapter 8). The GELOSE may just be more local, more ephemeral and more dependant on short-term external financial resources than the more extreme cases described by Scott (1998). Fortunately, this may provide a chance for its failure or its distortion and appropriation by local communities at their own advantage. This leads us to Muttenzer’s (2006) analysis of management transfers. For this author, management transfers would illustrate one more attempt of the state and postcolonial power to take control of customary rights and to redesign realities. But they would fail to do so as coercive approaches have failed before, due to their ignorance of some key challenges faced by local communities (such as the need to convert forest land), and to the fact that a complex and efficient system of negotiations and interactions between the legal and the legitimate has always been in place but remained ignored. This more optimistic perspective, which was also apparent in the IRD team’s outcomes, may be right if we consider the overall picture. Life in the Malagasy countryside was not significantly changed by the successive forest policies. Farmers kept on increasing their land by clearing forest, extracting natural resources, negotiating this extraction by bribing state representatives, migrating and recombining their social organizations during these migrations. I would contend, however, that state policies and project interventions do matter locally. Communities were removed to marginal land by the settlement of colonists or migrants, had their social organization weakened by the creation of new legal powers that brought corruption, had their
537
access to resources restricted by the gazetting of protected areas, had their land taken by urban people who managed to obtain legal title. Farmers submitted to these pressures always managed to find a piece of forest to clear in order to resist, but the weakest social groups were removed to the most marginal land (remote, less fertile, with steep slopes). If Management Transfers are just a continuation of this logic, it appears difficult to contest that they will have similar effects. Some social groups, those whose interest matches better with those of external actors, may be empowered, but others, those who depend on the unsustainable exploitation of natural resources, may end in becoming more marginalized. In short, the COBA, like the Kolo Harena association in Beforona (Chapter 6), may encourage resource capture by the local elites. All these critical analyses may be perceived as quite depressing for the many development and conservation actors genuinely committed to make the GELOSE approach work, to improve the well being of communities, and to preserve the remaining primary ecosystems. They could also be utilized by the enemies of community based natural resource management approaches. The recognized failure of CBNRM has indeed already provoked a movement of rejection among conservationists (Terborgh 1999; Oates 1999). The intention of this Chapter is clearly to oppose this movement, as did Brechin et al. (2002), Wilshusen et al. (2002), and Chapin (2004), by proposing new strategies for successful CBNRM. These new strategies, and their articulation with other approaches, will be presented in Part V, but I will anticipate a little bit on these conclusions here. First, I believe that we should abandon our attempts to design virtual perfect societies and that our role should be limited to provide more information, more opportunities, more capacities, and more means to all actors that want to engage in change, without targeting a specific group. This can be achieved through an equal or
538
less unequal access to education, health, market, infrastructures and political power. The GELOSE in its current form is not consistent with this logic of equal opportunity due to the high level of skills required for being an active COBA member or leader. Rather than defining criteria for a perfect COBA composition, we should reduce the skills required to become COBA leaders. More potential candidates would then be possible, and communities could choose the candidates according to their own criteria. The price to pay would be to decrease our expectations in terms of perfect designing, monitoring, management and social representativeness. Second, the fokonolona must be recognized as a legal body with a legal statute, as it was during the second republic, and its traditional institutions, such as the fokonolona assemblies, must be granted some legal form of power. This may indeed be the starting point for implementing genuine community based natural resource management in the context of Madagascar. Recently, the President of Madagascar made some declarations in favor of giving more responsibilities to the fokonolona to achieve development. This declaration must be translated into a formal political decision. Once fokonolona are legally recognized, they could be granted management rights like the communes, districts and regions, using article 24 of the forest law (Gouvernement de Madagascar 1997) as a legal framework. To the contrary of the GELOSE, this new type of contract could concern a territory and all its resources. Comanagement, involving the fokonolona and a partner, which could be an NGO or another category of actor, could also be implemented under article 24. The GELOSE legislation could still be utilized, but for granting single resource management rights to associations that would be controlled by the fokonolona. Indeed, the fokonolona could replace the commune in the tripartite contract, while the commune could play a role in coordinating several MTs implemented on its territory.
539
Third, the fundamental lack of congruence between local and global objectives (conversion versus conservation of forests) must be recognized, as well as its nonnegotiable character. In the present political and environmental context, it appears quite clear that the ban on forest clearing will remain in place. Even the strongest advocates of community based natural resource management never proposed to put an end to this ban. On the other hand, closing the forest frontier is a luxury that most Malagasy communities cannot afford. Once we would render these two points explicit, we could stop being manipulative and we could take distance from our dreams of natural harmony between communities and their environment. The illegitimate character of closing the forest frontier could come into full light and the necessity to put in place a system of compensation (I will come back on this issue in Chapter 15) could be recognized. Only by doing so can we put an end to the lack of congruence and create conditions favorable to genuine community based natural resource management. Under such a clear combination of control and supports designed to make local and global interests match together, communities could find their own way to develop. Access to forest land would be controlled, but land uses and social organizations would be freed from our design, not submitted to zoning, monitoring, planning, reporting and else. The communities and their environment could match with each other and evolve according to changing political, social, economic and physical contexts. And if we would further provide better education, communities would move toward a modernity they never rejected, entering into it at a pace dependant on their skills, not on our wishes.
540
CHAPTER 12 THE ANTIFIRE DISCOURSE
1. Introduction I showed in Chapter 2 that the antifire legislation was initially designed to fight against bush fires and uncontrolled fires and that slash-and-burn cultivation in Madagascar is regulated, but not forbidden. This land use can be practiced without authorization on nonligneous vegetation, requires the delivery of authorization to be practiced on dense ligneous vegetation and was usually tolerated on any type of secondary vegetation if fires were controlled by firebreaks. Chapters 5 and 6 showed that fire repression sometimes went beyond what could reasonably be expected given these regulations, revealing a demonization of fires (conclusion of Part III) which this chapter is now intended to analyze. The antifire discourse has been analyzed in detail by Kull (2004). His outcomes will be presented in this section, for not repeating his work, but a more personal stance will also be proposed. Section 2 will present Kull’s (2004) vision of the antifire received wisdom and will confront it to my own outcomes. I will show that the antifire discourse pervades in all other environmental discourses and explains the failure of the ASB and GELOSE approaches. Section 3 will present Kull’s interpretation of the impact of fires on the Malagasy environment, while Section 4 will present the attitude Kull (2004) recommends to have with regard to anthropogenic fires. In the conclusion, I will propose a slight departure from Kull’s analysis. This departure will be analyzed in more detail in Chapter 13, which concerns political ecology, the academic discipline that Kull (2004) adopted as a research framework.
541
2. The antifire received wisdom according to Kull (2004) Kull’s central thesis is that fires in Madagascar became a political problem that rendered them more harmful than they used to be when they were a mere tool for land management: Madagascar is not just aflame, it has a fire problem, a century-long conflict over natural resource use and protection. On the one side, a broad group of actors and institutions believe that the island sees too much fire, that fires should be stopped. They blame the fires for a variety of ills, including deforestation, desertification, rangeland impoverishment, soil degradation and accelerated erosion. Being powerful – the past colonial rulers, the Forest Services, the postindependence leaders, the international environmental agencies – they have used their power, including laws and propaganda, to enforce their views. On the other side, the Malagasy farmers and herders rely on fire as a tool of resource management, as a tool to meet their own livelihood needs, and they are trapped in the antifire laws. Being resourceful people who know and understand the intricacies of fire and the weakness of the state, they keep on burning, taking care to be out of sight, and letting fire do its work for them (Kull 2004, 4) [Hence] the “fire problem” of Madagascar is not that there are too many fires (though this is certainly true in certain places and moments); the real problem is a century-long conflict over appropriate resource use (Kull 2004, 11) An elite coalition … has criminalized burning through rhetoric and repression. However, 100 years of antifire repression have only succeeded in worsening the problem, polarizing the peasantry against the state, and giving a meaning to fire itself as a tool of resistance. Farmers and herders resent state interference with a practice they consider crucial for environmental management and for their livelihoods. They take advantage of fire’s complex character and of internal inconsistencies within the state in order to continue burning, leading to today’s stalemate. (Kull 2004, 12) The historical perspective I presented in Chapter 1, as well as my case study in Part III, are consistent with these outcomes. Fires have occurred in Madagascar for as long as people lived there (Chapter 1). We saw, based on the synthesis by Bertrand and Sourdat (1998), that their negative impacts were over-dramatized, providing arguments to justify colonial control over the peasantry. The tension between conservationists’ concerns and farmers’ logic did not stop after independence. A
542
culminating point was the transition period between the second and third phase of NEAP, in 2002, when the new regime of the Malagasy state, with support from donors and development and conservation projects, launched the antifire campaign, and when the Toamasina Province issued an order that forbid the use of fire in agriculture, in 2003 (Chapters 3, 5 and 6). A vicious circle of more repressive policies and stronger resistance through the ignition of fires may still be at play, with negative impacts on both ecosystems and livelihoods. We saw also that the demonization of fires shapes knowledge about agricultural development. It gives a justification to the development of sophisticated fireless alternatives to slash-and-burn cultivation, such as agroforestry technologies, mulching and composting, where an excessive focus on yield and biomass production leads to overlooking elementary economic criteria. It renders development organizations blind to endogenous pathways to intensification, leading to the exclusion of domains of knowledge associated with these realities. Using the concepts introduced in Chapter 9, we can say that during one hundred years of “commentary” about fires and their negative impacts, some “disciplines” (soil sciences and conservation biology) appropriated fire discourses and “excluded” other disciplines that could challenge their outcomes (anthropology, political ecology). Soil sciences elaborated a political “doctrine” (the alternatives to slash-and-burn paradigm described in Chapter 10) which justified efforts to “educate” farmers and convince them to adopt this doctrine, through communication, environmental education and repression. A “machine” was built which “recruited” more and more people (policy makers, staff from rural development and conservation projects and donor agencies, and young Malagasy students trained to adopt this discourse) and increased its “authority” through the building of a “strong rhetoric.” This machine even managed to spread the antifire rhetoric among local stakeholders, by connecting communities to the “authors”
543
of this rhetoric, through a new device: the COBA. The machine almost closed the “controversy” until a new generation of researchers, coming from the fields of social sciences, political ecology, and anthropology fortunately reopened the debate. Christian Kull (2004), clearly, is an essential contributor to this reopening. But Kull’s discourse may also suffer from some limitations, as I will now show.
3. Kull’s thesis about human impact on the environment Saying that fires are not demonic does not imply that they are not harmful to the environment. Once he identified the excesses of the antifire discourse, Kull (2004) thus had to take a position in the debate. In order to achieve this, he attempted to separate the myth from the reality, by focusing on a few empirical case studies. He showed for example, based on the analysis of aerial photographs and sketches drawn by early travelers, that the area covered by tapia 337 forests in the highland did not decrease during the last century, to the contrary of received wisdom that consider current management by local people as being destructive. These conclusions are consistent with the outcomes of research conducted in West Africa using similar methods (Fairhead and Leach 1996; 1998). It appears clear, in the light of these works, that environmental degradation during the past century was exaggerated by colonial discourses that needed to blame indigenous people in order to assert power over them. In many cases, such degradation did not occur and vegetation cover even increased (Fairhead and Leach1996; 1998). In spite of these outcomes, contemporary representations are still influenced by these discourses and excessively blame farmer for their negative impact on land. But Kull’s (2004) and Fairhead and Leach’s (1996; 1998) analyses also have a weakness that exposes them to the criticisms of conservation biologists. They are 337
Uapaca bojeri Baill. (euphorbiaceae), a fire resistant tree.
544
limited to the past century and cannot serve to argue about what happened on a larger time scale. It is still possible that significant environmental degradation occurred before the period covered by their analysis, over hundreds or thousands of years of human presence. Kull does not reject this possibility but is sometimes ambiguous about the position he takes. He makes for example the following assertion: the antifire received wisdom sees the tapia woodlands as “degraded remnants” of diverse prehuman forests, reduced by incessant burning to a fire-tolerant species that coincidentally had economic value to the locals. In contrast, I have argued … that fire is a key management tool used by rural Malagasy in the management of these woodlands, specifically for their livelihood goals tied to artisanal crafts in wild silk, trade in fruit, firewood supply, and food supplements. (Kull 2004, 144) It is not clear in this paragraph whether Kull rejects the “degraded remnant” thesis. It is not clear whether he attributes the term “degradation” to the ecological or the economical dimension of the forest, or to both. In another passage, he refers to authors that considered tapia forests as “a sign of previously grander forests” (GirodGenet 1898, in Kull 2004, 132) in a way that leads the reader to think that he is challenging this view. Kull, however, never makes this rejection explicitly. In sum, he presents environmental degradation as a myth but is not clear about where the myth starts and ends. The same ambiguity is actually apparent in the work of Fairhead and Leach (1996; 1998). The consequence is that these authors, as Kull, are often utilized for arguing that environmental degradation is a social construction, whereas they provide empirical evidence in favor of this thesis for the last century only. It is still possible that very significant environmental degradation occurred in more ancient periods in the areas studied by Fairhead and Leach (1996; 1998) and by Kull (2004), even if a relative environmental recovery occurred during the last century. This ushaped dynamics is now recognized by an increasing number of ecologists and agronomists (Figure 23 page 473). But if soils and biomass could partly recover in the process, the same can hardly be said for biodiversity as we saw in Chapter 5.
545
Kull, however, is aware that his data concern only the last century. His description of the historical transformation that led to the tapia forests is influenced by the recent outcomes of the work of archeologists and paleo-botanists: The presettlement highlands were a dynamic mosaic of grasslands, savanna, heath, woodland and riparian forests, shaped by a natural fires regime, grazing megafauna, and climate fluctuation. As humans arrived, their fires, set to clear land for pasture, agriculture and hunting, removed most woody vegetation. The tapia woodlands, found on certain soil types, were preserved because of their economic value and fire-ecological characteristics. Thus, the tapia woodlands should be seen as an anthropogenic forest, shaped through a complex history to meet local subsistence and economic needs. (Kull 2004, 143) This portrayal confirms rather than contradicts the thesis that tapia woodlands would be “degraded remnants”; that they would be “the sign of previous grander forests,” even if these forests did not cover the whole highlands as earlier botanists believed. In spite of this, Kull (2004), however, is still reluctant to see tapia woodlands as “degraded remnants.” I see two reasons for this. First is his belonging to the discipline political ecology, which, as any discipline, contributes to shaping knowledge, creating a tension between what the reality says and what the established representations of the discipline say. Chapter 13 will address this issue more in depth. Second, Kull (2004) attributes a significant value to the economic and cultural dimensions of ecosystems, whereas ecologists who defend the degradation thesis focus on ecological values. Certainly, cultural, social and economical dimensions all deserve attention. This entire dissertation is dedicated to showing the abuse and mistakes toward which we are directed if we fail to consider social issues. But Kull (2004) conflates together the natural and the cultural dimension of tapia woodlands. For this reason, he considers them as modified ecosystems, whereas empirical facts
546
(considering longer time span than the last century) rather show that they are degraded ecosystems. The consequences of this conflation will be analyzed in more detail in Chapters 13 and 14. But to say just a few words, it is clear, using the vocabulary of Latour (1993), that tapia woodlands, like most ecosystems, are hybrids of human and nonhuman objects. But precisely because of this hybrid character, a cultural and a natural dimension exist in them and have to be distinguished (but not separated). If not, ontological categories may disappear, the analytical framework may be weakened and interpretations would lose accuracy and clarity. This lack of clarity concerns for example what is degraded: is it the woodland regarded as a land use, or is it an initial ecosystem assessed in terms of biodiversity, biomass and basic ecological functions? In this confusion, dialogue between ecologists and social scientists would be rendered difficult, leading to sterile polemics and to the development of parallel representations. We will see in Chapter 13 that this may have already occurred and that this conflation of cultural and natural values is indeed a central issue in political ecology.
4. Kull’s conclusion: what attitude to take with regard to anthropogenic fires? The conclusion of Kull (2004, 76) with regard to fire is that fire is neither good nor bad, but complex. The ancient metaphor of the phoenix is quite apt: fire is at once destructive and creative. The black waste of a burned field gives way to the lush greenery of fresh resprouts. In human hands, however, fire is even more complex. That is because one person’s creation is another’s destruction. The creation of pastures, rice fields, and livelihood is the destruction of habitats, timber reserves, and biological diversity. This explains, perhaps, the vehemence of the conflict over fire. This perspective has the merit of putting a halt to the excessive demonization that led to the unrealistic and unethical policies and practices I described in Chapter 2 and 6, and to be more open to local answers to the problem. But policy makers at the
547
national level need to be more conclusive concerning the overall impact of fires, in order to take appropriate decisions without losing themselves in the inextricable complexity of local realities. These impacts can concern the environment but also the economy, the society, the politics. All these realms must be addressed distinctly for being understood in their unity. Only by doing so can we, eventually, understand what Madagascar gains and what it loses from wild fires, pasture fires and tavy fires ignited every year. And only then could wise fire regulations and policies be designed.
5. Conclusion In conclusion, Kull’s analysis has the great merit of debunking received wisdom and calling for more open-mindedness and flexibility. It shows that the fire problem is much more complex than what analyses that focus on its biophysical dimension could let us think, and that simplification favored a politicization of fires that amplified their destructive effects. Kull (2004) still sees the negative impact of fires on vegetation and soil as a reality supported by many empirical evidences, but not as the only reality that matters. He showed that a degraded ecosystem, such as the tapia woodland, can provide as much, or even more cultural and economical value than the virgin ecosystem that preceded it. But once he has identified the economical, cultural and natural dimensions of fires, he has a tendency to conflate them together. The consequence can be the replacement of a social construction (the supposed demonic character of fires) by another one (a supposed equivalence of socially constructed “natures”). In sum, Kull’s analysis may have slightly fallen into the trap of relativism (Chapter 9), which may not be the most appropriate answer to the antifire discourse that positivist science favored. Chapter 13 will show that relativist epistemologies are indeed quite influential in political ecology, the discipline to which Kull belongs.
548
CONCLUSION
Chapter 8 showed that Western development and environmental discourses are dominated by a conception of progress inherited from the Enlightenment, which envisions the future as a march to more technological societies. These discourses see the earth as a finite system whose carrying capacity has to be improved by technological innovation. The people, and not only the physical world, have to be rationally managed to achieve “progress,” as revealed by the high-modernist project analyzed by Scott (1998). This rational management pretends to solve the trade-off between conservation and agricultural development and to guide societies toward sustainable development. This perspective is consistent with the theory of Malthus (1798), which sees population growth as the consequence, not the cause, of technological innovation. But it is at odds with the observations reported in Part III, where changes emerge from inside, rather than being designed from outside. In Beforona, technological innovation seems to be the consequence of population growth and increased pressure on land, verifying Boserup’s (1965) model of agricultural development. This mismatch between discourses and realities may explain the failure of rural development projects (Chapter 6). Chapter 9 proposed a method to analyze more in depth environmental and development discourses and their translation into practices. This method borrowed concepts from Foucault (1970) and Latour (1987), in order to analyze the subjective dimension of the social construction of discourses. It also departed from these authors by contending that the real world, or the real nature, are accomplices of our knowledge, which renders possible a relative objectivity.
549
In Chapters 10, 11 and 12, I applied this method to three discourses that are central to the received wisdoms identified in Part III. These three discourses are the alternatives to slash-and-burn discourse, the participatory discourse and the antifire discourse. Doing this exercise, I attempted to separate discursive elements that match with realities from those that match only with themselves. In Chapter 10, I compared the alternatives to slash-and-burn paradigm to the building of a strong rhetoric in the sense of Latour (1987), but showed that the real world was in the end stronger than this rhetoric, verifying the epistemological model proposed in Chapter 9 (Figure 15 page 435). In Chapter 11, I showed the incredible pervasiveness of conservation discourses and their capacity to reshape alternative discourses designed to challenge them. I verified, in the case of community based natural resource management approaches, Foucault’s assertion that knowledge and power are intimately intertwined. I showed that unequal power distribution impedes consensus from being achieved by rational and transparent communication, which explains the failure of the GELOSE approach. In Chapter 12, mostly derived from Kull’s (2004) work, I showed that the demonization of fires may be the main cause of misrepresentation in other discourses, because antifire rhetoric pervades both the ASB rhetoric and the implementation of management transfers. I also showed, by critically analyzing Kull’s (2004) work, that anti-theses of dominant discourses are now emerging, but that these anti-theses are biased by social construction, as are the discourses they challenge. In sum, Part IV showed that the mismatch between representations and realities, which I observed in Part III, can be explained by the hegemony of conservation and development discourses embedded in Western culture and by the use of communication and knowledge as a medium to spread these discourses among local stakeholders. The consequence is the filling of the remaining free interstices in local systems of knowledge (a filling which is fortunately inefficient), a generalized
550
inefficiency of both conservation and development programs and the disruption of local practices. This conclusion is consistent with the outcomes of Part II (Chapter 2), where I showed that the ideal of the Malagasy Environmental Charter, which consists of reconciling humans and their environment, failed to be achieved, and that communication is used as a manipulative device aimed at providing illusions of success. It is now necessary to move beyond the dominant environmental discourse, the illusions it created and the emerging alternative discourses. This will be the endeavor of Part V, which will start by a larger review of political ecology, the discipline within which these alternative discourses emerged and within which this present dissertation may be located.
551
PART V SYNTHESIS: TOWARD NEW DISCOURSES, EPISTEMOLOGIES AND PRACTICES
552
INTRODUCTION
The realities described in Part III and the discourses analyzed in Part IV told us similar stories. The parallel is clear between the efforts of the LDI project to develop agrobiological technologies, such as “ecological ginger,” (Chapter 6) and the development of alternatives to slash-and-burn by the ASB Consortium (Chapter 10). I concluded Part III by saying that finding new alternatives is not what matters more. The ASB Consortium made the same conclusion when it shifted toward a new paradigm paying more attention to the political economy of deforestation. The case of the VMCRCC project, in Chapter 6, showed that, unfortunately, received wisdoms are pervasive and powerful and that this shift will not be easily translated into new practices. The parallel is clear also between the anti-tavy campaign, described in Chapters 5 an 6, and the antifire discourses analyzed in Chapter 12. Fire repression, because it is based on narrow analyses which consider nutrient balances and vegetation dynamics outside of their social and cultural contexts, hardens livelihoods and may lead to political protests, which will be counterproductive in the end. A parallel also exists between Chapter 6 and Chapter 11, which both described “participatory top down” methods at work. Chapter 6 showed how Kolo Harena associations are used to convince farmers to abandon slash-and-burn systems. Chapter 11 showed that COBA associations play a similar role. They are used by projects as a “Trojan horse” (Rahnema 1992) (Chapter 8) for spreading a western conception of “sustainability.” In this conception, the sustainability of ecosystems matters more than the sustainability of livelihoods. For farmers, the sustainability of livelihood matters
553
more and is achieved by exploiting a succession of resources. When one is depleted, another one is tapped. In sum, the stories presented in Parts II and III revealed a mismatch between representation and realities, but also between two systems of representations. On one side, there is the representation of farmers, which is deeply anchored in reality because if it did not match with it, there would not be sufficient rice to eat. On the other side, there is the representation of projects, whose main function is not to maintain rice production (the fate of projects does not depend on it). The consequence is that when farmers adopt the technological package proposed by projects, their production decreases. I verified this in the case of Beforona (Chapter 6) but also in Ikongo and Tsinjoarivo (Chapter 11). The oxymoronic character of the sustainability concept (Chapter 8) may be the main cause of this mismatch. This concept provides the illusion that humans and their environment can be easily reconciled, as required by the Malagasy Environmental Charter (Chapter 2). This opens the way to a schizophrenic world where an illusion of agreement hides a fundamental disagreement on priorities (ICDP approach in Chapter 2), or a fundamental lack of congruence (GELOSE approach in Chapter 11). New approaches can continue to replace each other, but they will all fail as long as this schizophrenic pattern dominates, as they could all succeed if it disappeared. We can wonder why this mismatch persists. The reason may be that project representations are aimed at matching with themselves rather than with the reality, because their main function is to justify the future existence of projects. Consulting firms, NGOs and research organizations are engaged in a competition to receive funding from donors. As failure is the norm and success the exception, the winner will be the institution that provides the best rhetoric and the most convincing illusions. Both the donors and the organizations that receive the aid are satisfied by this state of
554
affairs. Organizations can create illusions that satisfy their donor while donors use these illusions to satisfy tax payers. Moreover, all this is done without disturbing the political economy of international relations. It even maintains the status-quo. For example, behind the need to justify the eradication of slash-and-burn systems in Beforona, there is also the need to decrease carbon release in the atmosphere with minimal political and economic cost for developed countries. The main function of the ASB concept, in this context, is to create the illusion that communities can reduce carbon release for their own benefit. Carbon can even be sequestered (VMCRCC project in Chapter 6) and the profit generated by this sequestration captured by the state while “perfect” biophysical models (agroforestry technologies) create the illusion that communities can lose the usufruct on part of their land without being losers overall. In the same manner, behind the gazetting of protected areas, there may be the need to secure gene reservoirs for biotechnologies, which may be the future engine of economic growth when information technologies have finished playing this role. From this perspective, the unconfessed function of the GELOSE is to create the illusion that closing the forest frontier can lead to a win-win situation. But, as international relations and the macroeconomic context are the real cause of the environmental stalemate, not mismanagement, no solution can emerge. Facing the destructive social consequences of these macroeconomic forces, alternative discourses emerged. These alternative discourses were committed to the debunking of manipulative received wisdoms (Chapter 12) and advocated for a liberation from the hegemony of dominant discourses. But due to this political commitment, they may have ended in being biased as well. They created their own mismatch with realities. In the end, they will not help to impede the hegemony of
555
conservation discourses because they will not be taken seriously by biophysical scientists, and even, as we will see in Chapter 13, by some social scientists. To summarize, Parts III and IV showed that discourses are driven by politics rather than by science. They are a technology of power, which justifies using tools and concepts of poststructuralism and social constructionism to analyze them. But once the political driving forces of discourses are delineated, something must remain. Behind the representation, there is the reality which has to be found, even if, in an apparent paradox, this can only be done by creating a new representation. This is, indeed, the main departure that I took from the postmodernist stance. The objective of this fifth part is to explain more in depth this departure and to show how it can lead us to new strategies, new practices and new hopes for both the Malagasy ecosystems and the Malagasy people, and for both our cultural and our natural world. In order to achieve this, I will first synthesize my outcomes in the light of political ecology (Chapter 13), a discipline whose narratives are similar to the ones I developed in this dissertation. The work of Kull (2004) belongs to this discipline and I showed in Chapter 12 that I departed from Kull’s outcomes in certain aspects. This departure will be confirmed in Chapter 13. Second I will propose, in Chapter 14, an epistemological framework derived from the methodology I elaborated for analyzing discourses (Chapter 9). This epistemological questioning will help me justify my departure from political ecology. It will confirm that I adopted a Popperian perspective, and I will use Poppers’ (1963) conception of the advancement of sciences to synthesize again my outcomes. Third, I will propose, in Chapter 15, a series of actions, based on the outcomes of Parts III and IV and on my epistemological framework. This last chapter will show that questioning and changing our world are two inseparable processes, the
556
consequence being that science and politics need to be unified for achieving a single endeavor.
557
CHAPTER 13 TOWARD A POLITICAL ECOLOGY OF DEFORESTATION AND SLASH-ANDBURN CULTIVATION
1. Introduction In this chapter, I will present a young discipline, still in a process of maturation: political ecology. This discipline provided the conceptual and methodological framework of Kull’s (2004) work presented in Chapter 12. We will see that the main narratives encountered in this dissertation (the marginalization of poor farmers, the appropriation of resources by the most competitive groups) are central in political ecology. Despite arriving at this discipline by accident, I was not completely foreign to it. Comparative agriculture, whose tools and methods I used in Part III, can be regarded as a school of political economy, a broader discipline that encompasses political ecology. Section 2 will present Blaikie and Brookfield’s (1987) book Land Degradation and Society, a pioneer work in political ecology. It will be seen that this dissertation parallels this book in many points. This was an unintended convergence, as I read the book after the first draft of Parts I, II and III had already been written. Section 3 will review more recent work in political ecology and will present the main narratives developed by this discipline. It will confirm the affinities between political ecology and this dissertation. Section 4 will review recent political ecology research conducted in Madagascar, complementing the review of Kull’s (2004) work about antifire discourses in Chapter 12. It will confirm that political ecology presents some weaknesses when accounting for the biophysical dimension of environmental degradation. Section 5 will show that social constructions biased political ecologists’ representations as they biased those of ecologists. It will argue that in order to
558
depoliticize the debate, ecology and political ecology narratives need to be merged together. It will also show that my criticisms are not isolated and that other scholars have identified a revisionist tendency in certain forms of social environmentalism influenced by political ecology. The conclusion (Section 6) will summarize these outcomes. It will draw a transition to the two last chapters of this dissertation, where deconstruction will give way to the reconstruction of new epistemologies and practices.
2. A parallel with Land Degradation and Society (Blaikie and Brookfield 1987). Land Degradation and Society, edited by Piers Blaikie and Harold Brookfield and published in 1987, 338 is a key early text of political ecology, widely cited and highly influential on later work. The issues addressed by these authors, the methods they propose and their outcomes are very similar to those of this dissertation. This section will summarize the outcomes of this book by a series of quotes, in order to show these similarities. First, Blaikie and Brookfield (1987) made a statement about the failure of conservation policies: It has been forty years since some of the first seminal works on environmental degradation were written (for example, Jacks and Whyte 1939; Osborn 1948; Carson 1962; Commoner 1972).… Even now, the volume of literature and proliferation of national and international institutions flows unabated. In spite of this, whole United Nations agencies and a worldwide environmentalist movement have been unable to make more than a marginal impact upon the prevailing effects of the exploitation of nature for short-term gain. (Blaikie and Brookfield 1987, xvii) It would be naive to assume that the root of this state of affairs lies in intellectual failure alone. However, it is argued in this book that much of the literature on land degradation is beset by a fundamental theoretical confusion. (Blaikie and Brookfield 1987, xvii) 338
Shortly after The Political Economy of Soil Erosion in Developing Countries, by Blaikie (1985), which has similar content.
559
This statement of failure is similar to the one I made in my general introduction and the explanation provided by the authors (the fundamental theoretical confusion) reminds my hypothesis of mismatch between representations and realities. Blaikie and Brookfield propose three causes to explain this failure and the confusion associated with it: First, the nature of the debate itself between scientists, commentators and decision makers in government has hardly been critically examined. Since land degradation is par excellence an interdisciplinary issue, a comprehensive theory requires the combination of analytical tools of both the natural and social sciences. (Blaikie and Brookfield 1987, xvii) The second cause of confusion arises at a more fundamental and ideological level. There are profound differences of opinion on the overall significance of land degradation, which arise from opposing theories of social change. (Blaikie and Brookfield 1987, xviii) The third source of confusion in the literature of land degradation is related to the others, and arises from a failure to view degradation within a wide historical and geographical framework. (Blaikie and Brookfield 1987, xx) The first of these three causes (the need to combine analytical tools) is consistent with my attempts to go beyond disciplines. Despite the fact that twenty years have passed since the publication of Land Degradation and Societies, comprehensive studies still remain to be achieved. New areas of knowledge emerged at the intersection of disciplines, rather than from their inclusion, and these new areas may have become as narrow as their mother discipline. The a-disciplinarity ideal I adopted for this dissertation is an attempt to remedy this. Concerning the second cause (the diverging opinions among scientists from different disciplines), it reflects the tension between positivist and relativist epistemologies and could be analyzed in the light of the epistemological model I proposed in Chapter 9. Concerning the third cause, it leads to advocating more comprehensiveness at spatial and temporal scales, which is, again, consistent with the methodological choices I made for this dissertation.
560
Once they have listed these causes of failure, Blaikie and Brookfield propose new methodological pathways and conceptual frames to address land degradation. First, they propose to study realities and discourses over a chain of actors that they call the “chain of explanation.” This chain starts with the land managers and their direct relations with the land (crop rotations, fuel wood use, stocking densities, capital investment and so on). Then their next link concerns their relations with each other, other land users, and groups in the wider society who affect them in any way, which in turn determines land management. The state and the world economy constitute the last link in the chain. (Blaikie and Brookfield 1987, 27) I attempted to address this chain of actors in this dissertation. Part III was dedicated to the observation of local actors, analyzed through their practices, while Part IV concerned global actors, analyzed through their writings. Second, Blaikie and Brookfield (1987) point out the limits that quantitative data collection faces in addressing complex and comprehensive issues. Their arguments are similar to those I used to justify my preference for qualitative methods (Chapter 4): Measurement … is not an isolated process. First, somebody has to decide to do the measurement; set a working hypothesis for the measurement to test; choose a set of techniques; arrange a sampling programme and people to do the sampling; analyse the results and use judgement in the interpretation of these results; and decide how those results should be presented, and to whom. Then there is the recipient of the measurement who puts the data into context (or rejects them entirely) and who has to make value judgements as to the worth and applicability of the information. Finally, there is the end-user of the measurement; the person who makes the decisions, who bases a course of action on the results so presented. All these people have their preconceptions, misconceptions and ideologies, therefore, measurement is never neutral, never a pure service for science or policy. To quote Weatherall: “Man, as a scientist, is inescapably part of any experiment he conducts” (1968: 159). (Blaikie and Brookfield 1987, 49)
561
As an answer to this issue, Blaikie and Brookfield (1987) propose the acceptance of uncertainty and plurality, a stance that is also consistent with the methodological choices I made in this dissertation (Chapter 9): What then is our response to these demands for data which probably cannot be met? Presented with these problems it looks as if the task of explanation outruns the prospect of empirical verification. Part of the response is an adaptation and development of the ideas of Thompson and Warburton (1985a, b) who suggest ways of “getting to grips with uncertainty.” The first element in our approach is to accept plural perceptions, plural problem definitions, plural expectations and plural rationalities (Thompson and Warburton 1985a: 123). (Blaikie and Brookfield 1987, 16) Third, Blaikie and Brookfield identified and criticized received wisdoms similar to those listed in the synthesis and conclusion of Chapter 6, such as the overdramatizations of land degradation: What many alarmists would have seen more clearly from a wider perspective was that almost all natural landscapes are being continually modified and parts of them degraded, but this should not necessarily nor automatically engender alarm. Land degradation is a phenomenon not only of our time, and there have occurred “eco-disasters” (as our generation would describe them) over some thousands of years, and even before human use became a serious contributory factor. (Blaikie and Brookfield 1987, xx) Another received wisdom they identified is the supposed “backwardness” of traditional farmers: For a long time it was the fashion to decry the “stupidity” or the “conservatism” or the “uncaring idleness” of farmers [that resisted land management changes], or to stress their “ignorance.” (Blaikie and Brookfield 1987, 34) An alternative explanation for conservatism of land managers is an economic one, and rests on the often-observed risk-aversion behaviour of peasant and other farmers. (Blaikie and Brookfield 1987, 35) Consistent with this, they call for more reflexivity and for a thorough analysis of the forces that drive discourses and knowledge production: It … becomes necessary to examine critically the political, social and economic content of seemingly physical and “apolitical” measures such as
562
the Universal Soil Loss Equation, the “T” factor and erodibility. (Blaikie and Brookfield 1987, xix) Data are not reliable, they are constructed, and considerable attention [has to be] … devoted to their ideological nature, but this does not detract from the necessity to improve techniques of measurement. (Blaikie and Brookfield 1987, 16) In sum, they introduce key concepts of social constructionism and science and technology studies and construct a new method on this base. They call their new approach “political ecology,” which they define in the following way: The phrase “political ecology” combines the concerns of ecology and a broadly defined political economy. Together this encompasses the constantly shifting dialectic between society and land-based resources, and also within classes and groups within society itself. We also derive from political economy a concern with the role of the state. The state commonly tends to lend its power to dominant groups and classes, and thus may reinforce the tendency for accumulation by these dominant groups and marginalization of the losers, through such actions as taxation, food policy, land tenure policy and the allocation of resources. (Blaikie and Brookfield 1987, 17) A key narrative of political ecology is already revealed in this early definition: the marginalization of the poorest by the state or other dominant groups. This narrative was an essential outcome of this dissertation, for example in Chapters 1 and 2 (the marginalization of the Tanala people), in Chapters 5 and 6 (the unequal competition between recent migrants supported by projects and first settlers living in remote areas) and in Chapter 11 (the empowerment of more educated and better off people whose interests match better with those of external actors). Blaikie and Brookfield explain how this social process can lead to land degradation: Land managers can become marginalized through the imposition of taxes, corvee labour and other relations of surplus extraction. The responses they make may be reflected in land use and in investment decisions over the preservation of productivity of their land. Adversity of this sort can produce innovations which raise productivity - to pay for the extraction of surpluses - as well as safeguard future productivity. However, more extreme marginalization, often involving a whole number of readjustments particularly a loss of labour power (through war, conscription or emigration), has frequently led to changes in land use and the inability to
563
keep up longer-term investments in soil and water conservation (e.g., repair of terraces and cleaning of irrigation and drainage ditches). The land then becomes economically marginal and the result is a decline in capability and marginality of the agro-ecosystem. (Blaikie and Brookfield 1987, 22-23) This explanation leads them to identify the social dimension of the vicious circle of land degradation and poverty: If land degradation comes about as a result of either commercial exploitation or socially induced marginalization of land managers, a vicious circle of increasing impoverishment and further marginalization of land and land managers can sometimes result. Hence land degradation is both a result of and a cause of social marginalization. (Blaikie and Brookfield 1987, 23) In this new interpretation of degradation (through social marginalization processes), another received wisdom is put into question, the beneficial effects of markets, consistent with the conclusions of Chapters 5 and 6 and 10: Degradation is … often encapsulated in a web of surplus-extracting relationships, in feudal systems, in colonial/capitalist economic relationships, and almost wherever there is marked inequality. Neither centrally planned nor 'market' economies are free from such relationships. (Blaikie and Brookfield 1987, 241) Market opportunities add to population pressures and surplus extraction as a strong additional force which puts pressure upon resources. In many cases, locally enforced authority has been broken down. The state is forced to step in to protect resources, not only from pressured small farmers and pastoralists, but from contractors and opportunist farmers. The state is pulled into a vacuum created by pressures of the market, and the dissolution of informal, traditional, local authority which the market and capitalist relations have helped to dissolve. (Blaikie and Brookfield 1987, 244) Blaikie and Brookfield (1987), however, agree that market is a reality that cannot be avoided: Already the market is an essential element in the reproduction of peasantries throughout the developing world, the strains (Bernstein 1979). Also, despite these strains, most of the world's peasantry would not wish to withdraw from the market. (Blaikie and Brookfield 1987, 244-245) Eventually, Blaikie and Brookfield (1987) propose pathways to remedy land degradation. First, they put an emphasis on the necessity of investments to break the
564
vicious circle of marginalization and land degradation, which is consistent with my outcomes of Chapters 5 and 7: “Sustainable utilization” of land is both a technical and political economic concept and it can be achieved in a number of different ways, for example, with high inputs of chemical fertilizer and considerable earthworks, or substituting labour inputs and locally derived nutrients. Therefore, we argue for the necessity of including the future prospects of land managers for access to land, capital, extension, state assistance and so on, in any assessment of sustainable production. (Blaikie and Brookfield 1987, 242) But they call for cautiousness concerning certain modes of investments, such as subsidies in the form of food for work activities, which could divert farmers from their own strategies and lead them to dead ends. I indeed applied food for work and advocated for this approach (Chapter 7) but I evoked risks similar to those expressed by Blaikie and Brookfield (1987, 250): More positive steps such as bench-terracing aided by grants or soft loans, which could be quite the wrong technical solution, or even the provision of food-for-work aid to encourage farmers to divert inputs from production into land management may also not provide the answer. The whole complex of employment and income-gaining opportunity is involved, and while good management of the land can be achieved ultimately only at the interface where a spade or plough is dug into the soil, the whole of society and economy are involved in determining what will and what will not work. Despite this emphasis on economic solutions, they still consider that mismanagement can explain land degradation. They debate the classic Malthusian and Boserupian hypotheses: We adopt an open approach to the relation of population pressure to land degradation. Degradation can occur under rising PPR, 339 under declining PPR, and without PPR. We do not accept that population pressure leads inevitably to land degradation, even though it may almost inevitably lead to extreme poverty when it occurs in underdeveloped, mainly rural, countries.… PPR is something that can operate on both sides, contributing to degradation, and aiding management and repair. (Blaikie and Brookfield 1987, 34)
339
Population pressure on resources.
565
In the end, they conclude that land degradation is a complex process which has to be addressed with open-mindedness: There seem to be no short-cuts to an explanation of land degradation. Simply we must accept that explanation will be difficult and costly for policy-makers to achieve as a first step to effective action. (Blaikie and Brookfield 1987, 240). But they also acknowledge that the poorest pay the price of coercion, an argument which I use to justify compensation (Chapter 11, and soon, Chapter 15): Coercion of the kind which limits or denies access to resources is the most delicate aspect of intervention, though it has often been handled so crudely as to be wholly counter-productive, or has failed because of lack of will to enforce regulations.… Yet coercion may be necessary in cases where failure to protect remaining resources would lead to rapid overuse and degradation. There are no instant cornucopias awaiting even the most egalitarian reforms in such matters. Undue inequalities in access to resources lead to poaching and expensive protection, and simply diverts pressure on to resources elsewhere. Draconian measures by the state may be necessary but seldom work to the advantage of the poorer local people. (Blaikie and Brookfield 1987, 245-246) In conclusion, there are many similarities between my work and that of Blaikie and Brookfield (1987). My statements (the failure of environmental policies in Madagascar) and my hypothesis (the mismatch between representation and realities) are similar to their introductory statement (the failure of land conservation programs) and their explanation for this failure (the lack of comprehensiveness and interdisciplinary work, the confusions between representations). Their methodological recommendations (the analysis of a chain of explanation, the demystification of measurements, the deconstruction of received wisdoms) and their outcomes (the marginalization thesis, the perverse effects of markets, the importance of investments, the nonirreversibility of land degradation, the nonrejection of coercion) are also found in my dissertation. A summary of my own work would be indeed very similar to this summary of their work. As their research was implemented twenty years ago, one can wonder what progress has been made from that time. Did I produce a knowledge that
566
just leads us twenty years backward? The analysis of more recent work in political ecology will help answer this question.
3. Brief overview of political ecology work 3.1. The foundation of the discipline The books from Blaikie and Brookfield are not the only fundamental texts of political ecology. This field has its foundations in a wide range of disciplines such as: geography, anthropology, political sciences, development studies, social sciences, environmental sciences, cultural ecology, political economy, postcolonial studies, science studies and environmental history. These disciplines determined a diversity of schools but also converged to a common statement: understanding ecological processes is not sufficient to address environment issues, because social, cultural and political forces determine the way humans shape their environment. Taking this statement as its starting point, political ecology was designed to apprehend the environment as a complex object resulting from the interactions of ecological and social systems. The term political ecology was used for the first time by Eric R. Wolf (1972) “to signify the study of how power relations mediate human-environment relations” (Biersack 2006, 3). Neo-marxism and the dependency theory (Frank 1969; Wallerstein, 1974) were highly influential on early political ecology research (Biersack 2006). They orientated seminal work such as Nietschmann’s (1973) study of the subsistence ecology of the Miskito Indians of eastern Nicaragua; Watt’s (1983) analysis of the relationships between famine and the penetration of market economy in Nigeria; the Blaikie and Brookfield books (Blaikie 1985; Blaikie and Brookfield 1987); Little and Horowitz’s (1987) edited book Lands at Risk in the Third World: Local Perspectives; and Schmink and Wood’s (1984; 1992) Political Ecology of
567
Amazonia, which described how indigenous Amazonian production systems were disrupted by the demand created by the penetration of capitalism (Biersack 2006). These case studies focused on the links between unsustainable extraction of natural resources and relations of production. They all contributed to the development of the marginalization narrative. Beyond the social relations of production, political ecologists identified another cause of inequity in access to resources: the unequal distribution of power and the use of knowledge as a form of power. The influence of postmodernism has been a determinant of this new stance. The postmodern critique attacked “theories that read the trajectory of history in terms of an idea of progress” (Biersack 1987, 4). It deconstructed the modernization and industrialism paradigm, as we saw in Chapter 8, within which Marxism was embedded. Postmodernism also rejected the “dualistic view that nature existed outside the human realm” (Biersack 1987, 4). These differences resulted in the development of diverging schools. For those who adopted a postmodern standpoint, “political ecology … recognizes that reality, insofar as it is invested with meaning, is produced “discursively” through signifying practices of various sorts” (Biersack 2006, 4). The focus is put “upon the reciprocal impacts of nature and culture, using such terms as second, social, or humanized nature to signify a nature that is the by-product of human conceptualizations, activities and regulations – a nature, as it were, that is after nature (Escobar 1999)” (Biersack 2006, 4). Their view is consistent with Latour’s (1987) assertion that nature is the consequence of the settlement of controversies, not the cause (Chapter 9). The writing of Michel Foucault and Bruno Latour’s early work are highly influential on these schools. Stott and Sullivan’s (2000) political ecology is one of the most radical forms of this postmodern political ecology.
568
Others reject the relativist epistemologies which result from postmodernism and social constructionism. They see the environment “as having an ontological base and a dynamic role as an agent in its own right” (Zimmerer and Basset 2003b, 3). Nature’s agency is recognized but is regarded as “socially mediated” (Zimmerer and Basset 2003b, 3). Rather than on the deconstruction of discourses, the emphasis is put on the necessity to integrate various temporal and spatial scales as a way to favor “linkages and fusion of social and environmental analysis (Brunckhorst and Rollings 1999; Fry 2001)” (Zimmerer and Basset 2003, 18). These schools are influenced by a second period of Latour’s writing, where nature is given agency in order to escape from the relativist trap of postmodernism, as I will show in Chapter 14. This affiliation is, however, not always clear. In some writing, political ecologists rather seem to adopt a form of critical realism similar to the model I proposed in Chapter 9 and employed in this dissertation. If I was myself a political ecologist, I would consider that I belong to a trend influenced by critical realism, rejecting the radical forms of postmodernism and social constructionism as well as conceptions that “socialize” nature by granting agency to it. I will come back to these issues in Chapter 14. The existence of these two poles and the ambiguity of the second pole concerning its epistemological frame show that political ecology is still in a process of maturation. The main challenge of the discipline may be “to integrate realist biophysical prediction with social and political construction” (Forsyth 2003, 2). Two fundamental questions are indeed raised by the confrontation between the two schools: what is nature and what value can be given to it? I will elaborate more on this issue in Chapter 14. Despite this divergence, political ecologists all agree that environmental discourses are populated by false assertions, or received wisdom, which Kull (2004, 26) defines in the following way:
569
An environmental received wisdom is an idea that is held as correct by the politically powerful establishment, despite potential flaws and weaknesses, and which shapes the discursive field within which discussions of environmental change and resource management occur (Leach and Mearn 1996). Received wisdoms have also been termed “dominant narratives” (Roe 1991) or “Environmental orthodoxies” (Batterbury et al. 1997).… [They] often persist despite inconclusive evidence (Beinart 1996; Swift 1996, Head 2000), [because] they are persuasive, simple stories (Roe 1991) that serve the purposes of powerful groups (Bergeret 1993, Leach and Mearns 1996; Batterbury et al. 1997; Blaikie 1999), and because they are institutionalized into state bureaucracies (Espeland 1998). They serve to frame problems such that the answer justifies the actions of the interested parties (Ferguson 1994; Escobar 1995) and such have become a key element of neo-liberal development discourse (Basset and Koli Bi 2000). The concept of received wisdom matches quite well with what I called a strong statement or a taboo using the language of Foucault (1970), or a strong rhetoric using the words of Latour (1987) (Chapter 9). Having this concept at its core, political ecology led the environmental debate far beyond mainstream environmentalism. The question that remains to ask, now, is until what point the debate must go. Where does the deconstruction have to stop, what must remain after this deconstruction and who can decide what will remain. These questions force political ecology to solve the epistemological problem which determined the divergence of its schools. Chapter 14, which will push farther the epistemological model proposed in Chapter 9, will attempt to provide a contribution on this point.
3.2. The key narratives In this section, I will present more at length the key narratives developed by political ecologists, which were already apparent in the work of Blaikie and Brookfield (1987). In order to reflect the present state of the discipline, I will utilize a recent introduction to political ecology by Robbins (2004). Robbins (2004) presents four narratives. The first one is the marginalization thesis, which was central in this dissertation and was already evoked in Section 1. The
570
second narrative is the control thesis, with which the reader of this dissertation is also familiar but which will be now more precisely formulated. These two narratives are intimately intertwined and their interactions determine two other narratives: the environmental conflict thesis and the environmental identity and social movement thesis.
3.2.1. The marginalization thesis The marginalization thesis asserts that otherwise environmentally innocuous local production systems undergo transition to overexploitation of natural resources on which they depend as a response to state development intervention and/or increasing integration in regional and global markets. This may lead to increasing poverty and, cyclically, increasing overexploitation. (Robbins 2004, 131) This perspective improves the simplistic Malthusian model of a vicious circle between poverty and natural resources degradation. It adds a social mechanism: the weakest social groups are moved toward the margins by the most powerful ones. For this reason, the primary accumulation that could break the vicious circle, leading to more intensive systems through a Boserupian dynamic, does not occur. Rather than a lack of knowledge to improve resource management, the crisis could be explained by this social dynamic. Classic case studies of the marginalization thesis are, according to Robbins (2004), the analysis of deforestation in the Amazon by Hecht and Cockburn (1989) and the study of contract agriculture in the Caribbean by Grossman (1998). There is something trivial in the marginalization thesis. In all societies, there is social differentiation and stronger groups dominate weaker ones. It could not be otherwise. But what matters for political ecologists is the extent to which marginalization happens, the way this issue is addressed by policy makers and the environmental and social consequences. These can differ widely from one place to the other. For example, the majority of farmers lost in the competition for intensification
571
and abandoned their land throughout the history of agrarian systems in Europe. They moved to cities and provided a labor force for the industrial revolution. For this reason, social and environmental problems were mostly urban. In Madagascar, cities are already overwhelmed by poverty and cannot provide sufficient job opportunities. The farmers’ remaining option is to move to areas where there is less competition, i.e., to the forests, where land is perceived as being in free access or where this access is regulated by flexible customs. The marginalization thesis thus provides an explanation for environmental degradation. It can be used for the advocacy of human rights, but also for designing wise conservation policies.
3.2.2. The control thesis The control thesis mirrors the marginalization thesis. It is the consequence of the fact that marginalized people can only survive by tapping into natural resources. When such a situation occurs, states pay attention to the degradation of resources while the social dynamic of marginalization remains overlooked. The groups who are the cause of this marginalization are not ostracized whereas those who suffer from marginalization are regarded as the cause of environmental destruction. For this reason, they are targeted by control and repression. This is done by closing the frontier and by repressing practices that affect the environment, with negative consequences on their livelihoods. This occurred, for example, to the Tanala people of Ranomafana, when the national park was created (Chapter 2), and to the people of Ambodilaingo, when the state put a ban on tavy in 2002 (Chapter 5). In sum: “Control of resources and landscapes has been wrested from local producers or producer groups (by class, gender, or ethnicity) through the implementation of efforts to preserve “sustainability,” “community,” or “nature.” In the process, officials and global interests seeking to preserve the “environment” have disabled local systems of livelihood, production, and socio-political organization.” (Robbins 2004, 149-150)
572
Classic works about the control thesis, according to Robbins (2004), are St Martin’s (2001) analysis of New-England fisheries regulation, Kull’s (2004) analysis of antifire policies in Madagascar (Chapter 12) and Peluso’s (1992) analyses of colonial and contemporary forestry in Indonesia and India. The mechanisms to achieve control are more subtle than the simple state coercion paradigm inherited from the Yellowstone National Park, where militaries managed the land until a park service was created, in 1916 (Robbins 2004, 150). “Rather, efforts center on extending the discretionary conservation power of the state by causing individuals and social groups to “internalize” the coercive missions of the government, creating self-enforcing coercion” (Robbins 2004, 149-150). Bryant (2002) showed, in the case of the Philippines, how NGOs who claim to represent marginal communities extended the power of the state and “served to make certain state goals, like territorialization of protected areas and control of tribal groups, the internal goal of local opposition” (Robbins 2004, 150-151). The implementation of the GELOSE legislation in Madagascar is another illustration of this logic (Chapter 11). The control thesis has been criticized for being an oversimplification. Sivaramakrishnan (1996) observed, in a study of colonial forestry in Bengal, that “forestry field officers, observing and struggling with local practice, soon came into conflict with ecological experts in distant offices.” Finally, “local and state knowledge mixed and produced new outcomes in an internally divided conservation bureaucracy (Sivaramakrishnan 1996; 2000)” (Robbins 2004, 164). This analysis is consistent with Muttenzer’s (2006) optimism about the social impact of the GELOSE legislation in Madagascar (Chapter 11). Other authors (Agrawal 1995; Rangan 1997; Dove 2000) criticized the control thesis for its “normative distrust … of state science and expertise and its artificial conceptual separation from local knowledge and practice” (Robbins 2004, 164), which improved “neither the equity of environmental management nor its
573
ecological effectiveness” (Robbins 2004, 164). These diverging opinions probably reflect a variety of contexts, rendering necessary to consider both the control thesis and its criticisms, as a way to account for the diversity of situations.
3.2.3. The environmental conflict thesis According to the environmental conflict thesis, environmental conflicts would arise from a scarcity induced by the unequal appropriation of resources: Increasing scarcities produced through resource enclosure or appropriation by state authorities, private firms, or social elites accelerate conflict between groups (gender, class or ethnicity). Similarly, environmental problems become “politicized” when local groups (gender, class, or ethnicity) secure control of collective resources at the expense of others by leveraging management interventions by development authorities, state agents, or private firms. (Robbins 2004, 174) The environmental conflict thesis must be contrasted with its counterpart in mainstream environmentalism: the environmental scarcity thesis. This thesis was developed by Homer-Dixon (1998), whose work was highly influential in policy maker circles (Hartman 2001; Peluso and Watts 2001). For this author, the depletion and degradation of a resource are a function of the physical vulnerability of the resource, the size of the resource consuming population, and the technologies and practices this population uses in its consumption behavior. (Homer-Dixon 1998, 14) [They produce] a decrease in total resource supply or, in other words, a decrease in the size of the total resource “pie.” But population growth and changes in consumption behavior can also cause greater scarcity by boosting the demand for a resource.… In many countries, resource availability is being squeezed by both these supply and demand pressures.” (Homer-Dixon 1998, 15) This model is mostly influenced by the Malthusian survivalist discourse (Chapter 8). Environmental conflicts are seen as the consequence of disequilibrium between resource supply and demand, itself due to population increase. A pie whose size is fixed once for all is cut into smaller and smaller pieces until those who need the
574
resources use military force to appropriate more than their fair share. Homer-Dixon, however, recognizes that the unequal distribution of resources can also be a cause of conflict: Scarcity is often caused by a severe imbalance in the distribution of wealth and power that results in some groups in a society getting disproportionately large slices of the resource pie, whereas, other get slices that are too small to sustain their livelihood. Such unequal distribution – or what I call structural scarcity – is a key factor of virtually every case of scarcity contributing to conflict. (Homer-Dixon 1998, 15) In order to better explain these conflicts, Homer-Dixon develops the concepts of “ecological marginalization” (1998, 16) and “resource capture” (1998, 15), which present some similarities with the marginalization thesis and the control thesis of political ecologists. He evokes a “structural scarcity” that would be their consequence. But this does not divert his model from an excessive focus on biophysical aspects (the metaphor of the pie size and pie shares) and simple economic mechanisms (the supply and demand equilibrium). Since he sees marginalization and resource capture as a consequence of scarcity, rather than as a cause, he minimizes the importance of political economy issues. In the end, he claims that the maximization of social and technical “ingenuity,” which he defines as “ideas applied to solve practical technical and social problems” (Homer-Dixon 1998, 109), is the key to solve environmental problems. This view is Malthusian with no ambiguity: technology is the cause of changes and more technology is expected to lead to better resource management, and possibly population increase. Technology is the independent variable while population growth is the dependant variable, to the contrary of the Boserupian model where the relation is reversed. Ingenuity invokes the idea that something new has to be imagined or created. This diverts attention from endogenous technological innovations which could be the consequence, not the cause, of environmental change, as we saw in Chapter 5. As a result, existing solutions are not applied, as occurred in Beforona
575
(Chapter 6). In the end, the political economy of environmental degradation is not addressed and resources capture and ecological marginalization are seen as the consequence of a lack of ingenuity. Homer-Dixon, however, evokes what he calls social ingenuity to account for political economy issues. But he uses this concept only marginally. Peluso and Watts (2001), Hartmann (2001) and Fairhead (2001) formulated strong criticisms of Homer-Dixon’s work and of the “green warfare” discourse which it supports. Fairhead (2001) shows that Homer-Dixon’s thesis conflates environmental degradation, population growth and unequal social distribution of resources in one single concept, environmental scarcity, whereas it is “the very relationships [between these concepts] that are interesting to pull apart and explore” (Fairhead 2001, 219). 340 A first consequence of this conflation is that environmental degradation and resource scarcity end in being regarded as synonyms, whereas the reality is different and more complex. Resource scarcity, population growth and livelihood decline do not necessarily lead to environmental degradation, as shown by Murton (1997), Tiffen et al. (1994) and Lindblade et al. (1998, in Fairhead 2001, 218). Environmental recovery, a concept already evoked in this dissertation and which challenges simplistic Malthusian analysis (see the Boserupian hypothesis in Chapter 5), illustrates cases where resource scarcity leads to improving the state of the environment. A second consequence of this conflation is that it leads to overlooking global political economy issues (as I already mentioned above) and strengthens the received wisdom that land degradation is caused by mismanagement by local communities. It “localizes the blame” (Hartmann 2001, 50) on the poorest households:
340
Fairhead (2001) points out that even those who developed the green warfare discourse (for example, Benner 1991) recognize the importance of nonenvironmental factors such as “competition for resources, … government partiality in resource allocation, … differential political voices, and availability of arms” (Fairhead 2001, 217), to explain conflicts over resources.
576
Deploying environmental arguments, whether fortuitously or strategically, can play to the advantage of certain protagonists in … conflicts on the international stage because it deflects analysis away from international commercial concerns and national politics toward those who are environmentally marginalized. Environmental arguments can thus be wielded as political weapons on behalf on one side or another. (Fairhead 2002, 235) In short, “ “greenwar” and “environmental scarcity” approaches have a tendency to confuse, not clarify, the origins of conflicts. [They] put conservation on the peace agenda and on an equal footing with allocation issues” (Fairhead 2001, 220). Environmental mitigation becomes a central issue for national security, while the deeper causes of conflicts are overlooked and depoliticized. This may explain why Homer-Dixon’s thesis has been so influential in policy maker circles in the United States. In contrast with Homer-Dixon’s (1998) environmental scarcity thesis, a political ecology of environmental conflicts, such as the one elaborated by Peluso and Watts (2001), sees “violence as a site-specific phenomenon rooted in local histories and social relations yet connected to larger processes of material transformation and power relations” (Peluso and Watts 2001, 5). In this new frame, scarcity is no longer the core concept. Conflicts would be “less generated by resource poverty and bankruptcy than by resource value and wealth” (Fairhead 2001, 214). What matters is “the entitlements by which differentiated individuals, households, and communities possess or gain access to resources within a structured political economy” (Peluso and Watts 2001, 5). It is then necessary “to understand the changing contexts of nature transformation, who performs the labor, who bears the burdens, and how its benefits are claimed, distributed, and contested” (Peluso and Watts 2001, 5). All forms of violence, symbolic or physical, can be addressed in this frame. The technologies of power described by Foucault (1975), whose purpose is “to regulate, normalize, and
577
discipline – to forge docile bodies and subjects” (Peluso and Watts 2001, 6) are regarded as forms of violence. In conclusion, a political ecology of environmental conflicts, such as Peluso and Watts’ (2001) analysis of “violent environments,” investigates two realms of political economy. First are “the social relations of production,” which comprehend “the patterns and regimes of accumulation,” “the forms of access to and control over resources,” and “the actors (firms, workers, peasant, state operatives)” (Peluso and Watts 2001, 29). Second are “the social fields of power,” which can encompass “both physical and symbolic, organized and disorganized, state sponsored and “civil,” and highly mixed forms of violence” (Peluso and Watts 2001, 29). The articulation of this model with the marginalization and the control theses is clear. We recognize also some situations described in this dissertation. Ambodilaingo (Chapter 5) is a violent environment where marginalized farmers struggle to achieve primary resource accumulation. They clear forest land, dig canals, terrace bottom lands and search for favorable spots to plant banana trees, with the fear that more competitive groups will achieve this accumulation faster and take their land, for example if a road is built. The violence is also symbolic. It takes the form of messages broadcast on radios for discouraging farmers from the practice of tavy and for informing them of the arrests of those who violated the ban on this land use. This symbolic violence is a violence on their body as the abandonment of their tavy will reduce their caloric intake and have affect their health. The case of Madagascar further showed that environmental violence may enter in a new phase. An unprecedented event may have occurred recently that is changing the global situation with regard to access to natural resources: the closing of the frontier. Unequal access to land, social marginalization and appropriation of resources have always existed, for as long as humans have occupied the earth. But in the past,
578
the weakest groups always had the possibility to move to the forest frontier, where they found resources in “free access” that could sustain their livelihood. This is no longer possible. The increasing power of states and the implementation of conservation policies has closed access to these last available resources. The pieces of the pie are not smaller, but they have all been taken.
3.2.4. Environmental identity and social movements According to Robbins (2004, 189), changes in the environmental management regimes and environmental conditions have created opportunities or imperatives for local groups to secure and represent themselves politically. Such movements often represent a new form of political action, since their ecological strands connect disparate groups, across class, ethnicity, and gender. This thesis is a direct consequence of the former ones. Marginalization and control favor the emergence of an “ecological ethnicity” (Robbins 2004, 190), which gathers people for resistance. According to Robbins (2001), important research that illustrates this thesis are the study of agricultural modernization in the Andean highlands by Bebbington (1993; 1996), the analysis of the social consequences of the green revolution in India by Shiva (1991), the study of the Chipko movements in India by Guha (1990) and its criticism by Rangan (2000), the study of wetland agriculture in Peru by Zimmerer (1991), and the study of pastoral communities in Rajasthan by Robbins (1998).
4. The political ecology of deforestation in the Malagasy highlands Some political ecology work dealing with Madagascar has already been cited in this dissertation. Lucy Jarosz (1993), cited in Chapter 1, belongs to that field, as well as Kull (2004), whose work provided the base of Chapter 12 about fire discourses. The works of Ghimire (1994), Hanson (1996), Peters (1998; 1999) and
579
Harper (2002), cited in Chapter 2, also present some affinities with political ecology. In this section, I will review the recent works of Simsik (2002) and Klein (2004), which explicitly affiliate to this discipline and illustrate a new trend of environmental discourses which I already evoked in Chapter 12. Simsik (2002) analyzed the implementation of conservation policies in the Ambohitantely reserve, which protects one the last forest patches in the Malagasy highland. 341 He used a combination of concepts and methods of political ecology (mostly the work of Bryant and Bailey 1997) and science and technology studies (Latour 1993; 1999). His outcomes are similar to those of this dissertation: tensions between local and external actors, the unequal distribution and forms of power, and the influence of conventional wisdoms are the main causes of conservation failure: Local actors are frustrated by state sponsored conservation programs that unfairly and simultaneously victimize them, while criminalizing their traditional livelihood activities. These local actors are blamed for deforestation and then penalized by having their traditional lands taken away from them. These lands are given “protected area” status by the state yet made accessible to exploitation by powerful interests with the collaboration of state and IENGO resource managers. As one villager explained, “We work hard to collaborate with the park service on the protection of this reserve, yet we notice illegal cutting in the reserve, and the park officials turn a blind eye to it and instead, (state officials) end up turning a more critical eye on us (the local community.”(Simsik 2002, 237) Igniting fire is then viewed as the only way to resist, an argument also supported by Kull (2004) and illustrated by the following anecdote: “n one situation, 90 kilometers east of Ankazobe, people living near the Andranomay Forest, cleared twenty hectares of forest just after a team of research biologists from the World Wildlife Fund (WWF) arrived there to conduct an inventory of forest resources. When later asked by astonished WWF officials about why this had occurred, it became apparent that the local people equated the presence of foreign researchers (like the 341
This forest is one of the last forest patches of the highland forest. It is located close to Ankazobe, northwest from Antananarivo. See Langrand (2003), and Ratsirarson et al. (2003) for more details about this reserve.
580
colonialists before them) as the precursor to loss of natural resource access. Thus “local people cut and burned the forest to show land occupation, and as a consequence, appropriation” (Rakotondravony and Goodman 1998, 99). (Simsik 2002, 238) Simsik identified a series of conventional wisdoms that would explain these unsuccessful policies. They are quite similar to those identified in the end of Part III, in the case of Beforona. They are, for example, the stigmatization of tavy, the vicious circle between land degradation and poverty, the supposed ignorance of farmers, the belief that humans act against nature, and the belief that decentralization, privatization and individual land entitling would encourage better stewardship. Simsik (2002), however, pushes the deconstruction farther than I did: An important conventional wisdom underlying much of the biodiversity conservation efforts taking place today in Madagascar is grounded in the anthropocentric-ecocentric debate pitting humankind against nature. The implementation of conservation activities designed with this conventional wisdom in mind is guided by notions of Malagasy farmers and herders engaged in perpetual conflict against biodiversity conservation. Some Malagasy intellectuals have questioned the rationale for this problematization of humans against nature. They argue that such rationale is based in part on colonial perceptions of Malagasy farmers being both backward and lazy, using fire as an easy way to clear land (Esoavelomandroso 1988, 183). Another misconception views Malagasy as predisposed to pyromania (Randriamantena 342 in Rakotondravony and Goodman 1998, 99). (Simsik 2002, 239) This stance significantly differs from my conclusion of Part III and of Chapters 10 and 11. My objection to the work of Simsik (2002) are indeed similar to those I formulated with regard to the work of Kull (2004) in Chapter 12, and I apologize for this redundancy. I agree with Simsik (2002) that the history of the Malagasy environment has been over-dramaticized, that Malagasy farmers have been blamed for behaviors that are legitimate and that deforestation has to be addressed in a wider political economy frame. But I contend that the “perpetual conflict against biodiversity conservation” that Simsik (2002) seems to challenge is a reality and that this fact is 342
Reference not found. Not in bibliography.
581
even trivial. Malagasy customs clearly show a respect for the forests that shelter biodiversity. This is expressed by the rituals conducted by traditional chiefs when new tracts of forest are cleared: forgiveness is asked for from the spirits of the ancestors and from the animals and plants that live in the forest, before the vegetation is cleared. But the conflict exists, however, in terms of contradictory interests. The fact that communities ask for forgiveness from the forest is even a revelator of this conflict. Klein (2002) developed a similar counter-narrative to the dominant discourse about deforestation. Like Simsik, he argues that the deforestation rate has been exaggerated. 343 He challenges the idea that Madagascar is a “frightening example of the consequences of deforestation,” with its “red soils and erosion gullies” (Klein 2002, 191), and that the country would have been “totally forested by the time of human arrival” (Klein 2002, 191). He contends that these exaggerations serve to justify the hegemony of “centrally governed environmental policy based on Western expertise and Western ideas about the conservation of nature” (Klein 2002, 191). He also evokes “informational indolence” (Klein 2002, 196) to explain the slow transfer of new scientific outcomes and concepts which could debunk these received wisdoms. But Klein weakens his arguments by targeting the wrong discourse. He deconstructs a narrative that is still present in decision making arenas, but which is no longer accepted by conservation biologists. The myth of an original forest cover has now been abandoned in the light of the latest archeology and palynology results (Burney 1987a; Burney 1997; Burney 2003; Wright and Rakotoarisoa 2003). These works show that the Malagasy highlands used to be a mosaic of dense forests, woodlands, savannahs and grasslands maintained in this state by a natural fire regime
343
Klein (2002) provides the following examples from the aid literature: the part of the land that would have been deforested would be 80% according to USAID (1998), 60-85% according to the World Conservation Monitoring Center (1994), 70-75% according to Jenkins (1987) and IUCN and 80% according to the Madagascar Environmental Action Plan (World Bank 1988).
582
and by the presence of large herbivores. The figure of an 80% deforestation rate was thus replaced by the figure of an 80% conversion of primary ecosystems to secondary ecosystems, which puts an end to the myth of a completely forested island but does not refute the fact that the environmental impact of this conversion was huge. The “red soils and erosion gullies” that Klein (2002) refers to may be the consequence of this large scale ecosystem transformation. Klein further minimizes the importance of changes with arguments that lack rigor and may reveal a political bias in his discourse. He argues that as the groups of animals found in Madagascar differ largely from those found on the mainland, the same could occur for plant groups and one should not see as abnormal the low species diversity in grasslands. This comparison is highly disputable because in the case of animals, it is the absence of certain taxonomic groups that is pointed out, whereas in the case of plants, it is the absence of certain ecotypes. Herbaceous species can belong to a wide range of plant families and the fact that this ecotype dominates the landscape does not explain the low biodiversity of the Malagasy highlands. Klein further argues that the increase of grass pollen and charcoal sediments observed by Burney (1987a; 1987b) in the highland, in the seventh and eighth century, could not be explained by human presence, because no archeological traces are found before the twelfth century. This argument is also quite disputable, as a 500 year precision is quite honorable in archeology. According to Wright and Rakotoarisoa (2003, 116) “if the burning was the work of transhumant foragers and herders, such sites would be ephemeral, and their discovery will require an intensive and inspired search.” We saw in Chapter 1 that archeology and the study of charcoal sediments and pollens show, indeed, a clear correlation between human settlements, the frequency of fires and the extension of grasslands in the detriment of ligneous vegetation.
583
In short, Klein’s (2002) discourse does not stand up to the facts and may be as socially constructed as the received wisdoms he challenges. Klein argues that social constructionism must be applied to his own outcomes but contends that his counternarrative is “possibly a more accurate representation of reality than the traditional narrative” (Klein 2002, 198). He may be right if the traditional narrative he refers to was the one developed by Perrier de la Bathie (1921; 1936) and Humbert (1927; 1949). But this narrative has been already debunked in a more rigorous manner by Burney (1987a; 1987b; 1988; 1996; 1997a; 1997b; 1999; 2003). It may now be the wrong target if we are to improve our knowledge about the interaction between environment and human societies in Madagascar.
5. Toward a deconstruction of political ecology I contended in the conclusion to Chapter 9 that the real world is accomplice of our knowledge but is not the only accomplice. The two principles that shape knowledge (the “natural” asymptote, or match with the real, and the “cultural” asymptotes, or match with the self) must be identified. Their area of influence must be delineated and their effects must be distinguished if we are to produce more accurate and comprehensive representations of our world. In this section, I will try to disentangle discursive elements of political ecology and to identify the domains of exclusion of this discipline. Symmetrically, I will attempt to extract from received wisdoms some positive knowledge upon which they were built. This symmetrical deconstruction may help to produce, ultimately, a more inclusive and less politicized theory. Objective knowledge from both sides will be brought back to the surface while received wisdoms or biased representations will be debunked, on both sides. Mainstream ecology could then be rearticulated within a larger theory that would also integrate political ecology outcomes, and conversely.
584
For not going beyond the scope of the subject of this dissertation, I will limit this endeavor to the analysis of an ecological concept widely invoked by political ecologists addressing deforestation issues: nonequilibrium ecology.
5.1. From equilibrium to nonequilibrium ecology? Political ecologists often refer to nonequilibrium ecology to justify an equal valuation of anthropogenic landscape and primary ecosystems and to legitimate anthropogenic landscapes from an ecological standpoint: Ideas about vegetation community change have evolved from deterministic, equilibrium-based models that saw fire as an outside disturbance to nonlinear, nonequilibrium models that accept fire as simply another factor influencing the development of vegetation communities ... Resource management strategies that emerge from this new ecology necessarily have to be adaptive and opportunistic, not unlike many indigenous approaches … The collapse of a clear, legitimate goal (the climax vegetation) highlights the fact that landscape management is political, to be negotiated between different interest groups as much as the ecological system allows (Zimmerer and Young 1998; Scoones 1999; Basset and Koli Bi 2000; Zimmerer 2000; Leach et al. 2002). (Kull 2004, 28) I will argue, in this section, that nonequilibrium ecology does not justify the transformation of Malagasy primary ecosystems into anthropogenic landscapes. A first argument is that nonequilibrium ecology did not replace equilibrium ecology. It complements it, in the sense that these theories are not exclusive and that equilibrium and nonequilibrium states both exist in ecosystems. A second argument is that nonequilibrium ecology concerns ecological processes that occur at a different level than those that are impeded by frequent disturbance regimes, such as those encountered on anthropogenic landscapes.
585
5.1.1. The complementary character of equilibrium and nonequilibrium ecologies The debate about the equilibrium or nonequilibrium state of ecosystems is not new and still remains open. The "Autogenic model" of Clements (1916) was the first attempt to formalize a determinist conception of vegetation development. This author “believed that the community was literally a “superorganism,” that species were its organs and succession its ontogeny.… He argued that each species had an essential role to play in preparing the way for the next seral stage in the succession toward the equilibrium or “climax” plant community” (Hubbell 2001, 21). Clement (1916) was challenged as early as in 1926, by the "individualistic" model of Gleason (1926), according to which random assemblages made in the early stages are more determinant of the succession than affinities between species. It is true that Gleason’s (1926) model was paid significant attention only in the 1950s, but the confrontation between these two perspectives has remained a central issue in ecology since that time. The models have been criticized and refined by many authors (Tansley 1935; Whittaker 1951, 1956, 1965, 1975; Egler 1954; Odum 1969; Horn 1974; Connel and Slatyer 1977; Shugart 1984). Presently, there is a relative consensus to say that they each focus on one side of a continuum of situations and have to be unified into a single theory that would render account of all these situations. The works of Hubbell (2001) and Chase and Leibold (2003) are recent attempts for this unification, despite the fact that they still emphasize one or the other side of the continuum. Looking closer at the nondeterminist model developed by Hubbell (2001) confirms that the two approaches do not exclude each other. Hubbell’s neutral theory is an extension to nonisland situations of MacArthur and Wilson’s (1963; 1967) island biogeography theory. These authors argued that the number of species present on an island is dependant on the size of the island (which determines an extinction rate) and
586
of the distance between the island and the continent (which determines an immigration rate). For Hubbell, this theory broke away from the conventional neo-Darwinian view of ecological communities as co-adapted assemblages of niche-differentiated species residing at or near adaptive and demographic equilibrium. In its place it erected a brave new world in which ecological communities are seen as in turmoil, in perpetual taxonomic nonequilibrium, undergoing continual endogenous change and species turnover through repeated immigration and local extinctions. (Hubbell 2001, 14) From MacArthur and Wilson’s model, Hubbell developed a neutral ecological theory where all individuals in a given community would have equal probability to give birth, to die, to migrate and to speciate. When an individual would die, it would create a gap that would be colonized by another individual. All potential colonizers would have the same chance to fill the gap, leading to stochastic changes in the community composition, as occurs on an island through migration. The composition of the assemblage changes permanently, which can be regarded as an absence of equilibrium. Hubbell calls the plant community modification that results from this mechanism “ecological drift.” He does not deny the effect of biotic interactions (expressed in the existence of niches) but he minimizes their importance. MacArthur (1970), one of the authors of the island biogeography theory, was also a pioneer of the niche theory, which is the base of determinist equilibrium models that Hubbell challenges. Some authors (Schoener 1989; Loreau and Mouquet 1999) consider this a paradox. But the niche theory, according to which affinities between species contribute to determine assemblages, does not, indeed, lead to refute the island biogeography model. Among the species that migrate on an island, some may succeed to establish because they would find appropriate niches. Others, to the contrary, may fail because they would not fit with the available niches. The term of the mathematical equation that is believed to be a migration rate would then be an aggregate of migration rate and successful establishment rate. If the second term were constant, no
587
trace would remain of the species that failed to establish. In the same manner, the extinction rate can be influenced by the strength of the ties between species in island communities. Some links can be stronger than others but at the scale of the island, the strength of the average link can be a constant, rendering the niche effect invisible. Rather than contradicting it, the niche theory may even improve the island biogeography model. It explains for example why less specialized species, such as rats, colonize free habitats first. It is worth questioning why the unification of niche assemblage theories (which explain equilibrium states) and random assemblage theories (which explain nonequilibrium states) did not occur. For Chase and Leibold (2003, 180) ecologists are increasingly becoming specialists, not only within the subdisciplines of evolutionary, population, community, and ecosystems ecology, but also on the tools they use and the type of systems they study. A synthetic framework is critically needed to hold all this together. Chase and Leibold (2003) note that population genetics already unified genetic drift and natural selection into a single theory. They contend that ecological drift and the niche theory are the reflection of these population genetic concepts in the field of ecology, and that they need to be unified in the same way. Reductionism and specialization may not be the only explanations of the difficulties to unify theories. Choosing a theory, or a discipline, may be a political act. As Latour (1987) showed (Chapter 9), science works by the building of “machines” that connect laboratories and scientists to large networks of actors. As several networks often compete, strengthening a discipline or doctrine against the others can be a strategy to eliminate competitors, gain more authority, win controversies and channel more support. The ego of researchers can also play an essential role. No one can pretend to dominate all aspects of a comprehensive representation. Opening a system of knowledge to what is outside its boundaries ineluctably leads to losing
588
control over knowledge, and hence authority. Researchers prefer to choose the dimension of the object they can comprehend, one that makes them feel secure given their knowledge, skills and background. Once this choice has been made, they hide the subjectivity of their choice by heralding the objectivity of the methods provided by their discipline and of the resulting outcomes. Moreover, the choice for a certain discipline or doctrine can be subjective. Researchers prefer to adopt the theory that matches better with their personal political engagement. As political ecologists are often committed to putting an end to the hegemony of the Western conception of conservation, it is logical that they adopt the less determinist conception of how ecosystems function. The lesson that can be derived from these insights is that the first step for doing apolitical science would be to stop the battles between schools of thinking and work for the unification of theories, or at least to simultaneously adopt challenging theories and check, in a given reality, which of these theories has the highest explanatory value.
5.1.2. Distinguishing vegetation change and vegetation development My second argument concerns the insufficient distinction made by political ecologists between vegetation change and vegetation development. As I showed in Chapter 8, change is the most basic concept that can be used to describe system dynamics. It can occur in any direction. Development, to the contrary, implies the unfolding of new potentialities according to a program. The concept of secondary succession illustrates development in the case of ecosystems. Due to the existence of this program, vegetation development cannot be totally random, as shown by the capacity of farmers to predict vegetation dynamics in Ambodilaingo (Chapter 5). Soil properties in the initial stage, the seed bank content, and the seed rain are elements of
589
this program. In sum, nonequilibrium models may have more explanatory value with respect to the fluctuations of mature ecosystems while equilibrium models may still have significant explanatory value for describing the early stages of vegetation development, although stochastic events also matter. For this reason, an excessive focus on nonequilibrium ecology may lead to overlooking the importance of secondary succession for building up ecosystems. This oversight may have strong consequences because there must be something qualitatively different between an ecosystem perpetually maintained at juvenile stages by frequent disturbances (as in the case of most anthropogenic landscapes in Madagascar) and an ecosystem that can reach a stage of maturity, whether this mature stage is stable or not. For anyone who empirically observed the early seral stages of vegetation in Madagascar, whether he is a researcher or a farmer, there is no doubt that they represent something qualitatively different from a mature forest and that biodiversity and biomass are significantly lower. For this reason, the Malagasy anthropogenic landscape cannot be equivalent to the nonanthropogenic ecosystems that preceded them, from an ecological perspective. It is important to notice that it is not the disturbance itself that matters. Disturbance is not “good” or “bad,” even for ecologists. What matters is the geographical extension and the frequency of disturbances and its impact on ecosystems. Concerning the geographical extension of disturbance, a mosaic of undisturbed and disturbed vegetation has a very high biodiversity because different disturbance regimes lead to different species assemblages. Concerning the frequency of disturbance, the maximum biodiversity may not be found in the absence of disturbance, but at an intermediary level, as contended by the intermediate disturbance hypothesis (Connell 1978). A moderate human impact, such as shifting cultivation under very low population density, hunting and gathering of forest products, or pasture
590
fires limited in extension and frequency, could increase biodiversity. The calculation remains to be done. The intermediate disturbance hypothesis, however, cannot justify frequent disturbance regime such as those encountered in Madagascar. A time span of several centuries is necessary to lead tropical rain forests from juvenile to mature stages (Hallé et al. 1978), whereas disturbances occur every few years in most anthropogenic Malagasy landscapes.
5.1.3. Relativism and revisionism in social environmentalism In conclusion to this section, less exclusive theoretical frames are necessary to achieve a de-politicization of environmental discourses. Both equilibrium and nonequilibrium ecological models have to be adopted while attempting to understand the dynamic of Malagasy ecosystems and agro-ecosystems. We learnt from Hubbell’s (2001) neutral theory that the future assemblages of species, in the Malagasy landscapes, will be influenced by stochastic events, by a random selection of species in adjacent communities. But among the species that will be drawn, some may establish and some may not. The rate of success in this establishment is positively correlated to the diversity of available niches and negatively correlated to the frequency of disturbance. 344 Overall, anthropogenic landscapes have higher disturbance frequency and more uniformity, which reduces the diversity of available niches. This explains why they have less biodiversity. These facts, which are trivial for ecologists, cannot be refuted by the deconstruction of environmental discourses. They are what has to remain if we are not to fall into the relativist trap and the revisionist representations relativism could lead to. The criticisms I formulated about political ecology are not isolated. They illustrate a recent trend in the literature that has already been applied to the case of 344
Except in case of low frequency disturbance regimes.
591
Madagascar. Kaufmann (2006), following the arguments of Harper (2002) illustrates the difficulty of integrating anthropocentric and biocentric attitudes into anthropology work. This discipline mostly deals with humans but has a branch specialized in the study of primates. As an anthropologist, Kaufmann himself dedicated 20 months to the study of lemurs in the Beza Mahafaly special reserve in southwest Madagascar, and another 20 months “doing ethnographic fieldwork among primarily Mahafaly cattle herders” (Kaufmann 2006, 180). Kaufmann (2006, 180) considers his experience as unfortunately unique and sees “Primatologists placing lemurs first; ethnographers positioning Malagasy people first; and local people, who face enormous economic hardships, putting their children first.” As shown by Harper (2002) in the case of Ranomafana National Park, the two “sides” of academic anthropology at play … had hardened into ideologies of otherness, where fingers of blame and bias pointed at the other when and if they recognized each other at all. It was as if two people, after walking over the same ground, described their trek as occurring in alternative universes. (Kaufmann 2006, 180) Kaufmann (2006) accepts the dramatic figures of deforestation and land degradation described by biologists, such as the estimation by Du Puy and Moat (1996) that 10% of the initial primary vegetation cover remains on the island. He agrees that this figure led to excessively view humans as the enemies of nature (the Cartesian view that nature would have to be protected from culture) and to unwisely harden environmental policies. But he also criticizes the answer to this mainstream discourse, which he qualifies as being “revisionist” (Kaufmann 2006, 187). This revisionist tendency is now becoming quite influential in the development literature. In Madagascar for example, Allain (1997) and Bertrand and Lemalade (2003b, 135) distinguish “clearing tavy” 345 and “cultivating tavy,” 346 which implicitly suggests that
345 346
Translation by Pollini. Original text in French: “tavy défricheur”. Translation by Pollini. Original text in French: “tavy cultivateur”.
592
some tavy farmers do not clear the forest. These authors, however, are very explicit about the fact that “cultivating tavy” also leads to forest clearing, although at a slower pace. They argue that in the areas they investigated, cultivating tavy is “an agrarian system in equilibrium with the natural environment (less than 30%” of the land cleared in one century).” 347 It is true that the situation can be regarded as at equilibrium, in the sense that the pressure on the land is constant and deforestation proportional to population increase. But this type of equilibrium leads inexorably to clearing all forest and cannot be regarded as acceptable from an ecological standpoint. Hence this statement, by conflating together two types of equilibrium (at an agrarian system level and at an ecosystem level) opens the way to revisionist interpretation of human impact on the environment. The ambiguous character of Bertrand and Lemalade’s (2003b) statement further shows that the revisionist attitude is rather a political act which can hardly be sustained by the facts. For this reason, it is rarely made explicitly. It just appears in between more objective statements and is still in a process of maturation. Its influence is, however, already significant. Babin (2004, xxxv), for example, by asserting that “the global trend seems 348 to be toward the shrinking and degradation of tropical forests,” expresses a doubt which may have a strong impact upon those who do not directly experience tropical deforestation. Babin’s (2004) edited book reveals, indeed, an influence of relativist epistemologies similar to the one apparent in the works of Klein (2002) and Simsik (2002). The concepts of primary ecosystems, secondary vegetation, agroforestry systems, and plantation forestry are regarded as equivalent, or close to equivalent, and the distinction between nature and culture, and between economic value and ecological functions, is blurred. The same tendency can be observed in Horta’s (2000) attempts to demystify tropical deforestation, which 347
Translation by Pollini. Original text in French: “C’est un système agraire en équilibre avec le milieu naturel (moins de 30% du terroir défriché en un siècle).” 348 Highlighted by Pollini.
593
overlooks the fact that concerns about deforestation are more provoked by the risk of huge and irreversible biodiversity loss that can result from it than by the reversible destruction of forest cover. The laudable intention of Babin (2004) and Horta (2000) is to put an end to the hegemony of conservation discourses and practices that are harmful to indigenous populations. But their positions may be counter-productive, even in this endeavor, if they cast doubt on or are ambiguous about realities strongly supported by empirical facts, such as the fast pace of primary ecosystem clearing and the significantly lower biodiversity of the agro-ecosystems that replace them. In contrast to these positions, Kaufmann refers to Kottak and Costa (1993) and Bloch (1995) to show that “culture may as easily abuse nature as not” (Kaufmann 2006, 187). He wants to “redress romanticized assumptions of peasants being naturally good stewards of the land” (Kaufmann 2006, 187). He defends eventually a “middle ground position” where “culture and nature are best understood as a synergism and not as a dichotomy (Whitehead 1998), [and] … are not discrete terms but interactive ones, effecting change upon the other, never guaranteeing adaptation or equilibrium (Scoones 1999)” (Kaufmann 2006, 188). This middle ground position would “avoid the tendency to treat Malagasy people living in local environments as either external to conservation, or if they are internal, as either demonised or valorised agents” (Kaufmann 2006, 188). This methodological shift would lead to “listen[ing] to all sides.” and would “introduce us to Malagasy voices who accept their responsibility in degrading their environments but who lack the means to do otherwise (Feeley-Harnik 1995; Bloch 1995; Walsh 2005)” (Kaufmann 2006, 189). This perspective rather matches with the one defended in this dissertation, whereas the preference for nonequilibrium models and the tendency to regard ecosystems and agro-ecosystems as equivalent is also apparent.
594
6. Conclusion The work of political ecologists was, overall, constructive. It debunked a large number of received wisdoms that are still dominant in political circles and in the aid system. It demonstrated that farmers and herders are not to blame for environmental degradation; that blaming them diverts from addressing political economy issues that may explain the fast pace of deforestation during the last decades; and that the consequences are unethical policies and inefficiency, both in terms of rural development and conservation impacts. For all these points, the outcomes of political ecology are consistent with those of my dissertation. But beyond these outcomes, the discipline may still need to mature concerning the way it addresses ecological realities. There may be three narratives of the Malagasy environment challenging each other now. First is the myth of the green forested paradise, which is unfortunately still used to justify political decisions but is no longer accepted by the scientific community. Second is the reconstruction of ancient ecosystems based on recent research outcomes in palynology, paleontology, and archeology. The original Malagasy environment, before human settlement, would be a mosaic of forests, woodlands, savannahs, and grasslands. Hunting fires, pasture fires and the conversion of forest to agricultural land would have transformed it into an anthropogenic landscape with significantly lower biomass and biodiversity, except in areas not yet permanently occupied by humans. Third is a nascent myth of equallandscapes where human actions are seen as just one factor among others causing environmental changes that should not be a cause of great concern. At present, some political ecologists adopt a quite schizophrenic position between the second and third narrative. Empirical facts tend to attract them to the second narrative but political debates may bias them toward the third. Following the arguments of Kauffman (2006) and Muttenzer (2006), I qualify the third narrative of revisionist and contend that
595
future research concerned with the fate of Malagasy societies and their environments should not drift into it. This tendency was not apparent in Blaikie and Brookfield’s (1987) early work Land Degradation and Society. This is why I attached much importance to the presentation of this book in the beginning of this chapter. But political ecology later created its own domains of exclusion, like all disciplines do. Now that the myth of the first narrative and the received wisdoms derived from it have been broken, it may be time to produce a more comprehensive synthesis of the outcomes of social and biophysical sciences. The mistake of mainstream environmental discourses was to regard environmental degradation by Malagasy farmers as illegitimate, to blame marginalized communities and to overlook external factors that accelerate deforestation, such as market booms and resource capture. But the fact that farmers actually clear primary forests, at least at the pace of demographic growth, and that biodiversity is significantly lower once primary ecosystems are removed cannot be denied. Recognizing environmental degradation, however, is not the end of the story. A new landscape, less natural and more cultural, may emerge, leading to the recovery of fertile soils and the partial reconstruction of biomass. Woodland dominated by invasive trees of the genera Ziziphus, Acacia, Grevillea, Eucalyptus, and Pinus tend to partially replace the forests that have been cleared. In short, an environmental recovery may occur in the future. But this recovery has some limits. In these new landscapes that human societies will reconstruct, the original biodiversity may be lost forever unless it is conserved in some places. For this reason, protected areas are essential. Eventually, the biophysical environment is not the only aspect that will have to be balanced for designing future policies. For this reason, ecologists, in their turn,
596
have to open themselves and make a genuine move toward the social sciences. The transformation of landscapes produces some economic and cultural values that have to be put in balance with biophysical degradation, which raises the issue of nature’s and culture’s values, of the tradeoff between these two interdependent ontological categories. Nature and culture will have to be addressed simultaneously if we are to balance them in the most appropriate way, for our future and for the future of the world we live in. In order to achieve this, it is necessary to move again to epistemological considerations, in Chapter 14.
597
CHAPTER 14 TOWARD NEW EPISTEMOLOGIES
1. Introduction Chapter 13 synthesized the outcomes of my dissertation in the light of political ecology. It showed that the key narratives I developed (the marginalization of poor farmers and the appropriation of their land and resources by more powerful groups) are central in the literature of this discipline. It allowed me to formulate these narratives more precisely and to show that the dynamics I analyzed are far from being specific to Madagascar. Some divergences appeared, however, between my outcomes and those of political ecology. This discipline may be on the verge of producing a new representation that is as much politically or socially biased as the ones it challenges. The objective of this chapter is to propose an epistemological and conceptual framework that will help to adopt a more inclusive stance and to go beyond opposed social constructions. Section 2 will strengthen the epistemological framework proposed in Chapter 9. It will give special attention to Latour (1993) (Section 2.1), who influenced my methodology and is seen in the filigree of much political ecology work. But it will confirm the departure I took from Latour’s epistemology, showing the dead-end toward which it can lead. Actor-network theory, a method of inquiry elaborated with much input from Latour, will illustrate this dead-end (Section 2.2). Popper (1963) is another author to whom I will pay much attention. The model I proposed in Chapter 9 may be indeed quite close to his epistemology. Popper will allow me to refine this model and to give a more formal explanation of the mismatch between representation and realities (Section 2.3 and 2.4). Section 2.5 will introduce the work of Feyerabend
598
(1975; 1994; 2003), whose anarchist theory of science compensates for the limitation that an exclusively Popperian method could lead to. Eventually, Morin’s (1990) principle of dialogy (Section 2.6) will be used to demonstrate that there is no contradiction in framing my work within Popper’s (1963) and Feyerabend’s (1975; 2003) epistemologies simultaneously. In Section 3, I will utilize these outcomes, especially Morin’s principle of dialogy, to render more explicit my position concerning the value of nature and culture and the articulation of these two ontological categories. In the conclusion (Section 4), I will summarize these outcomes and apply them to the subject of this dissertation. It will reframe the questions I treated in Part III and Part IV, as well as the discourses associated with them, in a table structured according to the key concepts of Popper’s epistemology. This table will be the last step before Chapter 15, aimed at proposing new practices for solving the issue of deforestation in Madagascar.
2. Confirming and strengthening an epistemological framework 2.1. Bruno Latour’s parliament of things Latour’s work was highly influential in the social sciences, including political ecology. I used his concepts and tools in my analysis of discourses but showed that a radical interpretation of his early epistemology can lead to a cultural relativism which creates an ontological vacuum. Latour (2006) himself noticed the danger that resulted from a radical interpretation of social constructionism. In this section, I will analyze his more recent work, where he departs from social constructionism and proposes a new way to articulate the natural and the social worlds. In We Have Never Been Modern, Latour (1993) shows that we live surrounded by hybrid objects that associate elements of the social and the physical world, that enmesh nature and culture. Our civilization would create an increasing number of such
599
hybrids but would at the same time separate nature from culture in the representations it produces. It would proceed by creating “two entirely distinct ontological zones: that of human beings on the one hand; that of nonhuman on the other” (Latour 1993, 1011). In the first zone, there would be “a parliament of mutes, the laboratory, where scientists, mere intermediary, spoke all by themselves in the name of things” (Latour 1993, 142). In the second zone, there would be “naked citizens, unable to speak all at once, arranged to have themselves represented by one of their number, the sovereign, a simple intermediary and spokes-person” (Latour 1993, 143). Premodern cultures, on the other hand, would have avoided the proliferation of hybrids by integrating the cultural and the natural world into a single representation. Latour (1993) then questions the reasons for the separation of these two rooms. His answer is that separation offers several guarantees. First, as the scientific room “allows mute objects to speak through the intermediary of loyal and disciplined scientific spokespersons” (Latour 1993, 30), it let us think that humans do not make nature, that “nature always existed and was always already there” (Latour 1993, 30). We would just be “discovering its secrets” (Latour 1993, 30). The second guarantee is that the political room gives a voice to all citizens, whose words are computerized and translated into the decisions of the “sovereign.” In consequence, we have the illusion that “human beings, and only human beings, are the ones who construct society and freely determine their own destiny” (Latour 1993, 30). Latour (1993) then explains that this double separation did not work in practice and that hybrid human/nonhuman objects proliferated. A third guarantee then had to be provided to maintain this constitution: blocking the relationships between the two rooms. With this blocking, the hybrid objects or quasi-objects would be banished. They would continue to exist and proliferate but their hybrid aspect or “hybridity” would not be considered by the two rooms, which would instead “purify” them,
600
extracting either their social or their natural content to render them more legible. As a consequence of this purification, the scientific room can pretend to have more legitimacy to speak for “natural” objects, to know their transcendence and to impose modes of ruling based on these supposed truths. Latour (1999) compares this mechanism with the role played by Plato’s philosopher, who travels from the world of ideas to the cavern of ignorance in order to design perfect ruling and offer happiness to citizens. He rejects this model and contends that as our world is made of hybrid objects, it is necessary to break the separation between the scientific room (the laboratory), which focuses on natural objects, and the political room (the sovereign), which focuses on social objects. Latour’s (1993) analysis provides an explanation to the mismatch between representations and realities, which is the central hypothesis of this dissertation. Real objects (for example a landscape) are hybrids, but the representations produced about these objects are either social (a political system) or natural (an ecosystem), leading to an incapacity to understand reality. The consequence is irrelevant knowledge and inefficient action. Latour’s (1987) toolkit was useful in this dissertation because it allowed me to analyze the purification process that led to these misrepresentations. Latour’s (1993) answer to the separation of the two rooms is to create a new constitution that would associate human and nonhuman objects in the same collective. This collective, or “parliament of things,” would address simultaneously the human, the nonhuman and the discursive dimensions of our world. It would give a representation to the hybrid or quasi objects that were banished. In this manner, nature would contribute to the closing of controversies because nonhuman objects would be represented in the parliament, together with humans. These nonhuman objects would be “actants” of the social process. The third guarantee of the ancient constitution (the purification) would be cancelled and nature and society would not be separated
601
anymore. A mediation would be necessary to link the natural and the social. This mediation would be “the very center of the double power, natural and social” (Latour 1993, 139). This move would, however, not imply returning to a premodern state. For Latour (1993), we need to “keep the moderns’ major innovation: the separability of a nature that no one has constructed – transcendence – and the freedom of manoeuvre of a society that is of our own making – immanence” (Latour 1993, 140). In this way, Latour (1993) pretends to escape from the relativist trap. As he puts it, we can still belong to the enlightenment and believe in the sciences: there are no more naked truths …, no more naked citizens either. The mediators have the whole space to themselves. The enlightenment has a dwelling-place at last. Natures are present, but with their representatives, scientists who speak in their name. Societies are present, but with the objects that have been serving as their ballast from time immemorial. (Latour 1993, 144) Nature’s transcendence, its objectivity, and society’s immanence, its subjectivity, stem from the work of mediation without depending on their separation, contrary to what the constitution of the moderns claims. (Latour 1993, 140) In sum, the parliament is the formalization of networks that always existed (Chapter 9) but were unofficial. These networks are given a power of agency that the ancient constitution gave only to purely human institutions and an objective production of knowledge emerges from this agency. But the new parliament will not need to be physically created. It will just result from the fact that what we always did will be brought back to the surface, into the light, visible (Latour 1993). We have arrived here at a point where I take a different stance from that of Latour (1993), as I did already in Chapter 9. I prefer to contend that what needs to be clarified is the role of the two rooms in the networks. One must be dedicated to the production of knowledge, and we can call it the scientific room, while the other one must be dedicated to the taking of decision, and we can call it the political room.
602
Latour (1993) suggests this idea when he says that scientists speak in the name of nature in the parliament of things, but he remains unclear about the type of relationships that would link scientists to other humans in the parliament. I agree with Latour that the purification process is a paramount problem but think that in spite of this, both the natural and the social dimensions of objects are addressed by the two rooms even in the present situation. Both the “laboratory” and the “sovereign” address the social and the natural dimensions of our world and both are engaged in purification. The challenge, then, rather than bringing the networks back to the surface, would be to clarify the roles of the “laboratory” and the “sovereign” and to transform the cognitive modes they utilize, for a better articulation of the social and the natural on one hand, and of the production of knowledge and decision making on the other hand. Popper’s epistemology, in Sections 2.3 and 2.4, will help to design a method for achieving this. Latour’s (1993) model, by displacing agency to the networks, more precisely to the mediator, further raises several insolvable questions: who is going to be the mediator? Will it be a human, or a group of humans? Can the mediator be something else? How can the mediator lead to the transcendence of nature, to objectivity? Nonhumans have a form of agency but this agency is not of the same kind as in the case of humans. Nonliving objects do not make choices whereas humans and other forms of organisms do make choices, which are the expression of their intentions. Because humans make choices, they can defend particular interests instead of collective interests, and they can make mistakes. The physical world, on the other hand, does not make choices, nor mistakes. It just follows its laws. The consequence is that nature can “speak” objectively by itself while humans remain subjective. As choices, in the end, are always made by humans, the objectivity of nature is always subjectively interpreted by humans. This is why there must be a specially prepared
603
group of humans, the scientists, committed to read the signs of nature and speak for it without having to compromise. Latour (1993) is not clear about whether independence will be provided to scientists in the parliament of things. It is true that scientists, being humans, will still make choices and mistakes, but they will make fewer mistakes if they are well prepared and genuinely committed to what is expected from them. The problem, today, may be that these humans (the scientists) use inappropriate methods, confuse their role with that of politicians, 349 are asked to compromise, which is a totally nonscientific act.350 Hence they fail to speak about the real “nature” of things. The purification, which I recognize as being an essential problem, would be the consequence of these wrong cognitive methods and of this corruption. The following paragraph raises still more questions: At the end of the process, there is indeed a nature that we have not made, and a society that we are free to change; there are indeed indisputable scientific facts, and free citizens, but once they are viewed in a nonmodern light they become the double consequence of a practice that is now visible in its continuity, instead of being, as for the moderns, the remote and opposing causes of an invisible practice that contradicts them. (Latour 1993, 140) How can there be a nature that we have not made if nature is the consequence of these practices? What are these practices? The work of the collective? Of the mediators? Can they be something other than social practices? As I contended in Chapter 9, the consequence of social practices are representations. The real world, or nature, is one of the causes of representations because it can influence social practices (I symbolized this by the right magnet in Figure 15, page 435). It is true that representations can, in turn, change the real world, because the physical actions of humans are motivated by their representations. But this is a second step. The 349
They can play the two roles but must distinguish them (which does not mean to separate them). When a scientist changes his mind, this is not the result of a compromise. He changes his mind only because new facts changed his representation. Other-wise, he would be engaging in politics.
350
604
representations themselves do not directly modify the world out there. If they did, the world could not be one, mistakes would not exist and knowledge would have no meaning. Another objection is that if the social construction of knowledge produces only representations (which is what I contend), then controversies should never be closed and scientific facts should always remain disputable. Nonhuman objects are involved in the parliament of things but cannot create representations by themselves. This is why they cannot have their own agency or, if we call agency the fact that nonhumans interact with the world, it is a different sort of agency from that of humans. It is the consequence of mechanistic laws, not of intention and choices derived from representations. The nonhuman “agency” is totally objective and does not permit mistakes, because the physical world does not make choices. It is hence a neutral form of “agency” and nothing would be changed by conflating together human and nonhuman agency, in the concept of mediator. Nonhumans can only “speak” through the voices of humans who subjectively produce representations by observing them 351 . In sum, my disagreement is similar to the one I expressed in Chapter 9. In the same manner as nature cannot be the outcome of controversies (only the representation of nature can be), mediation or other social practices cannot define the transcendence of nature. Assuming that they could may just lead to the same radical forms of cultural relativism as social constructionism did before. Several truth could be established about the same reality, a process which I called cultural positivism in Chapter 9. To give an example of the potential consequences of this “cultural” or “social positivism,” I need to go back to my empirical materials. The people of Ambodilaingo (Chapter 5) are absent of the networks that decide their fate in the working groups of
351
Agency may, however, be shared with other conscious organisms, such as apes.
605
EP3 (Chapter 8). They resist this by igniting fires that enter into the networks. But controversies about who they are, how they think and how they behave are closed in their absence, the outcomes of these closures being received wisdom. In this dissertation, I used Latour’s (1987) concepts to debunk received wisdoms that result from political, social and cultural biases introduced by these networks. Latour’s parliament of things, by considering that the truth results from the settlement of controversies by these same networks, may end in legitimating these biased representations. Taking a different stance from Latour (1993), I contended that the mistake of the moderns was rather in their cognitive modes. More precisely, I think that this mistake was to pretend that certainty was attainable, which gave excessive power to certain forms of knowledge. The consequence is that certain domains of knowledge became isolated from democratic control and produced hegemonic doctrines. This was the case of Marxism and Liberalism and certain forms of environmentalisms may be the next hegemony to come. The case of the anti-tavy campaign of 2002 (Chapters 5 and 6) showed that this new hegemony is already present in Madagascar. Latour (1993) displaced the deliberation about truth from the scientific room to a larger network that involves not only science, but also other social actors. But Latour still accepts the principle of certainty. He gives to the new constitution the same power that the scientific room had before: to deliberate about the truth; to objectify nature; to close controversies. I contended that this may be indeed a form of return to positivism where the positivity would be constructed by a larger social group instead of being under the control of the scientists alone. This positivity could be as detrimental as the former one because it is not sure whether these social networks would be more objective than science to “establish” truth. Unless nonhumans would have an agency similar to that of humans? It would not be only our epistemology which would be
606
reversed in this case. It would be the whole understanding of the physical and metaphysical dimensions of our world, because physical objects could then make choices.
2.2. Actor-Network Theory Latour, associated with sociologists Michel Callon and John Law (Callon et al. 1986), created a new method aimed at applying his epistemology: actor-network theory (ANT). This method “intended to guide thinking and research about humannature relations” (Castree and MacMillan 2001, 211) and was widely used to study our relations with the environment. I present it briefly here in order to help to understand the impact of Latour’s epistemology when it is put into practice. This section may have some redundancy with the previous one, but this will help to clarify a complex issue whose outcome is crucial for the subject of this dissertation. As social constructionism did before it, ANT challenged “the belief that the natural and the social were ontologically different and distinct” (Castree and MacMillan 2001, 208). But it also claimed that social constructivist, in the end, had “an inability to imagine human-natural relations in a nondichotomous way” (Castree and MacMillan 2001, 210), because “bringing nature within the domain of the social simply shifts the causal and ontological arrows from one “side” of the social-natural dichotomy to the other” (Castree and MacMillan 2001, 210). This assertion is consistent with the idea, developed in Chapter 9, that social constructionism led to the replacement of a first form of positivity, controlled by scientists claiming objectivity, by another one, determined by the social construction of facts. ANT attempted to remedy this problem by encouraging “to imagine the unimaginable by doing away with dualistic categorical thinking” (Castree and MacMillan 2001, 214).
607
The core concepts of ANT clearly match with Latour’s epistemology and borrow several concepts to it. First is the principle of hybridity. The world is seen as made of hybrids of human and nonhuman objects, or “quasi objects.” This “ontology of not-quite-natural, not-quite-social entities rejects the pure transcendences of binarist thinking” (Castree and MacMillan 2001, 211). It encourages to “think relationally” (Castree and MacMillan 2001, 211), which corresponds to Latour’s (1993) concept of mediation. Second is the concept of network, also central in Latour’s (1993) work. ANT asserts that only after the analysis of networks can an explanation emerge, whereas the society-nature dualism provided an a priori explanation. The third principle of ANT is symmetry, equally important in the work of Latour (1993): the causal relations can go in both directions and all objects, human and nonhuman, are qualified as “actants.” This leads to minimizing the ontological distinction between humans and nonhumans: ANT “levels up” nonhumans to the status of humans and “levels down” humans to the status of nonhumans (Laurier and Philo 1999, 1060; in Castree and MacMillan 2001, 213). The social and the natural are “co-constitutive” (Castree and MacMillan 2001, 213) of the reality within complex networks. We recognize here Latour’s (1993) displacement of agency toward the network nodes. Agency is redefined as “a relational effect generated by … interacting components whose activity is constituted in the network” 352 (Whatmore 1999a, 28),” and it is argued that action “does not necessarily require speech or intentionality” (Castree and MacMillan 2001, 213). The fourth core concept of ANT is power, which is regarded as a “decentered” and “shared” capacity, a “relational achievement” which involves “myriad natural actants as much as social ones” (Castree and MacMillan 2001, 214). The three forms 352
As we saw in Chapter 13, this type of assertion is frequent in political ecology, showing the influence of Latour’s epistemology on this discipline.
608
of power which I evoked in this dissertation (coercive power, knowledge as power, and power of acting) are conflated into a single form located in the network nodes. Several comments can be made about ANT. First, it is not clear whether the natural and the social are regarded as ontologically distinct. ANT wants to put an end to “dualistic categorical thinking,” but the concept of hybridity implies duality. If the natural and the social can be subject to hybridization, they must be ontologically distinct. ANT does not deny this but makes assertions, such as the rejection of nature/culture duality, which contradict this principle. It may be that ANT confuses the concepts of duality (the fact that two things are distinct or different) and dichotomy (the fact that they are separated). I would contend, instead, that nature and culture are not separated but that they are distinct. As they are distinct, they are ontologically different and as they are ontologically different, their relations to each other cannot be of the same kind. If their relations are not of the same kind, ANT’s principle of symmetry cannot be valid. By defining agency as “a relational effect” between human and nonhuman, ANT avoids facing these contradictions and fails to address the differences between the cultural (intentional) and natural (nonintentional) forms of “agency.” Second, ANT may still be a form of social constructionism. Castree (2001) presents an ANT case study (the restoration of the River Cole) which illustrates this. He asserts that the identity of the river Cole as “social” or “natural” is established as an outcome of the restoration process, so there can be no question of predetermining whether the restored Cole would be one or other in the manner of the environmental realists or social constructionists. (Castree and MacMillan 2001, 216) ANT is not doing a social construction itself. It remains neutral. But according to the statement in the paragraph above, it considers that the qualities “natural” and “social” are the outcome of the process that it studies. This conception matches with
609
Latour’s (1987) statement that nature is the outcome of the settlement of controversies (Chapter 9). If humans and nonhumans had the same form of agency, then we could agree with ANT and say that the outcome would not be a social construction, or at least that it would not be totally a social construction. But if we considered that this is not the case (which is the position I defend), then there would be no difference between ANT and social-constructionism. As I contended earlier, humans have speech, intentions and tools while nonhumans do not. Both humans and nonhumans have power but these powers are not of the same kind. Nature can have a strong word to “say” in some cases. For example, the Mississippi River never “accepted” the management constraints imposed by humans. But trees do not have the power to refuse to be cut; species cannot refuse extinction if they are physically eliminated. By conflating together different forms of power and agency, ANT may legitimize the social construction of our physical environment, leading to denying the existence of objects that are not present in our representations or are just marginally paid attention. It is not only the marginal nature and the unknown nature of things that would be at peril. It is also the humans whose presence in the networks would be only marginal, such as the people of Ambodilaingo and all those rejected at the margin of our global world and of its representations. Castree and MacMillan identified this problem, but consider that it just concerns social constructionism. As they put it, “in the constructivist argument, political programs designed to protect or exploit nature can only be ever justified in human term” (Castree and MacMillan 2001, 220). “Consequently, natural entities have no voice: they are erased from the political landscape or, at best, spoken for by social actors” (Castree and MacMillan 2001, 220). We can hardly see why the situation would be different in ANT. Nonhuman objects have no voice anyway. They need someone to speak for them, to integrate them into representations which they cannot
610
produce themselves. The conflation of two forms of agency under the concept of “relational” agency may have led to the overlooking of these trivial facts. Castree and MacMillan, however, recognized some of the ontological limitations of ANT which I mentioned. They refer to the work of Laurier and Philo (1999, 1016, in Castree and MacMillan 2001, 221), who expressed concern about ANT’s tendency to conflate things. For these authors there is, ontologically, “the problem of installing a great indifference between the countless things of world… which arises when they end up being portrayed as potentially all the same.” ANT thus becomes a “flattening process” that obscures the differences between “noun chunks” of realities (Laurier and Philo 1999, 1014; in Castree and MacMillan 2001, 221). Castree also refers to Murdoch’s work (1997a, 750, in Castree and MacMillan 2001, 222), which contends that ANT may not be doing anything more than “describe[ing], in prosaic fashion, the dangerous imbroglios that enmesh us.” In consequence, ANT would “remain strangely agnostic about the actor-networks it seeks to describe and explain” (Castree and MacMillan 2001, 222). This would cause a political problem: “advocates of a strong ANT agenda risk ignoring the possibility that some actants “marshal” the power of many others and, in so doing, limit the latter’s agency and circumscribe their existence” (Castree and MacMillan 2001, 222). In sum, ANT does not provide any answer to the interference between power and knowledge, an issue that was so important for the turn toward postmodern epistemologies. Castree and MacMillan’s (2001) answer to this problem is to move to a “weaker” version of ANT that would “remain critical of binarist thinking, of asymmetry, of limited conceptions of agency and of centered conception of power” (Castree and MacMillan 2001, 222), but would recognize that the processes that drives actors are not similar; that the social relations “are often disproportionately directive” (Castree and MacMillan 2001, 222); that agents “vary greatly in their powers to
611
influence others” (Castree and MacMillan 2001, 222); and that “a politics of nature attuned to the needs and rights of both human and natural entities must ultimately be orchestrated through a putatively “social” actor” (Castree and MacMillan 2001, 222223). These recommendations are consistent with the criticisms I formulated above, but it seems that the authors refuse to face the full consequences of their findings. Recognizing “the needs and rights of both human and natural entities” (Castree and MacMillan 2001, 222-223) raises the question of their specific ontological nature, of their distinction, which puts into question the symmetry of their relations. One can question, then, if such a weak ANT would still be ANT at all. In conclusion to this section, Latour attempted to find a way to escape from the relativist trap into which his strong rhetoric led him. But actor-network theory, a method which claims having its roots in the most recent versions of Latour’s epistemology, still seems to be trapped in relativism. I will now introduce the epistemology of Popper, already evoked in the filigree of this dissertation (Chapter 9), which proposes another way to escape from positivism without falling into the relativist trap.
2.3. From Latour’s to Popper’s epistemology Latour’s epistemology and the methods derived from ANT may have failed to solve the problem of how to allow nature to be heard without giving hegemonic power to those who speak or pretend to speak for it. Another strategy may be necessary to achieve this goal. Rather than displacing to a larger collective the production of knowledge, I propose to define in another way what is knowledge, and more importantly, what is scientific knowledge, in order to decrease the power of those claiming they produce objective representations.
612
A simple way to achieve this is to accept uncertainty, i.e., to reject what Bernstein (1983) calls the “Cartesian anxiety”. According to this author, Descartes’ quest for certain knowledge may be more than a device to solve metaphysical and epistemological problems. It is the quest for some fixed point, some stable rock upon which we can secure our lives against the vicissitudes that constantly threaten us. (Bernstein 1983, 18) This quest enables an escape from “the forces of darkness that envelop us with madness, with intellectual and moral chaos” (Bernstein 1983, 18). Accepting to live with uncertainty, to the contrary, implies acceptance that controversies can never be closed, that all facts remain disputable. Latour’s epistemology contradicts this principle, which may explain why this author eventually developed a new form of positivism. As Bernstein (1983, 18) puts it, the Cartesian anxiety “still haunts us and hovers in the background of the controversies waged by objectivists and relativists.” Accepting the Cartesian anxiety would weaken positivist epistemologies. But it would not weaken science because knowledge would have the guarantee of remaining an open system and could be subject to permanent improvements. Nor would uncertainty imply a return to relativism, because relativism occurs when there is 50% uncertainty, whereas an open system of knowledge can function with 1% or less of uncertainty. Uncertainty further enables proposing a new way to articulate the work of the two rooms described by Latour (1993). Uncertainty, logically, should be accepted simultaneously in the domain of knowledge (by scientists) and in the domain of ruling (by politicians). This would prevent the “laboratory” and the “sovereign” from becoming hegemonic. The two rooms would work hand in hand and in this collaboration, they could control each other. Uncertainty is the condition for this mutual control to always remain possible.
613
I will show, now, that the acceptance of uncertainty is central in Karl Popper’s epistemology. Interestingly, Popper (1966) had the same concern as Latour with respect to the totalitarian potential of Plato’s philosophy. Politics of Nature, by Latour (1999) and The Open Society and its Enemies, by Popper (1966) both start with a critical analysis of Plato’s myth of the cavern. But Popper (1963) proposed a different pathway to break this myth, as we will see.
2.4. Karl Popper’s epistemology I find Popper’s epistemology very appealing because it makes possible to take a position that avoids the traps of both relativism and positivism, and because it can be applied to science as well as to political philosophy. The model I proposed in Chapter 9, to illustrate the way knowledge advances and to visualize my criticisms of Latour’s epistemology, was indeed quite Popperian. It legitimated my attempts to deconstruct representations, but was also consistent with the idea that progress in knowledge is possible. Indeed, it enabled me to consider deconstruction as a constructive act. I will try, now, to demonstrate this apparent paradox by introducing the central concepts of Popper’s epistemology.
2.4.1. The falsification principle According to Popper (1963, 48), “the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability.” In other words, a theory is scientific if it can be falsified, i.e., if facts can provide evidence that it is wrong. To the contrary, “a theory which is not refutable by any conceivable event is nonscientific. Irrefutability is not a virtue of a theory (as people often think) but a vice” (Popper 1963, 48).
614
With this falsification principle, Popper departed from the verificationist method, which constructs knowledge by the repeated verification of facts. However, Popper did not reject verification. His intention was rather to put falsification in the first position, whereas verification, which is usually put in the first place by positivist science, would come second. Popper justifies this reversal in the following way: It is easy to obtain confirmations, or verifications, for nearly every theory – if we look for confirmations. (Popper 1963, 47) Confirming evidence should not count except when it is the result of a genuine test of the theory; and this means that it can be presented as a serious but unsuccessful attempt to falsify the theory. (I now speak in such cases of “corroborating evidence”). (Popper 1963, 48) In sum, knowledge mostly advances by refutation. Verification can render a theory more secure, or less uncertain (it corroborates it) but it does not increase its content. Refutation, to the contrary, increases the content of theories because it consists in identifying false statements and in replacing them by new statements that have more chances to be true. Popper, however, accepts the idea that a theory can be saved from refutation by “introducing ad hoc some auxiliary assumption” (Popper 1963, 48). But he considers that this is done at the price of destroying, or at least lowering the theory’s scientific status. Based on the falsification principle, Popper considered certain theories, such as Freud’s psycho-analysis, as nonscientific, because “there was no conceivable human behavior which could contradict them” (Popper 1963, 49). Popper was often criticized for this narrow definition of science where scientific progress would be inseparable of thorough empirical observations. These criticisms, however, may result from a misunderstanding of his epistemology. By saying that Freud’s theory was not scientific, Popper did not mean that it was not meaningful. The problem that Popper tried to solve by proposing the criterion of falsifiability was neither a problem of meaningfulness or significance, nor a problem of truth or
615
acceptability. It was the problem of drawing a line (as well as this can be done) between the statements, or systems of statements, of empirical science, and all other statements. (Popper 1963, 51) Popper called this problem the “problem of demarcation” (Popper 1963, 51). He complained that he was misunderstood and that his detractors interpreted his work as being a theory of meaning. Aside from the realm of science as he defined it using falsification as the main criterion, Popper still recognized the possibility to develop meaningful theories, as shown by the fact that he extended his epistemology to political philosophy, as we will see in Chapter 15.
2.4.2. The problem of induction The main argument that led Popper to develop the falsification principle is his analysis of induction, a concept which is central in verificationist or positivist science. Popper followed Hume’s (1882) argument that induction is not the central process in the construction of scientific theories and can be misleading: There can be no logical arguments allowing us to establish “that those instances, of which we have had no experience, resemble those, of which we have had experience. (Popper 1963, 55, with a quote of Hume [1882]: part III Section vi and xii). In consequence, “theories can never be inferred from observation statements, or rationally justified by them” (Popper 1963, 56). But Popper goes further than Hume. He considers that induction is not working through the repeated observation of similar facts (which Hume thought it was), but through the repeated observation of facts regarded as similar by the observer. They are repetitions only from a certain point of view. (Popper 1963, 59) Instead of explaining our propensity to expect regularities as the result of repetition, [Popper] proposed to explain repetition-for-us as the result of our propensity to expect regularities and to search for them. (Popper 1963, 60) In other words,
616
Without waiting, passively, for repetitions to impress or impose regularities upon us, we actively try to impose regularities on the world. We try to discover similarities in it, and to interpret it in terms of laws invented by us. Without waiting for premises we jump to conclusions. (Popper 1963, 60) There is something Foucaldian in this declaration. This imposition of regularities on the world can be compared to the myth of the “universal mediation” (Chapter 9), according to which “the discourse is not much more than the reflection of a truth that is revealed under its own eyes” 353 (Foucault 1970, 51). Popper accepts all the consequences of this problem. He recognizes that a simple correspondence of theories with facts is impossible; that there is always, and necessarily, an interpretation by the observer; that pure induction is an illusion. Science can thus only advance by conjectures, which implies engaging in subjective and speculative processes which are necessarily social. At this point, Popper’s epistemology can be connected to the “social construction of fact” as analyzed by Latour (1986). But Popper sees the social construction as a bias, as a potential cause of necessary but temporary mistakes. Nature cannot be its outcome. Coming back to the principle of falsification, or refutation, it is the necessary counterpart of these conjectures. Through refutation, the conclusions toward which we subjectively jump can be discarded (Popper 1963). In sum, according to Popper, scientific theories [are] not the digest of observations, but they [are] inventions – conjectures boldly put forward for trial, to be eliminated if they clashed with observations; with observations which [are] rarely accidental but as a rule undertaken with the definite intention of testing a theory by obtaining, if possible, a decisive refutation. (Popper 1963, 61)
353
Translation by Pollini. Original text in French: “le discours n’est guère plus que le miroitement d’une vérité en train de naître a ses propres yeux.”
617
2.4.3. Dogmatism, critical attitude and uncertainty Popper’s falsification principle encourages a critical attitude “which is ready to modify its tenets, which admits doubts and demands tests” (Popper 1963, 65), while rejecting dogmatism, which “expects regularities everywhere and attempts to find them even where there are none” (Popper 1963, 64). For Poper (1963), dogmatic thinking would have been the characteristic of premodern societies, where myths provided the architecture of knowledge and where all observations were explained in the light of these myths. The verificationist attitude, which excessively trusts the outcomes of induction, would proceed exactly in the same way. For this reason, verificationism does not create a clear demarcation between science and mythology. The critical attitude, to the contrary, renders possible this distinction (Popper 1963). The critical attitude would have become a strong tradition in Hellenic cultures, where myths provided the basic material to expose to critics. The production of scientific knowledge, hence, did not start by repeated observation of facts (by induction) but by the critical analysis of myths (Popper 1963). Unfortunately, the critical method became a new dogma: The Greeks’ discovery of the critical method gave rise at first to the mistaken hope that it would lead to the solution of all the great old problems; that it would establish certainty; that it would help to prove our theories, to justify them. But this hope was a residue of a dogmatic thinking; in fact nothing can be justified or proved (outside mathematics and logic). The demand for rational proofs in science indicates a failure to keep distinct the broad realm of rationality and the narrow realm of rational certainty: it is an untenable, an unreasonable demand. (Popper 1963, 67) For Popper (1963), the answer to dogmatism is to accept theories as being only tentative, not definitive. This makes it impossible to objectively transcend the repetition of observations, but does not impede putting empiricism at the core of the scientific method. The critical attitude of conjecture and refutations, which Popper (1963) also calls the trial and error method, implies both uncertainty and empiricism:
618
Assume that we have deliberately made our task to live in this unknown world of ours; to adjust ourselves to it as well as we can; to take advantage of the opportunities we can find in it; and to explain it, if possible (we need to assume that it is), and as far as possible, with the help of laws and explanatory theories. If we have made this our task, then there is no more rational procedure than the method of trial and error – of conjecture and refutation: of boldly proposing theories; of trying our best to show that these are erroneous; and of accepting them tentatively if our critical efforts are unsuccessful. (Popper 1963, 68) In this passage, Popper’s epistemology appears indeed to be common sense. Beyond the science war and the battle between positivist and relativist epistemologies, it may be that everybody is Popperian in the way he conducts his life. The reason may be that we all experience that dogmatism and relativism are both misleading if we apply them to ourselves.
2.4.4. The growth of knowledge Once he established his method, Popper (1963) questioned how progress could be achieved in science. He considers that the growth of scientific knowledge is “the repeated overthrow of scientific theories and their replacement by better or more satisfactory ones” (Popper 1963, 293). He considers that this principle is applicable to all forms of knowledge, not only to scientific knowledge. In the light of these propositions, he proposes two criterion to evaluate the value of scientific theories. First is the “criterion of relative potential satisfactoriness, or of potential progressiveness” (Popper 1963, 294), which characterizes as preferable the theory which tells us more; that is to say, the theory which contains the greater amount of empirical information or content; which is logically stronger; which has the greater explanatory and predictive power; and which can therefore be more severely tested by comparing predicted facts with observations. In short, we prefer an interesting, daring, and highly informative theory than a trivial one. (Popper 1963, 294)
619
The second criterion is the probability that a theory is true, or degree of corroboration. It increases when the theory resists successive tests. For any given theory, a higher probability to be true (an ability to resist to more tests) is preferable. The key problem that Popper (1963) pointed out is that there is a tradeoff between these two criteria: the theories with higher content are also those that have a lower probability to be true. The reason is that trivial theories, made for example of a single statement “a,” can be easily tested and can resist a large number of tests. They have a high probability of being true, which results from their low informative content. Theories with more informative content, resulting from the articulation of several statements a, b, etc., will resist tests less, to the contrary, because all the statements they include will have to be verified by these tests. The more numerous the statements, the larger the content of the theory will be and the lower its probability to be true. Popper summarizes this situation with the following equations: Ct(a) ≤ Ct(ab) ≥ Ct(b) P(a) ≥ P(ab) ≤ P(b) Where a and b are statements, Ct is the content of statements and P is the probability that a statement is true. The consequence of this model is that only by keeping the same level of content can the probability that a theory is true be increased. As Popper (1963, 295) puts it: if growth of knowledge means that we operate with theories of increasing content, it must also mean that we operate with theories of decreasing probability.… Thus if our aim is the advancement or growth of knowledge, then a high probability.… cannot possibly be our aim as well: these two aims are incompatible. If we accept this idea that knowledge progresses by creating theories that have a lower probability to be true, then we must see the scientific activity as progressing to “problems of increasing depth” (Popper 1963, 301). The problem is that scientists
620
want to increase the certainty of their theories, rather than their content. They are reluctant to commit themselves to comprehensiveness and are not encouraged to do so. An article has more chances to be accepted in a journal if it has a low content and a high level of certainty than if it has a high content and a low level of certainty. The consequence is that science remains stuck on trivial problems such as the impact of fragmentation on biodiversity, the impact of fires on nutrient cycling or biomass maximization in agroforestry systems, while it does not treat the deep problems our future depends on, such as the causes of slash-and-burn cultivation and deforestation.
2.4.5. What is truth and can we know it? The idea of progress raises the issue of knowing what truth is. The answer I gave in this dissertation, in the general introduction (Part I) and in Chapter 9, is that truth is the perfect match between representation and realities. It exists as the reality does and is unique as the reality is. But despite the fact that the truth exists, only perfect spirits, which humans are not, can know it. This apparent paradox may explain why epistemologists are divided between those that contend that the reality can be known (the positivists) and those that contend that it cannot be known and that what matters is the representation we can made out of it (the relativists). But the fact that there is an absolute truth does not necessarily imply the possibility to know this truth, indeed. In order to avoid falling into the positivist and relativist traps, I took a position contending that it is worth trying get closer to the truth, even “knowing” that we cannot know it completely and cannot even know how far we know it (Figure 15 page 435). Popper’s (1963) epistemology matches with this model and helps to understand why it is not a paradox. Popper contends, as we saw, that there is an objective or absolute truth, which he defines as the correspondence of statements with
621
the facts, a concept similar to my model of a match between representations and realities. He actually considers that truth and correspondence with the facts are synonymous. But he raises the issue of the translation of a fact into a statement using a language. In this translation, the fact is never accurately translated, so the truth is never properly formalized in the form of statements. This is, however, not incompatible with the claim that the truth exists: a theory may be true even through nobody believes it, and even though we have no reason to think that it is true … [while] another theory may be false even though we have comparatively good reasons for accepting it.… Even if we hit upon a true theory, we shall as a rule be merely guessing, and it may well be impossible for us to know that it is true. (Popper 1963, 305) From this perspective, there are no criteria by which we can recognize truth – except perhaps tautological truth – [but] there are criteria of progress towards the truth. (Popper 1963, 306) In contrast with this perspective, which he calls objective theories of truth, Popper evokes subjective theories of truth which conceive knowledge “only as a special kind of mental state, or as a disposition, or as a special kind of belief, characterized, for example, by its history or by its relation to other beliefs” (Popper 1963, 305). These subjective theories of truth lead to consider that the truth is something that can be known, because if it is subjective (if it is centered on the subject) it might be attainable by these same subjects and some criterion could be defined to know when we hit the truth: If we start from our subjective experience of believing, and thus look upon knowledge as a special kind of belief, then we may indeed have to look upon truth – that is, true knowledge – as some even more special kind of belief: as one that is well-founded or justified. This would mean that there should be some more or less effective criterion, if only a partial one, of well foundedness; some symptom by which to differentiate the experience of a well-founded belief from other experiences of belief. It can be shown that all subjective theories of truth aim at such a criterion: they try to define truth in terms of the sources or origins of our beliefs, or in terms of our operation of verification, or of some set of rules of acceptance, or simply in 622
terms of the quality of our subjective convictions. They all say, more or less, that truth is what we are justified in believing or in accepting, in accordance with certain rules or criteria, of origins or sources of our knowledge, or of reliability, or stability, or success, or strength of conviction, or inability to think otherwise. (Popper 1963, 305) In this passage, Popper provides an explanation to the dead-end toward which Latour’s epistemology led. Latour can be regarded as adopting a subjective theory of truth. Nature, according to Latour (1987), may be a truth that, using the words of Popper (1963, 305) “we are justified in believing or in accepting, in accordance with certain rules or criteria.” “Science in the making” (Chapter 9), would provide this set of rules or criteria, leading to Latour’s (1987) assertion that nature is the consequence of the settlement of controversies. In this assertion, the closing of controversies is the objective criteria allowing saying that we hit the truth. Once he rejected the possibility that the truth can be known, Popper proposed an alternate concept that would play the same function as the truth used to play in positivist science. This concept is “verisimilitude.” It has “the same objective character and the same ideal or regulative character as the idea of objective or absolute truth” (Popper 1963, 317). It combines the two criterion of scientific progress that Popper proposed: the maximization of contents and the probability that these contents are true. The verisimilitude would be the difference between the “truth-content” and the “falsity-content” of a theory or a representation (Popper 1963, 317): (1) Vs(a) = TCt(a) – FCt(a) Where Vs = Verisimilitude, TCt = truth-content, FCt = falsity-content, and a = statement. In this conception, the objective of scientists will be to minimize the falsitycontent of theories, by falsification, and to increase overall content, by creating more comprehensive theories, even if this is at the cost of decreasing the probability that the overall theory is true. In other words, scientists must abandon their objective of
623
maximizing the probability that their theories are true (the quest for certainty), in favor of the maximization of both the content and the trueness of their theories (the quest for verisimilitude). Clearly, I adopted this program in this dissertation.
2.4.6. Summary: three epistemologies Once he has put his model in place, Popper (1963) summarizes his analysis by defining three fundamental types of epistemologist or philosophers: (1) the verificationists, who consider that induction and the verification of facts by observation are at the core of the scientific method; (2) the falsificationists, like himself, who put refutation at the core of the scientific method; and (3) the irrationalists or skeptics (the relativists), who adopt the subjective theory of truth. Concerning the verificationists, they belong to the positivist tradition and demand that we should accept a belief only if it can be justified by positive evidence; that is to say, shown to be true, or, at least, to be highly probable. In other words, they demand that we should accept a belief only if it can be verified, or probabilistically confirmed. (Popper 1963, 309) Popper rejects the epistemology of this group. Concerning the third group (relativists), Popper considers that its adoption of the subjective theory of truth is based on a logical mistake: For Popper, “the very idea of error or of doubt (in its normal straight forward sense) implies the idea of an objective truth which we may fail to reach” (Popper 1963, 307). As I argued in Sections 2.1 and 2.2, the opposite may be true as well. Social constructionism, by assuming that the “reality” that matters is the one that is represented (the one that results from the settlement of controversies), leads to the logical impossibility that errors exist and to the general equivalence of all representations, i.e., to relativism. ANT inherited this same logical mistake because it conflates the agency of humans and nonhumans into a single concept (the agency of the relation between humans and
624
nonhumans) whereas a fundamental distinction between humans and nonhuman objects (or, at least, between thinking and not thinking objects) is that the first can make errors whereas the second cannot. By consequence of this logical mistake, the rejection of the existence of an objective truth is not an appropriate criticism of positivism and Popper rejected the epistemology of this group. As an alternative, Popper (1963) defends the position of the second group (falsificationist, to which he belongs), which he also calls negativist to highlight the contrast. His definition of falsificationists provides us with a summary of his epistemology: [Falsificationists] discovered logical arguments [Hume’s criticism of induction] which show that the program of the first group cannot be carried out: that we can never give positive reasons which justify the belief that a theory is true…. The rationality of science lies not in its habit of appealing to empirical evidence in support of its dogmas …, but solely in the critical approach…. Science has nothing to do with the quest for certainty or probability or reliability. [Falsificationists] are not interested in establishing scientific theories as secure, or certain, or probable. Conscious of [their] fallibility [they] are only interested in criticizing them and testing them, hoping to find out where [they] are mistaken. (Popper 1963, 310)
2.4.7. Application of Popper’s model to the subject of this dissertation I must now come back to the subject of this dissertation, whose central objective is to understand why deforestation and slash-and-burn cultivation still occur in Madagascar. The dominant environmental discourse in Madagascar accepts the following theories: A. Tavy destroys the primary forests B. Fires degrade soils and vegetation C. Tavy is the main cause of deforestation D. Fires are the main cause of environmental degradation
625
Theories A and B are trivial. They have a low informative content and a high probability to be true. In Figure 15 (page 435), they would be located close to the bottom line and almost on the right asymptote. Theory A is corroborated by the repeated observation that farmers that practice tavy clear primary forests. We cannot conceive a way to falsify it. To do so, we would need to find cases where the primary forests reconstitute after tavy, but empirical observation always showed that the vegetation that develops between two tavy cycles is a juvenile stage of secondary succession, with very significantly lower biomass and biodiversity. In the same manner, theory B is corroborated by the repeated observation that fires degrade vegetation and expose soils to more erosion. We can say, without taking significant risks, that the probability that theories A and B are true tends to be equal to one. Theories C and D are not trivial. They have a much more informative content but also a lower probability to be true. In Figure 15 (page 435), they would be located in the upper part of the diagram and somewhere between the two asymptotes. If theory C was true, it would mean that deforestation would stop if tavy stopped. We saw that this would not necessarily occur, because the marginalization of poor farmers can lead them to be hungry for land, even if tavy was not their land use, and because farmers who practice other land uses are hungry for land as well. They take the land that was cleared by tavy farmers, but might find other ways to acquire new land to the detriment of forests if there were no tavy farmers. We can say, however, that tavy is the main proximal cause of deforestation, but this brings us back to statement A. Concerning theory D, stopping fires would very significantly reduce environmental degradation, but the distinction between proximal and distant causes must be done as in the case of statement C. Otherwise, the theory would have low content and would be equivalent to statement B. By saying that theory D is true, one would be tempted to focus on the proximal causes and use fire repression to solve environmental problems,
626
whereas fires have up-side causes which need to be addressed as well. In other words, theory D encourages a form of ignorance by conflating proximal and nonproximal causes of degradation. Unfortunately, because theories C and D are not trivial, and because they cannot be efficiently addressed by science (it would not be easy to obtain a high probability that they are true), they have not been tested. Chapter 10, about the alternative to slash-and-burn paradigm, showed how the preference of scientists for theories with lower content led them to prefer trivial problems such as the biophysical optimum of alternative systems to slash-and-burn, rather than deep problems such as the political economy of deforestation. But as the problem that mattered was indeed deforestation, theories A and C were conflated together and it was assumed, implicitly, that testing theory A would be equal to testing theory C. The consequence was the creation of misrepresentations aimed at maintaining the illusion that theory A was sufficient to solve the problems posed by the theory C. To give an example, once it was stated that the solution to deforestation was agroforestry and other high yielding agrobiological technologies, farmers had to be defined as being backward, blocked by tradition, and unaware of environmental issues in order to explain the failure of extension efforts. This is how received wisdoms are created. This creation is a totally subjective process but the subjectivity is hidden behind the irrefutable objectivity of the methods used to test theory A. We must now see how these theories about the impact of tavy and fires were addressed by those who believe in a subjective theory of truth. In Madagascar, social scientists, such as political ecologists (Chapters 12 and 13) centered their analysis on theories C and D. This allowed them to debunk received wisdoms and to make significant progress. But these authors may also have fallen into the same trap of conflating theories A and C, and B and D. So, when they concluded that C and D were
627
wrong, they also had the tendency to reject theories A and B, whereas they did not provide empirical facts that allowed this rejection. This explains the revisionist tendency of these authors (Chapter 13). They were encouraged in this rejection by the influence of relativist epistemologies, which led them to think that there is no objective truth and that we can consider representations produced from different standpoints as being equal. Political ecologists, however, were never explicit about the rejection of theories A and B, probably because they perceived that facts were against them. Beyond the influence of relativism, they also carried with them a positivist western culture and failed to solve the tension between these two epistemologies. They indeed adopted a verificationist attitude, searching for case studies which would corroborate their theory that environmental degradation was socially constructed. Consequently, once they debunked the key received wisdoms that dominate the first myth (the fact that although tavy farmers are only the proximal cause of deforestation, they are regarded as the main cause and are blamed for that), they started to build a new myth: the belief that a second nature with equal value and ontological signification will be constructed and will replace the previous one. I will provide more arguments for debunking this second myth in Section 3, where I will propose definitions of nature and culture. In conclusion to this section, science must commit to develop more comprehensive theories, that is to say, using the words of Popper, theories with more verisimilitude. The price to pay is accepting theories that have a lower probability to be true. If they adopted this perspective, researchers would be able to work in a more inclusive way. They would unify theories that were heretofore regarded as challenging each other, which would allow them to create theories with more content and to achieve significant scientific progress.
628
The problem is that subjectivity, which results from our history, our specific interests and our personal affects, could use the high uncertainty of these theories to bias the construction of representation. But we saw in this dissertation that subjectivity cannot be avoided in science anyway, unless we condemn ourselves to limit our inquiry and our action to trivial issues that can be translated into simple mathematical equations. The challenge, then, consists of controlling the biases created by these emotions. More than that, facing the complex, the global or, simply put, the real, the challenge may even be to use emotions and intuition in a positive way, as I contended in the general introduction of this dissertation. The epistemology of Feyerabend (1975; 2003), in articulation with Popper’s model, will help us envision pathways to achieve this.
2.5. Paul Feyerabend: beyond Popper? Paul Feyerabend proposed an epistemology that was often regarded as challenging Popper’s. I evoked his work in my introduction, particularly his advocacy for more anarchy in the scientific method and his idea that art and science should not be regarded as different. In this section, I will show that Feyerabend’s (1975; 2003) epistemology usefully complements the outcomes of Popper (1963). Feyerabend (1994) also strongly criticized positivist science. He rejected, like Popper before, the dominance of induction, the verificationist methods and the claim that these methods could provide proofs (Feyerabend 2003). He rejected the positivist epistemologies inherited from Plato’s myth of the cavern, as both Popper (1966) and Latour (1999) did. He asserts, for example, that “nature is an artifact, constructed by scientists and artisans, throughout centuries” (Feyerabend 1994, 87). This last statement is very similar to Latour’s (1987) assertion that nature is the consequence of the settlement of controversies (Chapter 9). But Feyerabend states
629
explicitly that this constructed nature is just “nature as described by our scientists” (Feyerabend 1994, 97) and is distinct from “Nature as She Is In and For Herself 354 ” (Feyerabend 1994, 98). He holds that the different ways humans interact with their world, and the different ways they represent nature to themselves, according to their specific culture, must be given consideration for their empirical significance, but that their philosophical significance is still another issue. In other words, he does not assimilate the equal value of representations to an equal value of the real objects which are reflected by these representations. This renders possible to consider that some ecosystems have more value than others in absolute terms, and that some values are universal, even if these universal values are useful in just a few contexts and present in certain representations only. This also renders possible making mistakes, avoiding the trap of relativism. Some cultures can be wrong and, for this reason, “not all of them survive attempts at stabilization” (Feyerabend 1994, 98) when they face major problems. It then becomes legitimate, for example, to question whether we will survive deforestation even if, in the representation of most cultures directly facing this issue, there is no doubt that deforestation is, to the contrary, the condition of immediate survival. This assertion that the value of empirical knowledge is mostly local, coupled with a recognition of the unfathomable character of our world, led Feyerabend (1994) to advocate breaking the frontiers between sciences and art and to put an end to the hegemony of rationalist thinking: Rationalists … like to nail things down. They are confused by change and they cannot tolerate ambiguity. But poets, painters, musicians cherish ambiguous words, puzzling designs, nonsensical movements, all instruments which are needed to dissolve the apparently so rigid and objective nature of scientists, to replace it by useful and changing appearance or artifacts 355 and in this way to give us a feeling of the 354 355
Capitalized by Feyerabend. Italicized by Feyerabend.
630
enormous and largely unfathomable powers that surround us. (Feyerabend 1994, 99) At this point, a connection can be made with Popper’s epistemology. Art may be an appropriate mode of enquiry to address complex objects of our real world and to produce theories with higher content and lower probability to be true, i.e., theories with higher verisimilitude. If we were just rational beings, some dimensions of our reality may never enter into our representations. We may, in the end, reshape our world as the image of these incomplete, mutilated representations, and let disappear all that is not rationally understandable and utilizable. Our world could end up having no beauty, in being unnatural and inhuman. By creating representation based on all that our world tells us, on all that we can perceive by our reason as well as by our emotions and intuitions, we may, on the other hand, allow the complete unfolding of our world into our representations. We may then respect it better, and thereby guarantee better its future and our future. To illustrate this issue, we can ask whether beauty, a concept which cannot be easily translated into rational equations, needs to be addressed and integrated into our representation. The statement “this painting is beautiful” has a very elevated content and a very low probability to be true, which may explain why positivist science was not significantly interested in it. A painting is the consequence of an infinite number of facts that depend on the whole identity, culture, and history of the painter, expressed in an infinite number of choices regarding the colors, the shapes, the objects, the geometry, the composition of the painting. The interaction between this infinite number of features and the observer determines the statement this observer makes. Due to the complexity of this interaction, all observers have a different appreciation. For this reason, it would be practically useless, and could even be dangerous to try to define an objective beauty. We can thus consider, in practice, that beauty is a subjective concept.
631
But this does not mean, philosophically, that beauty does not exist objectively, as a fundamental quality perceived by all human beings. An objective beauty could be more or less present in different paintings, but its perception would depend on the observer. What would be subjective would be the perception of beauty, not beauty itself. But as the real beauty cannot be perceived objectively (it can just be perceived by a subject), it appears to be nonsense to pay attention to this objective beauty. It may exist however, in which case not all paintings would be equivalent even in absolute terms. The fact that no society associates excrement to the concept of beauty is an argument in favor of a universal dimension of beauty. In a famous experiment (Fleming 1974), it has even been shown that apes spontaneously associate excrement to ugliness. It may be a mistake, hence, to think that because something is abstract, it is not real. Beauty is abstract but its perception depends on the existence of a body of flesh and on the interaction of this body with its environment. The objective beauty could be a property emerging from the complex combination of these interactions. As we are all born in the same world and have all seen the sun, rocks, animals, flowers, people, water and fires, there might be a universal principle of beauty which links us together. Coming back to the issue addressed in this dissertation, scientists in Madagascar, as any individual, can experience the beauty of nature, the beauty of people, the beauty of cultures. It may be that these different experiences result in different representations, different policies, and in the end in different worlds. If none of these beauties would be experienced, one could wonder what kind of representation would be produced and what kind of world would be created. If they would be experienced in an imbalanced way, the way we understand our world may be biased and may be inefficient to help us drive better our future.
632
The challenge of science, in order to favor knowledge growth, would then not be to suppress emotions, intuitions, and other skills that we oppose to rationality. It would be to use these skills positively in attempts to produce the most comprehensive theories. This dissertation, by showing the types of bias that constrain the growth of scientific knowledge and the resulting failure to solve crucial issues such as deforestation and poverty, provided empirical arguments in favor of this thesis. Chapter 11, for example, showed how the researchers from the ASB Consortium created relevant outcomes by confronting emotionally with their world, which helped them to free themselves from the methods within which their work was framed. One characteristic of conventional science should, however, be conserved: the experimental method. It would just need to be extended to larger and more complex objects. We do that already indeed. In every day life, we advance by a series of trials and errors. Policy making can also be regarded as a form of experimentation. This fact should just be recognized, as it should be recognized that the VMCRCC and the implementation of management transfers were nothing other than large scale experiments, with just one replication in the first case and more than 500 in the second case. Once this was recognized, we could set up more cautious monitoring tools and collect feed back that would allow adjustment of our actions. In summary, Feyerabend enables us to go beyond Popper’s epistemology but does not lead to the rejection of its key principles. Karl Popper understood that science needs to advance to less probable theories with more content, but he did not propose a method to do so. He remained framed in a conception of science which concentrates on issues that can be addressed by the ordinary tools of empirical research, by making a succession of experimentations aimed at refuting, rather than verifying, previous systems of representations. Feyerabend showed that, confronted by the increasing complexity of the objects it studies, science needs distance itself from these methods,
633
by being more anarchic (Feyerabend 1975) and by blurring its frontiers with art (Feyerabend 2003). His work enables us to realize that the problem of “demarcation” and the problem of “meaning” (Section 2.4) may indeed be the same problem. In other words, disorder may be necessary to let the door open to revolutions, paradigmatic change, and to create more intuitive but nevertheless meaningful representations. Popper did not deny that a-methodic production of knowledge could produce meaning, but he created a demarcation between a science that adopts a rigorous method and other modes of enquiry that free themselves from this method. Feyerabend, on the other hand, rejected this demarcation. But since Popper addressed the other side of this demarcation in his political philosophy work, as I will show in Chapter 15, the difference between the epistemologies of these authors may indeed not be significant. I will now introduce Edgard Morin’s epistemology, which will show that order and disorder are inseparable, as may be Popper’s and Feyerabend’s epistemologies.
2.6. Edgard Morin’s science of complexities The failure of environmental and rural development policies and programs in Madagascar; the huge quantity of knowledge available concerning one single issue such as deforestation; the complex structure of this dissertation, which is just one complementary attempt to address this issue; the difficult articulation of the social and the natural, of the local and the global, of the traditional and the modern; the fast pace of change and the globalization of our world; all these elements argue for saying that our world is becoming increasingly complex and that science has to provide appropriate tools to address this complexity. Popper allowed the scientific method to make significant progress. Feyerabend showed the importance of going sometimes
634
against methods. Edgard Morin (1973; 1990), whose work I will briefly present now, allows us to go farther in the design of new methods. Morin started his career with an analysis of change in a small commune of France, in the early 1960s, when French agriculture was entering into “modernity.” His book, La Metamorphose de Plozevet (Morin 1984), revealed a willingness to provide a comprehensive view of complex social changes. Morin (1990) himself recognizes that he was “never able to isolate a studied object from its context, its antecedents, its transformations” 356 (Morin 1990, 12). He always sought a “multidimensional thinking,” 357 believing that truths that appeared to be antagonistic were in fact “complementary, without ceasing to be antagonists” 358 (Morin 1990, 12). Morin (1973) later proposed a method to address complexity. The epistemological framework presented in the introduction of this dissertation was much influenced by this method. I will briefly present, here, the main principles that sustain this method. First, Morin contends that the modes of knowledge acquisition that currently predominate “mutilate rather than express the realities and phenomenon which they are accounting for” 359 (Morin 1990, 9). They “produce more blindness than elucidation” 360 (Morin 1990, 9). This is due to the dominance of a “simplification paradigm,”
361
formulated by Descartes, according to which truth has to be found in
“clear and distinct ideas.” 362 Morin (1990, 18) points out, here, the consequences of the Cartesian anxiety. He explains that this simplification paradigm provoked the 356
Translation by Pollini. Original text in French: “jamais pu isoler un objet d’études de son contexte, de ses antécédents, de son devenir.” 357 Translation by Pollini. Original text in French: “pensée multidimensionnelle.” 358 Translation by Pollini. Original text in French: “complémentaires, sans cesser d’être antagonistes.” 359 Translation by Pollini. Original text in French: “mutilent plus qu’ils n’expriment les réalités ou les phénomènes dont ils rendent comptent.” 360 Translation by Pollini. Original text in French: “Ils produisent plus d’aveuglement que d’élucidations.” 361 Translation by Pollini. Original text in French: “paradigme de simplification.” 362 Translation by Pollini. Original text in French: “idées “ claires et distinctes” .”
635
disjunction of science and philosophy, of arts and humanities, and led to reductionism. Science is now unable “to know itself, to reflect on itself, and even to understand itself scientifically” 363 (Morin 1990, 18). Hyperspecialization
leads to “tear apart the
complex fabric of realities and to believe that this arbitrary cutting up is the reality itself” 364 (Morin 1990, 19). Consistent with this, science operates “through the selection of significant data and rejection of insignificant ones.” 365 It “separates,” “unites,” “creates hierarchies,” and “centralizes,” 366 using logical principles which are indeed “commanded by “supra-logical” organization of thought or paradigms” 367 These paradigms “govern our vision of things and of the world without our being aware of it” 368 (Morin 1990, 16). Science, however, achieved great progress. But “its ultimate harmful consequences started to be revealed in the twentieth century” 369 (Morin 1990, 18), when societal problems were addressed by “obtuse doctrines that try to monopolize scientificness” 370 (Morin 1990, 20). These doctrines caused “supreme tragedies” 371 (Morin 1990, 21) when political powers used them to establish truths that help them to achieve their goals. Marxism, fascism, and the neo-classic model of free market may be the main doctrines that have created such tragedies. Chapter 6 showed us that the
363
Translation by Pollini. Original text in French: “de se connaître, de se réfléchir, et même de se concevoir scientifiquement elle-même.” 364 Translation by Pollini. Original text in French: “déchirer et morceler le tissu complexe des réalités, et donner a croire que le découpage arbitraire opéré sur le réel était le réel lui-même.” 365 Translation by Pollini. Original text in French: “sélection de données significatives et rejet de données non significatives.” 366 Translation by Pollini. Original text in French: “sépare,” “unit,” “hiérarchise,” “centralise.” 367 Translation by Pollini. Original text in French: “commandées par des principes “ supra-logiques” d’organisation de la pensée ou paradigmes.” 368 Translation by Pollini. Original text in French: “gouvernent notre vision des choses et du monde sans que nous en ayons conscience.” 369 Translation by Pollini. Original text in French: “Ses conséquences nocives ultimes ne commencent a se révéler qu’au vingtième siècle.” 370 Translation by Pollini. Original text in French: “doctrines obtuses qui prétendent monopoliser la scientificité.” 371 Translation by Pollini. Original text in French: “Tragédie suprêmes.”
636
most radical forms of conservation discourse may be the next strong doctrine to come during this century and may lead to new tragedies. For Morin, political strategies need, to the contrary, to work “with and against the uncertain, the unexpected, the game of interaction and retroactions.” This can be provided by what he calls “complex thinking,” 372 a scientific method whose ultimate goal would be to be “adequate to the objects” 373 (Morin 1990, 72), rather than to create objects that are adequate to science. A first principle of this method is the dialogic. It claims that we must overtake the separation of alternative terms and consider duality as an expression of unity. Concepts that used to be separated then become complementary inside a wider vision, while remaining antagonist and contradictory (Morin 1990). For example, unity and diversity, simplicity and complexity, stability and instability, order and disorder, equilibrium and disequilibrium, holism and reductionism, self-organization and entropy, nature and culture are inseparable pairs of qualities. They must exist together to have an effect and they must be addressed together to be understood. A second principle is the “recursive loop” 374 (Morin 1990, 99), a process where “effects are at the same time causes and producers of that which they produce” 375 (Morin 1990, 99-100). Reproduction is an example because each living organism is at the same time the cause and the product of it. In the same manner, individuals shape societies which shape individuals (Morin 1990). The principle of recursive loop departs from the idea of linear relations between causes and effects (Morin 1990) and authorizes self-organizational processes.
372
Translation by Pollini. Original text in French: “Pensée complexe.” Translation by Pollini. Original text in French: “Adéquat a l’objet.” 374 Translation by Pollini. Original text in French: “Boucle récursive.” 375 Translation by Pollini. Original text in French: “Les effets sont en même temps causes et producteurs de ce qui les produit.” 373
637
The third principle is the hologram. According to it, the whole is in the part. It complements the classical notions of system which emphasize that the whole is more than the sum of the parts but lacks insights about the relations from the whole to the parts. The three principles are linked. The recursive loop is an emerging property of the dialogy while the hologram may be produced by the recursive loop. A last key principle for this paradigmatic change, which results from the others, is the open system. As remarked by Maruyama (1974, in Morin 1990, 33), “to conceive all objects and entities as being closed lead to a vision of the world that is classificatory, analytical, reductionist, a linear causality.” 376 With the open system, to the contrary, larger chunks of the reality can be revealed by a never ending process of knowledge accumulation. This vision is consistent with the epistemology of Karl Popper, for whom open systems of knowledge and open societies are necessities, as we will see in Chapter 15. A last important insight from Morin (1990) is the distinction between rationality and rationalization: Rationality is the game, the never ending dialogue between our spirit that creates logical structures, which applies them to the world and which dialogs with this real world. When this world does not agree with our logical system, we have to admit that our logical system is insufficient, that it encounters just a part of reality. Rationality, as it were, never has the pretension to exhaust in a logical system the totality of the reality, but has the will to dialog with that which resists it. 377 (Morin 1990, 94) Rationalization, on the other hand,
376
Translation by Pollini. Original text in French: “concevoir tout objet et entité comme clos entraîne une vision du monde classificationelle, analytique, réductionniste, une causalité unilinéaire.” 377 Translation by Pollini. Original text in French: “la rationalité c’est le jeu, c’est le dialogue incessant entre notre esprit qui crée des structures logiques, qui les applique sur le monde et qui dialogue avec ce monde réel. Quand ce monde n’est pas d’accord avec notre système logique, il faut admettre que notre système logique est insuffisant, qu’il ne rencontre qu’une partie du réel. La rationalité, en quelque sorte, n’a jamais la prétention d’épuiser dans un système logique la totalité du réel, mais elle a la volonté de dialoguer avec ce qui lui résiste.”
638
consists of a will to enclose reality in a coherent system. And all that which, in reality, contradicts this consistent system is discarded, forgotten, put aside, seen as an illusion or appearance. Here, we realize that rationality and rationalization have exactly the same source, but that while they develop they become enemies one to the other. 378 (Morin 1990, 94) This analysis is also consistent with Popper’s epistemology and with his acceptation of uncertainty. Rationality would be compatible with uncertain and open systems of knowledge because it would guarantee a structured method of refutations. Rationalization, on the other hand, would be the enemy (Popper 1966) of uncertainty due to its attempt to enclose representations inside fixed boundaries. In sum, Morin’s epistemology is consistent with that of Popper but provides new tools to address specifically complex systems. Central in this new conceptual apparatus is the dialogical principle, which appears to be much more powerful than the hybridity principle developed by Latour (1993) as I will show in the next section about nature and culture. Morin (1990), furthermore, does not propose to abandon the principles of order, separability and logic that dominated conventional science. Instead he integrates these concepts into a wider framework that includes their opposites, creating dialogical relationship between these opposites. This allows the production of discourses that are “multidimensional but not totalitarian, theoretical but not doctrinal” (Morin 1990, 67). The production of knowledge, from a battle between challenging representations, becomes an attempt to fuse supposedly opposite perspectives. These visions, precisely because they seem to oppose, need each other to become meaningful. It appears clear that the dialogical principle was largely applied in this dissertation. For a smooth transition from these epistemologies to new practices, I will now analyze, in the light of my practical outcomes and of this epistemological framework, 378
Translation by Pollini. Original text in French: “consiste a vouloir enfermer la réalité dans un système cohérent. E tout ce qui, dans la réalité, contredit ce système cohérent est écarté, oublié, mis de coté, vu comme illusion ou apparence. Ici nous nous rendons compte que rationalité et rationalisation ont exactement la même source, mais qu’en se développant elles deviennent ennemies l’une de l’autre.”
639
two concepts that were central during this dissertation and that need to be clarified before I will propose new practices: nature and culture.
3. Nature and culture 3.1. Introduction Nature was present in several chapters of this dissertation but took many forms. The concept seemed to have unclear boundaries. For example, the distinction between what we call the environment and what we call nature was unclear. In order to avoid this confusion, I sometimes employed an elusive expression – “this thing that exists in the real world independently of the action or intentions of humans” – to designate what is commonly called nature. Before proposing new pathways and strategies to address the issues of deforestation and slash-and-burn cultivation in Madagascar, I need to provide a clear definition of nature, as well as of some concepts that are often opposed or associated with it: culture and the environment.
3.2. What is nature? For the Oxford English Dictionary, 379 nature is “the physical world, including plants, animals, the landscape, and natural phenomena, as opposed to humans or human creations.” Nature is also “the inherent qualities or characteristics of a person or thing.” These definitions, which evoke nature as “opposed to human creations” and refer to “inherent qualities,” are consistent with the expression I used in this dissertation. Nature would be “all those things whose existence is not dependent on our intentions.” This is the definition I will adopt, which I will now justify.
379
From the Oxford Dictionary On Line: http://www.askoxford.com/?view=uk.
640
First, I contend that it is necessary to designate these “things whose existence is not dependent on our intentions,” whether we use the word nature or not for this designation. Humans are an entity different from all others. They are self-reflexive, aware of themselves and of their world. They can make choices and errors, as we saw in Section 2.1 and 2.2. They have intentions and create and use tools that extend their power and serve their intentions. They can project themselves into the future and remember their past. They have rational thinking, not just instinct. All these characteristics allow them to produce something distinct from what other organisms produce. We call this thing culture. A cultural object is not necessarily physically different from a natural object. But the process which leads to it is different. To give just an example, the evolution of species in nature is the consequence of mainly two phenomenon: stochastic events that modify the genes and the selection of the most appropriate genes by the environment. This natural selection leads to increasing diversity and complexity. The recovery of biodiversity after each massive extinction in the past provides evidence of this characteristic. When the “evolution” of species is determined by human intentions, the rules are different. Only the genes that match better with some precise expectation are conserved, which can lead to simplification and less diversity, as occurred with cultivated crops. If we would use the word nature to designate cultural objects that are constructed following these different rules, the consequence could be an ontological vacuum. The particular outcome of processes that are not intended by humans would not be designated. It is important to notice that this definition of nature is not justified by the elevated human impact on environment. Rather, it is justified by the “nature” of this impact. For example, earthquakes, meteorites and other cataclysms can have more impact than human activities. But in this case, the impact is not intentional, which is
641
why it is qualified as natural. On the other hand, apes can fabricate tools (Whiten et al. 1999) and may be motivated to do this by intentions. For this reason, we can consider that apes have a culture, and that their environment is rendered less natural by this culture. The environmental impact of apes’ activities, however, is low, and for this reason, the concept of apes’ culture does not have much practical use. But ontologically and phylosophically, it is equal to the concept of human culture. It is true also that many societies did not create a distinction between nature and culture. But this does not mean that this distinction does not exist. The absence of a distinction in certain societies can be explained by the fact that when there is still a certain balance between nature and culture, this distinction would be of little practical use, as it is in the case of apes. Remembering the arguments of Feyerabend (1994), the fact that different representations exist has a different meaning from a practical standpoint than from a philosophical standpoint. When the basic constituents of nature are not at peril, there is no need to distinguish them from cultural productions. If we accept this definition I have proposed, nature should be seen as a dimension of things, rather than a thing. The exact same plant or animal could be more or less natural, more or less cultural, if its genome is the consequence of natural selection or of human engineering. Terrasson (2002) showed that nature is present everywhere, including in human productions, because it constitutes the basic material for all things. The lumber used to build a house still contains some nature, but less than the tree that has been cut to make it. A frozen fish stick contains some nature, but less than the fish caught to make it. Even a highway or a nuclear power plant contains some nature, in the structure of the sand and rocks that have been used to build them. As soon as humans remove some nature from objects, they put some culture in them. The destruction of nature, thus, is not necessarily the physical suppression of natural objects. It can be the diminution of their natural content in favor of cultural content. In
642
other words, the corollary of the destruction of nature is the production of culture. There is a tradeoff between the two. Or, as culture itself is a production of nature, we could say that nature has the capacity to destroy itself by the creation of more culture. Coming back to the subject of this dissertation, we can distinguish ecosystems according to their higher or lower natural and cultural content. Higher natural content is found in virgin forests. Then follow, according to a gradient of decreasing nature and increasing culture, primary forests, secondary forests, forest gardens, fallow vegetation, grasslands, multiple cropping systems, degraded grasslands, monocultures, high input agriculture, hydroponic culture of GMOs and urban environments. Until now, I addressed the problem of the quality of nature and culture, which must not be confused with the problem of the value of nature and culture. Maybe this confusion is frequent and is the cause of the excessive politicization of the debate. But the fact that nature is destroyed by human does not imply that the objects that replace it have less value. Some culture is produced in these objects and culture has value as well. I will try now to address this issue of the value of nature and culture. In other words, I will try to answer the question of whether we will win or lose from the creation of more culture to the detriment of nature.
3.3. The value of nature First, we can ask why we would call this process that leads to less nature and more culture “destruction,” rather than just calling it “change.” The reason is that cultural and natural objects are created following different rules, as I explained earlier. For this reason, they cannot be regarded as equivalent. One is destroyed while the other one is produced. A world designed by culture would have qualities different from a world designed only by nature. I evoked, concerning the diversity of living forms, the fact that nature increases their diversity while culture can decreases it. This
643
is certainly not the only difference. What would be the overall difference between a more cultural and a more natural world? We do not know, because not everything is known about nature. This idea that there is an unknown may be a central argument to justify keeping some nature in our world. Humans have intentions that are mostly determined by their knowledge and culture can be viewed as an expression of this knowledge. In a world where only culture mattered, unknown objects and functions would be at peril. In this world, humans would have an excessive arrogance. They could make everything revolve around their culture and their knowledge. They would transform objects according to what they can understand of them; according to their representations. Coming back to social constructionism, by defining nature as the consequence of the settlement of controversies, it can legitimate this arrogance, leading to a world where only our values matter. In the civilization we live in today, the value of things is mostly assessed by their utility and by their valuation by markets. Social constructionism could then lead to a triumph of this materialist paradigm, to the detriment of something that we do not know but that may have more significance. Social constructionists, however, are often the stronger detractors of the free market doctrine and of the commoditization of nature. They deconstructed development, industrialism and modernization with the same enthusiasm as nature. I contend that there is a contradiction here. Recognizing a nature out there and the necessity to conserve it may be the only way to avoid the complete commoditization of our world. It is not nature that had to be deconstructed, it is the representation of nature; the way it is managed, manipulated, appropriated. If we grant to nature a value which is not just that of a commodity, we have to recognize an intrinsic value to it, beyond the empirical and contextual value given to it by the
644
societies that utilize it, each of them in their own way. This value may be the value of the unknown and of the mystery. Conserving nature would be meaningful not only for our world, but for us as well. If nature would not be there, or if it would be significantly reduced, then it may be that part of our identity would be at peril as well. According to Terasson (2002), the natural world is mirrored in our unconsciousness. Maybe this is why nature remains fascinating and valuable to most people. It speaks to us because we are part of it and it is part of us. In traditional societies, nature is the domain of the spirits, the ancestors, the unknown. For modern people, it is a realm of esthetic and a cause of fear. Nature never leaves those who confront with it indifferent, except when it is controlled, managed, transformed into a cultural production.
3.4. The environment I must now introduce a last concept: the environment. First, this concept must not be confused with nature. As we saw, nature is a force, a quality, a principle more than a thing. The environment, on the other hand, has a physical reality. We can say that something is natural but we cannot qualify an object as being “environmental.” We just can say that it is “environmentally friendly,” which means that its relationship with other objects is respectful of environmental values. So, what is it then that we call the environment? The etymology of the word is meaningful. The environment is just what surrounds us. In French, “il nous environne” (it surrounds us). It is something multidimensional. We can speak of the natural environment but also of the cultural environment, the social environment, the urban environment and so on. When we simply speak about the environment, we speak about something that can have all these dimensions in it. The environment, then, can be socially constructed because the society makes it, in contrast to nature which is,
645
by definition, not socially constructed. Only the representations of nature are social construction. What social constructionists call the produced nature may indeed be the environment. An ontological vacuum was created by conflating together the concepts of nature and environment and this semantic mistake may explain the confusion that exists today in the debate about nature. If we used the term nature to designate the environment, we may end in having no word to designate that thing “whose existence is not dependent on our intentions.” And then we may lose it.
3.5. Nature in political ecology I can now go back to political ecology and to other narratives about environmental
management
and
policies
that
were
influenced
by
social
constructionism. These narratives were very useful for showing that nature is not the only thing that matters, and that nature should not be seen as separated from culture. But the problem is that these narratives may not be dealing with nature. Instead, they may be dealing with the environment. By doing so, they do not answer to the question of what room will or should be left to nature in our world. Political ecologists, however, are not enemies of nature. We can guess while reading their work that the physical world and the natural objects in it (the forests, the living organisms) matter to them as do culture. But they overlooked nature because they put into question the fact that nature and culture are distinct ontological categories. They conflated the two categories in reaction to the separation that ecologists had created.
3.6. The dialogy of nature and culture It is time, now, to call Edgard Morin to the rescue. The dialogy principle provides the solution to this issue. The relation between culture and nature may be
646
dialogical. If there is a culture, then there is something that is not a culture. So why not to call it nature? Nature and culture are, together, the identity of hybrid objects. They create unity in the duality. If there would be no nature, there could be no culture. The consequence would be a double ontological vacuum. This is why nature cannot be produced, and cannot even be social. It can just be associated with the social, with cultural productions. But inside this association, it remains nature.
4. Conclusion I showed in Section 2 that Latour’s (1993) concept of hybridity and principle of symmetry did not allow an escape from the relativist trap of social constructionism. They led to the conflation of human and nonhuman “agencies,” forming a unique principle which emerges from the relationships between humans and nonhumans. Due to this conflation, the ontological distinction of the natural and the cultural world are overlooked. This may explain why certain schools of political ecology, influenced by postmodernism, social constructionism and the concept of hybridity, had the tendency to consider ecosystems and agro-ecosystems as equivalent natures. I showed in Section 3 that the distinction between nature and culture, between ecosystems and agro-ecosystems, is indeed necessary. Otherwise, the dimensions of our world that are not part of our representations or that have no utility would be overlooked. They could even be denied the right to existence. But this distinction must not lead to the separation, by purification, of nature and culture, or of the physical and the social world. This explains why Latour’s (1987) concepts have been so useful to conduct this research (Chapter 9). In order to render possible ontological distinction without falling into the trap of separation or purification, we can use Morin’s (1990) concept of dialogy (Section 2.6). According to this principle, nature and culture have specific properties which make them distinct but this distinction is also the cause of
647
their unity. In this perspective, the specificities of each pole must be understood first, as a prerequisite to understand the properties of the whole. The concept of hybridity, to the contrary, led to a focus on the relations between the two poles and to conflation of the two concepts. But the most important outcome of this chapter was Popper’s (1963) concept of verisimilitude and the principles that sustain it (Section 2.4). Popper showed that induction does not allow theories to be improved: it just allows existing theories to be rendered more probable. Popper then proposed a paradigmatic shift where refutation (or falsification), rather than verification, would be at the core of the scientific method. We saw in this dissertation that verification is still the dominant mode of scientific inquiry. The reason may be that it allows increasing certainty for a given theory, which gives the illusion that truth can be known. It favors the accumulation of facts which are compatible with existing theories, which allows an escape from what Berstein (1983) called the Cartesian anxiety. Even research influenced by relativist epistemologies, such as the work conducted in Madagascar by political ecologists, has privileged verification through the accumulation of case studies that confirmed existing theories. With falsification, to the contrary, criticism is the condition for improving the representation of our world and the researcher is committed to find what is wrong in existing theories. Knowledge, then, grows by the addition of new content, not only by increasing the certainty of existing content, leading to representations of our world that have higher verisimilitude. When science addresses objects with very high content (when representations reach the upper part of the diagram in Figure 15, page 435), scientific methods are not sufficient for achieving progress, and a blurring of the frontier between sciences, philosophy and art becomes necessary, as shown by the insights of Feyerabend (2003). Even physics faces this necessity.
648
Unfortunately, there is a tradeoff between the content of theories and their probability to be true. When theories achieve more comprehensiveness, it becomes more challenging to provide evidence for all their content. This is why researchers usually favor theories with low content and a high probability of being true and privilege verification by induction of a few simple statements. But as theories with high content are also those that matter more, we consequently fail to understand the world we live in and to take the right decisions for our future. Decision makers, however, need to address the whole world, not only pieces of it. As they depend on the outcomes of work by scientists, they are often using the outcomes of theories with low content to give answers to problems with high content. In the process, additional content is injected subjectively in order to create an artificial match between the low content theory and the high content problem. Received wisdoms are created or are given more momentum during this process. In order to visualize this problem, I will now synthesize the main outcomes of this dissertation in a table organized according to Popper’s concept of verisimilitude (Table 22). Column A poses a series of question that relate to the issue treated in this dissertation. Column B shows the theory with low content and a high probability of being true that is commonly used to answer these questions. Column C shows the theory with high content and a low probability of being true that is usually derived from the theory in Column B. Column D shows the received wisdoms that are created, or supported in the process of moving from the theories in column B to the theories in column C. Column E proposes tentative theories having higher verisimilitudes, i.e., new interpretations made in the light of the outcomes of this dissertation. They are conjectures that need to be exposed to facts and falsified by future research work. This table, compared with the summary of Blaikie and Brookfield’s (1987) book Land Degradation and Society in Chapter 13, allows one to visualize the specific
649
Table 22: Comparison of theories in relation to their verisimilitude A. Question
650
Is deforestation significant at a global scale?
B. Theories with low content and high probability to be true
In most regions of the world where primary ecosystems remain, satellite imagery shows a fast reduction of their extension Once primary forests are cleared, human societies develop land uses that can leave significant room to trees and woodlands, especially in areas of high population density (more people, more trees)
C. Theories with high content and low probability to be true, derived from B Deforestation is very significant at a global scale
Deforestation is not so significant at a global scale
D. Received wisdoms created during the derivation from B to C
E. Theory with high verisimilitude
Human societies fight against nature
Human societies live in harmony with nature if they are not disturbed by forces from outside
Deforestation is very significant at a global scale, but is followed by reforestation, which allows the recovery of a significant part of the biomass, but not a significant part of biodiversity.
Table 22 (continued)
Are slash-andburn farmers the main cause of deforestation?
651
Does the Malthusian discourse make sense?
When primary forests are cleared, it is by slash-and-burn farmers
Slash-and-burn farmers are the main cause of deforestation
Many communities face the limit of the carrying capacity of their system, as shown by decreases in yield and increasing poverty The Malthusian discourse is hegemonic
Population control is necessary if we are to avoid a Malthusian crisis in Madagascar
Anthropogenic fires burn biomass and Are anthropogenic degrade soils fires degrading the Anthropogenic fires environment? are ignited by communities for managing their land
Slash-and-burn farmers must be educated and repressed if we are to stop deforestation People are the problem
The Malthusian discourse is a received wisdom in Madagascar
People are the solution, not the problem
Anthropogenic fires degrade the environment Anthropogenic fires do not degrade the environment when controlled
Anthropogenic fires are demonic Anthropogenic fires are a tool like another
Slash-and-burn farmers are only the proximal cause of deforestation. Marginalization and the search for new land for agriculture are the real causes, the ones that need to be addressed. People can be the problem or the solution, depending on the effect of other factors. A general rule may be that pressure on land is the cause of land-use change (Boserupian hypothesis) but that land-use change can further be constrained by economic factors (lack of investment capacity). Malthusian and Boserupian dynamics may constitute a dialogy and capitalization/decapitalization dynamics may be at the core of this dialogy. Anthropogenic fires are a powerful tool with high impact on the environment, but this impact is limited to certain boundaries when fires are controlled and associated with management practices. Anthropogenic fires create landscapes which have less nature, but more culture.
Table 22 (continued) Are agrobiological technologies appropriate alternatives to slash-and-burn cultivation?
652
Is community based natural resource management (CBNRM) a wiser approach than coercion to improve environmental management?
Agrobiological technologies allow higher yield and more sustainable production than slash-and-burn cultivation
Agrobiological technologies are appropriate alternatives to slash-and-burn cultivation
Coercion failed to favor efficient environmental management
CBNRM will work better than coercion
In Madagascar, there is less environmental degradation in areas where GELOSE and GCF are implemented
Agrobiological technologies are best-bet alternatives and research and extension efforts should concentrate on them.
Local communities have the capacity to manage their environment in a sustainable way In Madagascar, In Madagascar, CBNRM leads to the GELOSE less environmental and GCF degradation effectively transfer management rights to local communities.
Slash-and-burn farmers in a situation of crisis lack capacity to invest, cannot take large risks, have difficulty satisfying their short term needs and attempt to maximize labor productivity rather than yield. Considering these criteria, agrobiological technologies proved to be less profitable and less suitable than crop rotation, plowing, and integration with livestock husbandry in Beforona. Thus agrobiological technologies may not play a central role in agricultural intensification. In Madagascar, the implementation of GELOSE or GCF contracts has led communities to reduce forest clearing and vegetation burning in the place and during the time of this implementation. It remains to verify whether fires and clearing will not just be displaced in time and space. In a context where livelihood depends on access to forest resources and clearing of forest land, this behavioral change may result from a fear of the state and its partners, rather than from a genuine CBNRM. The communication device of the GELOSE may serve to channel power rather than rational and transparent negotiation. In sum, CBNRM is a wise approach but has not been genuinely implemented yet and cannot work alone.
outcomes of this dissertation. It shows that the narratives I developed are similar to those developed by political ecologists and that they were already apparent in early political ecology works such as Blaikie and Brookfield’s (1987) Land Degradation and Society. But it also shows that the construction of myths operates in political ecology as it did in mainstream environmental research before, which led me to depart from the outcomes of this discipline as well. I proposed an explanation for the mechanisms which lead to these myths. In fact, I did not propose any new explanation. The explanation was already provided seventy years ago by Popper (1959, 380 1963). But due to the influence of radical postmodern
philosophies
and
of
social
constructionism,
more
fashionable
epistemologies have developed recently, leading to the neglect of Popper’s insights. I hope that this chapter will contribute to moving beyond both the positivist and relativist traps and will result in more attention being given to Popper’s falsificationism. I will now try, in the next and final Chapter of this dissertation, to develop these theoretical outcomes into a program of action aimed at conserving both the Malagasy rain forests and its people.
380
First published in 1934.
653
CHAPTER 15 TOWARD NEW PRACTICES
1. Introduction Chapters 13 and 14 synthesized the outcomes of this dissertation in the light of the political ecology literature (Chapter 13) and of an epistemological framework aimed at favoring more inclusive theories (Chapter 14). These chapters made it possible to respond to the first and second research questions: why do slash-and-burn farmers still clear the Malagasy rain forests and why do policies and programs fail to address this issue (Part I). The answer I gave to the first question, in Chapter 13, is that social and economic dynamics marginalize the weakest social groups, while the forest frontier is the location where these groups can find a way to sustain their livelihood. The failure of environmental programs and policies (second question) results from the fact that conservation programs favor resource capture, which exacerbates this marginalization. They create new ways to generate profit from natural resources (ecotourism, scientific research, sustainable extraction, carbon sequestration, agricultural intensification) that the more powerful groups appropriate, and they closed the frontier and put a ban on tavy, without providing compensation or effective mitigation to the marginalized farmers who depend on the forest and on tavy. When the frontier is effectively closed, like in protected areas, the ecosystems are conserved but the weakest social groups pay the price for it. They become increasingly dependent on the unsustainable use of natural resources (the soil if not the biomass) which leads conservation discourses to blame them for degradation. This may provoke social conflicts which will threaten protected areas in the future. When the closing is not effective, the situation can still worsen because the efforts to close the frontier may encourage a faster land
654
appropriation by forest clearing. Ecosystems whose utilization and conversion was managed by local communities end up being in free access. In Chapter 14, I explained that these unwise strategies result from the unequal attention given to biophysical and social challenges. This imbalance results itself from a reluctance of science to address complex social objects (to address the social would imply favoring theories with more content and lower probability to be true) and from the advantage policy makers can take from this reluctance (they can orient science to the study of the parts of the reality that matter in their shortsighted view). I also showed that political ecology analyses, which focus on more comprehensive objects, improve knowledge about the social causes of environmental degradation but overlook certain biophysical aspects of environmental degradation. In this chapter, I will translate these outcomes into a series of practical recommendations aimed at stopping deforestation and reducing the negative impacts of slash-and-burn cultivation in the rain forests of Madagascar. I will hence answer my third research question: how to stop or reduce deforestation without requiring the marginalized groups to pay the price for it. Section 2 will present a concept which will be central in my arguments: metis, which is the answer Scott (1998) gave to the high modernist project (Chapter 8). Section 3 will propose a series of measures which could be tested in future programs and policies. It will provide a larger intervention framework to the development activities I proposed in Chapter 7. Section 4 will propose, based on Popper’s political philosophy, a new institutional approach for the implementation of these policies. In this framework, scientific research and policy making will be unified into a single process while still conserving their distinct role.
655
2. Metis The influence of the high modernist project described by Scott’s (1998) (Chapter 8) was manifest in both the realities (Chapter 6) and the discourses (Chapters 10 and 11) analyzed in this dissertation. High modernism was manifest in Beforona, as shown by the preference given to techniques that maximize yields whereas this criterion is not adequate for most farmers. It was also manifest in the forms of social organization and land management that were favored by the GELOSE legislation (Chapter 11). Scott’s (1998) reply to high modernism is the concept of metis, borrowed from Aristotle. For Scott, “metis represents a vast array of practical skills and intelligence acquired in responding to a constantly changing natural and human environment” (Scott 1998, 313). It is a form of knowledge that is applicable to “broadly similar but never precisely identical situations requiring a quick and practiced adaptation that becomes almost a second nature to the practitioner” (Scott 1998, 315-316). Metis resists simplification into deductive principles which can successfully be transmitted through book learning, because the environments in which it is exercised are so complex and nonrepeatable that formal procedures or rational decision making are impossible to apply. (Scott 1998, 316) Scott gives several examples to illustrate the concept. For instance, “when the first European settlers in North America were wondering when and how to plant new world cultivars, such as maize” (Scott 1998, 311), they were told by their native American neighbors “to plant corn when the oak leaves were the size of a squirrel’s ear” (Scott 1998, 311). This type of knowledge is metis because it is adapted to each locality and is based on experience. The date can change even if the rule remains the same, a fact which is rendered possible by a deep knowledge about local ecosystems, not only about the physiology of maize. A farmer’s almanac, to the contrary, “may suggest planting corn after the first full moon in May or after a specified date, such as May 20” (Scott 1998, 312). In this case, knowledge is not metis because a general rule
656
is claimed to be applicable to a local situation whereas not much is known about this local situation. The rule, which is based on logical deduction from simple principles (Scott 1998), belongs to a form of knowledge which contrasts with metis and is called techne. In sum, metis may be the type of knowledge which allows passing from the stage of “competent performer” to the stages of “proficient performer” and “experts” described by Dreyfus and Dreyfus (1986) (Part I). An important characteristic of metis is that it is “partisan knowledge” (Scott 1998, 318). It results from the conjunction of a local problem with a necessity to solve this problem, by those who are directly concerned by it. The holder of metis knowledge is “the immediate consumer of its own conclusion” (Scott 1998, 324). He “has a passionate interest in a particular outcome” (Scott 1998, 318). Due to this necessity, metis does not have the right to be wrong, which minimizes the risk of mismatch between representations and realities. Otherwise, the real existence of those who utilize metis would be at peril. A farmer, for example, uses metis because otherwise he would starve. The extension staff of rural development projects, to the contrary, do not need to use metis when dealing with agricultural intensification. They prefer to use techne, even if they accumulated metis knowledge, because they can still survive if the recommendations they make to farmers are wrong. Interestingly, the extension farmer, who is a hybrid of ordinary farmers and extension staff, applies techne on the land that serves to “demonstrate” new techniques to other farmers (supposedly) and metis on the land that will provide his food and income (Chapter 6). Due to its partisan character, metis is also highly flexible. It can adapt to a changing environment. It even has to. It may lack capacity to anticipate changes but is able to react quickly to change. “The big mistake of the rationalist –though it is not inherent in the method – is to assume that “tradition,” or what is better called “practical knowledge,” is rigid, fixed and unchanging – in fact it is “preeminently
657
fluid” (Oakeshott 1962, 31, in Scott 1998, 332). The case of Ambodilaingo, in Part III, confirmed this. It showed that traditions adapted to a changing economic context during the last decades (Chapter 5) and that farmers quickly adopted new practices as soon as they experienced that these practices were profitable and suitable to them (Chapter 7). Metis further leads to point out a new limit of Latour’s (1987; 1993) epistemological model (Chapter 14). The settlement of controversies cannot decide what nature is because different forms of knowledge can exist in parallel, resulting in settling the same controversies simultaneously and in different ways. In the global networks that link scientists and decision makers, techne is the favored mode of knowledge, while the networks that directly link agro-ecosystems and farmers favor metis. These two networks each close in their own way the controversy about what system would be the most appropriate alternative to slash-and-burn cultivation. Farmers opted for extending banana plantations, irrigating paddy fields, intensifying bottom land and maximizing labor productivity. Projects opted for fireless alternatives, sustainable systems on slopes and high yielding cropping systems. It is true that at the interface between the two groups, the controversies can be reopened like they are in this dissertation. Other points of contacts also exist like the extension farmers who practice simultaneously land uses designed according to techne (proposed by projects) and land uses designed according to metis (which they developed themselves). But overall, techne and metis coexist with limited links and it can be questioned whether the future nature (which will depend on the future land use) will be the consequence of the settlement of controversies relating to the first or the second type of knowledge. Eventually, metis is consistent with the epistemologies proposed in Chapter 14. By being contextualized, metis obliges integration of the whole in the representation of
658
the parts, which favors comprehensive representations where the world is a hologram (Morin 1990, see Chapter 14). Metis further requires open systems of knowledge and the acceptance of uncertainty, like in Popper’s epistemology. As Scott (1998, 320) puts it, “techne is settled knowledge” whereas metis remains open, calling for refutation by successive trials and errors. Metis is “the mode of reasoning most appropriate to complex material and social tasks where the uncertainties are so daunting that we must trust our (experienced) intuition and feel our way” (Scott 1998, 327). Scott (1998) postulates that ignorance is the starting point for designing human affairs, that development actors should employ flexible plans that “allow the largest accommodation to the unforeseen” (Scott 1998, 345) and the expression of human inventiveness. As Marglin (forthcoming, in Scott 1998, 344) puts it, the only certainty about the future is that the future is uncertain, if the only sure thing is that we are in for surprises, then no amount of planning, no amount of prescription, can deal with the contingencies that the future will reveal. One can wonder why the role of metis was marginalized in the development of scientific knowledge. The answer may be that positivist science, based on techne, was indeed an efficient way to develop technologies and to increase productivity, through the tailoring of industrial processes and the extension of this tailored logic to all spheres of our life (the high modernist project, Chapter 8). But now that this materialistic success has been achieved, human societies face other issues which techne may not be able to solve. Our social lives and our global environment must urgently be “fixed.” As long as decisions regarding social issues were local, they were able to escape from the technosphere and were taken by those directly concerned, using metis knowledge. But now, this escaping is no longer possible. Decisions are taken by actors that favor techne and are located further and further from the localities they impact. Paris and Washington, D.C. have significant weight to decide how Malagasy communities must organize (Chapter 11) and what land use slash-and-burn
659
farmers must implement (Chapters 6 and 10). Remote decision makers have tools that render them very efficient to study simple problems (the impact of fires and tavy on soils and vegetation), but not to address comprehensive realities (to understand the causes of land degradation and deforestation). As we saw in Chapter 14, they conflate simple problems solved by techne with nonsimple ones that techne is unable to solve. We may be in an era where metis urgently needs to be applied to the solving of nonlocal and nonpartial problems. This leads us to a last aspect: metis must not be confused with traditional or local knowledge. Metis refers to a way to apprehend the world and can be applied from the local to the global. It is a way to produce knowledge where the reality comes first to the method and matters more than the form of the outcome. Objects are not split into their parts. They are, to the contrary, contextualized. With metis, intuition plays a role in the apprehension of this context. Rationality is also used but the world is not rationalized, a distinction whose importance was shown by Morin (1990) in Chapter 14. For all these reasons, science needs metis if it is to address deep problems and there is absolutely no reason to limit the use of metis to local issues. Metis also has some weaknesses. Details of the object can be overlooked and interpretation of precise phenomenon can be false or their application can be limited to a precise context. Being contextualized both in time and space, metis knowledge may not enable anticipating the impact of contextual changes, nor the discovery of general laws. Metis is knowledge about what happens, not so much knowledge about why it happens. This is why the advocacy of metis must not be radical and techne still has to play its role. The two forms of knowledge may have a dialogical relationship (Chapter 14), each one depending on the other one. The world we live in created an imbalance between techne and metis but our attempts to reestablish an appropriate balance must not lead to the creation of an inverse imbalance. We will see, in the next sections, how
660
metis can be applied to new development and conservation policies and practices and how it can be linked to Popper’s (1966) political philosophy.
3. Propositions to solve the issues of deforestation and slash-and-burn cultivation in the eastern rain forests of Madagascar In this section, I will propose a series of practical recommendations aimed at reducing or stopping deforestation in Madagascar without requiring the weakest social groups to pay the price for it. First (Section 3.1.), I will clarify the link between deforestation and slash-and-burn cultivation in the light of my outcomes. Second (Section 3.2.), I will give my answer to the ethical question whether it is worth it or not to stop deforestation, as the controversy on this question is on the verge of being reopened (Chapter 13). Third (Section 3.3.), I will propose an ethical framework aimed at solving the tradeoff between conservation and development objectives. Fourth (Section 3.4.), I will describe the measures I propose. These measures concern three aspects which have to be articulated together: (1) the legislation (Section 3.4.1.), already briefly addressed in the conclusion of Chapter 11; (2) the supports to agricultural intensification (Section 3.4.2.), addressed in Chapter 7; and (3) compensation for restriction of access to forest land and resources (Section 3.4.3.), already briefly addressed in Chapters 6 and 10.
3.1. Are tavy farmers the main cause of deforestation? In this dissertation, I addressed simultaneously the issues of deforestation and slash-and-burn cultivation because empirical observations show that these two phenomenon are associated. Wherever there is forest clearing in eastern Madagascar, slash-and-burn farmers are those that practice this clearing, except maybe in a few particular cases.
661
But as we saw in Chapters 1 and 10, this does not mean that they are the most significant cause of deforestation. The real cause of deforestation is the search for new land to be used for agricultural production. Whatever the land use, land is necessary to increase agricultural production, when population grows, when soil degrades and when markets provide new opportunities. Tavy farmers, because they are located on the first line of the forest frontier, are those that actually clear the forest,381 but they clear it for the benefit of all types of farmers coming behind them and pushing them to more remote lands. They are only the proximal cause of deforestation while the real cause is the global macro-economic system. For this reason, tavy farmers must not be blamed and have a legitimacy to clear primary forests.
3.2. The reasons to stop deforestation When I chose the subject of this dissertation, I did not question whether deforestation was actually occurring or not in Madagascar, nor whether it would negatively impact biodiversity or other aspects of the global environment. I just accepted these facts as they are presented by the scientific community. I contended in Chapters 9 and 14 that knowledge must remain open and that all facts are disputable. But arguments must be provided for these disputes and I did not find arguments for refuting that deforestation occurs at a fast pace, will have a significant negative impact on biodiversity if all primary ecosystems are cleared or reduced to small fragments, and that this impact will be irreversible if primary ecosystems are replaced by frequently disturbed anthropogenic agro-ecosystems. Uncertainty tends to zero when no fact leads to refutation. In this case, theories must be regarded as certain in practice, in not in absolute terms. This is the case of the fast pace of deforestation in Madagascar. What is largely questionable, on the other hand, is the exact pace of 381
Not surprisingly, the region where the rain forest corridor is more fragmented is the area just facing the densely populated highlands (Map 1 in Chapter 1, page 26).
662
deforestation and its exact impact on biodiversity in the case that not all forests would finally be cleared. But as I contended earlier, I consider this debate as trivial with regard to geological time and to the ethical significance of the issue. The most reliable figures, anyway, converge to say that deforestation is fast and has accelerated during the last decades (Chapters 1 and 10). Once these statements are made, it can be questioned whether it is worth it to stop deforestation and conserve a significant part of the Malagasy biodiversity. This is an ethical question and I can just assert a personal opinion, which is that we must conserve primary forests 382 and their unique biodiversity. My main argument is that nature is ontologically distinct from culture (Chapter 14), which may provide it with a specific intrinsic value. We do not know what this value is made of because we do not know everything about nature. But precisely for this reason, and invoking the precautionary principle, I contend that some nature must be conserved in its primordial state or close to its primordial state, in some places if not in all places. I follow Serres’s (1990) arguments that a contract must be made with nature in the same manner as humans made a social contract (Rousseau 1762). As living forms are the elementary bricks of nature, I further contend that biodiversity should be conserved as an essential heritage. It is part of our world, part of our history and our future may also rely on it (Myers 1984; Wilson 1988; Whitmore 1992; Guruswamy and McNeely 1998). As it happens that the most significant biodiversity reservoirs are now located in the tropics, it is logical that the international community engages in efforts to conserve tropical forests. Accepting conservation in the name of the unknown is also a way to recognize that our knowledge has some limits, which is essential in order to escape from radical
382
This concept must not be confused with virgin forests, which are primary forests not occupied by humans. Biodiversity has no reason to be higher in virgin forests (in comparison with nonvirgin primary forests), because humans contribute to natural selection by coevolving with their environment.
663
positivist epistemologies. Moreover, as the value of things is based on what we know about these things and their uses, conservation may be necessary for avoiding the complete commoditization of our world. Conservation, then, becomes the consequence of the acceptance of the Cartesian anxiety and the remedy to the hegemony of markets. It is the guarantee that if there is something we do not know, we still have a chance to know it some day because it will still be there, in an unknown realm that we did not turn into a commodity. There is, however, a paradox in this assertion. Market, which is the cause of the destruction of nature, may also be the solution for conservation. “Nature has become an emporium, a commercial warehouse awaiting its brokers” (Zerner 1999, 4). I unfortunately see no way to escape from this contradiction. Nature has to be conserved for ethical and philosophical reasons, but unfortunately, money is necessary to achieve this and market hegemony renders it difficult to raise money by other ways. Markets can, however, be regulated. The internalization of externality is an approach given increasing importance by economists. I evoked it in this dissertation and Section 3.4.3., about subsidies will address it more in length. A last argument for conserving biodiversity and tropical forests is that all those having an experience in these ecosystems attribute a value to them which is not found in the modified landscape created once they have been destroyed. Local people see primary forests as the domain of their ancestors and other spirits. They developed rituals aimed at apologizing to the forest and its inhabitants before clearing it. Occasional visitors, on the other hand, appreciate primary forests for an esthetical value they can find nowhere else. For both, conservation has to be compatible with the satisfaction of materialistic needs, but some intrinsic value seems to be recognized.
664
3.3. Solving the tradeoff between conservation and local livelihoods The tradeoff between conservation and livelihood was a central issue in all chapters of this dissertation. This tradeoff is not specific to the case of Madagascar, as shown by the works of West and Brechin (1991), Ghimire and Pimbert (1997), Brechin et al. (2003), Sunderlin et al. (2005) and West et al. (2006), and by the political ecology literature in general (Chapter 13). Some essential aspects of communities’ livelihood, such as food security, are often threatened. Communities can also be subject to coercive state control that disrupt their culture and limit their freedom. Hence conservation programs often lead to cases of human right violation (Chapin 2004; Campese et al. 2007), which may explain the strong tension between discourses elaborated by biophysical scientists, which sustain the logic of the Malagasy National Environmental Action Plan (Chapter 2), and discourses elaborated by the social sciences, which led to the alternative visions I presented in Chapters 12 and 13 (political ecology). Facing this problem, I take a simple position which consists in putting human rights above nature’s rights, for moral as well as technical reasons. Only with this condition can we find solutions that will save nature without forcing the weakest social groups to pay for it. Such solutions certainly exist. We just need to be open, imaginative and avoid misrepresentations and simplifications. The current discourses about environmental conservation make the implicit assumption that local people can pay a high price for nature conservation, or reveal an unawareness of the price they pay. They also assume that local communities have no political weight in the bargaining about nature. But this is just an illusion, as local people are those who can ignite fires. The overlooking of these elementary socio-economic realities, rather than the lack of ingenuity, as would contend Homer-Dixon (2000), may be the cause of failure of conservation policies. Hence, putting human rights above nature rights is not
665
only ethically satisfying. It will oblige us to face social issues, creating an appropriate framework for finding solutions.
3.4. Proposition of solutions 3.4.1. The role of legislation The enforcement of a legal frame which forbids certain practices is an important aspect of any environmental policies. Chapter 10 showed that the conversion of forest land into agricultural land is, overall, more profitable than sustainable resource exploitation. Chapter 11 showed that the main reason why the GELOSE approach fails is the lack of congruence between the interests of the parties. The main function of primary ecosystems for communities is to serve as a reservoir of resources to be exploited, but more importantly, of land, whereas their main function for external actors involved in the GELOSE is to maintain for future generations a reservoir of biodiversity, carbon and other natural resources. By overlooking this contradiction, the GELOSE created the illusion that development and conservation objectives could be achieved by the negotiation of appropriate resource management. This idealistic vision of communities able to live in harmony with nature must be abandoned in the case of Madagascar. Communities, if appropriately supported, can only slow down forest conversion, by closing access to migrants. In the long-term, Malagasy farmers will convert most remaining primary ecosystems to other land uses, just like European farmers did before them. This is why there cannot be any conservation policy without the enforcement of a ban on forest clearing. Farmers, however, have reasons to stop or slow down forest clearing. They experience a variety of consequences such as increasing difficulty accessing land when forest resources are limited; local climatic changes and modification of hydrologic regimes; loss of products they use to build their houses, fabricate
666
traditional drugs, complement their diet and sustain their life in case of food shortage; violation of ancestors’ land and aggressions of the spirits of the natural world. For all these reasons, they have a commitment to manage their resources in as sustainable a way as possible and to conserve some, if not all forests. But this can last only as long as pressure by population growth or migration renders complete clearing unnecessary. Nambena (2007) showed that when farmers cleared the forest in Beforona, they left behind forest patches, but that when the frontier was closed, these patches were cleared one after the other due to the increasing pressure on land (Chapter 3). In practice, enforcing the ban means that forest guards must tour the country and show the authority of the state, like they did during the first republic (Chapter 2). As argued by Blaikie and Brookfield (1987), there might be some cases where coercion is necessary to conserve natural resources (Chapter 13). Coercion, also, may be a cheap approach. Forest guards would not cost much in training. An uniforms, a GPS, a decent salary and touring indemnities would suffice to provide them with good working conditions. They would also establish a link between the state and the Malagasy realities, a link which has unfortunately been lost because the reality toward which state services are directed, now, is the reality of the donors, their meetings, workshops, reports and lobbies. The ban, however, must concern only clear cutting, not access to forest land for the collection and commercialization of forest products, including the extraction and sale of wood. It is clear that conversion of forest land is a major threat to biodiversity but the same cannot be said about the extraction of natural resources. Human utilization of forest products modifies ecosystems but does not destroy them. It can even increase biodiversity (Hladik et al. 1993). The main problem of the enforcement of the ban, as we saw, is that the poor and the powerless are always the first victims of coercion. In case of relatively
667
prosperous economies, like in northern Madagascar where profitable cash crops such as cloves or vanilla provide farmers with significant income, repressive environmental policies do not necessarily result in social drama (Laney 2002). Farmers can invest their income in intensification and adapt to the new context. But for communities that do not have these opportunities, such as the people in Ambodilaingo, the Tanala from Ikongo and Ranomafana and a significant part of the population of the Fianarantsoa and Toamasina provinces, repression means a deep economic crisis. It would thus be unethical and politically, socially and economically unrealistic to implement these repressive policies alone. The ban must then be associated with strong supports to the local economy, on behalf of the fact that human rights have to be put above nature’s rights. The ban will be the medium that will link environmental policies to political economy issues. If it fails to do so, it will remain illegitimate and inefficient. In sum, the ban will define the duties of local communities, while the support associated with it will express the recognition of their historical rights over forest land and of their dependence on forest resources. In practice, this means that forest guards must not tour the country alone. They must be accompanied by experts who will identify alternatives, such as sites for creating hydraulic infrastructure or other improvements. In many cases, alternatives will not be sufficient, which justifies an approach based on compensation, rather than mitigation. In a context where failure of supports to agricultural intensification or other alternatives is the norm, direct compensation is a moral necessity. If it is not provided, mitigations will continue to produce illusions of development that falsely legitimate repressive conservation policies. The next sections will detail the measures which could be taken based on these insights.
668
3.4.2. Mitigation by supporting agricultural intensification The challenge of agricultural intensification is to make it favor Boserupian change before farmers have no choice, i.e. before the forest is completely cleared or closed to access and before a Malthusian crisis occurs. Several types of support are possible to achieve this. As they have already been discussed at length in Part III, and especially in Chapter 7, I will just quickly review them here. Agricultural intensification always requires investments, at least in the form of labor, to constitute the new asset that will replace the one that is degrading. The faster intensification occurs, the more investments are needed. The first role of projects aimed at mitigating the ban on forest clearing should be to favor these investments. The most efficient way to do so is by supporting farmers to increase the new assets they pay more attention to, such as water resources for irrigation, home gardens or livestock. Chapter 7 showed the fast results that can be obtained by supporting the creation of irrigation schemes. In general, farmers already know which assets represent their future. In Ambodilaingo for example, young people are aware that finding a good spot to establish a home garden is essential to guarantee their future (Chapter 5). All households further perceive that creating paddy fields is essential when access to fertile soils suitable for tavy is limited. In case the key asset cannot be identified and the new agricultural practices are still unknown, the comparison with what happened in other regions of the world with similar conditions provides relevant guidance. Boserup (1965) analyzed the unfolding of land uses in relation to population growth. She showed that once the fallow period of slash-and-burn systems is too short, farmers develop annual cropping and multiple cropping systems based on crop rotation, plowing and manure application (Chapter 5). The integration of agriculture with livestock husbandry substitutes for fallowing as a way to reproduce fertility. The challenge, hence, is to accelerate these dynamics by injecting capital in places where
669
investment capacity is low. The farmers of Beforona are already engaged in this change in the most densely populated villages (Chapters 3 and 6). They show us a pathway for intensification which may not be the optimal one, but which most farmers can afford if appropriately supported. More fancy technologies such as agrobiology and agroforestry could play a role if properly designed, but it would be highly risky to expect change to occur only from them. Besides these supports to intensification, other measures can have a significant impact, as I showed in Chapter 7. Roads can be created to facilitate access to markets, although in most cases this favors the intrusion of outsiders to the detriment of marginalized farmers who moved to the frontier. Establishing land tenure security, prior to the creation of the road, can be a way to remedy this. But the securing of land tenure must take account of the diversity of tenure modes which are associated with the diversity of land uses. Micro-credit can also have an impact, as well as the providing of basic services such as education, health or veterinary services. Eventually, agricultural intensification can be technically supported. But in order to avoid the generalized failure of extension efforts by projects (Chapter 6), the new techniques have to be created in the farming system context. On-farm research usually fails to do so because on-farm trials, in general, reproduce on farmer plots the artificial conditions of experimental stations. What matters is to implement research in the farmer’s conditions, not in the farmer’s location. The agricultural contest implemented by the NGO Zanaky ny Ala is an example of research implemented in farmer’s conditions. In this contest, metis was the type of knowledge that mattered and that farmers utilized and created.
670
3.4.3. Subsidies We saw that economic supports, rather than technical ones, are essential for the development of more intensive land uses in eastern Madagascar. We further saw that direct compensation must be preferred to mitigation in order to guarantee that farmers receive a benefit in exchange for their engagement in respecting the ban. Providing direct subsidies may thus be an efficient way to favor agricultural intensification. Subsidies could take the form of cash or food. In the context of Beforona, food may be more appropriate because cash circulation is still embryonic, especially in remote areas. But subsidies in the form of cash, on the condition that they would represent a significant but reasonable amount, could also be tried. Excessive amounts could disrupt the economy by introducing behavior that would not be economically rational. Farmers could behave like someone winning the jackpot at the lottery. But if the amount matched with what farmers need to solve ordinary economic problems (for example, buying rice during the food shortage period, buying animals to raise, paying laborers to develop irrigation schemes), they would have no reason to waste this money. At the scale of one household, a subsidy of 50 to 150 USD per year appears to be a reasonable amount to solve most ordinary economic problems. An objection often encountered with regard to subsidies, among conservation actors, is that farmers would receive a payment for doing something that is already prohibited (clearing forest). I reject this objection because it is based on the assumption that the ban on forest clearing is legitimate, which I disagree with as I said earlier. There are several reasons for saying that the ban is not legitimate. First, the benefit of stopping deforestation is received at a global level whereas its opportunity cost is born at the local level. For equity reasons, it cannot be accepted that local people continue to bear this cost. Second, the local people who pay the opportunity cost of stopping deforestation often belong to the weakest social groups. Their
671
presence on the forest frontier is usually the consequence of a marginalization process that has lasted for decades or centuries. Those who can immediately enjoy the benefits of stopping deforestation, on the other hand, are citizens of the richest nations. Third, deforestation is not the only global environmental issue. There is a law that forbids deforestation in Madagascar but no law limits the use of fossil fuels in any place in the world, which raises again equity issues. Fourth, reducing the use of fossil fuels (or more modestly, increasing taxes on them) would not lead to imperiling those concerned by this measure. Life comforts could be threatened but not basic needs. For the households that live on the forest frontier of Madagascar, it is the real livelihood, even the capacity to produce sufficient food, which is put into peril by the ban on clearing. Fifth, the burning of fossil fuels may be as significant a threat to Malagasy biodiversity as forest clearing. Some forests could escape from clearing due to infertile soils or to other factors that would render them unsuitable for agriculture. In the case of global warming, to the contrary, all ecosystems could be impacted. According to IPCC (2007), the risk of extinction will reach 20 to 30% of animal and plant species on Earth if world temperatures increased by 1.5 to 2.5 °C. There are already signs on all continents that global warming affects plant and animal species (Forcada et al. 2005; McClean et al. 2005; Wilson et al. 2005; Wing et al. 2005). There is also a large array of evidence showing that coral reef are threatened by carbon release and water temperature increase due to global warming (Glynn 1991; Kleypas et al. 1999; Hughes et al. 2003; Donner 2005; Lesser 2007). In the case of Madagascar, Ingram and Dawson (2005) showed that the first impact of global warming on ecosystems are already being felt. In sum, asking the weakest social groups, those whose survival depends on natural resources, to bear the main cost of conserving biodiversity is totally unethical. This is a very naïve advocacy but there is no other choice than being naïve or cynical
672
in this case. To compensate the cost of restricted access to forest land is both an ethical and a technical necessity. It is justified by ethics as well as efficiency. It is not only a way to respect the right of the people. It is also a way to guarantee that nature will be better preserved, because a nature that would be a vector of inequity would become an enemy to the people. It would be burnt, sooner or later. One can thus wonder why subsidies have not been implemented earlier. There may be a simple reason. Granting subsidies is usually a political decision which results from a negotiation between those who pay and those who receive the subsidy. In the case of slash-and-burn cultivation in Madagascar, there cannot be negotiation because the potential beneficiaries are not politically organized. They express the need to receive supports when they are visited, as did the farmers of Tsinjoarivo and Ambositra (Zafimaniry people), when they asked for compensation for the restriction on access to land which resulted from the implementation of management transfers (Chapter 11). But local communities lack efficient modes of political representation to claim subsidies in a more formal way. Associations, such as the COBAs, created for implementing GELOSE contracts (Chapter 11), could serve as a platform for this claim, but unfortunately often lack efficiency and representativeness. They are efficient, sometimes, but to favor the enforcement of the ban rather than to advocate for compensation (Chapter 11). By raising the issue of political representation, I am touching on an issue which I have not yet raised explicitly, but which may be central. In fact, projects may serve as a channel for the depoliticization of development and environmental issues. As Ferguson (1994) argued, the development apparatus fails to achieve its explicit objectives, but may successfully achieve implicit ones. Central in these implicit objectives may be the depoliticization of development, which is achieved by providing the illusion that problems can be fixed by technical solutions. The will for this
673
depoliticization may be the main cause of the maintenance of received wisdoms and of the mismatch between representations and realities. If the mismatch disappeared, the explicit objectives could be achieved to the detriment of implicit ones. Communities would develop, would empower, and could oppose the power of the state and international organizations. They would demand for compensation for the ban on forest clearing, for example. From this perspective, subsidies are a danger for the development apparatus, which may explain why there is so much reluctance to establish them, as shown in the case of Madagascar when organizations involved in NEAP refused the concept of compensation (Chapter 2). Subsidies would transform the anti-politic machine (Ferguson 1994) into a political one. Farmers receiving subsidies could develop new land uses according to their own strategies. They could free themselves from the discourses produced by projects and from the practices which attempt to reshape their productive basis and their social organization. If this thesis was true (and I believe it is), political activism, not science and knowledge, would be the solution to solving the environmental stalemate in Madagascar. This dissertation would then be useless unless it was read by activists. Granting subsidies further raises the question of who is going to pay for them. I would reply that if we, the global society, consider that the Malagasy biodiversity has more value than the agricultural production that is obtained after the forests are cleared, then we have to be ready to pay a price equivalent to the opportunity cost of losing this agricultural production. At present, this price is low, at least in countries like Madagascar, because farmers who clear forest still practice traditional forms of agriculture that have low productivity per unit of land. Considering an average yield of one ton of rice 383 per hectare and an average fallow period of five years, we can 383
In reality, tavy produces more than one ton per hectare during the first cultivation cycles and less than that once the land is degraded, and rice would not be the only product. I use here a very gross estimate because my intention is just to encourage more precise calculation aimed at exploring subsidies in more detail as a strategy for conservation.
674
quickly calculate that the nonclearing of the remaining 5.5 million hectares of the eastern rain forests of Madagascar could be compensated by 440 million USD. 384 The annual interest rate of a capital of about 7.5 billion USD would suffice to pay this annual compensation forever.385 This is much more than the half billion USD spent by NEAP up to now but is about the cost of a few subway lines, 386 which is quite cheap for the Malagasy biodiversity. This cost would be much lower if we considered only the forests directly threatened at present, i.e., the periphery of the remaining forest fragments. According a World Bank note (2007), the cost of restrictions on access to land concerns mostly villages located less than five kilometers away from a protected area limit. The same note proposes a formula to calculate the total area affected by these restrictions. 387 We can apply this formula to the whole eastern rain forest of Madagascar. If the 5.5 million hectares remaining was divided into 173 protected areas, the area affected by restriction would be 9,273,237 hectares, or 92,732 square kilometers. 388 With an average population density of 20/km2 (hypothesis of the World Bank note), this would represent 1,854,640 people, or about 370,000 households. A compensation of 100 USD per year per household would cost about 37 million USD per year in total. A 384
5,500,000 hectares (surface covered by rain forest) / five years fallow x 1000 kilograms (harvest) x 4000 FMg (price of one kilogram of rice) / 10,000 USD (change rate FMg-USD). 385 With an interest rate of about 6 %. 386 In New York City, a subway line project on Second Avenue, having a cost of 15 billion USD, is in discussion but failed to find a budget (http://www.thevillager.com/villager_34/thesecond.html.) Still in New-York City, an extension of line 7 (two new stations) was recently approved for a cost of 2.1 billion USD (http://www.nypost.com/seven/09292006/news/regionalnews/subway_station_to_cost_2_1_bil_regiona lnews_jeremy_olshan___transit_reporter.htm). 387 This formula is “area less than five kilometers off the protected area limit = 0.886 x area of the protected area + 25,480 hectares (World Bank 2007). 388 According to the World Bank (2007) note, there will be 143 protected areas in Madagascar in 2008, for a total of 4,539,000 hectares. This represents an average of 31,741 hectares per protected area. The eastern rain forest now covers 5,500,000 hectares. If it was divided into protected areas having the same average size (31,741 hectares), this would constitute 5,500,000 / 31,741 = 173 protected areas. The formula of the World Bank (2007) note can now be applied to calculate the area affected by mitigation measures if the whole rain forest was protected: 173 x (0.886 x 31,741+25,480)= 9,273,237 hectares = 92,732 km2.
675
trust fund of about 600 million USD with an interest rate of 6% would suffice to provide this annual compensation. This is much more than the trust fund (about 50 millions USD) constituted by the donors to sustainably finance the management of about 2 millions hectares by ANGAP, but is not much more than the total cost of NEAP over 15 years. In sum, the first figure (a 7.5 billion USD trust fund generating 440 millions USD per year) represents the upper limit of conservation cost, i.e., a case where the conversion of all the Malagasy rain forest into agricultural land would be compensated, whereas the second figure (a 600 millions USD trust fund generating 37 millions USD per year) represents the immediate cost of compensating land conversion by farmers living around the forest. If we extrapolated these figures to the 212,289,100 hectares of the 25 world biodiversity hotspots (Myers et al. 2000), it would be necessary to create a trust fund of about 289 billion USD 389 (generating 17 billion USD per year 390 ) in the high figure and 16 billion USD 391 (generating 102 million USD per year 392 ) in the low figure. The real cost would be indeed between the two figures because the opportunity cost of non clearing is more elevated in certain countries, like Brasil 393 . If the 500 millions richest citizens would contribute, this would represent an annual “biodiversity tax” of about 46 USD/year/person 394 in the high figure, and 0.2 USD 395 in the low figure. Again, this is relatively cheap for conserving the heritage that nature bequeathed to us. Even the highest figure (17 billion USD a year) appears reasonable if we compare it to other type of subsidies
389
212,289,100 hectares / 5,500,000 hectares x 7,500,000,000 USD = 289,485,136,363 USD. 289,485,136,363 USD x 6 / 100 = 17,369,108,181 USD. 391 212,289,100 hectares / 5,500,000 hectares x 440,000,000 USD = 16,983,128,000 USD. 392 16,983,128,000 USD x 6 / 100 = 101,998,768 USD. 393 In Brasil, the opportunity cost of nonclearing is much more elevated than in Madagascar because forest is cleared for the development of highly profitable agriculture systems, such as intensive soybean monoculture for export. On the other hand, a significant part of the forest is not directly threatened. 394 17,369,108,181 USD / 500,000,000 persons = 34.74 USD. 395 101,998,768 USD / 500,000,000 persons = 0.20 USD. 390
676
already put in place. To give just a few examples, in 2002, US farmers received 3.9 billion USD as subsidies in the season 2001/2002 for cotton alone (OXFAM no date). Between 1999 and 2001, OECD 396 countries spent an average 6.35 billion USD per year to subsidize sugar production (La Vina et al. 2007). In 2002, agricultural subsidies in OECD countries “totaled $235 billion, of which $100 billion was accounted for by the European Union and $40 billion by the US” (OECD 2003). Developed nations
engage these expenditures to guarantee their food security, maintain their competitiveness in the global market and slow the abandonment of agricultural land and the desertification of plains. But doing so, they also create unfair competition with farmers in developing countries (OXFAM no date), which decreases their capacity to invest and may increase their dependence on natural resources. Subsidies having perverse environmental effects also exist in other sectors. Myers (1998) estimated that they total 1.5 trillion USD per year, including 250 billion in the agricultural sector. On the other hand, according to data gathered by (James et al. 1999), about 6 billion USD are spent every year for managing protected areas world wide, including 1 billion only in the tropics 397 (Australia excluded) 398 . If subsidies were to decrease, these sums of money could be reallocated to biodiversity conservation programs (James et al. 1999), for example to compensate for the ban on forest clearing, which could enable the internalization of both social and environmental externalities. According to James et al. (1999), an additional 2.3 billion USD per year could fill the financial gaps of existing reserves (about 1.2 billion in the tropics), 14.5 more billion (including 7.5 billion in the tropics) would allow to extend the protected area network in order to
396
Organization for Economic Co-operation and Development (OECD).
397
In Latin America and Caribean, East Asia, Asia, Sub-saharan Africa and Pacific According to Myers (2000), 400 million USD have been spent annually on the 25 biodiversity hotspots from 1990 to 2000.
398
677
strictly protect 10% of the land, and 49 more billion (worldwide) could be used to compensate the communities whose access to resources is or would be restricted. Coming back to Madagascar, two to seven or eight billion USD is an elevated price in comparison with the amount currently spent for conservation. But the aid system faces two problems that play in favor of this new strategy. First, donors are more concerned by insufficient disbursement than by excessive expenditures. Second, the criterion to consider is cost-efficiency rather than total expenditures. NEAP is already at mid-term of its third and final phase and has not proven to be successful if we consider its overall impact on forest clearing (Chapter 1). A new estimation will be available soon and may provide a more encouraging figure. It has already been observed, indeed, that deforestation decreased after the antifire campaign of 2002. But this result, if confirmed, cannot not be called a success. The campaign just proved that when a few farmers are arrested for practicing tavy and when messages are broadcast nation wide to create fear, people stop clearing forest, at least for a few years. But considering ethical issues and the economic, social and political cost of the campaign (Chapter 5 and 6), this cheap strategy cannot be regarded as a success. A solution based on subsidies, to the contrary, creates a win-win situation. It is necessarily more efficient because if this win-win situation did not occur (if farmers continued to clear the forest), then the subsidy could be stopped and the capital be restituted to the donors. Subsidy for conservation, however, also raises ethical issues. To pay such a cheap price for conserving primary ecosystems can be perceived as a way to take advantage of poverty (Karsenty 2007). It is true. But in comparison with the current situation where clearing is forbidden with no compensation, more equity is, however, achieved. Subsidy for conservation thus remains a valid approach from a pragmatic perspective. In fact, the approach is increasingly paid attention by the international
678
community, under the appellation payment for environmental services (PES). 399 I showed in Chapter 10 that the ASB Consortium, in its conclusions (Sachs 2005), adopted a paradigmatic change where payment for environmental services replaced intensification as a privileged strategy to stop deforestation. Ferraro (2002a; 2002b) sees PES as a more cost-effective approach than the indirect support provided to communities through rural development projects, an argument which is consistent with the outcomes of Chapter 6. Wunder (2001; 2007) suggests that payments for environmental services offers opportunities to create situations where both poor households living on the forest frontiers and the international community can win from conservation. The approach, thus, will gain increasing momentum on the international agenda. The challenge is rather to direct the payments to the communities in need of compensation and avoid their capture by elites, as has occurred with other resources in the past. The case of the VMCRCC carbon sequestration project, in Chapter 6, unfortunately showed that this challenge will be a difficult one. The subsidy, furthermore, does not imply eliminating other economic activities. The core of the economy can remain agriculture and/or sustainable extraction of resources. The subsidy will be aimed at compensating for the ban on clearing, not other activities that are compatible with conservation or can be rendered compatible by improved management. Farmers could actually invest their subsidy in the development of their agricultural activities outside the forest and in the sustainable harvest of forest resources, including wood. The payment would be just a bonus, aimed at compensating the only point where no congruence of objectives can be
399
I prefer the term subsidy because PES are usually designed to compensate for more specific services. In Madagascar for example, PES are put in place for the monitoring of animal population (Durbin et al. 2001). Subsidies aimed at compensating the nonclearing of forest, however, can be regarded as PES because communities would have to guarantee that no clearing occurs on their land, which implies patrolling the area and can be regarded as a service.
679
achieved between local and global actors: biodiversity conservation and the nonconversion of forest into agricultural land. One could still object that future agricultural intensification will allow the land to generate more benefit than what is currently possible with slash-and-burn systems, and that subsidies could be used as a way to buy a “no development” situation (Karsenty 2004). I would agree with this objection if payment was fixed once and for all, and was associated with a ban on resource extraction, in addition to a ban on clearing. But the payment can be renegotiated over time in order to adjust to the opportunity cost of nonclearing. And if the ban concerned only forest clearing, local development could still occur. Development would just be based on the optimal use of biodiversity rather than on agriculture, or on a combination of resource extraction and agriculture. The discussion about the case of the VMCRCC project, in Chapter 6, allowed exploring this avenue.
3.4.4. Conclusion This section proposed the articulation of three strategies: the enforcement of the ban on forest clearing, support to agricultural intensification by injection of capital (which includes the creation of infrastructure), and subsidies, which are aimed at compensating the opportunity cost of not clearing primary forests. If the enforcement of the ban alone was applied, conservation policies would be unethical. If the ban was complemented by support for agricultural intensification, the situation could still remain illegitimate, or unfair, because failure is the norm concerning support for rural development. Rural development projects would continue to provide the same illusions aimed at legitimating conservation programs, as did the ICDPs. Compensation in the form of subsidy is thus a moral necessity if the ban is to be enforced. Indeed, support for intensification, in the form it takes in rural development
680
projects, could even be abandoned. Farmers would certainly be more efficient in the way they use money than projects that would be granted the same amounts.
4. Institutional approach Once a series of measures have been defined, it remains to determine the way they could be implemented by existing institutions. A first issue to consider is that idealized intellectual constructions, such as the one elaborated in this dissertation, can lead to unintended strategies when translated into actions. Institutions have their own understanding and impose their own logic, which can significantly distort the approach. In other words, this dissertation produced knowledge but development and conservation policies and practices are determined by much more than knowledge. Politics, i.e., the making of choices aimed at satisfying certain aims and interests, may be more determining. Chapter 11 showed how community based natural resource management in Madagascar was conducted away from its early ideal by the dominance of conservation discourse and top-down logic. The “anti-politic machine” described by James Ferguson (1994) provides another example of how implicit aims can detour the aid system from the realization of its explicit objectives. A second issue is that no representation can be perfect. The mismatch between representation and realities can never be reduced to zero, and there is no way to know for sure where a representation is located, between a perfect match and a total mismatch. The representation elaborated in this dissertation does not make the exception. How, then, to avoid that the outcomes of this dissertation will be wrongly distorted or perverted, if they are some day appropriated by the aid system? How, at the same time, to guarantee that they will be adjusted and improved, for aspects for
681
which they may be misleading? Popper (1966) helps in answering these questions. As I showed in Chapter 14, this philosopher defined a line of demarcation between science and other modes of inquiry but did not consider that science had a monopoly on the production of meaningful knowledge. Acknowledging that science alone could not decide the fate of human societies, he extended his epistemology to political philosophy. In The Open Society and its Enemies (Popper 1966), he developed two concepts which I will introduce here: the “open society” and “piecemeal engineering.”
4.1. The open society For Popper (1966, 173), the closed society is a “magical or tribal or collectivist” society comparable to an “organism.” The identity of each citizen is determined by the relations between one another. In order to guarantee that everyone plays its role, strong institutions such as castes and strong beliefs such as taboos guarantee that the ties remain strong. This provides a stable social environment which guarantees happiness, as long as citizens do not have access to forms of knowledge that could lead them to question the system. In such a closed society, there is no Cartesian anxiety. Myths and religions close knowledge, close the society on itself, which minimizes existential questioning, instability and insecurity. For Popper (1966), the transition of Western civilization from a closed to an open society started when the ancient Greeks invented democracy and is still ongoing. During this transition, unquestionable rules were terminated, leading to more responsibilities and insecurity. When the Greek society started this transition, it realized that the price of more freedom was less happiness, entered into a crisis, and attempted to find new ways to provide solid ground for its citizens. Philosophy played this function. It “replaced the lost magical faith by a rational faith” (Popper 1966, 188). Plato’s myth of the cavern played a central role in this endeavor. Based on this
682
myth, Plato committed to designing a perfect society and to arresting social change, allowing a return to a closed society. This myth provided the foundation for setting the epistemological framework of positivist science. In this perspective, science can be regarded as an attempt to return to the more comfortable and reassuring closed society. But for Popper (1966), there is no way to return to the closed society. “We can never return to [its] alleged innocence and beauty” (Popper 1966, 200). Once citizens are freed from the primordial myths and have access to more objective knowledge, closed systems of knowledge can only create utopias which lead to dictatorship. We must then “go on into the unknown, the uncertain and insecure” (Popper 1966, 201). It is interesting to notice that Popper’s (1966) The Open Society and its Enemies starts with a criticism of Plato’s myth of the cavern, as Latour’s (1999) The Politics of Nature does. But the concepts of open and closed societies led Popper to propose an alternative different from that of Latour (1999). Instead of proposing the constitution of large collectives that would deliberate democratically and close controversies, Popper (1966) contends that policies should abandon the objective of closing controversies. The search for happiness, for example, should be left to the private sphere because defining the good for each individual would create the risk of imposing a certain vision of existence. More modestly, policies should be designed to alleviate suffering. This leads us to a second key Popperian concept: piecemeal social engineering.
4.2. Piecemeal social engineering Popper’s (1966) “piecemeal engineering” consists of favoring incremental social progress aimed at alleviating suffering where it is more acute. The objective is more modest than the creation of happy societies. An ideal can still exist but it is regarded as a guidance, a direction to take, not a goal:
683
The politician who adopts [piecemeal engineering] may or may not have a blueprint of society before his mind, he may or may not hope that mankind will one day realize an ideal state, and achieve happiness and perfection on earth. But he will be aware that perfection, if at all attainable, is far distant, and that every generation of men, and therefore also the living, have a claim; perhaps not so much a claim to be made happy, for there are no institutional means of making a man happy, but a claim not to be made unhappy, where it can be avoided. They have a claim to be given all possible help, when they suffer. The piecemeal engineer will, accordingly, adopt the method of searching for, and fighting against, the greatest and most urgent evils of society, rather than searching for, and fighting for, its greatest ultimate good. (Popper 1966, 158) Popper (1966) opposes this approach to “utopian engineering,” which is an “attempt to realize an ideal state, using a blueprint of society as a whole” (Popper 1966, 159). Utopian engineering matches quite well with Scott’s (1998) high modernism. It “demands a strong decentralized rule of a few, and … therefore is likely to lead to a dictatorship” (Popper 1966, 159). The parallel between Popper’s epistemology (Chapter 14) and political philosophy is clear. The way he considers the truth in his epistemology is similar to the way he considers happiness in his political philosophy. The alleviation of sufferings by policy makers plays the same function as the falsification of theories by scientists. The progress of societies cannot be blocked by the imposition of a certain conception of happiness, in the same manner as the progress in science cannot be blocked by the imposition of pretended truths. In the open society, rationality and science have a role to play, but for debunking false representations rather than for building supposedly perfect intellectual edifices and establishing supposed truths. Outlining these concepts was necessary because the intellectual construction elaborated in this dissertation is not intended to be a new blueprint for the design of a perfect society. It is just intended to provide a critique of existing representations and new materials for future piecemeal engineering. No revolution is expected, nor wished to occur on the basis of the material I have provided, whether this material is wise or
684
not; whether the representation I elaborated matches or not with the reality. I just produced a new representation which, as any representation, would become a danger if it would be transformed into a new fashion or dogma. But as I did my best to maximize the match between this representation and the reality, I believe that it is worth translating it into policies, programs, and projects. I thus propose a new concept, the open project, which, I hope, will find a favorable ground for being translated into new practices.
4.3. The open project Consistent with Popper’s piecemeal engineering, new approaches must not pretend to do more than correct a problem that is currently a cause of failure or suffering. Mainstream approaches, which are determined by the current state of the society, must not miss the chance of having that problem solved, but their whole edifice of practices and representations must not collapse on behalf of this. In other words, the aid system may be unfair and unsustainable, but may be consistent with both the societies which sustain it and the societies it sustains. If it collapsed, these societies may lose their relative stability and fall into chaos. They would need new myths, new blueprints to be rebuilt, but Popper (1966) showed that new blueprints elaborated on clean canvas would not lead to better results. Hence, the system of practices and representation that currently prevails in Madagascar has to open itself to criticism and to experimentation with new practices. This will provide supplementary evidence that will lead to the falsification or corroboration both existing and new representations. Based on the new knowledge and experience that will be produced, the aid system will be able to engage in significant reforms. The world will then be used as a laboratory for cautious experiments, not as a battleground for ideological fighting.
685
One could consider that this approach is unethical because citizens are not laboratory guinea pigs. But the world is already a laboratory. The problem is rather that this fact is not recognized, which results in dogmatic, ill designed and poorly monitored experiments. The deleterious effects of these experiments can be significant but are not recognized. The structural adjustment programs implemented by IMF were a worldwide experiment that lasted a decade and cost several billion USD. Conservation and development programs can be viewed as other sorts of global experiments. If we recognized that we are indeed implementing nothing other than experiments, we could put in place more careful monitoring, design experiments and policies more cautiously and collect feedback from the people we target. Our responsibility would be completely engaged in these experiment/policies and there would be no certainty that could serve as a refuge. We would, eventually, become “virtuoso social actors” (Flyvbjerg 2001, 2) doing phronesis (Part I). This responsibility would oblige us to debunk mythologies, established truth, received wisdoms and excessive faith in doctrines. Appropriate policies may progressively emerge during this process. One can wonder why this has not been done before. The reason may be that the certainty of established truth is preferred to the uncertainty of experiments. If we made explicit that we are experimenting on the world, rather than pretending to rule it in an enlightened way, we would have to face the Cartesian anxiety (Chapter 14). If uncertainty is accepted, there is unfortunately no way to suppress the Cartesian anxiety. I contend, therefore, that we have no choice but to accept facing this anxiety. Unless we want to return to a premodern society and be bound again by myths and taboos? If we did that, our mistakes would result in more deleterious effects because we have more potential to impact our planet and our peers than our predecessors had,
686
and because the imbalance between techne on metis reduces our opportunities to learn from contextualized experience. In practice, and getting back to Madagascar, this approach implies that donors will have to open a second window in their practices. This will be a minority window. It is not necessary to invest hundreds of millions USD to test new ideas. But this second window will have to be totally open. It will have to receive projects established without guidelines, without frames. Only simple objectives, such as stopping deforestation while alleviating poverty, will have to be stated. Magical keywords such as participation, gender, governance, multidisciplinary approach, community based natural resource management, socio-organization, market, micro-credit, family planning, food for work, watershed management, integrated management, erosion control, agroforestry, fireless alternatives, zero tillage, mulching, composting, income generating activities, fish farming and bee keeping will not be required. The only criteria will be the overall consistency of the grant requests, the perspiration of the real world across and between the lines, the genuine commitment of the organization that would run the project and the impact in comparison with the cost, which will determine whether the project will be scaled up or not. All approaches would be tested at small scale, for example in a village, before being scaled up, cautiously, at communal, regional and national scales. Risks will have to be accepted in the move to the unknown of new practices, but will be minimized by the small scale of the first tests and by cautious monitoring. Expectation will concern the impact of activities and nothing else. There will be no other indicator than the state of the society and the environment before and after the project. Impacts will be carefully monitored and measured and the project will be modified, even stopped, in case of negative impacts. Constant adjustments would minimize the risk of stopping before completion of a project, meaning that there will be no rigid budget lines.
687
I have made propositions that represent a significant move away from the current practices. This move is the consequence of entering into the world of uncertainty. But institutions which manage aid should not fear this change, because it would concern just a second window, while most activities would continue to run as usual. No revolution is expected, precisely because the real idea of revolution is rejected. But the experiments conducted in the second window would provide new ideas, new results, new cases of successes and failures that would help to progressively reform practices in the first window. No major disturbance could result from this because the second window will have done the job of taking the main risk and of reducing uncertainty. The first window will just need to be open to a dialogue, allowing implementation, as efficiently as the reality and the discourses would allow, of what Popper (1966) called piecemeal engineering. Indeed, this second window may be created soon, by the appearance of new actors in the aid system, rather than by the opening up of existing actors. Private foundations, whose financial power now matches that of powerful organizations such as United Nations Agencies, may already be on the verge of creating this second window.
5. Conclusion In Section 2, I contended that the clue to implementation of more efficient development and environmental policies is to develop more contextualized forms of knowledge. Techne and metis are two complementary modes of knowledge production that need to be articulated together if we are to achieve this. Techne is necessary to isolate elements of the world, compare them, classify them and derive general rules. Metis is necessary because uncertainty always remains behind the general rules, even those most strongly established. Context, by the apparent disorder it bears, is the
688
counterpart of the order incarnated by rules. Techne is justified by the necessity to put some order in representations, while metis is justified by the necessity to address the disorder that surrounds this order. These two concepts may have a dialogical relationship (Chapter 14), like order and disorder have. Behind this dialogy, there may also be the dialogy of knowledge and ignorance. In an apparent paradox, what we often call ignorance and which I prefer to call uncertain knowledge, may need to be integrated into knowledge if we are to stop producing real ignorance. 400 In other words, some general rules must be identified, while flexibility must be allowed to question and constantly adjust these rules. Several of these general rules, applicable at least to the case of Madagascar, have been identified in this dissertation and served to make the propositions of Section 3. I will summarize them here. The first rule is that there is no “natural” harmony between humans and nature. Harmony can exist but is not intentional. It is a matter of circumstance. For this reason, if we are to conserve biodiversity, there has to be a ban on forest clearing and it has to be enforced, which constitutes the second rule. This ban, however, must not prohibit resource extraction, even the harvest and sale of wood, because granting access to these resources to local communities is a condition for the social acceptance of the ban and for minimizing its negative economic impact. The third rule is that the ban on clearing, although necessary, is not legitimate, because it is just one necessary measure among a larger series of environmental measures which are not taken (for example, significantly increasing taxes on fossil fuels, investing billions in renewable energy, and controlling the illegal international trade of tropical woods). The ban on forest clearing is just more easily subscribed to than these other measures because it has, supposedly, a lower political and financial 400
As I said earlier, I prefer to define ignorance as certainty about false statements. Uncertain knowledge is hence not true ignorance and this is why it has to be integrated to knowledge.
689
cost. This leads us to the fourth general rule: being illegitimate, the ban must be compensated. This compensation can be achieved by support for agricultural intensification. This support could be the continuation of rural development activities we are used to implementing. But as failure is the norm at this level, a significant departure is necessary, which leads to the fifth general rule: support for agricultural intensification has to shift from technical (training) to economic (credit or subsidies). In the case of Madagascar, these economic supports can help farmers to capitalize on new assets which will replace the assets (forest biomass and fertile forest soils) they lose as a consequence of the ban. This brings us to a sixth general law: if we brought some water to Malagasy farmers, it would decrease their dependence on forest biomass. This is not a discovery indeed. We just need to travel across the country to observe that Malagasy farmers move across the island in the search of two main assets that allow them to produce rice and sustain their livelihood. These assets are biomass (forest in free access which they can clear) and water (bottom land which can be irrigated and improved). They constitute the first and the second agricultural frontiers, as we saw in Chapters 1 and 5. The challenge, therefore, is to displace pressures from the first to the second frontier. 401 Creating hydraulic infrastructure could be a straightforward way to achieve this. But for this to work, it would need to be guaranteed that the poorest farmers would benefit from land improvements.
401
The potential for achieving agricultural expansion is indeed huge in Madagascar. Vast territories in the west and the middle-west could support much higher population density if infrastructure were created, security guaranteed, and subsidies for installation provided. These regions, however, are already occupied by groups of people who practice extensive livestock husbandry and who are already marginalized by waves of migrants. They resist by igniting fires and making raids to steal zebus. Expansion in these areas will hence face important social problems and improving land tenure security in favor of autochthonous communities will be a prerequisite. As I did not address the case of western Madagascar, I cannot provide more insight on these issues, which constitutes one of the shortcomings of this dissertation (a limit to my ideal of comprehensiveness).
690
Otherwise infrastructure would, to the contrary, contribute again to their marginalization, which leads us to the two last rules. As we saw in Chapter 5, communities in a situation of crisis live in a day to day economy and cannot invest significant effort in terracing, digging canals and other work required to create irrigation schemes or improve land in general. Experience shows that when irrigation schemes are created, the most competitive groups appropriate them, to the detriment of the weakest ones, who then abandon the land and move to the forest. If we are to allow marginalized farmers to get their share, we have to entitle them improved land, which constitutes a seventh rule, but we also have also to subsidize the factor of production that is available to all of them: labor force. This leads to the eighth general rule: energy must be subsidized. In practice, this means to purchase food for communities (or give money to buy this food) in exchange for the realization of land improvements. But to the contrary of the strategy I recommended to the NGO Zanaky ny Ala, farmers should be allowed to choose for themselves what type of land improvements to implement. And finally, the ninth rule is that subsidies should be preferred to mitigation measures because these mitigations usually have very low impact. Only in this way could farmers be guaranteed to receive compensation for the restriction of access to land. In sum I proposed in this chapter a major paradigmatic change that would lead to moving from technical to economic supports and from closed to open projects. Concerning the second point, I proposed to open a second window in order to guarantee radical changes without putting the existing aid system at peril. But for this second window to work, more opening will be necessary, not only at the level of relations between projects and beneficiary farmers, but also at the level of relations between donors and Malagasy institutions. These are the conditions for the design of
691
new projects that will escape from communicative illusions and from the many development buzzwords that close discourses on themselves, even when these projects or their activities are designed in a village meeting or in an international workshop. In order to achieve this opening, it is necessary to grant more autonomy to the “beneficiaries.” This does not mean that donors would have no word to say. Money is not free and cannot be obtained without commitment to something. But once these rights and duties are negotiated and agreed upon, the recipient would have to take responsibility and find his own ways to achieve results. In case of failure, he would know that the grant would be stopped. But once the rules are clear, he could use his own contextualized knowledge, his own combination of metis and techne, to design the most appropriate strategies. This would apply in the same way at the community level (the simple rule being the ban on forest clearing) and at the state service level (the simple rule being an objective measured in terms of impact, not in terms of realization). To make this work, it would be necessary to fix simple and more realistic objectives (to stop forest clearing, rather than to stop tavy) and simple and straightforward strategies (to tour the country rather than to create management transfers, or to build irrigation schemes rather than implement integrated watershed management). A supplementary advantage of this approach (but also a threat to the aid system) is that it does not need much external expertise. Success will mostly depend on the experience that teams of hydrologists, masons and forest guards will accumulate while touring the country and creating and maintaining land improvements, and on the opening of postal or communal offices where farmers will receive their subsidies. This is indeed a chance, because it is a guarantee that the Malagasy people, both the local communities and the state services involved, will hold development challenges in their own hands. This does not, however, mean that there
692
would be no watershed management and that tavy would be practiced forever. It would just mean that metis would do most of the job, beyond the enforcement of the ban, the construction of dams and roads and the distribution of subsidies.
693
CONCLUSION
I provided three syntheses of my outcomes in this fourth part. A synthesis in the light of other scholarly work (Chapter 13), a synthesis in the light of epistemological considerations (Chapter 14) and a synthesis in the form of a program of action (Chapter 15). In Chapter 13, I compared my outcomes to the narratives developed by political ecology, a discipline to which I may indeed belong, even if this was not intended. The marginalization thesis and the control thesis are central narratives in this discipline. This dissertation can thus be regarded as a supplementary case study corroborating these narratives. But Chapter 13 also showed the limits of political ecology and its difficulty addressing biophysical realities and integrating the outcomes of biophysical sciences. These difficulties can be explained by the influence of relativist epistemologies, which lead to conflation of concepts referring to distinct ontological categories, such as nature and culture. This statement forced me to raise again epistemological questions. The work of Popper (1963) provided me with appropriate tools and concepts to escape both positivist and relativist traps (Chapter 14). I reframed my outcomes in this new framework, which allowed the proposal of more comprehensive representations, i.e., theories with higher verisimilitude. The lesson learned doing this exercise is that scientists should attempt to create more inclusive theories in order to depoliticize their work. The dialogy principle of Morin further argued in favor of considering as distinct and interdependent some principles which are too often conflated (nature and culture) or considered as opposed (equilibrium and nonequilibrium). In Chapter 15, I translated my outcomes into a series of measures aimed at stopping deforestation in Madagascar without requiring the weakest social groups to
694
pay the price for it. These proposition mostly consisted of enforcing the ban on forest clearing, supporting agricultural intensification by providing significant investments, and compensating the opportunity cost of not being allowed to clear primary forests. Applying the first measure alone (the ban) would have an elevated political and social cost and would be totally unethical, because marginalized groups would pay the full price of conservation. Applying the first and second measures (ban and mitigation) would be insufficient because failure is the norm in the matter of support to agricultural intensification. Unless development practices accomplished significant progress? In order to achieve this, I proposed to open a new window at the donor level and to design “open projects” that would reflect the “piecemeal engineering” ideal of Popper’s (1966) political philosophy. I further contended that the concept of metis, as elaborated by Scott (1998) to answer to the high modernist project (Chapter 8), should be given paramount importance in these new practices. However, success could not be expected in the short term, and failure in support to agricultural intensification may still remain the norm for a long time. For this reason, the only viable solution appears to be the simultaneous implementation of the three measures (ban on clearing, intensification, compensation). The third measure is unfortunately unrealistic because the international community does not have the will to pay its cost at present. The Malagasy environment may then be doomed, unless activist voices are raised to defend its rights and the rights of its people. Until now, voices have been raised mostly to defend the rights of nature. As this imbalance may explain the failure of current policies and their drift to authoritarianism, new voices now have to rise to defend the rights of marginalized Malagasy farmers.
695
In the light of these last outcomes, it is time, now, to propose a general conclusion for this dissertation. This conclusion will start with a brief summary of the outcomes of each Part. This may create a few redundancies but is also a way to take a breath before the end of this journey inside the Malagasy environment, its people, the actors who watch them and the practices and representations that link all these together.
696
PART VI GENERAL CONCLUSION
697
1. Introduction This general conclusion will start with a brief summary of the main outcomes of Parts II through V (Section 2). This summary will not strictly follow the structure of the dissertation in order to emphasize the links between chapters. In Section 3, I will rearticulate the main empirical arguments in the light of my research questions and hypothesis. Eventually, in Section 4, I will delineate lines for future research.
2. Summary of outcomes 2.1. The conclusions from Part II The historical perspective drawn in Part II showed that for as far back as it can be remembered, there has always been a tension between poverty and deforestation in Madagascar. King Andrianipoinimerina launched the first environmental “policy” but advocated at the same time that the forests were the resources for the poor (Chapter 2). Throughout the history of Madagascar, groups of people whose land was taken or who fled from war, slavery, tax collection and forced labor moved to the agricultural frontier, where the forest offered them a refuge and resources to sustain their livelihood (Chapter 1). In spite of the issuance of an environmental charter that claimed that environmental policies had to “reconcile humans with their environment” (Gouvernement de Madagascar 1990), the tensions between poverty alleviation and deforestation remained apparent in the three phases of the national environmental action plan (Chapter 2). Market and migrations are now the new forces that push low resource farmers toward the forest frontier. But another significant change occurred recently: the international community is now committed to closing the forest frontier, due to the paramount importance given
698
to biodiversity conservation in the global environmental agenda. The Malagasy government and its partners are aware of the consequences, as revealed by the animated debate, in 2006, about compensation or mitigations of the negative impacts of protected areas on local communities (Chapter 2). In spite of this, no satisfactory solution has been applied at a significant scale, and marginalized communities continue to clear the forests or to pay the price of conservation.
2.2. The conclusions from Part III The case study drawn in Part III, showed that the economic impact of environmental policies on slash-and-burn farmer communities may be underestimated, while the economic level of these communities may be overestimated. The reason is that knowledge about this land use is mostly produced using data collected in villages located close to roads and having access to markets, whereas most slash-and-burn farmer communities live in remote areas. In these remote places, market opportunities are almost non-existent due to elevated transportation costs, unless some products with high value and limited weight can be sold. In Ambodilaingo, wood transportation plays this role but is not sufficiently profitable and secure to represent significant income. Most households live almost in autarky and are highly vulnerable, as shown by the impact of the influenza epidemic in 2002. Capacity to invest is about equal to zero and even energy, in the form of caloric intake, is insufficient to make land improvements. Due to this lack of investment capacity, combined with the closing of the forest frontier (by unfavorable ecological conditions as well as by forest control) the farmers fail to engage in intensification and the agrarian system remains in a situation of Malthusian crisis (Chapter 5). Part III also showed that solutions to escape the crisis are already present, visible in the landscape. They develop naturally as soon as pressure on land renders
699
the old systems unprofitable. Irrigated rice cultivation in bottom land, crop rotation, plowing, integration with livestock husbandry and specialization in the most profitable cash crops, such as ginger, may be the seeds of future land use. The central question, then, when dealing with slash-and-burn cultivation, is not whether to adopt a Malthusian or a Boserupian model (Chapter 5), but to understand what are the driving forces that lead to the first or the second dynamics. Population growth and agricultural intensification may have a dialogical relationship (Chapter 14). The key processes that create turbulences in this dialogy may be the decapitalization and capitalization of assets. These assets are the natural stock of fertility (the biomass that can be burnt); the fertile soils in forests (exploited by clearing the land) and in bottom land (exploited by plowing the land); water resources, which are rendered exploitable by irrigation schemes; perennial crops (home gardens), which can be extended without requiring significant maintenance efforts; and cash, which can be accumulated if significant market opportunities exist. I showed that farmers in Beforona, and even in Ambodilaingo, have strategies that relate to these assets. As closing the frontier signifies the immediate loss of access to forest biomass and fertile soils, the logical answer that rural development projects and policies should give is to favor capitalization upon alternative assets (water for irrigation, improvement of bottom land, access to market and hence to cash). Part III also showed that despite the fact that alternatives to tavy are already practiced by a significant part of the population, rural development projects do not significantly support their development. Instead, they invest research efforts in the development of systems that require still more investment. Farmers can ill afford to implement these alternatives. They try them on a small part of their land and continue to implement their own alternatives (traditional irrigated rice, ginger cultivation on plowed land, extensive home gardens) on the rest of their land. They sometimes
700
practice the “modern” alternatives a few years, but rather by curiosity or because they expect to obtain advantages from projects, such as access to jobs, social prestige or PSDR grants (Chapter 6). Ecological ginger and SRI provide the most striking examples. They are tried, but not adopted, because farmers quickly experience that they are more risky and less profitable than their own alternatives in terms of labor productivity. Part III also identified a series of myths, or received wisdoms, that can explain these awkward policies. One myth is the belief that geometry (sowing in line) and high yields are the keys to successful intensification, whereas in a context of abundant land and high vulnerability, the maximization of labor productivity and the minimization of risks are more essential. A second myth is the demonic character attributed to fires and the excessive faith in environmentally friendly techniques aimed at maximizing biomass production, such as agroforestry and agrobiological techniques (alleycropping, improved fallow, composting, mulching, cover-crops, zero tillage). A third myth is the supposed beneficial effect of markets. It is true that farmers living in remote areas could increase their income if they had access to markets. But as they are not the most competitive to seize market opportunities, increased access to markets often favors their marginalization. Part III further showed that the concentration of development activities along the road led projects to develop closer collaboration with a population of migrants and small entrepreneurs, who are easier to work with due to their higher education levels and their higher capacities to organize and adopt new techniques. Dynamic private entrepreneurs and traders can contribute to the development of their community and deserve some support. But as they are in competition with poor households for access to land and resources in general, focusing support on them may favor still more the dynamics of marginalization.
701
Eventually, Part III showed that projects in Beforona favored repressive policies against tavy. As the impact of these policies is not limited to the villages located close to the road, vulnerable farmers in remote areas, not targeted by projects, pay the high price of this policy. This contributes to keeping them in a state of poverty and explains their strong hostility toward outside interventions. In sum, Part III provided a case study that illustrates the social dynamic described in Chapter 1. Social differentiation leads to the appropriation of the most valuable resources by the most competitive farmers and to the marginalization of the weakest farmers, who continue to depend on natural resources. Development projects contribute to the empowerment of the elite while environmental policies create additional burdens on the poorest households. The consequences are not dramatic in the case of Beforona, because development projects overall are inefficient and because environmental policies (the ban on tavy) are only sporadically enforced. But the effects are sufficiently visible to say that the impact may be the reverse of what is intended. I contend, hence, that the key answer to the marginalization dynamic and the dependence on forest resources that results from it is to support investments in favor of the most vulnerable people. Remote villages have to be targeted first and supports have to be adapted to the poorest households. The experience of the NGO Zanaky ny Ala (Chapter 7) showed that this is technically feasible, that farmers are eager to invest their own efforts when supports are provided from outside and that significant results can be achieved at reasonable cost. The experience of Zanaky ny Ala, however, was too limited in duration and scope to definitively validate the activities that were implemented.
702
2.3. The conclusions from Part IV Part IV attempted to explain the genesis of the received wisdoms analyzed in Part III. It showed that the dominant paradigm around which rural development and conservation discourses and policies revolve are embedded in a Western culture inherited from the Enlightenment. A great partition occurred between romanticism and rationality, which led to the hegemony of positivist and reductionist science (Chapter 9). The consequence is that biophysical processes, which can be more easily addressed by this type of science, are paid more attention, as revealed by the research orientation of the alternatives to slash-and-burn consortium (Chapter 10) and by the hegemony of the antifire received wisdom (Chapter 12). System analysis, multidisciplinary methods and an increasing attention to social issues are recent attempts to move beyond these limitations. But these attempts failed, due to insufficient collaboration between researchers and practitioners, to difficulties in translating complex representations into simple programs of actions, to the influence of a “high modernist” project whose main objective is to render biophysical and social realities legible (Chapter 8), and to the subjective reshaping and filtering of research outcomes by political or “anti-political” (Ferguson 1994) forces. In Beforona, the BEMA and Terre-Tany projects produced relevant knowledge upon which wise development strategies could have been designed. But these strategies were never implemented (Chapter 6), despite the fact that collaboration existed between the BEMA/Terre-Tany and LDI projects. The researchers explored all potential pathways to change but development actors selected the pathways that matched better with their own objectives or dreams (to stop fires and to maximize biomass production). The episode of the GELOSE legislation (Chapter 11) further demonstrated the difficulty of translating social analysis into strategies that would genuinely reflect their insights.
703
Part IV (Chapters 8 and 9) also showed that the deconstruction of received wisdom cannot go beyond a certain limit. Development and modernization refer to existing realities and the deconstruction of these concepts should not lead to an ontological vacuum. Deconstruction is necessary, even vital, in order to understand the social, political and cultural biases that accompany the production of knowledge. But once the deconstruction is achieved, something has to remain. Dictionaries are simple tools that help to understand what is that thing that must remains, while encyclopedias reveal the socially constructed layers that are added upon the core meaning of concepts. In sum, when we do a deconstruction, we must ask: the deconstruction of what? This “what” is the thing that must be kept at the end of the process. Part IV further showed the necessity to take a slight departure from Foucault’s (1970) and Latour’s (1987) epistemologies. The real world is not the only accomplice of our knowledge, but it is an accomplice that has to be used to increase our chance of producing accurate representations. If not, we would fall into the relativist trap and deconstruction would be useless. It would only produce equal representations or ontological vacuum. Chapter 10 confirmed this. It showed how the ASB Consortium built a strong rhetoric, in the sense of Latour (1987), and how this led to overlooking the real cause of deforestation. But eventually, the real world, by unfolding its signs under the eyes of the ASB researchers, has been stronger than this rhetoric and favored a significant paradigmatic change. In spite of the limitation of these epistemological models, Part IV also confirmed the validity of the tools derived from them. The ASB rhetoric justified the use of Latour’s (1987) model, as I said above, while the case of the GELOSE validated Foucault’s conception of knowledge as an incarnation of power (Chapter 11). Rational and transparent communication was invoked by the designers of the GELOSE as a way to favor consensus among stakeholders. But in the end,
704
communication served to channel knowledge, and hence power, in order to impose decisions (to stop forest clearing and to harvest resources in a sustainable way) which were already taken by the international community and which are not negotiable. Eventually, Part IV showed that new discourses emerged which challenged the hegemony of mainstream environmentalism (Chapter 12). Kull (2004) put into question the antifire discourse and showed that beyond environmental degradation, humans create new landscapes with new cultural contents and economic values. He further showed that antifire policies created resistances, and that wildfires, which are more destructive than pasture fires or agricultural fires, are the main weapons of these resistances. Kull’s narratives, however, also had its own limits, which was confirmed in Part IV (Chapter 13).
2.4. The conclusions from Part V In Part V, I proposed three syntheses of my work. The first synthesis (Chapter 13) was aimed at comparing my outcomes with other work conducted on similar subjects. I established a parallel with a book published twenty years ago: Land Degradation and Society, by Blaikie and Brookfield (1987). The parallel was clear with regard to the general statement (the failure of conservation
policies),
the
methodological
choices
(comprehensiveness,
demystification of quantitative approaches), the main narratives (received wisdoms, marginalization, control, resource capture) and the recommendations (importance of investments, need to be open-minded, non-rejection of coercion). Science and technology studies and political ecology were two disciplines visible in the background of this book, as they are in my dissertation. A larger review of the political ecology literature confirmed that my work has strong affinities with this field. But Chapter 13, by analyzing recent political ecology work conducted in Madagascar, also
705
showed that this discipline may now be creating its own domains of exclusion, leading to representations that are as socially constructed as the representations it challenges. The reason may be the influence of relativist epistemologies derived from an a la lettre interpretation of Foucault and Latour. Some schools of political ecology, influenced by post-modernism and social constructionism, are contending that nature is socially constructed, or that there is a “second nature.” This leads to ignore the ontological distinction of the natural and cultural dimensions of our world and results in a conflation of the concepts of nature and environment. The consequence is that political ecologists often have the tendency to consider anthropogenic and nonanthropogenic landscapes as equivalent (Chapter 13), which they could be in terms of practical value, but which they cannot be ontologically, as they are designed with different aims and following different rules. Hence the parts of the world which exist independently of human intentions and choices may be overlooked by political ecologists. The second synthesis, in Chapter 14, was aimed at projecting my outcomes into a new epistemological framework that would help to escape both the positivist trap (which favors the hegemony of received wisdoms) and the relativist trap (which leads to create supposedly equivalent representations). Chapter 14 may be indeed the core chapter of this dissertation, the point of crystallization of my outcomes for the emergence of a new interpretative scheme. It started by an analysis of the answer given by Latour (1993) to escape the relativist trap. Latour proposed to displace the deliberation about nature into a larger collective whose agency would be the effect of a mediation between humans and non-humans. Agency, in this model, would involve non-humans (through the mediation), which would guarantee that controversies would be closed with some form of objectivity. This would be a departure from social constructionism and would allow an escape from the relativist trap. I criticized this
706
model because it still recognizes the possibility of closing controversies, which can be regarded as a new form of positivism, and because it tends to conflate human and nonhuman agency, which forces the acceptance of a significant metaphysical revolution. As I rejected this revolution, I contended, in the end, that this model is not different from social constructionism: only humans can mediate between the human and the non-human world, by producing representations. The consequence is that relational agency may indeed just be human agency. In Chapter 14 I also explored another way to escape from the positivist and relativist traps. This way was proposed by Popper (1959, 402 1963). For this author, knowledge advances by a series of conjectures and refutations, which contrasts with the positivist perspective that puts verification by induction at the core of its method. Verification serves to corroborate theories, i.e., to render them more certain, but certainty is not the only criterion for saying that one theory is better than another. For Popper, knowledge also grows by the creation of theories with more content, which does not necessarily imply that their certainty increases. To the contrary, there is a tradeoff between increasing content and increasing certainty because theories with more content are also less probable. In order to address this tradeoff, Popper proposed the concept of verisimilitude (Chapter 14) to measure the growth of knowledge in consideration of these two criteria (content and certainty). A theory with higher verisimilitude can be less certain than another one with lower verisimilitude but can represent some progress, because it can reflect reality better.403 Another essential principle in Popper’s epistemology, which illustrates the departure from both positivist and relativist epistemologies, is the acceptance of the “Cartesian anxiety” (Bernstein 1983). In other words, a truth exists, because the reality is unique, but this truth cannot be known. Accepting uncertainty is indeed the 402 403
First published in 1934. It can even be questioned whether realities with low content exist.
707
condition for the advancement of knowledge, because accepting certainty is equal to closing a system of knowledge to future advancement. Both positivist and relativist epistemologies refused to accept uncertainty. They envisioned that controversies could be closed and that nature would be the cause or the consequence of this closing. Popper’s model provided an explanation for the production and strengthening of received wisdoms. Scientists favor certainty over content because they do not like the Cartesian anxiety, and because a theory with high certainty and low content has more chance of being published by a journal than a theory with low certainty and high content. On the other hand, decision makers need theories with high content because they are dealing with the real world, not with isolated parts of it. Theories with low content, produced by scientists, are thus transformed into theories with high content by decision makers that need to justify their policies. This is particularly true in places like Madagascar because science is utilized to justify political choices made by foreign actors who are not supposed to meddle in political affairs. In this process, the theories with high content are not controlled by scientists (this dissertation was aimed at remedying this). They are hence socially constructed to an extent that allows little room for observing the signs unfolded by the real world. But they are given an illusion of certainty by their assimilation to theories with more certainty and less content. Scientists are sometimes even accomplices in this process. This subjective drift from theories with low content to theories with high content may be the key mechanism that leads to received wisdoms and ignorance. 404 Popper’s model also provided an explanation for the revisionist tendency of certain schools of the social sciences (Chapter 13). These schools attempted to answer to the hegemony of theories with low content by addressing larger and more complex
404
From a Popperian perspective, I defined ignorance as certainty about a theory that is indeed false. Ignorance, like truth, exists but cannot be known. Ignorance can however be avoided, by refusing certainty.
708
objects embedded in the social. They created theories with higher content and lower probability to be true, contributing to a significant expansion of knowledge. But once these theories were created, these schools often reversed the drift mentioned above. They derived new theories with low content from their theories with high content, challenging biophysical theories that indeed had high probability to be true and were sustained by an important heritage of empirical facts. This is how revisionism appeared. These schools, however, had a sense that the facts were against them, and for this reason, their “revision” remained ambiguous, mostly motivated by the political necessity to break the myths created by the drift that operated the other way. But Popper may have failed to see that it is not only the epistemology that has to be changed. As shown by Morin and Feyerabend, we may also have to change the scientific method in a more radical way, and even work against this method if we are to address the complexity of our world. In theories with high content, the whole world may need to be present, with its esthetic and spiritual dimensions in addition to its materiality. We may need to be artists and philosophers, not only scientists, if we are to elaborate such theories and design better our future. My third synthesis, in Chapter 15, is a translation of the outcomes of this dissertation into a program of action aimed at stopping deforestation without requiring the marginalized social groups to pay the price for it. This program was based on nine rules or principles: 1. There is no “natural” harmony between humans and nature. 2. There has to be a ban on forest clearing (but not on resource extraction), and this ban has to be enforced if we are to conserve biodiversity. 3. This ban, although necessary, is not legitimate. 4. Being illegitimate, the ban must be compensated.
709
5. Support to agricultural intensification has to shift from technical (training) to economic (credit or subsidies). 6. If we brought some water to Malagasy farmers, it would decrease their dependence on forest biomass. 7. Land with potential for improvement must be entitled, with a priority to marginal areas. 8. Energy (caloric intake) must be subsidized to accelerate land improvement. 9. Direct compensation in the form of subsidies must be preferred to mitigation by support to agricultural intensification. These rules, despite the fact that they are very precise, must not be seen as fixed and immutable laws. They reflect tendencies that were revealed by careful observation of realities. But they are operating in complex social and biophysical systems that provide them with a context. This context is highly variable and can be highly influential on the final outcomes, contradicting these rules in some cases. For this reason, projects must be designed with flexibility and must be ready to accept outcomes they did not predict. The concept of Metis (Scott 1998, see Chapter 8) helps to achieve this. If metis knowledge is to be favored, the rules I mentioned can serve to fix general objectives but must not lead to designing the details nor to defining the way to achieve these objectives. In practice, this means: •
To put a ban on forest clearing but to let local communities organize control themselves, if they are in charge of this control.
•
To provide credit to farmers but to let them design the activities they will conduct with this credit.
•
To provide subsidies to farmers, but to let them use this subsidy in the way they like.
710
•
To build a dam that has the potential to irrigate bottom land but to let communities find their own way to improve the land and to organize for water management.
•
To title land in the name of a lineage elder when land is not clearly individually appropriated and to let customary rights manage land distribution at the individual level.
•
To establish “food for work” contracts for land improvement but to let farmers chose the type of land improvement they will do.
The nine rules and the measures derived from them can be translated into a policy that should be tested locally, before being scaled up. This precaution is necessary to avoid the construction of new fashions or dogma based on these nine rules, and to debunk the mistakes of the representations I proposed. These tests, if successful, would provide more evidence of the correspondence of these nine rules with the facts, increasing their level of certainty. In case of failure, some facts would be refuted, allowing construction of a new theory with more true content and less false content, i.e., a theory with higher verisimilitude. To make this approach work, science and decision makers would be forced to collaborate, each providing feedback to the work of the other. The acceptance of uncertainty, on both sides, is the condition for this collaboration. In practice, this approach would mean that decision makers, supported by donors, would open a second window aimed at funding “open projects” for testing new approaches. This second window would be freed from received wisdoms and from the many buzzwords that asphyxiate development practices. It would be open to radically new practices, but the projects implemented within it would be carefully monitored. The outcomes, in case of success, would be integrated into the first window, leading to the progressive reform of practices. Scientists would then be
711
“virtuoso social actors” (Flyvbjerg 2001, 2) doing phronesis (Part I) and society could engage in doing what Popper (1966) 405 called piecemeal social engineering.
3. Conclusion In this last section, I will come back to my general introduction (Part I) in order to assess to what extent this dissertation answered my research questions and verified my hypothesis. I will reorganize my empirical outcomes according to the research questions and hypothesis, at the risk of creating redundancies with the previous section. My research questions were: •
Why do Malagasy slash-and-burn farmers still clear the Malagasy eastern rain forest?
•
Why do the current policies and programs dealing with this issue fail?
•
What needs to be done to stop forest clearing and to decrease the poverty of slash-and-burn farmers?
The hypothesis I adopted for answering these questions is that there is a mismatch between the reality of slash-and-burn farmers and the representation of this reality by the actors who act upon it. This dissertation provided a large amount of evidence verifying this hypothesis in the case of Madagascar. For example, •
According to the dominant representations, there is a vicious circle of land degradation and poverty which can only be stopped by technological innovation, appropriate management and population control. Resources are granted once and for all and must not be depleted if sustainability is to be achieved. In the real world, farmers
405
First published in 1949
712
exploit natural resources and capitalize assets to sustain their livelihood. Resources can be forest biomass, fertile soils, water or deep soils on bottom land. Capitalization can concern land improvement by irrigation or terracing, livestock or plantation of perennials. When one resource is depleted, farmers shift to another one and when one asset is lost, they attempt to capitalize upon another one. This is the strategy they use to achieve the sustainability of their livelihood. •
According to the dominant representations, the market is a panacea and its development is expected to increase farmers’ income, allowing investments for the development of sustainable land uses. In the real world, market opportunities are limited and are captured by the most competitive groups, who already have the higher investment capacity. Markets hence contribute to the marginalization of the weakest groups and their move toward the forest frontier.
•
According to the dominant representations, there is no alternative to tavy available yet. Farmers clear forest because they are backward and do not know appropriate techniques. They have to receive environmental education and be trained about environmentally friendly techniques. In the real world, alternatives to tavy exist already. Farmers have the required knowledge to develop more intensive land use but fail to do so because they lack capital and energy, and because they are pushed toward marginal lands by more competitive groups.
•
According to the dominant representations, rural development projects achieve results. They increase farmers’ assets and yields and target the poor first. In the real world, they flatter a clientele which flatters them in return. They work with a rural elite which channels to itself the main
713
benefits of the aid. They create technical dreams that depoliticize development issues and their real impact is never measured. •
According to the dominant representations, the repression of tavy and fires is paramount. Fire controls have to be enforced in order to allow projects to have impacts and in order to create incentives to land use change. In the real world, tavy repression is harmful. It creates anger, politicizes fires and leads to overlooking the real causes of policy and project failures.
•
According to the dominant representations, farmers are seen either as a threat to ecosystems or as living in harmony with them. They have to be controlled and guided to sustainable land uses, or they have to be given rights to manage resources themselves. In the real world, farmers regulate the extraction of their resources, but the main function of this regulation is to provide land to future generations, through conversion of forest to agricultural land. Clearing stops only when farmers encounter marginal land whose marginal production is not profitable in comparison with other options. When farmers are marginalized themselves, alternatives are nonexistent or unsuitable and even the most marginal lands get cleared.
This last point leads us to what may be the main empirical conclusion of this dissertation. The Malagasy primary forests may be the last locations where the marginal people and the marginal natural resources – those having no value on the market – encounter each other, and where the first taps into the second to survive. In this tradeoff, the international community has chosen to protect the natural resources first, to the detriment of the people. It is not a conscious choice. It is the choice of ignorance. It is the consequence of the mismatch between representations and realities.
714
The consequence will be that both the people and the resources will be lost. Three forms of power determine the fate of Malagasy forests: (1) coercion, which is used by the state for ruling; (2) knowledge, which is used by the international community to lobby states and other national or regional organizations (through working groups and workshops) and people (through participatory methods); and (3) actions on the ground, for example by igniting fires, which are used by local people to shape their environment and resist to the other forms of power. 406 Action will be the most efficient form of power in the end. Farmers will decide the fate of the forest by clearing and burning it or not. Unless the other forms of power engage in confrontation? As this confrontation would be based on a system of knowledge which contains much ignorance, this strategy would just lead to new forms of totalitarianism. In order to escape from this trap, I proposed to internalize marginal resources and marginal people together in the world economic system. Technically, the solution seems to be simple. Marginalization results from the appropriation of the most valuable resources, such as the forest biomass and the fertile bottom land, by the most powerful social groups to the detriment of the weakest ones. This trend just needs to be reversed, or at least stopped. We saw that three ways are possible for this: (1) securing resource ownership, with a priority given to farmers living in marginal areas; (2) injecting capital in the form of investments, in order to favor intensification and reduce dependence on unsustainable resource exploitation; and (3) providing subsidies that will compensate for the restrictions on access to resources, in case the other strategies would not suffice. The first strategy justified the creation of the GELOSE legislation, but failed (Chapter 10) because actors in charge of implementing it believed that negotiation alone could lead to win-win solutions. The second strategy failed as well because it proposed ill designed alternatives, and because it did not 406
These forms of resistance are described in detail by Scott (1976; 1985). I have not read these two books yet but I can guess they provide relevant insights to understand this form of power.
715
properly balance conservation and development objectives. The third strategy (subsidies) may then be the main source of hope that remains. The aid system, indeed, claims to be already engaged in this new strategy. After the failure of ICDPs during EP1 and the failure of the GELOSE at present, Payment for Environmental Service (PES) emerges as a new promising strategy. It may quickly become a new fashionable concept while the Clean Development Mechanism of the Kyoto protocol is expected to provide an abundance of resources to finance these payments. But Chapter 6 showed that projects emerging from this new approach (the VMCRCC project) can focus on the same ill-designed development alternatives and fail to avoid resource capture by elites. Behind the apparent simplicity of solutions, it may be extremely difficult to escape from the trap of received wisdoms, shortsighted politics and unequal power relations. Knowledge can progress quickly in an apolitical context but can immediately lose its main content when introduced in the political arena. The theoretical background that sustained the ICDP approach may have been wise, but the ICDP ideal was violated by the prioritization of conservation objectives over development. In the same manner, the designers of the GELOSE approach anticipated most problems to come, but in the end, the GELOSE transferred rights to associations, not to communities, which changed everything. The subsidy approach could become the victim of similar political distortion and ill designed implementation. New bureaucratic systems could be built to manage the distribution of subsidies. They would define complex criteria and modalities of distribution. Those who can open bank accounts, organize and negotiate with NGOs and government staff would capture these subsidies, pushing illiterate farmers out of this new opportunity. Moreover, we saw that subsidies would cost more than existing approaches, and we can expect a strong reluctance of the international community to pay for them unless a clear additional benefit can be measured, which may explain
716
why the term PES becomes so fashionable while the concept of compensation is feared (Chapter 2). In other words, researchers can contribute to improving knowledge but the implementation of their outcomes follows rules that are out of their control. Would this dissertation then be useless? The answer will be given by decision makers so I will end this conclusion by a few recommendations in their direction. First, science and politics must not be separated but must remain clearly distinct. Their functions must not be confused. The aim of scientists is to “know” while the aim of politicians is to “decide”. It may be that the main reason for our misrepresentations is not incapacity to produce knowledge but incapacity to make a distinction between our will to know and our will to decide. Knowledge is a form of power so as soon as we have it, we want to utilize it to take decisions. In order to remedy this, science and politics must be distinguished but must also exert a mutual feed-back to each other. When decision makers go against empirical facts, scientists should be there to correct their trajectories. When scientists fail to address complexity and prefer to recreate oversimplified virtual worlds, and when the real world itself is reshaped in the image of these models, democracy should exert a control on science. The real social world could hence jump into the face of scientists and bring them back to the realities they overlooked (Chapter 10). Second, we arrive at a time where politics can no longer be for the defense of particular interests. I am not raising here a question of legitimacy. Everybody has the right to defend their interests and one cannot expect things to be otherwise. But the world has now become global and there may be only one way to defend the interests of the particular: to defend the interests of the whole. Stated more simply, if poor farmers continue to be required to pay for conservation, they will continue to burn the forest.
717
Third, people and biodiversity both matter. In this dissertation, I contended that people matter even more than biodiversity, which is a personal choice determined by my personal ethic. Some could argue that biodiversity matters more because it may be destroyed forever, whereas human rights can be violated only locally and temporally. Another argument is that human societies would pay a higher price from biodiversity loss than from local human rights violations. I disagree with these arguments because I believe that human rights should not be computed, but also because the fundamental character of human rights is the cause of failures of approaches that would violate them. People can resist environmental policies and biodiversity will become their enemy if they can feel that more importance is given to it than to them. Fourth, biodiversity matters more than what can be guessed from the still limited commitment of the international community to conserve it. The fact that biodiversity is often put above people does not mean that it is paid significant attention by the nations that fund conservation programs. We saw in Chapter 15 that biodiversity, indeed, may matter less than guaranteeing cheap food or export opportunities to agricultural producers, or than improving transportation systems in large cities. If a solution was found to conserve efficiently the world’s biodiversity, the international community would have to pay significantly more than what it has paid until now to achieve this. At present, we have the illusion that conserving biodiversity is cheap but this is because marginal farmers worldwide pay a large part of conservation costs. Fifth, and as a consequence of the third point, it may be that science alone will not be sufficient to orient policies in the most appropriate way. Marginalized communities are expected to pay part of the cost of conservation because their voices are absent from workshops and working groups. In order to remedy this, communities have to be given a voice in formal institutions, which can be done by empowering
718
grassroots movements. The GELOSE was intended to do this but failed because it channeled power in the opposite direction: from the centers of decision making to local communities, using communication devices (Chapter 11). Subsidies aimed at compensating for conservation costs could be another form of empowerment. We saw that the ban on forest clearing is not negotiable at present. But if this ban was compensated, negotiation would be possible with respect to the amount of this compensation. Communities would then be empowered because there would be a bargain in which they could engage. The rejection of such compensations by organizations engaged in conservation (Chapter 2) may indeed reveal a fear of this empowerment. Sixth, the faith in science, and especially in biophysical sciences, should be tempered. Science itself must remain paramount and my dissertation is intended to be a strong advocacy for science. But an excessive faith in biophysical sciences serves as an excuse for not addressing deep social problems located in the global governance of resources and markets. Science can provide us with illusions that progress can be made while maintaining a political status-quo. Agroforestry played this role, as we saw in Chapters 6 and 10, and ecoagriculture and agrobiology (McNeely and Scherr 2003; Altieri 2002) may take its succession in the future. As I mentioned in Chapter 6 when referring to agroforestry alternatives, I am not criticizing these approaches per se. But problems may arise from the excessive faith given to them (faith in technology in the case of ecoagriculture , faith in harmony in the case of agrobiology), which can lead to overlooking whether or not alternatives systems are profitable or suitable in the economic context of poor households, and to maintenance of a blind eye to macroeconomic and political economy issues. Seventh, and this brings us back to the main epistemological outcomes of this dissertation, society and the knowledge it produces must remain open. We must accept
719
living with uncertainty and resist the Cartesian anxiety. Only doing this can we listen to what our world says to us, pragmatically improve knowledge and take a chance to reduce mismatches between representations and realities. Only doing this can we reconcile the heritage of positivist science with its post-modern deconstruction and escape from the traps created by hegemonic doctrines. Decision makers, as well as scientists, face this challenge and this is why they have to work together while not confusing their tasks. By doing so, we may have a chance to understand the world we live in, to be respectful of both the people and the environment and to find a way to reconcile our nature and our culture.
4. Future research This dissertation provided a holistic interpretation of the issues of deforestation and slash-and-burn cultivation in Madagascar. It attempted to integrate into a single vision the perspectives provided by different disciplines. What is to be done next? To answer this question, I need to come back to the epistemological model of Popper (Chapter 14). My research was aimed at elaborating a theory with very elevated content, i.e., a theory that would include a large set of statements. In consequence, and given the limited time available to conduct my work, I had to accept that each statement regarded separately would have a relatively low level of certainty. This choice was motivated by the fact that until now, most scientists preferred to proceed the other way: by looking at one aspect of the reality and by attempting to maximize the certainty of their representation with regard to this aspect. These research works were relevant and I used them to increase the level of certainty of certain of my outcomes. But in the end, due to my emphasis on comprehensiveness, I believe that my dissertation is a more relevant explanation of the causes of
720
deforestation and slash-and-burn cultivation in Madagascar. In sum, I claim to have elaborated a theory with higher verisimilitude despite the fact that most of the statements I have made are based on relatively weak empirical evidence. Future research, logically, should then be dedicated to improving the certainty of these statements, or to refuting them. This, however, must not be done to the detriment of the overall verisimilitude, meaning that scientists who will work on certain aspects of the representation I elaborated should not isolate them from the big picture. They will have to improve the whole theory in the end. Otherwise, they may again provide only illusions of progress. Concerning the way to increase the certainty of specific statements, I have already argued (Chapter 15 and previous section) that the experimental method should be given paramount importance and that experiments have to be conducted in the real world, by being linked to the implementation of policies. The open project was designed to achieve this. The simplest experiment that comes to my mind, following the outcomes of my dissertation, is to implement conservation subsidies 407 on a pilot site and to measure the impact at the level of agricultural intensification, dependence on natural resources, natural resource management, biodiversity conservation, poverty and well being, at a spatial and temporal scale that would go beyond the site of this implementation. Metis and adaptive management are the key concepts that the designers of this experiment will have to keep in mind. But research detached from politics and focusing on certain aspects of the reality must also be conducted. I believe that comprehensiveness is necessary and that more synergies should be established between science and politics, but this approach should not become a new fashion and should not be exclusive.
407
I prefer the term subsidy to that of payment for environmental services because I justify the payment by the existence of a tradeoff between social and environmental externalities, rather than by the providing of a particular service.
721
For instance, I am interested in the quantitative methods used by Levitt and Dubner (2006), which proved to be efficient tools to debunk received wisdoms. Levitt and Dubner (2006), However, mostly addressed microeconomic problems and it remains to determine whether their methods can be used to solve macroeconomic issues. If they could not, qualitative approaches addressing the global and quantitative methods measuring the local may need to complement each other. The quantitative surveys I conducted in Beforona 408 were, indeed, intended to complement and test my qualitative outcomes. Quantitative economic studies have also been conducted in Madagascar, by IFPRI, 409 by the ILO 410 program 411 and by the SAGA 412 project, 413 with funding from USAID. They did not focus specifically on slash-and-burn agriculture and deforestation but provided precise measurements of certain aspects of the reality I investigated, such as the relationships between market expansion, poverty and environmental degradation (Barrett 1999; Lapenu et al. 1998; Zeller et al. 2000; Paternostro et al. 2001; Dorosh et al. 2003); 414 the relationships between agricultural intensification and a series of variables such as risk, prices and labor constraint (Barrett 1999; Barrett et al. 2003; Minten et al. 2006; Minten and Barrett 2006); the effect of remoteness on agricultural competitiveness and development (Moser et al. 2006).
408
Due to a lack of time, I did not use the results of this survey in this dissertation (Chapter 4), except to estimate rice production per household (Chapter 5). I intend to use the results of these enquiries and to complement them with other surveys in the future. 409 “Improved economic analysis for decision making in Madagascar,” funded by USAID and implemented by Cornell University. 410 International Food Policy Research Institute, involved in a research program in partnership with Cornell University and FOFIFA. 411 For an overview of their work, see FOFIFA-Programme ILO (2003). 412 Strategies and Analysis for Growth and Access, “A project of Cornell and Clark Atlanta Universities for research and technical assistance in Africa, funded with a cooperative agreement with USAID” (http://www.saga.cornell.edu/saga/annreport/mad1003.html) 413 Some of the work of these programs was mentioned in Chapter 1, section 4.1. 414 I introduced some of these works in Chapter 1.
722
These studies provide relevant results but could also be biased by the focus on certain variables and certain sets of data. For example, Minten and Barrett (2006) show that communes with higher rate of adoption of new agricultural technologies have more healthy economies. Hence, these authors recommend supporting agriculture intensification as a key strategy for alleviating poverty and food insecurity. But this correlation could be exaggerated by the fact that farmers unable to intensify (the losers in the competition) may move to other communes (to the frontier) or to remote villages less represented in the sample (their paper raises this sampling issue). It could also be explained by the fact that a more healthy economy favors adoption (the causal relationship could then be the other way around). Eventually, the authors focus on certain indicators, such as rice yield as a proxy of agriculture performance, and the use of “inorganic fertilizers, off-season crops, early rice seedling transplanting, improved rice seeds, SRI, and agricultural equipment (plows and harrows)” (Minten and Barrett 2006, 23) as indicators of the adoption of improved agricultural technologies. We saw in the case of Beforona (Chapter 3, 5 and 6) that intensification can use very different pathways. In sum, the conclusion of these authors (the rejection of the Boserupian hypothesis) could be explained by the method, as well as by the reality. In order to achieve future progress, it is hence necessary to articulate the outcomes of this type of research with more comprehensive analyses. This could be done by implementing the methodologies of these authors in the sites I investigated, which could decrease the uncertainty of some of my key statements, or by conducting comprehensive qualitative studies in the sites they investigated, which could help to identify the most relevant variables to measure and to limit the risks of spurious correlations. Future research could also be aimed at pushing comprehensiveness still farther. I already noticed, for example, that favoring migrations to the grasslands of the Middle
723
West of Madagascar can be a way to reduce the pressure on primary forests, but that these grasslands are inhabited by pastoral communities already marginalized by immigrant groups. The links between deforestation and migration on this “second frontier” (Chapter 1) hence need to be better addressed. I also argued (Chapter 15) that deforestation is not the only threat to Malagasy biodiversity. In the long term, global warming may have a very significant impact as well, directly, by putting stress on ecosystems, and indirectly, by increasing climatic risks and perturbing agricultural production. Vulnerable communities living on the forest frontier may be the first victims of regional climatic change and of the increasing risks, which will increase their dependence on the unsustainable use of natural resources. Future theories with higher verisimilitude and dedicated to the “phronesis” ideal (Part I) will have to include this aspect of the problem. Eventually, I must ask what might be the place of art and philosophy in future research, or, more practically, what might be the place of ethics. I already argued that the issues addressed in this dissertation cannot be isolated from ethical considerations. Ethics appears when we ask the question of whether the dimension of the world that exists independently of our representation, which I called nature (Chapter 14), is worth conserving or not (Chapter 15). Ethical issues are raised also when we face the tradeoff of biodiversity conservation and local livelihoods. In the present political economy, this tradeoff tends to be the rule, more than the exception. Ethic is hence inescapable, for moral reasons but also for practical ones, which led me to argue in favor of subsidies for conservation (Chapter 15). As I argued in several chapters, if nature became the enemy of people, it would have more reasons to be destroyed. The environmental stalemate in Madagascar argues in favor of this interpretation (Bertrand and Sourdat 1998; Kull 2004). Fire is the simplest, the most universal and one of the most powerful tools. It would be unrealistic to believe that it could be controlled by
724
repression alone. I hence encourage future researchers working on environmental degradation, including those focusing on biophysical problems, to state clearly which ethic they adopt and to envision a way to test if their ethical choice helps to achieve results for the problems they intend to solve. I finish here my travels into realities and discourses of environmental degradation and conservation in Madagascar. I hope that other scholars will continue this trip, taking the holistic representation I created and submitting it to tests. I hope also that decision makers and all those having the power to shape realities will use these outcomes. It may be that making the right decision now is not so much a matter of knowledge. Relevant knowledge about the human impact on the environment and its causes has existed for a long time. Chapter 13 showed that most outcomes from my dissertation were expressed twenty years ago by Blaikie and Brookfield (1987). Our global society has huge financial resources that could be invested to stop the movement that leads to the exclusion of marginal natural systems and marginal people. To date, we have not made such investments, as shown by the comparison with the amount of money spent on agricultural subsidies (Chapter 15). It would be interesting to calculate the exact “benefit” (in the short term) that an average world citizen receives from deforestation and social marginalization, two externalities that Polanyi described in 1944 using different words. If this calculation was done, it may confirm that what is required from better-off world citizens, indeed, is not to pay for biodiversity conservation, but to stop receiving the short term benefits of environmental destruction while marginalized farmers worldwide are asked to pay the main costs of conservation. I encourage future research to engage in this endeavor, in the continuation of my work.
725
APPENDIX
EP3’s logical framework. Source: World Bank (2004a, 57)
“EP3 Purpose: Natural resources are conserved and wisely utilized in support of sustainable economic development and a better quality of life. Strategic objective 1.Sustainable management systems of renewable natural resources and of biodiversity conservation are adopted and owned by the populations in program interventions. Specific objective 1.1: Sustainable development programs are implemented. Output 1.1.1: Community development plans (PCDs) and inter-communal schemes incorporate an environmental dimension of development. Activities: a – Support communes in preparing « green » PCD and inter-communal development schemes b – Promote inter community and regional exchanges, as well as with other development programs Output 1.1.2: Sustainable development alternatives are implemented to reduce environmental pressures. Activities: a – Improve management of previously deforested areas b – Support management transfer of livestock grazing areas c – Promote conservation and sustainable use of water and soils d – Implement activities that provide alternatives to pressures in protected areas peripheral zones Output 1.1.3: Value of biodiversity sector production chains is sustainably enhanced. Activities: a – Implement new management and revenue-sharing schemes b – Reorganize management of traded and tradable species Output 1.1.4: The use of equipment and fuels that reduce pressure on forest resources is scaled-up. Activities: a – Identify available and locally appropriate renewable energy resources (for heating) ; b – Promote substitution of biomass fuels c – Promote alternative energy uses Output 1.1.5: Urban environment management is improved.
726
Activities: a – Support development and implementation of municipal environmental programs b – Promote prevention and reduction of pollution in urban areas Specific objective 1.2: Forest ecosystems and water resources are sustainably managed: Output 1.2.1: Forest resources are managed rationally Activities: a – Refine forest zoning b - Intensify forest management transfers c - Rationalize forest exploitation d - Promote and apply standards/norms concerning forests products (ecocertification, …) e - Implement complementary economic and regulatory tools f –Develop effective and sustainable forest management systems Output 1.2.2: Forest cover and carbon storage capacity is preserved Activities: a – Continue creating Land Reserves for Reforestation (RFR) b – Promote reforestation c - Manage carbon sequestration pilot sites Output 1.2.3: Wood fuel management is improved and contributes to communal development Activities: a – Support improved techniques of carbonization b – Promote energy uses that consume less wood fuel Output 1.2.4: Occurrence of natural forests fires is diminished Activities: a – Raise public awareness and motivate population to fight against fires b – Better enforce laws against bush fires c – Establish a fire satellite monitoring mechanism Output 1.2.5: Humid areas and water reserves are managed in a sustainable manner Activities: a - Promote sustainable management and conservation of fresh water lakes b – Intensify protection of water basins Specific Objective: 1.3: Protected areas are effectively managed and generate economic benefits
727
Output 1.3.1: Representation of ecosystems is PA system is improved Activities: a – Reclassify some protected areas b – Create new land protected areas and conservation sites c – Develop marine park system d – Re-demarcate some protected areas Output 1.3.2: Biodiversity conservation and proper management of ecological process in PAs is ensured Activities: a – Ensure ecological monitoring of habitat, species, pressures as well as conservation measures b - Ensure surveillance and control c - Establish conservation infrastructures and effectuate zoning d - Develop biodiversity research e- Support the CROs f - Establish voluntary PAs Output 1.3.3: Eco-tourism in protected areas continues to grow and generate revenues Activities: a - Improve and expand visitor infrastructure in PAs b – Put in place / improve ecotourism services c – Promote protected areas d - Obtain PA visitor feedback Specific objective 1.4: Marine and coastal ecosystems are sustainably managed: Output 1.4.1: Management of coastal area and marine resources is subject to sustainable development planning framework Activities: a – Develop and implement inter community development schemes b – Build capacity of actors in terms of GIZC Output 1.4.2: Value and equitable and sustainable management of coastal and sea resources is enhanced Activities: a – Intensify the transfer of marine and coastal zone resources to communes b – Promote certification of marine and coastal zone resources products Output 1.4.3: Marine and coastal zone ecosystem biodiversity and functions are maintained Activities: a – Promote conservation sites outside protected areas to enable species stock renewing b – Promote protection of marine and coastal endangered species
728
c - Promote marine ecotourism Output 1.4.4: Prevention and reduction of coastal and sea pollution and degradation are initiated Activities: a – Develop and support implementation of inter community pollution prevention and reduction plans b – Put in place inter community pollution and degradation observatories c – Reinforce erosion prevention and reduction upstream sea and coastal zones subject of special management Strategic objective 2.Sustainability of natural and environmental resources management at national level is ensured. Specific objective 2.1: Positive change in behavior vis-à-vis the environment is observed: Output 2.1.1: Environmental information and decision-making support systems support design and implementation of sustainable development activities Activities: a – Manage environmental management dash-board at national and decentralized level and facilitate exchange of information b – Ensure the monitoring of terrestrial and marine ecosystems and management of Malagasy biodiversity data c – Develop environment economic accounting systems Output 2.1.2: National capacity for effective environmental management is strengthened Activities: a – Improve knowledge about the environment b – Support environmental communication c – Support environmental education and training d - Promote attitudes in favor of conservation of protected areas and conservation sites Specific objective 2.2: Basis for the sustainable financing of the environment is established Output 2.2.1: Specific tools for financial sustainability are developed Activities: a – Put in place a « trust fund » for protected areas
729
b – Develop other financing instruments and optimize interface with other sectors Output 2.2.2: Financial management and control is set on a more rational and transparent footing Activities: a - Improve existing financial management systems b - Develop management and administrative capacities c - Strengthen mechanisms for institutional coordination d – Optimize cost structure of implementing agencies Output 2.2.3: Local financing systems for the environment are put in place Activities: a – Develop local taxation mechanisms b – Support establishment of local sustainable investment funds Specific objective 2.3: Environmental governance systems and policies are strengthened: Output 2.3.1: Environmental dimension is incorporated in Madagascar's development policies Activities: a – Develop environmental management tools and ensure consistency of legal texts and procedures b – Ensure compatibility of investments with the environment c – Ensure environmental monitoring, control and complaint management mechanism d- Incorporate environmental dimension in land use planning policies e – Incorporate environmental dimension into energy and water management policies Output 2.3.2: Environmental institutional framework is better coordinated Activities: a – Reinforce mandate of institutions (CIME, CNE) b – Develop partnership with environmental structures at sector and communal level Output 2.3.3: Capacity of environmental administration is strengthened Activities: a – Improve institutional capacity of MEnvEF b – Ensure implementation of international conventions ratified by Madagascar c – Ensure interface with Ministry of Plan (PIP) and Finance (RPI, taxes,..) Output 2.3.4: Improve forest governance Activities: a – Reinforce forest administrative department b – Put in place an autonomous structure to ensure forest operations c – Build stakeholder's capacity in forest management
730
d – Extend control and information mechanisms initiated by OSF e – Support forest control”
731
WORKS CITED Agarwal, D. K., Dewar, R. E., Gelfand, A. E., Mickelson, J. G., and Silander Jr., J. A. (2005). Tropical deforestation in Madagascar: analysis using hierarchical, spatially explicit, Bayesian regression models. Ecological Modelling, 185, 105–131 Agrawal, A. (1995). Dismantling the divide between indigenous and scientific knowledge. Development and Change, 26(3), 413-439. Alegre, J. C., and Cassel, D. K. (1996). Dynamics of soil physical properties under alternative systems to slash-and-burn. Agriculture Ecosystems and Environment, 58(1), 39-48. Allain, G. (1997). Les voies d'amélioration du tavy: conditions et moyens d'une gestion viable des agrosystèmes dominés par l'agriculture sur brûlis dans la région d'Andapa (nord-est de Madagascar). Institut National Agronomique Paris-Grignon, Paris. Althabe, G. (1969). Oppression et libération dans l'imaginaire, les communautés villageoises de la cote orientale de Madagascar. Paris: F. Maspero. Altieri, M.A. (2002). Agroecology: the science of natural resource management for poor farmers in marginal environments. Agriculture, Ecosystems and Environment, 93, 1–24 Amelung, T., and Diehl, M. (1992). Deforestation of tropical rain forests: economic causes and impact on development. Kieler Studien, Int. fur Welwirtschaft. Kiel, Germany. Amin, S., and Pearce, B. (1977). Unequal development: an essay on the social formations of peripheral capitalism. New York: Monthly Review Press. Andriambololona, A. H., Ranaivoson, O. A., and Rasamoelina, M. S. (2004). Evaluation des programmes forestiers environnementaux dans le cadre de la mise en oeuvre de la politique forestière à Madagascar: Rapport final. Antananarivo: Agence Japonaise de Coopération Internationale (JICA) and Ministère de l’Environnement, et des Eaux et Forêt (MINENVEF). Andriamialisoa, F., and Langrand, O. (2003). The history of zoological exploration of Madagascar. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 1-15). Chicago and London: The University of Chicago Press. Antona, M., Motte Bienabe, E., Salles, J. M., Pechard, G., Aubert, S., and Ratsimbarison, R. (2004). Rights transfers in Madagascar biodiversity
732
policies: achievements and significance. Environment and Development Economics, 9, 825-847. Aristotle. (1976). The nicomachean ethics. Harmondsworth: Penguin. Arrow, K., Costanza, R., Dasgupta, P., Folke, C., Holling, C. S., Jansson, B. E., Maler, K. G., Perrings, C., and Pimental, D. (1995). Economic growth, carrying capacity, and the environment. Science, 268, 520-521. Association Nationale pour la Gestion des Aires Protégées (ANGAP) and MINENVEF (Ministère de l'Environnement, des Eaux et Forêts (2003). Cadre de procédure pour la mitigation des impacts de la création des aires protégées, des sites de conservation, et des réserves foncières. Antananarivo: Gouvernement de Madagascar. Association Zanaky ny Ala. (2003). Identification de projet: aménagement des bassins versants dans le fokontany d'Ambodilaingo. Antananarivo: Association Zanaky ny Ala. Astuti, R. (1995). People of the sea: identity and descent among the Vezo of Madagascar. Cambridge: Cambridge University Press. Aubert, S. (1999). Gestion patrimoniale et viabilité des politiques forestières a Madagascar: vers le droit a l'environnement? Unpublished Phd Dissertation, Université de Paris I Pantheon Sorbonnes, Paris. Aubert, S. (2003). Autorités coutumière et régulation sociale. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar: les dynamiques des tavy sur la cote orientale (pp. 107-132). Antananarivo: CIRAD-CITE-FOFIFA. Aubert, S., and Razafiarison, S. (2003). Le tavy et la maîtrise des moyens de subsistance. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar: les dynamiques des tavy sur la cote orientale (pp. 53-75). Antananarivo: CIRAD-CITE-FOFIFA. Aubert, S., Razafiarison, S., and Bertrand, A. (2003). Déforestation et systèmes agraires a Madagascar: les dynamiques des tavy sur la cote orientale. Antananarivo: CIRAD-CITE-FOFIFA. Babin, D. (2004). Beyond tropical deforestation: from tropical deforestation to forest cover dynamics and forest development. Paris: UNESCO-CIRAD. Barck, S., and Moor, P. (1998). Les conditions dynamiques du développement rural dans la région de Beforona. In Projet Terre-Tany and BEMA (Ed.), Les stratégies endogènes et la gestion des ressources naturelles dans la région
733
de Beforona: les résultats des recherches pluridisciplinaires de la phase 1995-1998 (Vol. 8, pp. 11-20). Antananarivo: Projet Terre-Tany / BEMA. Barnett, H., and Morse, C. (1963). Society and growth: the economics of natural resource availability. Baltimore, Md: John Hopkins University Press for Resources for the Future. Barrett, C. B. (1999). Stochastic food prices and slash-and-burn agriculture. Environment and Development Economics, 4, 161-176. Barrett, C. B., Moser, C.M., Barison, J., and McHugh, O. V. (2003). Better Technology, Better Plots or Better Farmers? Identifying Changes In Productivity And Risk Among Malagasy Rice Farmers. Paper presented at the annual meetings of the American Agricultural Economics, Association, July 27-30, 2003, Montreal, Quebec, Canada. Barthod, C., and Ollagnon, H. (1991). Vers une gestion patrimoniale de la protection et de la qualité biologique des forêts. Arbres, Forêts et Communautés Rurales, Bulletin No 3, 32-35. Basset, T. J., and Koli Bi, Z. (2000). Environnemental discourses and the Ivorian svanna. Annals of the Association of American Geographers, 90(1), 67-95. Batterbury, S., Forsyth, T., and Thomson, K. (1997). Environmental transformations in developing countries: hybrid research and democratic policy. Geographical Journal, 163, 126-132. Beaujard, P. (1983). Princes et paysans : les Tanala de l'Ikongo : un espace social du sud-est de Madagascar. Paris: Editions L'Harmattan. Beaujard, P. (1985). Rice from Heaven, Rice from the Earth - Ideology, the PoliticalSystem and Rice Production in the Tanala Kingdoms of Ikongo (Southeastern Coast of Madagascar). Etudes Rurales (99-100), 389-402. Bebbington, A. (1993). Modernization from below: an alternative indigenous development? Economic Geography, 69(3), 274-292. Bebbington, A. (1996). Movements, modernizations, and markets: indigenous organizations and agrarian strategies in Ecuador. In R. Peet and M. Watts (Eds.), Liberation ecologies (pp. 86-109). New York: Routledge. Beinart, W. (1996). Soil erosion, animals, and pasture over the long term: environmental destruction in Southern Africa. In M. Leach and R. Mearns (Eds.), The lie of the land (pp. 54-72). Portsmouth, N.H.: Heinemann.
734
Belsky, J.M. (1999). Misrepresenting communities: the politics of community-based rural ecotourism in Gales Point Manatee, Belize. Rural Sociology, 64(4), 641-666. Bergeret, A. (1993). Discours et politiques forestières coloniales en Afrique et a Madagascar. Revue Francaise d'Histoire d'Outre-Mer, 79(298), 23-47. Bernstein, H. (1979). African peasantry: a theoretical framework. Journal of Peasant Studies, 6(4), 420-444. Bernstein, R. J. (1983). Beyond objectivism and relativism: science, hermeneutics and praxis. Philadelphia: University of Pennsylvania Press. Bertrand, A., and Lemalade, J.-L. (2003a). Riziculture de tavy et sécurité alimentaire. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar. Les dynamiques des tavy sur la cote orientale (pp. 75-85). Antananarivo: CIRAD-CITE-FOFIFA. Bertrand, A., and Lemalade, J.-L. (2003b). Démographie galopante, tavy et déforestation. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar. Les dynamiques des tavy sur la cote orientale (pp. 133-135). Antananarivo: CIRAD-CITEFOFIFA. Bertrand, A., and Randrianaivo, D. (2003). Tavy et déforestation. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar. Les dynamiques des tavy sur la cote orientale (pp. 9-31). Antananarivo: CIRAD-CITE-FOFIFA. Bertrand, A., and Sourdat, M. (1998). Feux et déforestation a Madagascar: revues bibliographiques. Antananarivo: CIRAD/ORSTOM/CITE. Biersack, A. (1987). Moonlight: negative images of transcendence in Paiela pollutions. Oceania, 57, 178-194. Biersack, A. (2006). Reimagining political ecology: culture/power/history/nature. In A. Biersack and J. B. Greenberg (Eds.), Reimagining political ecology (pp. 3-43). Durham: Duke University Press. Biersack, A., and Greenberg, J. B. (2006). Reimagining political ecology. Durham: Duke University Press. Bignell, D. E., Tondoh, J., Dibog, L., Huang, S. P., Moreira, F., Nwaga, D., Pashanasi, B., Guimaraes Pereira, F., Susilo, F.-X., and Swift, M. J. (2005). Belowground biodiversity assessment: developing a key functional group approach in best-bet alternatives to slash-and-burn. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture:
735
the search for alternatives (pp. 119-143). New York: Columbia University Press. Blaikie, P. M. (1985). The political economy of soil erosion in developing countries. London ; New York: Longman. Blaikie, P. M. (1999). A review of political ecology: issues, epistemology, and analytical narratives. Zeitschrift fur Wirtschaftsgeographie, 43(3-4), 131147. Blaikie, P. M., and Brookfield, H. C. (1987). Land degradation and society. London and New York: Methuen. Bloch, M. (1995). People into places: Zafimaniry concepts of clarity. In E. Hirsch and M. O'Hanlon (Eds.), The anthropology of landscape: perspective on places and spaces (pp. 63-77). Oxford: Clarendon. Bookchin, M. (1986). The modern crisis. Philadelphia, Pa: New Society. Borrini-Feyerabend, G., and Dudley, N. (2005a). Elan Durban... Nouvelles perspectives pour les aires protégées a Madagascar: IUCN. Borrini-Feyerabend, G., and Dudley, N. (2005b). Les aires protégées a Madagascar: bâtir le système a partir de la base: IUCN. Boserup, E. (1965). The conditions of agricultural growth: the economics of agrarian change under population pressure. London: Earthscan Publication Ltd. Brady, N. C. (1996). Alternatives to slash-and-burn: a global imperative. Agriculture, Ecosystems and Environment, 58, 3-11. Brand, J. (1997a). L'agroclimat et le cycle de l'eau. In Projet Terre-Tany and BEMA (Ed.), Un système agro-écologique dominé par le tavy: la région de Beforona, falaise Est de Madagascar (Cahier Terre-Tany numero 6, pp. 418). Antananarivo: Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brand, J. (1997b). La region de Beforona et les recherches de Terre-Tany/BEMA. In Projet Terre-Tany and BEMA (Ed.), Un systeme agroecologique domine par le tavy: la region de Beforona, Falaise-Est de Madagascar (Cahier Terre-Tany numero 6, pp. 1-3). Antananarivo: Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brand, J., and Rakotovao, W. L. (1997b). La dégradation des sols. In Projet TerreTany and BEMA (Ed.), Un système agro-écologique dominé par le tavy: la région de Béforona, falaise Est de Madagascar (Cahier Terre-Tany
736
numero 6, pp. 49-59). Antananarivo. Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brand, J., and Randriamboavonjy, J.-B. (1997). L'utilisation des sols. In Projet TerreTany and BEMA (Ed.), Un systeme agroecologique domine par le tavy: la region de Beforona, Falaise-Est de Madagascar (Cahier Terre-Tany numero 6, pp. 89-103). Antananarivo: Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brand, J., and Zurbuchen, J. (1997). La déforestation et le changement du couvert vegetal. In Projet Terre-Tany and BEMA (Ed.), Un système agroécologique dominé par le tavy: la région de Béforona, falaise Est de Madagascar (Cahier Terre-Tany numero 6, pp. 59-68). Antananarivo: Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brand, J., Randrianarisoa, D., and Rakotovao, W. L. (1997). Impact des cyclones. In Projet Terre-Tany and BEMA (Ed.), Un système agro-écologique dominé par le tavy: la région de Béforona, falaise Est de Madagascar (Cahier Terre-Tany numero 6, pp. 19-34). Antananarivo. Projet Terre-Tany / BEMA, CDE/GIUB and FOFIFA Madagascar. Brechin, S. R., Wilshusen, P. R., Fortwangler, C. L., and West, P. C. (2002). Beyond the square wheel: toward a more comprehensive understanding of biodiversity conservation as social and political process. Society and Natural Resources, 15, 41-64. Brechin, S. R., Wilshusen, P. R., Fortwangler, C. L., and West, P. C. (Eds.). (2003). Contested nature: promoting international biodiversity conservation with social justice in the twenty-first century. New York: State University of New York Press. Brosius, J.P., Peter, J., Tsing, A., Lowenhaupt, A., Zerner, C. (1998). Representing communities: histories and politics of community-based natural resource management. Society and Natural Resources, 11(2) Brown, M. (2000). A history of Madagascar. Princeton [N.J.]: Markus Wiener Publishers. Brown, S., Noble, I., and Bosquet, B. (In Press). Financing options for carbon sequestration projects under the clean development mechanism. Agricultural Systems. Brunckhorst, D. J., and Rollings, N. M. (1999). Linking ecological and social functions of landscapes: I: influencing resources governance. Natural Areas Journal, 19(1), 57-64.
737
Bryant, R. L. (1997). The political ecology of forestry in Burma, 1824-1994. Honolulu: University of Hawai Press. Bryant, R. L. (2002). Non-governmental organizations and governmentality: 'consuming' biodiversity and indigenous people in the Philippines. Political Studies, 50(2), 268-292. Bryant, R. L., and Bailey, S. (1997). Third world political ecology. London ; New York: Routledge. Burney, D. A. (1987a). Late Holocene Vegetational Change in Central Madagascar. Quaternary Research, 28(1), 130-143. Burney, D. A. (1987b). Late Quaternary stratigraphic charcoal records from Madagascar. Quaternary Research, 28(2), 274-280. Burney, D. A. (1988). Modern pollen spectra from Madagascar. Paleogeography, Paleoclimatology, Paleoecology, 66, 63-75. Burney, D. A. (1996). Climate change and fire ecology as factors of the quaternary biogeography of Madagascar. In Lourenco. W.R., (Ed.), Biogéographie de Madagascar (pp. 49-58). Paris: Edistions ORSTOM. Burney, D. A. (1997a). Tropical islands as paleoecological laboratories: Gauging the consequences of human arrival. Human Ecology, 25(3), 437-457. Burney, D. A. (1997b). Theories and facts regarding holocene environmental change before and after human colonization. In S. M. Goodman and B. D. Patterson (Eds.), Natural change and human impact in Madagascar. Washington D.C.: Smithsonian Institution Press. Burney, D. A. (1999). Rates, patterns, and processes of landscape transformation and extinction in Madagascar. In Macphee, R. D. E. (Ed.), Extinction in near time: causes, contexts and consequences. New-York: Plenum. Burney, D. A. (2003). Madagascar's prehistoric ecosystems. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 47-51). Chicago: University of Chicago Press. C3EDM. (2005). Indicateurs Economiques pour l'Evaluation des Transferts de Gestion - maquette d'ouvrage technique- rapport final. Antananarivo: C3EDM (Centre d’économie et d’éthique pour l’environnement et le développement à Madagascar). Cabinet Colin Rabenitany (CCR). (1995). Rapport d'évaluation par les bénéficiaires du Programme Environnemental phase I: ANGAP - ANAE -D.D. Antananarivo: Office National de l'Environnement - Banque Mondiale.
738
Callon, M., Law, J., and Rip, A. (Eds.). (1986). Mapping the dynamics of science and technology : sociology of science in the real world. Basingstoke, Hampshire: Macmillan. Campese, J., Borrini-Feyerabend, G., de Cordova, M., Guigner, A. and Oviedo, G. (Eds.) (2007). Conservation and human rights. Policy matter, 15. IUCN Commission on Environmental, Economic and Social Policy. Carpentier, C. L., Vosti, S. A., and Witcover, J. (2005). Coffee, pastures, and deforestation in the Western Brasilian Amazon: a farm-level bioeconomic model. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 233247): Columbia University Press, New York. Carret, J.-C., and Loyer, D. (2003). Comment financer durablement le réseau d'aires protégées terrestres a Madagascar? Apport de l'analyse économique. Antananarivo: The World Bank and Agence Française de Développement. Carson, R., Darling, L., and Darling, L. (1962). Silent spring. Boston and Cambridge, Mass.: Houghton Mifflin Riverside Press. CAST. (1995). Evaluation par les bénéficiaires du projet environnemental phase I zone est. Antananarivo: Office National pour l'Environnement. Castree, N. (2001). Socializing nature: theory, practice and politics. In N. Castree and B. Braun (Eds.), Social nature (pp. 1-22). Malden, USA and Oxford, UK: Blackwell Publishing. Castree, N., and Braun, B. (2001). Social nature. Malden USA and Oxford UK: Blackwell Publishing. Castree, N., and MacMillan, T. (2001). Dissolving dualism: actor-networks and the reimagination of nature. In N. Castree and B. Braun (Eds.), Social nature (pp. 208-225). Malden USA and Oxford UK: Blackwell Publishing. Cattanao, A., and San, N. N. (2005). The forest for the trees: the effects of macroeconomic factors on deforestation in Brasil and Indonesia. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slashand-burn agriculture: the search for alternatives (pp. 170-198). New York: Columbia University Press. Chambers, R., Pacey, A., and Thrupp, L. A. (1985). Farmer first : farmer innovation and agricultural research. London: Intermediate Technology Publ. Chapin, M. (2004). A challenge to conservationists. World Watch Magazine. November-December 2004.
739
Chase, J. M., and Leibold, M. A. (2003). Ecological niches: linking classical and contemporary approaches. Chicago and London: The University of Chicago Press. Chauvet, B. (1972). The forests of Madagascar. In R. Battistini and G. RichardVindard (Eds.) Biogeography and Ecology in Madagascar, pp. 191-199. The Hague: W. Junk. CIA. (2007). The World Factbook: Madagascar. Retrieved, from the World Wide Web: https://www.cia.gov/cia/publications/factbook/geos/ma.html Clements, F. E. (1916). Plant succession: an analysis of the development of vegetation. Washington: Carnegie Inst. Washington. Publ. Cochet, H. (2003). Agrarian dynamics, population growth and resource management: the case of Burundi. Geojournal, 0(1), 1-10. Cochet, H. (2005). L'agriculture comparée: genèse et formalisation d'une discipline scientifique. Paris: Institut National Agronomique Paris-Grignon. Cole, J. (2001). Forget colonialism? Sacrifice and the art of memory in Madagascar. Berkeley: University of California Press. Commoner, B. (1972). The closing circle: confronting the environmental crisis. London: Cape. Conklin, H. C. (1957). Hanunoo agriculture: a report on an integral system of shifting cultivation in the philippines (FAO Forestry Development Paper No 12). Rome: Food and Agriculture Organization of the United Nations. Connell, J. H. (1978). Diversity in tropical rain forests and coral reefs. Science(199), 1302–1310. Connel, J. H., and Slatyer, R. O. (1977). Mechanisms of succession in natural communities and their role in community stability and organisation. American Naturalist, 111, 1119-1144. Cooke, B., and Kothari, U. (2001). The case for participation as Tyranny. In B. Cooke and U. Kothari (Eds.), Participation: the new tyranny? London: Zed Book. Cormier-Salem, M.-C., Boutrais, J., Roussel, B., and Juhé-Beaulaton, D. (Eds.). (2002). Patrimonialiser la nature tropicale : dynamiques locales, enjeux internationaux. Paris: IRD Ed. Craig, D. and Porter, D. (1997). Framing participation: development projects, professionals, and organizations. Development in Practice, 7(3), 229-236.
740
Cronon, W. (1983). Changes in the land : Indians, colonists, and the ecology of New England. New York : Hill and Wang, Damasio, A. R. (1994). Descartes’ error : emotion, reason, and the human brain. New York: G.P. Putnam. De Coignac, B., Bailly, Malvos, Hueber, and Ramanahadry, M. (1973). Essai d'aménagement des terres dans la zone forestière de l'est de Madagascar: expériences des villages de Marolafa et Andranomody. Bois et Forêts des Tropiques, 152. Demeritt, D. (2001). Being constructive about nature. In N. Castree and B. Braun (Eds.), Social nature (pp. 22-41). Malden USA and Oxford UK: Blackwell Publishing. Deschamps. H. J. (1959). Les migrations intérieures passées et présentes à Madagascar. Paris: Éditions Berger-Levrault. Deschamps H. J. (1972). Early kingdoms in Madagascar, 1500-1700. Journal of African History, 13(1), 151-152. Dewar, R. E. (2003). Relationship between human ecological pressures and the vertebrate extinctions. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 119-123). Chicago: University of Chicago Press. Dewar, R. E., and Wright, H. T. (1993). The Culture History of Madagascar. Journal of World Prehistory, 7(4), 417-466. Direction du Département charge de la sécurité civile et de la conservation de l'environnement. (2003). Arrêté provincial No 037-MIRA/DS/PA/TOA.fixant les interdictions, alternatives et sanctions relatives à la pratique des feux de brousse et de végétation (TAVY). Antananarivo: Délégation Spéciale de la Province Autonome de Toamasina. Dixon, J., Gulliver, A., and Gibbon, D. (2001). Farming systems and poverty. Improving farmers' livelihood in a changing world: Food and Agriculture Organization, Rome. Donner, S. D., Skirving, W. J., Little, C. M., Oppenheimer, M., and Hoegh-Guldberg, O. (2005). Global assessment of coral bleaching and required rates of adaptation under climate change. Global Change Biology, 11(12), 22512265. Dorosh, P., Haggblade, S., Lungren, C., Razafimanantena, T., and Randriamiarana, Z. (2003). Economic Motors for Poverty Reduction in Madagascar1: Cornell University, ILO program.
741
Dove, M. R. (1983). Theories of swidden agriculture, and the political economy of ignorance. Agroforestry Systems, 1(2), 85-99. Dove, M. R. (2000). The life cycle of indigeneous knowledge, and the case of natural rubber production. In R. Ellen and P. Parkes and A. Bicker (Eds.), Indigenous environmental knowledge and its transformations: critical anthropological perspective (pp. 213-251). Amsterdam: Harwood Academic Publisher. Dreyfus, H. L., and Dreyfus, S. E. (1986). Mind over machine. New York: The Free Press Macmillan. Dryzek, J. S. (2005). The politics of the earth: environmental discourses: Oxford University Press. Duffield, M. (2001). Governing the borderlands: decoding the power of aid. Disasters, 25 (4), 308-320. Dufils, J.-M. (2003). Remaining forest cover. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 88-96). Chicago and London: The University of Chicago Press. Dufumier, M. (1996). Les projets de developpement agricole: manuel d'expertise. Paris: CTA Karthala. Du Puy, D. J., and Moat, J. (1996). A refined classification of the primary vegetation of Madagascar based on the underlying geology: using GIS to map its idstribution and to assess its conservation status. In W. R. Lourenco (Ed.), Biogeographie de Madagascar (pp. 205-218). Paris: Editions ORSTOM. Du Puy, D. J., and Moat, J. (2003). Using geological substrate to identify and map primary vegetation types in Madagascar and the implication for planning biodiversity conservation. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 51-74). Chicago: University of Chicago Press. Durbin, J. Andrianarimisa, A. and P. J. DeCosse. (2001). Le potentiel des contrats de conservation pour contribuer à la conservation de la biodiversité à Madagascar. Projet d’Appui à la Gestion de l’Environnement and International Resources Group, Ltd., Antananarivo. Egler, F. E. (1954). Vegetation science concepts I. Initial floristic composition, a factor in old-field vegetation development. Vegetatio, 4, 412-417. Ehrlich, P. R., and Ehrlich, A. H. (1970). Population, resources, environment: issues in human ecology. San-Francisco: W.H. Freeman and Company.
742
Ehrlich, P. R., and Ehrlich, A. H. (1996). Betrayal of science and reason: how antienvironmental rhetoric threatens our future. Washington D.C. and Covelo, California: Island Press and Shearwater Book. Ellis, W. (2004). Three Visits to Madagascar During the Years 1853 to 1856 Including A Journey to the Capital with Notices of the Natural History of the Country and of the Present Civilization of the People. Kila, Mt: Kessinger Publishing. Encyclopedia Britannica (2002). The New Encyclopaedia Britannica, 15th edition Escobar, A. (1995). Ecountering development: the making and unmaking of the third world. Princeton: Princeton University Press. Escobar, A. (1999). After nature: steps to an antiessentialist political ecology. Current Anthropology, 40(1), 1-30. Esoavelomandroso, M. (1988). La destruction de la forêt par l'homme malgache: un problème mal posé. Antananarivo: Ministère de la Recherches Scientifique et Technologique pour le Développement. Espeland, W. N. (1998). The struggle for water: politics, rationality, and identity in the American Southwest. Chicago: Chicago University Press. Fairhead, J. (2001). International dimensions of conflict over natural and environmental resources. In N. L. Peluso and M. Watts (Eds.), Violent environment (pp. 3-38): Cornell University Press. Fairhead, J., and Leach, M. (1996). Misreading the African landscape: society and ecology in a forest-savanna mosaic. Cambridge: Cambridge University Press. Fairhead, J., and Leach, M. (1998). Reframing deforestation: global analysis and local realities: studies in West Africa. London and New-York: Routledge. Fara Lala, R. (1999). Analyse de l'interface humain forêt et directive d'aménagement du bassin versant de Vohidrazana, falaise est de Madagascar. Université d'Antananarivo, Antananarivo. Faramalala, M. H. (1995). Formations végétales et domaine forestier national de Madagascar. Antananarivo: Conservation International. Fay, C., De Foresta, H., Sarait, M., and Tomich, T. P. (1998). A policy breakthrough for Indonesian farmers in the Krui damar agroforests. Agroforestry Today, 10(2), 25-26.
743
Feeley-Harnik, G. (1995). Plants and people, children or wealth: shifting grounds of “choice” in Madagascar. Political and Legal Anthropology Review, 18(2), 45-64. Ferguson, J. (1994). The anti-politics machine: "development,” depoliticization and bureaucratic power in Lesotho. Minneapolis, London: University of Minnesota Press. Ferraro, P. J. (2002a). The local costs of establishing protected areas in low-income nations: Ranomafana National Park, Madagascar. Ecological Economics, 43(2-3), 261-275. Ferraro, P. (2002b). The cost-effectiveness of conservation payments. Land Economics, 78(3), 339-353. Ferraro, P. J., and Kiss, A. (2002). Direct payments to conserve biodiversity. Science, 298(5599), 1718-1719. Feyerabend, P. (1975). Against method. London: New left books. Feyerabend, P. (1994). Art as a product of nature as a work of art. World Futures, 40, 87-100. Feyerabend, P. (2003). La science en tant qu'art. Paris: Albin Michel. Flacourt, E. (1658). Histoire de la grande isle de Madagascar avec une relation de ce qui s'est passe em annees 1655, 1656 et 1657 non encore vue par la premiere. Paris: Chez Pierre Bien-fait. Fleming, J. D. (1974) The state of the apes. Psychology today, 7, 31-38. Flyvbjerg, B. (2001). Making social sciences matter: why social enquiry fails and how it can suceed again. Cambridge: Cambridge University Press. FOFIFA and Programme ILO (2003). Agriculture et pauvreté: conférence du 20 mars 2003, Antananarivo. Antananarivo: FOFIFA and Programme ILO. Food and Agriculture Organization. (1974). Shifting cultivation and soil conservation in Africa. Papers presented at the FAO/SIDA/ARCN regional seminar on shifting cultivation and soil conservation in Africa. Ibadan, Nigeria, 2-21 July 1973. Soils Bulletin. Roma: FAO. Food and agriculture Organization (FAO). (1984). Changes in shifting cultivation in Africa. FAO Forestry Paper, 50. Food and agriculture Organization (FAO). (1985). Tropical Forestry Action Plan. Rome: FAO.
744
Food and Agriculture Organization (FAO). (1997). State of the world's forests. Roma: FAO. Forcada, J., Trathan, P. N., Reid, K., and Murphy, E. J. (2005). The effects of global climate variability in pup production of Antarctic fur seals. Ecology, 86(9), 2408-2417. Forsyth, T. (2003). Critical political ecology : the politics of environmental science. London ; New York: Routledge. Foucault, M. (1961). Folie et déraison: histoire de la folie à l'age classique. Paris: Plon. Foucault, M. (1963). Naissance de la clinique: une archéologie du regard médical. Paris: Presses universitaires de France. Foucault, M. (1969). L'archéologie du savoir. Paris: Gallimard. Foucault, M. (1970). L'ordre du discours: leçon inaugurale au collège de France prononcée le 2 décembre 1970. Paris: Gallimard. Foucault, M. (1975). Surveiller et punir. Paris: Gallimard. Foucault, M. (1980). Power/knowledge: selected interviews and other writings 19721977. Brighton: Harvester. Frank, A. G. (1969a). Capitalism and underdevelopment in latin America: historical studies of Chile and Brasil. New-York: Monthly Review Press. Fry, G. L. A. (2001). Multifunctional landscapes: toward transdisciplinary research. Landscape and Urban Planning, 57(3-4), 159-168. Galeano, E. (1974) Open veins of Latin America; five centuries of the pillage of a continent. New York: Monthly Review Press. Gautier, L., and Goodman, S. M. (2003). Introduction to the flora of Madagascar. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 229-250). Chicago: University of Chicago Press. Geist, H. J., and Lambin, E. F. (2002). Proximates causes and underlying driving forces of tropical deforestation. BioScience, 52(2), 143-149. Ghimire, K. B., and Ghai, D. (1994). Parks and people: livelihood issues in national parks management in Thailand and Madagascar. Development and environment: sustaining people and nature. Development and Change, 25(1), 195-229.
745
Ghimire, K. B., and Pimbert, M. P. (Eds.). (1997). Social change and conservation: environmental politics and impacts of National Parks and protected areas. London: Earthscan Pubilication Limited. Gillison, A. N. (2005). The potential of abovre-ground biodiversity indicators in assessing best-bet alternatives to slash-and-burn. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 83-118). New York: Columbia University Press. Girod-Genet, L. (1898a). Renseignements forestiers, tournée du chef de service des forêts de Tananarive, Antanifotsy, Ambositra, Fianarantsoa. Journal Officiel de Madagascar, 294, 2349-2350. Girod-Genet, L. (1898b). Renseignements forestiers, tournée du chef de service des forêts de Tananarive, Antanifotsy, Ambositra, Fianarantsoa. Journal Officiel de Madagascar, 295, 2356. Gleason, H. A. (1926). The individualistic concept of the plant association. Bulletin Torrey Botanical Club, 53, 7-26. Gleick, J. (1987). Chaos: making a new science. New-York: Viking-Pinguin. Glynn, P.W. (1991) Coral reef bleaching in the 1980s and possible connections with global warming. Trends in Ecology and Evolution, 6, 6, 175-179. Gockowski, J., Tonye, J., Diaw, C., Hauser, S., Kotto-Same, J., Njongang, R., Moukam, A., Nwaga, D., Tiki-Manga, T., Tondoh, J., Tschondeau, Z., Weise, S., and Zapfack, L. (2005). The forest margins of Cameroon. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-andburn agriculture: the search for alternatives (pp. 305-331). New York: Columbia University Press. Gonzales, G. A. (2001). Corporate power and the environment: the political economy of US environmental policy. Lanham, Md: Rowman and Littlefield. Goodman, S. M., and Benstead, J. P. (2003). The natural history of Madagascar. Chicago: University of Chicago Press. Goodman, S. M., and Patterson, B. D. (Eds.). (1997). Natural change and human impact in Madagascar. Washington D.C.: Smithsonian Institution Press. Gouvernement de Madagascar. (1960). Ordonnance No 60-127 fixant le régime des défrichements et des feux de végétation. Antananarivo: Gouvernement de Madagascar.
746
Gouvernement de Madagascar. (1962). Ordonnance n° 62-121 Portant modification de l'ordonnance n° 60-127 du 3 Octobre 1960 relative au régime des défrichements et des feux de végétation. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (1990). Loi n° 90-033 relative à la Charte de l’environnement malagasy. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (1996). Loi 96/025 relative a la gestion locale des ressources naturelles renouvelables. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (1997). Loi 97/017 portant révision de la législation forestière. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (1998). Décret n° 98-610 réglementant les modalités de la mise en œuvre de la Sécurisation Foncière Relative. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2000a). Décret No 2000/028 relatif aux médiateurs environnementaux. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2000b). Décret No 2000/027 relatif aux communautés de base chargées de la gestion locale des ressources naturelles renouvelables. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2001). Décret No 2001/122 fixant les conditions de mise en oeuvre de la gestion contractualisée des forêts de l'état. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2002). Décret N 2002-793 définissant les mesures incitatives a la prévention et a l'éradication des feux de brousse. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2004a). Arrêté interministeriel No19560 Portant suspension de l'octroi de permis minier et de permis forestier dans les zones réservées comme "sites de conservation.” Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2004b). Arrêté No 21694 relatif à la suspension de toute activité extractive de ressources ligneuses dans les zones réservées comme Sites de Conservation. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2005a). Arrêté portant définition des critères de classification et des règles de gestion des périmètres soumis a des régimes spéciaux. Antananarivo: Gouvernement de Madagascar.
747
Gouvernement de Madagascar. (2005b). Décret 2005/848 du 13 Décembre 2005 appliquant l'article 2 alinéa 2 de la loi No2001/15 portant Code des Aires Protégées. Antananarivo: Gouvernement de Madagascar. Gouvernement de Madagascar. (2006). Loi N° 2006-031 du 24 novembre 2006 fixant le régime juridique de la propriété foncière privée non titrée. Antananarivo: Gouvernement de Madagascar. Green, G. M., and Sussman, R. W. (1990). Deforestation history of the eastern rain forests of Madagascar from. Science, 13, 212-215. Greenwood, d. J., and Levin, M. (1998). Introduction to action research: social research for social change. London: Sage Publication. Gross, P. R., and Levitt, N. (1994). Higher superstition: the academic left and its quarrel with science. Baltimore and London: The John Hopkins University Press. Grossman, L. S. (1998). The political ecology of bananas: contract farming, peasants, and agrarian change in the eastern Caribbean. Chapel Hill: University of North Carolina Press. Groupe Vision Durban (2004). Document de travail du Groupe technique et Juridique pour la Gestion et Classification des Aires Protégées: Sites de Conservation. Version du 4 mars 2005. Antananarivo: Groupe Vision Durban, Ministère de l’Environnement, des Eaux et Forêts. Groupement Semis Direct Madagascar (GSDM). (2006). Atelier sur la diffusion des techniques agro-ecologiques a Madagascar (12-13 Octobre 2006 au Hilton Madagascar): rapport final. Antananarivo: GSMD, MAEP, AFD, ECR and CIRAD. Guha, R. (1990). The unquiet woods : ecological change and peasant resistance in the Himalaya. Berkeley: University of California Press. Guruswamy, L. D., and McNeely, J. A. (Eds.). (1998). Protection of global biodiversity: converging strategies. Durham and London: Duke University Press. Hairiah, K. (2005). Sustainability of tropical land use systems after forest conversion. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 143-169). New York: Columbia University Press. Hallé, F., Oldeman, R. A. A., and Tomlinson, P. B. (1978). Tropical trees and forests: an architecture analysis. New-York: Springer Verlag.
748
Hanson, P. W. (1996). Coming to terms with the people of Madagascar. Retrieved, from the World Wide Web. Hanson, P. W. (1997). The Politics of Need Interpretation in Madagascar's Ranomafana National Park (Conservation, Economic Development, Tanala). Unpublished PhD Dissertation, University of Pennsylvania. Hardenbergh, S. H. B. (1996). Behavioral quality and caloric intake in Malagasy children relative to international growth reference. American Journal of Human Biology, 8, 207-223. Hardenbergh, S. H. B. (1997). Why are boys so small? Child growth, diet and gender near Ranomafana, Madagascar. Social Science and Medicine, 44(11), 1725-1738. Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243-1248. Hardin, G. (1974). Lifeboat Ethics - Case against Helping Poor. Psychology Today, 8(4), 38-and. Hardwood, R. H. (1996). Development pathways toward sustainable systems following slash-and-burn. Agriculture, Ecosystems and Environment, 58, 75-86. Harper, J. (2002). Endangered species : health, illness and death among Madagascar's people of the forest. Durham, N.C.: Carolina Academic Press. Harrison, E. (2002). The problem with the locals': partnership and participation in Ethiopia. Development and Change, 33(4), 587-610. Hartmann, B. (2001). Will the circle be unbroken? a critique of the project on environment, population and security. In N. L. Peluso and M. Watts (Eds.), Violent environment (pp. 39-64). Ithaca and London: Cornell University Press. Hastie, J. (No date). Le voyage de Tananarive en 1917. Hawkins, F., and Horning, N. (2001). Forest cover change in USAID and control areas: a preliminary evaluation and report to USAID and partners. Washington DC: Projet d'Appui a la Gestion de l'Environnement, IRG. Head, L. (2000). Second nature. Syracuse: Syracuse University Press. Hecht, S. B., and Cockburn, A. (1989). The fate of the forest : developers, destroyers, and defenders of the Amazon. London and New York: Verso.
749
Hladick, A., Linares, O. F., Hladik, C. M., and Pagezy, H. (Eds.). (1993). Tropical forests, people and food : biocultural interactions and applications to development. Paris: UNESCO, Carnforth, UK, Pearl River, N.Y. and Parthenon Pub. Group. Hockley, N. J., and Razafindralambo, R. (2006). A social cost-benefit analysis of conserving the Ranomafana-Andringitra-Pic d'Ivohibe Corridor in Ranomafana. Antananarivo: University of Wales and Conservation International. Homer-Dixon, T. (1998). Environment, scarcity, and violence. Princeton and Oxford: Princeton University Press. Homer-Dixon, T. (2000). The ingenuity gap: can we solve the problem of the future? : Vintage Canada Edition, Random House Canada. Horn, H. S. (1974). The Ecology of Secondary Succession. Annual Review of Ecology and Systematics, 5, 25-37. Horta, K. (2000). Rainforest: biodiversity conservation and the political economy of international financial institutions. In P. A. Stott and S. Sullivan (Eds.), Political ecology : science, myth and power (pp. 179-203). London and New York: Arnold, Oxford University Press. Hubbell, S. P. (2001). The unified neutral theory of biodiversity and biogeography. Princeton and Oxford: Princeton University Press. Hughes, T. P., Baird, A. H., Bellwood, D. R., Card, M., Connolly, S. R., Folke, C., Grosberg, R., Hoegh-Guldberg, O., Jackson, J. B. C., Kleypas, J., Lough, J. M., Marshall, P., Nyström, M., Palumbi, S. R., Pandolfi, J. M., Rosen, B., and Roughgarden, J. (2003). Climate Change, Human Impacts, and the Resilience of Coral Reefs. Science, 301(5635), 929 - 933. Humbert, H. (1927). La destruction d'une flore insulaire par le feu; principaux aspects de la végétation à Madagascar, documents photographiques et notices. Tananarive: Impr. Moderne de l'Emyrne G. Pitot et Cie. Humbert, H. (1949). La dégradation des sols à Madagascar. Mémoires de l'Institut de Recherche Scientifique de Madagascar, 1, 33-53. Humbert, H., and Cours Darne, G. (1964). Carte internationale du tapis végétal et des conditions écologiques, République Malgache. Pondichéry: Institut français de Pondichéry. Humbert, H., and Cours Darne, G. (1965). Notice de la carte Madagascar. Pondichéry: Institut français de Pondichéry.
750
Hume, D. (1882). A treatise on human nature; being an attempt to introduce the experimental method of reasoning into moral subjects. London: Longmans, Green. Huttington, S. (1996). The clash of civilizations and the remaking of world order. New-York: Simon and Shuster. Ingram, J. C. and Dawson, T. P. (2005) Climate change impacts and vegetation response on the island of Madagascar. Philosophical Transactions of the Royal Society A, 363, 55-59 Institut Geographique National - Paris (annexe de Tananarive). (1963). Carte au 1/50000eme. Lakato Feuille S-47 Sud Intercooperation, MIRAY, and Conservation International. (2000). Les premiers pas de la gestion contractualisée des forêts a Madagascar. Rapport de l'atelier de Mantasoa du 28 au 30 novembre 2000. Antananarivo. Intergovernmental Panel on Climate Change (IPCC). (2007). Climate Change 2007: Impacts, Adaptation and Vulnerability. Genf: Intergovernmental Panel on Climate Change (IPCC). International Emissions Trading Association (IETA). (2006). State and trends of the carbon market 2006. Washington DC: The World Bank. International Resources Group (IRG). (2001). Atelier national sur le mécanisme de développement propre: un outil pour financer le développement durable durable [CD ROM]. Antananarivo: Ministère de l'Environment, Madagascar. IUCN (World Conservation Union). (1994). Guidelines for Protected Area Management Categories. Gland, Switzerland and Cambridge, UK: IUCN. IUCN (World Conservation Union). (2004). Speaking a Common Language: An investigation into the uses and performance of the IUCN System of Management Categories for Protected Areas. Jacks, G. V., and Whyte, R. O. (1939). The rape of the earth: a world survey of soil erosion. London: Faber and Faber ltd. James, A. N., Gaston, K. J. and Balmford, A. (1999). Balancing the eart’s accounts. Nature. 401, 323-324. Jarosz, L. (1993). Defining and explaining tropical deforestation: shifting cultivation and population growth in colonial Madagascar (1896-1940). Economic Geography, 69, 366-379.
751
Jenkins, M. (1987). Madagascar : an environmental profile. Gland, Switzerland: IUCN Conservation Monitoring Centre. Johnson, B. L. (1999). The role of adaptive management as an operational approach for resources management agencies. Conservation Ecology, 3(2). Jolly, A. (1980). A world like our own : man and nature in Madagascar. New Haven, Connecticut: Yale University Press. Jolly, A. (2004). Lords and lemurs : mad scientists, kings with spears, and the survival of diversity in Madagascar. Boston: Houghton Mifflin Co. Jolly, A., and Leonard, H. J. (1989). The Madagascar challenge: human needs and fragile ecosystems. In H. J. Leonard (Ed.), Environment and the poor: development strategies for a common agenda. New Brunswick, USA: Transaction Books. Julien, G. (1909). Institutions politiques et sociales de Madagascar. Paris : Guilmoto. Juo, A. S. R., and Manu, A. (1996). Chemical dynamics in slash-and-burn agriculture. Agriculture, Ecosystems and Environment. 1996. 58 (1). 49-60. Kant, I. (2003). Critique of pure reason. Houndmills, Basingstoke, Hampshire, New York: Palgrave Macmillan, 2003. Karsenty, A. (2004). Des rentes contre le développement? Les nouveaux instruments d'acquisition mondiale de la biodiversité et l'utilisation des terres dans les pays tropicaux. Mondes en Développement, 127(3), 59-72. Karsenty, A. (2007). Questioning rent for development swaps: new market-based instruments for biodiversity acquisition and the land use issue in tropical countries. International Forestry Review, 9(1), 503-513. Kaufmann, J. C. (2006). The sad opaqueness of the environmental crisis in Madagascar. Conservation and Society, 4(2), 179-193. Kellert, S.R., Mehta, J.N., Ebbin, S.A. and Lichtenfeld, L.L. (2000). Community natural resources management: promise, thetoric and reality. Society and Natural Resources, 13, 705-715. Kightlinger, L. K., Seed, J. R., and Kightlinger, M. B. (1996). Ascaris lumbricoides aggregation in relation to child growth status, delayed cutaneous hypersensitivity, and plant anthelmintic use in Madagascar. Journal of Parasitology, 82(1), 25-33. Kightlinger, L. K., Seed, J. R., and Kightlinger, M. B. (1998). Ascaris lumbricoides intensity in relation to environmental, socioeconomic, and behavioral
752
determinants of exposure to infection in children from southeast Madagascar. Journal of Parasitology, 84(3), 480-484. Klein, J. (2002). Deforestation in the Madagascar highlands - established 'truth' and scientific uncertainty. Geojournal, 56(3), 191-199. Klein, J. (2004). Fiddling while Madagascar burns. Deforestation discourses and highland history. Norwegian Journal of Geography, 58, 11-22. Kleypas, J. A., Buddemeier, R. W., Archer, D., Gattuso, J. P., Langdon, C., and Opdyke, B. N. (1999). Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science, 284(5411), 118-120. Koechlin, J., Guillaumet, J. L., and Morat, P. (1974). Flore et végétation de Madagascar. Lehre: J. Cramer. Kothari, U. (2005). Authority and Expertise: The Professionalisation of International Development and the Ordering of Dissent. Antipode. 37(3), 425-446. Kottak, C. P. (1971). Cultural Adaptation, Kinship, and Descent in Madagascar. Southwestern Journal of Anthropology, 27(2), 129-147. Kottak, C. P. (1980). The past in the present : history, ecology, and cultural variation in highland Madagascar. Ann Arbor: University of Michigan Press. Kottak, C. P., and Costa, A. C. G. (1993). Ecological awareness, environmentalist action, and international conservation strategy. Human Organization, 52, 335-343. Kramer, R. A., Richter, D. D., Pattanayak, S., and Sharma, N. P. (1997). Ecological and economic analysis of watershed protection in Eastern Madagascar. Journal of Environmental Management, 49(3), 277-295. Kuhn, T. (1970). The structure of scientific revolutions. Chicago: The University of Chicago Press. Kull, C. A. (1996). The evolution of conservation efforts in Madagascar. International Environmental Affairs, 8(1), 50-86. Kull, C. A. (2000). Deforestation, erosion, and fire: degradation myths in the environmental history of Madagascar. Environment and History, 6(4), 423450. Kull, C. A. (2004). Isle of fire: the political ecology of landscape burning in Madagascar. Chicago: University of Chicago Press.
753
Landscape Development Intervention (LDI). (2000). Landscape Development Intervention: annual report July 1999 - June 2000. Antananarivo: Chemonics International, LDI-USAID with Cornell University. Landscape Development Intervention (LDI). (2001). Landscape Development Intervention: annual report. Overview and LDI regular activities July 2000 - June 2001. Antananarivo: Chemonics International, LDI-USAID with Cornell University. Landscape Development Intervention (LDI). (2002). Landscape Development Intervention: annual report Volumes I and II, July 2001 - June 2002. Antananarivo: Chemonics International, LDI-USAID with Cornell University. Landscape Development Intervention (LDI). (2003). Landscape Development Intervention: annual report. Overview and LDI regular activities July 2002 - June 2003. Antananarivo: Chemonics International, LDI-USAID with Cornell University. Laney, R. M. (2002). Disaggregating induced intensification for land-change analysis: a case study from Madagascar. Annals of the Association of American Geographers, 92(4), 702-726. Langrand, O. (2003). Reserve speciale d'Ambohitantely. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of madagascar (pp. 1472-1476). Chicago and London: The University of Chicago Press. Lapenu, C., Zeller, M., Minten, B., Randrianaivo, D., Randrianarisoa, C., Ralison, E. (1998). L’extensification agricole et son impact sur les pratiques agricoles et les ressources naturelles : résultats bases sur les données communautaires. Cahier de la Recherche sur les Politiques Alimentaires, 11. Antananarivo : IFPRI and FOFIFA. Larson, B. A. (1994). Changing the Economics of Environmental Degradation in Madagascar - Lessons from the National Environmental Action Plan Process. World Development, 22(5), 671-689. Lassoie, J. P., and Buck, L. E. (1999). Exploring the opportunities for agroforestry in changing rural landscapes. Paper presented at the Fifth Conference on Agroforestry in North America, Cornell University, Ithaca, NY. Latour, B. (1986). Laboratory life : the construction of scientific facts. Princeton: Princeton University Press. Latour, B. (1987). Science in action: how to follow scientists and engineers through society. Cambridge, Massachussets: Harvard University Press.
754
Latour, B. (1993). We have never been modern. Cambridge, Massachussets: Harvard University Press. Latour, B. (1999). Politiques de la nature: comment faire entrer les sciences en démocratie. Paris: La Découverte. Latour, B. (2006). Why has critique run out of steam? from matters of facts to matters of concern. Critical Inquiry, Winter 2004, 225-248. Laurier, E., and Philo, C. (1999). X-morphizing: a review essay of Aramis. Environment and Planning A, 31, 1047-1073. Lavauden, L. (1934). Histoire de la législation et de l'administration forestière a Madagascar. Revue des Eaux-et-Forêts, 72, 949-960. La Vina, A., Fransen, L., Faeth, P. and Kurauchi, Y. (2007). Agricultural subsidies, poverty and the environment: supporting a domestic reform agenda in developing countries. WRI Policy Note (1). Washington, DC: World Resources Institute. Leach, M., Fairhead, J., and Amanor, K. S. (2002). Science and the policy process: perspectives from the forests. IDS Bulletin, 33(1). Leach, M., and Mearns, R. (1996). The lie of the land. Portsmouth, N.H.: Heinemann. Leach, M., Mearns, R., and Scoones I (1999). Environmental Entitlements: Dynamics and Institutions in Community-Based Natural Resource Management. World Development 27(2), 225-247. Le Bourdiec, F. (1974). Hommes et paysages du riz à Madagascar : étude de géographie humaine. Antananarivo: Imprimerie FTM. Lee, K. N. (1993). Compass and gyroscope: integrating science and politics for the environment. Washington, D.C: Island Press. Lee, K. N. (1999). Appraising Adaptive Management. Conservation Ecology, 3(2). Le Roy Ladurie, E. (1969). Les paysans du Languedoc. Paris: Flammarion. Lesser, M. P. (2007). Coral reef bleaching and global climate change: Can corals survive the next century? Proceedings of the National Academy of Sciences of the United States of America, 104(13), 5259-5260. Levitt, S.D. and Dubner, S.J. (2006). Freakonomics : a rogue economist explores the hidden side of everything. New York : William Morrow.
755
Lindblade, K., Carswell, G. and Tumuhairwe, J. K. (1998). Mitigating the relationship between population growth and land degradation - Land-use change and farm management in southwestern Uganda. Ambio, 27 (7), 565-571. Little, P. D., and Horowitz, M. M. (1987). Lands at risk in the third world : local-level perspectives. Boulder: Westview Press. Locatelli, B. (2000). Pression démographique et construction du paysage rural des tropiques humides : l’exemple de Mananara (Madagascar). Unpublished PhD Dissertation, Ecole Nationale du Genie Rural et des Eaux et Forêts, Montpellier. Lomborg, B. (2001). The skeptical environmentalist: measuring the real state of the world. Cambridge: Cambridge University Press. Loreau, M., and Mouquet, N. (1999). Immigration and the maintenance of local species diversity. American Naturalist, 154, 427-440. Lowry II, P. P., Schatz, G. E., and Phillipson, P. B. (1997). The classification of natural and anthropogenic vegetation in Madagascar. In S. M. Goodman and B. D. Patterson (Eds.), Natural change and human impact in Madagascar (pp. 93-123). Washington D.C.: Smithsonian Institution Press. Mac Phee, R. D. E., and Burney, D. A. (1991). Dating of modified femora of extinct dwarf Hippopotamus from southern Madagascar: implications for constraining human colonization and vertebrate extinction events. Journal of Archaeological Science, 18, 695-706. MacArthur, R. H. (1970). Species packing and competitive equilibrium for many species. Theoretical Population Biology, 1(1-11). MacArthur, R. H., and Wilson, E. O. (1963). An equilibrium theory of insular zoogeography. Evolution, 17, 373-387. MacArthur, R. H., and Wilson, E. O. (1967). The theory of island biogeography. Princeton, NJ: Princeton University Press. MacPhee, R. D. E., and Burney, D. A. (1991). Dating of Modified Femora of Extinct Dwarf Hippopotamus from Southern Madagascar - Implications for Constraining Human Colonization and Vertebrate Extinction Events. Journal of Archaeological Science, 18(6), 695-706. Maldidier, C. (2001). La décentralisation de la gestion des ressources renouvelable a Madagascar: les premiers enseignements sur les processus en cours et les méthodes d'intervention. Antananarivo: Office National pour l'Environnement.
756
Malthus, T. R. (1798). An essay on the principle of population, as it affects the future improvement of society. London: J. Johnson. Marglin, S. A. (Forthcoming). Economics and the social construction of the economy. In S. Gudeman and S. A. Marglin (Eds.), People's ecology, people's economy. Oxford: Oxford University Press. Maruyama, M. (1974). Paradigmatology, and its application to cross-disciplinary, cross-professional and cross-cultural communication. Cybernetika, 17, 135-156. Mayeur, N. (1785). Collection des voyages de Mayeur dans l'interieur de Madagascar: quatrieme voyage, voyage du pays d'ancove par le pays baizangounzangous. Antananarivo: Collection des ouvrages anciens concernant Madagascar. Mazoyer, M., and Roudart, L. (2001). Histoire des agricultures du monde: du neolithique a la crise contemporaine: Editions du Seuil, Paris. McClean, C. J., Lovett, J. C., Kuper, W., Hannah, L., Sommer, J. H., Barthlott, W., Termansen, M., Smith, G. E., Tokamine, S., and Taplin, J. R. D. (2005). African plant diversity and climate change. Annals of the Missouri Botanical Garden, 92(2), 139-152. McConnell, W. J. (2002). Misconstrued land use in Vohibazaha: participatory planning in the periphery of Madagascar's Mantadia National Park. Land Use Policy, 19(3), 217-230. McConnell, W. J., Sweeney, S. P., and Mulley, B. (2004). Physical and social access to land: spatio-temporal patterns of agricultural expansion in Madagascar. Agriculture Ecosystems and Environment, 101(2-3), 171-184. McNeely, J.A. and Scherr, S.R. (2003). Ecoagriculture: strategies to feed the world and save wild biodiversity. Washington D.C.: Island Press. Meadows, D., Randers, J., and Behrens, W. H. (1972). The limits to growth. New York: Universe Book. Messerli, P. (2003). Alternatives a la culture sur brûlis sur la falaise est de Madagascar: stratégies en vue d'une gestion plus durable des terres. Unpublished PhD Dissertation, Université de Berne, Suisse, Bern. Meyers, D. (2001). Makira forest project. Antananarivo: Ministere des Eaux et Forests, International Resources Group, Programme d'Appui a la Gestion de l'Environnement and USAID.
757
Meyers, D. and Berner, P. O. (2001). Carbon sequestration: Maroantsetra carbon project progress report. Antananarivo: Projet d'Appui a la Gestion de l'Environnement (PAGE). Miara-Mita. (1995). Evaluation par les bénéficiaires du Programme Environnemental 1 (PE1): ONE, DEF, ANGAP, ANAE, FTM, Domaines. Antananarivo: ONE. Michon, G. (1985). De l'homme de la forêt au paysan de l'arbre : agroforêts indonésiennes. Unpublished PhD Dissertation, Université des Sciences et Techniques du Languedoc, Montpellier, France. Minten, B. (2003). Compensation and cost of conservation payments for biodiversity: Cornell University, ILO Program. Minten, B. and Barrett, C.B. (2006). Agricultural technology, productivity, poverty and food security in Madagascar. Strategies and Analysis for Growth and Access Working Paper (January 2006). Antannarivo: USAID and SAGA project. Minten, B., Randrianarisoa, J-C. and Barrett, C.B. (2006). Productivity in Malagasy rice systems: wealth-differentiated constraints and priorities. Strategies and Analysis for Growth and Access Working Paper (June 2006). Antannarivo: USAID and SAGA project. MIRAY. (2004). MIRAY final technical report. Antananarivo: MIRAY program for ecoregion-based conservation and development, Conservation International, PACT, World Wildlife Fund. Montagne, P. and Rakotondrainibe, P. (2006). Les politiques forestières et environnementales: brève revue historique. In P. Montagne and Z. Razanamahoro and A. Cooke (Eds.), Le transfert de gestion a Madagascar; dix ans d'efforts. Antananarivo: Centre International de Recherche Agronomique pour le Développement (CIRAD), Ecole Supérieure des Sciences Agronomiques de Antananarivo, FOFIFA and RESOLVE Conseil. Moor, P. (1998a). Les champs d'activités paysans et l'organisation du travail. In P. T.T. BEMA (Ed.), Les stratégies endogènes et la gestion des ressources naturelles dans la région de Beforona: les résultats des recherches pluridisciplinaires de la phase 1995-1998 (Vol. 8, pp. 28-42). Antananarivo: Projet Terre-Tany / BEMA. Moor, P. (1998b). Les stratégies endogènes dans la région de Beforona. In P. T.-T. BEMA (Ed.), Les stratégies endogènes et la gestion des ressources naturelles dans la région de Beforona: les résultats des recherches
758
pluridisciplinaires de la phase 1995-1998 (Vol. 8, pp. 53-66). Antananarivo: Projet Terre-Tany / BEMA. Moor, P., and Rasolofomanana, L. (1998). Economie des ménages paysans dans la région de Beforona. In P. T.-T. BEMA (Ed.), Les stratégies endogènes et la gestion des ressources naturelles dans la région de Beforona: les résultats des recherches pluridisciplinaires de la phase 1995-1998 (Vol. 8, pp. 42-53). Antananarivo: Projet Terre-Tany / BEMA. Moran, E. F. (1981). Developing the Amazon. Indiana University Press, Bloomington. Morin, E. (1973). La methode. 1. La nature de la nature. Paris: Points Essais. Morin, E. (1984). La métamorphose de Plozevet : commune de France. Paris: Le Livre de Poche. Morin, E. (1990). Introduction a la pensee complexe. Paris: Points Essais. Moser, C. M., and Barrett, C. B. (2003a). The complex dynamics of smallholder technology adoption: the case of SRI in Madagascar. Working Paper of the Department of Applied Economics and Management, Cornell University. Moser, C. M., and Barrett, C. B. (2003b). The disappointing adoption dynamics of a yield-increasing, low external-input technology: the case of SRI in Madagascar. Agricultural Systems, 76(3), 1085-1100. Moser, C. M., and Barrett, C. B. (2006). Agricultural Technology, Productivity, and Poverty in Madagascar. Saga Working Paper. Antananarivo: USAID. Moser, C. M., Barrett, C.B. and Minten, B. (2006). Spatial integration at multiple scales: rice markets in Madagascar. Saga Working Paper. Antananarivo: USAID. Moynot, G. (2005). Mise en cohérence des indicateurs des actions des bailleurs de fonds avec ceux du cadre logique du programme environnemental 3. Antananarivo: Secrétariat Multi Bailleurs. Murdiyaso, D. a. (2005). Greenhouse gas fluxes in slash-and-burn and alternative land use practices in Sumatra, Indonesia. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 64-83). New York: Columbia University Press. Murdoch, J. (1997). Inhuman/Nonhuman/Huma. Environment and Planning D: Society and Space, 15, 731-756. Murton, J. (1997). Coping with more people. Unpublished PhD Dissertation, University of Cambridge.
759
Muttenzer, F. (2006). Déforestation et droit coutumier a Madagascar: l'historicité d'une politique foncière. Unpublished PhD Dissertaion, Universite de Genève, Genève. Myers, N. (1984). The primary source: tropical forests and our future. New York: W.W. Norton. Myers, N. (1991a). Tropical forests: present status and future outlook. Climatic Change, 19, 3-32. Myers, N. (1991b). Tropical forests: the main deforestation fronts. Environ. Conserv., 20(1), 9-16. Myers, N. (1998). Lifting the veil of perverse subsidies. Nature, 392, 327-328. Myers N, Mittermeier R. A., Mittermeier C. G., da Fonseca G. A. B., and Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403, 853858. Nambena, J. (2004). Analyse de la subsistance paysanne dans un système de production en crise et identification participative de stratégies durables d'adaptation: cas de Beforona, versant oriental de Madagascar. Unpublished PhD Dissertation, Université de Heidelberg, Heidelberg. Nambena, S. (2007). Régionalisation de l'utilisation des ressources naturelles à partir du cas de Beforona. Unpublished PhD Dissertation, Universite d'Antananarivo, Antananarivo. Neuman, R. P. (1998). Imposing wilderness: struggles over livelihood and nature preservation in Africa. Berkeley, Los Angeles and London: University of California Press. Nietschmann, B. (1973). Between land and water: the subsistence ecology of the Miskito Indians, eastern Nicaragua. New York: Seminar Press. Nietzsche, F. (2003). The genealogy of morals. Mineola, N.Y.: Dover Publications. Nye, P. H., and Greenland, D. J. (1960). The soil under shifting cultivation. Farham Royal, Bucks, England: Commonwealth Agriculture Bureaux. Oakeshott, M. (1962). Rationalism in politics and other essays. New York: Basic Books Pub Co. Oates, J.F. (1999). Myths and realities in the rain forest: how conservation strategies are failing in West Africa. Berkeley: University of California Press. Odum, E. (1963). Ecology. New York: Holt,Rinehart and Winston.
760
OECD (2003) Subsidies to Agriculture: Why? OECD in Washington (46) Office Statistique et Informatique pour la Programmation du Développement (OSIPD). (1995). Evaluation par les bénéficiaires du PE1 dans le Faritany de Fianarantsoa. Antananarivo: ONE. Office Statistique et Informatique pour la Programmation du Développement (OSIPD). (2000). Evaluation par les bénéficiaires du Plan d'Action Environnemental - phase II (PE2). Rapport détaillé, version finale. Antananarivo: ONE. Olson, S. H. (1984). The robe of the ancestors : forests in the history of Madagascar. Journal of Forest History, 28(4), 174-186. Ophuls, W. (1977). Ecology and the politics of scarcity. San-Francisco, California: W.H. Freeman. Osborn, F. (1948). Our plundered planet (1st ed.). Boston: Little Brown. OXFAM (no date). Cultivating poverty: the impact of US cotton subsidies on Africa. OXFAM Briefing Paper (30) Oxford University Press (1998). The new Oxford dictionary of English. Oxford: Oxford University Press. Palm, C. A., Swift, M. J., and Woomer, P. L. (1996). Soil biological dynamics in slash-and-burn agriculture. Agriculture, Ecosystems and Environment. 1996. 58 (1). 61-74. Palm, C. A., Vosti, S. A., Sanchez, P. A., and Ericksen, P. J. (2005a). Slash-and-burn agriculture: the search for alternatives. New York: Columbia University Press. Palm, C. A., van Noordwijk, M., Woomer, P. L., Alegre, J. C., Arevalo, L., Castilla, C. E., Cordeiro, D. G., Hairiah, K., Kotto-Same, J., Moukam, A., Parton, W. J., Ricse, A., Rodrigues, V., and Sitompul, S. M. (2005b). Carbon losses and sequestration after land use change in the humid tropics. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-andburn agriculture: the search for alternatives (pp. 41-63). New York: Columbia University Press. Partohardjono, S., Pasaribu, D., and Fagi, A. M. (2005). The forest margins of Sumatra, Indonesia. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 291-305): Columbia University Press, New York.
761
Paternostro, S., Razafindravonona, J., and Stifel, D. (2001). Changes in poverty in Madagascar: 1993-1999: INSTAT, Cornell University, World Bank and USAID. Peet, R., and Watts, M. (2004). Liberation ecologies : environment, development, social movements (2nd ed.). London and New York: Routledge. Peluso, N. L. (1992). Rich forests, poor people : resource control and resistance in Java. Berkeley: University of California Press. Peluso, N. L., and Watts, M. (2001). Violent environment. In N. L. Peluso and M. Watts (Eds.), Violent environment (pp. 3-38): Cornell University Press. Perrier de la Bâthie, H. (1921). La végétation malgache. Annales du Museum Colonial, 9, 1-266. Perrier de la Bâthie, H. (1936). Biogéographie des plantes de Madagascar. Paris: Société d'éditions géographiques maritimes et coloniales. Peters, J. (1998). Transforming the integrated conservation and development project (ICDP) approach: observations from the Ranomafana National Park Project, Madagascar. Journal of Agricultural and Environmental Ethics, 11(1), 17-47. Peters, J. (1999). Understanding conflicts between people and parks at Ranomafana, Madagascar. Agriculture and Human Values, 16(1), 65-74. Peters, W. J., and Neuenschwander, L. F. (1988). Slash and burn. Farming in the Third World forest. Moscow, Idaho: University of Idaho Press. Pfund, J-L. (2000). Cultures sur brûlis et gestion des ressources naturelles: évolution et perspectives de trois terroirs ruraux du versant est de Madagascar. PhD Dissertation. Ecole Polytechnique Fédérale de Zurich: Zurich. Phillips, A. (2002). Management Guidelines for IUCN Category V Protected Areas Protected Landscapes/Seascapes (Best Practice Protected Area Guidelines Series No. 9). Gland: IUCN, World Commission on Protected Areas (WCPA). Plodprasop, S. a. (2005). Northern Thailand: changing smallholders land use patterns. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 355-386). New York: Columbia University Press. Polanyi, K. (1944). The Great Transformation. Boston: Beacon Press.
762
Pollini, J. (1993). Analyse des réalités agraires de la Province de Bukeye, Burundi. Unpublished Professional Master, Institut National Agronomique Paris Grignon, Paris. Pollini, J. (1994). L’intérêt de l’agroforesterie pour la conservation et la restauration de la fertilité des sols au Burundi. PRPE project, Institut des Science Agronomiques du Burundi, Bujumbura. Pollini, J. (1999). Analyse d'un térroir villageois : Ban-Tintok, Province de LuangNamtha, Laos (Report for a consultancy mission). Vientiane: European Community-Integrated Rural Development Project of Luang-Namtha. Ponting, C. (1991). A green history of the world : the environment and the collapse of great civilizations. New York: St. Martin’s Press. Popper, K. (1959). The logic of scientific discovery. New York: Basic Books. Popper, K. (1963). Conjectures and refutations : the growth of scientific knowledge. London: Routledge and Kegan Paul. Popper, K. (1966). The open society and its ennemies. London: Routledge. Popper, K. (1981). La quête inachevée. Paris: Calman Levy. Programme de Transition Ecoregional (PTE). (2004a). Final report January 2004 September 2004. Antananarivo: Chemonics International and PACT Madagascar. Programme de Transition Ecoregional (PTE). (2004b). Interim progress report January 2004 - May 2004. Antananarivo: Chemonics International and PACT Madagascar. Programme POLFOR, and Equipe MIRAY. (1998). Rapport de l'atelier aménagement forestier et gestion participative du 14, 15 et 16 octobre 1998 a Antananarivo, Madagascar. Antananarivo. Project design document form for afforestattion and reforestation project activities (CDM-AR-PDD) version 02. (No date). The Vohidrazana-Mantadia Corridor Restoration and Conservation Carbon Project (VMCRCC) - draft. UNFCCC/CCNUCC - CDM - Executive Board. Purnomosidhi, P., Hairiah, K., Rahayu, S., and van Noordwiik, M. (2005). Smallholder options for reclaiming and using Imperata cylindrica L. (alang alang) grasslands in Indonesia. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 248-264). New York: Columbia University Press.
763
Pyne, S. J. (2001). Fires: a brief history. London: The British Museum Press. Rabetsimamanga, R. S. (2005). Contribution à l'étude de la mise en place d'une aire protégée de catégorie V UICN à Madagascar - Enjeux, contrainte, perspective- Cas du pays Zafimaniry- district d'Ambositra- région de l'Amoron' Imania.” Unpublished Master Thesis, University of Antananarivo, Antananarivo. Radimilahy, M. d. C. (1998). Mahilaka : an archaeological investigation of an early town in northwestern Madagascar. Uppsala: Department of Archaeology and Ancient History. Raharilantsoa, A. Y. (2002). Diversification et commercialisation de la production agricole dans la region de Beforona: un avenir prometteur. Université d'Antananarivo, Antananarivo. Rahnema, M. (1992). Participation. In W. Sachs (Ed.), The development dictionary: a guide to knowledge as power (pp. 116-131). London: Zed Books. Raintree, J. B., and Warner, K. (1986). Agroforestry pathways for the intensification of shifting cultivation. Agroforestry Systems, 4(1), 39-54. Raison, J.-P. (1984). Les hautes terres de Madagascar et leurs confins occidentaux. Paris: Karthala. Rakotoarijaona Razafimbelo, N. M. (2005). Intégration des paysans dans un projet de développement dans le fokontany d'ambodilaingo, commune rurale de Beforona, district de Moramanga. Unpublished Master thesis, Département d’Histoire. Antananarivo: Université d'Antananarivo. Rakotondravony, D., and Goodman, S. M. (1998). Inventaire biologique: forêt d'Andranomay, Anjozorobe. Antananarivo: Ministère de la Recherche Scientifique. Ralinoro, A. (1998). La dynamique des systèmes de commercialisation. In Projet Terre-Tany / BEMA (Ed.), Les stratégies endogènes et la gestion des ressources naturelles dans la région de Beforona: les résultats des recherches pluridisciplinaires de la phase 1995-1998. Antananarivo: Projet Terre-Tany / BEMA. Ramakrishnan, P. S. (1992). Shifting agriculture and sustainable development: an interdisciplinary study from north-eastern India. Carnforth: Parthenon Publishing Group and UNESCO. Ramanantsoavina, G. (No date). Histoire de la politique forestière à Madagascar. Antananarivo.
764
Randrianarisoa, J. D. D. (1998). Le rôle de la RN2 sur la gestion des ressources naturelles dans la région de Beforona. In Projet Terre-Tany / BEMA (Ed.), Stratégies endogènes et gestion des ressources naturelles dans la région de Beforona: résultats des recherches pluridisciplinaires de la phase 1995 a 1998 (Vol. Numero 8, pp. 121-129). Antananarivo: Projet TerreTany/BEMA, FOFIFA and GDE/GIUB. Randrianasolo, J. (2000). Rapport de synthèse du mandat capitalisation des expériences en gestion contractualisée des forêts a Madagascar. Antananarivo: Intercoopération Délégation Madagascar Projet POLFOR. Rangan, H. (1997). Indian environmentalism and the question of the state: problems and prospects for sustainable development. Environment and Planning A, 29, 2129-2143. Rangan, H. (2000). Of myths and movements: rewriting Chipko into Himalayan history. London: Verso. Rapport d'atelier sur la gestion décentralisée des ressources ligneuses contribuant à l'approvisionnement énergétique de la ville de Mahajanga. (2004). Antananarivo. Rarivoarivelomanana, J. (2001). Le stockage du carbone et ses avantages à travers le protocole de Kyoto: cas des forêts de l'est de Madagascar. Antananarivo: Ministère de l'Environnement. Ratner, J. (Ed.). (1939). Intelligence in the modern world: John Dewey's philosophy. New York: The Modern Library. Ratsimijanona, H. L. (2006). Politiques de répression des feux de brûlis et paysans: le cas du fokontany d'Ambodilaingo, commune rurale de Beforona (cote centre est malgache). Unpublished Master Thesis, Université de Antananarivo, Antananarivo. Rausser, G. (1980). Economics of soil conservation from the farmers' perspective: discussion. American Journal of Agriculture Economics, 62, 1093-1094. Ravelomandeha, F. M.-P. (2003). Essai de lutte naturelle contre les insectes ravageurs des cultures maraîchères a Beforona. Unpublished Mémoire de fin d'études pour l'obtention du diplôme d'ingénieur agronome, Université d'Antananarivo, Antananarivo. Razafiarison, S. (2003). La construction sociale a l'épreuve du temps. In S. Aubert and S. Razafiarison and A. Bertrand (Eds.), Déforestation et systèmes agraires a Madagascar: les dynamiques des tavy sur la cote orientale (pp. 89-106). Antananarivo: CIRAD-CITE-FOFIFA.
765
Razanaka, S., Grouzis, M., Milleville, P., Moizo, B., and Aubry, C. (Eds.). (2001). Sociétés paysannes, transitions agraires et dynamiques écologiques dans le sud-ouest de Madagascar. Antananarivo: CNRE-IRD. Razakanirina, D. (1986). Contribution à l'étude de la végétation d'Andasibe Périnet. Phd. Dissertation. University of Madagascar, Antananarivo. Redman, C. L. (1999). Human impact on ancient environment. Tucson: The University of Arizona Press. Republika Malagasy. (1972). Rapport national de la république malgache pour la conference des nations unies sur l'environnement. Antananarivo. Republikan'I Madagasikara. (2003). Document de stratégie pour la réduction de la pauvreté. Antananarivo: Republikan'I Madagasikara. Republique de Madagascar, Ministère de l'Economie, des Finances et du Budget. (2004). Enquête périodique après des ménages 2004: rapport principal. Antananarivo: MEFB, INSTAT, UNDP, Union Européenne. RESOLVE - PCP - IRD. (2004a). Evaluation et perspectives des Transferts de Gestion des Ressources Naturelles dans le cadre du Programme Environnemental 3. Rapport Final phase 1 - Etat des lieux. Antananarivo: Ministère de l'Environnement. RESOLVE - PCP - IRD. (2004b). Evaluation et perspectives des Transferts de Gestion des Ressources Naturelles dans le cadre du Programme Environnemental 3. Rapport Final Phase 2: analyse qualitative. Antananarivo: Ministère de l'Environnement. RESOLVE - PCP - IRD. (2005). Evaluation et perspectives des Transferts de Gestion des Ressources Naturelles dans le cadre du Programme Environnemental 3. Rapport Final de synthèse. Antananarivo: Ministère de l'Environnement. Restore the Earth Project. Catalyzing rainforest restoration in Madagascar: AndasibeMantadia corridor. Retrieved, from the World Wide Web: ww.restoreearth.org/restearth.samproj01.html Revelle, R., and Suess, H. E. (1957). Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric CO2 during the past decades. Tellus, 9, 18-27. Ribot,
J.
C. (2002). Democratic decentralization of natural resources: institutionalizing popular participation. Washington DC: World Resources Institute.
766
Rist, G. (1997). The history of development: from Western origins to global faith. London and New York: Zed Book. Rivolalao, T. (2006). La paupérisation paysanne et les rapports sociaux dans le fokontany d'Ambodilaingo, commune rurale de Beforona, District de Moramanga. Master thesis. Département des sciences sociales du Développement. Antananarivo: Université Antananarivo. Robbins, P. (1998). Authority and Environment: institutional landscapes in Rajasthan, India. Annals of the Association of American Geographers, 88(3), 410-435. Robbins, P. (2004). Political ecology : a critical introduction. Malden, MA: Blackwell Pub. Roberts, J. T., and Hite, A. (Eds.). (2000). From modernization to globalization: perspectives on development and social change. Oxford: Blackwell Publisher Ltd. Roe, E. (1991). Development narratives, or making the best of blueprint development. World Development, 19(4), 287-300. Rollings, N. M., and Brunckhorst, D. J. (1999). Linking ecological and social functions of landscapes: II. Scale and modeling of spatial influence. Natural Areas Journal, 19(1), 65-72. Rose, D. (1991). Elite Discourse of the Market and Narrative Ethnography. Anthropological Quarterly, 64(3), 109-125. Rossi, G. (2003). L'ingérence écologique: environnement et développement rural du nord au sud. Paris: CNRS Editions. Rossi, G., and Donque, G. (1982). Importance, causes et consequences de la crise morphoclimatique actuelle a Madagascar. Bulletin de l'Academie Malgache, 56(1-2), 99-104. Rostow, W. W. (1960). The stages of economic growth: a non-communist manifesto. Cambridge: Cambridge University Press. Rousseau, J.-J. (1762). Du contract social, ou principes du droit politique. Amsterdam: Chez M. M. Rey. Sachs, J. D. (2005). Foreword. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. ix-x). New York: Columbia University Press. Sachs, W. (1992). The development dictionary. London: Zed Book.
767
Saint Martin, K. (2001). Making space for community resources management in fisheries. Annals of the Association of American Geographers, 91(1), 122142. Sanchez, P. A. (1996). Special issue: alternatives to slash-and-burn agriculture. Introduction. Agriculture, Ecosystems and Environment. 1996. 58 (1). 1-2. Sanchez, P. A., and Hailu, M. (1996). Special issue: alternatives to slash-and-burn agriculture. Agriculture, Ecosystems and Environment. 1996. 58 (1). vi + 1-86. Sanchez, P. A., Simmons, A. J., and Place, F. J. (1998). More people, more trees: the future of trees is on farm in Africa. Agronomy abstract, ASA, Madison, Wisconsin. Sanchez, P. A., Palm, C. A., Vosti, S. A., Tomich, T. P., and Kasyoki, J. (2005). Alternatives to slash-and-burn: challenge and approaches of an international consortium. In C. A. Palm, S. A. Vosti, P. A. and Sanchez P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 3-37). New York: Columbia University Press. Schmink, M. and Wood, C. H., (1984). Frontier expansion in Amazonia. Gainesville: University of Florida Press. Schmink, M., and Wood, C. H. (1992). Contested frontiers in Amazonia. New York: Columbia University Press. Schoener, T. W. (1989). The ecological niche. In J. M. Cherrett (Ed.), Ecological concepts. Oxford: Blackwell. Scoones, I. (1996). Range management science and policy: politics, polemics and pasture in Southern Africa. In M. Leach and R. Mearns (Eds.), The lie of the land (pp. 34-53). Portsmouth, N.H.: Heinemann. Scoones, I. (1999). New ecology and the social sciences: what prospects for a fruitful engagement? Annual Review of Anthropology, 28, 479-507. Scott, J. (1976). The moral economy of the peasants: rebellion and subsistence in Southeast Asia. New Haven: Yale University Press. Scott, J. (1985). Weapons of the weak: everyday forms of peasant resistence. New Haven: Yale University Press. Scott, J. (1998). Seeing like a state: how certain schemes to improve the human condition have failed. New-Haven and London: Yale University Press. Serres, M. (1990). Le contrat naturel. Paris: Editions F. Bourin.
768
Shiva, V. (1991). The violence of green revolution. Penang, Malaysia: Third World Network. Shugart, H. H. (1984). A theory of forest dynamics : the ecological implications of forest succession models. New York: Springer-Verlag. Shugart, H. H. (1998). Terrestrial ecosystems in changing environments. Cambridge: Cambridge University Press. Shyamsundar, P., and Kramer, R. (1997). Biodiversity conservation - At what cost? A study of households in the vicinity of Madagascar's Mantadia National Park. Ambio, 26(3), 180-184. Simon, J. (1981). The ultimate resource. Princeton: Princeton University Press. Simon, J. (1996). The ultimate resource 2. Princeton, NJ: Princeton University Press. Simsik, M. J. (2002). The political ecology of biodiversity conservation on the Malagasy highlands. Geojournal, 58(233-242), 233-242. Sivaramakrishnan, K. (2000). State sciences and development histories: Encoding local forestry knowledge in Bengal. Development and Change, 31(1), 6189. Sivaramakrishnan, K. S. (1996). The politics of fire and forest regeneration in colonial Bengal. Environment and History, 2, 145-194. Soule, M. E., and Lease, G. (1995). Reinventing nature? responses to postmodern deconstruction. Washington DC: Island Press. Stonich, S. C., and Vandergeest, P. (2001). Violence, environment, and industrial shrimp farming. In N. L. Peluso and M. Watts (Eds.), Violent environment (pp. 261-287): Cornell University Press. Stoop, W. A., Uphoff, N., and Kassam, A. (2002). A review of agricultural research issues raised by the system of rice intensification (SRI) from Madagascar: opportunities for improving farming systems for resource-poor farmers. Agricultural Systems, 71, 249–274. Stott, P. A., and Sullivan, S. (2000). Political ecology : science, myth and power. London and New York: Oxford University Press. Styger, E. (2004). Fireless alternatives to slash-and-burn agriculture (tavy) in the rainforest region of Madagascar. Unpublished PhD Dissertation, Cornell University, Ithaca.
769
Styger, E., Rakotondramasy, H. M., Pfeffer, M. J., Fernandes, E. C. M., and Bates, D. M. (2007). Influence of slash-and-burn farming practices on fallow succession and land degradation in the rainforest region of Madagascar. Agriculture Ecosystems and Environment, 119(3-4), 257-269. Sunderlin, W. D., Angelsen, A., Belcher, B., Burgers, P., Nasi, R., Santoso, L., and Wiunder, S. (2005). Livelihoods, forests and conservation in developing countries: an overview. World Development, 33(9), 1383-1402. Suraswadi, P., Thomas, D. E., Pragtong, K., Preechapanya, P., and Weyerhaeuser, H. (2005). Northern Thailand: changing smallholder land use patterns. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-andburn agriculture: the search for alternatives (pp. 355-386). New York: Columbia University Press. Sussman, R. W., Green, G. M., Sussman, L. K., and Behrens, C. A. (1994). Satellite imagery, human ecology, anthropology, and deforestation in Madagascar. Human Ecology 22(3), 333-354. Swift, J. (1996). Desertification: narratives, winners and losers. In M. Leach and R. Mearns (Eds.), The lie of the land (pp. 73-90). Portsmouth, N.H.: Heinemann. Tansley, A. G. (1935). The use and abuse of vegetational consepts and terms. Ecology, 16, 284-307. Terborgh, J. (1999). Requiem for nature. Washington, DC: Island Press and Shearwater Book. Terrasson, F. (2002). La peur de la nature. Paris: Sang de la Terre. Thomas Jr, W. L. (Ed.). (1956). Man's role in changing the face of the earth: an international symposium under the co-chairmanship of Carl O. Sauer, Marston Bates, Lewis Mumford. Chicago: The University of Chicago Press. Thompson, M., and Warburton, M. (1985a). Uncertainty on a Himalayan Scale. Mountain Research and Development, 5(2), 115-135. Thompson, M., and Warburton, M. (1985b). Knowing Where to Hit It - a ConceptualFramework for the Sustainable Development of the Himalaya. Mountain Research and Development, 5(3), 203-220. Tiffen, M., Mortimore, M., and Gichuki, F. (1994). More people, less erosion: environmental recovery in Kenya. New-York: Wiley.
770
Tinker, P. B., Ingram, J. S. I., and Struwe, S. (1996). Effects of slash-and-burn agriculture and deforestation on climate change. Paper presented at the Agriculture, Ecosystems and Environment. 1996. 58 (1). 13-22. Todaro, M. P., and Smith, S. C. (Eds.). (2003). Economic development: Pearson Addison Wesley. Tomich, T. P., Cattanao, A., Chater, S., Geist, H. J., Gockowski, J. a., Kaimovitz, D., Lambin, E. F., Lewis, J., Ndoye, O., Palm, C. A., Stolle, F., Sunderlin, W. D., Valentim, J. F., van Noordwijk, M., and Vosti, S. A. (2005). Balancing agricultural development and environmental objectives: assessing tradeoffs in the humid tropics. In C. A. Palm and S. A. Vosti and P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 415-440). New York: Columbia University Press. Turner II, B. L., Clark, W. C., Kates, R. W., Richards, J. F., Mathews, J. T., and Meyer, W. B. (Eds.). (1990). The earth as transformed by human action: global and regional changes in the biosphere over the past 300 years. Cambridge: Cambridge University Press. UCFB. (2006). Projet de Conservation et de Restauration du Corridor forestier Ankeniheny-Mantadia-Zahamena. In C. C. PEIII (Ed.). Antananarivo: Ministère de l'Environnement des Eaux et Forêts. United Nations Conference on Environment and Development (UNCED), (1992). Agenda 21. Rio: United Nations. USAID. (1998). The USAID FY 1998 congressional presentation. Retrieved, from the World Wide Web: http://www.info.usaid.gov/pubs/cp98/afr/countries/mg.htm Valentim, J. F., and Vosti, S. A. (2005). The Western Brazilian Amazon. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 265-290). New York: Columbia University Press. Van Schaik, C, Terborgh, J. and Rao, M. (2002). Making parks work: strategies for preserving tropical nature. Washington, D.C.: Island Press. Vernadski, V. I. (1986). The Biosphere. Oracle, Arizona: Synergetic Press. Vosti, S. A., and Witcover, J. (1996). Slash-and-burn agriculture - household perspectives. Paper presented at the Agriculture, Ecosystems and Environment. 1996. 58 (1). 23-38. Vosti, S. A., Gockowski, J., and Tomich, T. P. (2005). Land use systems at the margins of tropical moist forest: addressing smallholder concerns in
771
Cameroon, Indonesia, and Brazil. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 387-414). New York: Columbia University Press. Wallerstein, I. (1974). The modern world system I: capitalism, agriculture and the origins of the European world-economy in the sixteenth century. New York: Academic Press. Walsh, A. (2005). The obvious aspects of ecological underprivilege in Amkerana, Northern Madagascar. American Anthropologist, 107(4), 654-665. Watts, M. (1983). Silent violence : food, famine, and peasantry in northern Nigeria. Berkeley: University of California Press. Weatherall, M. (1968). Scientific method. London: English University Press. Weber, J. (1994). L'occupation humaine des aires protégées a Madagascar: diagnostic et éléments pour une gestion viable. Paper presented at the Colloque international occupation humaine des aires protégées, Mahajunga, 21-25 novembre 1994, Antananarivo. Wells, N. A. (2003). Some hypotheses on the Mesozoic and Cenozoic paleoenvironmental history of Madagascar. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 16-34). Chicago: University of Chicago Press. Wells, N. a. A., Benjamin R. (1997). Extreme gully erosion in Madagascar and its natural and anthropogenic causes. In S. M. Goodman and B. D. Patterson (Eds.), Natural change and human impact in Madagascar (pp. 45-74). Washington D.C.: Smithsonian Institution Press. West, P., Igoe, J., and Brockington, D. (2006). Parks and people: the social impact of protected areas. Annual Review of Anthropology, 35, 251-277. West, P. C., and Brechin, S. R. (Eds.). (1991). Resident people and national parks: social dilemmas and strategies in international conservation. Tucson: The University of Arizona Press. Whatmore, S., and Thorne, L. (1998). Wilderness: reconfiguring the geographies of Wildlife. Transactions of the Institute of British Geographers, 23, 435-454. White, D., Arca, M., Alegre, J. C., Yanggen, D., Labarta, R., Weber, J. C., SoteloMontes, C., and Vidaurre, H. (2005). The forest margins of Cameroon. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slashand-burn agriculture: the search for alternatives (pp. 332-354). New York: Columbia University Press.
772
Whitehead, N. (1998). Ecological history and historical ecology: diachronic modeling versus historical explanation. In W. Balee (Ed.), Advances in historical ecology (pp. 30-41). New-York: Columbia University Press. Whiten, A., Goodall, W.C., McGrew, T., Nishida, V., Reynolds, Y., Sugiyama, C. E. G., Tutin, R. W., Wrandham, R. W., and Boesch, C. (1999). Cultures in chimpanzees. Nature, 399, 682-685. Whitmore, T. C., and Sayer, J. A. (Eds.). (1992). Tropical deforestation and species extinction. London: Chapman and Hall. Whittaker, R. H. (1951). A criticism of the plant association and climatic climax concepts. Northwest Science, 25, 17-31. Whittaker, R. H. (1956). Vegetation of the Great Smoky Mountains. Ecological Monographs, 26, 1-80. Whittaker, R. H. (1965). Dominance and diversity in land plant communities. Science, 147, 250-260. Whittaker, R. H. (1975). Communities and ecosystems. New-York: Macmillan. Wibawa, G., Hendratno, S., and van Noordwiik, M. (2005). Permanent smallholder rubber agroforestry systems in Sumatra. In C. A. Palm, S. A. Vosti, P. A. Sanchez and P. J. Ericksen (Eds.), Slash-and-burn agriculture: the search for alternatives (pp. 222-233). New York: Columbia University Press. Williams, M. (2006). Deforesting the earth: from prehistory to global crisis, an abridgment. Chicago and London: The University of Chicago Press. Wilshusen, P.R., Brechin, S.R., Fortwangler, C.L. and West, P.C. (1999). Reinventing a square wheel: critique of a resurgent “protection paradigm” in international biodiversity conservation. Society and Natural Resources, 15, 17-40. Wilson, E. O. (Ed.). (1988). The current state of biological diversity. Washington D.C.: National Academy Press. Wilson, R. J., Gutierrez, D., Gutierrez, J., Martinez, D., Agudo, R., and Monserrat, V. J. (2005). Changes to the elevational limits and extent of species ranges associated with climate change. Ecology Letters, 8(12), 1346-1346. Wing, S. L., Harrington, G. J., Smith, F. A., Bloch, J. I., Boyer, D. M., and Freeman, K. H. (2005). Transient floral change and rapid global warming at the Paleocene-Eocene boundary. Science, 310(5750), 993-996.
773
Wolf, E. (1972). Ownership and Political Ecology. Anthropological Quarterly, 45(3), 201-205. World Bank (no date). Biocarbon Fund. Washington DC: The World Bank. World Bank. (1988). Plan d'Action Environnemental. Antananarivo: Gouvernement de Madagascar. World Bank. (1990). Rapport d'Evaluation, Republique Democratique de Madagascar, Programme Environnemental. Washington DC: The World Bank. World Bank. (1991). The forest sector (World Bank Policy Paper). Washington DC: World Bank. World Bank. (1996). Staff Appraisal Report Madagascar Second Environment Program. Washington DC: The World Bank. World Bank. (2001). Project appraisal document on a proposed credit in the amount of SDR 69.2 million (USD 90.05 million equivalent) to the republic of Madagascar for a rural development support project. Washington DC: World Bank Africa Regional Office. World Bank. (2003). Madagascar Environment Program Phase 2, Draft ICR, Mission to Madagascar July 8-29, 2003. Washington DC: The World Bank. World Bank. (2004). Madagascar third environment program support project: project appraisal document. Washington DC: The World Bank. World Bank. (2005). Crédit IDA H-0870 et Don GEF 053226: Troisième Programme Environnemental de Madagascar. Première Mission de Supervision (10 Mai -24 Juin, 2005) Aide-mémoire. Antananarivo: The World Bank. World Bank. (2006a). Troisième Programme Environnemental de Madagascar, Supervision Conjointe (10 - 30 Janvier, 2006), Aide-mémoire. Antananarivo: The World Bank. World Bank. (2006b). Troisième Programme Environnemental de Madagascar, Troisième Mission de Supervision (11 Septembre - 16 Octobre 2006) Aide-mémoire. Antananarivo: The World Bank. World Bank. (2007). SAPM et sauvegardes: coût de création et gaps de financement. Note explicative des estimations (draft, BM-EESD, Janvier 2007). Antananarivo: The World Bank. World Commission on Environment and Development. (1987). Our common future. Oxford ; New York: Oxford University Press.
774
World Conservation Monitoring Center. (1994). Biodiversity Data Source Book. World Conservation Press. World Resources Institute. (1989). Natural Endowments: financing resource conservation and development. Washington, DC: World Resources Institute. Wright, H. T., and Rakotoarisoa, J. A. (2003). The rise of Malagasy societies: new developments in the archeology of Madagascar. In S. M. Goodman and J. P. Benstead (Eds.), The natural history of Madagascar (pp. 112-119). Chicago: University of Chicago Press. Wright, H. T., Verin, P., Ramilisonina, Burney, D. A., Burney, L. P., and Matsumoto, K. (1996). The evolution of settlement systems in the Bay of Boeny and the Mahavavy River valley. Azania, 31, 37-73. Wright, P., and Andriamihaja, B. (2002). Making a rain forest National park work in Madagascar: Ranomafana National Park and its long-term research commitment. In J. T. Terborgh and C. V. Shaik and L. Davenport and M. Rao (Eds.), Making park work: strategies for preserving tropical nature. Island Press. Wunder, S. (2001). Poverty alleviation and tropical forests : what scope for synergies. World Development, 29(11), 1817-1833. Wunder, S. (2007). The efficiency of payments for environmental services in tropical conservation. Conservation Biology, 21(1), 48-58.
Zeller, M., Lapenu, C., Minten, B., Ralison, E., Randrianaivo, D. and Randrianarisoa, C. (2000). Pathways of rural development in Madagascar: en empirical investigation of the critical triangle of environmental sustainability, economic growth, and poverty alleviation. FCND Discussion Paper No 82. Washington DC: International Food Policy Research Institute Zerner, C. (1999). Toward a broader vision of justice and nature conservation. In C. Zerner (Ed.), People, plants and justice: the politics of nature conservation. New-York: Columbia University Press. Zimmerer, K. S. (1991). Wetland production and smallholder persistence: agricultural change in a highland Peruvian Amazon. Annals of the Association of American Geographers, 81(3), 443-463. Zimmerer, K. S. (2000). The reworking of conservation geographies: nonequilibrium landscapes and nature-society hybrids. Annals of the Association of American Geographers, 90(2), 356-369.
775
Zimmerer, K. S. (2006). Globalization and new geographies of conservation. Chicago: University of Chicago Press. Zimmerer, K. S., and Bassett, T. J. (2003a). Political ecology : an integrative approach to geography and environment-development studies. New York: The Guilford Press. Zimmerer, K. S., and Bassett, T. J. (2003b). Approaching political ecology: society, nature, and scale in human-environment studies. In K. S. Zimmerer and T. J. Bassett (Eds.), Political ecology : an integrative approach to geography and environment-development studies (pp. 1-28). New York: The Guilford Press. Zimmerer, K. S., and Young, K. R. (1998). Nature's geography: new lessons for conservation in developing countries. Madison: University of Wisconsin Press. Zurbuchen, J. (No date). Demographiche entwicklung und landnutzungsrechte am ostabhang. Institut de Géographie, Université de Berne, Berne.
776
