I use the term 'sociological variant' to highlight the fact that IR theory has long been interested in/engaged with an attendance to practice, however more recently this engagement has drawn specifically from the sociological theory of Pierre Bourdieu.
For example, see: Richard Ashley, (1984) "The Poverty of Neorealism." International Organization 38(2): 225-286.
See: Kenneth Waltz, Theory of International Politics, 1979.
The phenomenon of some states wishing to remain neutral while simultaneously willing to modify their sovereignty by joining the European Community (Carlsnaes, 1992, 247).
At this juncture I want to point out that my border drawing of Doty and Wight as the 'last portion' of the debate is largely related to space constraints and to enable this article to move quickly into a discussion of the Bourdieusian practice turn. Important contributions to the agent/structure debate continued to appear throughout the 2000s.
Wendt, Dessler, and Carlsnaes
Sparking a side debate: Doty (1999) and Wight (2000).
Here it seems we may be returning to a positional ontology that Dessler tried to counter, however on the individual, rather than state level. This brings up all sorts of messiness in terms of agency/structure, levels of analysis, and meta-theory (epistemology/ontology) that are outside the scope of this review.
E.g. poststructuralism, postcolonialism, critical theory, critical feminism, etc.
For another example of how scholars in IR have proposed to bridge this gap see Vincent Pouilot (2010) International Security in Practice: The Politics of NATO-Russia Diplomacy. (Cambridge: Cambridge University Press).
Kirkpatrick – Revisiting Structure and Agency in IR
Revisiting Structure and Agency in International Relations – Contextualizing the Practice Turn
Erika M. Kirkpatrick
ekirk040@uottawa.ca
University of Ottawa
2013
AbstractThis article questions the general utility of a Bourdieusian approach as a solution to the agent-structure debate in International Relations Theory. It provides an historical overview of the agent-structure problem in order to furnish a critical reflection of how the 'practice turn' has appeared as a potential solution. The impetus behind the article is recent arguments regarding the productive capacity of Bourdieu to assist IR scholars in breaking with the polemic of structure-action. To respond to the question of whether or not Bourdieu offers a general solution to the agent-structure problem this article is organized into two sections. First there is a review of the literature that pays particular attention to the evolution of the debate. The second considers the proposed potential of Bourdieu and reviews an example of what such a solution would look like. In conclusion it is suggested that a Bourdieusian approach provides a very specific, rather than general, solution to the structure-agency problem.AbstractThis article questions the general utility of a Bourdieusian approach as a solution to the agent-structure debate in International Relations Theory. It provides an historical overview of the agent-structure problem in order to furnish a critical reflection of how the 'practice turn' has appeared as a potential solution. The impetus behind the article is recent arguments regarding the productive capacity of Bourdieu to assist IR scholars in breaking with the polemic of structure-action. To respond to the question of whether or not Bourdieu offers a general solution to the agent-structure problem this article is organized into two sections. First there is a review of the literature that pays particular attention to the evolution of the debate. The second considers the proposed potential of Bourdieu and reviews an example of what such a solution would look like. In conclusion it is suggested that a Bourdieusian approach provides a very specific, rather than general, solution to the structure-agency problem.
Abstract
This article questions the general utility of a Bourdieusian approach as a solution to the agent-structure debate in International Relations Theory. It provides an historical overview of the agent-structure problem in order to furnish a critical reflection of how the 'practice turn' has appeared as a potential solution. The impetus behind the article is recent arguments regarding the productive capacity of Bourdieu to assist IR scholars in breaking with the polemic of structure-action. To respond to the question of whether or not Bourdieu offers a general solution to the agent-structure problem this article is organized into two sections. First there is a review of the literature that pays particular attention to the evolution of the debate. The second considers the proposed potential of Bourdieu and reviews an example of what such a solution would look like. In conclusion it is suggested that a Bourdieusian approach provides a very specific, rather than general, solution to the structure-agency problem.
Abstract
This article questions the general utility of a Bourdieusian approach as a solution to the agent-structure debate in International Relations Theory. It provides an historical overview of the agent-structure problem in order to furnish a critical reflection of how the 'practice turn' has appeared as a potential solution. The impetus behind the article is recent arguments regarding the productive capacity of Bourdieu to assist IR scholars in breaking with the polemic of structure-action. To respond to the question of whether or not Bourdieu offers a general solution to the agent-structure problem this article is organized into two sections. First there is a review of the literature that pays particular attention to the evolution of the debate. The second considers the proposed potential of Bourdieu and reviews an example of what such a solution would look like. In conclusion it is suggested that a Bourdieusian approach provides a very specific, rather than general, solution to the structure-agency problem.
Introduction
According to both Alexander Wendt (1987) and David Dessler (1989) the agent-structure problem is based on two 'truths' about human existence that are themselves based on the fundamentally social nature of human beings. The first truth is that humans act – on each other, and on their environment – to build societies and enduring social and political structures; they may be perceived to do so independently and of their own volition. The second truth is that those selfsame social and political structures, based on 'independent' and 'willed' actions, endure through time and impact on individual human relationships, and the relationships between larger collectivities of people (nations, states, societies, firms, corporations, institutions, etc.). To sum up, the agency side of the debate proposes that independent human action is of utmost importance, while a focus on structure emphasizes their constraining and conditioning effect on human action. The most radical proposed solutions follow this self-same dichotomy: methodological individualism sees agents constructing their world outside the influence of structural constraint/enablement; while methodological structuralism views agents as automatons (re)acting within a structurally determined space.
Encountering this problem is not unique to International Relations, but rather is rooted in both the social and natural sciences, echoing out from the metatheoretical-metabehavioural conundrum of nature vs. nurture. An overview of the social science debate on this issue writ large is outside the scope of this article. Rather, herein I provide an historical overview of the agent-structure problem's incarnation as a 'debate' in the discipline of IR. Particularly, the contributions of a variety of major players in the debate: Alexander Wendt, David Dessler, Martin Hollis and Steve Smith, Walter Carlsnaes, Vivian Jabri and Stephen Chan, Roxanne Doty, and Colin Wight – their positions and points of critique – are reviewed. This overview traces the main lines of debate throughout the 1980s, 1990s and early 2000s – and contemplates its reverberations in the 2010s, it also serves as a reminder as what was/is perceived to 'be at stake in the agent-structure debate' (Dessler, 1989). As such, the first section is presented as a vehicle for considering the emergence of the latest round of engagement with this intractable problem – that is, a sociologically informed variant of the practice turn.
Engagement with the notion of 'practice' in IR has taken several forms, most notably through a poststructural attendance to the nuances of subjectivity and intersubjectivity, but also through a postmodern rejection of the grand narrative as an organizing principle for social life. Of particular concern here is the way that recent turns to practice have proposed a use of the sociological thought of Pierre Bourdieu as a 'solution' to the agent-structure problem. Although referencing Bourdieu is not a new phenomenon in IR, what is relatively novel for the discipline is the depth of engagement that is being made with his methodological tools. Specifically I want to engage with the observation made by Peter Jackson (2008; 2009) that one of the key targets of Bourdieu's "intellectual project…is the commonly held assumption that it is necessary to take sides on the question of agency and structure" in the social sciences (2008, 163; 2009, 105). If this side taking is in fact unnecessary then how do we overcome and move past it? Many in IR have suggested that we do so by turning to practices, specifically turning to 'practices' as understood through a Bourdieusian framework. Whether such a framework indeed assists us in addressing the agent-structure problem is of key concern here. As such, this article is perhaps most fruitfully read as a review of one of our discipline's most intractable debates, rather than as a proposed solution in its own right. Following the historical overview of the agent-structure debate in IR in section I, the second half of this article attempts an analysis of the perceived productive capacity of Bourdieu in assisting IR theorists in breaking with the polemic of agency-structure/structure-agency.
The structure-agency debate in International Relations has been described as having 2 'waves:' the first representing responses in International Relations theory to developments elsewhere in the social sciences, and the second representing poststructuralist challenges to the first wave (Bieler and Morton, 2001). Maintaining Bieler and Morton's two wave description of the structure agent debate in IR, below I trace the lines of debate through Wendt (1987) and Dessler's (1989) critiques of structural theories of international politics via engagements with scientific realism. To Hollis and Smith's entry into the debate around: their distinction between explaining and understanding (1990, 1994), their warnings against the importation of social theorists into IR through a critique of Walter Carlsnaes (1991), and their further engagements with both Wendt and Dessler, among others, over the philosophical foundations of the structure/agent problem (ontology vs. epistemology), following up with Jabri and Chan's response. From the philosophical critique, I move into a consideration of the 'second wave' of the agent-structure debate, especially the entry of poststructural IR theory through the work of Roxanne Lynn Doty and the debate it sparked between Doty and Colin Wight in the pages of EJIR in the late 1990s – which begs the question of whether the bullet aimed at the dead horse really did put the beast down for good. Alas, with a resounding 'no' on that last point, the early 2010s mark a return to the agent/structure problem with the introduction of the sociological practice turn into the fray. Scholars, exemplified by Didier Bigo, ignored Hollis and Smith's tongue in cheek admonishment to 'beware of gurus,' with their insistence on Pierre Bourdieu's theory to smart bomb the bullet ridden horse.
I
"The Agent-Structure Problem in International Relations Theory" (1987) is perhaps best viewed as Wendt's first foray into (what eventually became highly successful) theory construction in the discipline. He uses the structure-agent problem as a vehicle to make an important critique of the inability of the two dominant structural IR theories at the time (neorealism and world systems theory) to satisfactorily account for systemic change in light of the ending of the Cold War. Furthermore, this critique acted as a point of access into the broader meta-theoretical debate over epistemology and ontology that was ramping up at the end of the 1980s. In particular, he advocates a scientific realist philosophy of science over a narrow empiricism as a way to ameliorate research in International Relations. Drawing on Giddens' structuration theory, and Bhaskar's scientific realism, Wendt's overall goal was not to "convincingly demonstrate" the "utility of structuration theory as a meta-theoretical framework for international relations" (1987, 337), but rather to indicate 'necessary changes' that needed to be made in International Relations research. These necessary changes refer to the ways in which IR theorizes "the two basic building blocks of international relations theory, states and international system structures" (1987, 362). In effect, Wendt introduces a tension into IR theory around 'explaining and understanding,' proposing that what is really important in terms of the agent/structure problem is how thinking about it forces IR scholars to "make the actual behavior and properties of states and state systems 'problematic' rather than simply accepting them as given" (1987, 363). As such, his article is more fruitfully read as a call for debate over hitherto accepted ways of knowing in International Relations, rather than a fully developed 'solution' to the structure-agent problem, and debate is certainly what it sparked – starting with Dessler's query of the stakes of such a debate, and Hollis and Smith's insistence that no matter the stakes we should 'beware of gurus' and recognize that there are always 'two stories to tell.'
Like Wendt, Dessler is primarily concerned with applying scientific realism to international politics in order to remedy perceived deficiencies with structural theories of International Relations. In his 1989 article, Dessler equates the agent-structure problem with the inherent difficulties of theory building along the lines proposed by Kenneth Waltz. Thus, he attempts to use the external analytical framework of scientific realism, which argues that unobservables can be said to exist based on the causal relations between observable entities – that is, in the case of (neo)realism, states. In effect his critique of Waltz is aimed at producing a transformational model of political realism based on a relational ontology and a scientific realist understanding of empirical data that escapes the agent/structure problem through its more comprehensive (as compared to Waltz's positional model) ontology. In effect, Dessler wants to counter Waltz's positional model in terms of its conceptualization of structures as unintended/spontaneous, irreducible to unit actions/attributes, and impervious to transformation (1989, 450). Rather, the transformational model suggested by Dessler points toward the mutual constitution of structures and agents via a relational ontology that highlights the connections and relations of structures and agents in terms of the fundamental ways that structures enable/constrain action, as well as structures' existence as both a medium and outcome of action (1989, 452). In opening neorealism to a wider range of phenomenon beyond the position of states in the system, Dessler presents his model as "a more promising basis for progressive theoretical research" (1989, 444). According to Dessler, more phenomenon can be investigated in neorealism by accepting that it is the interaction of ontological categories (states and systems) rather than their (states) relevant power positions in the system that is truly important. A move made possible by a scientific realist epistemology. This is one disciplinary conundrum that fed into Hollis and Smith's publication of their IR guidebook Explaining and Understanding in International Relations, and the two stories it suggested we tell.
According to Hollis and Smith, stories that seek to explain world politics are those "told in the manner of a natural scientist seeking to explain the workings of nature, whilst stories that aim to aid or develop an understanding of world politics are "told so as to make us understand what…events mean, in a sense distinct from…unearthing the laws of nature" (1990, 1).The saliency of these stories for the agent-structure debate in IR are highlighted by Hollis and Smith (1994) through their critique of Walter Carlsnaes' 1992 article. In which he suggests that the agent/structure problem has two interrelated aspects – one epistemological and one ontological. However, Carlsnaes prefers to only deal with ontology, the "more fundamental" issue as it goes beyond the basic properties of structures/agents (Carlsnaes 1992, 248). According to Carlsnaes, an ontological focus highlights the relationship between structures and agents – a recurrent theme in articles dealing with the agent-structure problem. Hollis and Smith find this assumption that agents and structures are somehow ontologically comparable, as distinct and separate objects, problematic (Hollis and Smith, 1994, 244).
Carlsnae's priming of ontology is not a unique stance in the agent/structure debate; however his article represents an anomaly in at least two respects. In the first respect, because it is written from the perspective of Foreign Policy analysis, focusing on a practical case in European politics. In the second more theoretically pertinent respect, is his reliance on Margret Archer's concept of morphogenesis as a 'solution' to the agent-structure problem. At its core, the morphogenetic approach proposes that "structural factors…logically both predate and postdate any action affecting them…[and]…an action…logically both predates and postdates the structural factors conditioning it" (Carlsnaes 1992, 260). In effect, Carlsnaes introduces the issue of temporality (Doty, 1997) to the agent/structure debate, wherein actors and structures interact over time. This allows for a return to "analytical dualism" (Hollis and Smith, 1994, 243) by presenting a "strategy of uncovering morphogenetic cycles that can be analytically broken up into intervals in order to penetrate the relations between structure and action" (Carlsnaes 1992, 258). In effect, Carlsnaes argues for an approach to the agent/structure problem that refuses to 'bracket' either structures or agents, but rather prefers to bracket intervals of time. This is in order to tease out the relations between structures and action, and reveal specific causal dynamics at any given moment in processes of foreign policy decision making. Despite this valiant effort, Hollis and Smith insist that there are still 'two stories' to tell, particularly around the ontological grounding of Carlsnaes argument. They argue that asserting agents/structures as distinct objects that have relations with each other is, in effect, an epistemological move. Furthermore, it is a move that allows for an argument about casual conditioning which Hollis and Smith see as conflating ontological differences between different levels of analysis (Hollis and Smith, 1994, 244 -245).
Running concurrently to Hollis and Smith's critique of Carlsnaes' position, there was a brief exchange between them and Jabri and Chan over this same issue. Like Carlsnaes, although perhaps more forcefully than in his arguments Jabri and Chan push forward the notion of ontology as preceding epistemological concerns (1996, 107). In brief, their article argues that ontology must precede epistemology in International Relations theory in order to better understand the ontological underpinnings of theoretical difference (1996, 107). This side-lining of epistemology is based on their assertion that the primary function of epistemology is to attempt a discernment of causal relationships between specific human actions and specific aspects of social structures (Jabri and Chan 1996, 108). It is, in effect, the focus on causality that is under attack by Jabri and Chan because such a focus undermines a vision of structure and agency as mutually constitutive and instead requires a bracketing of either one or the other – a phenomenon that Hollis and Smith equate to the notion of a relational ontology between agents and structures as distinct entities (1994, 245). A caveat must be introduced into this argument – in discussing epistemology Jabri and Chan are referring to a particular variant of universal epistemology that functions to negate difference (1996, 107). It is exactly this point that Hollis and Smith seize upon in their response, proposing that the latter "seem to have made the bizarre assumption that epistemology is a peculiarity of positivism" (Hollis and Smith, 1996). Conversely, for Hollis and Smith ontology can only ever appear as central within a theory after the epistemological issue has been put to rest. This is due to their claim that disputes over ontology (over what exists) are always inherently concerned with epistemology (how we know) as this knowing is precisely what is in dispute (1996, 112). Thus, although Jabri and Chan argue that "the ontological basis" (1996, 109) of Giddens' structuration theory is what moves IR toward considering the mutual constitution of the ontological categories of structure and action, Hollis and Smith (as noted above) reject the conceptualization of structures and agents as ontologically distinct and thus comparable. In this sense these two sets of authors seem to be talking at cross purposes.
Nevertheless Jabri and Chan's critique of Hollis and Smith works to push the agent-structure debate in IR to an important point, one that is echoed in subsequent articles – that is, toward a focus on practice. Indeed their insistence that Giddens' structuration theory is most relevant in IR for the way in which it highlights how institutional practices and social continuities are spatially and temporally reproduced, while simultaneously acknowledging how meanings and practices are historically situated (1996, 108), pushes the notion of practice to the foreground. The explicitly ontological focus of structuration theory, according to Jabri and Chan, moves IR toward considering how the "production and reproduction of practices…become the basis for exclusionist dominant discourses and differential forms of enablement and constraint" (Jabri and Chan 1996, 109). In short, how the (re)production of social practices, based on agents' shared knowledge, feed into structural contingencies. As already noted, the issue of practice has been an underlying one in nearly every article that is part of the agent/structure debate. However, I assert that Jabri and Chan's arguments mark the commencement of the 'second wave' of the agent structure debate; wherein the ontology vs. epistemology, and levels of analysis problems, that are more salient in the early years, largely fade into the background. In effect the late 1990s marks a point of relative commensurability in which: 1) ontology emerges as primary in discussing structure and action, regardless of author's underlying epistemological commitments; and 2) structures and agents are accepted as ontologically comparable. Both features are apparent in the last portion of the debate discussed in this paper between Colin Wight and Roxanne Lynn Doty.
In her 1997 article, Doty revisits the early writings in the agent/structure debate to argue that neither presents a convincing solution to the problem. Finding all three deficient, she attempts to subject the agent/structure problem to a 'poststructural gaze' in order to offer "a radical, decentered understanding of practices" (1997, 366). In brief, Doty highlights three problems within the early agent/structure writers: i) the framing of the problem in terms of an oppositional logic that presumes agents, structure, or some combination of the two are the most significant forces in social and political life (replication of original problem) (1997, 368); ii) scientific realism is used to maintain scientific legitimacy and in the process upholds an essentialized notion of structure (1997, 370); and iii) morphogenesis necessitates that scholarly focus be either on agents or structures at any single point in time (thus again replicating the original problem) (1997, 373). These form the aporia of her title: "a self-engendered paradox beyond which it [the agent-structure 'problématique'] cannot pass" (1997, 375). This paradox may be impassable, but Doty argues that confronting it nonetheless points towards the centrality of practice in all three of the scholars she references – a centrality that is "asserted to be key, [but] is not really taken seriously enough" (1997, 376). Rather, Doty presents their notions of practice as restricted and conservative because they locate the meaning and sources of practice in pregiven "subject[s] or generative structural principles" (1997, 376). Conversely, she insists that practices must be decentered by investigating the ways in which meanings are imposed and constructed. That is, by attending to the question of power (1997, 376). In effect, Doty argues that the 'structure/agency writers' will never be able to move beyond the alternatives of structural determinism or autonomous subject 'solutions' because they follow in the modernist tradition. Her response is to turn toward discursive/representational practices in order to question the production of what structures and agents are. In other words, to turn toward "the question of determining which…repetitive and widely disseminated practices produce the effects of agency and structure" (1997, 387). This argument represents a significant move toward a serious discussion of practice in terms of the agent-structure debate. However, it was not received without criticism.
The bulk of that criticism was presented by Colin Wight in his 1999 article which suggested that Doty's enthusiasm for the indeterminacy of the agent-structure problem was akin to overkill (do they really shoot dead horses?). He rejects Doty's framing of the early writers as oppositional, but rather proposes that they offer alternatives to dominant IR conceptualizations, rather than solutions to the agent-structure problem as such. The bulk of Wight's criticism seem to have been directed at Doty's self-designation as a poststructuralist and, flowing from this, her perceived deficiency in reading Wendt, Carlsnaes, and Dessler (and the social theorists upon whom they draw). However, his critiques of Doty do enable Wight to present a rather convincing account of the lack of an attendance to agency in the agency/structure debate. He presents a layered notion of agency (three layers in this account, but by his own admission more layers are possible - 1999, 133), wherein practices become important in positional terms. To summarize, there are three layers of agency: agency1 is the most basic level, all humans are agents in terms of being born and growing up in particular socio-cultural contexts; agency2 is when anyone becomes an agent of that socio-cultural system; and agency3 refers to "positioned-practice-places which agents1 inhabit on behalf of agents2" ( Wight 1999, 133-34). The important category that can be taken from it is that of 'positioned practices' which are "structural properties that persist irrespective of the agents that occupy them" (Wight 1999, 134). Here Wight seems to have pushed the agent-structure division to its breaking point by showing that agency may itself be a structure. He reaches this conclusion through an argument that highlights the role of the position of agents within a social field – an argument that is picked up on by theorists applying the sociological tools of Bourdieu in International Relations theory.
II
Before moving onto a specific discussion of the ways IR theorists have suggested that the sociology of Pierre Bourdieu may be used to counter, or dissolve, the agent/structure problem, I want to provide a brief overview of thee of Bourdieu's key 'thinking tools' that are implicated in suggestions of the utility of his thought for IR: 'habitus,' 'field,' and 'practice.' In his major work Distinction: A Social Critique of the Judgment of Taste Bourdieu presents the notion of habitus as both a 'structuring structure' and a 'structured structure.' In effect habitus is defined as a relationship between generative principles and systems of classification, when these are internalized they work to generate meaningful practices and meaning giving perceptions that help form an individual's habitus (Bourdieu 101-103, 1984). In summary, habitus can be defined as the internalization of external social structures by social agents. These enable those agents' effective function within a social field. Thus, habitus is made up of, on the one hand both unconscious and conscious "learned experiences" and on the other by "the cumulative impact of practices" (Jackson 2009, 108). This presents habitus as inseparable from the notion of field as a "'particular social universe'" (Jackson 2009, 108) in which a habitus exists.
In simplified terms, the notion of field may be presented as a "gaming space" (Bourdieu 1996, 265) or a playing field in which players intuitively know the rules of the game and what is at stake in choosing to play. The game analogy provides an intuitive understanding of the sphere of play as an ordered universe wherein competence is unequal between players (Mahar, Harker and Wilkes 1990, 70-72). Thus, a field can be more abstractly understood as "an arena of struggle where actors compete for various forms of material and symbolic power resources" (Jackson 2009, 108). Bourdieu and Wacquant assert that "the field structures the habitus which is the product of the incorporation of the immanent demands of the field…[and] the habitus contributes to the constitution of the field as a world of meaning" (1992, 119). This inseparability leads to a Bourdieusian notion of practice. Therein, practices emerge or arise from interrelationships between field(s) and habitus. This is where the perceived power of a Bourdieusian approach lies in terms of the agent/structure problem: it helps us to move past an individual-society dichotomy, pointing instead towards an understanding of habitus and fields as engendering practices which then feed back into the original concepts.
Central to Bourdieu's 'theory of practice' is an attempt to unite "'subjectivist' agent-centered approaches…with 'objectivist' accounts that emphasize the role of structural conditions in shaping social life" (Jackson 2009, 102). This is a good summary of the perceived utility of Bourdieu in terms of the particular debate at hand. If overcoming the structure/agent problem was at the core of Bourdieu's own sociological project, than ipso facto it becomes a way of overcoming the same problem in IR theory. However, Roy Bashkar, Anthony Giddens, and Margret Archer were all also preoccupied with the agent/structure problem and as shown above, this does not mean that use of such theorists in IR necessarily 'solves' the agents/structure problem. Perhaps this is where Hollis and Smith's admonishment to 'beware of gurus' deserves repeating. Their suggestion is not that social theorists are of no value for IR theory; on the contrary I think they were trying to highlight the need for caution in the uptake of such theorists by IR scholars. Thankfully, this has been the case – as evidenced by the many well thought out and profitable engagements with Bourdieu by scholars in the field of IR. Thus, I do not wish to suggest in this article that Bourdieu is of no utility in confronting the agent/structure problem, rather I want to critically assess IR scholars that draw upon Bourdieu specifically around this issue.
According to Peter Jackson in his chapter on Bourdieu in a recent (2009) book on the utility of critical theorists for IR, Bourdieu's usefulness lies in his "focus on the inter-relationship between the material and symbolic dimensions of power in social life" (102). One such inter-relationship has been the preoccupation of the above literature review, that is, the debate over structure vs. agency – both of which contain elements of materiality and symbolism. By honing in on the relational aspects (which has been highlighted throughout the agent-structure debate) it is suggested that side taking can be avoided (Jackson, 2009, 105-106). In effect, for the sociological practice turn the first step in overcoming the agent-structure problem is to recognize that any division between the two categories is by and large artificial (Jackson, 2009, 106). It is so because, as the proceeding quote from Bourdieu highlights, structures and agents are not merely mutually constitutive, they are inseparable in practice:
On the one hand, the objective structures that the sociologist constructs in the objectivist moment, by setting aside the subjective agents, form the basis for these representations and constitute the structural constrains that bear upon interactions; but on the other hand these representations must also be taken into consideration particularly if one wants to account for the daily struggles, individual and collective, which purport to transform or to preserve these structures (1989, 15).
Here we see that it is the scholar who imposes a division in theory, to simplify the complexity of human practice. Thus, Bourdieu attempts to overcome the agent-structure problem by recognizing that it is a consequence of the scholarly endeavor of understanding and explaining the complex world in which we live. To do so, he proposed the categories of 'habitus' and 'field' to provide tools for thinking about and researching the "interplay between the subjective perspectives and predispositions of social actors, their habitus, and the structural conditions of the particular social context in which they are acting, the field" (Jackson, 2009, 106). Thus, the oppositional logic that Doty ascribes to the early structure-agency writers is revealed as false through a reference to the Bourdieusian logic of practice (Bigo and Madsen, 2011, 221). All this being said, Bourdieu designed his theory of practice for the academic field of sociology; despite geopolitical influences on his life and work (notably the Algerian War) he was above all a social theorist focused on domestic relationships (domestic in two sense – in one state, France, and in the homes, schools and workplaces of French people). It has largely been within the past 5-10 years that IR scholars have set out to discover how best to apply his thinking tools in their own scholarship.
A new volume, released early this year, attempts to do just that. Edited by Rebecca Adler-Nissen, Bourdieu and International Relations provides an excellent overview by scholars in the field of the utility of Bourdieu, and the challenges involved in taking up his legacy outside of social theory. I want to highlight Adler-Nissen's comments on Bourdieu as regards the structure-agent problem, before moving on a discussion of one Bourdieu inspired solution put forth by Bigo in 2011. Three points made by Alder-Nissen are worth repeating here. First, Bourdieu can help us with the agent-structure problem by "dissolving it in convenient analytical units" – habitus and field. Second, the use of these units allows us to "map political units as spaces of practical knowledge on which…agencies position themselves and…shape international politics" – problematizing the state; and. Third, Bourdieu eschews a "simple relationship between the individual and society [and]…substitutes [it with] the constructed relationship between habitus and field(s)" (2013, 8). Despite the resounding cheers by Adler-Nissen and Jackson on the ability of Bourdieu to help us out of the agent-structure problem, neither tells us what such a solution may look like in analytical terms. Thus, I turn to Bigo for a more specific discussion.
Bigo begins his article with what is by now familiar trope in IR, asserting that the discipline has long been organized in oppositional terms between an objectivist-empiricist mainstream and a subjectivist-idealist constructivism ( Bigo 2011, 226). He, rightly, points out that Bourdieu viewed this objectivist-subjectivist opposition as the "'rock bottom antimony upon which all the divisions of the social sciences are ultimately founded" (Bigo 2011, 229 [Bourdieu, 1988, 778] emphasis added); all the divisions – including those between structures and agents. Thus, overcoming the fundamental opposition between objectivism and subjectivism is part and parcel of dealing with the agent-structure problem. Indeed, breaking "with the antimony or dialectic of agents and structures" is "central" to the Bourdieusian project of putting objectivism and subjectivism in "symbiotic relation" (Bigo, 2011, 235). To do so, to break the dialectic, Bigo proposes that we first rethink the traditional dichotomy of the agent-structure problem wherein structures are presented as "collective and abstract" as opposed to agents as individual, speaking subjects that are "permanently conscious of" their own "free will" (Bigo 2011, 235). Rather, Bigo emphasizes the thinking tools of field and habitus to interrupt this dichotomy. Fields present us with a more specific and concrete conceptualization of 'structures,' while habitus pushes us to conceive of agents as more than simply conscious individuals, but rather as actors with complex dispositions, that they are not always aware of, which impact their actions. In short, the purpose of Bigo's article is to propose how these thinking tools may allow us to escape the structure-agent problem through a relational approach.
Bigo differentiates his relational approach from the common "positions of individualism, structuralism, and interactionism" (2011, 235) which have contributed to the agent-structure problem in IR theory. In order to avoid the "structure-agency dilemma" (Bigo 2011, 235) presented in section I of this article, Bigo proposes a relational approach made up of two key characteristics. First, such an approach focuses on "the invisibility of the relations between agents rather than the visibility of these same agents" (Bigo 2011, 235). To clarify: rather than only focusing on agents' salient features such as formal agreements, formal positions/titles, concrete capital (e.g. liquid assets) or any other 'visible' characteristic, the notion of 'invisible relations' suggests that researchers focus on uncovering, detecting, and problematizing the tacit inter-agential relationships beneath, alongside, and above agents' 'visibility.' Secondly, such an approach serves to interrupt the simple mutual constitutions of the relational ontologies traditionally presented in the structure-agency debate by rejecting their tendency to "presuppose the existence of fully constituted agents" (Bigo 2011, 237). Rather, Bigo wants us to begin "with the moment of the making of the action" and to "consider the agents only when they act in relation to each other" (2011, 237). Through these two moves it is suggested that researchers should be able to avoid the structure-agent problem.
Specifically, this allows those following such a relational approach to avoid the 'interactionist' presumption of fully constituted objective agents, and allows – through an ontology of practice – for the object of analysis to shift from the agents themselves to the relations between them. As for the structure side, Bigo goes on to propose an attention to dynamism and specificity as an escape chute. Accordingly, "a relational approach in Bourdieusian terms has to examine change and transformation of specific processes and at a specific time (and duration)" (Bigo, 2011, 237). By so doing, the researcher is able to avoid an invocation of grand narratives and totalizing explanations – thus, "avoiding any idea of structure" (2011, 237). Avoiding the structural trap by refusing the siren's call of "explaining the entire history of humankind" (Bigo, 2011, 237) is a direct admonishment to ahistorical explanations like Marxism and neorealism – in short, one issue that I view then as defining this latest round of the agent/structure debate seems to be an issue of scale as to what counts as structure.
Conclusion
The structure-agent debate in international relations has, like many other intra-disciplinary debates reached the status of disciplinary mythology, and has to a large extent become part and parcel of the 'lore' of International Relations. The predicament that Wendt has so succinctly summarized flows from two opposing approaches in the social sciences, structuralism and intentionalism (Bieler and Morton 2001, 7) – the first giving nearly insurmountable power to structures over social life, the other insisting on the intractable capacity of human agents regardless of the structures in which they are located. Beyond this simple dichotomy, I would add that the structure-agent problem in IR follows in part from the very complexity of our subject matter. When dealing with 'big picture' questions of war and peace, equality and exploitation, governance and chaos – on a global scale – parsing out, bracketing, or otherwise simplifying the 'variables' involved is to an extent unavoidable.
The strength of a Bourdieusian approach thus seems to lie in its specificity. Indeed if one browses the works of scholars using Bourdieu they should observe at least two common features: a commitment to empirical research that brings them into close proximity to the practitioners of international politics, and a specificity and precision of subject matter that sees them focus on concrete sites and issues within international politics. Coupled with these is a commitment to reflexivity by these scholars that sees them problematize their own place as IR scholars in these research endeavors and the dynamics of international politics that they are studying. In effect, a Bourdieusian approach does assist researchers interested in the dynamics of specific practices at specific times. However, as noted above, one of the challenges involved is that throughout its history as a discipline International Relations has very much been preoccupied with bigger questions that are notable for their lack of specificity – this being the case a Bourdieusian approach may assist some in moving past structure and agency in their specific research projects but such an approach may not be applicable to all research projects in IR. Thus its utility for the discipline as a whole in overcoming structure and agency is equal to the various 'solutions' that have been proposed throughout the agent-structure debate.
References
Adler-Nissen, Rebecca. (2013) Bourdieu in International Relations: Rethinking Key Concepts in IR. (Routledge: New York).
Bigo, Didier. (2011) "Pierre Bourdieu and International Relations: Power of Practices, Practices of Power." International Political Sociology 5: 225-258.
Bieler, Andreas and Davis Morton. (2001) "The Gordian Knot of Agency-Structure in International Relations: A Neo-Gramscian Perspective." European Journal of International Relations 7(1): 5-35.
Bourdieu, Pierre. (1984) Distinction: A Social Critique on the Judgment of Taste. Translated by Richard Nice (Harvard University Press).
Bourdieu, Pierre. (1989) "Social Space and Symbolic Power." Sociological Theory 7(1): 14-25.
Bourdieu, Pierre. (1996) The State Nobility. (Stanford: Stanford University Press).
Bourdieu, Pierre and Loïc Wacquant. (1992) An Invitation to Reflexive Sociology. (Chicago: Chicago University Press).
Carlsnaes, Walter. (1992) "The Agency-Structure Problem in Foreign Policy Analysis." International Studies Quarterly 36: 245-270.
Dessler, David. (1989) "What's at Stake in the Agent-Structure Debate" International Organization 43 (3): 441-473.
Doty, Roxanne Lynn. (1997) "Aporia: A Critical Exploration of the Agent-Structure Problématique in International Relations Theory." European Journal of International Relations 3(3): 365-392.
Doty, Roxanne Lynn. (1999) "A Reply to Colin Wight." European Journal of International Relations 5(3): 387-390.
Hollis, Martin and Steve Smith. (1990) Explaining and Understanding in International Relations. (Oxford: Oxford University Press).
Hollis, Martin and Steve Smith. (1991) "Beware of Gurus: Structure and Action in International Relations." Review of International Studies 17(4): 393-410.
Hollis, Martin and Steve Smith. (1992) "Structure and Action: Further Comment." Review of International Studies 28(2): 187-188.
Hollis, Marin and Steve Smith. (1994) "Two Stories about Structure and Agency." Review of International Studies 20: 241-251.
Hollis, Martin and Steve Smith. (1996) "A Response: Why Epistemology Matters in International Theory." Review of International Theory 22(1): 111-116.
Jabri, Vivian and Stephen Chan. "The Ontologist Always Rings Twice: Two More Stories About Structure and Agency in Reply to Hollis and Smith." Review of International Studies 22(1): 107-110.
Jackson, Peter. (2008) "Pierre Bourdieu, the 'Cultural Turn' and the Practice of International History." Review of International Studies 34: 155-181.
Jackson, Peter (2009) "Pierre Bourdieu." In Critical Theorists and International Relations edited by Jenny Edkins and Nick Vaughan Williams. (Routledge: New York): 102-113.
Mahar, Cheleen; Harker, Richard and Chris Wilkes. (1990) An Introduction to the Work of Pierre Bourdieu: The Practice of Theory. (Macmillan: Houndmills, Basingstoke, Hampshire)
Wendt, Alexander. (1987) "The Agent-Structure Problem in International Relations Theory." International Organization 41(3): 335-370.
Wight, Colin. (1999) "The Shoot Dead Horses Don't They?: Locating Agency in the Agent-Structure Problematique." European Journal of International Relations 5(1): 109-142.
Wight, Colin. (2000) "Interpretation all the Way Down?: A Reply to Roxanne Lynn Doty." European Journal of International Relations 6(3): 423-430.
13