Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
168 168
Appendix
those grooves—as we did at Nemea with the modern runners—we assured an equal starting stance for every runner. The start of the races was accompanied by a verbal command from the aphetes. Our version version was a threethr ee-part part shout using using the words words "έτοιμοι," "έτοιμ οι," "εμπρός," "απιτε," "απιτε ," the t he 34 2 last of which was dictated by the documented use of the word in antiquity. It became clear that the athletes tried to synchronize a rocking motion with the rhythm of the commands from the "aphetes" so that the upper parts of their bodies would be moving forward—no matter how slightly—with the third command. It thus seems likely to me that such a system of rhythm—whether introduced by verbal commands or by the accompaniment of a flute—was used in antiquity. It was certainly clear that the invention and use of the hysplex in no way obviates the need for an audible starting signal. But the major question that was answered by the experiment at Nemea was whether the feet of the runners would become tangled in the cords that were thrown down in 343 front of them by the ankones of the hysplex. The answer is that there were absolutely no problems. To be sure, a couple of the runners—obviously very concerned with the possibility of tripping—hopped or jumped out over the cords. But most simply put their heads down and ran over and beyond the cords without fear and without mishap 34 4 (figs. 125-128). Given the much greater speed of the fall of the cords compared
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
HYSPLEX The Starting Mechanism in Ancient Stadia A Contribution to Ancient Greek Technology
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
Trusted by over 1 million members
Try Scribd FREE for 30 days to access over 125 million titles without ads or interruptions! Start Free Trial Cancel Anytime.
UNIVERSITY OF CALIFORNIA PUBLICATIONS: CLASSICAL STUDIES Editorial Board: John K. Anderson, David Blank, Richard Janko, Donald Mastronarde, Ronald Mellor, Jo-Ann Shelton Volume 36
UNIVERSITY OF CALIFORNIA PRESS BERKELEY AND LOS ANGELES, CALIFORNIA UNIVERSITY OF CALIFORNIA PRESS, LTD. LONDON, ENGLAND © 1999 BY THE REGENTS OF THE UNIVERSITY OF CALIFORNIA PRINTED IN THE UNITED STATES OF AMERICA
Contents
List of Fig ures
ix
Prologue
xv
Abbre viatio ns and Bibliography
xvii
Introduction
1
Docu ment s and Docu menta tion of the Mechanism
3
Wri tten tt en Sources and Terminology Archit ectura l Features of Stad ia The New New Pan ath ena ic Ampho ra The Parts of the Hysplex and Its Means of Operati on Reconstruction of the Form of the Hysplex The Operat ion of the Mechanism Mechanism The Chronology of the Hysplex Representations of a Hysplex in Monuments of a Roman Dat e Observ ation s on th e Star ting Lines of Various Stad ia Mainland Greek Stadia
3 10 20 31 35 44 49 53 57 57
viii
Contents
Th e Finish The The Cost Conclusion
169 169 172 172 173
Su mm ar y in Greek
175
Glossary
179
Index
181
List of Figures
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Reconstruction of the hysplex system syste m at Nemea frontispiece Isthmia. Pla n of the balbis and reconstruction of the hysplex system with the sta rt of the runners in Stadium I 7 Isthmia. Plan of the balbides and bases for the hyspleges 8 Epidaur os. Plan of the balbides and bases for the hyspleges 8 Corinth. Plan of the balbides and bases for the hyspleges 9 Nemea. Plan Pl an of the balbides and bases for the hyspleges 9 Isthmia Isth mia.. Pla n and cross-section cross-section of the th e single single base of the hysplex (the left left one) excavated excav ated in Sta dium diu m II 10 Epidaur Epi dauros. os. Plan and cross-section cross-section of the right base of the th e hysplex . . 11 Corinth. Corin th. Plan and cross-section cross-section of the two bases of the th e hysplex 12 Corinth. Corin th. Plan Pl an of the left left base of the hysplex 13 Corint Cor inth. h. Left Left base of the th e hysplex, from from the th e west 13 Corinth. Corin th. Pla n, side side elevations, and cross-section cross-section of the th e right base of the hysplex 14 Corinth. Cori nth. Right base of the hysplex, from th e west 15 Corint h. Plan Pla n and cross-section cross-section of the central base of the hysplex. . . . 16 Corinth. Cori nth. Central Cent ral base of the hysplex, from from the th e west 17 Nemea. Plan Pl an of the bases bases of the hysplex of the southern star ting line. 18 Nemea. Left Left base of the th e hysplex, from from the nor th 18 Nemea. Right base of the hysplex, from from the nor north th 19
χ
List of Figures
30 31 32 33 34 35
36 37 38 39 40 41 42 43
Drawing showing showing terms and materials materia ls of of the basic elements of ancient stone-throwing machines Schematic drawing of the neura and the metallic elements that made up the moving parts of the hysplex Wooden scale model of the hysplex Corinth. Corint h. Plan of the two bases of the hysplex showing the brackets that would would create crea te the cord barriers barrie rs in an application applica tion like the th e Nemea system. syste m. Corinth. Corint h. Suggested Suggested schematic reconstructi recons truction on of all par ts of the hysplex. The ankon was significantly taller tall er than th an shown here Nemea. Plan of the sphendone of the stadium. The semicircular depres sion for the aphetes is visible near the top of the semicircle of the water channel Codex Vaticano Latino Lat ino 3439 Mosaic at Gafsa Mosaic at Tebessa Olympia. Brick Brick foundation next to block block y at the southern sout hern end of the western balbis with cuttings for the receipt of the hysplex Olympia. Detail Detail of of block block i in the eastern balbis with the semicircular cutting intended to receive the hysplex Didyma. Plan Pla n and longitudinal longitudina l crosscross-sectio section n of the bases of the balbis and hysplex Didyma. Schematic reconstructi recons truction on of the balbis and the hysplex. ... Plan of the bases of the fir phase of the line
40 41 43 45 46
47 54 55 56 58 60
66 66 67
List of Figures
57
58
59 60 61 62 63 64 65 66 67
Rhodes. Rhodes . The remains of slabs of the balbis of the third phase and the base of the hysplex of the second phase on the northern end of the stad st adiu ium, m, from from the th e west, 1973 Rhodes. Rhodes . The remains of slabs of the th e balbis of the third phase and the base of the hysplex of the second phase on the northern end of the sta dium, diu m, from from the east, 1973 1973 Rhode s. The left left base of the southern sout hern hysplex of the second phase, from the east , 1973 1973 Rhode s. The left left base of the southern south ern hysplex of the second phase, from th e west, 1973 Rhodes. Plan of the remains of the southern star ting line line Rhodes. Rhodes . The operating opera ting channel and circular pit for for the aphetes at the southern sout hern sta rti ng line of the thi rd phase, from from the th e sout h, 197 1973 3 Rhodes. Rhodes . The operati oper ating ng channel and circular pit for for the aphetes at the southern sout hern sta rti ng line of the thi rd phase, from from the th e nor th, 197 1973 3 Rhodes. Rhode s. The circular pit for for the th e aphetes, from from the nor northe the ast , 1973 1973.. . . Rhodes. Rhode s. Poros underpinning underpi nning of the left left hysplex of the southern starting line of the th e thir th ird d phase, phas e, from from the th e east, eas t, 1992 1992 Rhodes. Schematic reconstruction of the southern star ting line line of the third phase of the stadium Rhodes. Rhodes . The channel for for the meta, from from the th e south, sou th, 1973
xi
82
83 84 85 86 88 89 90 92 93 94
List of Figures
xii
79 80 81 82 83 84 85 86
87 88 89 90 91
Epidau Epi dauros ros.. Bot tom and one side of the base of the th e centr al half column. Epidau Epi dauros ros.. Bot tom of the base of the central centr al half half column Epi dau ros . Front view view of the base of th e centr al half column set set in place on its foundation found ation,, from from the west west Epidau Epi dauros ros.. Rear view view of the base of the centr al half column set set in place on its foundation, founda tion, from from the th e east Epidau Epi dauros ros.. Rear view view of the first first dru drum m of the th e central centra l half column. . . Epidau Epi dauros ros.. Rear view view of the base of th e centr al half column Epida uros. Reconstr uction of of the star ting mechanism on one side of a half column Epidau ros. Schematic reconstruction of the hysplex of phase II at the eastern end of the stadium, with wood barriers and wood epistyle thr oug h which which ran a cord (do (dotted tted line) line) Pri ene. ene . Plan of th e bases of the th e first first and second second phases of the start sta rtin ing g line Priene. Pri ene. Elevat ion of of the base and cross-section cross-section of th e epistyle of one of the pillars of the th e sta rtin rt ing g line Prien e. Schematic reconstruction of the mechanism for for the introduct ion of torsio tor sion n on one side of th e pillar Prien e. Schematic reconstruction of the hysplex with the two systems for for the th e intr odu oducti ction on of torsion and its release Kos. General vi of the architectur al remains of the balbis-hysplex,
106 106 106 106 107 107 107 108 108 109 109 110 110
112 112 113 114 114 115 115 117 117
List of Figures
106 106 107 107 108 108
109 109 110 110 111 112 112 113 114 114 115 115
Kos. Detail of the half-cylindrica half-cylindricall cutting cutti ng on the back par t of the th e third thi rd of the ext ant half columns, from from the th e east Kos. Drawing of the th e back side side of the front, half-round, element of the th e half column of the hysplex Kos. Reconstructi Reconst ruction on of the th e mechanisms for for the th e introdu int roducti ction on of of torsion torsio n (broken line) and its release (dotted line) in the interior of a half column. Metallic Metalli c elements elemen ts in heavy solid black line Kos. Schematic reconstruct recons truction ion of the th e hysplex with its two systems for the th e introd int roduct uction ion of torsion torsi on (broken line) line) and its release (dotted (dot ted line). . Nemea. Modern Moder n wooden post inserte ins erted d in ancient anci ent lead-lined socket of balbis Nemea. Wooden column and frame of hysplex inserted in right (eastern) support supp ort base, from from the th e north nor th Nemea. Petr Pe tros os and Voula Briles working wood to fit fit into left left (western) (wester n) support supp ort base Nemea. Le Left ft (western) hysplex from the northwest with wooden blocks A, B, and Γ (from (from rig right ht to left) left) and column at rear rea r Nemea. Right (eastern) (east ern) hysplex from the northeast with wooden blocks Ζ, Ε, and Δ (from (from left left to right) righ t) and column at rear rea r Nemea. Detail of right hysplex from the north showing ankon inserted in neura (rope) above cut-down part pa rt of central cent ral block Ε
xiii
134 134 136 136
137 139 139
146 146 147 147 149 149 150 150 151 152 152
Prologue
Nothing shows so clearly the state of research on the starting mechanism in the ancient stadium as the following quotations from two important students of the stadium and of ancient athletics in general, 0. Broneer and J. Jüthner. The first introduces the problem to us gently: "Architecturally "Architec turally the Greek stadiu sta dium m appea rs to be a simple structure, the principal features and the use of which can be readily understood. Actually both its construction and its functioning have given rise to many questions that still remain unanswered. The most difficult problem has to do with the lines and devices for regulating the start and finish of the various forms of foot races. In
spite of much study on the part of several competent scholars, it is still a matter of uncertainty what signal or mechanical operation gave the runners the cue for the start of the race at any given period." 1 The second is more laconic and authoritative: "Es gibt wenige Einzelfragen der Altertumskunde, bei denen eine solche Fulle wertvoller Nachrichten zu Gebote steht und wo doch so wenig sicheres zu ermitteln ist, wie bei der Untersuchung der Ablaufeinrichtungen im griechischen Stadion." 2 In fact, despite our relatively good state of knowledge about various aspects of ancient Greek athletics, areas of darkness exist which continue to stimulate further study, discussion, and research. One of these dark areas is the means of starting the
XVI XV I
Prologue
I could not have finished this study without the help of many friends and colleagues to whom I wish to express my gratitude. N. Prokopiou granted me permission to study and publish the vase despite her intention to study Panathenaic amphoras. My friend the architect Ntinos Kyriakopoulos, with his great knowledge of ancient technology, his talent for synthesis and for drafting, and his—all too rare in our day—sincerely al truistic generosity, helped to check and solve many problems and so made it possible to provide very detailed reconstructions of the mechanisms. Also, I was helped greatly by the discussions I had from time to time with M. Korres, A. Matthaiou, and M. Tiverios, who read and made important comments on an earlier version of this study. Finally, for every kind of information, assistance, and criticism, I must thank V. Bardani, Ch. Bouras, E. Brouskari, G. Despinis, S. Halatsis, Ch. Kantzia, Th. Karagiorga, V. Kontorini, E. Kondylaki, M. Kreeb, Ch. Kritzas, M. Michalake-Kollia, I. Ninnou, F. Pachyianne, I. Papachristodoulou, M. Philimonos, E. Stasinopoulou-Kakarougka, M. Theochari, D. Tsouklidou, and A. Yiannikouri. I also benefited from the comments and suggestions of three anonymous readers for the University of California Press and the editorial assistance of R. A. White and S. Berg. The photographs of the vase are by G. Maravelias, and the drawings, aside from the reconstructions by Nt. Kyriakopou los, are by L. Vranopoulou. J. Parsons prepared the camera-ready copy of this book. I thank all warmly.
Abbreviations and Bibliography
In addition to the standard internationally recognized bibliographic abbreviations, the following are used here: Arete = S.G. Miller, Arete: Greek Sports from Ancient Sources (second edition, Berke ley, 1991). Aupert = P. Aupert, Fouilles de Delphes II, Le Stade (Paris, 1980). Baatz Baa tz (1979) (1979) = D. Baatz, "Teile "Teile Hellenistischer Hellenistischer Geschiitze aus Griechenland," Griechenl and," AA (1979) 68 ff. Baatz (1982) = D. Baatz, "Hellenistische Katapulte aus Ephyra (Epirus)," AM 97 (1982) 211 ff. Baatz Baa tz (1985) (1985) = D. Baatz, Baa tz, "Katapu "Kat apult lttei teile le aus dem Schiff Schiffswr swrack ack von Mahdia Mahd ia (Tunesien)," A A (1985) 679 ff. Bean = G. Bean, Aegean Turkey, an Archaeological Guide (London, 1966). Broneer = 0. Broneer, Isthmia II, Topography and Architecture (Princeton, 1973). Fiechter = E. Fiechter, "Stadion, der Bau," RE IIIA2 (1929) 1967 ff. JHS S 23 (1903) 261 if. Gardiner = E.N. Gardiner, "Note on the Greek Foot Race," JH Harris (1960) = H.A. Harris, "Stadia and Starting Grooves," Greece & Rome 7 (1960)
XVIII
Abbreviations and Bibliography
Miller (1980) = Stephen G. Miller, "Turns and Lanes in the Greek Stadium," AJA 84 (1980) 159-166. Morgan = C. Morgan, II, "Excavations at Corinth, 1936-37," AJA 41 (1937) 549 if. Morricone = L. Morricone, "Scavi e Ricerche a Coo (1935-1943). Relazione Preliminare," Boll.d'Arte 35 (1950) 222 if. Nemea: A Guide = S.G. Miller (ed.), Nemea: A Guide to the Site and the Museum (Berkeley, 1990). Patrucco = R. Patrucco, Lo Sport nella Grecia Antica (Florence, 1972). Proceedings = W. Coulson and H. Kyrieleis (eds.), Proceedings of an International Symposium on the Olympic Games, 5-9 September 1988 (Athens, 1992). Romano = D.G. Romano, The Stadia of the Peloponnesos (Diss. U. Pennsylvania, 1981). Roos = P. Roos, "The Start of the Greek Foot Race," OpAth 6 (1965) 149-156. Schilbach = J. Schilbach, "Olympia, die Entwicklungsphasen des Stadions," in Pro ceedings 33-38. Schramm I = E. Schramm, Die antiken Geschutze der Saalburg (1918) reprinted with an introduction by D. Baatz in Beiheft zum Saalburg-Jahrbuch (1980) Schramm II = E. Schramm, "Poliorketik," in J. Kromeyer and G. Veith (eds.), Heerwesen und Kriegfuhrung der Griechen und Romer, HdA 4, 3, 2 (1929) 209 ff. Shear = T.L. Shea r, Jr. "The Athenia n Agora: Excavations Excav ations of 1973 1973-74, -74,"" Hesperia 44
Introduction
The need to guarantee a simultaneous start for all the footraces and especially for the stadion race was certainly one of the more important problems that the officials of the competitions in ancient Greek city-states and sanctuaries had to confront and 4 solve. This was because the execution of the games was a reflection of the authority, the reputation, and the more general prestige of the various competitions. It was not by chance that the whole of ancient Greek literature—from Herodotus to Plutarch and Julian—is filled with references to early, or false, starts by athletes and to their 5 consequent punishment. The oldest and best known is found in Herodotus (8.59) and concerns the discussion by the Greek generals about their plan of action shortly before the battle of Salamis. In response to the proposal by Themistokles for immediate action, th e Corinthian general Adeimantos said: said: "Themistokles, in the games those who start too soon get flogged." To which the Athenian general gave his instant and immortal reply: "But those who start st art too late lat e do not get crowned." crowned." This passage is usually cited as evidence that in that period barriers had not yet been set up in front of the runners, even though such an interpretation assumes that false starts ceased to exist once barriers came into existence. Clearly, the barriers did not prevent false starts; rather, they prevented subjective interpretations by the judges about whether a false start had occurred.
2
Introduction
8
the whole of antiquity. It is also not impossible that the sound of the salpinx was was used, not only because of its impressiveness but also for practical reasons since this was perhaps the only way for the signal to be heard above the cheers and shouts of the spectators. The great significance of the salpinx in ancient Greek games is well known. We are certain that it was used during the proclamation and awards of prizes 9 to the victors and also during the course of many competitions. A possible indication that the start of the race, at least in some times or circum stances, was given by the salpinx might come from the lively description by Heliodorus (Aeth. 4.3) of the start of a hoplitodromia at Delphi with the handsome Thessalian Theagenes Theage nes as protagon prot agonist ist under the th e gaze of the th e love-smitten Charikleia: "He spoke thus and jumped out, and going into the middle of things, he announced his name and his nationality, and was allotted his position on the track. Donning the panoply, he stepped on the balbis, panting for the race and hearing the warning blast of the 10 trumpet and hardly able to wait—a solemn and notable spectacle." It seems, however, that these means did not turn out to be effective, despite the punishment suffered by the falsestarters, with the result that problems during the course of the games threatened to weaken their prestige and popularity. And so the authorities conceived a kind of barrier that, set in front of the runners, impeded false starts since athletes could begin the race only at the moment the barrier was removed 11
Documents and Documentation of the Mechanism
Written Sources and Terminology The sources offer little information on the barriers that impeded false starts. Most that 12 doe exist come from scholiasts and lexicographers cited under the words hysplex and 13 balbis. However, there are many simple mentions of these two words throughout ancient Greek literature. We note that both words first appear in Aristophanes, who is accustomed to using metaphors and similes from the athletic world and especially from the competitions in footraces and wrestling. Balbis first appears in the Hippeis (line 1159), which was produced in 424 BC, where it is used with its literal meaning 14 since it refers to the starting place of a race between Allantopolis and Paphlagon. The word hysplex makes its earliest appearance in the Lysistrata (line 1000), which was produced in 411 BC, as a metaphor for the characteristic of unanimity, or accord, 15 which was needed in a specific project of the women. In other words, we see in the 5th century BC use of the two terms without any definition. This suggests that both terms were already well established in popular parlance. The same undifferentiated use is to be found in the later lexicographers who give us descriptions of the system. It is referred to as a hysplex by the scholiast (Eustathius?)
4
Hysplex
17
start them running." It is referred to as balbis also in Bekker: "two wooden pieces pieces for the runners, from which the cord is stretched, which is called balbis, since the 18 competitors should run thence." Finally, in the Etymologicum Magnum we read: "balbis the th e apheteria . . . is the star st artt of the footraces. They were two pieces pieces of wood wood 19 from which the cords were stretched." If we combine these definitions and leave aside the question of nomenclature for the moment, we may conclude that we have to do with a barrier that was located in front of and parallel to the apheteria and that consisted of two pieces of wood, between which was stretched a cord or a piece of wood that impeded the start of the race. The start of the race automatically and immediately followed the removal of that rope or 20 piece of wood. If we accept, then, that the same system is represented by two different words, we still need to elucidate an apparent confusion between the terms balbis and hysplex. We may note that the word balbis is used almost always only in contexts relevant to runners and never to horse races and, to be sure, always with regard to the feet of 21 the th e runne ru nners rs (e.g., (e.g., "βαλβίδα "βαλβίδα ποδός θέτε θέτε πόδ πόδα α παρ παρά ά πόδα πό δα " ) and with prepositions and 22 23 verbs t ha t mean mean "on top to p o of" f" (e.g., "έπί "έπί της βαλβ βα λβίδ ίδος ος"" or "έφεστήκει τήι βαλβίδι" ). Clear definitions also occur: "Balbides, the bases cut into for the starts, upon which 24 the runners stand so that they may have an equal stance." It is therefore proper that the word balbis is usually applied to the row of slabs that define the starting line
Documents and Documentation of the Mechanism
5
The word hysplex, in contrast, appears in regard to both footraces and horse races, a fact that favors the use of this term for some element that is indispensable both in stadia and in hippodromes. It is reasonable to hypothesize that the original concept of a kind of barrier at the start began with horses and was later adopted for runners. It might be possible to control the start of humans without a barrier but not, I 26 think, the start of animals. Moreover, since verbs that signify some motion, such as σχάζειν ("to separate") and πίπτει ν ("to fall"), are connected with the hysplex, it seems that the hysplex is a movable device, and that interpretation agrees entirely with the barrier and its characteristics as defined by the lexicographers. We have many good instances of verbs of motion, usually downward, that are connected with 27 the th e hysplex, for example, "The hysplex dropped and the track stretched out" or 28 "while the hysplex fell straight down" or "At the moment the hysplex fell I was 29 already proclaimed as having won." A final important detail is that as the hysplex fell fell it made a sudden sudd en and harsh har sh sound: "The "Th e noise noise of the th e hysplex was in our ears and 30 already someone else was being crowned." With this evidence from ancient authors as a base, we can unreservedly attribute the word hysplex by itself to some kind of barrier that was in front of the runners and that, at the moment of the start, fell or was forced down with the help of a mechanism, 31 leaving the field free to the athletes. work I have not thought it proper to discuss it fully and at length. I refer to it only where there
6
Hysplex
That the ancients themselves often felt no need to distinguish between balbis and hysplex is shown by the fact that they frequently assigned the characteristics of one to the other as, for example, in associating the verb σχάζει ν ("to separate") with the 32 33 balbis, or in using the phrase "on the hysplex" instead of "on the balbis." In any event, this thi s "confusion" should not surprise surpri se us since, as we we shall see, see, balbis and hysplex are parts of one and the same system, they are to be found close to one another, and the working of the one depends on the existence of the other. To be sure, in some cases—in later Hellenistic starting lines—balbis and hysplex are actually parts of the same construction with the result that it is not possible to distinguish which 34 parts belong to one and which to the other. It is typical that references from the lexicographers and scholiasts display the close relationship and interdependence of the two parts: "Balbis takes its name from the racers. It was a line made beneath the 35 hysplex so that the runners could take their stance upon it." Also typical is the tendency tenden cy to equate equat e the th e two ter terms, ms, as in "As "As if from from a single hysplex instead of an 36 aphesis, a balbis. In front of an apheteria was a kanon and a kampter." Here it should be noted that of the two starting lines in the stadia, one was the more important because it was the start of most of the footraces and the finish of 37 all. For this reason the words balbis and hysplex can also sometimes mean terma ("finishline"). Typical are the references of Hesychios (βαλβιδοϋχος = τερματοϋχος) and of the Roman orator L. Annaeus Cornutus, who uses the image of the end of
Documents and Documentation of the Mechanism
7
Figur e 1: Isth mia. Plan of the balbis and reconstruction of the hysplex system with the start of the runners in Stadium I (after K. Iliakis in History of the Olympic Games (Athens, 1976) figs. 67-68).
8
Hysplex
Documents and Documentation of the Mechanism
9
Hysplex
10
0
0.50
1
Figu re 6: Ist hmi a. Pla n and cross-section cross-section of th e single base of th e hysplex (the left one) excavated in Stadium II (after Broneer, pl. 96a).
proper reconstruction of the mechanism of Stadium I but also in his association of the stone blocks added to the starting line of Stadium II at Isthmia with those in other
Documents and Documentation of the Mechanism
11
Figure 7: Epid auro s. P la n and cross-se cross-section ction of the right base of the hysplex (after Kavvadias, pls. Β 6 and 6')·
example of the stadium at Epidauros with its two precisely similar constructions—for both the balbis and the hysplex —at —at each of the two ends shows that there was no difference in the method of starting the various types of footraces regardless of at 51 which end of the stadium they began.
12
Figure 8:
Hysplex
Cori nth. P la n and cross-section cross-section of the two bases bases of the hysplex (after Broneer, pl. 98).
Figu re 9:
Cor inth . Plan of the left left base of the hysplex.
Hysplex
Documents and Documentation of the Mechanism
15
Documents and Documentation of the Mechanism Mechan ism
17
Figur e 15: Nemea. Pla n of the bases of of the hysplex of the southern starting line (drawing J. Burden).
Documents and Documentation of the Mechanism
19
20
Hysplex
(9) The elimination of the central lane of the balbis can also be seen at Nemea, where there was a realignment of the posts that marked the lanes for the runners. This realignment appears to have happened after the addition of the support bases for the hysplex as is shown by the fact that the new post sockets do not extend right or left of the support bases to the ends of the balbis. The realignment must have been caused by the need to redivide the length of the balbis, which remained usable after the addition of the hysplex. Had the redivision not taken place, four lanes would have been lost and only nine runners could have participated at one time. Still, although the width of the lanes was reduced in this later phase from 1.64 to 1.30 m., the lane in the middle of the balbis (between the sixth and the seventh socket) was widened to 65 2.22 m. We have seen, then, that the central lane in the balbides of Corinth and Nemea was eliminated with the placement of the hysplex, evidently because it was needed for a central post. 66 However, in some stadia at least the central lane had already remained 67 free before the introduction of the hysplex of the type at Stadium I at Isthmia. 68 Therefore Fiechter's hypothesis that the central lane was reserved for the starter seems probable. Indeed, in some instances, as we shall see below, is the only one 69 possible.
Documents and Documentation of the Mechanism
21
on which is portrayed a starting system the existence of which had been a necessary, but ra ther a bstract bstr act,, hypothesis based based on written source sourcess and architectural archite ctural remains in ancient stadia. The vase is a Panathenaic amphora (fig. 18) that was discovered in 1970 in a salvage excavation in Athens. It became more widely known only in 1989 when it appeared appe ared in the exhibition "Mind "Mind and Body." 73 Although the inscription with the name of the archon a rchon has been lost, this vase is to be dated to th e year of the archon Lykiskos (344/3 BC) because of the representation on the main side on the columns of the main side of a seated Demeter and a standing nude Ploutos. 74 #17, who hypothesized the existence of the hysplex in two examples: (a) on a skyphos from Capua formerly in the Collection Bourguignon (now in Hillsborough, California; see Jüthner 46, fig. 8 and I.K. Raubitschek, "The Heart Hillsborough Vases (Mainz, 1969) 63 #16: close to the Pan Painter 2 (ARV 561.11) with the representation of the start of a hoplitodromos (Hauser identified the little column with the notch and two sockets at its top as a hysplex post) and (b) on a kylix then in his own collection (now in Leipzig; see Jüthner 65, fig. 16) by Pheidippos (ARV 2 165.5) where he identified a line between the feet of the athlete as a cord from a hysplex. See Jüthner 52, 65 ff. Two years later A. de Ridder, "L'hoplitodrome de Tubingue—Questions d'Agonistique," BCH 21 (1897) 211 ff., esp. 233), and later Gardiner, 263, 283, expressed fundamental objections to such an interpretation. However, Patrucco, 113 ff., continues to support the interpretation of the notch in the little column of the skyphos as intended to receive a hysplex. An abbreviated representation of a hysplex is probably to be understood in the little columns with holes at the top, exactly like that on the Hillsborough skyphos, which are present in several representations of the start of a footrace. See also Maffre 357. The representation on the Hillsborough skyphos frequently appears in scholarship particularly
24
Hysplex
75
On the second side of this vase (fig. 19) the start of the hoplitodromia is depicted. Three nude hoplitodromoi, each wearing a helmet and holding a shield, stand with the left leg slightly bent forward and the right stretched out toward the rear, ready 76 to start the race. In action and appearance, with their bodies twisted about threequarters to the right and with right hands at their sides or raised out in front, they display all the characteristics that ancient athletic theoreticians prescribed for this 77 competition. The unique detail in the depiction is that in front of the athletes are stretched two horizontal cords, one at the knees and the other at the waist, the ends of which are tied to two thin perpen pe rpendicu dicular lar posts pos ts at the th e sides of the th e scene (figs (figs.. 19-22). As the representation at the left edge (figs. 20, 22, 23), which is preserved in good condition, shows us, these posts are not driven into the ground but their bottoms are pierced by a horizontal element that is held at its two ends by cubes. Immediately next to the right post is a short column (?) the height of which is roughly equal to the waists of the athletes (figs. 21-23). We have, then, a representation of a start of the hoplitodromia with the athletes shown from the front. Until now many scenes of the start of various races were known but only with a single athlete shown from the side, with his two feet close together ("πόδα παρά πόδα"), the left foot slightly forward of the right, the knees always bent, 78 and the body leaning forward. We have here for the first time a representation of the start of a race shown from the front, that is, from a view that is not typical for
Documents and Documentation of the Mechanism
25
the end hoplitodromos. On the other side the depiction of the object is not preserved except for the outer of the two cubes. But from this cube we see yet another horizontal element that passes in front (?) of the column and ends in a third cube that is shown by means of a light incision on the black silhouette of the column. It is worth noting here that the right post has a pronounced slant to the upper right, evidently to show that it did not stand vertically but slanted slightly to the back or to the side, and most probably that its top was tied to the column that was either 79 behind or to the side of it. This detail seems to have had special significance since at this point, and even though it was covered by the shield, the vase painter indicated with great care that the post ended very near to the column. Prom this representation it follows that, in order for there to have been need to tie the upper end of the post, there must have existed at the bottom torsion that, when released, would throw the post to the ground. We can make the following observations with regard to the means by which torsion could have been produced in that era. Based only on the depiction of the vase we could reconstruct a very simple mechanism that could work well enough to fulfill the need for a fast fall of the two vertical posts that hold the horizontal cords. This mechanism (fig. 24) could have consisted of a thin rectangular bronze strip the ends of which would have been held in a horizontal position a little above the ground by two wooden
Documents and Documentation of the Mechanism
29
30
Hysplex
Another exceptionally interesting detail is that the vase painter, aside from the two thin posts at the ends, had included in his preliminary sketch (figs. 22, 23) three more thin vertical objects (two between the runners and the third just next to the left post) that he did not depict in the final painting. That these details were placed exactly between the runners shows that they were intended to depict the rectangular wooden posts that stand in the balbis as is indicated by the corresponding sockets that are preserved at regular intervals in the balbides of nearly every ancient Greek stadi um. 84 These posts were once thought to be a part of a starting system corresponding to Stadium I at Isthmia, but more recently they have been assigned various roles. I would agree with those who believe that their basic role was to demarcate the place of each athlete on the balbis, although it is not impossible in middle-distance races, like the diaulos and the hoplitodromia (two stadia in length), that they should be identified with the kampteres, that is, with the point at which the runners made a turn of 180° in order to return to the apheteria, which was also the terma.85 Finally, I think it quite likely likely that th at these wooden wooden posts of the balbis played one more role: they served as the starting point for cords that marked out lanes in the stadium, which 86 must have continued for the whole length of the track to show each runner his path. 84. Since these represent the posts of the balbis, the farthest to the left, the end of which is preserved (on the other two the end is hidden by the shields of the runners), could give us an idea of their height. Thus, while the hysplex posts reach about waist height, the posts of the balbis were a little higher and reached the middle of the chest. Compare the reconstructed drawing in Nemea: A Guide 181, fig. 64 and the opinions of ft. Patrucco, Lo Stadio di Epidauro (1976) 109 109 ff ff. For rep re sen tat ion s of
Documents and Documentation of the Mechanism
31
Figure Figu re 25: Schemat ic reco nstr uction uct ion of the whole hysplex based on the representation of half of it on the Panathenaic amphora (#A6374).
Thanks to the representation on this vase of so many details of both the balbis and the hysplex we can establish a nearly complete agreement between the painters' depiction and the excavated remains, which show that the balbis remained a necessary part of the system even after the placement of the added bases for the hysplex. The placement of the hysplex, in other words, did not make the posts of the balbis obsolete, as would have been the case had the latter some connection with a starting gate as 87 was previously believed. This conclusion agrees with the excavated evidence from Nemea, where not only are the posts of the balbis retained after the addition of the 88 hysplex, but their arrangement is somewhat altered. The posts of the balbis and the posts of the hysplex, then, make up two different parts of the same system. That the vase painter did not actually show those balbis posts could be a result of his increasing awareness during the course of the painting that while he had originally
32
Hysplex
παραστάδας ύσπλήγων τέτταρα[ς] και κίονας δύο. σύριγγας των ύσπλήγων δύο." In the second inscription we read "ϋσπληγος άνκώνας III. [παρα]στάδας ύσπλήγων ΠΠ." In the first text, four parts are mentioned: three ankones, four parastades, two columns, and two syringes. In the second, only the first two appear: three ankones and four parastades. It appears, then, that the two catalogs contain inventories of the same objects by different, perhaps successive, committees or officials responsible for this task. This conclusion is reinforced by the fact that the other objects that are mentioned in the two catalogs are similar but written in a different order. It is also telling that in the second inscription, aside from the two parts of the hysplex (two columns and two syringes), there are missing other dedications, supplies, and tools 91 that are included in the first inscription. Until now it has not been possible to make use of the Delian inscriptions, because we were completely completel y ignorant igno rant of the form of the th e mechanism. mechani sm. Of all the th e mentioned mention ed parts, only the ankones were associated with the horizontal barriers of the hysplex of 92 Stadium I at Isthmia, while the syringes have been interpreted as tubes through which the cords passed and with which the starter made the start, by analogy once 93 again with the staples nailed in the paving of the system in Stadium I at Isthmia. Now, however, we have as a starting point for our understanding and use of these inscriptions, as well as for the final, total reconstruction of the system and its means
Documents and Documentation of the Mechanism
33
produce torsion by means of a system of twisting together sinews, which were released 95 to give motion to a vertical arm. Let us now return to the study of the inscriptions themselves. The initial question that comes to mind is why the word hysplex is used in both the singular and the plural. I think this can be explained only if we note that the three ankones are "of "of the hysplex"; that is, they belong to the whole starting system considered as a single unit. The four parastades, the two columns, and the two syringes are "of "of the th e hyspleges," as is shown by their even number, and they belong to the two separate mechanisms that make up the hysplex; that is, they belong to the two vertical posts at the opposite ends of the balbis with the elements that they have below and 96 next to or behind them as they appear on the vase. This, together with the number of parts that are mentioned in the inscriptions and the meanings of these terms in ancient literature, especially with regard to military machines, allows us to make the following observations about the significance of these elements for the hysplex: (1) (1) Ankones (Αγκώνες): The only elements that belong to the whole hysplex and can be three in number and that actually are three are the vertical posts, as we saw in Corinth and on the Panathenaic amphora and as we will see frequently below. That is, there is one on each end and one in the middle of the balbis. We should not overlook the fact that there are mechanisms only at the two ends; the middle post was only to help support the horizontal cords. It is most likely, then, that we have three posts in
34
Hysplex
Documents and Documentation Documentation of the Mechanism
35
(4) Parastade (Παραστάς): None of the known meanings of this word has a close resemblance to any of the parts of the hysplex that we have attempted to reconstruct: neither the very specific meaning the word has in an architectural context nor the very 104 general "anything that stands beside." The related word παραστάτης was used in stone-throwing machines for the rectangular wood frame of the sinew stretchers to indicate the exterior wood of this frame that the ancients called πλαίσια (frames) or 105 πλινθία (plinth; fig. 26). For example, two external παραστάτες and two internal 106 μεσοστάτες were necessary for the frame of the euthytonon. We should imagine such wood frames for our system as well, because they were needed to hold the neura and the ankon. These wood frames would have been wedged into the support bases of th e hysplex, as the trapezoidal (in section) cuttings in many of the bases have shown us. Further, as the use of the number four indicates, it is possible that the two basic parts of the wood frames for each of the two hyspleges were called parastades by 107 analogy with the parastates of stone-throwing machines. In concluding this section I would like to note also that I thought it a good work ing hypothesis that the nomenclature of the hyspleges of Delos probably referred to a stadium hysplex, of a type more or less like that which is depicted on the Panathenaic amphora and which was supported on the bases we have seen in various stadia. How ever, since we do not know the form of the hysplex of the stadium or the gymnasion of
36
Hysplex
38
Hysplex
understandable if we consider its advantages, especially the ease with which it can be worked to fit neatly and join securely with the other materials that made up the system (i.e., stone, cord, etc.). The placement of wooden elements in the stone base was due to the need to set up the whole system only during the course of the games; when not in use it was removed and stored in some closed area, as we have seen in 111 Delos. And this occurred not only so that the perishable parts of the mechanism would not be destroyed by rain, freezing, and so on, but also—as also occurred with catapults—so that proper servicing could take place: the wooden parts and sinews 112 could be separated and placed in special oil so that they would not become brittle. The base of the hysplex was a rectangular wood frame that was wedged into the corresponding cutting in the stone support bases in the stadia. The need for wedging is obvious: the torsion and quivering caused by the sinews and the corresponding movement of the ankones brought the danger that the whole system might come loose from the ground. This wedging was helped, as we have seen at Nemea and Corinth, by the trapezoidal (in section) undercutting in the interior sides of the large cutting 113 and the smaller exterior ones that we have seen in the bases of Corinth. It is obvious that no wood base could be inserted as a complete unit in the stone base intended to receive it but had to be inserted in pieces and with the same prin 114 ciples as that of the lewis used to lift stone blocks. Given this principle, as well as
Documents and Documentation of the Mechanism
39
40
Figu re 30: machines.
Hysplex
Drawing showing te rms and mat eri als of the basic elements of ancient stone -throwi ng
Documents and Documentation of the Mechanism
41
Figure 31: Schematic drawing of the neura and the metallic elements that made up the moving parts of the hysplex.
as we call them], katazygides, and thus the tonos, wrapped around these, is stretched 122 tight by the whole frame." This is the only way to ensure that the system does not malfunction—that the sinews do not rub on the wood and thus provoke a loss 122 of torsion in and damage to the neura. The Th e neura probably would not have been visible at all, since it was hidden in the horizontal cylindrical casing in the interior of the wooden frame. Perhaps it could be discerned a little at the middle of the frame, in the opening where the heel of the ankon was fixed (figs. 27-28), which thus could receive the torsion produced by the twisted neura. The form of the elements that produced the torsion and were hidden in the frame of the hysplex were as shown in the analytical analyti cal drawing (fig fig. 31). The basic basic shape shap e of the th e hysplex, then, must have resembled resembled clos closely ely the th e frame frame of of the th e huge catapu cat apult lt (onager) that is described clearly by
42
Hysplex
difference in the hysplex, however: the neura was not twisted along its whole length from one end to the other but only from the exterior side to the middle, that is, to the point where the ankon passed through. From that point to the other end the sinews were straight and parallel to one another. That happened because in our case, and unlike the case of the stone-throwing machines, the act of twisting the sinews happened only from one side where it was necessary to have the twisting element, as we shall see below. The ankon in the middle stopped the twisting from extending the whole length of the neura, and this then left half without a twist. But by pre-twisting t h e neura before inserting the ankon, the latter's certain position at the center of the 1 26 hysplex could be preserved. Th e neura first was set in the wood frame without torsion, which was introduced by means of twisting. This must have been produced by a system that evidently was located in the narrow channel of the stone support base that we have seen in all the stadia and in the rectangular cutting on the outside of the base that we have seen only in Corinth. This hypothesis is reinforced by the elongated projection that exists from the second to the third cube of the vase painting and from the reference 127 in the Delian inscriptions to the two syringes that we have interpreted as pipes. This hypothesis is further supported by references to military machines in texts. For example: "When the ankones are passed through the middle of the tonoi, it is necessary to turn the choinikides with an iron mochlos that has a socket, into which is inserted
Documents and Documentation Documentation of the Mechanism Mechanism
43
44
Hysplex
As the cuttings in the support bases that we have seen in the various stadia show, and especially the trapezoidal (in section) external cubic sockets in the two Corinthian 1 34 bases, it was not possible for the syrinx and the crank to be in contact with a stone casing, but these elements were also encased in a wooden housing that was wedged into the corresponding cuttings of the stone and was, of course, connected with the wood frame on the base that we have already discussed (figs. 27, 28). All of these wooden parts could not have been entirely hidden in the stone bases but must have projected enough to allow sufficient internal space for the workings of the various parts of the mechanism. Finally, it should be noted that all the parts of the mechanism, as well as all the places where wooden parts came into contact with parts of some other material (stone, cord, etc.), must have been greased so that would be avoided and the mechanism would work smoothly. Aside from these basic parts of the mechanism proper, fixed and stable columns or posts to hold the taut ankon when the hysplex was upright were also necessary (figs. 26, 27). The existence of this element is confirmed by the vase painting, the Delian inscriptions, and the remains of the stadia. In Corinth there are deep rectangular sockets next to the whole system; in the other cases there are semicircular sockets in th e balbis, that is, set behind the system. It is probable that the top of the ankon was
Documents and Documentation of the Mechanism
45
Figure 33: Cori nth. Pl an of the two bases of the hysplex showing the brackets that would create the cord barriers in an application like the Nemea system.
48
Hysplex
at Isthmia, Nemea, and Epidauros, where the column is directly behind the vertical ankon. But in the case of Corinth, the cuttings for the fixed columns—wooden posts here—are to the side and at a distance from the ankones and more or less in the same line as they are (fig. 33). A possible solution for this instance might be the placement of a post of a height at least equal to that of the ankon, on the top of which was a light horizontal board. This board would have held the ankon in its vertical position and would have released it by being lifted slightly by a cord that was passed through a ring at the top of the post. But this solution provokes two problems: (a) the post would have to have been sufficiently higher than the ankon so as to allow the aphetes to remove easily the horizontal board that released the ankon, and (b) since the posts are a little in front of the line of the two ankones, there is the danger that the two horizontal barrier cords might touch the posts and thus form slight bends in their otherwise straight line (fig. 33). 13 9 Thus we are led to another solution that avoids these two problems. We place in the deep cutting a very low post (fig. 34) from the top of which a light board extends diagonally upward so as to hold the ankon and still be easily removed by the aphetes. At the same time the low post does not block the horizontal barrier cords, which, as we have seen, were at heights of about 0.60 and 1.00 m. from the ground. Given, then, that the release of the ankones was effected by the removal of the rings
Documents Documents and Documentation Documentation of the Mechanism
49
an unfair advantage, or disadvantage, for a given athlete is the existence of the neura, which, like a spring, produced the immediate and fast fall of the barrier, unlike the system of Stadium I at Isthmia where the fall was produced only by the weight of the 14 1 barrier—the horizontal wooden arm—itself. Finally, it is clear that whenever they wanted to replace the ankones in their original position, they lifted them by hand until the top touched the column, which existed in most stadia, where they were attached by means of a ring or a loop. Correspondingly in the case at Corinth, the ankon would have been held by the diagonal light board hypothesized above (fig. 34). This lifting of the ankon by hand perhaps is relevant to the passage—which has been difficult to explain heretofore—in the "Alexandra" of Lykophron, line 21: "κ' από γης έσχάζουσαν υσπληγας" which is used metaphorically 142 for the difficulty with which a ship is freed from a rocky shore. It seems probable that the poet wanted to compare the strained and slow but sure movement of the rowers pulling on their oars with the corresponding action of those who lifted the 143 ankon of the hysplex from the ground. This action must have occurred often and in a short time, for example, in the successive preliminary heats of the stadion as well as in the other footraces when these were run all together and in succession on a specific 144 day of the various games.
50
Hysplex
the start of horses 14 6 —or, more likely, that they refer to an older type of hysplex, as that of Stadium I at Isthmia, which, based on archaeological evidence, is to be dated in the second half of the 5th century BC and which perhaps should not be thought, as 14 7 some scholars hypothesize, to have been a unique example. It should also be noted that the form of the hysplex that is shown on the Panathenaic amphora can be associated with physical remains in four stadia of the northeastern Peloponnesos. Of course, the support bases for the hysplex located in those stadia cannot be dated precisely. But we do know that they are additions and therefore later than the balbis, thus providing us with a terminus post quem on the assumption that th e balbides can be dated, albeit not always with accuracy. Generally there has been a trend in the scholarship, "although still unproven," which is based on the observations of Broneer, to date the addition of the hyspleges to the preexisting balbides in the 3rd century BC "or later." 1 4 8 If we examine closely the relevant facts of each monument we will see that the securest chronological evidence for the placement of the type of hysplex discussed here comes from Nemea, where all the construction work of this phase of the stadium is to be dated in the period from ca. 330 to ca. 270 BC. 14? In Corinth the balbis of this phase 150 was set in place after 280 BC and remained in use until 146 BC. We have a more general chronological framework for Stadium II at Isthmia, which appears to have
Documents and Documentation of the Mechanism
51
Thus, since no work took place at Nemea after ca. 270 BC, the hysplex must already have been in place by that time. Again, in Corinth the new balbis was not put in place before 280 BC. These chronological facts show that the first installation of the hysplex in the stadia of this region must have taken place late in the first third of the 3rd century BC. Likewise, the pronounced similarities in the various hysplex bases, and therefore in the machines that once rested in them, show that their dates cannot 153 be far apart. Broneer hypothesized that Philon himself had probably installed the hysplex in his native Corinth before doing that at Epidauros, and it seems he was correct since the preserved details of the bases at Epidauros resemble closely those of the right base at Corinth. Broneer's hypothesis is supported by the fact that the mechanism at Corinth seems to be less developed than the others since the support bases of the hysplex are set at the ends of the balbis, a detail that is not repeated 154 at any other stadium. But the excavation data would place it a little later than Nemea; one might hesitate to assign such precise dates in this period. With regard to Athens, however, based on the new information of the vase, we can safely raise the date of introduction of the new mechanism to the middle of the 4th century BC. Direct evidence for the existence of a hysplex in the Panathenaic stadium—but from the 2nd century BC—comes from an inscription from the Acropolis in which, together with other repairs to sanctuaries and public buildings of the city, 155 mention is made of "starts from the hyspleges of the Panathenaic stadium." In
52
Hysplex
With regard to the older type of hysplex (i.e., that of Stadium I at Isthmia), and in addition to the substantial differences in the working of the mechanisms that we have already seen, our hysplex has yet another difference. While the barriers at Isthmia I were located above the balbis, in the new type they stand on other blocks that project from the balbis, with the result that the barrier is some 0.50-0.60 m. in front of the balbis. Again, these balbides, and especially the type with the double parallel groove and the rectangular cuttings at regular intervals, were placed in their stadia at the same 159. time as or a little earlier than the hyspleges. This nearly synchronous placement brings us to the notion that perhaps the one provoked the other; that is to say, perhaps the new type of balbis provoked the invention of the new hysplex, or vice versa. We should remember that just as the barrier cord of the hysplex obliged all the athletes to stand in the same line, so the double grooves of the balbis forced their feet into similar positions, with the same distance between the front and rear foot of each. Together cords and grooves reinforced a complete equality at the start of the race, and the victory would belong to the athlete whose actual running of the race on the track was the best. A straight balbis without continuous grooves and sockets (fig. 4) was constructed in the Forum of Corinth after 280 BC; it has exactly the same characteristics as the version 159 that is roughly two hundred years older (although the earlier balbis was curved). That the new type of balbis with the two parallel grooves and the sockets for posts was
Documents Documents and Documentation Documentation of the Mechanism Mechanism
53
the th e balbis, the new type of hysplex followed the placement of the new balbis with two l6 2 grooves.
Representations of a Hysplex in Monuments of a Roman Date Two monuments of Roman Imperial date present portrayals of a type of hysplex like that with which we are concerned. The first of these is in the Codex Vaticano Latino 3439 (formerly the Codex Coburgensis), which was written in the middle of the sixteenth century after Christ, but the picture that interests us here was copied, as is sta ted te d explicitly explicitly in the margin, from from a now now lost lost relie relieff th at then t hen was "ret "r etro ro Belvedere" 163 (fig. 36). The representation shows four runners, two beardless and two bearded, with the "cirrus in vertice" hairdo typical of the professional athlete of this period standing in front of a hysplex that is shown here in a simplified form as a Π, with two vertical posts holding a horizontal round stick. It is interesting that the two end athletes hold the horizontal stick with one hand and with the palm up. On the right side of the representation there is preserved the right side of a himation- wearing figure 164 with woven sandals, apparently some official of the games, and on the left there is 165 a herm.
54
Hysplex
Documents and Documentation of the Mechanism
Figur e 37:
55
Mosaic at Gafsa (after M. Khano ussi , AntW 1991, figs. 5, 6, 13).
17 2 part (fig. 38). We see here depictions of a hysplex just like that in the codex, which leaves no doubt about the interpretation of the representation in the codex that we
56
Hysplex
Observations on the Starting Lines of Various Stadia
From the data that we have studied to this point and from the conclusions that we have reached concerning the shape and the operation of the Classical hysplex, we may now proceed to observations about some characteristics in the starting lines in other stadia that seem to be relevant to the hysplex mechanism.
Mainland Greek Stadia OLYMPIA
Two constructions were added at the ends of the last phase of the western balbis. These project and make a right angle with the balbis, and they have created problems for the excavators of the most famous stadium of the ancient world. 17 6 At one end this addition consists of a brick foundation about 1.50 m. long (fig. 39) that continues the line of the balbis southward. In front of the end of the brick construction there has been placed upside down, apparently in a second or third use, a stone plinth (block y)
Observations on the Starting Lines of Various Stadia 59
stadium that we see today (phase HIB or IIIC, recently dated by Schilbach to 367/618 2 360 BC and 340-330 BC, respectively) there are semicircular cuttings that, in their dimensions and their shape, resemble the corresponding cuttings of other balbides, to which were added a hysplex and which we have interpreted as intended to receive the 193 fixed column of the hysplex. Such are the cuttings in blocks c and i of the eastern 184 balbis (fig. 40). It is worth noting that block c of the eastern balbis corresponds to the position of the second and next to last runner—that is, they correspond to the place where the support bases of the hysplex were added to the balbis at Nemea and in Stadium II at Isthmia. Moreover, the cutting in block c measures about 0.30 x 185 0.09 m., which is about the same as that at Isthmia. Also, in the projections added to the west balbis, which as already noted belong to 186 the final phase of the stadium, there are indications of their use. The end of the northern projecting wall and block y of the southern end (with its cutting for the insertion of a wooden post) are on exactly the same line, which is 0.40 m. away from and parallel to the balbis. This is exactly the same distance between the ankon and the 187 first groove of the balbis at Nemea. All these details provide very strong indications that there were starting mechanisms in the stadium at Olympia in all its phases from the 4th century BC and later. This does not preclude, of course, the possibility that mechanisms existed in even earlier phases together with the balbides no traces of which 18 8 survive.
60
Hysplex
Observations on the Starting Lines of Various Stadia
6l
193
to little columns or little pillars that bear inscriptions including that of ΤΕΡΜΩΝ. It is believable that these representations of little columns have a metaphorical sig 194 nificance, as J.M. Moret has suggested; that is, these are not depictions of real pillars or columns but imaginary columns used as a means of showing the inscriptions. Despite this, however, I believe that in the back of his mind the vase painter saw the corresponding real columns of athletic places, especially the posts of the balbis, that constituted the starting place of the race as we know from the remains of balbides and an d from the customary presence of such elements in scenes of the start of races. Now if in some cases the sense of the word ΤΕΡΜΩΝ is that of the end of life and is to be connected with the eschatological character that the vase painter wanted to give to all these scenes, then what more appropriate iconographie symbol could he use to show the th e end of life life of men th a n the th e end of the th e "footr "fo otrace ace"" (αγώνα (αγών α δρόμου) of life life?? We see then that in a series of representations of the finish, the terma is indicated by an element that also exists at the start. Thus we are led to the conclusion that the finish did not have some especially characteristic mark that was to be found in the line of the balbis-hysplex. For this reason, when the vase painters wanted to show the finish, they adopted something from the illustrative elements that they used also for 195 the start. This conclusion is reinforced also by the archaeological evidence. In the stadium at
62
Hysplex
The only indication that the balbis of the stadium at Delphi provides for the place ment of a hysplex is the reduction before the middle of the 3rd century BC of the positions for runners from 20 to 17, which probably was caused by the placement of the three bases of the hysplex at the two ends and in the middle of the balbis in the 199 same way that we saw in the Peloponnesian stadia. Unfortunately, nothing was discovered in situ and the stones that were used as the base of the system do not bear any of the elements that we have seen in the bases of the late Classical hysplex. Thus the representation of the hysplex in his plate 22 is, as Aupert himself admits, a simple 20 0 conjecture. Nonetheless, the lack of adequate evidence cannot in any way be taken to mean that there was no starting mechanism at Delphi. The principles and the practices of the ancients indicate that such an element was present in every stadium, especially in 20 1 one of the most significant stadia of the ancient world. It is, moreover, very possible that no trace of the hysplex would remain because, as we have seen, the support bases of the system were surface constructions, only one block deep, without foundations, at least in the pre-Hellenistic period. It is also not impossible that in some stadia these bases, since they were additions to the balbis, were thought to be added in later antiquity and that the purist tendencies of the original excavators did not leave them in place.
Observations on the Starting Lines Lines of Various Stadia
63
Certainly the construction of such completely portable systems was possible only after a probable change of the means of effecting the torsion in the support bases of the hysplex. This change must have occurred with the replacement of the bulky and heavy neura by a smaller and much lighter but equally effective system made from bronze 20 6 strips as happened around the middle of the 3rd century BC with war machines. THE ATHENIAN AGORA
In the early 1970s, on the north side of the Athenian Agora, immediately northeast of the Altar of the Twelve Gods, there was uncovered a row of five limestone bases with dimensions of 0.47 χ 0.38 m., each of which bears a square cutting, 0.12 m. on a side, on its top surface. These bases are set in a straight line with a distance of 1.38 m. from base to base (or 1.85 m. on centers). The race that we know to have occurred in 207 the Agora was that for the epheboi of the ten Attic tribes. Based on these remains, there has been drawn a starting line with ten such bases, as well as a "dromos" with a length of 600 feet, that split the Agora from the northwest to the southeast and 208 reached the east end of South Stoa I. The use of this starting line, as revealed by 20 9 excavation data, lasted only from the middle until the end of the 5th century BC. In any case, these bases—as they appear and by themselves—cannot have worked 210 as a proper starting line. In stadia and tracks where there was no continuous balbis,
64
Hysplex
211
"obviously" been removed. This pit surely had something to do with the starting line because it is precisely in the same line as the five bases and at the same distance from the last of them. For this reason it has been interpreted as the probable position 212 of a kampter. However, the pit is very large for such a use and, given its perfectly circular form and well-preserved edges—which would not be expected if something had been pried from it—we might well suppose that it formed a part of the system and was not for the receipt of a base. Based on these hypothetical details, another possible use for it might have been as follows. In many cases we have already noted that the central position on the starting line was wider than the rest and that this probably had to do with the presence there of the aphetes, who checked the position of the runners and who started the race. Here, then, I would suggest that the pit was the position for the aphetes who could, from this place, not only have looked left and right at the runners but also have held a horizontal cord that, connected to the tops of the wooden posts, could have made up some type of hysplex. We have seen such a 213 pit—albeit behind the runners—in Stadium I at Isthmia, and we will also see below the example at Rhodes where the well-like construction for the aphetes has a similar 214 diameter of 1.20 m.
Stadia of Asia Minor and the Islands
Observations on the Starting Lines of Various Stadia
65
measuring 0.28 m. on a side. Although they have been disturbed, it appears that these blocks were originally placed in two parallel rows: the first row consisted of the six small blocks placed in two groups of three with a canonical width left between them for the runners and a much larger space in the middle between the two groups. At a distance of 0.57 m. in front of this row of smaller blocks was placed the row of larger blocks, one well beyond each end of the row of small blocks and one in the central larger space. The excavators and first scholars who studied the area interpreted correctly the working of the two types of bases. They theorized that the small blocks composed the th e balbis and were intended only for the posts that separated the runners and the large blocks placed 0.57 m. in front were remains of the hysplex and were used for the insertion of "special posts between the top of which there is to be assumed a single, cord 216 barri barrier er was was str et che d. " We see, then, that here too were all the requisite items for the existence of a hysplex with three ankones as we have seen in Corinth and in the Delian inscriptions. A further interesting element is the existence in the central large base of a horizontal hole, rectangular in section ("rechteckige Durchbohrung"), that measures 0.10 χ 0.09 m. This hole pierces the right (south) side of the block and extends to the rectangular vertical cutting. This hole led the excavators to place the operating 217 mechanism of the system in the central base. The existence of the mechanism in the central base alone, however, would force us to replace the horizontal cord barrier
Observations on the Starting Lines of Various Stadia
Ο
5
10
67
15M.
Figure 43: Prien e. Pl an of the bases of th e firs firstt phase of the sta rti ng line (after (after Wiegand and Schrader , fig. 264).
having undergone the appropriate hammering, could be pre stressed and could hold their torsion for a long time. With this new system, which was much cheaper and less bulky bulky than th an th e earlier one, one, "the "t he heel heel of the ankon had been riveted on the lepides," 219 th a t "having been divided with great gr eat force, force, lifte lifted d the heel of the th e ankon." It seems that this new discovery was not installed in the large machines since the gigantic
68
Hysplex
Figure Figur e 44: 44: Miletos. State St ate plan and separated plan pla n of the bases of the first phase of the start st arting ing line at the eastern end of the stadium (after von Gerkan, pl. 5c).
MILETOS (STADIUM I
Observations on the Starting Lines of Various Stadia
69
Figure 45: Miletos. Cross-sections Cross-sections thr oug h the bases bases of the star tin g lines lines at the easter n end of the stadium (after von Gerkan, pl. 6).
The publication of the stadium at Miletus by von Gerkan suggested a probable reconstruction of the system with the placement on the three large bases of the frames for the hysplex: ... it is then probable that higher posts, but with a different shape, stood in the middle and at the ends of the barrier, as is to be concluded from the sockets that are larger and deeper than any others, and that the backsides are cut so that the ends and the middle of the bar could be lifted with the help of strings to the desired height. The way the strings worked can not
Hysplex
70
0
Figu re 46: 46:
5
10 M.
Mileto s. General schematic reconst ructi on of of the first first phase of the balbis an d hysplex.
since only the central one has provision for a release (by means of the lower horizontal cutting) is one I cannot resolve. Finally, we should note that a line drawn along the front of the smaller post sockets of the balbis aligns with the center of the three "column" cuttings of the larger bases, and this would have hindered the fall of the wooden bars of the hysplex, since they would have struck the wooden posts of the balbis. A possible alternative would be to restore an ankon in each of the three bases with a frame supporting a metallic lepis as the agent of throwing the ankon to the ground as suggested above for Didyma (fig. 42). The barrier must have been a cord fastened to
Observations on the Starting Lines of Various Stadia
71
the Hellenistic hyspleges, regardless of whether it is contemporary with or even later than some of the latter. Now the small bases for the posts of the balbis and the three large bases of the hysplex disappear with the rise in ground level of about 0.15 m. The large bases, however, are replaced by other similar bases that are set immediately in front of the 22 7 older examples. These new bases consist of very short columns in a second use that have been cut on their back sides (and were not made originally as half-columns), with a height of 0.95-1.25 m. that rest on foundations of different heights but whose upper surfaces are at the same level above ground (fig. 45). On their upper surfaces are rectan rec tangul gular ar cutting cutt ingss 0.35 0.35 χ 0.44 0.44 m. with a depth dep th of 0.17 0.17 m. m. The Th e two end columns columns have on their back flat side a vertical channel that is 0.04 m. square in section; the central column has in a similar position and arrangement two cuttings, with lengths of 0.10 m. for one and 0.035 m. for the other. Finally, at the upper rear corner of the foundations of the outside columns are horizontal cuttings, rectangular in section, that von Gerkan believed were sockets for the receipt of wooden boards that bordered the edge of the track. The existence of three vertical members but with a channel only in the two end ones indicates, at first glance, that we still have here the use of a single barrier for all the runners that was operated only from the two ends. The dimensions of the cuttings in the top of the columns, which are similar to those of the previous phase of the same
72
Hysplex
O
5
10 M
Figure Figu re 47: Miletos. Gene ral schema tic recon struc tion of the second second phase of th e hysplex starting mechanism.
metallic bar barriers and for releasing those bars to fall to the ground, allowing the runners to start. Secondly, the epistyle would have provided a means of support for the center bar, thus eliminating the problem of sag. The outside ends of the bars would have been raised by cords running through the vertical cuttings on the back of the outside columns while the inside ends of the bars were supported by cords suspended on the sides of the central column as can be seen in the drawing (fig. 47). To avoid
Observations on the Starting Lines of Various Stadia
73
Unfortunately the stadium at Rhodes (figs. 48-50) remains completely unknown in scholarly research because its excavation by the Italians has not been published and 230 the records of its subsequent reconstruction have been lost. The information we have at our disposal and on which this study is based, in large part comes essentially from photographs and drawings that were made about 1972 or 1973 in connection with excavation research and cleaning and that were graciously provided to me by the Ephoreia of Antiquities in Rhodes. Despite the meager information available in the bibliography, I will attempt to make use of the available data so as to reconstruct the phases of the balbides and the hyspleges in the stadium at Rhodes. The chronology given is especially to be treated with caution. (1) The older remains were uncovered in a pit a little south of the northern starting line. They consist of two poros limestone constructions (figs. 51E, 52, 53). One of these is formed of two adjacent blocks, of roughly the size of the bases of the Classical hysplex. Their interiors are hollowed out to form a place for the insertion of a cylindri cal, probably wooden, element that most likely held the mechanism. The horizontal cylindrical cutting in one of the two stones contributes to this interpretation, for it is an indication of the insertion of some horizontal element like the neura of the Clas sical hysplex. The same role must have been played by the slightly later (?) second poros construction (fig. 53) with a large but shallow cutting the rim of which is open only toward one short end, and that in the same direction as the ankon of the earlier
Figure 48:
Rhodes. Rhodes . Plan of the th e stadiu sta dium m (based (based on the plan of S. lakovides in Ρ Mvlonas ΠεοΙ '
σταδίων (1952) pl. 47).
Figure 52: Rho des . Rem ain s of the left left base of the hysplex of the first phase, from the southwest, probably 1973.
Observations on the Starting Lines of Various Stadia
79
for the feet of the runners like those we have seen in the racecourse at Corinth and also some system of hysplex at its ends the mechanism of which must not have been very different from that of the Classical hysplex. This location for the stadium and its starting system must have been used until the great earthquake of 227 BC. (2) In the following years, and probably in the 2nd century BC, the construction of the new stadium must have taken place in the location where it survives today. It represents a part of the basic changes of the architecture of the city during this period 23 3 which occurred simultaneously with the general remodeling of the akropolis. The Th e new stadium was constructed on the eastern side of the akropolis parallel to the retaining walls of the latter. This followed on a large-scale cut into and removal from the akropolis of large quantities of earth, that were subsequently used for the eastern embankment of the stadium. This, in turn, was used to support the seats on this side of the track. This location had benefits beyond the obvious practical ones, for it 234 complied with basic principles of Rhodian public architectural and town planning. Remains of the bases for the hyspleges of this phase are preserved at the two nar row ends of the stadium, but the balbis cannot be described accurately without new excavations (figs. 48B, Γ). On the northern side, at the entrance, is preserved a series of limestone slabs that make up a section of the balbis and probably belong to the next phase (figs. 55, 56). At the western end of this series of slabs is still preserved a construction consisting of four blocks reused (probably from the earlier balbis) in this
80
Hysplex
84
Hysplex
Observations on the Starting Lines Lines of Various Stadia
87
Near each of the two ends of the starting line are a group of four or five poros blocks that form the foundation for a larger base (fig. 61). The face of these two foundations is 0.30-0.45 m. in front of the face of the row of individual projecting blocks but is in line with the face of the large central base. We have here, then, a starting system that consists of three large bases in between which extends a continuous row of blocks, and in front of the latter are placed a total of 14 smaller blocks. Not in situ but at a short distance away lie two square blocks of hard ha rd gray limestone limes tone that have on their upper surface dimensions of 0.4 0.41 1 χ 0.44 0.44 and an d 0.57 0.575 5 χ 0.50 0.50 m. and an d a rectangul rect angular ar deep cutti cut ting ng measuring measu ring 0.11 0.115 5 χ 0.115 0.115 and an d 0.115 0.115 χ 0.10 m., respectively (figs. 56, 65). In other words, they have the same characteristics as similar bases that held the wooden posts that divided the positions of the runners as we have seen in other stadia in Asia Minor and in the Athenian Agora. On the basis, then, of the the data from the stadium of Rhodes, the corresponding elements that we can provide from similar constructions and especially that of Sta 24 2 dium I at Isthmia for the operating system and that of Stadium I at Miletos for the 24 3 starting system, together with what we know of the practices of the ancients, we might reconstruct the mechanism of the hysplex at the stadium of Rhodes as follows (fig. (fig. 66). On top of the long continuous foundations of the start st arting ing line there should be stone slabs of marble like that of the channel and of the well-like construction. These slabs would have worked as a continuous balbis like those we saw in mainland
Observations on the Starting Starting Lines Lines of Various Stadia
91
As we have said, in the third phase of the Rhodian stadium, balbis and hysplex are securely documented only on the end toward the sphendone. On the other, entrance, end of the stadium, at the north, one can identify with certainty only a row of slabs for the foundations of the balbis (see (see above p. 79), alt a lthou hough gh the existenc exi stence e of bases for for the hysplex is not impossible (see n. 233). But the existence of a channel and operating pit should be considered impossible since the monumental propylon was placed east of the balbis during this period. In general, space is not available for the operation of a system from behind since the balbis is exactly at the entrance of the stadium (fig. 48). And whatever might have existed cannot be compared with the system at the southern end. However, at the same, northern, end, west of the balbis and on the long axis of the stadium there was constructed during this same period a channel constructed of limestone limeston e slabs wi th a length lengt h of 12.6 12.60 0 m. and a nd a dep depth th of 0.07 0.07 m. (figs. (figs. 48 Δ, 67). A characteristic of this channel is the trapezoidal cross-section of the interior sides and widening toward the southern end. These details, as well as its location precisely on the long axis of the stadium, indicate that the channel was used for the insertion of a parapet, probably of wood, that must have resembled a shortened spina of the 248 Roman circus. This interpretation is also reinforced by the widening of the channel toward its southern end, as well as the many sockets around it that apparently would 249 have received and supported some taller element like the meta in the circus. It is
Observations bservations on the Starting Starting Lines of Various Stadia
93
Figure Fig ure 66: Rhodes. Schematic Schematic reconstructi on of the southern starti sta rting ng line of the thir th ir d p phase hase of the stadium.
Observations on the Starting Lines Lines of Various Stadia
95
an inscription on the base of a dedication of the 2nd century after Christ where the names of various religious associations are written. Implicit in the name of the last association associat ion "of those who who worked together together on the hippo hi ppodro drome" me" (τώ (τών συνεργαξα συνεργαξαμένω μένων ν 255 τον ίππόδρομον) might be the construction or reconstruction of the hippodrome.
Monumental Hyspleges Of Later Helle Hellenistic nistic Date Beginning in the 2nd century BC impressive monumental installations were created for the balbis and the hysplex even while the classical hysplex system was maintained at older sites. These new installations occurred in mainland Greece and in Asia Minor. These naturally satisfied the practical need for a proper start for the races and, at least from a technical viewpoint, were equipped with even better systems. But at the same time they satisfied the new spirit that then reigned in architecture for monumental 25 6 entrances and impressive forms in general. A characteristic common to all these constructions is the use of the large stone pillar-half columns that stand on isolated 257 bases or on a continuous balbis. These have on their side and back surfaces sockets, channels, and other kinds of cuttings, in which were placed and within which moved the cords of a new hysplex system. Similar channels also exist in the horizontal members of
96
Hysplex
Figure Fig ure 68: 68: Ep ida uro s. P la n of the second second phase phase of the starting line at the eastern eastern end of the stadi um (after Kavvadias pl. Bl-4).
barriers fall. It is also interesting that in the new system each individual athlete had his own barrier; we have returned, that is, at least with regard to this detail, to the
Hysplex
98
262
half columns is a vertical rough-picked stripe that is, for the most part and in most cases, cases, completely complet ely superfici super ficial; al; only onl y on one side of two differ different ent columns is this "stri "s tripe" pe" a l ittl it tle e deeper, formi fo rming ng a recessed area are a 0.068-0.07 0.068-0.070 0 m. wide wi de and an d 0.01 0.01 m. deep. dee p. (figs. (figs. 71.3 71.3,, 72). Al A l l of the the drums are provided provide d wit h matching matc hing dowel holes holes measuring measuri ng 0.0 0.055 55 χ 0.030 0.030 m. on their thei r upper and a nd lower surfaces. Aside from these, and only on the upper horizontal and immediately adjacent ver tical surfaces of the second drum, there are two additional cuttings, one on each side of the drum and exactly above the rough-picked stripe on the side, that are of special interest and importance (figs. 71.3, 73). These cuttings have, in section, the shape of a Γ. They are roughly 0.02 m. wide and 0.085-0.105 m. in length, seen from above (fig. 74) and from the sides (fig. 75). The ends of the Γ have holes cut at right angles to the side and the top of the block which extend, like the hook of a clamp, to a greater depth in the stone. Were they extended slightly farther, they would intersect 26 3 each other. It would appear that all these cuttings existed on all five half columns of both starting lines. However, some additional interesting details appear on the central half column of the eastern starting line. Starting from the bottom, we note that the foundation base
Observations Observations on the Starting Lines Lines of Various Stadia
101 101
a cord passed through the east-west channel of the stone base of the cen tral half column, and that the cord is likely to have attached to a sliding metal element on the rear vertical face of the same central half column. It seems possible that this vertical element made attachments to horizontal elements, cords or more likely bars, which made some connection to the shallow vertical cuttings in the sides of the four flanking half columns. The horizontal bars would have fallen to the level of the racecourse floor between the front of the grooved starting line blocks and the rear of the bases for the Ionic half columns, all of which projected slightly above the floor level. Thus the horizontal bars would not have been stumbled over 26 8 by the starting athletes. Th e basic basic principle princ iple in Romano's Roman o's suggestion suggestion for for the workin g of the whole system system with wi th a cord that passed though the pipe in the middle half column is generally correct. I also agree that the barriers should have been "horizontal bars", probably wooden. But it seems that the barriers cannot have been behind the half columns but only in between them as is indicated by the fact that the half columns project from the balbis (0.60 (0.60 m.) m.) as much muc h as the support supp ort bases bases of the older type typ e of hysplex (fig. 68) while the vertical channels in the sides of the half columns are as far away from the balbis as 269 the ankones of the older hysplex type. Finally, the argument that the bars behind
Figure-86: Epidau Epi dauros ros.. Schem Schematic atic reconstruction reconstructi on of the hysplex of phase II at the eastern end of the stadium, with wood barriers and wood epistyle through which ran a cord (dotted line).
27 1
of ± 3.20 m. are huge for single runners. Based on the new hypothesis—that is the attachment of a barrier to every Γ-shaped cutting of a half column (i.e., two for every intercolumniation)—I think that ten runners could have run simultaneously. This is a
112
Hysplex
27 3
as appears to have been the case at Isthmia I a few centuries earlier. That the Hellenistic hysplex of Epidauros follows on an older system in which a mechanism was used for the creation of torsion-produced force, in the context of the great technological advances of the period, leads us to the reconstruction of a more advanced system in 27 4 which were used the lepides of the chalkotonon of Ktesibios discussed above. I would, then, restore at each Γ-shaped cutting such a lepis, attached to the "heel" 275 of each ankon barrier, in order to drive it down with force (fig. 85). Further, we can now suggest the following concerning the operating system in the central half column. As we have already concluded, in that part of the channel on the back of the column that is semicircular there passed a central cord to which was attached the individual cords coming from each of the barriers through metallic rings attached to the wooden epistyle (figs. 77.7, 84). Through the lower part of the channel, which is larger and of rectangular section, it appears that together with the cord was some other wooden or metallic element like a plug that went up and down with the 276 cord. This element (the plug) ought to have been tied to the cutting in the center of the bottom surface of the half-column or at the lower horizontal hook-shaped metallic clamp. This tie was necessary because the cord was pulled with force by the ankones thanks to the pressure of the lepides. When the aphetes released the plug, it flew up and allowed the ankones to fall,
116
Hysplex
(2) We do not know for certain if all these channels or only some of them ran the whole height of the pillars, as the vertical holes and the continuous channel on top of the geison indicate. (3) Since the construction is preserved only to the bases of the pillars, we do not know if other elements were present in their shafts or in what way they bore the horizontal barriers of the hysplex. Since the lack of information precludes suggesting another solution, we can only indicate on the drawing (fig. 89), and by analogy with Stadium II at Epidauros, the placement of horizontal wooden elements at a height of about a meter above the balbis which are held by vertical wooden posts wedged into the lower part of the interior channel of each gate. Again, it seems necessary to place here metal spring plates for the introduction of torsion in the barriers, as we did above also in Stadium II at Epidauros. Finally, since the four right gates have their channel on the opposite parastade from the four left gates, the whole system must have had 280 an overall image like that shown in figure 90. In the event that these vertical channels were intended only for the receipt of wooden posts, they might not have reached all the way to the epistyle, and then the cord that suspended the horizontal barriers would have hung free in the air between the pillars
Observations on the Starting Lines of Various Stadia
119 11 9
Afterward, he would pull the two cords that came from below, from the channels of the foundation, to release the ankones, which, propelled by the torsion of the spring plates, would fall downward with force. Now that we have placed these two systems in the appropriate channels, there remain unassigned only the channels in the outer sides of the two end gates and in the inner sides of the central wider intercolumniation. Since they enclose the four gates on each side of the balbis, these probably had running through them cords to connect the cords of the balbis and the geison, tying them together and organizing all the cords into two parts of a single unified system. The lack of upper elements at Priene hinders us in conceptualizing in greater detail the mechanism and the way the system worked. I believe, nonetheless, that an on-thespot study of the architectural remains together with a correlation with the hysplex and of other possible parts of the upper construction would lead to a clearer understanding 28 2 of these problems. KOS
We are fortunate to have preserved in Kos not only the base of a monumental and relatively complicated starting system but also all of its architectural members, from
124
Hysplex
Alongside the half columns was found a section of Doric epistyle that preserves the fascia with regulae and guttae, and that evidently comes from this construction (fig. 95). Its dimensions (1.505 m. long, 0.53 m. wide, and 0.315 m. high) are similar to the interaxial dimensions of the "gates", and there are in addition holes 0.18 m. on a side cut into the ends of the block with a point (fig. 96). These correspond precisely 28 5 to the vertical holes of the pillar-half columns. On the upper surface of the same epistyle is another, oval, hole resembling a funnel (fig. 97; dimensions at the top: 0.20 χ 0.14 m.) which was also hastily cut with a point. It ends at the bottom in two small holes, aligned parallel to the long axis of the block, each with a diameter of 0.035 m., and each opened with a drill. Lying together with the epistyle are two pieces of Ionic geison (figs. 95, 96) the surfaces of which are so broken that their original dimensions cannot be recovered. It is, however, very probable that they come from this same construction since in this place and at that time the combination of Doric epistyles with Ionic geisons was the most common practice. A visit to the archaeological areas of Kos where the buildings are reconstructed by the Italians provides many examples of the phenomenon. L. Morricone associates with the monumental elements of the starting line at Kos
Observati Observations ons on the Start Starting ing Lines Lines of Various Stadia
129
In addition to these elements that we have seen in other cases, there are also ex tant in Kos, because of the very good state of preservation of the lower parts of the construction, some other characteristics that allow us now to hypothesize with a great deal of probability not only the general principles of the working of the mechanism of the hysplex but also the reconstruction of the system in considerable detail. These el ements come from a horizontal semicircular cutting (figs. 103-107) with a total length of 0.425 m. and a diameter—in its longer part—of 0.08 m. chiseled in the back of the pillar and thus visible in the empty space between the pillar and the orthostate. This cutting, which has a narrower "mouth" at that side of the half column with a diameter of 0.067 m., is at a height of 1 m. from the stylobate—that is, at a height that we have seen associated with the barriers in many hyspleges of various types and times. In one of the half columns, at the far end of the semicircular cutting, there is pre served an iron ring, slightly smaller in diameter than the cutting, leaded into the "shoulder" of the cutting (figs. 106, 107). Corresponding traces indicate the existence of similar rings in all the half columns. On the bottom side of this cutting and project ing from it is another cutting with a quarter-circle section and measuring 0.055 m. in length (from the horizontal semicircular cutting) and 0.04 m. in width. The long axis of this cutting is exactly on the vertical axis of the half column. Along the interior edges of the upper surfaces of the pillar and the orthostate there is a matched set of
130
Hysplex
that were hung for the protection of the mechanism when the games were 29 0 not in session. Although I essentially agree with the very general principles for the working of the mechanism that Krinzinger establishes, I will proceed with an analysis of all the data and offer a new hypothesis for the detailed working of the hysplex. In fact, we are sure that the system operated by means of cords that moved through the vertical holes of the pillar and the horizontal channels of the base and—probably—of the geison. All these cords must have been tied ti ed together int i nto o two halves (one (one to the left and a nd one to the right), as at Priene, and the ends of each would have come together at the hole of the central epistyle and at the interruption of the channel in the foundation at the central wider intercolumniation where the aphetes apparently would have stood. It is also obvious that the parts of the mechanism as well as the barriers themselves must have been located in the empty space that is formed by the two lower members of the pillar-half columns that communicated above and below with the system of cords. Studying the details of the empty space, starting from the top and proceeding to the bottom, I propose the following:
Observations on the Starting Lines of Various Stadia
137
shaped cuttings that Krinzinger suggests were used for the placement of wooden lids 292 so that the mechanism could be closed up when not in use. But as we have seen in every case, and as appears to have been true at Kos as well, all the elements of the mechanism were movable and were set in place only for the duration of the games. For the rest of the time they were removed, which means that nothing remained in the empty space that had to be protected. Nonetheless, I do agree that the L-shaped cuttings were intended to receive pins protruding from the sides of boards, the bottom edge of which were held in place by the oblong sockets that appear on the stone bases in the empty space. The only use that this perpendicular board could have had was to hold the ankon of the hysplex (fig. 108 D, E). In other words, it is also a basic part of the mechanism with the ability to be set in place and removed. It must have extended to the height of the trigger of the roller, and at its top there must have been a slot in which, by means of a simple wooden pin, the ankon was held. The ankon thus had the ability to move from a horizontal to a vertical position and back again. The swift and sure movement of the ankon could have been achieved by means of the appropriate bronze spring plate, which would have been nailed in place near the joint of the two wooden elements, as we suggested was also the case at Epidauros, with one end in the vertical support and 294 the other in the ankon. The existence of two openings and of two pairs of facing L-shaped cuttings in every
Observations on the Starting Lines of Various Stadia
141
the two systems—either that of the torsion production or that of the release—must have ceased to be wholly within the architectural elements and have been replaced by those cords that were in front of the athletes in every intercolumniation. It is, finally, self-evident that, to avoid malfunctions in the system, there must have been at every corner either small and sensitive rollers or a working down of the stone members so that with a good greasing they could facilitate easy movement of the cords.
Conclusion
With these complicated systems we have come to the end of a series of mechanisms that, set in front of or on top of the balbides of ancient stadia, had the purpose of achieving an equal and fair start for all the runners in the races. We should give to all these systems the name, known from many sources, hyxplex since, despite their differences and clear evolution, all had as a basic and common characteristic the existence of barriers in front of the athletes which fell suddenly and with a bang, thanks to the help of a system of cords, and thus allowed the simultaneous start of the runners. Such systems must have first appeared in the hippodromes and have been adopted with some adaptations in stadia in the second half of the 5th century BC, as we are shown by the oldest preserved system at Isthmia and by a reference in the comedies of Aristophanes. These first hyspleges were simple. They consisted of individual wooden posts, one for each runner, fixed in sockets in the balbis. These posts held, at their upper parts, a horizontal wooden barrier that, on the appropriate action of the aphetes, fell only due to its own weight, thus allowing the runners to start. A depiction on a Panathenaic amphora of the archon Lykiskos (344/3 BC) shows that from the middle of the 4th century there was conceived a new system with two end posts, and a third central post
144 144
Hysplex
amphora, with the existence of a single unified barrier for all the athletes, as opposed to the individual barriers of Stadium I at Isthmia, as well as with the sudden fall of the barrier thanks to the use of torsion produced by the twisted neura, it was possible not only to reduce favoritism or the opposite, whether intentional or not, toward a certain athlete but also to produce a greater theatrical impression at the significant moment of the start of the race. Given the principles and the practices of the ancients a similar type of barrier must have existed in every Greek stadium. Traces of it are to be noted at Olympia, doubts about its specific reconstruction at Delphi are expressed, and hypotheses are made relative to the starting lines of some stadia in Asia Minor in which it is possible that there was a simpler completely-movable-as-a-whole mechanism that was placed on top of a heavy base and was transported to the front of the balbis only during the period of the games, just as took place otherwise with the main part of the mechanism of t he hysplex as we have seen. Again, such a movable system no traces of which remain, might have existed in the gymnasia —in both the xystoi and the paradromides. Such a light system could have been constructed after the replacement of the neura for the production of torsion in war machines by a smaller system of metal springs that was invented in the middle of the 2nd century BC. In later Hellenistic years, at the same time that the principles of the classical system
Appendix The Rebirth of the Hysplex at Nemea
The problem of the form and method of functioning of the ancient hysplex has vexed me ever since the discovery of the stone bases for it at Nemea in 1975, but a reading of an early draft of Panos Valavanis's study convinced me that he had solved all, or nearly all, the problems of that mechanism. I was, in fact, so convinced of the correctness of his theories that I asked to be allowed to do an English translation of his work. During the translation a question nagged: Could the hysplex as hypothetically re stored by Valavanis actually work? The idea of a practical demonstration—a life-size working model based, literally, on the ancient blocks in the stadium at Nemea—was thus born. Valavanis graciously, if with some anxiety, agreed to allow me to put his theories to the test. During preliminary discussions, a day spent on the site at Nemea with Valavanis, Manolis Korres, and Petros Briles 29 7 was crucial since certain ques tions were resolved, at least temporarily, so that decisions could be made about the actual reconstruction. The latter took place during the first two weeks of July 1993.
146 146
Appendix
2
in antiquity might have been elm, ash, or boxwood, we were limited by the wood available on the market in a size sufficiently large for the various parts, particularly for 300 the post of the left end of the balbis. We were also concerned with resistance to rot and with ease of working, as well as with strength. The final decision was for African mahogany, and even then it was necessary to laminate large planks to produce blocks 30 1 of the size necessary for the lower frame and the columns. Of course, had we taken more time, we might well have been able to search out larger single blocks of wood of an appropriate type. The metallic elements were all of iron or steel, and they did not, as will be seen below, always reproduce exactly the ancient forms as we understand them. This was due, mostly, to a desire to use items that were readily available in today's marketplace. Still, 302 the basic elements of syrinx, choinikis, epizygis, and mochlos will be recognized in their modern steel reincarnations, although in forms that are not absolutely authentic. For the cords we used hemp rope with a diameter of 0.005 m. I suspect that this is fairly close to the ancient cords. The remaining element, the neura, was the most difficult. Again motivated by a de 30 3 sire for the easily and quickly attainable, and lacking long-haired women and sinews, 304 we settled on hemp rope again, but this time with a diameter of 0.01 m.
Appendix
147
Since most of the holes were badly weathered and broken as well as missing their lead lining, 10 of the 12 posts were pinned in place by wooden wedges. But in the two cases in which the lead was preserved, we discovered that by leaving the posts out in the sun to dry and shrink, it was possible to slip them easily into the lead sheathing in the early evening when the lead had cooled and also shrunk. By the next morning, the dew from the night had caused the posts to swell and fill the holes tightly and securely. Deliberate dampening during the day, aided by the expansion of the lead in the heat of the sun, would also have created this tight fit. We further noted that the sides of the lead lining are smooth as if from wear and do not possess some of the sharp edges that one might have expected of lead that was poured while molten around a post that subsequently eroded, leaving its negative impression in the surface of the lead. This raises, again, the question about the use of the lead in these holes. Broneer had stated that the purpose of the lead in the post holes at Isthmia (and by extension elsewhere) was to fasten turning posts "in the sill so firmly that the runner could grab his post with his left hand and swing around with a minimum loss of time and momentum." 30 7 Sweet, realizing that such a use for the posts would be awkward, at best, for the hoplitodromoi, hypothesized that those runners reversed the normal—and so far as I know, universal—counterclockwise turn to the left to make a clockwise turn to the right. 30 8 As I have pointed out elsewhere, there is no ancient evidence for any turn in a clockwise fashion or for swinging on the posts. 30 9 After our experience with slipping the posts into and out of the lead lining (fig. 110),
First, we note that there is considerable difference of dimension from one wooden block to the next within a single base unit and that the central blocks of each unit are quite different from one another. But, even though the outer and inner blocks of a single unit (A and Γ, Δ and Z, respectively) are also very different, the outside blocks of both units (A and Z) and the inside blocks of both units (A and Γ) are quite similar to one another in their widths (0.171 vs. 0.178; 0.220 vs. 0.230). Finally, the widths of the inside blocks are about 0.05 m. greater than the widths of the outside blocks. An explanation for this will be offered below, but at least we can already say that there appears to be some system at work which was not so apparent from the cuttings alone. It needs also to be noted that both of the central blocks, although retained at their full height at the rear of the system where they are wedged up against the columns, were cut down for a depth of 0.105 m. to a point 0.105 m. behind (south of) the axis
Appendix
155
this pre-twisting was important to keep the ankon more or less centered in the frame so that it could move freely without contact with any surfaces. But we discovered that the pre twisting, if done in the proper direction, also increased significantly the force of the torsion. 31 8 Now the ankon could be inserted between the two groups of rope and the rope twisted by means of the pipe and pin. It was usually possible to make two or three turns by hand, but the final twisting was done by means of a pipe (of an arbitrary length but sufficient to provide good leverage) the inner diameter of which was just slightly larger than the outer diameter of the pin so that by slipping this thin pipe over the pin one was able to exert a great deal of pressure. When the limits of possible pressure were reached, a wedge of wood placed between the pin and the stone below held the tensed rope in torsion. The next step was to lift the ankon into a vertical position, positio n, thu s "cocking" "cocking" it for for release release by the add ition of a quar ter tu rn of torsion torsio n on on the ropes. Several additional details about this process should be presented here. (1) The ankon was a round piece of wood, 0.055 m. in diameter with a height of 1.15 m. Given that the central wooden block was cut down 0.105 m., it thus protruded 1.05 m. (allowing 0.005 m. for movement of the "heel" of the ankon) above the level of the balbis, as did all the vertical members in our reconstruction. The first few times we released it, there was a "hesitation" at the point of release. This was due to the flat bottom of the post the corners of which caught on the recessed upper surface of
158
Appendix
simple, for the syrinx-choinikis element was surely a specially manufactured element of greater stability than our version even if the sources that mention it are rather later in date than our machine (see above, n. 315). But the point here is, what part of the ensemble is going to take the pressure of holding the torsion in check? I do not think that the earth, on the one hand, in a supposed crank pit is a very secure candidate, and I suspect that the wooden elements, on the other, ought not be be subjected to the sort of pressure that would tend to split them. But the stone base itself is clearly capable of undergoing such pressure, whether in the form of our simple wooden wedges or some more elaborate device. If this line of reasoning is valid, then the pipe-turning element ought not to extend beyond the line of the stone base, and the purpose of the cutting that runs to the outside of the support bases is revealed as providing space for the movement necessary to introduce torsion into the ropes as well as for a base to withstand the pressure of holding that torsion in check. (4) The modern ground level at the balbis is slightly lower than the ancient level (because of test trenches), but it is still too high for the hysplex. On our initial trials, t h e ankon came down to hit, at a single point in its length, the corner of earth in front of the stone support base with the obvious danger that the ankon might snap at that point. It was therefore necessary to excavate a narrow "trough" (not yet done at the time when the photographs presented here were taken), the bottom of which
Appendix
159
The Release System We duplicated closely the system suggested by Valavanis 32 4 with a short cord secured at one end to a nail in the top of the column and at the other end to an iron ring that slipped over an iron pin in the top of the ankon (fig. 118). This is the ring that held the force of the torsion in check and that released it when pulled by another, much longer, cord that was attached to the ring at one end and that ran back to the apex of the isosceles triangle to the hand of the aphetes. The actual execution of the release proved interesting for while it is possible to perform it faultlessly repeatedly, to do so requires skill and strength. Four different people—including the two whose names appear on this work—tried their hand, and each failed to cause both ankones to fall together on the first try. Because it is nec essary to have the rings at the extreme top of the pins of the ankones in a relatively treacherous position, the long pull cords must be raised deliberately and carefully off the ground. Next, the aphetes must be sure that the pull cords have as little sag as possible and that whatever sag exists is equal for each cord. Needless to say, the length of the pull cords—that is, the distance from aphetes' hand to ankon's ring—must be the same for each. When he releases the ankones t he aphetes must pull the cords upward equally and
Appendix
161
Immediately north of the settling basin at the west end of the Nemea starting line is a large block the placement of which caused a westward shift in the location of the water channel north of the basin. 32 6 This change in the water channel brings quickly to mind the similar change in the posts on the balbis occasioned by the addition of the th e hysplex.327 It is clearly tempting to think that both changes were the result of the same cause—the newly added hysplex. We tied one guy at the bottom of this block so that the cord would allow the ankon to fall flat on the ground (see above), and the block worked very well well for for the th e anchor anch or of the th e guy at this th is end e nd of the th e line (fig (fig.. 119). I am now confident that this was the original purpose of this block. The corresponding water basin and channel at the east end of the starting line also show clearly that the channel there was moved eastward from its original position. 32 8 About 5 m. directly north of this location there was discovered in 1976, no longer in place, a block that is very similar in size and shape to that at the west end of the starting line.32 9 We had no qualms in placing (replacing?) it in a position analogous to its western mate, and it served equally well as an anchor for the guy at this end (fig. 120). We have, then, resolved the purpose of these two blocks in the stadium at Nemea. By tightening the guys, it was possible to bring the two horizontal cords to an almost perfectly straight line (fig. 121), and the whole system was thus taut with torsion and tension waiting for release by the aphetes. I suspect that some similar system for the horizontal cords existed in other stadia, and the suggested use of wooden bars as barriers can be discarded. It is improbable in any event. First, the bars restored by