CLASS
PHILO A JournalDevotedto Researchin ClassicalAntiquity
VOLUME THE
99
NUMBER
UNIVERSITY
1 OF
JANUARY
CHICAGO
PRESS
2004
Shadi Bartsch, Editor
David Wray, Book Review Editor Elizabeth M. Adkins, Managing Editor Robert J. Germany, Editorial Assistant Maureen E. Mahowald, Editorial Assistant ASSOCIATE Danielle S. Allen Michael I. Allen Elizabeth Asmis Helma Dik Christopher Faraone Jonathan M. Hall Nancy Helmbold W. Ralph Johnson EDITORIAL
EDITORS David G. Martinez Richard T. Neer Mark Payne James Redfield D. Nicholas Rudall Richard Saller Peter White
BOARD
Mary Depew, University of Iowa Dewald, University of Southern California Carolyn P. E. Easterling, Newnham College, University of Cambridge Denis Feeney, Princeton University
Brad Inwood, University of Toronto Lisa Kallet, University of Texas at Austin Robert A. Kaster, Princeton University Marilyn A. Katz, Wesleyan University
C. E Konrad, Texas A&M University Charles E. Murgia, University of California, Berkeley Robert Renehan, University of California, Santa Barbara David Sansone, University of Illinois at Urbana-Champaign Laura M. Slatkin, New York University James E. G. Zetzel, Columbia University OF EDITORIAL POLICY STATEMENT Classical Philology welcomes both longer articles and short notes or discussions that make a significantcontributionto the study of Greek and Roman antiquity.Any field of classical studies may be treated, separatelyor in relation to other disciplines, ancient or modern. In particular,we invite studies that illuminate aspects of the languages, literatures,history, art, philosophy, social life, and religion of ancient Greece and Rome. Innovative approaches and originality are encouraged as a necessary partof good scholarship.Illustrationsmay be used as appropriate.Book reviews are published by invitation only. Cover Illustration:Reverse of bronze medallion, showing Horatius Cocles. Roman, c. 140 c.E. de 1'Empireromaindepuisle regned 'Auguste Lesme'daillons jusqu'l DrawingfromW.Froehner, Priscus Attale (Paris, 1878), 60.
CLASSICAL VOLUME
99
PHILOLOGY
NUMBER
1
JANUARY
2004
ARTICLES Exemplarity in Roman Culture:The Cases of Horatius Cocles and Cloelia Matthew B. Roller An Oracle of Apollo at Daphne and the Great Persecution Elizabeth DePalma Digeser NOTES
1 57
AND DISCUSSIONS
Impersonal and IntransitiveEITIZHMAINEI Daryn Lehoux
78
BOOK REVIEWS The Sleep of Reason: Erotic Experience and Sexual Ethics in Ancient Greece and Rome. Edited by MarthaC. Nussbaum and Juha Sihvola. Craig Williams
86
Greek Literature and the Roman Empire: The Politics of Imitation. By Tim Whitmarsh. CliffordAndo
89
NOTE
TO
CONTRIBUTORS
Please addressall correspondenceto the Editor, Classical Philology, 1010 East 59th Street, The Universityof Chicago,Chicago,IL 60637. Submissionsshouldbe paginated,with amplemargins, double-spacedthroughout,andwith footnotesgatheredat the end of the paper.Please submitthree copies. All submissions are refereedanonymously.No indicationof the author'sidentity should appearon the typescript;referencesto the author'sown work should take the same form as references to the workof others;acknowledgmentsor otherremarksthatcould revealthe author'sidentity shouldbe withhelduntil the paperhas been accepted.
Classical Philology (ISSN 0009-837X) is publishedquarterlyin January,April, July,and October by The University of Chicago Press, JournalsDivision, 1427 East 60th Street, Chicago, Illinois 60637-2954. Yearlysubscriptionrates (print& electronic),U.S.A.: institutions,$183.00; individuals, $50.00; students,$31.00 (with copy of validatedID); Canada(includes$5.00 postage and7% GST): institutions,$200.81; individuals,$58.50; students,$38.17; rest of world(includes surface postage; airmail rates available): institutions, $188.00; individuals, $55.00; students, $36.00. Electronicversion for institutionsonly, World:$165.00. Multi-yearrates available:see web site (below). Japanesesubscriptionagent: KinokuniyaCompanyLtd. Single copy rates:institutions, $50.75; individuals,$17.50. All correspondenceconcerningsubscriptionsshould be addressedto The Universityof ChicagoPress,JournalsDivision, P.O.Box 37005, Chicago,Illinois 60637. Tel.: (773) 753-3347 or toll-free in U.S. andCanada(877) 705-1878. Fax: (773) 753-0811. E-mail:
[email protected] should be made payable to Classical Philology. Back issues older than7 years are availablefromPeriodicalsService Co., 11 Main Street,Germantown, NY 12526, www.periodicals.com.All volumes areavailablein microfilmfromProQuestInformation and Learning,300 NorthZeeb Road, Ann Arbor,Michigan48106. The articlesin thisjournal are indexed in the HumanitiesIndex. WebSite: http://www.journals.uchicago.edu/CP CopyingBeyondFair Use: The code on the firstpage of an articlein thisjournalindicatesthe copyright owner's consent that copies of the articlemay be made beyond those permittedby Sections 107 or 108 of the U.S. CopyrightLaw providedthatcopies are made only for personalor internal use, or for the personal use or internal use of specific clients and provided that the copier pay the stated per-copy fee throughthe CopyrightClearance Center (CCC), 222 Rosewood Drive, Danvers,MA 01923. To requestpermissionfor otherkindsof copying, such as copying for general distribution,for advertisingor promotionalpurposes, for creating new collective works, or for resale, kindly write to PermissionsDepartment,The Universityof Chicago Press, 1427 East 60th Street,Chicago,Illinois 60637-2954. If no code appearson the firstpage of the article,permission to reprintmay be obtainedonly from the author. ANSIStandard:The paperused in this publicationmeets the minimumrequirementsof American National Standardfor InformationSciences-Permanenceof Paperfor PrintedLibraryMaterials, ANSI Z39.48-1984. @TM Postmaster:Send addresschanges to Classical Philology, The Universityof Chicago Press, Journals Division, P.O.Box 37005, Chicago, Illinois 60637. Periodicalspostage paid at Chicago, Illinois, and at additionalmailing offices. ?2004 by The Universityof Chicago.
EXEMPLARITY IN ROMAN CULTURE: THE CASES OF HORATIUS COCLES AND CLOELIA MATTHEW B. ROLLER
I. INTRODUCTION N DISCUSSING it seems fittingto begin with an example of one. EXEMPLA, In a celebrated passage, Polybius describes for Greek readershipsome of the itocaoif-habitsor customs-that enabled the Romans to overcome the Carthaginians.First he points to the aristocraticfuneral (6.53-54), in which wax masks representingancestors of the deceased are conveyed in the cortege, and a eulogizer, himself a scion of the family, recounts these ancestors' deeds on the community's behalf. All this pomp, says Polybius, is socially efficacious. Young men who observe this spectacle are fired with the desire to endure and risk everything for the community, in order to win such immortalityand renown for themselves (?53.9-54.3). Indeed, he adds, many Romanshave done noteworthydeeds withjust this end in view (?54.45). He continues (?54.6-55.4): irap&'Pwpaifotg Iv 6' &pKcoOv toXit p&yvoyv Katinept inoXXWvitTopeitat rotaizra Onapyv irr' 6vo6aTCog l0• azrat Kai CirtzsEogVEKEV.(?55.1) K6aTpbg 6Tb 6OrtosidyLarog KXV V yap kysctat zbv 'pzartov isrtlKlcrivra, 8tay'(vti6pcvov npb; 860 TiOvbirnvavitov (?54.6)
T KazcvztKPv zflg y7tpu6pagitcpazt zTg tini 7TOTtPlIptSog, i" KEItat pob rflg 6 Irti T 7OX•Sg, pti tadcipcvot s86pSv rzoi 7okEpiotq, s PorlP0ovz0•v ioCsavra nrapairtowotvsig i•v nCr6tv,poov 9intozpapsvzTa toKi Kaz6ntvt agzTaoq ava0oppjoavra~q 6taonrtv 7iv y'iqupav. (2) zdv 6~ 7rt0aprlocvrTov, iog p.tv o`zot 8ticr0sov, birtePve t&OpCov, o06 oi•rgO zpauLtd-ov rT•l0o0qdva6cX6pOEvogKai Staaicciozs -nilv irttpoop&vrTvr Uvatv0 v rzdv i tiqzliv irDt6craotv a6toD Kai zt6paTv Kaznctlht• zivv pXLwy .nrvavzrtov 6~' ,i yFytppaq,oi l V rtnoXkptot (3) 6taonaOcsioalg Z•g 6ppltqi9io0A60lav, 6 R K6~hXSq TVv rV soZactptvZoiv istyaq iqkots Katt rtpoaipcsatv psEikaSE z6v Piov, t7api auqa v Ei Zi ignRatpi6So 7totrlodapEvoqTilv q 6pdXetav Kai zliv toopTvrlv jpEWdT raszra xspi r•,ciovog 1Oi Kai rtTO ou piou. (4) zotatrll zt~, tbg aorbv efiKsetav Zrit T taktropJC 0ol tapoiq Kazo cotce, trv tyyEvvarat roig vsotg 6ppjl KCai•plott•Zpia tpoq r& &8tt trap' a6Trotg 6OtoCttov Epywv. KakX ztrv EnrEt rXfi0o0g ;rttpEp6pEvov
This paper has been long in the making, and many institutions and individuals had a hand in shaping it. I thank the Institute for Research in the Humanities in the University of Wisconsin and the American Council of Learned Societies for providing the resources for a sabbatical leave during which these ideas took shape; also audiences at the APA annual meeting in 2001, at Indiana University, the University of Texas at Austin, and the Archaeological Institute at the University of Munich. I further thank Hans Beck, Karl-JoachimH61keskamp,Bob Kaster, Chris Kraus, Ellie Leach, Ralf Von den Hoff, and two exceptionally engaged referees for CP for their careful readings and advice on particularissues; also David Wray for sharing a valuable manuscriptof a work in progress. All translations are my own.
Classical Philology 99 (2004): 1-56 2004 The of by University Chicago. All rights reserved] 0009-837X/04/9901-0001$10.00 [?
1
2
B. ROLLER MATTHEW (?54.6) Many such stories concerning many men are related by the Romans, but one notable instance will suffice for the present, offered as an example and as proof. (?55.1) It is said that one Horatius Cocles was fighting against two adversaries on the opposite end of the bridge over the Tiber that lies before the city. When he saw a large force of enemy reinforcements approaching,fearing that they would force a passage and storm into the city, he turned to those behind him and shouted that they should withdraw immediately and tear down the bridge. (2) While they did as he bid and tore it down, he stood fast, receiving a large numberof wounds, and checked the onslaught of the enemy, his adversaries being astounded not so much by his strength as by his resolution and boldness. (3) Upon the collapse of the bridge, the enemy was prevented from attacking and Cocles, hurlinghimself into the river in his armor,purposefullygave up his life, reckoning the safety of his fatherland and the renown that would accrue to him thereafter more valuable than his currentexistence and the portion of his life remaining. (4) Such, it seems, is the impulse and love of honor regarding noble deeds that is engendered in Roman youths by their customs.
Polybius implies that Horatiusacted in imitation of unnamedpredecessors: to such actions are young Romans stirredwhen they hear the deeds of past heroes narratedduring a funeral (?54.4-5). Yet he also acted in hopes of winning similar glory for himself (?55.3); and the fact that Romans often tell his story (?54.6) suggests that he succeeded, providing a model for imitation to future generationsjust as he himself acted in light of existing canons. Other literary texts bear out Polybius' suggestion that Horatius' story was resonant: I know more than thirty narrativesof or references to this deed in Roman literature.Sometimes, as here, there is a full-scale narrative, while other times his name is mentioned in passing, with the expectation that the reader can supply, from his preexisting knowledge of the narrative,whatever details are pertinentto the context. Now, while Polybius has detached this narrativefrom all historical context-he does not inform his readerwhen, under what circumstances, or to what political or military end Horatiusdid his deed-he does supply crucial cultural context: he implies that the Romans themselves tell the story as a "stand-alone"(i.e., without narrativehistorical context) just as he himself has done; and he implies that such narrativizinghas an ethical aim. In section III below, I discuss furtherthe exemplum's capacity for historical decontextualizationin the service of ethics. In the meantime,let us bringLivy's version of the story into consideration, since it does provide a narrativehistorical context. According to Livy, Horatius' deed dates to the second year of the Republic, the consulship of P. Valerius Publicola II and T. Lucretius (2.9.1).1 Tarquinius Superbus,deposed as king of Rome the previous year, had appealed to Lars Porsenna, king of Clusium, to reinstall him by force. Porsenna agreed, and led his army against Rome the next year. First he seized the Janiculum, ejecting a Roman garrison.These soldiers fled across the one bridge 1. Likewise Plut. Publicola 16.3; this consular year corresponds to 508 B.C.E.on Varro's chronology, and 504 on Livy's. Dionysius of Halicarnassus, meanwhile, places the deed in the following year, the consulship of Valerius Publicola III and Horatius Pulvillus II (5.21.1, cf. 5.35.3)-a year unrecordedby Livy. On the confusion in the consular lists for these years (and hence in the dating of Porsenna's attack) see Broughton 1951, 1.6-7; on early chronology generally see Cornell 1995, 218-23, 399-402.
EXEMPLARITY IN ROMAN CULTURE
3
that spanned the Tiber, the pons sublicius or "bridge on piles," with Por-
senna'stroopsin hot pursuit.At this point (2.10.1-4) Horatius,in the rear guard,declaresthathe will hold up the Etruscanattackwhile his companions attemptto break down the bridge. The narrativecontinues (2.10.5-11): (5) vadit inde in primumaditumpontis, insignisqueinterconspectacedentiumpugnaeterga obversis comminus ad ineundum proelium armis, ipso miraculo audaciae obstupefecit hostes.... (8) circumferens inde truces minaciter oculos ad proceres Etruscorumnunc singulos provocare, nunc increpare omnes: servitia regum superborum, suae libertatis immemores alienam oppugnatumvenire. (9) cunctati aliquamdiusunt, dum alius alium, ut proelium incipiant, circumspectant;pudor deinde commovit aciem, et clamore sublato undique in unum hostem tela coniciunt. (10) quae cum in obiecto cuncta scuto haesissent, neque ille minus obstinatus ingenti pontem obtineretgradu,iam impetu conabantur detruderevirum, cum simul fragorruptipontis, simul clamor Romanorum,alacritateperfecti operis sublatus,pavore subito impetum sustinuit.... (11) ita sic armatusin Tiberim desiluit multisque superincidentibus telis incolumis ad suos tranavit, rem ausus plus famae habituram ad posteros quam fidei. (12) grata erga tantam virtutem civitas fuit; statua in comitio posita; agri quantumuno die circumaravit,datum. (5) Then he strode to the very entrance of the bridge. Among those who showed their backs as they withdrew from the conflict, he was conspicuous for turning his weapons at close quartersto face the approaching battle, confounding the enemy with the very spectacle of his boldness .... (8) Then, casting his fierce eyes threateninglytoward the Etruscan forefighters, he now challenged them individually, now derided the lot: they were slaves of arrogant kings, and heedless of their own freedom they had come to attack someone else's. (9) The Etruscanshesitated for a while, while each looked to another to begin the battle; then shame roused them to attack, and raising a shout they hurled their javelins from all sides against their solitary enemy. (10) But when all had stuck in his protective shield, and he was holding the bridge no less resolutely with his magnificent stance, they were then trying to dislodge the hero with a charge, when the crash of the severed bridge, and at the same instant the cheer raised by the Romans thanks to the speedy completion of their task, checked the Etruscanonslaught with sudden dread .. . (11) In full armor as he was, he leaped into the Tiber and, through a thick shower of javelins, swam safely across to his companions, having dareda deed that would gain more glory than credence with future generations. (12) The state was grateful for such great bravery: a statue was erected in the comitium, and as much land as he plowed aroundin one day was granted him.
Considering this narrative together with Polybius', I would note four aspects that are more or less prominent in both. First, Horatius' deed is spectacular: he stands conspicuously alone on the bridge, under the Romans' gaze from behind and the Etruscans' from before. In both versions, Horatiusmaintains verbal and visual contact with the Romans who work to demolish the bridge, and is an object of astonishmentto the Etruscans. Indeed, in Livy, and in other accounts too, the emotional engagement of these audiences of witnesses receives as much attention as the performer's own actions, if not more.2According the audience such prominenceforegrounds 2. Forthe audience'sengagement,notethattheEtruscansfocalizemuchof Livy'saccount:in ?5, it is to theirgazethatmostRomansturntheirbacks,butHoratius(to theiramazement) turnshis weapons;in ?9 thehesitationandshamearetheirs.ButtheRomanstoo areengaged;notetheircheer(?10).
4
B. ROLLER MATTHEW
the fact that the action is meaningful and relevant with reference to the values and interests of the community (or communities) before whose eyes the action is done.3 Second, the audience's evaluation of the deed is central. In Livy, the concrete expression of that evaluation forms the narrativeclimax: the Romans indicate their gratitude(grata erga tantam virtutemcivitasfuit) by grantinga statue and land. In Polybius, it is partlythe positive report(EsKkEta)he expects among the Romans that makes him willing to die. And in both versions, even the enemy acknowledges the hero's qualities.4 Third, commemorationof the deed is functionally and thematically importantin both accounts. The narrativeitself is a self-conscious form of commemoration, remarkinghow futureaudiences will receive the deed: "therenown that would accrue to him thereafter"(Polybius); "boundto gain more glory than credence with posterity" (Livy).5 Livy's account also adverts to a second commemorative device, a statue-erected in the comitium, no less, the political heart of the republicanForum and the most frequented,looked-upon place in the city. Fourth, imitation: in Polybius, this narrativeis adduced as evidence that Romans imitate the glorious deeds recounted in funeral orations, and in Livy's history, another actor in due course is said to imitate him (Cloelia, at 2.13.6-8; see section III below). I have just summarizedthe main features of what I call "exemplary"discourse in Roman culture, a discourse linking actions, audiences, values, and memory. The four principal components of this discourse can be schematized more generally, as follows: 1. An action held to be consequentialfor the Roman community at large, and admittingof ethical categorization-that is, regardedas embodying (or conspicuously failing to embody) crucial social values. In Horatius' case, this category is normallystatedor implied to be virtus, or somethingrelated.6 Virtus, etymologically "behaviorappropriateto a man,"can be a capacious ethical category, especially when used to translate pUEti1in philosophical contexts. In a long-standing Roman ethical vernacular,however, it encompasses a narrower,more specific range of consequential action: a soldier's 3. "Spectacular"events, in which a judging audience figures prominently, are common in Livy and Polybius. For Livy see, e.g., the battle of the Horatii and Curiatii, the Verginia episode, and duels such as Torquatusand the Gaul, with discussion by Feldherr(1998, 127-31, 203-12, 82-111, and passim); for Polybius, see Davidson's overview (1991, 11-18). 4. In Livy the Etruscansmarvel at his audacia; in Polybius, at his ibn6aToat and t6)rpa (cf. n. 2 above). Val. Max. 3.2.1 stresses both the Roman and Etruscanevaluations: unus itaque tot civium, tot hostium in se oculos convertit, stupentes illos admiratione, hos inter laetitiam et metum haesitantes . . . quapropter discedentes Etrusci dicere potuerunt "Romanosvicimus, ab Horatio victi sumus." 5. Two accomplices, Larcius and Herminius, are attachedto Horatiusin some versions of the story (e.g., Livy 2.10.6-7; Dion. Hal. 5.24.1; Plut. Publicola 16.6; Serv. in Aen. 11.642; see Fugmann 1997, 42). But they do not much detract from his individual glory, as they are said to withdrawbefore the collapse of the bridge, leaving Horatius alone to fight and leap into the river. 6. Virtus is Livy's category (2.10.12: grata erga tantam virtutemcivitasfuit); also Culex 358-61, Sen. Ep. 120.7, Quint. Inst. 5.11.10. Other categories: this story falls under the rubric de fortitudine in Val. Max. 3.2.1 (similarly Cic. Leg. 2.10); his &hvpeiais praised at Dion. Hal. 5.25.4; and at Plut. Mor. 317D-E he is an acolyte of Aprtsi, in contrast to those who follow also at Plut. Publicola 16.9, Dion. T6X•l (d&pEti Hal. 5.25.3). At Frontin. Str. 2.13.5 he exemplifies the proper way to retreat(de effugiendo).
EXEMPLARITY IN ROMANCULTURE
5
bravery or steadfastness in battle. This is how the term functions in Horatius' case.7 2. An audience of eyewitnesses who observe this action, place it in a suitable ethical category (e.g., virtus orpietas or gratia), andjudge it "good" or "bad"in that category; I call this the "primary"audience. In most cases this audience is a subset of the Roman community, the group for whom the action is most consequential-but in military contexts, as we have seen, the enemy too may be invoked as a valid judging audience. These audiences, by their very spectatorshipand evaluation, constitute the action as consequential for the community, and thereby transformit into a socially and ethically significant "deed,"a res gesta. 3. Commemorationof the deed-that is, commemorationnot only of the action, but of its consequence to the community, and of the ethical evaluation it received from the primaryaudience.Commemorationoccurs by means of a monument,a device that calls the deed to memory; monumentsinclude narratives,statues, scars or other bodily marks, toponyms, cognomina, and even rituals, to name just a few.8 Monuments aim to make the deed more widely visible by constructing "secondary"audiences-persons who were not eyewitnesses, but who learn of the deed throughthe monument(e.g., by reading the narrative,looking at the statue, or inquiring about the scar). The witnessing of a secondary audience thereforehas a broaderscope than that of the primary audience: for included in the secondary audience's field of view is both the action itself and the primaryaudience's prior evaluation of it. That is, secondary audiences typically see the "deed"alreadyconstituted as such by the primary audience's judgment, freighted with social consequence and ethical significance. Secondary audiences therefore form their own judgments in full knowledge of what the primary audience thought. Clearly, a monument (such as an honorific statue or laudatorynarrative)invites secondary spectatorsto concur with the primaryaudience's judgment, to agree that the action was done well or badly in the pertinent ethical category. But as we shall see, secondary spectators have minds of their own, and do not slavishly accept their predecessors' verdicts. 4. Finally, imitation: any spectator to such a deed, whether primary or secondary, is enjoined to strive to replicate or to surpass the deed himself, to win similar renown and related social capital-or, for negative examples, to avoid replicating an infamous deed. How Romans determineddegrees of similarity, and evaluated or ranked deeds relative to one another, will be discussed below. For now, suffice it to say that the imitator typically seeks to become "the new X" or "anotherX," or at least something comparableto 7. On the ethical domain(s) encompassed by the category virtus in the late Republic and Empire, see McDonnell 2003; also Roller 2001, 20-26; Barton 2001, 34-43, 281-83. See Lendon 1999, 304-16, for (traditional, military) virtus in Caesar; Moore 1989, 5-17, for virtus and related categories in Livy; and Eisenhut 1973 for a survey of occurrences in many Latin texts. 8. See H1lkeskamp 1996, 302-8, on the varieties of monuments implicated in "monumentalmemory." Narratives are regularly classified as monumentain Latin writing: see OLD, s.v., senses 4-5; also Feldherr 1998, 21-37, p. 156 and n. 27, and Jaeger 1997, 15-29. H61keskamp(2001) and Horsfall (1996) discuss how memory inheres in topography.
6
B. ROLLER MATTHEW
X in the socioethical significance of his or her action.9With imitation, then, we come full circle to (1), actions of potential social consequence done before a judging audience of witnesses. Here we perceive a cyclical dimension to exemplary discourse: deeds generate other deeds, spawning ever more audiences and monuments, in an endless loop of social reproduction. This scheme requiresthree immediate comments. First, the community of Romans involved in doing, witnessing, evaluating, and monumentalizing deeds is the populus Romanus at large, not just elites. The actors of monumentalized deeds tend to be elites (victorious generals and the like), but are not elites exclusively. For instance, the most valorous Roman soldier of all, Siccius Dentatus, is represented as sub-elite. Again, certain monumental forms, such as honorific statues and cognomina, were apparentlyreserved for elites. But other, equally compelling forms, like narratives and scars, could be attached to actors of any status; and all these forms could be interpretedby observersof any status.10Even when narrativeswere ensconced in literary texts-which in general presuppose a literate, leisured readership, and functionedprimarilyas medium of communicationamong elitesthey could still be made available to illiterate Romans through recitations and other types of performance." The quantity,variety, and accessibility of monumental forms suggests that actors of every status took care to submit their actions to the scrutinyof a broadcross-section of the people, in whose collective name and interest they normatively acted. Exemplary discourse, then, encompasses all of Roman society, from the loftiest aristocratsto the humblest peasants, laborers, and slaves. Second, exemplary discourse has powerful ideological effects. The schematic form, traced above, exposes what Romans from the late Republic onward took to be the normal or normative way in which social values were established and instilled, deeds were done and evaluated accordingly, and social reproductionoccurred. In some cases, this schema may describe the actual unfolding of an action in the public eye, the evaluation it receives, 9. Literarytexts often present the simple desire for glory as the impulse for imitation (e.g., Polybius). However, some texts present a more complex engagement between spectator and spectacle: see Feldherr 1998, 82-111, on Livy (highly relevant to this paper), and Bartsch 1994 on emperors and aristocratsin imperial Rome. 10. Exemplary sub-elite actors: for Siccius see Gell. 2.11, Val. Max. 3.2.24, Plin. HN 7.101, Dion. Hal. 10.36-38; Caesar's centurions, depicted in B Gall. and B Civ., are similar. Also, Val. Max. 3.2.6 attributes to elites in the good old days (here, the second century B.C.E.)anxiety about being outdone in virtus by people beneath them in dignitas. Hl1keskamp (1996, 303-12, esp. 305, 310) shows how monuments implicate both elites and sub-elites in the same ideological structure. See also Horsfall 1996, 109-14 (and passim) on the range of cultural production consumed by both elites and sub-elites; also Bell 1999, 27376. In general, elite and sub-elite values overlap but are hardly identical, and may coexist uneasily in various social spheres: see, e.g., Alston 1998 and Lendon 1997, 237-66, on the army; also de Libero 2002, 179-85, and Leigh 1995, 200-205, on the class and status implications of wounding. 11. Habinek (1998, 45-59) discusses the acculturativeeffects of early Roman literatureupon elites; but see Bell 1999, 264-67 (and passim), on recitations of Vergil and the availability of the Aeneid to nonreaders. Within literary texts, the audiences described as observing and judging an action are often representative of the Roman people as a whole: for instance, "spectacular"events in Livy (n. 3 above) are observed by the populus Romanus, whether assembled as an army in the field, a mob in the Forum, or even the voting tribes and centuries at the elections (which are anothervenue for the witnessing-and-judgingphenomenon).
EXEMPLARITY IN ROMANCULTURE
7
and its transmutationinto monumental and imitable form. More important, however, is that Romans assumed that actions, audiences, monuments, and social value were or should be linked in these ways. They often acted with a view toward being observed, evaluated, monumentalized, and imitated, and assumed that other people did likewise-even if most actions did not actually achieve so glorious an afterlife. Similarly, any object that looked "monumental"was likely, at some point, to be assimilated into exemplary discourse. For as we will see, the expectation that the elements of exemplary discourse stuck together was so strong that any given element could attractor spawn the others. Thus the ideological salience of Horatius' story is not compromised by the recognition-even in antiquity-that it is fabulous.12On the contrary,this and many other narrativeswere socially efficacious because they manifest a narrativestructurein which action, judging audiences, commemoration, and imitation are all present and all work together. Each element presupposes and implies the others; it is the ensemble that gives meaning to each part. Third, while this (simplified) scheme may leave the impression of monolithic, seamless coherence, in fact the productionof exemplary discourse is beset at every turnby instabilities,contradictions,andcontestation.An action may be evaluated positively in one ethical category, but negatively in another; or perhaps different aspects of an action carry divergent value. How are these conflicting judgments to be reconciled or weighed? An object that some viewers interpretas monumental,hence part of exemplary discourse, can be rejected by other viewers, who contend that its appearanceis deceptive and it has no monumentalquality at all-or there may be disputes about precisely what, or whom, a certain monumentcommemorates.Finally, how one goes about imitating an exemplary deed-what constitutes legitimate imitation, and whether a given actor has produced one-is often fiercely disputed. To produce an exemplum, then, is to struggle constantly to establish or disestablish a particularinterpretationof an action's value, a monument's reference, or an imitator's success, and alternativereadings threaten to (or do) proliferate at every instant. But far from underminingthe ethical cogency of the exemplum, these ubiquitous opportunities for debate and contestation are the lifeblood of exemplary discourse-this is how every example can be made anew, or deployed in a novel way, to meet the requirements of any new contingency. My aim in what follows is twofold. First, I seek to demonstratethat exemplary discourse, as schematized above, describes an actual Roman way of confronting the past, of giving it value and purpose. Indeed, I argue that the socioethical dynamics of exemplarity are fundamentalto Roman historical consciousness itself. For a Roman, the question of what the past is, and 12. Doubts: rem ausus plus famae habituram ad posteros quamfidei (Livy 2.10.11); tunc illa tria Romani nominis prodigia atque miracula, Horatius Mucius Cloelia, qui nisi in annalibus forent hodie fabulae viderentur(Flor. 1.4.3). But the phrasesplus famae habituramad posteros and prodigia atque miracula show that these deeds are admired, monumentalized, and imitable despite their questionable facticity.
8
MATTHEWB. ROLLER
what it is for, is closely tied up with the monuments that mediate his or her encounters with the past-monuments in narrative,plastic, or other formthat sort actions into various ethical categories and then re-present them as deeds to later audiences in an injunctive, hortatoryrhetoricalmode. Second, I argue that this model of exemplary discourse enables us to grasp the social and ideological functioning of exemplary figures in any given present, and in any particularcontext, in which they appear. I pursue this double aim by examining two exemplary figures, as they appearin narrativesand other monumentalforms from the middle Republic onward: Horatius Cocles in section II, and Cloelia in section III. Individually and as a pair, these figures admit of especially productiveanalysis. They are among the most frequently adduced exemplary figures in surviving Roman texts, and were commemoratedin a varietyof monumentalforms. Moreover, each actor's deed was done in the same mythistoricalcontext, the war with Porsenna of the second or third year of the Republic. This, together with several structuralparallels between their deeds, caused Roman writers to pair and comparethem frequently.Because of their frequentattestationin many different contexts, these figures enable us to study in some detail the exemplary discourse in which they participate-how that discourse functions in the various contexts in which it appears; also, where and why it generates contestation and instability in producingthese figures as exempla. Throughthese figures, then, we can learn much about how and to what ends Roman society constructed and consumed its own past. The relationship of this work to previous scholarship also warrantsbrief discussion, since the idea that actions, audiences, values, monuments, and memory are interrelatedis not radically new. Within Roman studies, numerous scholarlyworks have examined linkages among these phenomena,focusing on particular periods, problems, authors, or objects-examining, for instance, the "theatricality"or "spectacularity"of Roman public life, or the ethical implications of human action, or the audiences for and "messages" of honorific statues. Beyond classics, social and culturaltheorists have formulated the concept of "collective (or social) memory,"and philosophers have examined the problematics of arguing from part to whole, from specific to general.13For Roman studies, however, the work of examining these elements together in their cultural wholeness has barely begun. The structure of subfields within the discipline of classics is partlyresponsible:while texts and visual objects (obviously) were originally parts of a cultural continuum, classicists are typically trainedto work with just one corpus of material or another.For the student of exemplarity, then, one challenge is to bring what archaeologists and art historians know together with what literary and historical scholars know, to comprehend the relationship and con13. For "collective memory"or "social memory"the locus classicus is the work of Maurice Halbwachs, with recent investigations by Aleida and Jan Assmann (extensive bibliography in Hl1keskamp 1996 and especially 2001); in cultural history, Pierre Nora's Les lieux de m~moire deserves pride of place, along with the journal History and Memory. The philosophical investigation of examples is most closely associated with Immanuel Kant, with recent interventions by Jacques Derrida. I will discuss elsewhere questions these works raise for Roman exemplarity.
IN ROMANCULTURE EXEMPLARITY
9
tinuitybetweentwospeciesof monument-namely imagesandtexts-in
Romanculture.A secondchallengeis to bringdiscreteculturalphenomena,
suchas the"theatricality" of Romansocietyorthe"messages" of images,
into focus as partof a largerwhole. Karl-Joachim Hilkeskamp'sprelimiarticle"Exemplaund mos maiorum:Uberlegungen narybut pathbreaking zumkollektivenGedachtnisderNobilitat"(1996)showsthepromiseof pursuingthis syntheticapproach. This paper'saims also differfrommost earlierstudiesof the traditions the warwith Porsennaandits Romanheroes(includingHorasurrounding tius and Cloelia).Threeapproachesto this topic have predominated. The
firstis Quellenkritik--the to sortthemanyversionsof thesenarraattempt tivesintofamilygroupsbasedon linealdescentfromotherversions,often withtheaimof recovering theseearlierversions,oreventheoriginalone. variouscomponents of thenarratives, considerQuellenkritik distinguishes earlier some and others or some better andothersworse.Regardlater, ing
ing the legendsof the warwithPorsenna,somescholars(forinstance)have soughtto identifyandathetizecertainelementsas "late"accretions,introducedintoa preexistingformof thelegendsby thehistoriographer Valerius scholarsof comparative Antias.14Second,structuralist religionhavelinked thesestorieswithlegendsfoundin othercultures,seekingto identifyshared structureswithinIndo-European (andother)mythandreligion.ThusHoratius Cocles has been connectedwith otherlegendaryfigureswho lack an
or eyeand/orlimb,andareseeneitherasembodying aspectsof sovereignty
as resisting tyranny.15This approachcan be combined with Quellenkritik, so that comparativereligion provides the ur-version that some source critics posit. A third approachseeks to recover the historical or cultural actualities that lay behind the production of these legendary tales. Thus it is argued that the stories of Horatius,Cloelia, and Mucius Scaevola provide a fig leaf by which Romans dissembled the fact that Porsenna captured the city; or that these stories describe archaic religious rituals in a transmuted,misremembered form; or that these stories received their form and content from a mid-republicannarrativeoral tradition;and so on.16These approachesall 14. For Quellenkritiksee, e.g., Munzer 1913, cols. 2331-36 on Horatius Cocles, esp. col. 2336.12-15: "Fuirdie Entstehungund den Wert der Oberlieferung ist aus solchen spaten Zutaten und Anderungen nichts zu lernen, ebensowenig aus ihren zahlreichen Anfuihrungenbei spiteren Autoren" (i.e., some versions are "late"and "worthless"for establishing the origins of the story). Wiseman (1998) discovers numerous fabrications by Valerius Antias in the legends of early Rome (esp. 83, on Horatius). Fugmann (1997) offers a comprehensive, nonjudgmentalsource-critical analysis of the Horatius and Cloelia legends (37-49, 60-67, respectively). 15. Dumezil ([1940] 1988, 143-48) argues that Horatius Cocles and Mucius Scaevola have parallels in Norse mythology and elsewhere, and that such figures are expressions of an original Indo-Europeanoneeyed and one-handed god representing magic and legal aspects of sovereignty. Moeller (1975) offers a Dumizilian analysis of various one-eyed and one-legged figures from Roman mythistory (including Horatius). Lincoln (1991, 244-58) critiques Dum6zil's approaches and conclusions, arguing that such figures representresistance to, or embody the antithesis of, illegitimate kingship. 16. For the view that Porsenna,if historical, actually capturedRome (so Plin. HN 34.139; and Tac. Hist. 3.72), see, e.g., Fugmann 1997, 38; Alfildi 1965, 72-75; cf. Cornell 1995, 217-18. Gag6 (1988, 242-45) challenges the legend's historicity in a different way, arguing that there was no war with Porsenna, but an economic crisis misrememberedas a military siege. Scholars who see the origin of these stories in oral poetry that incorporated elements of Indo-Europeanmyth (thus accepting Dum6zil's conclusions in some
10
B. ROLLER MATTHEW
seek origins and univocality: the single earliest form of the legend, the period and sociohistoricalcontext of the legend's formation,the Indo-European gods or heroes, the actualevents. Scholarsemploying such approaches,therefore, seek to discover the original events, structures,and meanings that they believe are implicit in, but concealed, misremembered, or transmutedby our texts, due to the lapse of time or distortion in the tradition. For my part, I will not seek to discover original versions and lost meanings. My aim is to grasp the meanings and functions of Horatiusand Cloelia in precisely those phases of Roman society in which our surviving narrative accounts (and other monumentalforms) were produced and consumed. For the object of my study is Roman exemplary discourse itself, as manifested in surviving texts and other representations-the discourse within which Romans encompassed these legendary figures. This discourse, in all its cogency, complexity, and contradiction, lives in and through the multivocal narratives,the welter of contested variations, and the multiplicity of uses to which Romans, at any given moment, put their exemplary figures. II. HORATIUS COCLES Monuments We have examined two detailed accounts of what Horatius did, and noted the prominence of eyewitnesses (Romans and Etruscans)who attest that the action is consequential for their community and provide its initial positive evaluation. Furtheraspects of the action itself and the primary audience's engagement will emerge below. But let us pursue here the third component of exemplary discourse, as schematized above: namely, commemorationhow knowledge of the action andof the evaluationit initiallyreceives is transmitted to wider audiences, who respondto that knowledge in turn.For Horatius' deed, as for most exemplarydeeds, narrativesare arguablythe premier monumental form. This is not only because narratives happen to survive. There can be no commemoration of any sort without narrative,since even nonnarrativemonumental forms explicitly refer to, or implicitly require, a narrativethat accounts for their occasion." But just as nonnarrativemonuments typically imply a narrative,so narrativesoften refer to nonnarrative monuments. This "cross-referencing"suggests that any single monumental form was seen as just one element in a systematic marshalling of resources for preserving and transmittingmemory, each element of which reinforces
form), and consider the "historicization"of these myths a later development, thereby bracket all questions of historicity (e.g., Ogilvie 1965, 258; Forsythe 1994, 253; Frier 1979, 59). That the stories of Horatius and/or Cloelia conceal/reveal ancient rituals involving the Tiber or the pons sublicius: Gage 1988, 236, 241; 1973, 10-12; Ogilvie 1965, 258. For recent surveys of scholarly opinion see Cornell 1995, 215-18 (on the whole end-of-the-Tarquinssaga), and Forsythe 1994, 252-57 (on Horatiusand Cloelia in particular). 17. So Habinek 1998, 49-50. Source critics sometimes dismiss narrativesthat account for other monuments as "aetiological" myths, that is, secondary, inferior accretions that are merely retrojectedfrom these monuments, and carry no independent explanatory value. In exemplary discourse, however, such narratives and the monuments they explain are parallel means of commemoration-indeed, the narrative has ideological primacy because it is explanatory,regardless how and when it was actually produced.
EXEMPLARITY IN ROMAN CULTURE
11
and complements the effects of the others.18In the next few pages I examine five monumental forms associated with Horatius-all but one "crossreferences" in narrative accounts-considering how they function qua monumentswithin exemplary discourse, and also how they function in their particularsocial and literary contexts. 1. To begin with the most obscure monument: Propertiusin Poem 3.11 implies that the temple of Apollo at Leucas, which commemorates Augustus' victory at Actium, eclipses all prior monuments and the great deeds they represent. He names several particular monuments that the temple overshadows: among these are (59-62) "the spoils of Hannibal,""the monument of defeated Syphax," "the broken glory of Pyrrhus,"the Lacus Curtius, and "the path of Cocles [that] bears witness to the cutting of the bridge" (Coclitis abscissos testatur semita pontes, 63). This "pathof Cocles" is otherwise unknown, but must have been a reasonably familiar feature of the urbanlandscape, one whose name linked it with Horatius' deed and so called that deed to the memory (testatur) of those who knew of it.19 2. Similarly, the pons sublicius itself may serve as a monument to Horatius' deed. The elder Pliny notes-tangentially, in a discussion of marblethat the bridge was built entirely of wood and had no iron nails, "because it was torn down with such difficulty when Horatius Cocles defended it."20 Thus this curious feature of the city's built environmentserves as a monument to the deed, summoningit to the memory not only of Pliny's (probably elite) reader,but of anyone at all, of any status, who might inquire about the bridge's distinctive construction. And just as the bridge commemoratesthe deed, so the deed explains the bridge. This explanation, as Pliny gives it, is purely pragmatic, though it also accounts for the taboo against the use of iron in the bridge's construction: because iron fastenings so hindered the work of Horatius' companions, an ironless building technique was adopted thereafter,in case the need should arise again.21Modern scholars reject this explanation for the bridge's construction, and the Horatiusconnection with it.22But my point is that at least some Romans were preparedto comprehend 18. For narratives referring to honorific statues, see below. For honorific statues accompanied by explanatory inscriptions, the summi viri of the Forum Augustum are famous instances (among innumerable others). A more complex form of cross-referencing is seen on coins that carry images of honorific monuments, thus commemorating not so much the original deed as the fact that this deed has already received monumental commemoration. See, e.g., a denarius minted by Marcius Philippus (56 B.C.E.),whose reverse shows the arches of the Aqua Marcia surmountedby an equestrian statue (discussion in Crawford 1974, no. 425/1, and Bergemann 1990, 35); also a group of coins dating to the 110s C.E.showing Trajan's column (Claridge 1993, 15-16). 19. The text of Prop. 3.11.59-70 is vexed, leaving many readings and the ordering of the couplets in doubt. Nevertheless, the thematic importance of monuments, memory, and witnessing is clear: see Gurval 1995, 203-7, and Shackleton Bailey 1956, 174-75. 20. Plin. HN 36.100: quod item [sc. being constructed sine ferreo clavo] Romae in ponte sublicio religiosum est, posteaquam Coclite Horatio defendente aegre revolsus est. 21. Dion. Hal. (9.68.2) offers a similar explanation, though without reference to Horatius:the Tiber, he says, could only be crossed by means of a bridge, i1 iv iv TjP t6rs Xp6v(vp piaS0u66ppaKrro,i)v ~huov v Tzo! nokipots. Cf. Dion. Hal. 3.45.2, 5.24.1, and Serv. in Aen. 8.646, all implying that the bridge was already constructed of wood in Horatius' day; also Plut. Num. 9.2-4. 22. Some scholars prefer a religious explanation, and regardHoratius' leap either as a devotio or piaculum for the bridge itself, or as a doublet for the throwing of the argei: see Delcourt 1957, 177-78; Le Gall 1953b, 80-86 (esp. 81-82 on the wooden construction); Le Gall 1953a, 78-82 (on its religious import).
12
B. ROLLER MATTHEW
this otherwise obscure curiosity as a monument, connecting it with an exemplary deed that was bound to be imitated in due course ("easier to tear down next time ..."). That is, they made sense of it by integrating it into exemplary discourse. 3. A third form of monumentconsists in a physical trace of the deed that the hero carries aroundwith him ever after. While Polybius has him die in the act and Livy has him escape unscathed (likewise Val. Max. 3.2.1 and Sen. Ep. 120.7), most accounts assert or presuppose a different outcome: that he indeed survived, but suffered a crippling wound to his hip. Servius, explaining why Horatius was depicted on Aeneas' shield, tells the story as follows (in Aen. 8.646): et cum per sublicium pontem, hoc est ligneum, [sc. Porsenna] transire conaretur,solus Cocles hostilem impetum sustinuit, donec a tergo pons solveretur a sociis: quo soluto se cum armis praecipitavit in Tiberim, et licet laesus in coxa, tamen eius fluenta superavit: unde est illud ab eo dictum, cum ei in comitiis coxae vitium obiceretur "per singulos gradus admoneor triumphimei." ... and when [Porsenna]sought to cross over on the pons sublicius, that is, the wooden one, Cocles alone held up the enemy's attack until the bridge could be torn down behind him by his companions. Once it was down he threw himself into the Tiber wearing his armorand, though wounded in the hip, still overcame its current.Hence the famous bon mot of his, when during the elections his hip injury was held against him: "With every step, I am reminded of my triumph."
The point of Horatius' bon mot is that his detractorstoo should be reminded of his great deed when they see him limping along. War wounds, and scars in particular,appearfrequently in Roman texts as markersof valorous conduct. Wounds incurred in the proper way, located in the right part of the body, inscribe into the living flesh of the hero the record of his valor, for all the world to see, so long as the hero lives. Narrativesof the conflict of the orders, for instance, portrayveterans who are at risk of debt bondage baring their scar-covered chests to public view; likewise in Cicero and in the rhetorical treatises we hear of defendantswhose lives or propertyare at risk exposing their scars to the judges. The aim of such display is to persuade the audience that this person's record of valiant military service to the state, attested by his scars, should earn him favorable consideration in the current situation.23Horatius' case, as Servius presents it, is similar. By alluding to the limp's origin, Horatius transformsit into a monument to his deed, one that should summon his defense of the bridge to the memory of his interlocutors. Yet, a wound is a contestable, unstable monument.It does not convey (as many monuments do) whether the action it attests was deemed consequential for the community, or what evaluation it received from a pri23. Displayof scarsby personsfacingnexum:Livy2.23.3-7,2.27.1-2,4.58.11-13,6.14.3-8. By those
standing trial: Livy 6.20.8-9; Cic. De or. 2.124, Verr.2.5.3, Pro Rabirio perduellionis reo 36; Sen. Con-
trov.9.4.7; Quint.Inst. 6.1.21, 30. Also Curt.9.3.10, 10.2.12;Tac.Ann. 1.35 (soldiersdemandingdischarge);andDion. Hal. 10.36-38 (plebeiandemandingagrarianreform).De Libero(2002, 175-79, and passim)andLeigh(1995,205-12) discussthevaluesandsocialaimsassociatedwithdisplayingscars;also Gag6(1969, 196-202).
EXEMPLARITY IN ROMANCULTURE
13
mary audience. Presumably, his limp was "held against him" by a viewer who inserted the wound into a different discourse, one in which physical imperfection correlates with moral and social inadequacy.Indeed, this ambiguity-Is the disfiguredperson turpis orfortis? Are the markson his body dehonestamentaor decora?-is endemic in Roman discussions of wounds; the wounded person is always on his mettle to dress his injury persuasively in the discourse of exemplarity,lest others make it a grounds for despising him.24Providing a narrativeof origins, as Horatiusdoes here, achieves two key ends at once: it inserts the injury into exemplary discourse by asserting its monumentality,and seeks to stabilize thatdiscourse by supplyingwhat the wound inherentlylacks as a monument,namely,thatit does not by itself convey whetherthe deed was consequentialand how it was originally evaluated. To what end? The reference to elections suggests that Servius imagines that Horatiuswas standing for public office, when an opponent asserted that his disability made him unsuitable. By "monumentalizing"his limp, however, Horatius turns his detractor's argument on its head, converting the limp from a liability to an advantage in the eyes of the voters; for the populus Romanus, assembled into its constituent tribes or centuries for voting, is the main audience for Horatius' words (and limp). It is they who cast the votes, and who constitute the community for whom his deed is consequential-particularly so in an electoral context, since they would not be voting at all had he not preserved the res publica. Like those who display their scars in court or underother circumstancesof risk and opportunity,Horatius is presented here as attemptingto convert one form of social capital-the deference and respect that the populus Romanus grants him in return for meritoriousservice on their behalf-into anotherform, in this case the honor of holding a magistracy, which he hopes will be conferred upon him now, in gratitude and as furtherreciprocationfor his deed. Servius has reason to imagine that Horatius would have sought to exchange his social capital in this way. In Sallust's Bellum lugurthinum,Marius, in a speech as consul, reminds the people that they elected him not for his noble ancestry,but for his military decorations and "scarsin the front of the body" (cicatrices advorso corpore, lug. 85.29). Likewise Plutarch, in
24. For these alternative and competing discourses about wounds, cf. Publ. Sent. N 12: non turpis est cicatrix quam virtus parit; also Cic. De or. 2.249 and Plut. Mor. 331B-C, where persons disabled by wounds initially feel shame about going forth in public, but are reassured that an honorable reading can be maintained by allusion to the wound's origin; cf. Plaut. Curc. 392-400, Curt. 4.14.6. Excellent discussion of these matters in de Libero 2002, 183-87. Scars convey "inherent"information about their origin only in certain locations: wounds to the front are "good," showing one faced the enemy; those to the back are "bad,"suggesting one fled-or, if they are whip marks, evincing slavish treatment(cf. Leigh 1995, 19699). Wounds located elsewhere, such as the hip (Horatius) or the testicles (M. Servilius Geminus Pulex, Livy 45.39.16-20, Plut. Aem. 31.8-9), are therefore especially ambiguous, even more in need of narrative buttressing. One tradition locates Horatius' wound in his buttocks (Dion. Hal. 5.24.3, Plut. Publicola 16.8), which Wiseman interprets (1998, 83) as an invention of Valerius Antias-an attempt to make Horatius look undignified so that Valerius Publicola would look better by comparison. His Quellenkritikhere is sheer conjecture, and implausible anyway: as the Servilius Geminus example shows, even a wound in an "undignified" place could potentially be narrativized into exemplary discourse and so constructed as a monument to a great deed.
14
MATTHEWB. ROLLER
his Quaestiones Romanae, suggests a reason why office seekers used to canvass in togas without wearing tunics underneath:perhapsthey wanted their scars to be visible to the voters (Mor. 276C-D; cf. Coriol. 14.1-3).25 As these texts show, the idea that elites whose bodies attest their martial valor might on this ground seek election had broadercultural currency,so its application to Horatius is unsurprising.And Horatius, indeed, seems to have been imagined as an elite. The Horatii were a patrician clan, to whom tradition ascribed other prominentmembers in the regal period and early Republic; Dionysius of Halicarnassus(5.23.3) makes Horatiusthe nephew of the consul Horatius Pulvillus, and a descendant of the surviving member of the Horatii who defeated the Curiatii in the reign of Tullus Hostilius, thereby securing Rome's ascendancy over Alba Longa.26 Thus Horatius could be imagined as consular material by birth as well as for his achievements. Servius does not report the election's outcome, but other accounts say that his crippling wound left him unfit for the duties of high office, wherefore he never held the consulship or other military command.27Perhaps his absence from the consularfasti, despite high birthand achievement, spurredthe fabrication of a reason why he was never elected (so Miinzer 1913, col. 2334)-especially since Larcius and Herminius, his forgettable accomplices in some accounts, appearin one chronology as the consuls of year four (Dion. Hal. 5.36.1). At any rate, while his wound constitutes the social capital on which he was imagined to trade in standing for office, it also provides a pragmaticreason why this exchange was not consummated. While Servius presents Horatius as monumentalizinghis own wound, in other texts other social actors invoke that wound to summon his deed to memory. In Book 45 Dio Cassius places in Cicero's mouth a "Philippic" against Marc Antony, ostensibly delivered in the senate early in 43 B.C.E. Speaking of the incident at the Lupercalia where Antony offered Caesar a 25. See Leigh 1995, 195-96, 202-3, on these passages. He argues that the rhetoric of military achievement, attested by scars, surroundingfigures like Marius and the elder Cato (e.g., Plut. Cat. Mai. 1.6-10), is necessitated by their novitas, as they have only their achievement, and no distinguished ancestors, to recommend them for election. True, but aristocrats of distinguished lineage are also said to display scars to secure favorable outcomes: see on M. Servilius Geminus, cos. 202 (Livy 45.39.16-20 and Plut. Aem. 31.7-10), M'. Aquilius, cos. 101 (Cic. De or. 2.124, 194-96, Verr. 2.5.3; Quint. Inst. 2.15.7), and, for a much earlier period, M. Manlius Capitolinus (Livy 6.14.6-8). 26. For prominent Horatii of the early Republic, see Miinzer 1913, cols. 2328-31, 2400-2404 (s.v. "Horatius,"nos. 6, 7, 8, 13, 14, 15)-Pulvillus is no. 15. Dionysius of Halicarnassus ascribes lofty birth to the Horatii triplets: 3.13.3, 17.4-5. 27. So Dion. Hal. 5.25.3: XPrlotog88' E Eig ot&xot itpdiypata Tig S8td lV nTlrpoatv trig Ldot)E rn6•esg Kai&&d tilv oup(popdvzTanrTv o06i'bitaCifa oir' ikkrlg orparntrKcqig o8Eptsqg CruXev; also Appian, jyltEoviag Basilike, frag. 10. Cf. Dion. Hal. 9.13.4 for a consul resigning before the end of the year, because bedridden with a wound. These are pragmaticexplanations: the demands of office exceed these persons' physical capacities (discussion and parallels at de Libero 2002, 179-80, 187-88). One may also wonder about a ritual explanation, that is, that a magistrate with religious functions-a consul, for instance, may take the auspices-must, like priests, be free of bodily defects (see Dion. Hal. 2.21.3; Sen. Controv.4.2; Gell. 1.12.3; Fronto p. 146 van den Hout; Plut. Mor. 281C; with furtherdiscussion by Garland[1995, 63-65]; Wissowa [1912, p. 491 and n. 3]; and Mommsen [1887, 1.493-94 and n. 1]). Moreover Lincoln (1991, 248-49) provides comparative evidence that a king's body must be whole for ritual and other reasons. Yet this explanation cannot hold uniformly since, as just observed, war wounds could also be presented as a positive recommendationfor high office. A middle path may be visible at Plin. HN 7.105: M. Sergius Silus, heavily debilitated by war wounds, held the praetorshipbut was barredby his colleagues from religious functions (. .. cum in praetura sacris arceretur a collegis ut debilis); discussion by de Libero (2002, 172-75, 187-90).
EXEMPLARITY IN ROMANCULTURE
15
crown, Dio's Cicero demands, "Have we, who expelled the Tarquinsand rejoiced in Brutus' deed, who threw Capitolinus from the rock and killed Spurius, have we commanded you to install someone as king over us ... ? No, by the rods of Valerius and the law of Porcius, no, by the leg of Horatiusand the hand of Mucius, no, by the spear of Decius and the sword of Brutus!" Here Horatius' deed is indexed by reference to his (wounded) leg, which therefore serves as an icon of freedom, along with Valerius Publicola's loweredfasces, Mucius Scaevola's incineratedright hand, and the rest: each is a monumentcalling to memory the great freedom-preservingdeed in which it was involved, and which Antony's actions threatento undo.28In Book 46, Antony's ally Fufius Calenus replies at length, also before the senate. He says that Antony, in fact, made certain that Caesar would reject the crown, because he chose to offer it in a way and in a place that made it impossible to accept. Thus he secured Rome's freedom from this would-be tyrant:"Antony did not break his leg for nothing so that he himself might escape, nor burn off his hand to scare Porsenna,but ended Caesar's tyrannyby wisdom and cunning, surpassing the spear of Decius and the sword of Brutus."29 Calenus' meaning is not entirely clear when he implies that Horatius"broke his leg for nothing so that he himself might escape," but he must mean, at least, that Horatius' action was somehow ineffectual, and besides aimed for self-preservation rather than a greater good. In any event, he is rejecting Cicero's use of the injury as an icon of libertas, and imposing some lesser meaning upon it. By contrast, the really effective defender of libertas is Antony, whose actions at the Lupercalia far surpass any of these overrated ancient deeds. Thus Antony is presented either as actually doing what the ancient models are wrongly credited with having done, or as doing it better. On some accounts, a second monument to Horatius' valor is inscribed into his body-and attached to his name as well. For the cognomen Cocles is explained as meaning "one-eyed,"either by derivation from co- and ocAnd why this cognomen? ulus, or as a corruptionof the Greek K•UjK)W.30 Several accounts say that Horatius earned it by losing an eye in an earlier battle; thus he already bore an onomastic (and bodily) monument to his martial valor before ever stepping onto the pons sublicius.31The claim that Horatius fought heroically once before is only ever found as an explanation 28. Dio Cass. 45.32.3: . . o.06 P6 z 1 i6Poug rs ph o6cb T Oakepiou Kaicbv v6pLovt6v 1opKiou, o06P oG &T 66pu TOAKioDouKai Tzb (pogTbBpo6zou(cf. 45.31.2). og T6 'OpaTiou Kai Tilv XClpa Tlv MouKiou, 06 Also Verg. Aen. 8.646-51, where Horatius and Cloelia, shown together on Aeneas' shield, are said to have defended libertas against Porsenna/Tarquin(Gurval [1995, 223-24] discusses the theme of freedom on the shield); similarly Livy 2.10.8; and Juv. 8.262-65. 29. Dio Cass. 46.19.8:... o0 Yvaa'it p06yfl, o66 XEipaKraKac aov-og Y'va koXXsg KqKaTrdavTog o6hKog 6h& zilv tupavvi6a TilvToO Kaiocapog oo0pia Kai 7rEptlrXviot, Kai Intp T 66pu z6 Hoporvvav (popiloni, AeKiouKai 6p o Tr6 i(pogzb Bpo6tou, ra6coavTog. 30. Derived from oculus: Varro,Ling. 7.71, ab oculo cocles, ut ocles, dictus, qui unum haberet oculum Plut. Publicola 16.7; Enn. (cf. Suda e 1610, K 1921, o 118). Derived from or synonymous with KU6Kd•dC: Sat. 67 (apud Varro, Ling. 7.71). For codes simply meaning "one-eyed" (with no specific etymology implied), see, e.g., Plaut. Curc. 392-94; Plin. HN 11.150 (distinguishing the cognomen Cocles as "one-eyed from birth" from Luscinus, "having lost an eye"); Serv. in Aen. 8.649 (cocles as the older word for what we now call luscus). Furthercitations at TLLOnomasticon, s.v. "Cocles." 31. Dion. Hal. 5.23.2: 8' 'Opdaitog6 Kako6pEvogK6Kk;9gKTzoi KcTa TlJv6itv v PtdXD tov -TpovI6rtXtoS alio proe6(pea•p6v;De vir. ill. 11: Horatius Cocles illo cognomine quod in.akzTTopmcTO oKKorTCoiqg lio oculum amiserat; Suda E 1610; Plut. Publicola 16.7 (quoted n. 33 below).
16
MATTHEWB. ROLLER
of his cognomen, suggesting that this earlier deed is indeed an "aetiological myth,"retrojectedfrom the name for which it accounts. Miinzer, indeed, declares this back story a "lamentable evasion" ("eine klagliche Ausflucht," 1913, col. 2335) necessitated because no one could find a way to connect the cognomen to his defense of the bridge, the only deed for which he was known.32What this explanation illuminates, however, is the Romans' propensity to order their world according to the cognitive frameworkof exemplary discourse. Upon encountering an object, like a cognomen or wound, that might be monumental (because such objects commonly were), they were disposed to comprehendit by means of a narrativethat confirmed its monumentalityby integratingit into exemplary discourse. Thus, the "battle wound" explanation of the cognomen Cocles asserts that the hero was already decoratedwith a monumentalname, which simultaneously pointed to a visible wound and averredthat this wound was received in a consequential act of military valor.33In defending the bridge, then, he imitates his own prior deed of valor, which itself provided the model and standardfor his future actions; he initiates and terminates his own exemplary loop. Thus, throughexemplary discourse, the Romans renderthe otherwise obscure cognomen comprehensible, transformingit into a monument freighted with social and ethical weight. 4. Only one monument commemorating Horatius' deed survives independently of literarytexts. The object is a bronze medallion (fig. 1), minted by Antoninus Pius between 140 and 144 C.E.,as the legend cos iii indicates. Numerouscoins and medallions from this periodfeatureearly Romanthemes on the reverse; many scholars explain this thematic cluster as anticipating the nine hundredthanniversaryof the city's founding in 148.34On the reverse, the legend COCLESidentifies the hero explicitly, lest anyone miss the reference. As viewers of this scene we look downstream, so that the Etruscans are on the right and the Romans on the left. The pons sublicius-shown arched, yet clearly supported by piles-passes between the belligerents. 32. Just one account, to my knowledge, connects the lost eye with the defense of the bridge: Plutarch (Mor. 307D-E [Parallela Graeca et Romana]) says that an Etruscan arrow struck his eye before he swam back to the Roman side. This event presumablyexplains the cognomen, though Plutarchdoes not make the connection explicitly. This unique version (differing even from Plut. Publicola 16.7-8) was likely generated to bring the story into line with the Greek "parallel,"given immediately before (307D): Philip II of Macedon suffering an arrowto the eye while swimming a river near Olynthus. 33. By this explanation Cocles is a precise parallel for Mucius Scaevola, whose cognomen is said to refer to his missing right hand, thus both cross-referencing the mark already present on the body and attesting that the primaryaudience approved the deed whereby he acquired that mark. Plutarch (Publicola 16.7) offers both this and an alternative explanation for the cognomen Cocles: 6 86''Opdttog bv K6Kkrlv ~aw0v6[tov oXEv, isv TiCZv 6tgs'TCp)v 06tEpov KKOnEi7Si( 6' i''Tpot •t•it tf tvb6 kEyooot, This oI6Ptrl•y nOkX.p Giorc KaliT&s6(ppUqouyfceKuoart. alternative, where 6v&E&UKuirgs, prlbT18v c6b&opiyovTl&5ppara t •'vat physiognomic anomaly, deprives it of monumental status and removes it the cognomen marks a simple from exemplary discourse. 34. See, e.g., Toynbee 1925; alternative explanations by Coarelli (1999, 112) and Krumme(1995, 21220). Discussion and photographs in Krumme 1995, 134-36 (and generally at 203-20), cat. no. 69/1 and fig. 137; Gnecchi 1912, 2.9 and pl. 43.4 (image reproducedat Lexicon TopographicumUrbis Romae 4.436, also appears on fig. 38); Le Gall 1953b, p. 81 and pl. 4; and Banti 1983, 2.3.58, no. 52. The legend COCLEs the obverse of a restoration denarius of Trajan (after 107 C.E.), accompanying a head of Roma; however, the original coin of the early second century B.C.E.lacks this legend, which is therefore a Trajanicmodification. On this coin see Krumme 1995, 138, 190-92, cat. no. 48/1; also BM Coins, Rom. Emp. 3.138.
EXEMPLARITY IN ROMAN CULTURE
".-:f .~iik?rc~,
I-r
f c3
r
C ~Jt; d~? 6r~
a?i *. P ,,i.l;
~
~
,,
r:
~s;e ?1~3 ~t~~c~ rC
''?Ij) J1
;3
i
"il-~~ P*l/i~*?rC3 P
2~"~
r,
?~~
~,~3 ,, ~,' c??
c~i~3~
t~
3
i
cr h?u~ia
rrf~
r
?t~ "
t,
~9"1 'zP
(1.
Ii
rB, ??m';;f
r ~
;ii~t*r
:~t
c:
~..s'3~
C?I ca C'r ;rr
t~:
~I
$?
1?,
~1
I'
I,~?J~-l 1
"4:.. ?~
,.,.~?? ?
17
C __Ya~t~-?le~e?l-4?.~?" rrr" ,?1.,
FIG.1. Bronze medallion, mintedby AntoninusPius between 140 and 144 C. E.,from W. Froehner, Les midaillons de 1'Empireromaindepuis le regne d'Augustejusqu'a Priscus Attale (Paris, 1878), 60.
The attention that these opposing forces lavish upon the hero in the water brings out the spectacularityof his action, constituting the "public eye" in which it occurs. One Etruscan poises a javelin, while Horatius swims throughsubstantialwaves to the Roman side, his shield slung over his back, his helmet on his head, and apparently(at least as drawn here; the photographs are less clear) still wearing his lorica. On the left bank, one Roman wields the axe with which he has just chopped a gap in the bridge. Another at the far left extends his right hand, palm upward, toward the momentous scene unfolding before him, perhaps a demonstrativegesture ("look!") but also one of reception and welcome for the returninghero.35This gesture makes clear that we are not merely seeing Horatius in action before an audience of eyewitnesses: we see the moment at which that primaryaudience (or at least the Romans) marks his action as noteworthy and, by welcoming him, judges it beneficial to the collective. With this gesture, the primaryaudience constitutes the action as a "deed."Thus this medallion is a perfect monument, as defined above: it records the action itself, along with the eyewitnesses' positive evaluation of the action as consequential and good; it then places all this information before our-the (secondary) spectators'eyes, inviting us to reproducethe primaryaudience's approval. Who the original secondary audience, actual or intended, may have been is unclear. Medallions were were not intended to circulate like coins, and were normally struck in limited numbers for narrowdistribution.Intended 35. For similar, roughly contemporarygestures of reception and welcome (adventus), cf. the Dacians' of Trajan'scolumn (Brilliant 1963, p. 126 and fig. 3.53). Krumme reception of Trajan,in scenes LXXXIX-XC (1995, 136) seems to interpretthis gesture as an adlocutio, but in this situation Horatius is unlikely to be paying attention to any address. Besides, in adlocutio scenes (see Brilliant 1963, 130-32) the speaker, with arm extended, is normally of higher status than the addressee, as when an emperor addresses his troops. Here, if any status differentials may be imagined among the Romans, the aristocratic Horatius is likely to be superior.
18
B. ROLLER MATTHEW
recipients, when identifiable, seem normally to have been elites, and the objects themselves were often used as jewelry or for other decorative purposes. Perhaps, then, this object was intended for an elite viewership, and may not have come before the eyes of a wider population. Yet the image accords closely enough with key features of a widespread narrativethat it must have been recognizable to a viewer of any status-especially when which would have assured identification by glossed by the legend COCLES, all who were even minimally literate.36 Yet we must note specifically what aspect of the deed this image commemorates. Though we may read into it the whole story-his solo fight against great odds, his plunge into the river, possibly a grievous wound to his leg or hip-none of this is actually depicted in this scene. What it shows, rather, is the hero swimming through rough water while burdened with a full complement of armor.In the narrativeaccounts too, we find that his armorconspicuouslyaccompanieshim in the water.Thusin Livy (2.10.11) he swims across in full armor,unwoundedbut amidst a shower of Etruscan javelins (perhaps implied also on the medallion, where an Etruscanpoises his javelin), while in Plutarch(Publicola 16.8) he swims across in armornot only while wounded,but with an Etruscanspearlodged in his buttock;meanwhile Seneca (Ep. 120.7) stresses that he emerged from the river with all his armor,"as safely as if he had crossed by way of the bridge."Otheraccounts are similar.37Why this widespread insistence that his armor accompanied him in the river? Probably its explanation is found in the long-standing Graeco-Romanwarriorethic, wherein flight from battle is disgraceful, and a soldier who returns without all of his arms gives reason to infer that he fled. For a soldier would only drophis heavy equipmentin orderto run away expeditiously; in a proper "fightingretreat,"where he withdrawsfacing the enemy, he would need his full complement of weapons and armor.Moreover, the enemy could be expected to gather up abandoned equipment as spoils, erecting trophies on the battlefield and adorning the temples and houses of their native city, as monumentsto their victory and to the greater shame of the defeated.38Now, Horatiusundeniablyretreatsbefore the Etrus36. For the recipients and uses of medallions see Toynbee 1944, 112-21; also 1925, 171-72, on the possible audience for the "early Rome" medallions in particular. 37. Livy 2.10.11: ... ita sic armatus in Tiberimdesiluit multisque superincidentibus telis incolumis ad & TavT5nXsOv iqCaUTvy si TVnov suos tranavit. Plut. Publicola 16.8: Ka 7aap6v, binvijEV?ao oi{rs0 pErT 0(PFS '5X%0, yXouT6v.Sen. Ep. 120.7: ... iecitque se in npoo•GisPtE 1 TupprlvKi rTbv t?npav PE•p•Xrllpvot 86patt praeceps et non minus sollicitus in illo rapido alveofluminis ut armatus quam ut salvus exiret, retento armorumvictricium decore tam tutus redit quam si ponte venisset. Other passages: Dion. Hal. 5.24.3: Ka5dkXoiv ToictntXoctsi; tbv sTsats ntozrstI v KCai T6b••pla XaXsic; xtvu ... EiqP1v 8tavrlcd1isvoq ,siKohX6Plsrlosv 9v j) VEIV topaXlv. Frontin. Str. 2.13.5: ... deiecit se in alveum eumque et armis yifv ous6vvTrvvisrXOv et vulneribus oneratus tranavit. Flor. 1.10.3: Horatius Cocles... ponte recisso transnatat Tiberim nec arma dimittit. Serv. in Aen. 8.649: ... quo soluto se cum armis praecipitavit in Tiberim et licet laesus in coxa tamen eius fluenta superavit. De vir. ill. 11: ... cum quo [sc. ponte] in Tiberimdecidit et armatus ad suos transnavit. Val. Max. 3.2.1: armatus se in Tiberim misit... nec pondere armorumpressus . .. tutum natandi eventum habuit. See also Ampelius 20.4, and Dio Cass. 45.31.1. 38. The disgrace of leaving arms on the battlefield, indicating defeat and flight, is widely attested in Greek and Roman texts; Nisbet and Hubbard(1978, 113-14) collect numerous references. Note that, in our case, Seneca makes Horatius value his armor no less than his life (Ep. 120.7): non minus sollicitus... ut armatus quam ut salvus exiret. For the fate of abandoned arms as booty in the city of the victor, see, e.g., H1lscher 1978, 318-24 ("Beutedenkmaler"). For Roman warrior/combat values in general, see Barton
EXEMPLARITY IN ROMAN CULTURE
19
can onslaught, and in the water even turnshis back to the enemy (as the medallion shows)-an action that, on the face of it, demands censure for cowardice, not praise for noteworthy valor. Retaining his armor,however, guaranteesthat the retreatwas "proper,"and can be recuperatedas a display of valor. The strenuous insistence in these accounts and on the medallion that he took all his armorwith him, then, betrays a difficulty in the production of Horatiusas an exemplum ("He retreated..."), even while doing the work necessary to stabilize the problem and smooth it over ("... but was
valorous anyway"). Finally, for Roman readers and viewers of the late Republic and Empire, the swimming itself enhances the glory of the deed. Swimming is frequently representedin literary texts as a manly pursuit, and is often associated with other athletic activities-riding, footraces, javelin throwing-that were part of a soldier's training.39Indeed, feats of swimming appear in battle narrativesas noteworthy elements of performances that are striking on other grounds as well. In Vergil's Aeneid the Rutulianchampion Turnus, fighting alone and pressed hard by the Trojans, leaps into the Tiber in full armorand is carriedto safety (Aen. 9.815-19). To Romans of the early Empire, this vignette likely recalled the paradigmof Horatius,even while purporting to give Turnuschronological precedence. In a more historical vein Q. Sertorius, wounded and unhorsed at the battle of Arausio, reportedly swam the Rhone in full armor(Plut. Sert. 3.1); and Suetonius says that Julius Caesar, cast overboardduring the battle of Alexandria, swam a considerable distance holding up his record books with one hand to keep them dry and "holding his cloak in his teeth, lest the enemy take it as a spoil."40If swimming had broad cultural currency as an indicator of valor, as these texts suggest, then Horatius'feat of swimming may itself have been thought to contributeto the overall valor of the deed (and the paradigmof Horatius may in turn have informed these other accounts: see below under "Imitation"). Indeed, if we ask why Horatius needs to survive his plunge into the river, the valor associated with swimming-enhanced by doing so in armor, in a turbulentriver, and perhaps while wounded, all with the result of bringing his armor home-may provide the answer.41Under such circumstances he might have been expected to drown, as he does in Polybius' 2001, 38-56; Lendon 1999, 304-16, and 1997, 237-66; Alston 1998; Goldsworthy 1996, 264-82; Leigh 1995; Oakley 1985; and Harris 1979, 9-41. 39. Swimming as military training: Hor. Carm. 1.8 (with Leach 1994, 335-38), 3.7.22-27, 3.12 (also Reis 1994, 47-60, on the athletics in these poems); Plut. Cat. Mai. 20.6; Veg. Mil. 1.10. German warriors are said to excel at swimming rivers in full armor:Tac. Hist. 4.12, Ann. 2.8.3; ILS 2558 (cited by Konrad 1994, 43-44; q.v. for further references and discussion). In general see also Mehl 1931, cols. 861-63. I thank JonathanChicken and Ellie Leach for discussion of these matters. 40. Suet. lul. 64: Alexandriae circa oppugnationempontis eruptione hostium subita conpulsus in scapham pluribus eodem praecipitantibus, cum desilisset in mare, nando per ducentos passus evasit ad proximam navem, elata laeva ne libelli quos tenebat madefierent,paludamentum mordicus trahens ne spolio poteretur hostis; cf. Plut. Caes. 49.7-8; [Caes.] B Alex. 21.2-3. 41. Turbulence: note the large waves on the medallion; also Dion. Hal. 5.24.3, nIepsy&p Toi i0XrpEi(i.e., he is caught up in rapids and opaot TCyv oavi•&v oXt(6pevog 6 05oi56bvgi v cai bivasg qroist p•ydaXag whirlpools), and Val. Max. 3.2.1, nec ullo verticis circuitu actus (i.e., he manages to avoid the whirlpools). At Sen. Ep. 120.7, the water merely flows fast.
20
B. ROLLER MATTHEW
version. This is no disgrace, but does precludehim from displaying his valor in the river.42 5. Horatius' statue, alone among the monuments examined here, has attractedconsiderable scholarly attention, since several of its attested characteristics pose problems for archaeologists and art historians. Much information about this statue provided by literary texts43 is consistent, or can plausibly be interpretedas consistent. First, the statue was erected by public authority,to honor his deed and/orto recompense him for the disabilities he suffered. That is, the populus Romanusreciprocatedhis service on its behalf with a monument that not only called the deed to memory, but also reified their collective positive judgment that the deed served their interests.44Second, the statue was erected in the most public of places: either in the comitium proper, or in the Volcanal on the southern flank of the comitium. Because this corner of the republican Forum is described as the most visible, most frequented, or most importantplace-as stands to reason, since the comitium, the curia adjacent to it on the north side, and the rostra adjacent on the south side constituted the political heart of the republicanRoman state-then the value of this gift by the collective was enhanced by the care taken to ensure its visibility. Since the statue's job qua monumentis to construct secondary audiences for the great deed, it could collect the most eyes, from the widest range of Romans of every age, sex, and status, in precisely this location.45Moreover, Pliny asserts that it still stood in his day (quae durat hodieque, HN 34.22). Third, the statue was bronze, and represented Horatius in armor.46 42. For exhausted, heavily laden soldiers drowning, see Livy 5.38.7-9: circa ripam Tiberis quo armis abiectis totum sinistrum cornu defugit, magna strages facta est, multosque imperitos nandi aut invalidos, graves loricis aliisque tegminibus, hausere gurgites; also Livy 22.6.6-7 and Polyb. 3.84.9. Polybius may have his own, programmaticreasons for making Horatius die; his version need not be laid to an alternative tradition (contra Voisin 1992, 261-66, who argues unconvincingly that traces of the Polybian version can be seen in other accounts too). Having described how young Romans are inspired to glorious deeds by narratives they hear at aristocraticfunerals, and having invoked Horatius as an instance of such a youth, Polybius may kill him off to close the exemplary loop-to imply that this deed too will be duly narratedat the upcoming funeral, inspiring other Romans in turn. 43. Cic. Off. 1.61; Livy 2.10.12; Dion. Hal. 5.25.2; Plin. HN 34.22; Plut. Publicola 16.9; Gell. 4.5, De vir. ill. 11. 44. Three passages specify the dedicator as the community at large: Livy 2.10.12: grata erga tantam virtutem civitas fuit; statua in comitio posita. Dion. Hal. 5.25.2: eiic6va vonkov 6 8ij?log ?Eorlev Xa.XK11v 6 t o3rTotqEi6va ... .pb a6broiTf; dyopb.9 Tvzi~ Kpatiaot. Plut. Publicola 16.9: [sc. 'PoMlalotainavtre] iv •votrloav [az6T] v ri itepj toO 'Hypatorou,Piv yEvoEvrlv TK toO tpalO6a.toq(•sX6trlIta Tij aX?lK tv8pi as 7t Livy and Plutarch also mention gifts of food and land, presenting the statue p•tEtd p;l aprryopoIvtE;. one in a broader act of component reciprocation. For these gifts cf. De vir. ill. 11.2; Plut. Mor. 820E; discussion by Sehlmeyer (1999, 92-93); Miinzer (1913, cols. 2333-34). For the gift-exchange dynamic between individual and collective in which an honorific statue takes part (objectifying a relationship of mutual obligation), see Tanner2000, 25-30. 45. Placement in comitium: Livy 2.10.12 (see previous note), Plin. HN 34.21-22 (apparently);in Volcanal: Plut. Publicola 16.9, De vir. ill. 11.2. Dion. Hal. 5.25.2 (previous note) locates it "in the most im9v Tir portant place in the Forum." For this area's high visibility see Dion. Hal. 1.87.2, tf?g yopiq... q) picp rtapdroli Kpatiort ?LpP6Xotq(i.e., the Volcanal; again at 3.1.2); 2.29.1, 9v rT <pavepo(rdp trifg dyopd;q(the comitium); Plin. HN 34.24, quam oculatissimo loco (the Rostra). Gell. 4.5.1-4 reports that the statue was relocated from the comitium to the Volcanal, apparentlya very short move. On the statue's location(s) see Sehlmeyer 1999, 94-95; Coarelli 1983, 1.168, 174-75; Lahusen 1983, 12-13, 33-34; Vessberg 1941, 87-88. 46. Bronze: Dion. Hal. 5.25.2; and Plut. Publicola 16.9 (quoted n. 44 above). Armor:Dion. Hal. 5.25.2; perhaps Cic. Off. 1.61 (see below, pp. 21-22).
IN ROMANCULTURE EXEMPLARITY
21
But archaeologists question this representation. Whether an honorific statue could have been erected in the late sixth century B.C.E.is far from clear, since no certainly historical honorific statues are attested until the late fourth century.Thus it is argued that Horatius' statue, if properlyhonorific, must have been erected at least two centuries afterthe traditionaldate of his deed, thoughit was subsequentlymisrememberedas being contemporarywith it; or, if in fact archaic, it was a cult or votive statue that was subsequently misunderstoodas honorific.47For this paper,questionsconcerningthe statue's actual origins can be left out of account, being irrelevant to what the Romans of the late Republic and Empire made of this monument. The writers through whom we know of this statue believed it to be honorific, set up shortly after Horatius' great deed (insofar as they credit the deed at all), and their own understanding of their contemporary world is my focus here. Within the frameworkof this "honorific"interpretation,archaeologistsraise a typological question:was it a statua loricata, the "normal"type of armored statue known from the imperial period? Or did it have a form distinctive to itself?48 This typological/iconographic question is highly pertinent to my study, since it considers how Romans might have comprehendedHoratius' statue in terms of the iconographical types that they themselves deployed through most of the period from which our evidence comes. What, then, might Roman viewers-the secondary audiences to Horatius' deed, constitutedas such by encounteringthis statue-have made of it? Two writers, Dionysius and Cicero, provide what may be regardedas interpretations of the object's iconography.In De officiis 1.61, Cicero describes how one can blame or praise deeds done badly or well. Regarding praise, he writes: contraquein laudibus, quae magno animo et fortiterexcellenterque gesta sunt, ea nescioquo modo quasi pleniore ore laudamus. hinc rhetorumcampus de Marathone,Salamine, Plataeis, Thermopylis, Leuctris, (hinc no>ster Cocles, hinc Decii, hinc Cn. et P. Scipiones, hinc M. Marcellus, innumerabiles alii, maximeque ipse populus Romanus animi magnitudine excellit. declaraturautem studium bellicae gloriae quod statuas quoque videmus ornatu fere militari.
47. On the emergence of honorific statuaryin Rome around300 B.C.E.,and the problem of statues supposedly commemorating earlier figures, see Sehlmeyer 1999, 41-109, esp. 109, and H61scher 1978, 32444 (332-35 on Horatius). Conversely, Hafner (1969, 27-33) accepts the Horatius legend as entirely historical, including the erection of the honorific statue, and identifies a sixth- or fifth-century warrior head, known in three copies, as that of Horatius. Other "early" possibilities: Hill Richardson (1953, 98-101) points to sixth-century Etruscan votive statues and statuettes, some bronze, representing armed men (thus Holscher [1978, 332] suggests an under-life-size bronze); Coarelli (1983, 1.168, 174) argues for an early cult image of Vulcan later misunderstood as honorific. (For Horatius' connections with Vulcan see Camassa 1984, 829-31, 835, 840; Gage 1973, 10-13; for an older view Pais 1905, 157-61.) Gellius' notice (4.5) of a lightning prodigy involving this statue, cited from Book 11 of the Annales Maximi, has spurred scholarly discussion about dates: Frier (1979, 58-64) accepts a fifth-century date for both the statue and the contents of Book 11; Hdlscher (1978, p. 334 and n. 97) rejects an early dating of the contents of Book 11, hence any necessarily early date for the statue; Forsythe (1994, 253) dates both to c. 300 B.C.E. Cf. Vessberg 1941, 88; Mtinzer 1913, col. 2334. 48. See Stemmer 1978, 142, 145, on the relationship of Horatius' statue to the later Panzerstatue (= statua loricata) type; also Lahusen 1983, pp. 51-52 and n. 51, and Sehlmeyer 1999, pp. 93-94 and n. 292. The earliest imperial statua loricata is associated with Julius Caesar (Plin. HN 34.18); see Sehlmeyer 1999, 230; Lahusen 1983, 51-52; Stemmer 1978, 145.
22
MATTHEWB. ROLLER ... andconverselyin praising,we praisewitha kindof "fullervoice,"as it were,deeds donebravelyandoutstandingly, withgreatspirit.Hencethe battlesof Marathon,Salamis, Plataea,Thermopylae,and Leuctraare an exercisegroundfor the rhetoricians; henceour own Cocles,the Decii, GnaeusandPubliusScipio, MarcusMarcellus,and innumerable others, and most of all the Roman people itself, are outstanding in the greatness of their spirit. That we see statues in basically military garb also testifies to a zeal for military glory.
Whetherthese statuescommemoratethe specific heroes mentionedjust above is unclear, though certainly Cicero is referringto what he deems a typological class of statues. Assuming Horatius' statue was in armor,as Dionysius says, then Cicero, who has just mentioned Horatius,may have his statue in mind here as a member of that class of "statuesin military costume, more or less" (fere). Whatever doubts or exceptions may lurk in this fere, Cicero is confident of the costume's meaning: all persons so commemoratedshare a zeal for military glory; one can read off from the costuming the values the commemorandsembraced and put into action. To look at the statue of Horatius through Cicero's eyes, then, is to see a statue whose ornatus militaris (whatever form this may take) causes it to share with other statues a certain physical form and, consequently, a specific sociocultural meaning. Now let us turn to Dionysius, who stresses that Horatius went into and came out of the river wearing his armor (5.24.3). He goes on to list the honors and rewards he received publicly and privately (5.25.1-2), including the "bronzestatue in armor"(iK6vcvaXaXKlV vonkov, 5.25.2). He adds his own praise to these earlier acclamations, bracketing this list of honors with declarations in his own voice that Horatius had achieved something singular: he won "deathless renown," and was "the most enviable of any Roman at that time."49Since Dionysius gives Horatius' armorprominence in his account both of the battle and of the swim in the Tiber, he seems likely to mean that the armoron the statue is Horatius' own, that very same armorthat by (still) being on his body is a towering monument to his valor. This interpretationseems the more likely since Dionysius also says that Horatius' achievement outstrippedthose of all other contemporaryactors, and was thereforereciprocatedby a striking set of honors. To depict Horatiusin his armoris, precisely, to mark his distinctive and surpassing achievement. It follows that to look at Horatius' statue throughDionysius' eyes is to see the hero's specificity; the armoris an individualizing,particularizingelement in the statue's iconography.50Yet for Cicero, we saw that it is a generic element. These two texts, then, appearto provide the following answer to the typological question posed above: the statue of Horatius admits either a generalizing or a particularizinginterpretation,depending on the circumstances of the interpretiveact. Cicero aims to provide a stock of similar examples that one can invoke while praising, and therefore declares that any armoredstatue attests a praiseworthyhero. Conversely Dionysius, stressing 49. TorTo r6 Epyov Qdv0vaTov a6t~ 686kav eipydoarTo... (5.25.1); Opdttog Piv 6il Tota6rlv &TCO&StdEV TCrj T6 pO vAiEcK rig ... (5.25.3). POqPai(v p (rlksrb64 &PETilV •Pvog Xpo6v• ttqdog 7vTvoro 50. The allusion to this statue at Dio Cass. on pp. 34-35 below, may also require a par45.31.1, quoted ticularizing interpretationof the armor.
EXEMPLARITY IN ROMAN CULTURE
23
Horatius' surpassing valor, clothes him in the significant, distinctive armor that sets him apartfrom all others. In either case, the statue is a monument that calls the hero's deed to memory, and so constructssecondary audiences. But the specific interpretationthat any given secondary spectator, such as Cicero or Dionysius, imposes upon the statue depends upon the specifics of the rhetorical situation in which the statue is adduced. By examining how different monuments commemorate Horatius' deed, we have seen how each form foregroundsdifferent aspects of the deed, and constructsfor itself differentsecondary audiences in differentcontexts so as to produce differentrhetoricaland social effects. We have also seen how the productionof exemplary discourse is at every point contested and unstable, allowing it to be turnedto many different,even opposed, ends. But we have not yet examined one particularlyimportantend of exemplary discourse, which is to authorizeand promote certain patternsof action. Hence we turn to the fourth element of the scheme set out above: imitation. Imitation Within exemplary discourse, "imitation"entails the production of a (new) action in the public eye in light of a previous deed it resembles in some way, and the submission of this new action to various audiences for judgment and commemoration, with a view to spurring still further imitation in due course. Any action thus potentially looks both backwardto some previous deed in light of which it was done, and forwardto a subsequentimitation of itself, once constituted as a deed. To examine imitation regardingHoratius, then, is to ask two questions: What deeds does he imitate? And what deeds imitate his? Although Horatius stands in the tradition as one of the earliest Roman military heroes, exemplary discourse does not permit his deed to be a first. In fact, several accounts identify models that he could be regarded as imitating and striving to surpass. We have seen that Polybius makes Horatius an imitator of the sorts of deeds that young Romans hear narratedduring aristocraticfunerals,thoughno particularmodels are specified.Otheraccounts are more specific. In the first place, those that explain the cognomen Cocles as "one-eyed" because he had lost the other in a prior battle make him a model for himself. Just as the cognomen, and the injury it indexes, indicate that he displayed virtus or fortitudo on that earlier occasion, so he does again in defending the bridge. Perhaps, however, the wounds he suffers on each occasion enabled Romans to see a further parallel: here is a soldier who habitually accumulates disabling injuries as monuments to his valor, yielding one body part after another to the ever-greater glory of the parts that remain.51Provisionally, then, we may identify two distinct ways in which deeds can be "alike":their resemblance may be categorical, in that 51. So Sallust writes of Sertorius (Hist. 1.88M, Reynolds' text): magna gloria tribunusmilitum ... fuit, multaque ductu eius peracta ... incelebrata sunt: quae vivosfacie sua ostentabat aliquot advorsis cicatricibus et effoso oculo. quin ille dehonestamento corporis maxume laetabatur neque illis anxius, quia relicua gloriosius retinebat (similarly at Dem. De cor. 67, on Philip II of Macedon-a passage Gell. 2.27 says Sallust imitates).
24
B. ROLLER MATTHEW
Romans could place both under a single ethical rubric like fortitudo (Valerius Maximus' collection of exempla is organized entirely by such categories); or the resemblance may be structural, in that one deed reproduces specific features of another. In contrast, one account of Horatius' deed-Seneca's, in Epistula 120has a stake in denying that there was an earlier valorous performancesetting the model for the later one. Here Seneca argues, in a Stoic vein, that no external observer can assess an agent's moral state-whether that agent is a (Stoic) good or wise man, or not-without observingmany actions done over time and in various circumstances. For only long-term patterns of action, exposing the agent's overall consistency or inconsistency in making moral choices, reveal the virtuousness or viciousness of his disposition. In this context Seneca gives a brief, admiringnarrationof Horatius' defense of the bridge (?7): a brilliant deed, as Seneca admits, but he insists that one cannot extrapolate from such a deed the agent's overall moral state, for no single action provides sufficent evidence. That is, Seneca adduces Horatius as an example of an agent about whom we know only one thing, and about whom any properly moral judgments (as opposed to simply evaluating his one action) are therefore premature. In making such an argument, Seneca is rejecting the long-standing patterns of moral evaluation embedded in exemplary discourse, with its focus on the single great deed. This move, as I and others have argued,is partof a larger Senecan project to offer theorized Stoicism, in place of traditionalethics, as a means by which Roman elites may address certain ethical binds imposed by the emerging imperial regime.52 So if Seneca accepted that the cognomen Cocles attested a prior valorous deed, his conclusion might seem less cogent, since some readers who concede his point about one deed might disagree about two-believing that two valorous deeds do constitute a pattern and offer grounds for inferring that the agent's moral state is virtuous overall. Admittedly this is an argumentfrom silence. But I speculate that Seneca omits the explanation of the cognomen found elsewhere in the tradition because his immediate argument is best served by presenting Horatius as a one-deed marvel. Dionysius provides Horatius with other models by supplying a genealogy. He says that Horatius was the nephew of Horatius Pulvillus, suffect consul in year one of the Republic and ordinary consul in year three, the year in which he locates Cocles' deed. Moreover, Cocles was descended from the surviving Horatiuswho had, along with his two brothers,defeated the Alban triplets to secure Roman ascendancy over that city.53A prominent feature of Roman exemplary discourse is that the most compelling models for imitation often come from within the actor's own family. Certainly no actor is restrictedto familial models, but the idea that certain patternsof behavior do or should run in families-that is, that the deeds done by members 52. Roller 2001, 64-126 (88-97 on Senecan exempla); Wray forthcoming; in a different vein, Inwood 1995. 53. Dion. Hal. 5.23.3: o06og dsZX(Pt6o0g piv v 'Opa-rtouMdprKou 1 -rb rT&v 0tr•pou irtcTov, E y7vog Ka-riyev d"p'ivb6 qtsv rptS6poyv'Opariou MdpKoutoO vt1'roavrTog TobqAaXavoubTpt66Iouo, 6ic tpi tfgq Oi iyEsLoviaq0la6 t Ei sn6XovovKaTCaoCrToatuovvrlav.
EXEMPLARITY IN ROMANCULTURE
25
of a single gens demonstratestructuralas well as categoricalresemblancesis widespread in Roman culture.54In Cocles' case, the surviving member of the triplet Horatiiprovides not only a categorical model of "valor"and "endurance,"but the structuralparallel that both fought initially in a threesome, of whom two fell away leaving the one to face down a more numerousenemy alone. It is also tempting to find a categorical model in Horatius Pulvillus (exemplary for his fortitudo animi in Val. Max. 5.10.1, and displaying robur animi in Livy 2.8.8), who persisted in dedicating the temple of Jupiter despite being apprised,in the middle of the ritual,of his son's death.55Lacking any martialelement, however, this deed representsa somewhat different dimension of fortitudo. So much for Horatius' own models. Let us turn now to the second question: when and how is he himself invoked as a model for others to imitate? To be sure, any monument commemoratinghis deed presents him as a potential model for imitation, since exemplary discourse posits that those who learn of his deed through a monument should seek to reproduce it to gain similar renown. But let us consider cases where he is clearly invoked as a model or standardof evaluation for others. We have already encountered one case: in the "Philippic"that Dio Cassius places in Cicero's mouth, Cicero suggests that Antony, in offering a crown to Caesar at the Lupercalia, has fallen short of Horatius' standardin the category of "defending freedom" (Dio Cass. 45.31.2). SubsequentlyFufius Calenus, replying to Cicero, maintains that Antony in fact defended freedom on that occasion more successfully than Horatiushad done (46.19.8). Here the deeds in question have no structuralresemblance; the comparison is entirely categorical: who has preserved libertas most effectively through actions of one sort or another? Horatiusis also invoked as a canon in contexts where the similarities run deeper. Valerius Maximus relates the following anecdote under the rubric de amicitia. When the tribune C. Gracchus was being pursued throughthe city by his enemies, a steadfast friend named Laetoriusbarricadedthe pons sublicius and held up the pursuit until Gracchus had crossed the bridge 54. To list only a few instances: several Scipionic epitaphs speak of the deceased as rivaling or surpassing his ancestors' deeds, or ennobling his stock, or causing the ancestors to "rejoice that he was born"(ILS 4, 6); Polybius offers to teach the young Scipio Aemilianus how to "speak and act in a way worthy of his ancestors" (31.24.5, 10; see Habinek 1998, 50-51); three generations of Decii Mures reportedly die in battle as consuls, and at least two "devote" themselves in almost identical ways (e.g., Cic. Tusc. 1.89, Fin. 2.61; Dio Cass. apud Zonar. 8.5); several generations of Appii Claudii in the early Republic fill identical roles as fearsome enemies of the plebs; and M. lunius Brutus reportedly felt pressure to measure up to the "tyrannicide"heritage of his ancestor, L. lunius Brutus (e.g., Cic. Phil. 2.26; Dio Cass. 44.12; App. B Civ. 2.112; see Macmullen 1966, 7-10). Rich reading along these lines in Sen. Controv. 10.2, passim. I know no systematic treatmentof these familial traditions, unless it be G. D. Farney, "AristocraticFamily Identity in the Roman Republic" (Ph.D. diss., Bryn Mawr, 1999), which I have not been able to examine. For selective treatments see Bernard 2000, 167-96; Flower 1996, 128-58 (esp. 139); and Hilkeskamp 1996, 316-26 (esp. 319-21, 325). Scholars have noted that both Livy and Augustus (in his forum) strive to "nationalize" these familial traditions, to claim them for broader use by other Romans: Jaeger 1997, 107-24; Feldherr 1998, 97-100. 55. Dionysius (5.35.3) has Pulvillus dedicate the temple in year three, the year of his ordinaryconsulship and Cocles' deed, while in Livy he dedicates it as suffect consul in year one, the year before Cocles' deed (see n. 1 above for these authors' different chronologies). Dionysius alone makes Pulvillus Cocles' uncle, yet does not mention the death-of-the-son incident when reporting the dedication of the temple. So he, at any rate, does not seem to suggest even a categorical similarity along the lines of fortitudo.
26
MATTHEWB. ROLLER
safely. Then, says Valerius, "overwhelmedby force of numbers, he turned his sword againsthimself and with a rapidleap sought the depthof the Tiber. The fondness for his entire fatherlandthat Horatius Cocles had once demonstrated on that bridge, Laetorius gave to a single friendship, and added his voluntary death."56On the one hand, Valerius here draws a structural parallel: Laetorius imitates Horatius in his one-against-many fight on the pons sublicius, concluding with a leap into the river. Indeed, in saying that he "added"something, namely a suicide, Valerius may imply that he surpassed Horatius,making the deed even more noteworthy(though in another respect it is less so, since the suicide precludes him from rivaling the swim in full armorwhile wounded).57On the other hand, the categorical parallels are problematic. Valerius continues, "Whatfine soldiers the Gracchi could have had, had they been willing to follow their father's or maternalgrandfather's way of life! With what force and perseveranceof spirit would Blossius, Pomponius, and Laetorius have facilitated the Gracchi's trophies and triumphs, since they were such energetic accomplices to their mad undertakings... !",,58 By comparing Laetorius' actions favorably to a soldier's, he seems to acknowledge that this deed (like Horatius') manifests courage and steadfastness. Yet unlike such soldiers, and Horatius before them, he insists that Laetorius does not act in the collective interest. On the contrary, he stakes his life only to his friendship with Gracchus, whose "madundertakings" are (on this account) utterly contraryto the collective interest, and entirely at odds with the "trophies and triumphs"that would have marked valued service to the res publica. So, however close his imitation of Horatius in structuralterms, Laetorius perversely acts not for but against the common good. For this reason Valerius deprives him of the positive evaluation that successful imitators might expect to receive. Laetorius becomes an exemplary figure in his own right, as the narrativehere and in Velleius attests-but he is exemplary only for (misplaced) amicitia and fides, and not for Horatian-stylevirtus orfortitudo, outstandingvalor on behalf of the res publica at large. Even when not named explicitly, Horatiusmay provide an implicit model for others' heroic deeds. Consider Q. Sertorius,who Plutarchsays swam the 56. Val. Max. 4.7.2 (de amicitia): Laetorius autem in ponte sublicio constitit et eum, donec Gracchus transiret, ardore spiritus sui saepsit, ac vi iam multitudinis obrutus converso in se gladio celeri saltu profundum Tiberispetiit, quamque in eo ponte caritatem toti patriae Horatius Cocles exhibuerat, unius amicitiae adiecta voluntaria morte praestitit. Note that Gracchus goes the opposite direction from Horatius, passing westward through the porta Trigeminaand onto the bridge: on his route see Coarelli 1988, 31-34. Veil. Pat. (2.6.6), giving the friend's name as Pomponius, draws the structuralparallel equally precisely: quo die singularis Pomponii equitis Romani in GracchumJidesfuit, qui more Coclitis sustentatis in ponte hostibus eius gladio se transJixit.See also Plut. Ti. Gracch. 38.1; De vir. ill. 65. 57. Likewise in Polybius' version, where death precludes the valorous swim. Voisin (1992, 264) suggests that Laetorius' death, as described by Valerius, preserves a trace of the Polybian version. But Valerius' phrase adiecta voluntaria morte presents this death not as reproducingHoratius' death (Ala Polybius), but as an "addition,"a respect in which Laetorius' deed is new and different from Horatius'. That is, Valerius has in mind the "standard"version in which Horatius lives, not the Polybian one where he dies. 58. Val. Max. 4.7.2 (continuing the passage quoted in n. 56): quam bonos Gracchi, si aut patris aut materni avi sectam vitae ingredi voluissent, habere milites potuerant! quo enim impetu, qua perseverantia animi Blossius et Pomponius et Laetorius tropaea ac triumphos eorum adiuvissent, furiosi conatus tam strenui comites....
IN ROMANCULTURE EXEMPLARITY
27
Rhone,woundedbutretaininghis breastplateandshield,duringthe Roman defeatat Arausio(105 B.C.E.).Evenafterattaininghigh command,he continuedputtinghis body at risklike a commonsoldier:for instance,he lost aneye, andwasproudthatthe woundofferedvisibleproofof his valor.59 As evidence of valor, the swimmingfeat and eye loss suggest the Horatian andcategorically,thoughHoratiusis not named.60 modelboth structurally are Equallysuggestive the deeds of CassiusScaeva,one of JuliusCaethebattleof Dyrrachium, credits sar'scenturions.Caesarhimself,narrating Scaevawithwardingoff a Pompeianattackon a fortification,andsays that his shield suffered120 perforations(B Civ. 3.53.3-5). Lateraccountsare moreelaborate,mentioningnot only the perforatedshieldbut also wounds to the shoulderand leg, and the loss of an eye.61Lucan'sversionmakes Scaeva'sperformanceexplicitlyspectacular,as Scaevahimself calls upon the opposingcommandersto witnesshis deed (6.158-60), and a groupof young soldiersgatheraroundto watchandjudge (6.167-69). At the end, theylift up theirwoundedchampionandproclaimhim "thelivingimageof greatvirtus"(6.251-57), thoughLucanhimself,as a secondaryspectator, rejectsthese eyewitnesses'positiveevaluationandimposesa negativeone of his own (257-62), a phenomenonwe considerfurtherbelow.62 Severalaccountsalso creditScaevawithanearlierdeed,duringCaesar's campaignin Britain.ValeriusMaximus(3.2.23) reportsthathe defendeda narrowspit of land,extendingbetweenan islandanda rock,againstmany enemiesuntilhis fellow soldiersescapedbehindhim.His helmetdislodged by blows, his shield "consumedby holes,"and having sufferedseveral wounds,he providedan astonishingspectaculumto bothBritanniandRoman spectators.Finallyhe plungedinto the sea wearingtwo breastplates (oddly)andswamto safetyunderthe eyes of Caesarhimself-begging his general'sforgivenessfor the armorhe left behind.63It is hardto imagine 59. Plut. Sert. 3.1:... KaKCSg Kai rporflg yevopnvrlq, tnopep3XrKcci; zv r(•v 'PsOmaiwV dyewvoc•apvWv Kai KaTaTeTp(~( SvogTb odpa Tv 'Po8svbv 8stnpaocv, aU(bzT?rcOT 06paKt Kai OupE4 ivavriov kIYnOV srpb6g 0ESba n7ok6 VlX6pOEvog....(??3-4) o6b lytV b piCKato fT oTpaTOt1WtKf?g T6k7p eiq pockXrlXOudc &,ioMpa t Kai Xstp6 &7doStKVv6Cevoqpya o aupLaoTaKai To' o pa zoig•Ty&otv&cptS6g lysLp6voq, &XXk isrtstSo6g, iv IEOWv tzlv 9zTpav KKO0tEitOV. itri tZOtZq) 8 Kai Kao7n•MtPE6voq d(ei StET4kt ... a6)T 8&tf • IVdldtiraXF 4 rI a Kai Tiq oujupopaq OeardS.SimipayaOtfa yvopiopaTa, zobqiaboioS~xovrt rXOV &psi 4iPE Sall. napa•Clvetv Hist. 51
1.88M (n. above). larly, 60. Though no ancient text, to my knowledge, explicitly links Sertorius and Horatius, modern scholars have done so: Konrad1994, 43-44; Africa 1970, 532-35; Gage 1969, 197-98. However, Aulus Gellius (2.27) notes a parallel between Sertorius and Philip II of Macedon (comparing Sall. Hist. 1.88M and Dem. De cor. 67); while Plutarch (Mor. 307D-E) compares Philip and Horatius. It may therefore be only an accident of survival that no text directly compares Horatius and Sertorius. 61. Val. Max. 3.2.23; Lucan 6.140-262; Plut. Caes. 16.3-4; Suet. lul. 68; App. B Civ. 2.60 (where the perforatedshield and lost eye are attributedto a Minucius). Discussion of these passages in relation to Horatius in Capdeville 1972, 602-11. 62. Leigh (1997, 158-90) discusses well the spectacular elements of this episode in Lucan, pointing to the "double audience" of viewers within the text and readers of the text (164-65, 184-primary and secondary spectators, in my terms), rightly noting that the viewers within the text construct the exemplum as such (181-84). He also connects Scaeva with Horatius (174-75), and observes parallels to other exemplary deeds (166-72, 182-90). Thus, while the Scaeva episode is assuredly an epic aristeia, and as such participates in a literary tradition, Leigh shows how this episode is meaningful and effective in distinctively Roman cultural terms (181-82). 63. Other accounts: Plut. Caes. 16.5-7 (explicitly noting Caesar's witnessing of the deed), Dio Cass. 37.53.2-3 (deed done in Lusitania ratherthan Britain); discussion by Capdeville (1972, 611-15).
28
B. ROLLER MATTHEW
that Valerius did not intend this narrative to be compared closely with the account of Horatius provided at the beginning of this same chapter (3.2.1). Not only are both broughtunder the same categorical rubricde fortitudine (3.2) and share obvious structuralparallels, but Scaeva seems to "cap" and thus surpass Horatius' deed: for although he emerges from the water without all his equipment,he has the shame and presence of mind to ask Caesar's pardon for the loss. Thus, says Valerius, he was "great in battle, but greater in his recollection of military discipline," and was rewarded "not only for his deeds, but especially for his words"-thus achieving something beyond the (mere) valor of Horatius. Scaeva's engagement with the Horatian model is beyond question here, even though Valerius does not draw the parallel explicitly.64 That the texts through which we know of Sertorius and Scaeva clearly present them as imitators of Horatius, yet equally clearly feel no need to name him as the exemplary model, seems to indicate that, for authors and readers who had internalized and naturalizedthe discourse of exemplarity, and for whom seeking out models for imitation and then seeking to set a new model in turn was the fundamentalrhythmof action in the public eye, there was no need to point out the obvious by making the model explicit. III. CLOELIA Among the figures constructed as imitators of Horatius, Cloelia is perhaps the most unexpected. Her story goes as follows. It is still the year of Horatius' deed-year two or three of the Republic, depending on the chronology-but somewhat later.65Porsenna,his initial assault stymied by Horatius, is now encamped on the Janiculum, laying siege to the city. Mucius Scaevola's exemplary deed has occurredin the meantime, leading to a truce. The Romans guaranteethis truce by sending as hostages to Porsennathe young sons and daughters of leading families. Among these children is a girl-a virgo-named Cloelia.66Here is Livy's account (2.13.4-11): (4) his condicionibus composita pace exercitum ab Ianiculo deduxit Porsenna et agro Romano excessit. (5) patres C. Mucio virtutis causa trans Tiberim agrum dono dedere, quae postea sunt Mucia prata appellata. (6) ergo ita honorata virtute, feminae quoque 64. Val. Max. 3.2.23: cum laude mereris veniampetisti, magnus proelio sed maior disciplinae militaris memoria. itaque ab optimo virtutis aestimatore cum facta tum etiam verba tua centurionatus honore donata sunt. For Horatius and Scaeva see Capdeville 1972, p. 619, n. 1; also Leigh 1997, 175. Capdeville (1972, 615-18) demonstrates even more extensive parallels: in Valerius and elsewhere, Scaeva is coupled with another Caesarian soldier named Acilius, who loses his right arm attacking a ship in the naval battle at Massilia (Val. Max. 3.2.23; Suet. lul. 68; Plut. Caes. 16.2; cf. Lucan 3.609-26). Thus the early imperial tradition of Caesar's wars seems to present Scaeva and Acilius together as a recent analog to Horatius and Mucius Scaevola, the heroic duo of the wars with Porsenna. 65. Livy (2.13.6-11) places both deeds in year two, the consulship of P. Valerius Publicola II and T. Lucretius (likewise Plut. Publicola 16.3); Dion. Hal. (5.33) puts them in year three, the consulship of Valerius Publicola III and Horatius Pulvillus II (see n. 1 above). 66. The gens Cloelia is included by Livy (1.30.2) and Dion. Hal. (3.29.7) among the leading Alban families enrolled in the patriciate after the cities unified; indeed, a Cluilius is among the last kings/leaders of Alba (Livy 1.23.4; Dion. Hal. 3.2.1), and a Cloelius is consul at Rome several years after Cloelia's deed. On the family see Fugmann 1997, 62; Forsythe 1994, 254-55; Arcella 1985, 35, 39-40; and Miinzer 1901, cols. 109-10 (s.v. "Cloelius," nos. 6, 8, 10-12-all early republican).
EXEMPLARITY IN ROMAN CULTURE
29
ad publica decora excitatae, et Cloelia virgo una ex obsidibus, cum castra Etruscorum forte haud procul ripa Tiberis locata essent, frustratacustodes, dux agminis virginum inter tela hostium Tiberim tranavit, sospitesque omnes Romam ad propinquos restituit. (7) quod ubi regi nuntiatumest, primo incensus ira oratores Romam misit ad Cloeliam obsidem deposcendam: alias haud magni facere. (8) deinde in admirationemversus, supra Coclites Muciosque dicere id facinus esse, et prae se ferre quemadmodum si non dedaturobses, pro rupto foedus se habiturum,sic deditam
inviolatamque ad suos remissurum. (9) utrimque constitit fides; et Romani pignus pacis ex foedere restituerunt,et apud regem Etruscumnon tuta solum sed honorataetiam virtus fuit, laudatamque virginem parte obsidum se donare dixit; ipsa quos vellet legeret. (10) productis omnibus elegisse impubes dicitur; quod et virginitati decorum et consensu obsidum ipsorum probabile erat eam aetatempotissimum liberari ab hoste quae maxime opportuna iniuriae esset. (11) pace redintegrataRomani novam in femina virtutem novo genere honoris, statua equestri, donavere; in summa Sacra Via fuit posita virgo insidens equo. (4) With a truce arrangedon these terms, Porsenna led his army down from the Janiculum and withdrewfrom Roman territory.(5) Because of his virtus, the senate gave Gaius Mucius land on the other side of the Tiber, which afterward was called the "Mucian meadows." (6) Consequently, with virtus held in such esteem, women too were incited to public honors. A maiden named Cloelia, one of the hostages, eluded her guardssince the Etruscan camp by chance was located not far from the Tiber's bank-and leading a band of maidens swam across the Tiber amidst the enemy's javelins, and restored them all safely to their kinsmen in Rome. (7) When this was announced to the king, he was at first infuriated, and sent ambassadorsto Rome to demand Cloelia back as hostage: about the others he cared little. (8) Then, his anger turning to wonder, he said that it was a deed surpassing the likes of Cocles and Mucius, and let it be known that if the surety were not returned,he would deem the truce broken, but that if she were given over, he would return her unharmed and inviolate to her people. (9) Fides was maintained on both sides: the Romans restored the guarantee of the truce as the agreement required, while the Etruscan king not only kept her virtus safe but even held it in esteem: after praising the maiden he declared that he was making a gift to her of a portion of the hostages; she herself should choose which ones she wanted. (10) It is said that, when all had been brought forth, she chose those who had not reached puberty;it both befitted her virginitas, and in the hostages' own collective view was commendable, that the age that was most at risk of violation be freed from the enemy first and foremost. (11) The truce reestablished, the Romans rewarded this novel virtus in a woman with a novel form of honor, an equestrian statue; at the top of the Sacred Way was placed a virgo sitting upon a horse.
Livy introduces Cloelia as an imitator:he says that, with the virtus of Mucius receiving such acclaim, even women were stirred to act in the public eye for the benefit of the community. Yet on his account Cloelia successfully imitates both Mucius and Horatiusin categorical terms, since all three are honored publicly for their virtus (Horatiusat 2.10.11, Mucius at 2.13.5, Cloelia at 2.13.11)-that is, Livy insists that the community deemed all these actions consequential, evaluated them positively in the category of virtus, and constructedmonumentsin each case. In Cloelia's case, Porsenna too concurs with the Roman judgment: he esteems virtus (honorata etiam virtusfuit, ?9) just as they do (ita honorata virtute, ?6), and he expressly declares that her deed surpassed those of Cocles and Mucius (?8). But her imitation of Horatius is structuralas well as categorical, inasmuch as she
30
B. ROLLER MATTHEW
swims the Tiber from the Etruscan to the Roman side, under a shower of enemy javelins (2.13.6, cf. 2.10.11). Othertexts too stress this structuralresemblance, suggesting that her parallel with Horatius figured more broadly in the tradition. Valerius Maximus not only places Cloelia under the rubric de fortitudine (3.2.2), like Horatius,but narratesher story immediately following his, linking them closely with the assertion that "she dared her famous deed at almost the same time, against the same enemy, and in the same Tiber."67Thus he takes pains to note a (structural)parallel going beyond what his regular (categorical) organizing principles already assert. Returningto Livy, we note finally that she obtains the desired acclaim. Her countrymenmonumentalize her action as consequential and ethically positive througha laudatorynarrativeand an honorific equestrianstatue, placed at the summit of the Sacred Way at the opposite end of the Forumfrom Horatius' statue. Even Porsenna,an enemy but perforce a close observer of Roman valor, praises her, deems her a successful imitator of earlier heroes, and gives a gift (?8-9). Thus the exemplary loop closes, with the commemoration of her deed in monumentalforms that will inform and inspire future imitators,just as (according to Livy and others) she herself imitated earlier heroes. The figure of Cloelia is only slightly less resonant than Horatius in Roman culture, to judge by her textual trace: there are about a half dozen extended narrativesof her deed, and anotherdozen or so sparernarrativesor mere references. The earliest surviving reference is a fragment of the historian CalpurniusPiso (transmittedby the elder Pliny), dating probably to the later second century B.C.E.;next comes a probable hidden reference in Cicero, followed by a profusion of references and fuller accounts from the Augustanage onward.These accountspresentnumerousvariations:she seized a fortuitous opportunity to escape, or actively tricked her guards; she escaped alone, or with other hostages; she crossed the Tiber by day with many observers watching, or invisibly by night; in a few accounts she does not swim but crosses on horseback. Finally, some accounts say that as she and the other hostages returnedto Porsenna, they were ambushedby Tarquin's forces; here anotherhostage, Publicola's daughterValeria, is said to escape, while Porsenna's troops come to the rescue.68 67. Val. Max. 3.2.2: immemoremme propositi mei Cloelia facit, paene eadem enim tempestate, certe adversus eundem hostem et in eodem Tiberi inclutum ausa facinus. Valerius links these deeds even more closely by causing them, as a pair, to interrupta narrativeof Romulus, which begins at 3.2.praef. and concludes at 3.2.3. Also, whereas Livy categorizes the deeds of Horatius, Mucius, and Cloelia all under virtus, Valerius banishes Mucius to the next chapter (3.3.1) underpatientia. Other pairings of Cloelia and Horatius: Verg. Aen. 8.650-51 (without Mucius; likewise Serv. in Aen. 8.646) and Dio Cass. 45.31.1-2. All three are linked at Manilius 1.779-81; Juv. 8.264-65; Flor. 1.4.3. Extensive narrationof all three deeds (as in Livy) at Dion. Hal. 5.23-35; Plut. Publicola 16-19; De vir. ill. 11-13. 68. Earliest reference: Piso frag. 20 Peter = frag. 27 Forsythe = frag. 7.22 Beck-Walter (apud Plin. HN 34.29); discussion at Forsythe 1994, 252-57. Differences from the Livian account: she tricks the guards at Dion. Hal. 5.33.1, De vir. ill. 13, Schol. in luv. 8.264; she crosses alone at Flor. 1.4.3, Serv. in Aen. 8.646; by night at Val. Max. 3.2.2; on horseback at Val. Max. 3.2.2, Flor. 1.4.3 (alternative versions-swimming or crossing on horseback-given at Plut. Publicola 19.2, 8, Mor. 250C-F), De vir. ill. 13. Tarquinsattack returninghostages: Plut. Publicola 19.4-6, Mor. 250D-F, Fetialis apud Plin. HN 34.29 (all naming Valeria as escapee), Dion. Hal. 5.33.3-4.
IN ROMANCULTURE EXEMPLARITY
31
These variationshave spurredsource critics, as in the case of Horatius,to scrutinize the tradition for earlier and later (or "better"and "worse") elements, in the attemptto reconstructearlier versions of the legend.69In this case too, origins have sometimes been sought in early Roman cult and ritual, of which the story as found in surviving texts is seen as a transfigured or misrememberedrepresentation.70Even if such approaches do, in some cases, recover aspects of the legend's actual origins or conditions of production, they do so by seeking meanings that were largely or entirely lost to the authorsand intended audiences of our texts. Whateverthe origins of the legend, I agree that the Romans of this later period had largely forgotten, misremembered,or transfiguredthem. My own approach,again, is different. By examining this legend's representationand deployment in and through the monuments that survive from, or are attested for, the late Republic and Empire, I argue that Romans of this period comprehendedthe story and its variants within the discourse of exemplarity.They thereby put the figure of Cloelia into an ethical dialogue with their own day, with importantsociocultural consequences. Here I will not illustrate phases in the schema of exemplary discourse, as I did for Horatius. Instead, I focus on certain complexities in this discourse that are especially interesting or problematic in Cloelia's case. First I examine the moment at which secondary spectators encounter a monument;I consider what these spectators do and experience in the course of this encounter. Then I examine a particularethical and social paradoxthat Cloelia presents to these spectators:the paradox of the "manly maiden." Continuity and Analogy We observed above that Polybius' narrative of Horatius' deed (6.54-55) removes it entirely from its own sociopolitical circumstances: not a word about Porsenna,the expulsion of the Tarquins,or the like-information that Livy and Dionysius do supply as frames for the deed, and that many modern scholars would deem essential for proper historical understanding.Yet Polybius intimates that the Romans themselves recounted such stories in isolation from their historical circumstances-at funerals, for instance-just as he himself does in this passage. He furtherimplies that such accounts have primarily ethical force, since they endorse certain deeds as consequential and valuable to the community, and thus provide canons of value for future 69. Quellenkritik:see, e.g., Miinzer 1901, cols. 110-11 (s.v. "Cloelius" 9); Ogilvie 1965, 267. At least as old as Peter 1865, 49-50, is the idea that Valerius Antias invented the ambush episode, to give an heroic role to a member of the gens Valeria; Wiseman (1998, 84) has recently reasserted this. 70. E.g., Gag6 (1963, 60-62, 271-72) suggests that "Cloelia" and "Valeria"are originally not gentilic names but ritualfunctions, respectively "purification"and "health";similarly Gag6 1988, 241. Coarelli (1983, 1.86) accepts this view; Arcella (1985, 33-40) expressly rejects it. Taking a Dum6zilian line, Arcella sees the three heroes as "trickster"figures (27-31), on which the city must rely for its defense against a more powerful enemy. He argues that Cloelia's traversals of the Tiber establish and defend a principle of endogamy, but at the same time bring Romanjides into question (see pp. 43-44 below), which the actions of Valeria repair.In a different vein, Wiseman (1999, 198-99) suggests that the legend-at least the "Tarquin ambush"version-originated as a risque mime at the ludi Florales.
32
B. ROLLER MATTHEW
actors.That is, these accounts establish values and effect social reproduction. Polybius is right;throughoutRoman literaturewe find tellings of greatdeeds out of historical context. Valerius Maximus provides a massive collection, grouping his narrativesunder ethical rubrics and so making clear that his work's concerns are, above all, moral. Not just narratives, however, but monuments as such, are capable of historically decontextualizing action in this way. Many scholars have observed, for instance, that the honorific statues that filled the Forum by the late Republic were located in no chronological order:if they were, the statues of Horatius and Cloelia would have stood close together, and not at opposite ends of the Forum with (eventually) countless other monuments in between. Yet the fact that such monuments tell no coherent chronological story does not detract from their individual ethical force. Furthermore,the meanings that statues and other physical monuments may be taken to have individually are augmented and nuanced by their topographical location, their juxtaposition with other monuments, and so on.71But let us first consider how an individual Roman viewer encounters and engages a single monument. A Roman who encountersa monumentcommemoratinga deed-whether a narrative,a statue, or some other object-experiences a temporaldislocation. For the monumentis, in some sense, an outpost of the past: it has been thrust forward in time from the moment of the deed (or rather,of its commemoration) to meet the Roman in his or her own day, and to present the deed to this Roman's eyes or ears. As this Roman engages the monumentby reading or listening to a narrative,analyzing a statue's iconography,registering the type, number,and location of scars, or the like-the monument draws him or her back to the moment of the deed, thus constituting him or her as a secondary spectatorwho looks (as it were) over the shoulder of the primary spectators. At this point, our spectator experiences the social, ethical, and psychological effects already discussed: this person is invited to replicatethe primaryaudience'sjudgment,therebyaffirmingthe monument's ethical integrity; he or she should also feel the impulse to imitate, to gain similar glory. The monument,then, makes our spectatorcomplicit in a temporal collapse of past and present. For the spectator is pulled backwardin time, requiredto evaluate a past action by the same criteria that he or she would use in evaluating a contemporaryaction, and finally dispatched back to his or her own present with the idea that that deed thereby discovered is ethically relevant to one's own choices and actions, and those of one's contemporaries.72From this description, we can see that exemplary discourse and the monuments it encompasses presuppose both ethical continuity and 71. See Holkeskamp 1996, 308-15, 323-24, on the loss of chronological structurearoundexempla, and the ethical recontextualizing of events that this often enables. Jaeger (1997, 27, 50-53) argues that Livy, in constructing his narrative,systematizes the miscellaneous characterof a Roman's subjective experience of the Forum; he effectively "rearranges"its monuments, juxtaposing them in particularways, to confer particular meanings upon them. For statues furnishing models for imitation, see Bergemann 1990, 33; H61scher 1978, 340. 72. This description of an encounter with a monument is based on Jaeger 1997, 15-18, itself an interpretation of a difficult passage in Varro(Ling. 6.49).
EXEMPLARITY IN ROMANCULTURE
33
performativeanalogy. On the one hand, a monumentcan have such an effect only if the criteria of ethical judgment, and the social values that underpin these criteria,remain tolerably constant over time-constant enough for the actions and judgments of one period to remain comprehensible in another. On the other hand, to believe that a past action is ethically relevant to a contemporaryaction is to posit an analogical relation between these actions, to regardthem as sharing one or more properties-that is, as having structural or categorical similarities-that renderthem directly comparablein ethical terms, in light of the (assumed) condition of ethical continuity.73 Texts that discuss Cloelia illustrate profusely how past and present meet through the conduit of monuments, and illuminate the assumptions about continuity and analogy just described. Let us examine three texts in which Cloelia is explicitly rankedrelative to her contemporaries.Manilius, in the first book of his Astronomica, lists some Roman heroes who he says reside in the Milky Way. These include all the kings (except TarquiniusSuperbus), the threeHoratii,Mucius Scaevola, Cloelia-whom he calls "amaidengreater than the men"-and HoratiusCocles.74Here Manilius enters into the ranking debates of the remote past by asserting that Cloelia surpassed those male contemporarieswhom he lists along with her; she is presented as the greatest hero of early Rome. The elder Pliny enters this same debate, but with a different result. He remarks dyspeptically on Cloelia's equestrian statueand the honorit represents:"as if it were not enough that she be clothed in a toga, when equestrian statues were not granted to those who expelled the kings, Lucretia and Brutus, thanks to whom Cloelia had been included among the hostages."75Apparently,he takes the statue to mean that Cloelia's contemporaries-the primary audience-ranked her deed above Lucretia's and Brutus', since the latter do not have such statues.76But having been drawnbackwardin time by this statue to become a secondaryspectator to her deed, looking over the primaryspectators' shoulders as they evaluate her, he rejects their evaluation on the ground that, in his view, Lucretia's and Brutus' deeds were of greater consequence to the collective. Because he assumes that her deed lies as transparentlyopen to his own scrutiny and 73. My use of the terms "continuity"and "analogy"is indebted to Knapp 1989, esp. 129-32. On the historical continuity-even changelessness-presupposed by exempla, see Hilkeskamp 1996, 312-15; Stemmler 2000, 145. 74. Manilius 1.777-81: Romanique viri quorum iam maxima turba est, / Tarquinioqueminus reges et Horatia proles, / tota acies partus, nec non et Scaevola trunco / nobilior, maiorque viris et Cloelia virgo, / et Romanaferens quae texit moenia Cocles. ... Here I treat Manilius not as the creator of a monumental text by which his readers may encounter these early heroes, but as a secondary spectator himself to Cloelia's deed, who has encountered this and the other heroes' deeds through other, unspecified monuments, and records his own judgments in this text. 75. Plin. HN 34.28: pedestres sine dubio Romaefuere in auctoritate longo tempore; et equestriumtamen origo perquam vetus est, cumfeminis etiam honore communicato Cloeliae statua equestri, ceu parum esset toga eam cingi, cum Lucretiae ac Bruto, qui expulerant reges, propter quos Cloelia inter obsides fuerat, non decernerentur. 76. While no statue of Lucretia is attested from any period, one of Brutus stood on the Capitol, along with statues of seven kings, in the late Republic (anti-Caesariansquibs were written on it). The posture and dress of this statue are not reported. However, the kings were almost certainly pedestrian and togate; that Brutus was included in this group may suggest that his statue was similar. See Sehlmeyer 1999, 68-74; H1lscher 1978, 328-31.
34
B. ROLLER MATTHEW
evaluation as it did to her contemporaries,he can disagree with them and suggest that Cloelia should have received only a togate statue (presumably pedestrian),which he implies is a lesser honor and better suited to the deed's actualimportance.This phenomenon-the secondaryspectatorwho disagrees with his predecessors' judgments-is a key instability in the production of exemplary discourse, which makes it possible always to reappropriatean exemplum in a new way to meet new exigencies. Now consider again how Livy presentsPorsenna'sevaluation of Cloelia's deed. On this account, Porsennawas not among the primaryspectators,since the deed was reportedto him (quod ubi regi nuntiatumest, 2.13.7)-a secondary spectator, then, but contemporarywith the deed. Still, he evaluates it against contemporaryrivals and judges it superior (supra Coclites Muciosque ... id facinus esse, 2.13.8). This is precisely the evaluation that Manilius, Livy's younger contemporary,makes in his own voice, and that Pliny assumes (from the statue) that the primary audience made-though Pliny, as we have seen, contests this judgment. Thus it is evident that Livy imagines (and assumes his audience imagines) that Porsenna would have gone about evaluating Cloelia's deed in the same way and on the same standards as Livy and his contemporarieswould, and as Manilius and Pliny in fact do: namely, by comparing her deed to those of her contemporariesin the pertinentethical categories. The ideology of exemplarity makes no distinction between secondary spectators contemporarywith a deed and such spectators at a large temporal remove, regardinghow they evaluate a deed and its ethical implications for Romans of any era. These three passages illustratehow exemplary discourse presupposes the principle of ethical continuity.When Romans of the late Republic and Empire pass judgment on an ancient deed, they assume that both they and the original judging audience are playing the same game by the same rules. They may disagree with the original verdict, but they nevertheless assume that both parties pass judgment in light of persistent ethical standardsthat obtain equally in both eras. Moreover, it is a monument, of whatever form, that provides the conduit whereby later spectators can access and enter the past that they (re)evaluate.But recall that a monumentitself is dislocated in the opposite direction: it projects a deed forwardin time, to meet that spectator in his or her own day; the past is thus made available in any given present to inform the choices and actions taken by oneself and one's contemporaries.To see a monumentat work in this way, we returnto the "Philippic" of Cicero composed by Dio Cassius. Cicero complains that Antony, as consul, gave a speech from the Rostra while "naked,"having stripped to a loincloth to serve as Lupercus on the occasion of the Lupercalia (45.30). Declaring Antony's (un)dress an outrage to the dignity of the consulship (?30.2, 5), Cicero makes the following comparison (?31.1): ~ tlv KXot7iav -dca y' &vo-ro? 1TOv "OpdTtovtOv 7naXatv KEiVOVK1tI ti1v tpXaiav Iov 6 6 Kalt CpLttJifLctoLo, V i tEvc7'vkflccta traoav •v6E6UKUla TO'v IrOTaOiv 6t~VfllaTO, ' •V h v itSnkov?q 1-cr Kai to •autbv 't6v y& (o6 ydp;) tou tv Ec6va i5c3.Ota pu-rtdX EvjLpaksv. & KEt'v t? dyopr yuitv6W aCfioat,'Yv'(6 v KaliCtV TtpTtpt6tdcnktoat&voq,6 R 6p~tro.
EXEMPLARITY IN ROMAN CULTURE
35
Perhaps he imitated the famous Horatius of old, or indeed Cloelia of old, of whom the latter swam across the river in a fully clothed state, while the former threw himself into the stream with his armor on as well. It would be fitting indeed (how could it not be?) for a statue to be erected of Antony too, so that the one might be seen in armoreven in the Tiber, and the other might be seen naked even in the Forum.
Despite the assertion "perhapshe imitated ... ," Antony's actions described here-serving as consul, addressingthe people, playing the Lupercus-have no resemblance to the actions ascribed to Horatius or Cloelia. Rather, "Cicero's" comparison among these figures turns on their sartorial states. He invokes a moral hierarchyin which being clothed in the Tiber (like Horatius and Cloelia) evinces high virtue, and being nakedin the Forum(like Antony) indicates low vice. Presumably the other two combinations-naked in the Tiber, clothed in the Forum-are ethically unmarkedand neutral,being the "normal"sartorial states of persons who swim or engage in civic life, respectively. At any rate, monuments to the two archaic heroes provide the moral canons of which Antony's recent behavior falls short. These monuments are, first, the brief narrativesthat Cicero himself provides here, describingthe heroes' costumes while in the river;and second, Horatius'statue, which provides a strong moral contrastwith the imagined statue of Antony. For just as the former commemorates Horatius' (virtuous) swimming the Tiber in armor,so Antony's will commemoratehis (vicious) speaking from the Rostra naked. In this passage, then, monumentsof Horatiusand Cloelia vicariously bring these heroes into the present, making them available as standardsof ethical conduct against which today's actions can be evaluated. The principle of ethical continuity is clearly taken for granted.But this passage also illustrates, in an indirect way, the principle of performativeanalogy. Even though Cicero is ironic when he suggests that Antony "imitated" Horatius and Cloelia (for he focuses on how Antony diverges from the archaic heroes, not how he resembles them), this very irony attests the strength of the expectation that a contemporaryaction will and should have morally salient similarities to an ancient one, and that these actions can be weighed in a moral balance on the basis of their similarities. Indeed, Cicero carries his point precisely by imagining a statue of Antony in the Forum-clearly "analogous"to that of Horatius-whose juxtaposition with the image of the hero will (paradoxically) show not how nearly Antony rivals him, but how far he has fallen short.77 Cicero himself never makes such an argumentin the extantPhilippics. But in De officiis he does confirm that Cloelia can provide an authoritativestandard for evaluating contemporaryaction. At ?1.61, he discusses techniques for praising and blaming. The "praising"portion was quoted and discussed 77. By what iconographical difference might this moral contrast be carried? If Horatius' statue portrayed him in armor-and here again it must be understood as his own armor,the very armorthat marks his virtus in the river-then Antony's envisioned statue would have to portray him "naked," which might mean wearing only the perizoma that was the priestly costume of a Lupercus. Pliny (HN 34.8) mentions statues Lupercorumhabitu as a recent innovation in honorific statuary,and several such statues from the first to third centuries C.E.have recently been identified (Wrede 1983). So Dio Cassius may be imagining a statue of Antony in a specific iconographical form familiar to himself and his intended audience.
36
B. ROLLER MATTHEW
on pages 21-22 above; let us here consider the "blaming"portion. Cicero writes, "And so, one is best equipped to make reproaches if one can say something like this: 'for you young men have a womanly spirit, but that maiden has a man's'" (itaque in probris maxime in promptu est si quid tale dici potest: "vos enim iuvenes animumgeritis muliebrem,illa virgo viri"); he then goes on to discuss praising, for which one may invoke Horatius, the Decii, and so on. Now, the "reproach"that Cicero quotes here is a line of verse, probably a septenarius;Ribbeck tentatively assigns it to Accius' Meleager, in which case the virgo is Atalanta.78But because Cicero contrasts this "reproach"with the praise that arises from invoking Horatius(and other heroes), I suspect he imagines that his audience will understandthis reference as being to Cloelia-that she is the maiden with the manly spirit, whom the orator can invoke as a standardof which certain people fall short here and now. A strikinginstance of Cloelia doing exactly the job this verse describesprovidinga canon of manlinesssurpassinganythingtoday's men can musteris found in Seneca's Consolatio ad Marciam, addressedto a woman whose father has recently died. About midway through this treatise, Seneca adduces examples of men who courageously endured the deaths of family members, refusing to be bowed by grief (12.4-15.4). He then imagines his addressee Marcia making an objection as follows (16.1-2): (1) scio quid dicas: "oblituses feminam te consolari, virorumrefers exempla."quis autem dixit naturammaligne cum mulierum ingeniis egisse et virtutes illarum in artumretraxisse? par illis, mihi crede, vigor, par ad honesta, libeat (modo), facultas est; dolorem laboremque ex aequo, si consuevere, patiuntur.(2) in qua istud urbe, di boni, loquimur? in qua regem Romanis capitibus Lucretia et Brutus deiecerunt: Bruto libertatem debemus, Lucretiae Brutum; in qua Cloeliam contempto et hoste et flumine ob insignem audaciamtantumnon in viros transcripsimus:equestri insidens statuae in Sacra Via, celeberrimo loco, Cloelia exprobratiuvenibus nostris pulvinum escendentibus in ea illos urbe sic ingredi in qua etiam feminas equo donavimus. (1) I know what you are saying: "You have forgotten you are consoling a woman; you offer up exempla of men." But who said that naturedealt parsimoniously with women's spirits and constrainedtheir virtutesto a narrowdomain?Believe me, they have the same energy, the same capacity for honorable deeds, if they please; they endure pain and toil on equal terms, once they have grown accustomed to it. (2) In what city, by the good gods, are we uttering this? One in which Lucretia and Brutus threw down a king from the Romans' necks (we owe freedom to Brutus, and we owe Brutus to Lucretia); one in which we have all but enrolled Cloelia as a man because of her outstanding boldness, despising both the enemy and the river. Sitting upon her equestrian statue in the Sacred Way, in a heavily frequented place, Cloelia reproaches the young men of today as they climb up onto their litters, that they go about thus in a city where we have honored even women with a horse.
Here Seneca develops the genderinversionposited in the verse Cicero quotes. Cloelia has all but been reclassified as vir on account of her "outstanding boldness" and has indeed received a manly honor, sitting atop a horse like 78. TRF2,locus incertus,fabula incerta 210.
IN ROMANCULTURE EXEMPLARITY
37
a malehero;meanwhile,today'syoungmenareconveyedin littersin a stereotypicallyeffeminatemanner.ThusSenecanot only adducesCloeliaas a but also hypostatizesthe injunccanonfor evaluatinghis contemporaries, tionherstatueimplicitlydelivers,by makingherspeakfromherpositionon horseback,whichso conspicuouslymarkshermanlyquality.The"reproach" she casts at today'syouthis "youmustdo as I did,in orderto get this kind of mount."The pastcouldhardlybe broughtinto the presentmorevividly: evalthe archaicherois givenvoice, becominga spectatorandauthoritative to meauatorof today'ssocialactors.Farfromleavingit to contemporaries suretoday'sactionsagainstcanonicaldeedsof thepast,theheroherself,the doer of one such deed and thereforethe most authoritative judge imaginable,assumesthatrole.The statueitself enablesthis fiction,as the figureof Cloelia upon the horse, in a "verycrowded"spot on the SacredWay,is her,where imaginedto be aliveandsensibleto all the activitiessurrounding she can see andjudgeeverythingthatoccursin the Forum,just as everyone therecansee her.OnSeneca'saccountshe reallyis in the present,projected in asserting forwardby her monumentfromher own day.79Furthermore, thatshe speaksto "our"youth(exprobratiuvenibusnostris),andthat"we" havehonoredherwiththis statue(etiamfeminasequodonavimus),Seneca to the primaryspeccompletelyassimilateshimself andhis contemporaries tatorswho originallyapprovedherdeed.Thispassagestrikinglyillustrates how a specificmonumentconstitutesthe relationshipbetweenpresentand past as one of ethicalcontinuityandperformativeanalogy.The conditions the evaluationof the actorsareassumedto persistunchanged underpinning fromCloelia'sdayto thepresent;whatconnectsthe actors-the morallysalientpropertytheyshare-is thattheyareall mountedon someformof conveyance.Wherethey differ,andwhereSeneca(/Cloelia)drawsan ethical distinction,is in the precisecharacterof thatconveyance,sincehorsesand littershave differentmoralvalencesin this context. We have now seen, in Cloelia'scase, how monumentsprovidea conduit betweenthetimeof a deedandanysubsequenttime,in theserviceof ethics. By meansof a monument,Romansin any given presentcan travelbackwardto the momentof the deed and,as secondaryspectators,validateor second-guessthe evaluationthattheprimaryspectatorsbestowed.Theycan also bringthatdeed forwardto theirown presentas an ethicalcomparandum for contemporary actions.Indeed,monumentsof every sort function this way because,by theirverynature,they dissolvethe stricturesof chronology and sociohistoricalcontingency.A monument'sraisond'etreis to thrusta deedbothoutwardin spaceandforwardin time,makingit available 79. This ideologically potent inversion, where the exemplary hero is revivified to sit in judgment on posterity, can be observed in many monumental forms. See, e.g., the epitaph of Cn. Cornelius Scipio Hispanus (ILS 6 vv. 3-4): maiorum optenui laudem ut sibei me esse creatum / laetentur; likewise at aristocratic funerals the maiores sit in judgment on the newly deceased and on his son or other relative who delivers the oration (Polyb. 6.53-54, with Habinek 1998, 53; Hblkeskamp 1996, 321-22; Flower 1996, 128-31). Also Cicero, in a striking prosopopoieia (Cael. 33-34), summons up Appius Claudius Caecus to rebuke his descendant Clodia by comparing her doings to his own and those of other Claudii (e.g., ?34: ... ideone ego pacem Pyrrhi diremi ut tu amorum turpissimorumcotidie foedera ferires?). Wray (forthcoming) discusses the Senecan passage (Consolatio ad Marciam 16) further.
38
B. ROLLER MATTHEW
and comprehensible to persons elsewhere and elsewhen. These characteristics of monuments, in turn,expose an importantaspect of Roman historical consciousness. In the discourse of exemplarity, the past is by no means a "foreign country," but is ethically and culturally homogeneous with the present. The "Manly Maiden" Every monument, commemoratingevery doer of deeds, can and must transcend historical contingency in the mannerjust described; Cloelia's monuments are illustrativebut not distinctive in this respect. But in anotherrespect Cloelia is strikingly distinctive, and unlike any other exemplary figure: she is a female-in particularan unmarriedone, a virgo-who displays "manliness," virtus. The texts that attributevirtus to her clearly intend it to carry its full etymological force, for they often formulate the resulting paradox with tight rhetoricalpoint: Livy makes the Romans remarkupon the novelty of virtus in a woman (novam in femina virtutem,2.13.11); Valerius Maximus describes her as "carryingthe torch of virtus for the men, though a girl" (virispuella lumenvirtutispraeferendo,3.2.2); Maniliusmakes her a "maiden greater than the men" (maiorque viris et Cloelia virgo, 1.780); and Florus exclaims that even maidens had virtus (ecce et virginum virtus, 1.4.3)-a tongue-in-cheek figura etymologica, as if virtus were derived from virgo, not vir. Such wordplay even occurs in Greek, since &v6pciaparallels virtus (in the narrower,military sense) both in its ethical force and in its etymological connection to "man."Thus Polyaenus says that Porsenna praised "the manliness of the maiden"(Tzb 0vSpiov zi; K6pril,8.31.1), and Plutarch extends the pun by declaring that, on account of her manliness (tzb vSpd(6cS azoTfl), she received a statue (v6Spt`goatzil, Publicola 19.8).80 But what does it mean for late republican and imperial Romans to categorize Cloelia's actions under the rubric of virtus (or &v6psda)in the first place? Manifestly, an ideology of gender is encoded in the use of words derived from "man"to label socially consequential, ethically valued feats of military courage. To assign this value to a female is, inevitably, to challenge this ideology in at least one of two ways. Namely: does "womanly virtus" problematize the concept of virtus by asserting that there is a specifically womanly sort of virtus-a category of socially valued actions bearing the name virtus, but not in fact identical in content to the virtus of men? Or does it problematize the category of vir by extending it to include a female, so that the criteria employed for categorizing her actions under the rubric virtus are exactly or largely the same as those employed in regardto viri? Both approachescan be found in late republicanand early imperial texts that ad80. Scholars accept that virtus (in its narrower,traditional usage) and avapeialabel the same (primarily) military value and encode the same gender ideologies: so McDonnell 2003, 235-36; also (apparently) Wray forthcoming; and McInerney 2003. This assumption seems largely correct to me, though certain differences exist (Eisenhut 1973, 13, 175) and a systematic study is needed. In contrast, the relation between virtus (in its broader, philosophical usage) and dpeTrshas been studied in some detail: McDonnell 2003, 241-43, 247-58; Eisenhut 1973, 14-22; also Wray forthcoming; Roller 2001, 22-26.
EXEMPLARITY IN ROMANCULTURE
39
dress gender and ethics, including those involving Cloelia.81In her case, I will argue that our texts do not consistently prefer one approachover the other, but tend to mix them up even in the same account. This tendency affects the social and ethical dynamics of exemplarity in Cloelia's case, and also signals a complexity in the evolving discourse of gender at Rome in this period. As a first approachto Cloelia's "manliness,"we may note how she compares to other Roman women of exemplary status. As recent scholarshiphas convincingly demonstrated,traditionalGraeco-Romandiscourse on the virtues of women tends to focus on women's bodies, and on behaviors associated with the body-especially sexual conduct, and the matter of which men do or should have sexual access to them. Exemplary women from early Rome, such as the Sabine women, Horatia,Lucretia, and Verginia, function this way: theirbodies are objects of contestationamong rival groups of men. But this contestation has furtherconsequences: threatenedor actual sexual violations of these women's bodies echo, or constitute, threatenedor actual political violations of the civic body. Women's bodies function in these ways because they are conduits for both lineal descent and marriagerelations, which often exist in tension (to the point of snappingviolently) in the legends of early Rome. I cannot pursue these mattershere, but would stress that Cloelia, the virgo, functions differently. Being ex hypothesi too young for marriage,the potential or actual cognate/agnatetension that crystallizes in other female figures is absent in her case. She does not function like other exemplary women because she is not (yet) a woman.82In fact, she shares with properly constituted men the characteristicof never having been sexually penetrated,which perhaps forges a conceptual link between the categories of virgo and vir.83 Being "notpenetrated"may be a necessary condition for "manliness,"but is hardly sufficient. The "manly" ethical categories of virtus and av8psia must, like most other Roman ethical qualities, be won and maintainedby the performanceof consequentialactions in the public eye, in this case normally 81. The first alternative is conservative in respect to gender categories-keeping men's and women's deeds, hence social roles, distinct-but radical in altering the content of the ethical category virtus. The second is radical in respect to gender categories (a female is gendered as a vir) but conservative in regard to the ethical category virtus, making its contents invariantregardless of the actor's sex. Wray (forthcoming) examines these alternatives, and their implications for early imperial politics and gender discourse, in Seneca and Valerius Maximus; McInerney (2003) examines a similar dilemma in Plutarch'sMulierum virtutes (Mor. 242E-263C). 82. On the primarilybodily virtues of women see Mclnerney 2003, 328-41; Wray forthcoming.On early Roman exemplary women and their functions see, e.g., Miles 1995, 190-96, 207-12 (and chap. 5 passim); Joshel 1992, 121-28; Konstan 1986, 210-13. The figures of Tarpeia,Tanaquil, and Tullia can be analyzed similarly. Since the legends surroundingthese early women reveal tensions between endogamy and exogamy, perhaps a similar anxiety hovers aroundthe Cloelia legend: thus Arcella 1985, 36-38, contends that Cloelia's flight across the Tiber confirms a principle of endogamy. Yet Cloelia's undisputed status as a virgo who is not yet marriageablewould seem to minimize this anxiety; even if she were sexually violated, this would not result in problematic cognate relations. See Bernard 2000, 214-19, on Livian portrayalsof women, noting Cloelia's exceptionality. 83. Late etymologies connect virgo with vir through words like vis and viridis (Barton 2001, 41-42). But whether Romans of the late Republic and early Empire regardedthese words as connected etymologically, or saw any other intrinsic link between these social categories, is uncertain.
40
B. ROLLER MATTHEW
(if not exclusively) throughdisplays of valorin combat.84Therefore,to understand what Romans meant when they attributed"manliness" to Cloeliadid they mean that she did exactly what a (real) man would have done, or that she did some other kind of deed that the category virtus was extended to encompass?-we must closely examine her action and the evaluations our texts say that it received. On the one hand, the evaluation is always positive, and she is almost always credited with "manliness"in some formwhether by having virtus or &v6pEiaascribed to her, or by being compared favorably to men. On the other hand, the accounts vary regardingwhat, precisely, she did. Livy, for instance, lists five distinct achievements: she deceives the guards, leads the other girls, avoids the javelins, swims the river, and restores the hostages safely to their families (frustratacustodes dux agminis virginum inter tela hostium Tiberimtranavit, sospitesque omnes Romam ad propinquos restituit, 2.13.6). Somewhere here "manliness"resides, for on these grounds Porsenna ranks her above Horatius and Mucius (?8), and both he (?9) and the Romans (? 11) honor her virtus. Dionysius, meanwhile, lists only three elements: she tricks the guards rather elaborately, asking them to withdrawso that the hostages can modestly undressto bathe in the river; then she leads them in swimming across (5.33.1). "Manliness" is here too, for Porsenna subsequently praises her as "having a spirit surpassing her natureand age"-that is, surpassingher female natureand childish age: the spirit, in short, of an adult male-and deems the city blessed "not only for rearing good men, but also maidens equal to the men."85Still other texts mention only one or two elements. Yet even in these spare accounts Cloelia's actions are often called "manly"in one of the ways described above; this evaluation comes either from her own contemporaries (the Romans or Porsenna),or from the authorhimself in propria voce-the secondary spectator thus reaffirmingand validating the original audience's positive evaluation.86 84. Scholars interested in ancient discourses of gender have recently demonstratedthat "manhood"does not devolve upon a person by virtue of biological sex or age, but is a prized, precarious quality that must be won competitively and maintained vigorously through social performance (e.g., Gleason 1995; Barton 2001, 38-43). This understandingis entirely consistent with the discourse of exemplarity described in this paper-indeed, it is through exemplary discourse that virtus (like most other social values) gets assigned by judging audiences to specific social actors on the basis of actions done under their gaze. 85. Dion. Hal. 5.34.3: payv 6~inape0vov Ktitv 6jupipv, bp' ig i••siopoav ai Xotnai 8tavti•aoeat 6bv noTae6v, linatvi~aagd KpeizrovEouoav (pp6vrlqpa Tfg ze c 6seoKai rg filXtKig, Kai ilv r6XtyvpaKapiaag KaiIrape~vougdvSpdatv6poiag. Plut. (Publicola 19.2, 8) ZT pil p6vov &v6pa d&yaOo Kp•(6PEstv,&kX•& rti also mentions her leadership and her swim (though here she does not deceive the guards, but opportunistically flees when they are not watching), and has Porsenna deem the performance "manly:"(K56' oirTertvd T6pov.... 6ppifv EaXov npbq EOslpa nrok6Kai 8iva pa3Os{ag. Evtot 6 (pauotpav &tnovA14ao0at pu),Kifv a6eifv al; (kkat veo6eats Kai itapa0app66vopa KkotMav i'nnu 68tssekdaoatv rn6pov, ZT cyKEX•eopvIlv vouoeav.... Tripoat Tb abe6rgrbv Tupprv6v (similarly at Mor. 250C-D). &v6p8e•S of 86. Swimming and leadership hostages, with explicit judgment of "manliness": Polyaenus, Strat. 8.31.1: pia 6t 4 abzTv KkothXa po3 ivagtq~a0iatq ... ro rtO b agoT ' tpoi'rpTWEv toaaqS 8taviao••a0atT~6pla Kai TKlv 86onopov. itsi 6~ 6teviavro, Popiot z•l Pv dpE~ilv apv6tv &vSpiav 0a6paaouav... Flopoivag 6nEpayao0icgT6bdvSpEbovrTf K6plg. Swimming (or riding) and deception of guards, with judgment of "manliness": Val. Max. 3.2.2: custodiam egressa equum conscendit celerique traiectu fluminis . .. viris puella lumen virtutis praeferendo; Flor. 1.4.3: elapsa custodiam Cloelia per patriam flumen equitabat. et rex quidem tot tantisque virtutumterritus monstris; De vir. ill. 13: deceptis custodibus noctu castris eius egressa equum ... arripuit et Tiberim traiecit ... cuius ille [sc. Porsenna] virtutem admiratus. ... Only
EXEMPLARITY IN ROMAN CULTURE
41
Let us examine the three most common elements-crossing the river, leading the girls, and deceiving the guards-for their potential "manliness," whether individually or in combination. Since the river crossing, with swimming sometimes replaced by a ride on horseback, is present in every account of her deed, it seems a good candidate for the irreducibly "manly" aspect. We saw in Horatius'case that swimming is a stereotypically"manly" exercise associated with military training and, occasionally, featuring in a battle narrative.Thus, girls who swim might be categorized as "manly"even without being wounded or wearing armorlike Horatius and his imitators.87 The variants in which she crosses on horseback may betray a rationalization, in that a girl might be thought more likely to succeed this way than by swimming unassisted. But this scarcely affects the action's ethics, for imperial texts praise the bravery and enterprise of German horsemen who cross rivers with their mounts-this too, then, falls within the ambit of virtus.88Besides, in this version she can be credited with the opportunisticseizure of a horse (Val. Max. 3.2.2, De vir. ill. 13). The river crossing, in whatever form, therefore appearsto be at least one locus of "manliness."It is also a specifically masculine sort of virtus, as Cloelia has done exactly what men do to be so categorized. What of her leadershipof other girls? Livy's account gives her deed a decidedly militaristic color: she is a dux agminis virginum, and escapes inter tela hostium(2.13.6). As dux virginumshe resembles aristocraticmales who, as generals (duces), command troops of men in wartime; these aristocrats, discharging their magisterial duties in combat, engage in activities central to the category virtus. That the band of girls is described as an agmen, and thatjavelins fly aroundthem, furtherreinforcesthe militaryovertones;Livy's account all but places Cloelia and her followers in combat.89Otheraccounts of her leadership are less militaristic, crediting her instead with compelling rhetorical skills by which she urged the other hostages along.90Either way, her leadership falls into a characteristicallymale, aristocraticpattern.Texts that representher action this way are again extending the category of vir to encompass this virgo, because she has done what men do to be credited with virtus. Yet, many accounts that do not mention her leadership still credit her with "manliness."Thus this aspect of her deed, when present, may contribute to an overall "manly"effect, but does not constitute that effect by itself.
river crossing (swimming) noted: Dio Cass. 45.31.1, Juv. 8.264-65, Sil. 13.828-30 (with "manliness" judgement). Unique is Sil. 10.496-98: facta virum sileo. rege haec et foedere et annis / et fluvio spretis miranteminterrita Thybrim/ tranavit. 87. Some accounts, however, make her achievement more impressive by insisting (as for Horatius) that the water was rough and the swim difficult: Plut. Publicola 19.2, Mor. 250 C-D; Polyaenus, Strat. 8.31.1. 88. Whether these troopers remain mounted while their horses swim, or dismount and swim alongside their horses, is unclear: Tac. Agr. 18.4, Hist. 4.12.3; Dio Cass. 69.9.6. 89. Livy may include the javelins to make her deed more like Horatius', who also (on Livy's account, 2.10.11) swam safely to the Roman side amidst a shower of Etruscan shafts. 90. For her skills at persuasion and exhortation, note the verbs with which her leadership is described: 7ei0o (Dion. Hal. 5.34.3), ayKlsk6oplatand napaOapptvs (Plut. Publicola 19.2, Mor. 250D), npocpi•n7o (Mor. 250C, Polyaenus, Strat. 8.31.1).
42
MATTHEWB. ROLLER
Now consider the third common element, the deception of the guards. Roman attitudes toward military trickery and deceit vary greatly, so there need be no surprisethat Cloelia's stratagemis nowhere condemned as such (though discomfort attendsher incidental violation of the terms of the truce; see below). Nevertheless, stratagemsare by definition alternativesto direct, open confrontationon the battlefield,traditionallythe primarylocus of masculine virtus. Thus, her stratagem seems unlikely to be a manifestation of such virtus.91Perhaps,however, a Roman could hold that deception constitutes a "womanly" sort of virtus, through which a woman pursues an end that a vir would pursue through(say) violence. This seems to be Silius Italicus' view, as he distinguishes Cloelia's achievement from that of a hypothetical male actor: "if naturehad changed her sex, perhapsPorsennawould not have returnedto his own territory"-implying, I take it, that as a man she would have killed him ratherthan merely escaping him.92 Some accounts include a furtherelement that softens the masculine character of her leadership and swimming. The story goes that Porsenna, upon her return to his camp, allowed her to choose any hostages she wished to take back to Rome; she chose the impubes, as being most at risk of sexual exploitation. While concern for the bodily integrity of freebornchildrenwas of course widespreadin Roman society, and by no means limited to any particular status group, Livy pointedly says that Cloelia's choice "befittedher status as virgo" (virginitati decorum, 2.13.10)-implying that it is particularly appropriatefor a freeborn impubis to seek to protect other freeborn impubes from what most threatens their status. This story appears also in Servius Auctus and De viris illustribus, though without the overt declaration that it befitted her own status.93To the extent that her virtus resides in this action, it is (on Livy's account) a fittingly childish form of virtus, rather than masculine or feminine-even while, in these very same accounts, her virtus also clearly resides in her (masculine) swimming and/or leadership. Looking at all these accounts as a whole, then, Cloelia's virtus seems to have been imagined to reside principally in masculine sorts of achievements (crossing the river, leading the girls), making her a gender deviant-that is, an honoraryman who does deeds such as men do. But her virtus could also, or alternatively,be seen as residing in achievements considered appropriate to her specific sex or age (deceiving her guards;saving the impubes) so that 91. The contrast between stratagemand virtus (on the battlefield) is manifest in the stories of the Faliscan schoolmaster, whose attempted betrayal is shunned by Camillus (e.g., Livy 5.27, esp. ??5-8: "non ad similem " inquit "tui nec populum nec imperatoremscelestus ipse cum scelesto munere venisti. . .. ego Romanis artibus, virtute opere armis, sicut Veios vincam"), and that of FabriciusLuscinus, who alerts Pyrrhus to a poisoning plot (e.g., Plut. Pyrrh. 21, esp. ?4: ... 6irsoqPlilbT obv rd0og l?iv stanloXilv6vKyKTI Kal Ch 86ho 86eopev, dkpEr fiPl 8UVdipvot,Kat'Epydinoat tbvy nTh•epov).On the moral values associated with in Roman see 50-92. Wheeler warfare, 1988, stratagem esp. 92. Sil. 10.499-501: cui si mutasset sexum natura, reverti /forsan Tyrrhenastibi non licuisset in oras, / Porsena. On trickery and deceit as distinctively female spheres of action see McInerney 2003, 333-35; in a different vein, Arcella 1985, 27-29. 93. Serv. in Aen. 8.646: qui admiratus virtutempuellae dedit ei optionem ut cum quibus vellet rediret. illa elegit virgines, <Servius Auctus: quae iniuriae poterant esse obnoxiae,> unde Porsenna hoc quoque miratus concessit; De vir. ill. 13: cuius ille virtutemadmiratus cum quibus optasset in patriam redire permisit. illa virgines puerosque elegit, quorumaetatem iniuriae obnoxiam sciebat.
EXEMPLARITY IN ROMANCULTURE
43
she remains identified as a female and/or child, doing actions that are merely dignified with the label virtus and not at all identical to what men do. These alternative visions of "womanly virtus" can coexist in a single account, as when she both swims and tricks the guards. In Livy the juxtaposition is especially striking. He writes that, when Cloelia was returnedto Porsenna,"hervirtus was not only kept safe, but even held in esteem" (apud regem Etruscumnon tuta solum sed honorata etiam virtusfuit, 2.13.9). The virtus that Porsenna keeps safe must be her bodily integrity as a freeborn Roman child, fulfilling the promise of inviolability he had made earlier (sic deditam inviolatamquead suos remissurum,?8). But the virtus he honors must be that she displayed in leading the girls across the river, the masculine deed for which he ranksher above Horatiusand Mucius (?8) and which the Romans later honor with a statue (Romani novam in femina virtutem novo genere honoris, statua equestri, donavere, ?11). Thus Livy, by a kind of zeugma, asserts both the "childish virtus" and "masculinevirtus" aspects of her deed simultaneously. Such accounts show that her deed is a composite of two or three distinct ways of comprehending virtus, and as such poses a conundrumfor received gender and ethical categories. Inseparable from her "manliness" is a further ethical complexity. The hostages were pledged as security for a truce; their flight, whatever virtus it displays, abrogates the truce and undermines Roman credibility. Plutarch says that Publicola "neither marvelled nor rejoiced" at the hostages' safe arrivalin Rome, "but was vexed that he might appearworse than Porsenna in his trustworthiness(RtioG), and that the daringof the maidens might give cause for accusing the Romans of trickery."Thus Publicola holds that the girls' action harmedthe collective in a key respect, wherefore he evaluates it negatively in the category of fides/ntioTtG.94In other accounts it is the Romans collectively, or Porsenna, who object that Roman trustworthinesshas been tainted, even while they praise the virtus/&v6pfiaof the escape itself-thusjudging her deed negatively in one category but positively in another.95 Her glory is therefore tarnishedunless the breach of Jides can be repaired. Hence the cumbersome exchange whereby the Romans returnthe hostages to Porsenna(as the agreementrequires), who immediately releases them (in acknowledgment of their valor): in this way the Romans reclaim a positive evaluation for from themselves and Porsenna, and thus sweep fides/niort• away the negative ethical repercussions of Cloelia's deed so that she, and the city, can reap the benefits of its positive repercussions. Far from nullifying Cloelia's deed, then, her returnto Porsenna enables it to enter upon its full, beneficial effect for both actor and community.96The deed itself can then be monumentalized, closing the discursive loop in which she 94. Plut. Publicola 19.3: OK 066' 'yarloEv, X ' ilvid6Or, rst lopaivva i0a6tpacEv ?pavsicat, KaiTb T6kpipa TCvnapOivov aitiav i'Et KaKoupyrg7a 'Popai•v ysyovivat.
Iv v xioTt KaKiow
95. Mixed judgments by Romans: Plut. Mor. 250D: irnciU omo0Efa;gSiov oi 'PToPaiot, , rilv Alyv perTv 6ripetvav iv nio~st XEipovEg iv6&vb6pbg KCi tyTIv6kpacv 90a6eLaovV, Tflv 55 KOptSiC v 06K1cydinroav o068"' ysvio0at (likewise Polyaenus Strat. 8.31.1). By Porsenna:Livy 2.13.8-9; Dion. Hal. 5.33.2, 34.3. 96. Breach effaced: Livy 2.13.9: utrimque constitit fides; et Romani pignus pacis ex foedere restituerunt, et apud regem Etruscum non tuta solum sed honorata etiam virtusfuit; Dion. Hal. 5.34.3: 6 8i Tv
44
MATTHEWB. ROLLER
successfully imitates previous exemplary actors and is herself installed as a model for future imitation.97The community not only adds to its stock of exemplary deeds, but also escapes its immediate crisis, as the consuls and Porsenna conclude a final peace. If Cloelia can be blamed for breaking the truce, then, she can also be credited with ending the war.98 These competing judgments on Cloelia's performance,supposedly by her own contemporaries,illustrate yet anotherinstability and complexity in the productionof exemplary discourse. Many deeds admitof evaluation in multiple ethical categories, with conflictingresults. A primaryaudiencemay split in its evaluation, and any secondary audience can turna primaryaudience's judgment on its head by locating and exploiting such fissures. Livy portrays an enemy of Rome doing exactly this in Book 9, where the Samnite general Pontius complains that the Romans are violating the terms of their surrender at Caudium: "Will you never lack a reason for not abiding, in defeat, by your agreements? You gave hostages to Porsenna, and smuggled them out throughtrickery... ." By omitting to mention Cloelia's own actions and the subsequent returnof the hostages (which supposedly effaces the breach of fides), this enemy (on Livy's presentation)can invoke her as an exemplum not of Romanvirtus,but of the very perfidyof which he now accuses them.99 Perhaps the most important index of her "manliness," however, is her association with a particularkind of honorific statue. The texts that discuss this statue provide information that is roughly consistent, in two respects. First, its location is said to be on the summit of the Sacred Way, or where the Sacred Way enters the Forum, or opposite the temple of JupiterStatorin the vestibule of TarquiniusSuperbus' house. Interpretedgenerously, these descriptionscan be taken as referringto the same location. Second, the statue representeda woman on horseback,and accordingto Dionysius (5.35.2) was made of bronze.'00 These accounts also diverge on key points, however.
Pa -cdv KXO46caq&ohiOGaCGtij -6tndt(p Eictiv, Tupprlvwvpaat••e5 r&6p•Lpa /t 'Pst0aiov 7ntTO' la npoaXcOvat & t n7Tdorl6VIrpeia?sKpsFiTovalyEzat Tilv iazttv tri; n6kesEg.Also, implicitly, at Plut. Mor. 250D; De vir. ill. 13; Serv. in Aen. 8.646. McInerney (2003, 334) is incorrect to say that her deed is nullified by her return to Porsenna, though he discusses well the deed's fides repercussions. See also Arcella 1985, 29-31, 36, 40-41. 97. An imitator of Cloelia appears in Silius Italicus, Book 10-a young cavalryman named Cloelius, dying on the battlefield after Cannae. His valor is ascribed to his descent from the family of the legendary Cloelia (472-502), though his deeds are not narratedand therefore no structuralparallels are evident. Silius probably invented Cloelius from whole cloth; therefore the connection he forges to Cloelia offers striking evidence that Romans of the late Republic and early Empire (namely, Silius and his intended audience) readily assumed that notable deeds ran in families. 98. Val. Max. 3.2.2: non solum obsidio se sed etiam metu patriam solvit; Sil. 13.828-30 (Scipio in the underworld, surveying the shades of heroic women): illa est quae Thybrim,quae fregit Lydia bella [i.e., she swam the Tiber and ended the war ...] / nondumpassa marem [. .. though just an unmarried/unpenetrated girl .. .], qualis optabit habere / quondam Roma viros, contemptrix Cloelia sexus [... and so has done what real men should do]. 99. Livy 9.11.6: numquamnecausa defiet cur victi pacto non stetis? obsides Porsinnae dedistis, furto eos subduxistis. Cf. Chaplin 2000, 40. Another such inversion is Lucan's evaluation of Scaeva's deed (6.257-62), in contrast to the evaluation Scaeva's own fellow soldiers bestow (251-56): the latter judge his deed positively as an astonishing display of virtus, while the narrator,without disputing the judgment of valor, condemns him as acting contrary to the community's interest, bringing slavery upon it. 100. On the statue's form and location: Livy 2.13.11: statua equestri donavere; in summa Sacra Via Enti iTsp1q E51oorxv, fuitposita virgo insidens equo; Dion. Hal. 5.35.2: oziotv siKovog yakKfq yV zig &v9O•EOav
IN ROMANCULTURE EXEMPLARITY
45
Several report a variant in which the honorand is not Cloelia but Valeria, Publicola's daughter who escaped the ambush that (on some accounts) the Tarquinslaid for the hostages. However, the surviving texts always report this alternative as a variant, suggesting that our authors-or the tradition they inherited-had already constructed a hierarchy of versions in which the Cloelia identification was preferred.10'Also, different dedicators of the statue are reported:the senate, the populus Romanus, or the hostages whom Cloelia led to freedom (or their fathers).102In any of these cases, however, the dedicators represent a group of evaluators whose interests are those of the community at large, and who deem the action both consequential and beneficial for this community. In this case, as in Horatius',archaeologists question whether an honorific bronze equestrianstatue could have been erected in the sixth century B.C.E. The earliest attested honorific equestrian statues that are undoubtedly historical date to the late fourth century B.C.E.Therefore, if this statue was indeed honorific, it must have been erected long after her deed (c. 300 B.C.E.). By the time of our texts the actual conditions of its erection had been forgotten, and the statue was mistakenly assumed to be contemporarywith her deed. Alternatively, if the statue really dated to the sixth century, it must have been the cult statue of a goddess, say Venus Equestris or Venus Cloacina or Vica Pota; in time its original meaning was lost, and a new, honorific meaning was constructed in connection with Cloelia (making this another instance, then, of an obscure object renderedcomprehensibleby integration into exemplary discourse). To complicate mattersfurther,Dionysius reports that the statue had been destroyed by fire prior to his own day and no longer stood (5.35.2); yet Seneca, Pliny, Plutarch,and Servius speak of it as being visible to their eyes. Scholars reconcile these accounts by conjecturingthat the statue was reerected-plausibly by Augustus, who restoredother ancient monuments,erected the statues of other republicanheroes in the ForumAugustum, and so on.103
Eig oiV&pdtv pepo6or; Fetialis apud Plin. HN 34.29: [statuam] equestrem contra lovis Statoris aedem in•ivestibulo Superbi domus; Sen. Consolatio ad Marciam 16.2: equestri insidens statuae in Sacra Via, celeberrimo loco; Plut. Publicola 19.8: 8E Ei 'VlV iEpbv 686v &vdIcst•at nopeuooE•voP; flaaidsTtov avSptdga6tilg ipqtrnog(cf. Mor. 250F); De vir. ill. 13: huic statua equestris in foro posita; Serv. in Aen. 8.646: statua equestris quam in Sacra Via hodieque conspicimus. 101. Nevertheless, the Valeria version is carefully preserved and transmitted,if only as a variant. See Fetialis apud Plin. HN 34.29; Plut. Publicola 19.8, Mor. 250E For Wiseman (1998, 84), the Valeria variant is a fabrication of Valerius Antias; Forsythe (1994, 255-56) explains how Valeria could have been the original identification, later displaced by Cloelia. 102. Dedicator: Romani, Livy 2.13.11; hanc [sc. statuam] publice dicatam crediderim, Plin. HN 34.29; populus Romanus (at Porsenna's request), Serv. in Aen. 8.646. Dion. Hal. (5.35.1-2) says that the senate decreed it, but the hostages' fathers actually erected it (i.e., bore the cost?), while Piso (apud Plin. HN 34.29) says the other hostages dedicated it (perhaps implying a private dedication). See Forsythe 1994, 256. & 103. Statue destroyed: Ta6Trlv KEt9vrlyVs•popsv, IX7yrTo8' plCpi)cstOgrsdS L•tq •?V OUCKEtt npi .rloiov oiKia y7EvogjivrIg r1'pavio0at(Dion. Hal. 5.35.2); cf. e.g. Serv. in Aen. 8.646: cui data est statua equestris... quae hodieque conspicimus (he supposes that the statue he sees, in the early fifth century C.E.,is some nine hundredyears old). On the problem of early honorific statuaryat Rome, cf. n. 47 above. On the dating, original identity, and posited reerection of Cloelia's statue see Sehlmeyer 1999, 98-101; Fugmann 1997, 65; Papi 1995, 226; Forsythe 1994, 254-56; Flory 1993, 289; Bergemann 1990, 32-33, 157 (L11); Verzir 1980, 58-61; H61lscher1978, 332, 334-35; Vessberg 1941, 88. 6806Ti
46
B. ROLLER MATTHEW
As in Horatius' case, here too we can pass over these questions, despite their intrinsic archaeological and historical interest, as being irrelevant to this paper's purposes. The writers who inform us about this statue unanimously accepted that it was honorific, commemorating the deed done by Cloelia (though some would say Valeria) during the war with Porsenna in the first years of the Republic, and that this monumentplaced her deed before the eyes of posterity to rediscover, reevaluate in light of the original evaluation, deploy as a canon of value, and imitate in due course. It is these beliefs about the statue, and the social functions that follow from these beliefs, that I investigate here-not its actual origins. Furthermore,the statue reerectedby Augustus, if real, would have replacedthe original statue, whatever its actual date and original function, with a new one representingwhat the original statue was thought to be: that is, the Augustan statue would really have been an honorific equestrian statue commemoratingthe Cloelia of legend,104and it is this statue that imperial authorswould have seen and interpreted. Augustus might even have attached an identifying label and explanatory inscription, as he did for another set of honorific statues he erected-the summi viri of the Forum Augustum. Now, to regardthe statue as honorific is to raise a typological question: was it regardedas a "normal" honorific equestrianstatue of the type well attestedfrom the late fourthcenturyB.C.E.onward,and taken as ascribingoutstandingvirtusto the honorand? Or does the monument instead represent the uniqueness of her particular deed? Like the parallel question for Horatius' statue, we will see that our authorsconfront this question repeatedly.For them, interpretingthe statue's iconography was inseparable from knowing and evaluating her deed, and pertinent to comprehendingher "manliness." Several texts describing the statue insist that it was a "manly" honor. Servius is most explicit: he says that Porsennaadmiredher virtus and asked the Roman people to decree "somethingmanly" (aliquid virile), whereupon she was given an equestrian statue.105Plutarch reports that Porsenna gave her a horse as a gift; some people, he says, explained that Porsenna"admired her strengthand daring as superiorto that of a woman, and deemed her worthy of a gift befitting an adult male warrior.At any rate, an equestrianstatue of a woman stood on the Sacred Way. .. ." The implication is that the gifthorse honored her "manliness,"and that the equestrianstatue representsher upon that very horse.106The statue's "manliness"is stressed in earlier texts too. Livy says that the equestrianstatue was "a novel honor for novel virtus in a woman;" Seneca sees the horse in the statue as proof of Cloelia's allbut-male status; and Pliny remarksabout Cloelia that even women could receive an equestrian statue-implying that, in his view, equestrian statues were overwhelmingly associated with men.107 104. Astutely observed by Sehlmeyer 1999, 100-101. 105. Serv. in Aen. 8.646: qui admiratus virtutempuellae ... rogavit per litteras populum Romanumut ei aliquid virile decerneretur: cui data est statua equestris .... 106. Plut. Mor. 250F, quoted below (p. 49). 107. Livy 2.13.11: Romaninovam infemina virtutemnovo genere honoris, statua equestri, donavere. Sen. Consolatio ad Marciam 16.2: tantum non in viros transcripsimus: equestri insidens statuae ... Cloelia
IN ROMANCULTURE EXEMPLARITY
47
Leaving Cloelia and the gender conundrumsshe poses aside, these authors have good reason to gender equestrian statues as "masculine"ex hypothesi. Hundreds of such statues honoring men are attested in literary and epigraphic texts, by plastic remains, and on coins; yet there is no evidence for any other equestrianstatue honoring a woman, from any period, in Rome itself or anywhere in the Roman world. Already in Livy's day-the earliest author ascribing "manliness"to Cloelia's statue-many equestrian statues honoring male aristocratsstood in the city's public places, including a recent group nearthe Rostra.The occasions for their dedication, when known, are the honorands' success as military leaders or in discharging high magistracies, activities falling within the normal ambit of virtus for male aristocrats.108In the imperial period this gendered pattern was reinforced, as equestrian statues honoring male members of the imperial family and their close associates were erected throughoutthe city. They were also erected elsewhere in the empire, together with equestrian statues honoring local magistrates.109Imperialauthorsthereforehad even strongercause than Livy to associate this type with aristocraticmales and their characteristicpublic activities, hence to regardCloelia's statue as anomalous. The paradoxof the "manly maiden" that Cloelia's story poses is thus posed independently by the unique iconography of the statue.110 How did Romans of the late Republic and Empire understandthis anomalous monument in their midst? Evidently, it predisposed some viewers to comprehend her deed in terms of the charateristicallymale achievements, military or magisterial, which all other equestrian statues commemorated. On this view, hers was a "normal"equestrian statue such as men otherwise had, where the horse merely signified the actor's outstandingvirtus. So Livy interpretsthe statue when he says that it markeda novel honor for novel virtus (2.13.11: Romani novam infemina virtutemnovo genere honoris, statua exprobrat iuvenibus nostris [note equestri in emphatic position: this is the key evidence for her masculinity]. Plin. HN 34.28: et equestrium tamen origo perquam vetus est, cumfeminis etiam honore communicato Cloeliae statua equestri [similarly telling is etiam at Sen. Consolatio ad Marciam 16.2:... in qua etiam feminas equo donavimus]. 108. Republican equestrian statues in Rome itself are attested only in literary texts (Bergemann 1990, nos. L9-10, 12-25) and on coins, where certain images of horses and riders likely portrayequestrian statues (Bergemann 1990, nos. MI-21b); see id., pp. 14-20, on the attested locations of these statues, the status of the honorands, and the occasions for erection. Also Lahusen 1983, 56-61. 109. Imperial equestrian statues in the city of Rome: see Bergemann 1990, nos. P27, 51 (presumably); E1-5; L27-28, 30-32, 34-40. 110. If we survey other forms of honorific statuary, Cloelia's appears only slightly less exceptional. Flory (1993, 287-92) shows that prior to 35 B.C.E.only three other honorific statues of women are reported from Rome, for only one of which is any iconographical information known: Cornelia mater Gracchorum was seated (also Sehlmeyer 1999, p. 99, n. 325). Beyond Rome, fragments of apparentlyhonorific statues of women (standing) are known from the late Republic in Italian towns (Bergemann 1990, no. P1 and pls. 9-10), and Plin. (HN 34.31) says that Cato in his censorship objected to the erection of statues of Roman women in the provinces (Forsythe 1994, 256-57). From the Augustan age onward, many honorific statues of the spouses and female relations of emperors are attested in literature, and many portraits survive (see Flory 1993, 293-306, on statues of Livia and Octavia; Bartman 1999, passim, on Livia's portraits; and Lahusen 1984, 66-67, 70-71, 73-74, 76, 81-82, for the literary references). That a woman had an honorific statue at all, then, must have seemed stranger to Livy and Dionysius than to authors of the early Empire and later, who were more accustomed to seeing women so commemorated. But even then, Cloelia's association with an equestrian statue put her in exclusively male company (as far as we know).
48
MATTHEWB. ROLLER
equestri, donavere), implying that the horse signifies the (high) general level of her achievement. Pliny too interpretsit as a "normal"equestrian statue, for he introduces it as the earliest example of the type (HN 34.28: et equestrium tamen origo perquam vetus est, cumfeminis etiam honore communicata Cloeliae statua equestri).His declarationthatthe monumentis excessive, and that she deserved only a togate statue (... ceu parum esset toga eam cingi, cum Lucretiae ac Bruto ... non decernerentur),furthershows that he takes the horse (and the toga) as ascribing to her a certain, generic level of Finally, Seneca interpretsthe statue this way, and deploys achievement."'11 it in supportof a philosophical polemic. In section 16 of his Consolatio ad Marciam, quoted above, he develops the Stoic argument that women and men are equal in their capacity for virtue: specifically, they have the same vigor, facultas ad honesta, and tolerance for pain and toil (?1). To corroborate this assertion (?2), he invokes briefly Lucretia, then at greaterlength Cloelia: he declares that "we have all but enrolled her as a man" on account of her "outstandingboldness,"which he describes in such a generalized way that it could indeed apply equally to Horatius: "despising the enemy and the river" (contempto et hoste et flumine). Finally, to put the fully masculine quality of her performance beyond doubt, and thus prove the point aboutthe ethical equality of the sexes, he declares outrightthat "we have given equestrianstatues even to women" (sc. right along with the men whom such statues normally honor), and that her statue hurls a reproachat today's underachieving youth. That is, he takes the statue to indicate the fully masculine quality of her performance, not essentially different from the performanceof men so honored."12 In these authors'views, then, Cloelia's 111. For equestrian statues as the most prestigious honorific type, surpassing (standing) togate and loricate statues, see Bergemann 1990, 20. Bergemann (157) interprets Pliny's phrase ceu parum esset toga eam cingi as meaning that the monument represented Cloelia both mounted and togate, the toga (he suggests, 32) emphasizing her citizen status. But Pliny clearly contrasts toga (note its emphatic position in its clause) with statua equestri, a contrast that is only meaningful in the context of alternativetypes of honorific statue. Thus he means, "she has a statua equestris, as though it were not enough to have a (standing) statua togata," i.e., the lesser honor, which implies nothing about her costume on the actual equestrian statue. Gabelmann (1985, 517-25) and Goette (1990, 5-6, 80-83, 158-59, pl. 70) have shown that citizen girls of every status likely wore the toga praetexta in the late Republic and early Empire, at least on ceremonial occasions, just as boys did; only at marriagewould the (now) woman assume the stola. Thus Pliny may have understood this (imagined) togate statue of Cloelia in the same way he understoodtogate statues of adult males, with each figure wearing his or her normal citizen's costume. Because Cloelia is not an adult, the toga does not imply she is a prostitute:for togas marking transgressivesexuality in adult women, see McGinn 1998, 156-71, 208-11, with furtherbibliography. Regardless of Pliny's meaning, Bergemann may be right that the equestrian statue itself showed Cloelia togate. If indeed the statue was reerected under Augustus, it might well have renderedthis citizen virgo in the standardformal attire for such a person in this period, just as equestrian statues for men represent them in appropriatecostumes-whether togas, military garb, or heroic nudity. 112. See Wray (forthcoming) on Seneca's articulationof the Stoic doctrine of the ethical equality of the sexes, including this passage in particular.It is notable that Seneca's rhetoric in this passage is at odds with his overarching philosophical point. In constructing the gender inversion of the woman who sets a standard of masculine achievement to which today's effeminate youth cannot measure up, he eo ipso appeals to his readership's ingrained presumption of ethical inequality of the sexes-the very view he is formally contesting. Throughout his ethical prose Seneca makes similar, apparently self-contradictory appeals to the very "common conceptions" he is out to overturn;he does so precisely to heighten the effectiveness of his rhetorical persuasion among readers who, at least initially, accept those "common conceptions" (see Roller 2001, 75-77, 84-88).
EXEMPLARITY IN ROMANCULTURE
49
monumentis a "normal"equestrianstatue such as men otherwise have, signifying the same type and level of achievement.113 Other viewers, however, coped with the anomaly in a differentway. Consider the alternativeinterpretationsoffered by Plutarch (Mor. 250E-F): H yaoOCi06 1opaiva;
EK&sEUcEV I rHEOVdX6flvat KeKootrPl~C&vov EinpErn., KIal Tr Kkotkti
8•p0lodpacevo; •E6Psv&CpKalav0p6rowg KCai rdomg. rToiro notovrtat atrllGtiov oi &Trtns&lTev nokhoi zoi filvKkotkiavi'rrt~ Staekdioat rtotTsap6v-oi 6' oi05 aotv, d&kk& Tilv 1vby •iplyrlv zitv z6Laov KpitTTova yuvatKbc &v6ppi rokFe0autdoavza KTai • oped•g at6T-l; t&Ptuat YV yuv PltoTl rtpetnouorlg.av'cVECtTO oJG (tpt7tog yatKbg i~;i Ttg0 680o tiP; itEpdg eyoi•V elvat. itvg, fiy oi P.yv Z Kkotkhitg oi 6' tig Ouakepia;s XCyo•otv
In admiration,Porsennaordered a horse to be brought, fittingly adorned, and presenting it to Cloelia sent all the maidens away in a kindly, well-disposed manner. Most people take this as an indication that Cloelia crossed the river on horseback. Others deny this, saying that he marvelled at her strength and daring as being greater than a woman's, and judged her worthy of a gift befitting an adult male warrior.At any rate (yoiv), an equestrian statue of a woman stood on the so-called Sacred Way, which some say was of Cloelia and others say was of Valeria.
The particle youv here is crucial, in its "partproof" function: the equestrian statue authorizes the inference of "most people" that she crossed the river on horseback, and of "others"that she was given the horse as a gift. These viewers do not comprehendthe monumentalhorse as generically certifying that she did a certain sort of deed at a high level, but instead connect it with one or another version of the story of her deed-in fact, Plutarchsurmises (yoiv) that they retrojecttheir preferredversions of her story from the monument itself.114 Thus it is clear-more so for Cloelia than for Horatius-that her statue's iconography admittedboth generalizing and particularizinginterpretations, which in turn carried divergent ideological and ethical freight.115" The generalizing interpretation,regarding her monument as a typical equestrian 113. Some modernscholars also take this view. Sehlmeyer 1999, 101: "Das Pferd ... lies deutlich werden, daB die dargestellte Frau eine militiirischeLeistung vollbrachthatte, denn das Pferd symbolisierte ... einen konkreten Erfolg im Krieg." Flory 1993, 288: "Because there was no cultural tradition or public context for statues of women, when later Romans saw public statues of women or found records of them. .. they created stories to explain the statues in terms of the situation for men."
114.Otheraccountstoo sayCloeliacrossedon horseback(Val.Max.3.2.2;De vir.ill. 13)or wasgiven
a horse as a gift (Dion. Hal. 5.34.3; Polyaenus, Strat. 8.31.1; Dio Cass. frag. 14.4; both alternatives again at Plut. Publicola 19.7-8), though only Plutarchin the passage quoted expressly connects these versions to the horse in the monument. Some scholars take a similar view, either accepting that the deed involved the horse that is depicted in the statue (Bergemann 1990, 32-33, "Im Falle Cloelias hatte der Ablauf ihrer Tat, niimlich ihre Flucht zu Pferd, die Aufstellung der Reiterstatue motiviert.... Es war also ein Handlungsbildnis .. ."; cf. Fugmann 1997, 66), or that the versions of her story involving a horse are retrojectedfrom the statue (e.g., Forsythe 1994, 254; Gag6 1988, 238; Arcella 1985, 30). 115. Roman art often represents the general in and through images that may also appear to represent particularevents: see, for example, Fittschen's (1972) discussion of Trajan'sarch at Beneventum, arguing that its scenes articulatekey ideas and values associated with the emperor ratherthan simply showing particular events of recent years. But as the statues of Cloelia and Horatius show, Roman viewers themselves could move between generalizing and particularizinginterpretations,and they debated the ideological implications of their iconographical practices just as modern scholars do-or rather,modern scholarship replicates the interpretivedivide already found in ancient texts.
50
MATTHEWB. ROLLER
statue marking a generic level of (stereotypically masculine) military or magisterial achievement, makes Cloelia into a gender deviant-ethically a real vir-just as the "swimming" and "leadership"components of many of her narrativesdo. Like these aspects of her deed, this interpretationof the statue bestows a characteristicallymasculine form upon her virtus. She can thus be made to challenge the established ethical hierarchyof the sexes, as Seneca shows: she rivals, surpasses,even becomes one of the men, while they are effeminized and must struggle, in their degeneratecondition, to emulate a virgo's exemplary virtus. Conversely the particularizing interpretation, making the horse part of her story and the monument a unique instance rather than a general type, thereby regards its resemblance to a "normal" equestrian statue as fortuitous, and so eliminates the gender deviancy. Like the "deceiving the guards" and "saving the impubes"aspects of her deed, the monument on this interpretationcredits her not with a manly form of virtus, but with some other form more fitted to the particularitiesof her age, status, and circumstances. At any rate, the conundrumof gender and ethics posed by the figure of the "manlymaiden"is articulatedand worked out in similar ways both in the interpretationof her actions (as described in the narratives)and in the interpretationof her statue's iconography. IV. CONCLUSIONS In part, this paper has argued that it is useful to assemble several cultural phenomena that are well attested throughRoman texts and images-consequential action in the public eye; the evaluation of such action in ethical terms; actions and their evaluations being commemorated in monumental form; encounterswith monumentsspurringemulation-and to identify them as collectively constituting a discourse. Such an identification,with its fourpart schematic structure,is useful because it stands to provide us moderns with a heuristic device by which we can perhaps better understandcertain aspects of Roman historical consciousness. In and throughthis discourse, I suggest, Romans of the late Republic and Empire encountered their past, gave it value and meaning, and deployed it in the service of the present. Through it they also gave value and meaning to contemporaryactions, in the expectation that these actions would have repercussions in the future just as past actions were having repercussions in the present. Of course, "discourse"and "historicalconsciousness" are modern concepts, and when imposed upon ancient ways of thinking necessarily do a degree of violence to them. No ancient text, to my knowledge, puts the four elements together to construct a discursive loop as I have, even though the individual elements and the pairwise links between them are abundantly attested in texts and images (as we have seen). It is hard to know, then, whether a Roman would recognize her or his culture in the analysis presented here. Nevertheless, to assemble these elements in this way helps us address questions that we find pressing (such as, "Whatdid the Romans make of their own past?"), even if these questions are not the Romans' own. Being unable to inhabit their
EXEMPLARITY IN ROMANCULTURE
51
culture ourselves, we can still come to understandthem better, at least in our own terms. What, then, did the Romans make of their past? I have sought to show, by analyzing Horatius and Cloelia as protagonists in a discourse of exemplarity, that the past had a thriving, evolving, ideologically efficacious life in any given present. This contention contains a hidden polemic, as it is sometimes claimed that exemplary figures from the republicanera became "fossilized" underthe Empire.Throughhabitualdeploymentin particularmodes, their meanings fixed by compilers like Valerius Maximus, particular figures came to be so closely associatedwith particularvalues as to become mere metonyms or personifications-hence unattainable,incontestable,eminently "dead"ideals (so the argumentgoes).116Certainly,Horatiusand Cloelia are closely associated with values like fortitudo and virtus. Yet the association is complex: a constellation of specific actions and monuments constitutes each exemplum. For Horatius there is his defense of the bridge, his swimming, his armor, his wound, his statue, his possible earlier deed, and his subsequent attemptto convert his valor into public office. Any given invocation of the hero as a canon of value may engage a differentone (or more) of these aspects, leading to divergent and sometimes contested evaluations. These contestations, these instabilities in the productionof exemplary discourse, are precisely what make exempla so good for Romans to think with. Consider the disagreement Dio Cassius stages between Cicero and Fufius Calenus regarding how Antony and Horatius measure up as defenders of libertas; or the debate (of which Valerius Maximus and Velleius Paterculus give us one side) regarding whether Laetorius imitated Horatius properly; or how Livy's Samnite general invokes Cloelia as an example of Roman perfidy; or the many texts that propose divergent ways of understanding how a virgo can display virtus, and what such a display means.'"17 All of these texts, of course, were composed by literary artists. Yet I have little doubt that, in and through debates like these, actual Romans living in any given present addressedcontemporaryissues of fundamentalsociopolitical importance, such as articulating gender roles and the shifts they undergo, defining the collective interest and how one properly serves it, considering how the claims of the collective and individual should be balanced, and so on. Exemplarydiscourse has an importantrole in such debates precisely because of the instabilities and contestations that entangle its production. These are what make the discourse flexible, useful, and vital in the social debates of any given historical moment. 116. This view goes back at least to Litchfield 1914; see his table (28-35) correlatingheroes with virtues. Hilkeskamp (1996, 314-20, 323-26) rightly insists on a more complex correlationbetween exemplary figures and abstract values. But in suggesting that exemplary figures are ethically uncontested, and that they set "bindingrules" (317), he seems to renderthem monolithic-dominating any given present-rather than serving as flexible ethical touchstones by which that present can interrogateitself, as I see them doing. 117. For more on exempla whose meanings are contested in various contexts, see the salutary discussion of Chaplin 2000, 31-49 and 73-105 (on Livy).
52
MATTHEWB. ROLLER
Finally, a reflection on the project of examining exemplarity in general through the investigation of two specific exemplary figures. Scholars who study examples, whether in classical or other contexts, sometimes distinguish two functionalmodes, illustrativeand injunctive.An example deployed illustratively is, or purportsto be, an utterly typical instance of a series of similar objects, a "one among many."Conversely, an example deployed injunctively is singled out as distinctive, as crucially unlike other objects, especially in its ethical import (that is, it is uniquely good or bad), and to single it out amounts to demanding that other objects should be like or unlike this one.118This distinction is useful, though in practice the two modes often intermingle. Thus, an example purportingto be illustrative may subsume and include a normativeelement that in fact amountsto an injunction. Conversely, an example used injunctively is presented as distinctive, but aspires to become illustrative: it aims to generate a new series of objects like itself, therebyreducing itself to a "one among many."This paperhas largely been concerned with the injunctive mode, examining how Horatius and Cloelia are invoked as standardsand models for the actions of others. "Exemplary discourse,"as defined here, is mostly concerned with the injunctive mode. However, several passages discussed above in fact deploy these figures in an illustrative way: Seneca adduces Horatius and Fabricius as typical, illustrative instances of the class of people who are admiredfor just one or two brilliant deeds (Ep. 120.6-7); and Cicero lists Horatius, the Decii, the Scipios, and Marcellus as typical instances within a much larger class of figures who are outstanding for their magnitudo animi (Off. 1.61). Yet if my analytical focus has been on the injunctive deployment of Horatius and Cloelia, my own rhetoric in this paper is otherwise. I adduce these figures illustratively, as being typical instances of injunctive exempla-that is, as representativesof a larger class of mythistoricalfigures that Romans liked to deploy injunctively, in the context of exemplary discourse. Hence, I imply that the analytic approachdeveloped here could equally be applied, and with similar results, to other figures whom the Romans used injunctively: Fabricius, the Decii, Regulus, Fabius Cunctator,Cato the Elder, and so on. Except, of course, that I chose to examine Horatius and Cloelia because they seemed exceptionallygood "typicalinstances."Being so heavily attested, they illustrate a particularlywide and rich range of exemplary phenomena, and pose striking problems unique to themselves (to judge from surviving representations). So how typical of this class are they, really? Thus, my own examples, like so many others, entail a mixing of the illustrative and injunctive modes. For I present them, in part, as models for how other, less richly attested exemplary figures might-ought to-have functioned socially and ethically, and might be seen to function, had more monuments survived. In working with exemplarity,this kind of self-reflection is 118. The terms "injunctive"and "illustrative"were suggested to me by Noel Carroll.For more on these two modes (though with different terminology) in a Roman context, see Chaplin 2000, 137-40; and Stemmler 2000, 157-58.
EXEMPLARITY IN ROMAN CULTURE
53
difficultto avoid:willy-nilly,one is investigatingthe foundationsof one's own argumentation, andof mentalideationitself.119 Johns Hopkins University of suchself-reflection,andan over119.See Goldhill1994 for an engagingdiscussion/performance on examples. view of thephilosophicalscholarship
LITERATURE CITED Africa, T. W. 1970. The One-Eyed Man against Rome. Historia 19:528-38. Alf61ldi,A. 1965. Early Rome and the Latins. Ann Arbor, Mich. Alston, R. 1998. Arms and the Man: Soldiers, Masculinity and Power in Republicanand Imperial Rome. In WhenMen WereMen: Masculinity, Power and Identityin Classical Antiquity, ed. L. Foxhall and J. Salmon, 205-23. London. Arcella, L. 1985. Il mito di Cloelia e i Valerii. SMSR9:21-42. Banti, A. 1983. I grandi bronzi imperiali: Sesterzi e medaglioni classificati secondo il sistema Cohen. 4 vols. Florence. Bartman,E. 1999. Portraitsof Livia:Imagingthe ImperialWomanin AugustanRome.Cambridge. Barton, C. 2001. Roman Honor: The Fire in the Bones. Berkeley, Calif. Bartsch, S. 1994. Actors in the Audience: Theatricality and Doublespeakfrom Nero to Hadrian. Cambridge,Mass. Beck, H., and U. Walter,eds. 2001. Die friihen r6imischenHistoriker.Vol. 1, VonFabius Pictor bis Cn. Gellius. Darmstadt. Bell, A. J. E. 1999. The PopularPoetics and Politics of the Aeneid. TAPA129:263-79. im iffentlichen Bereich. Mainz. Bergemann,J. 1990. RcimischeReiterstatuen:Ehrendenkmailer Bernard, J.-E. 2000. Le portrait chez Tite-Live: Essai sur l'dcriture de l'histoire Romaine. Brussels. Brilliant, R. 1963. Gesture and Rank in Roman Art. New Haven, Conn. Broughton, T. R. S. 1951. The Magistrates of the Roman Republic. 2 vols. New York. Camassa, G. 1984. Sull'origine e le funzioni del culto di Volcanus a Roma. RSI 96:811-54. Capdeville, G. 1972. Le centurion borgne et le soldat manchot. MEFRA84:601-21. Chaplin, J. D. 2000. Livy's Exemplary History. Oxford. Claridge, A. 1993. Hadrian's Column of Trajan.JRA 6:5-22. Coarelli, E 1983. Il Foro Romano. 2 vols. Rome. . 1988. II Foro Boario: Dalle origini alla fine della repubblica. Rome. . 1999. Pons Sublicius. In Lexicon TopographicumUrbis Romae, ed. E. M. Steinby, 4:112-13. Rome. Cornell, T. J. 1995. The Beginnings of Rome: Italy and Romefrom the Bronze Age to the Punic Wars, c. 1000-264 BC.London. Crawford,M. 1974. Roman Republican Coinage. 2 vols. Cambridge. Davidson, J. 1991. The Gaze in Polybius' Histories. JRS 81:10-24. Delcourt, M. 1957. Horatius Cocles et Mucius Scaevola. In Hommages ai WaldemarDeonna, 169-80. Brussels. de Libero, L. 2002. Mit eiserner Hand ins Amt?-Kriegsversehrte r6mische Aristokratenzwischen Recht und Religion, Ausgrenzungund Integration.In Res publica reperta:Zur Verfassung und Gesellschaft der r6mischen Republikund desfriihen Prinzipats, ed. J. Spielvogel, 172-91. Stuttgart.
54
B. ROLLER MATTHEW
Dum6zil, G. 1988. Mitra-Varuna:An Essay on TwoIndo-EuropeanRepresentationsof Sovereignty. Trans.D. Coltman. New York.Originally published as Mitra-Varuna:Essai sur deux reprdsentationsindo-europdennesde la souverainetd, Paris, 1940. Eisenhut, W. 1973. Virtus Romana:Ihre Stellung im r6mischen Wertsystem.Munich. Feldherr,A. 1998. Spectacle and Society in Livy's History. Berkeley, Calif. Fittschen, K. 1972. Das Bildprogrammdes Trajansbogenszu Benevent. AA 4:742-88. Flory, M. 1993. Livia and the History of Public Honorific Statues for Women in Rome. TAPA 123:287-308. Flower, H. 1996. Ancestor Masks and Aristocratic Power in Roman Culture. Oxford. Forsythe, G. 1994. TheHistorian L. CalpurniusPiso Frugi and the Roman Annalistic Tradition. Lanham, Md. Frier, B. W. 1979. Libri annales PontificumMaximorum:The Origins of the Annalistic Tradition. Rome. Froehner,W. 1878. Les midaillons de l 'Empireromaindepuis le regne d Augustejusqu'a'Priscus Attale. Paris. Fugmann,J. 1997. Kinigszeit undfriihe Republikin der Schrift "De viris illustribus urbis Romae." Vol. 2.1. Frankfurtam Main. Gabelmann, H. 1985. Romische Kinder in toga praetexta. JDAI 100:497-541. Gage, J. 1963. Matronalia:Essai sur les ddvotions et les organisations cultuelles desfemmes dans 'ancienne Rome. Brussels. . 1969. Infirmes, bless6es de guerre et marques corporelles dans l'ancienne Rome. REL 47:184-208. . 1973. Une consultation d'haruspices: Sur les tabous 6trusques de la statue dite d'Horatius Cocles. Latomus 32:3-22. 1988. Les otages de Porsenna. In Res sacrae:Hommages a Henri le Bonniec, 23645. Brussels. Garland, R. 1995. The Eye of the Beholder: Deformity and Disability in the Graeco-Roman World.London. Gleason, M. 1995. Making Men: Sophists and Self-Presentation in Ancient Rome. Princeton, N.J. Gnecchi, E 1912. I medaglioni romani. 3 vols. Milan. Goette, H. R. 1990. Studien zu rdmischen Togadarstellungen.Mainz. Goldhill, S. 1994. The Failure of Exemplarity.In Modern Critical Theory and Classical Literature, ed. I. J. E De Jong and J. P. Sullivan, 51-73. Leiden. Goldsworthy,A. 1996. The Roman Army at War, 100 B.C.-A.D.200. Oxford. Gurval,R. A. 1995. Actiumand Augustus: ThePolitics and Emotions of Civil War.Ann Arbor, Mich. Habinek, T. 1998. The Politics of Latin Literature. Princeton, N.J. Hafner, G. 1969. Bildnisse des 5. Jhs. v. Chr. aus Rom und Etrurien.MDAI(R)76:14-50. Harris, W. V. 1979. Warand Imperialism in Republican Rome, 327-70 B.C. Oxford. Hill Richardson,E. 1953. The EtruscanOrigins of Early Roman Sculpture.MAAR21:77-124. H61keskamp,K.-J. 1996. Exempla und mos maiorum:Oberlegungenzum kollektiven Gedichtnis der Nobilitat. In Vergangenheitund Lebenswelt: Soziale Kommunikation,Traditionsbildung und historisches Bewusstsein, ed. H.-J. Gehrke and A. M6ller, 301-38. Tiibingen. . 2001. Capitol, Comitium und Forum: Offentliche Riume, sakrale Topographie und Erinnerungslandschaftender r6mischen Republik. In Studienzu antikenIdentitaiten,ed. S. Faller, 97-132. Wiirzburg. H61scher,T. 1978. Die Anflinge r0mischer Reprisentationskunst.MDAI(R) 85:315-57. Horsfall, N. 1996. The CulturalHorizons of the Plebs Romana. MAAR41:101-19. Inwood, B. 1995. Politics and Paradox in Seneca's De Beneficiis. In Justice and Generosity, ed. A. Laks and M. Schofield, 241-65. Cambridge. Jaeger, M. 1997. Livy's WrittenRome. Ann Arbor, Mich.
EXEMPLARITY IN ROMAN CULTURE
55
Joshel, S. R. 1992. The Body Female and the Body Politic: Livy's Lucretia and Verginia. In Pornography and Representation in Greece and Rome, ed. A. Richlin, 112-30. Oxford. Knapp, S. 1989. Collective Memory and the Actual Past. Representations 26:123-49. Konrad, C. 1994. Plutarch's "Sertorius": A Historical Commentary.Chapel Hill, N.C. Konstan, D. 1986. Narrative and Ideology in Livy: Book I. ClAnt 5:198-215. Krumme, M. 1995. Romische Sagen in der antiken Miinzpriigung.Marburg. Lahusen, G. 1983. Untersuchungenzur Ehrenstatue in Rom: Literarische und epigraphische Zeugnisse. Rome. . 1984. Schriftquellenzum rdmischenBildnis L Bremen. Leach, E. W. 1994. Horace Carmen 1.8: Achilles, the Campus Martius, and the Articulationof Gender Roles in Augustan Rome. CP 89:334-43. Le Gall, J. 1953a. Recherches sur le culte du Tibre. Paris. . 1953b. Le Tibre: Fleuve de Rome dans l'antiquite'.Paris. Leigh, M. 1995. Wounding and PopularRhetoric at Rome. BICS 40:195-212. . 1997. Lucan: Spectacle and Engagement. Oxford. Lendon, J. E. 1997. Empire of Honour: The Art of Governmentin the Roman World.Oxford. . 1999. The Rhetoric of Combat:Greek Theory and Roman Culturein Julius Caesar's Battle Descriptions. CIAnt 18:273-329. Lincoln, B. 1991. Death, War,and Sacrifice: Studies in Ideology and Practice. Chicago. Litchfield, H. W. 1914. National Exempla Virtutisin Roman Literature.HSCP 25:1-71. MacMullen, R. 1966. Enemies of the Roman Order. Cambridge,Mass. McDonnell, M. 2003. Roman Men and Greek Virtue. In Andreia:Manliness and Courage in Classical Antiquity, ed. R. Rosen and I. Sluiter, 235-61. Leiden. McGinn, T. A. J. 1998. Prostitution, Sexuality, and the Law in Ancient Rome. Oxford. McInerney, J. 2003. Plutarch'sManly Women. In Andreia:Manliness and Courage in Classical Antiquity, ed. R. Rosen and I. Sluiter, 319-44. Leiden. Mehl, E. 1931. Schwimmen. RE Suppl. 5:847-64. Miles, G. B. 1995. Livy: ReconstructingEarly Rome. Ithaca, N.Y. Moeller, W. 0. 1975. Once More the One-Eyed Man against Rome. Historia 24:402-10. Mommsen, T. 1887. R6misches Staatsrecht3.3 vols. Leipzig. Moore, T. 1989. Artistry and Ideology: Livy's Vocabularyof Virtue.Frankfurtam Main. Miinzer, E 1901. Cloelius. RE 4.1:108-11. . 1913. Horatius. RE 8.2:2328-36, 2400-2404. Nisbet, R. G. M., and M. Hubbard.1978. A Commentaryon Horace: "Odes,"Book II. Oxford. Oakley, S. P. 1985. Single Combat in the Roman Republic. CQ 35:392-410. Ogilvie, R. M. 1965. A Commentaryon Livy, Books 1-5. Oxford. Pais, E. 1905. Ancient Legends of Roman History. Trans. M. Cosenza. New York. Papi, E. 1995. Equus: Cloelia. In Lexicon TopographicumUrbis Romae, ed. E. M. Steinby, 2:226. Rome. Peter, H. 1865. Die Quellen Plutarchs in den Biographieen der Romer. Amsterdam. Reis, M. 1994. Sport bei Horaz. Hildesheim. Roller, M. 2001. ConstructingAutocracy: Aristocrats and Emperors in Julio-Claudian Rome. Princeton, N.J. Sehlmeyer, M. 1999. Stadtrimische Ehrenstatuender republikanischenZeit. Stuttgart. Shackleton Bailey, D. R. 1956. Propertiana. Cambridge. Stemmer, K. 1978. Untersuchungenzur Typologie,Chronologie und Ikonographieder Panzerstatuen. Berlin. Stemmler, M. 2000. Auctoritas Exempli: Zur Wechselwirkung von kanonisierten Vergangenheitsbildernund gesellschaftlicher Gegenwartin der spAtrepublikanischen Rhetorik.In Mos Maiorum: Untersuchungenzu den Formen der Identittitsstiftungund Stabilisierung in der rdmischenRepublik, ed. B. Linke and M. Stemmler, 141-205. Stuttgart. Tanner,J. 2000. Portraits,Power, and Patronagein the Late Roman Republic. JRS 90:18-50.
56
MATTHEWB. ROLLER
Toynbee, J. M. C. 1925. Some "Programme"Coin-Types of Antoninus Pius. CR 39:170-73. . 1944. Roman Medallions. New York. Verza.r,M. 1980. Pyrgi e l'Afrodite di Cipro. MEFR(A) 92:35-86. Vessberg, 0. 1941. Studien zur Kunstgeschichte der romischen Republik. 2 vols. Lund. Voisin, J.-L. 1992. Deux archetypes de la mort volontaire: Lucrece et Horatius Coclks? In La Rome des premiers siecles: Legende et histoire, 257-66. Florence. Wheeler, E. L. 1988. Stratagem and the Vocabularyof Military Trickery.Leiden. Wiseman, T. P. 1998. Valerius Antias and the Palimpsest of History. In RomanDrama and Roman History, 75-89. Exeter, England. . 1999. The Games of Flora. In The Art of Ancient Spectacle, ed. B. Bergmann and C. Kondoleon, 195-203. Washington, D.C. Wissowa, G. 1912. Religion und Kultus der Rimer2. Munich. Wray,D. Forthcoming.Manly Matronsand Stoic Virtue in Seneca and Valerius Maximus. AJP. Wrede, H. 1983. Statua Lupercorumhabitu. MDAI(R)90:185-200.
AN ORACLE OF APOLLO AT DAPHNE AND THE GREAT PERSECUTION ELIZABETH DEPALMA DIGESER CCORDING TOTHEEMPEROR Constantine(306-37 C.E.),the immediate cause for Rome's "GreatPersecution"was a Pythian oracle's complaint that Christianswere preventing accurateprophecies. Constantine's claim, articulatedin a 324 edict "To the Eastern Provincials" (apud Euseb. Vit. Const. 2.48-60), has usually been conflatedwith an episode from Lactantius'De mortibuspersecutorum (10.1-6). According to the latter account, early in 303 the emperorDiocletian (284-305 C.E.)was pressuredby his junior colleague Galerius to launch a general persecution.To resolve the issue, Diocletian sent a haruspex,or soothsayer,from his palace in Nicomedia to consult with Apollo at Didyma. Lactantiussays that the oracle answered as an "enemy"of God, and as a result Diocletian issued the edicts that increasingly targetedChristiansin the general population (De mort.pers. 1115). Despite the long history of linking these two accounts, however, clues in the emperor'sletter and several other fourth-centurytexts-both Christian and pagan-suggest that Constantinewas describing a separate and earlier prophecy from an oracle of Apollo at Daphne near Antioch in 299. This prophecy,in turn, triggereda purge of Christiansoldiers from the army,1an event that both Lactantiusand Eusebius of Caesareamarkas the true beginning of Diocletian's persecution.2 Although it seems a minor adjustmentto the historical record, locating Constantine's oracle at Daphne in 299 has significant implications for the history of the early fourth century. For example, pagan and Christiantexts from this period, from Porphyry's Philosophy from Oracles to Lactantius' 1. Although A. H. M. Jones dated the army purge to 297 (PLRE 1.955), 299 'is now more generally accepted. See T. D. Barnes, Constantine and Eusebius (Cambridge, Mass., 1981), 18-19; P. Davies, "The Origin and Purpose of the Persecution of A.D. 303," JThS 40 (1989): 66-94 at 91-93; and R. W. Burgess, "The Date of the Persecution of Christiansin the Army,"JThS 48 (1997): 471-504. 2. Euseb. Hist. eccl. 8.1.7: 9K TCv9v orpateiats 6hekXpiv ToO&Uoyloi (Greektext of Euseb. Hist. eccl. throughoutis that of E. Schwartz's GCS editionKac•pXojpLvou as reprintedin K. Lake's 1959 Loeb trans.). In Lactantius' case, he says that Diocletian's failure to read the auspices at the palace (De mort. pers. 10.1-4) led to his decision to purge Christians from the army and punish those in the court. Subsequently (10.6), he says that, in doing so, Diocletian had begun to persecute ([Galerius]... ad persequendos Christianos instigaret [Diocletianum], qui iam principium fecerat). Moreover, in Div. inst. 4.27.4-5, Lactantius says that these aborted auspices drove the emperors to "storm the temple of God" (expugnarent dei templum), an expression he uses throughouthis works to refer to the persecution (e.g., De mort. pers. 33.11). Latin text of Lactant. De mort. pers. throughout is from the edition of J. L. Creed (Oxford, 1984); that of Lactant. Div. inst. is from S. Brandt's edition, CSEL 19 (Vienna, 1890). Translationsthroughoutare my own, except for those from Lactant. De mort. pers., for which I use Creed's translation.
Classical Philology 99 (2004): 57-77 [? 2004 by The University of Chicago. All rights reserved] 0009-837X/04/9901-0002$10.00
57
58
ELIZABETHDEPALMADIGESER
Divine Institutes and Eusebius' Preparation for the Gospels, demonstrate an extraordinarypreoccupationwith oracles that has never been addressed, much less explained. Porphyry's concern for oracles is more easily understood, however, if the emperorshad begun to fear that Christianswere disruptingthe sources of divination upon which the regime relied for guidance. In this view, the Daphnic Apollo's complaint would have provided the stimulus and context for the pagan literary effort to defend traditionaltheology and returnChristiansto the traditionalcults that W. H. C. Frendonce called a "PropagandaWar"3-a campaign in which Porphyrytook part.4Likewise the reason for the Christian authors' fixation with oracles becomes clear if Apollo's prophecies had not merely triggered the general persecution of 303, but had also initiated the series of steps leading toward that end, from the purge of the army throughthe "propagandawar."'The account of Constantine's oracle and the other sources that corroborateit also help to answer perennial questions about the persecution: Why did Diocletian, who had appointedChristiansto key positions in his court, turn against them toward the end of his reign? What role did Galerius, Diocletian's Caesar,play in bringing persecution to the general population?And to what extent were long-standing anti-Christianfactions-especially in Antioch and Didymainstrumentalin fanning the flames of persecution? The link between these anti-Christianfactions and the oracles raises an importantissue regardingthe functioning of these sites in general and their role in the persecution in particular.Whether they employed the enthused priestesses of Delphi and Didyma or an attendant(or theurgist) whose rites (TEcGtrucl)gave voice to the prophetic statue at Daphne,6those who consulted and staffed these oracles considered them to be the means by which a god communicated to the world.' Some Christians, during and after the Great Persecution, accused not only the oracularpersonnel at Didyma and Antioch but also the Caesar Galerius of having deliberately fabricatedantiChristianprophecies.8The oracle workersconfessed undertortureto having done so (Euseb. Praep. evang. 4.2; Hist. eccl. 9.11.6), but this does not mean that the Christianclaim should be accepted at face value. For even the sceptical E. R. Dodds acknowledged that "to ascribe such manipulationsin general to conscious and cynical fraudis ... to oversimplifythe picture."9Priests 3. W. H. C. Frend, "Preludeto the Great Persecution: The PropagandaWar,"JEH 38 (1987): 1-18. 4. For the date of Porphyry's text, see R. Wilken, The Christians as the Romans Saw Them (New Haven, Conn., 1984), 65-69; M. B. Simmons, Arnobius of Sicca (Oxford, 1994); E. Digeser, The Making of a Christian Empire: Lactantius and Rome (Ithaca, N.Y., 2000), 1-17, 91-114; and T. D. Barnes, "Monotheists All?" Phoenix 55 (2001): 142-62, at 159. 5. For the date of Lactant. Div. inst., see n. 71 below. Eusebius' concern with the Great Persecution is evident throughoutthe Praep. evang. (e.g., at 1.2) although he must have written some of it after the events of 313 that he describes (see Praep. evang. 4.2). 6. Although Proclus would call such wonder-workers"theurgists"and their rites such termiand 14-20 [Kroll], In nology may not have been current in the late third century (Procl. In R. 1.39.9-10 z•Xoaztrci, Ti. 1.51.25-27 [Diehl]); see C. van Liefferinge, La thdurgie des "Oracles chalda'ques" a Proclus (Liege, 1999), 12-13, 87-97, 268-74. For the priestesses at Didyma and Delphi, see n. 39 below. 7. For just two examples, see Plutarch's De def or. and his De Pyth. or. 8. Euseb. Praep. evang. 4.2; Lactant. Div. inst. 7.17.5. 9. E. R. Dodds, The Greeks and the Irrational (Berkeley and Los Angeles, 1951), 74.
THE GREAT PERSECUTION
59
and priestesses with particularprejudices, say against Christians,were probably predisposedto interpretambiguouscues in a certainway whenever they believed that they were communicating with and for the god.10 That people with known prejudices could be later accused of fraudby Christianconspiracy theoristsdoes not necessarily mean thatthe oracle's staff acted cynically, but that they simply saw and heard what they were primed to perceive. In the case of the hapless functionaries of Didyma and Antioch, their prophecies were prosecuted as "frauds"in 313, when Licinius, imperial ally of the newly ChristianemperorConstantine,defeated Maximin Daia, the last persecuting emperor(Euseb. Praep. evang. 4.2; Hist. eccl. 9.11.6). Clearly what had seemed genuine to Diocletian and Galeriusappeareddifferentlyto many Christians and to Licinius' agents. More than ten years after his defeat of Maximin Daia, Licinius himself fell to Constantine,who then gained sole controlof the RomanEmpire.In the immediate aftermathof this 324 victory, Constantineissued several edicts.11 One of these, his letter "To the Eastern Provincials" &vaTo(Entapxtobait ktKolt), describes a Pythianoracle's role in the GreatPersecution.The edict, preserved in Eusebius' Life of Constantine, begins by asserting that those emperors who persecuted Christians all came to a bad end (2.48); it then tells the story of the oracle. According to Constantine, Diocletian and two of his co-rulers "rekindled"persecution,12or "civil wars" (•pl(pu•iou? rtokhiwhen "all human and divine matters"(ntdvzov 6Voi zUTOv EOLov zTKai lpoug), were "at "Under (2.49).13 peace" avpoprEnivo7Tpaylp6acov) (siprlveuo~pivov) these circumstances" (t6b'rlvticaira), he claimed, "they" (presumably the 10. In accounting for the role that "humanintelligence" played in the oracular"process,"Dodds, Greeks and the Irrational (n. 9 above), argued that "an amazing amount of virtual cheating [could] be done in perfectly good faith by convinced believers." 11. For the date, see Eusebius "Lifeof Constantine,"trans. A. Cameron and S. G. Hall (Oxford, 1999), 233-44. 12. Constantine excluded his father from this list (apud Euseb. Vit. Const. 2.48). 13. 2.49: scot ~ Xotnroiol6 Kai bytaivovTres T; (ppEvagdtypt6TrlTo; pd•hov ~ t6pa6trlro; ~rs~X7kovTo, 7 TCov Ta6utrl E-'peov kni ihiav Katp~ov bTv d&XiOifX6yov 8staapE'povTsg, trig novrlpiag abtoi i1 &•p06v(o, 8etv6trlgEi; Tooo3rTovk9p'rTsTo, ikRdvrov 6tAoi riT6v 6EioWte Kai dv0pirtvov xpay7gdrav siprlvEuooPjv•0 50 Tbv Ant6kXo oin' KEivOvitok0iLoud&vapptrxiso0at. Epaoav g4vrpou rtvb6 Kai TXb TrqvtvKarTa •cpuiou4o ' i~ dv0ppcinou Xpiloat, d;pa oi tRiTi? ;tiKatot •gn68tov EIEv o dhXtk0e86etv oloTou puXo6 ob06i Yfl T&vrptt68OWv rdTSavEita abr6v, Kai 8th roi3to ydp rot 7l ispria abTro, KaTTrp ti; ro~toytOu6SEri rtotsiGOatL. d&X' Tb 9v Av0pdnot; KaK6v Tob rsko7KdPoug &vsioa nb6pavia; T' 9XauvopVVrl, TraMTra h•r•066pero. •&•Oev rTOTKoCtLis 7tai; i•t b7dpx(ov, Eig 6noiov Triko;i6KEtXh. 51 Xi viv tbvu1j6yatoTov0sv Kak"I flKpo06lAlv 6 6 ar' KEIvoKatpoG p Toi4 "PioLaiwv ,t;g CXvzvTrnp(OTia, 8Eikatog,6X&rl0Xj abToKpdaTopotV 6eikatog, rtapo hpa Eldv oi irpb; tfi yj~ iKatot, rokuabr6v, Irckvn T•ivyuXi~v •a7trlaT1avo;, ncapat6v 0opucpopo6vTw•v T•ISN 6 npaygoviiv inruvOdvsEo,KaituT;T@vrtspia6tbv 0urt6rbXov 7ir 6&trOKptOsi;,XptorTtavoi 86i•ou0Ev,I(rp. 8 t•iv tiv •tKrlldTWrov tr KaTrid KptotViortsp tt opox0ioa;piXkt Eps0~v9ta KaTd tig 6venthiFIrntoo 6ot6trltog i•pr X TS ;6) 6i 6tWKaOTai oUv ETStvev. abiTiKca 8taTdtytlaTa60pov ptatcp6votR 8irsiv •KsKatig;ouv&aUVtTT, Toi5 KatvoTpskOVKTvstv .... 54 6i K5v TilvKaT6 p60otv6yXivotav cig EijpEotv KoXaoTrlpio•v stapKE•s•6sto o• i e TvTo i~ E KEivql phl abToigou•3E•piKFet, dosfE Huioou pavTsia KiP8•rlov86valptv 9o~rKct. • To XprljZlpio)v Greek text of Constantine's edict is from I. A. Heikel's edition (Leipzig, 1902) of Eusebius' Vita Constantini. Where Heikel's edition says that the prophecy came "not out of a human being" (ob6X8' d&v0paiou, 2.50), that of E Winkelmann (Berlin, 1975) has "not out of heaven" (ob6i 6' E o6pavoi). Only one exemplar of the nine key copies that Winkelmannconsulted-Cod. Vatic.gr. 149-gives this reading (as o U). Although Heikel consulted this text (p. ix), he did not reportthis variant. My sense is that Winkelmann (together with the Vatican scribe), unfamiliar with Lucian's De dea Syria 36 (see p. 65 below), was perplexed about what prophesying "not out of a human being" could mean and so chose the alternatereading. I have chosen Heikel's edition since it seems to follow more closely the weight of the manuscripttraditionhere.
60
ELIZABETHDEPALMADIGESER
aforementionedemperors) "were saying that"Apollo spoke from "a certain but inner chamber and dark alcove" (' ivz'pouztvbq Kai tuXou), oicoToiou 6' said that he a human was out of "not 4 being" (o5'i Apollo &vOpcotnou). because, as Con"composing false oracles from the tripods"(zTv zptn6&vm), were an imstantine put it, "the 'righteous' on earth (oit ni Tzg 6iKatot) y•lg pediment to his" telling the truth.14"This,"Constantineclaimed, by which he seems to be indicating the so-called righteous people, was what Apollo's priesthood (fl ipsia) :"were lamenting as the evil within human society" (T6 Av&vpo~rtotKCKbyV pcEo) (2.50).15 Accordingto the emperor,this in&tx0O cident"droveheadlong"(5(6KetX,) towardthe use of force againstChris-
tians.Suggestingthathe was an eyewitness,or at the veryleasta court
insider,Constantineclaims thathe "heardthenhow" (?IKpocCyrlv z6TEz... natq)Diocletian"keptclosely questioninghis bodyguards"(6opu(poporvzov), wantingto knowwho the so-calledrighteouswere.Finally,one of his
a haruspex-saidthattheywere,of course, diviners(Ounrpt6Xov)-probably
the Christians(XptoatavoitiTnooucv). Accordingto Constantine,"on the very spot" (azUiKa), Diocletian "was drawingup edicts of blood and venom" (6tak60pp.v... ouvETazzE)(2.51). Finally, after briefly describing the Tz•ypara sad end to which each of the persecutingemperorscame, Constantineclaims, "this would not have happened to them, if that impious prophetic power of the Pythian oracles (Tzovto~3 IeOioo XPIrlarlpiopv) (CEPic7.i... avrzia) had not achieved a spurious force" (2.54).16 Comparedto Constantine's description of Diocletian and the oracle, the account with which it is usually identified is significantly less detailed. Like Constantine,Lactantiusalso writes as an insider:for he was a court-appointed professor of rhetoric, called to Nicomedia probably in 299.17According to Lactantius' On the Deaths of the Persecutors, Diocletian and Galerius, his Caesar, held private discussions in Nicomedia's palace all throughthe win14. Although it is technically possible to read the 6g clause at 2.50 as dependent on Eupaoavfrom the main clause (especially given the use of acT6v and not autz6v in the bg clause), such a reading is impossible, given what follows. For if the emperors (and not Apollo) were claiming that "the righteous" were an impediment to the gods' truthfulness, then Diocletian's wanting to know the identity of "the righteous" makes no sense. In translating the 6g passage as dependent on Xpfioat, I have followed the example of Cameron and Hall (n. 11 above) as well as the translatorsof the Nicene and Post-Nicene Fathers, series 2, vol. 1 (Edinburgh, 1890). 15. Note that Constantine refers not to Apollo's priestess, i1 but to the god's priesthood, i1 iUptta, kpcia. See LSJ and Cameron and Hall, "Lifeof Constantine" (n. 11 above), at Euseb. Vit. Const. 2.50. 16. Throughoutthis analysis the use of the term "oracle"can denote both the site at which people believed that a god communicated to humanity and the message itself. In the sources discussed here this notion is conveyed in a variety of ways. For example, Constantine says that Apollo prophesied that "the righteous" were affecting his truthfulness (Euseb. Vit. Const. 2.50 t0bv An6itho . . . paoav .. pfoat .. .). The site is the object of study here; the message is Apollo's claim. Lactantius, conversely,.. says that Diocletian's haruspex consulted Apollo at Miletus-who answered (De mort. pers. 7: imperator ... misitque aruspicem ad Apollinem Milesium. Respondit ille .. .). Oracles need not occur at a site specifically devoted to this purpose, but in order for an oracle to be attributedto a particulargod (in this case Apollo), I am assuming that something either in the prophecy itself or the site at which it was delivered (i.e., the temple, or perhaps a statue) links the communication to the god. 17. For his position, see Hier. De vir. ill. 80. Lactantius says that he journeyed to Bithynia as the temple of God was being overturned, an expression that links his arrival with the emperors' purge of Christians from the army and court in 299. For his arrival's coinciding with the army purge, see Lactant. Div. inst. 5.2.2 ("ego cum in Bithynia oratoriaslitteras accitus docerem cotigissetque ut eodem tempore dei templum everteretur"),and see n. 2 above.
THE GREAT PERSECUTION
61
ter of 302/3 (De mort.pers. 10.6-11.3).18 In this account, written soon after Constantine's conversion of 312,19 Lactantius claims that Galerius hoped these discussions would bring to the general population the persecution that had alreadytargetedthe army and court in 299 when imperial priests in Antioch had been unable to read the auspices (10.6).2oAfter many intense sessions with Galerius in the Nicomedian palace, Diocletian was unable to "restrainthe madness of his headstrongcolleague," accordingto Lactantius. Consequently the senior emperorthen sought advice from some civil magistrates and military officers (De mort.pers. 11.4-5). These men, at least in Lactantius' report, responded to pressure from Galerius and supported the Caesar's argument (11.6). Still unwilling to proceed, Diocletian then "decided it would be best to consult the gods." So he "sent a haruspex to the Apollo" of Didyma, nearMiletus, where,Lactantiussays, the oracle answered as "one would expect of an enemy of God's religion" (11.7). And thus, Lactantius concludes, "Diocletian was drawnover from his purpose,"and on 23 February303 began to issue the edicts that levelled churches, burnedScripture, arrested clergy, and ultimately called for sacrifice from the whole population (11.8-15.7).21 Perhapsbecause oracles were once thought to be nearly defunct by 300,22 most scholars dealing with the history of the Great Persecution simply 18. De mort.pers. 10.1: Cum ageret in partibusorientis, ut erat pro timore scrutatorrerumfuturarum,immolabatpecudes et in iecoribus earumventuraquaerebat.2 Tumquidem ministrorumscientes dominumcum adsisterent immolanti, imposuerunt frontibus suis immortale signum; quo facto fugatis daemonibus sacra turbatasunt. Trepidabantaruspices nec solitas in extis notas videbant et, quasi non litassent, saepius immolabant. 3 Verumidentidem mactataehostiae nihil ostendebant,donec magister ille haruspicumTagis seu suspicione seu visu ait idcirco non respondere sacra, quod rebus divinis profani homines interessent. 4 Tunc ira furens sacrificare non eos tantum qui sacris ministrabant,sed universos qui erant in palatio iussit et in eos, si detrectassent, verberibus animadverti,datisque ad praepositos litteris, etiam milites cogi ad nefanda sacrificia praecepit, ut qui non paruissent, militia solverentur. ... 6 Deinde interiecto aliquanto tempore in Bithyniam venit hiematum eodemque tum Maximianus quoque Caesar inflammatusscelere advenit, ut ad persequendos Christianos instigaret senem vanum, qui iam principium fecerat.... 11.3 Ergo habito inter se per totam hiemem consilio, cum nemo admitteretur.... 4 Nec tamen deflectere potuit praecipitis hominis insaniam. 5 Placuit ergo amicorum sententiam experiri .... 6 Admissi ergo iudices pauci et pauci militares, ut dignitate antecedebant. Quidam proprio adversus Christianos odio inimicus deorum et hostes religionum publicarum tollendos esse censuerunt, et qui aliter sentiebant, intellecta hominis voluntate vel timentes vel gratificarivolentes in eandem sententiam congruerunt.7 Nec sic quidem flexus est imperator, ut accommodaret assensum, sed deos potissimum consulere statuit misitque aruspicem ad Apollinem Milesium. Respondit ille ut divinae religionis inimicus. 8 Traductusest itaque a proposito et quoniam nec amicis nec Caesari nec Apollini poterat reluctari, hanc moderationem tenere conatus est, ut eam rem sine sanguine transigi iuberet, cum Caesar vivos cremari vellet qui sacrificio repugnassent. The date of the Didymaean oracle derives from the fact that the result immediately prompted the general edicts, which Euseb. Hist. eccl. 8.2.4 dates to 303. 19. For the date, see T. D. Barnes, "Lactantiusand Constantine,"JRS 63 (1973): 29-46. 20. I discuss this portion of Lactantius' testimony more fully below. 21. See also Euseb. Hist. eccl. 8.2.4. 22. E.g., E. Krebs, "Die Religionen im R6merreich zu Beginn des vierten Jahrhunderts,"in Konstantin der Grosse und seine Zeit, ed. F J. D61ger (Freiburg, 1913), 6, drawing on the testimony of Plutarch and Clement of Alexandria. It is now generally accepted that the oracles continued to be consulted, at least into the reign of Constantine. A good example of such a consultation in the late third century is the oracle with which Porphyryin 300 begins his Life of Plotinus; cf. L Brisson, "L'oracle d'Apollon dans la Vie de Plotin par Porphyre,"Kernos 3 (1990): 77-88 (contra R. Goulet, "L'oracle d'Apollon dans la Vie de Plotin," in Porphyre: La "Vie de Plotin," ed. L. Brisson, M.-O. Goulet-Caz6, R. Goulet, and D. O'Brien [Paris, 1982], 1.369-412). For analysis of the continued popularity of Late Antique oracles, cf. D. E. Aune, Prophecy in Early Christianityand the Ancient MediterraneanWorld(GrandRapids, Mich., 1983), 24; P-F Beatrice, "Un oracle antichr6tienchez Arnobe," in Mimorial Dom Jean Gribomont, ed. Y. de Andia et al. (Rome, 1988),
62
ELIZABETHDEPALMADIGESER
assumed that Constantine and Lactantius were describing the same event from two different perspectives.23Henri Gr6goire, in fact, used Constantine's quotation of Apollo's oracle to restore a very fragmentaryinscription from Didyma that referred to Christians.24Nevertheless, some have been troubled by differences in the two accounts. Joseph Fontenrose, for example, observed that Constantine's oracle did not follow the conventions at Didyma, where a priestess communicatedthe prophecy.25And a few scholars have gone so far as to suggest that Lactantius and Constantinewere referring to distinct events: Johannes Geffcken, for one, simply assumed that the accounts were different, although he thought that Constantine's oracle had also come from Didyma.26The rest of those who differentiatethe two, however, do so because they assume that in referringto a "Pythian"oracle, Constantine must be talking about Delphi,27 the quintessentially Pythian Apollo.28 For Late Antiquity, however, Constantine's allusion to the "Pythian"god is not enough to distinguish his oracle from Didyma's,29for fourth-centurytexts routinely apply the epithet to a number of sites, from Delphi to Didyma to Daphne.30
107-29; Cameron and Hall, Life of Constantine, 245; J. Fontenrose, The Delphic Oracle: Its Responses and Operations, with a Catalogue of Responses (Berkeley and Los Angeles, 1978), 5; P. de Labriolle, La Rdactionpaienne: Etude sur la poldmique antichrdtiennedu ler au Vie siecle (Paris, 1934), 132; T. E. Gregory, "Julian and the Last Oracle at Delphi," GRBS 24 (1983): 355-66, at 363; and P. Athanassiadi, "The Fate of Oracles in Late Antiquity,"Deltion Christianikes Archaiologikes Etaireias 115 (1989-90): 271-78. 23. H. Gr6goire, "Les chr6tiens et I'oracle de Didymes," in Mdlanges Holleaux: Recueil de Mimoires concernant l'antiquiti grecque offert a Maurice Holleaux (Paris, 1913), 84; N. H. Baynes, "The Great Persecution," in The Cambridge Ancient History 12, ed. S. A. Cook et al. (1929-39; reprint Cambridge, 1981), 665; de Labriolle, Reaction palenne (n. 22 above), 319; A. Rehm, "Kaiser Diokletian und das Heiligtum von Didyma," Philologus 93 (1938): 74-84, at 74; W. H. C. Frend, Martyrdomand Persecution in the Early Church (New York, 1965), 363; S. Williams, Diocletian and the Roman Recovery (London, 1985), 175; J. Fontenrose, Didyma: Apollo's Oracle, Cult and Companions (Berkeley and Los Angeles, 1988), 206-7; idem, Delphic Oracle (n. 22 above), 425; Davies, "Persecution of A.D. 303" (n. 1 above), 77; R. Lane Fox, Pagans and Christians (New York, 1989), 595; S. Levin, "The Old Greek Oracles in Decline," ANRW2.18.2 (1989): 1626-27; W. Portmann,"Zu den Motiven der diokletianischen Christenverfolgung," Historia 39 (1990): 212-48, at 217; Simmons, Arnobius (n. 4 above), 41. 24. The inscription is D 1 306 in Fontenrose's collection (Delphic Oracle, 425); Gr6goire, "Chr6tiens et I'oracle" (n. 23 above), 81-91. Most scholars have since rejected Gr6goire's creative restoration (e.g., Baynes, "GreatPersecution" [n. 23 above], p. 665, n. 3; Fontenrose, Delphic Oracle, 425), Lane Fox, Pagans and Christians (n. 23 above), 595, being a notable exception; for an alternative interpretationof the inscription, see Athanassiadi, "Fateof Oracles" (n. 22 above), 274. 25. Fontenrose, Didyma (n. 23 above), 297; Davies, "Persecutionof A.D. 303," 78. Gr6goire("Chr6tiens et l'oracle"), 86-87, thought that Constantine's reference to the oracle that "ne r6ponditpas par l'interm6diare accoutum6 d'une bouche humaine"necessarily meant that Eusebius-whom he thought had invented Constantine's edict-had confused Constantine's oracle with a later one, namely the talking statue of Zeus Philios erected by Theotecnus in Antioch during the reign of Maximin Daia (311-13); on this incident, see p. 72 below. 26. J. Geffcken, The Last Days of Greco-RomanPaganism, trans.S. MacCormack(Amsterdam,1978), 32. 27. This was the first assumption of Fontenrose, Delphic Oracle, 425. He subsequently changed his mind in Didyma, 206-7, but still maintained that Constantine, "rely[ing] on an oral tradition that had already distorted the content and circumstances,"thought that the oracle of 303 was Delphic; Cameron and Hall, "Lifeof Constantine,"245; Barnes, "Monotheists"(n. 4 above), 158-59 (a reversal of his position in Constantine and Eusebius [n. 1 above], 21). 28. Cf. H. V. Geisau, "Pythios,"RE 24 (1963): 566f. 29. Fontenrose, Delphic Oracle, 89, 151, 191. 30. Cf. Euseb. Praep. evang. 4.4, for the sense that the term "Pythian"was used by his opponents to refer to oracles that were "true"and "divinely inspired."For references to the Daphnic Apollo as "Pythian," see Libanius Or. 11.228 and Theodoret Hist. eccl. 3.7; Fontenrose (Didyma, 24) discusses a similar reference to Didyma (D 1 159 = Bericht 8:23).
63
THE GREAT PERSECUTION
EVIDENCE
FOR AN ORACLE
FROM APOLLO
AT DAPHNE
IN 299
Although the term "Pythian"is not proof enough that the emperorand Lactantius recorded two different oracular events, evidence in their accounts and other ancient sources does indicate that Constantine's oracle was not delivered in Didyma in 303 but at Daphne in 299. First, the context for Constantine's oracle does not fit the situation in 303 when Diocletian consulted Apollo at Didyma. Before narratingApollo's complaint, Constantineclaims that persecution was "rekindled"by Diocletian, Galerius, and Maximian when "all human and divine matters" were "at peace" (apud Euseb. Vit. he continues in Const.2.49). 31"Underthese circumstances"(b TIrvtmKaua), the next sentence, "they were saying that Apollo" complained about the socalled "righteouson earth"(2.50). Once Diocletian understoodto whom the prophecy referred,"on the very spot" (a'TziKa)he began "drawingup edicts of blood and venom" (2.51). Humanaffairs, however, were not as much "at peace" in the East in 303 as they had been a few years prior.For in 298, the eastern emperorshad just put down usurpersin Egypt and defeated the Persians;32but in 302-3, Galerius and Constantinehimself were busy fighting Sarmatians and Carpi.33Moreover, at least from a Christian perspective, which Constantine certainly shared by the time he issued his edict in 324, "divine affairs"had been disruptedsince Diocletian purged Christiansfrom the army and court in 299. Both Lactantius and Eusebius believed that this was when the persecution began.34As a member of the eastern imperial entourage in 299, Constantine would not only have known about this order, but was probably with Diocletian and Galerius at the palace in Antioch whence the order was issued;35he thus would have witnessed the whipping of those who confessed to be Christian. Thus, from his point of view, the order to whip Christiancourtiers would certainly have been the first bloody edict that Diocletian "began to draw up."36The letters that Diocletian sent to his army officers requiring sacrifice within the ranks would have contained the second edict (Lact. De mort.pers. 10.4), and more would follow in 303.37Constantine,writing from the vantagepoint of 324, has compressed 31. See n. 13 above for the text of Constantine's account. 32. T. D. Barnes, The New Empire of Diocletian and Constantine (Cambridge,Mass., 1982), 54-55, 63, drawing upon Hier. Chron. 226e; Eutr. Brev. 9.22-23, 25; Lactant. De mort. pers. 9.6; Vict. Caes. 39.34; Festus Brev. 25; and Amm. Marc. 24.1.10. 33. See Barnes, New Empire (n. 32 above), 64, which draws on Lactant.De mort.pers. 13.2; MattinglySydenham, RIC 6.510; and Anonymus Valesianus 1.2-3. See also T. D. Barnes, "Sossianus Hierocles and the Antecedents of the 'Great Persecution,'" HSCP 80 (1976): 239-52, at 250-51; and W. Ensslin, "Maximianus (Galerius),"RE 11 (1930): 2516f. 34. See n. 1 above for the date, n. 2 above for Lactantius' and Eusebius' treatment of this event, and n. 18 above for the text of Lactantius' account. 35. For the Antiochene venue and the presence of Diocletian and Galerius at the time the purges were ordered, see Lactant. De mort. pers. 10 and Div. inst. 4.27.4-5 (printed here in n. 88 below); for Constantine's presence, see Barnes, "Sossianus" (n. 33 above), 250-51, drawing on Euseb. Vit. Const. 1.19; Anon. Val. 1.2-3; Constantine Oration to the Saints 16; Pan. Lat. 9(4).21; and The "Chronicle" of Joshua the Stylite, trans. W. Wright (Cambridge, 1882), p. 6. 36. ouvtarsts (Euseb. Vit. Const. 2.51); note the imperfect; see n. 13 above for the text of Constantine's account. 37. Datis adpraepositos litteris: see n. 18 above for text, n. 13 above for the text of Constantine'saccount, and n. 21 above for the edicts of 303-4.
64
ELIZABETH DEPALMA DIGESER
his chronology, to be sure.38Nevertheless, the peaceful context he claims for his oracle is ill suited to the Didymaean prophecy of 303. In saying that the oracle interruptedpeace in divine affairs and brought about legalized bloodshed, Constantine more likely describes an event just prior to the purge in 299 of the army and Antiochene court. Next, as Fontenrosehad observed, the way in which Constantine's oracle was conveyed also does not conform to the traditions followed at Didyma or Delphi, but ratherwith those at Daphne. The key point is Constantine's claim that Apollo spoke from a dark alcove and an inner chamber (or antron) and not through a human being (Euseb. Vit. Const. 2.50). Delphi, Didyma, and the other oracles of Apollo in Greece or Asia Minor were, conversely, famous for the human priestesses through whom Apollo was believed to communicate his messages.39And while Delphi, at least, offered other types of divination, it is hard to reconcile its haruspices or lot oracles with the lengthy statement that Constantine reports from Apollo (Euseb. Vit. Const. 2.50).40 In Syria, however, were Pythian (i.e., "Apolline") oracles that fit Constantine's description more closely. Here, the prophecy was not conveyed through a human priestess; rather,Apollo's oracle was associated with the god's statue. Both Fontenrose and Gregoire, in fact, had assumed that Con38. See H. A. Drake, "Suggestions of Date in Constantine's Oration to the Saints," AJP 106 (1985): 335-49, at 344, for examples of similarly compressed chronology in the Oration to the Saints. Constantine would have been likely to review events quickly in the 324 edict because his main point was not the events of the Great Persecution; rather,as the western emperor who had just invaded the East-ostensibly to defend Christians from the emperor Licinius-Constantine's purpose in alluding to the demise of the earlier persecutors was to legitimize his recent conquest. 39. For late third- and early fourth-centuryaccounts of Didyma's priestess see Porph. Aneb. frag. 142 (according to the system of Porphyry,Lettera ad Anebo sulla teurgia, ed. and trans. G. Faggin [Genoa, 1982]) and lamb. Myst. 3.11; and for late third- and early fourth-centuryaccounts of Delphi's priestess, see the same, in addition to the earlier accounts of Luc. 5.97-101; Plut. De def or. 414b, 435b-d, 437c38d, De Pyth. or. 397b-d, 404e, 405c-e, 408c-d, f; and Paus. 10.5.4. The most reliable modern studies of Didyma and its priestess are Fontenrose, Didyma; H. W. Parke, The Oracles of Apollo in Asia Minor (London, 1985); idem, "Apollo and the Muses, or Prophecy in Greek Verse,"Hermathena 130-31 (1981): 99112; and D. Potter, Prophets and Emperors: Human and Divine Authorityfrom Augustus to Theodosius (Cambridge, Mass., 1994), especially at 41-42. Although Fontenrose was uncertain of the Didymaean prophet's gender when he wrote Delphic Oracle (p. 228), by the time Didyma was published he had decided that Iamblichus' account could be trusted (pp. 78-85). For modern accounts of the priestess at Delphi, see Dodds, Greeks and the Irrational, 70-71; Fontenrose, Delphic Oracle, 196-97, 218; Gregory, "Julian and the Lost Oracle" (n. 23 above), 355-66; H. Lloyd-Jones, "The Delphic Oracle," G&R 23 ' (1976): 60-73; Potter, Prophets and Emperors, 40-41; G. Rougemont, "Techniques divinatoires Delphes: Etat present des connaissances sur le fonctionnement de l'oracle," in Recherches sur les artes 4 Rome, ed. J. M. Andre (Paris, 1978), 152-54. For general studies of the oracles of Apollo, see Dodds, Greeks and the Irrational; Parke, Oracles of Asia Minor; and Y. Hajjar,"Divinites oraculaires et rites divi inatoires en Syrie et en Phenice l'apoque gr6co-romaine,"ANRW2.18.4 (1990): 2458-508. Not long ago, a team of geologists determined that ethylene vapors could have helped the Pythia achieve her trancelike state in antiquity. This gas, which until recently was used for general anesthesia, "produces disembodied euphoria, an altered mental status and a pleasant sensation" (toxicologist Henry Spiller quoted by W. J. Broad, "The Delphic Oracle-A Real Gas," The Montreal Gazette [23 March 2002]: H14; see also G. Gugliotta, "A High Calling for Priestesses at Delphi," The Los Angeles Times [7 February2002]: E9). 40. For the presence of haruspices at Delphi, see Paus. 10.6.1; Plin. HN 7.56.203; P. Amandry,La mantique apollinienne a Delphes (Paris, 1950), 59-60; and E W. H. M. Myers, "Greek Oracles,"in Hellenica: A Collection of Essays on Greek Poetry, Philosophy, History, and Religion (London, 1880), 388-449, at 407; for lot oracles at Delphi, see Amandry,Mantique apollinienne, 183; and Lloyd-Jones, "Delphic Oracle" (n. 39 above), 66; Fontenrose (Delphic Oracle, 223) doubts, however, whether lots were used there. Gregory ("Julianand the Lost Oracle,"360-61) argues for prophetic springs at Delphi.
THE GREAT PERSECUTION
65
stantine was describing this type of prophecy.41There are several examples of such Syrian oracles. For instance, according to Lucian, priests would ask questions of the statue of Apollo from Hierapolis. They believed that the god communicatedby moving his statue as well as by speaking through it. In fact, Lucian's language echoes Constantine's in saying that the god "speakswithout priests or prophets.This god takes the initiative himself and completes the oracle of his own accord."42While Lucian provides evidence for propheticApolline statues in Syria, the prophecies at Hierapolis lack the dark recesses of Constantine's account.43At Daphne near Antioch, however, was an oracle of Apollo that functioned as Constantinedescribed. The evidence for how the statue of Apollo at Daphne prophesied comes from a well-documented incident during the reign of Julian (361-63). Despite the strong similarities linking Julian's experience at Daphne with Constantine's description of his oracle, the connection between the two has never been drawn-probably because all modern studies of prophecy at Daphne have concentratedon the Castalia, the prophetic springs.44The emperorHadrian had closed this source of divination, however, after it had foretold his accession.45According to Ammianus, the Castalia remainedclosed until after Julian's accession (Amm. Marc. 22.12.8). Since Libanius records a prophecy at Daphne foretelling Julian's rise to power (Or. 60.5), however, sources of divination other than the springs must have been available at the site.46
41. Fontenrose, Didyma, 207; Gr6goire, "Chr6tienset l'oracle," 86-87. Egypt was also home to talking oracular statues, a phenomenon that continued into the fourth century. Nevertheless, none of these sites (i.e., the oracle of Re-Harmaclus,the bull of Kom el-Wist, and the temples of Sobek in the Fayyum) could have been described as "Pythian";see D. Frankfurter,Religion in Roman Egypt: Assimilation and Resistance (Princeton, 1998), 150-51, 174-75; see also G. Loukianoff, "Une statue parlante ou oracle du dieu R6-Harmaklos,"ASAE 36 (1936): 187-93; L. Habachi, "Finds at K6m el-Wist," ASAE47 (1947): 285-87; and G. Brunton, "The Oracle of K6m el-Wist," ASAE47 (1947): 293-95. n & Kalinap' AiyunTiotot, T& 81KKai vAVt, 42. 36: Ptavivta rnoX• Kai v t Vtpv nap' "EXXrlao, noX 8&a6T6g KtvFstat ntpo~prlit•v (p0yyovTat, 5686 Tr 81 Aoin noXod iortv. &XkkX& rTp oi'tEipi'v dveu oUTtE T•e Greek text and translationis from Lucian "De dea Syria," ed. H. W. IcT Tilv Iavrtlirv qTC•Xoao6Tooupyist. Attridge and R. A. Oden (Missoula, Mont., 1976). Note that Lucian indicates the distinctiveness of Apollo at Hierapolis as compared to Greek, Egyptian, Libyan, and Asian (i.e., Western Turkish)oracles; he does not compare this oracle to others in Syria. 43. See Hajjar,"Divinit6s oraculaires"(n. 39 above), 2290, 2268, n. 239. Lucian's account (De dea Syria 34-35) makes it clear that the statue of Apollo is prominentlydisplayed in the temple at Hierapolisjust be81 hind Helios' throne (iv a6riO81&TCV v~i ot6vra vv ptorspj KieTattp~ra plav0p6vo4'HXiou. .... .•. t T6v Op6vovzobrov K•iEat46avov An6kX),voq.. .). Given the antics of this statue that he describes here, it could not be described as prophesying from a dark alcove and inner chamber. Other reports of prophetic statues include Numenius (apud Origen Cels. 5.38), Asclep. 23, 24, and 38; and Macrob. Sat. 1.23.13 (Van Liefferinge, "Oracles chaldaiques" [n. 6 above], 90, 94-96); see also Eusebius' account of the statue to Zeus Philios, erected in Antioch probably during the summer of 311 (Hist. eccl. 9.3 and p. 72 below). 44. Amandry,Mantiqueapollinienne (n. 40 above), pp. 132-33, p. 137, n. 2, p. 183; A. Bouch6-Leclercq, Histoire de la divination dans I 'antiquit&,vol. 3 (Paris, 1880), 267; G. Downey, A History of Antioch in Syria (Princeton, 1961), 222; Gregory, "Julian and the Last Oracle," 357-58, 360; Hajjar, "Divinit6s oraculaires," 2283-84; Potter, Prophets and Emperors, 13-15. 45. Amm. Marc. 22.12; Sozom. Hist. eccl. 5.19; Downey, Antioch (n. 44 above), p. 222, with n. 103, p. 387; idem, "The Water Supply of Antioch on-the-Orontes in Antiquity,"Annales Archdologiques de Syrie 1 (1951): 171-87 at 182; Hajjar, "Divinit6s oraculaires,"2284; E W. Norris, "Antioch on-the-Orontes as a Religious Center. I: Paganism before Constantine,"ANRW2.18.4 (1990): 2338-39. 46. Downey (Antioch, p. 222, n. 103, p. 387) explains the pre-Julianicoracles by positing that Ammianus invented the story of Hadrian's prophecy and closure, since they are not mentioned by John Malalas in his discussion of Hadrian'sre-engineering of the waterworksthere (11.14). But it is not clear what Ammianus
66
ELIZABETH DEPALMA DIGESER
A close examination of the accounts describing Julian's advent to Daphne makes clear that Apollo's prophecies were associated with the statue itself. Preparing in Antioch for his upcoming campaign against the Persians, Julian had gone to the suburb of Daphne to visit the famous temple of Apollo (Amm. Marc. 22.13.1). 47According to Ammianus, Julian wanted to reopen the nearbyCastaliansprings (22.12.8).48Julianprobablydid so, since later accounts describe how the Daphnic Apollo prophesied through these springs. Indeed, the only extant accounts of the Castalia's prophecies come from the period after Julian.49But the emperor's efforts must not have been immediately successful:50 For Ammianus says that Julian then spoke to the god (deum adfatus).51And Julian himself says that the god's statue (Tb iyakla) gave him signs (FFt•ilCflrvF),a description consistent with how Proclus says theurgistsreceived divine messages as they practiced"telestic" rituals before a god's statue.52According to Ammianus, after this encounter in the temple, Julian decided to purify the temple grounds, in part by moving some bodies buried nearby (22.12.8). Christianaccounts claim that one of these corpses belonged to Babylas, an Antiochene bishop martyredunder Decius and interredin the sacred precinct when Julian's brotherGallus was Caesarin Antioch underConstantiusII (337-61).53 These later sources vary, however, regardingwhat had transpiredin the temple: Socrates, for example, says that "the daemon dwelling in the temple did not reply,"muteness that he attributesto Apollo's fear of Babylas (Hist. eccl. 3.18).54 In Sozomen's account, conversely, as soon as the emperorenteredthe temple Apollo would gain by this deception. On the contrary,Daphne was near Ammianus' home turf, and he would have been familiar with local lore. Moreover, Philostratus' account of Apollonius' visit (V A 16) may anachronistically reflect the silence of these springs after Hadrian(Norris, "Antiochon-the-Orontes"[n. 45 above], 2338-39). 47. The best description of Antioch's suburb of Daphne, a site sacred to Apollo since the Seleucids (Lib. Or. 11.86-93), is to be found in Lib. Or. 11.86-98, 228; see also Philostr. VA 16; Amm. Marc. 22.12.8; 22.13.1; John Malalas 11.8, 14, 30; 12.38, 47; 13.19; Sozom. Hist. eccl. 5.19-20; and Suda, s.v. "Kastalia." 48. 22.12.8: "Haecquedum more pacis ita procedunt, multorumcuriosior lulianus, novam consilii viam ingressus est, venas fatidicas Castalii recludere cogitans fontis, quem obstruxisse Caesar dicitu Hadrianus mole saxorum ingenti, veritus ne (ut ipse praecinentibusaquis capessendam rempublicamcomperit), etiam alii similia docerentur:deumque adfatus circumhumatacorpora statuit exinde transferri,eo ritu quo Athenienses insulam purgavit Delon" (Latin text is from J. C. Rolfe's 1950 Loeb edition). 49. Suda, s.v. "Kastalia";Eudocia Violarium520; EvagorasHist. eccl. 1.16; Gregory of Nazianzus Disc. 5.32; Nonnus Abbas Hist. Juln. 21; Procopius of Gaza Ep. 66.11; Sozom. Hist. eccl. 5.19. 50. Evagrius Scholasticus Hist. eccl. 1.16: 'IouXhltvb6g v ydp 6 d&tTilptog,1 0eooauyOTlg uTpavvig, 6 cpavilvKai spopsrlziav PTlvKaorakiav aKyOV Kai TnetSi6? 6 AavpvaiogAsr6Xmov, Xsov, p•toTt6plvog, p• K dvF•.Eiv TI86vato t pat3aoi XprloXrlptagopvp,, Bapljhazo dtyiou navToitog yEt~t6voV 9Xtotopi ovto;, it text is from the edition of J. Bidez and L. Parmentier [Amsterdam, Ttp tpercarOsatb6v (Greek iyttov.... Hist. eccl. 10.36. 1964]); see also Rufinus 51. See n. 48 above for the text of Ammianus' account. 52. Misopogon 361c: Aoi lp?vo v 866KEtKcai 6 6 086g, cepb toI tupbg Tbv veciv tnoe•o•otvat 'Hxtov rrEorlrlvE T6 ~ iytpa, Kai To6TOU ldptupa KaX6tIT ydp eiotk06vt LPotrTp•rTov tbv pLyav tpbg 6 rtomTobv"TobgVan Lieftext is from Kroll's 1969 Loeb edition); Proclus In R. 1.39.9-10 and 14-20; ag. .... (Greek feringe, "Divinit6s oraculaires,"273-74. 53. Rufinus Hist. eccl. 10.36; Sozom. Hist. eccl. 5.19; Socrates Hist. eccl. 3.18; Theodoret Hist. eccl. 3.6; Evag. Hist. eccl. 1.16; Philostorgius Hist. eccl. 7.8. 54. Td yap Kart tilv Avzt6XetaviEsp& zdv 'EXXlveovdvotyqvat KEXKe6oa;,Xprlpolbv apesivnap&rob v 6~6 votKOc)vtz iepcj 8aiPo)v T6v ycUzova 8S80otK6g,X y) 861BapukdvTvb Adcpvt Axr6•Xovo;g or'u6v.. '0; (Greek text of Socrates is from Migne, PG 67). Padptupa, o•K dTEKpivaTo
THE GREAT PERSECUTION
67
complained that "'the place is quite full of the dead' and that because of this the oracularresponses were prevented from going forward."Sozomen, who knew of the Castalia's closure by Hadrian,adds that "the daemon had 6"h6yog X ceasedto utteroracles"('E KCiWcvou pilXPrlOlpo0t Ouvi•0omg -b
afterGallustranslatedBabylas'relicsthere(5.19)-further evi:atpi6vtov) betweenHadrian'sclosureandJulian's dencethatthestatuewasprophesying
of thesprings.55 about uncertainty Despitetheancientaccounts' reopening
whetherJulianreceivedhis oracle,then,they all still indicatethat,before the Castaliareopened,people could-and did-consult Apollo at Daphne throughhis statuein the temple.GregoryNazianzencompletesthe argument,observingin his seconddiscourseagainstJulian(Disc.5.32) that,after the emperor's death, "Apollo is again a mute statue"(indtv &vpti&g tc0ovog 6 An4i6oWv).56 Thus, Constantine's description of how Apollo prophesied fits the fourth-centurytradition at Daphne; but it is inappropriatefor Didyma, the site that Diocletian consulted in February303.57 Constantine's Oration to the Saints provides the fourth hint that Daphne was implicated in the persecution. Not far into the speech, delivered after his conversion, Constantinedeclares his intention to testify to the divine nature of Christ by using pagan sources, one of whom was the Erythraean Sibyl. Constantine's aim is to introduce a Sibylline oracle that he considers a divinely inspired prophecy (Oration to the Saints 18), but before he does so, he complains about Apolline prophecy in general. "Her parents,"Constantineclaims, "devoted"the Sibyl "to this ridiculousservice, throughwhich shameful furies and nothing very worthy of respect comes into being-and
55. Kai ydp 'IouXtavoi p6vou KpaTOUVTOg Tflg'Popais)v oiKoupJiVrg, onov63v Kai Kvioorlg KLai a60oviag Kai TbreXtuaiov Xpliorag,ij•sye Kai ai6bg tfig 7por~paq 0updTowvpwrT•ov, o6hv irTTov flppEt" 6 o•s(irig ntspaOijvattoTOvd 6E savTfiov, rilv airiav. 'EirtSi' ydp sPEpo6XCuro Paaost4bg rEpi 7v oi s 66c•et Kai E6iTo nEpi 7v napayev6pevog sig 6biep6v, &va0ipaot Kai 8uoiatg ziptc TO '6tp6vtov, is fromMigne,PG 67). AlthoughSozomendeptXOTip.0s ono66ac pil•p•csiv (Greektextof Sozomenthroughout
the earlierauthor'sworkwith his own pendedheavilyon Socrates,here he has clearlysupplemented and research;see G. F Chesnut,TheFirst ChristianHistorians:Eusebius,Socrates,Sozomen,Theodoret, Evagrius(Paris,1977),197-98;forsimilaraccountsof Julianandtheoraclesee alsoEvag.Hist.eccl. 1.16; JohnChrysostom St. Babylas2; Philostorgius Hist.eccl. 7.8; TheodoretHist.eccl. 3.6; andRufinus Hornm. Hist.eccl. 10.36. 56. Ocr Tt(p0fyyerTat 6pig, tkiprlgplavTrEerat,OUrc•ttl-uOiatrlApoiTrat o6K oib' ivrtvwv kfilv . fldXtvo••iTn i KaozatiFooiyiraTKtKaioty Kai ii6•p oktiv o0p)avTzu6pcvov,6•&• pl60av aiXt.ppCdT(VO 6 rA6XX(ov,IrdXtv (Greektext ~1AdhpvrI ysF.X6pvovrtdXtv dv6ptsdg&povog pyuT6vozstvPlU0p Oprvo6CiEvov is fromGregoryof Nazianzen,Discours4-5: ContreJulien,ed. andtrans.,J. Bernardi, SC, 309 [Paris, on Gregory'sdiscourse(Hist.Juln.22; PG 36) ad1983]).NonnusAbbasin his sixth-century commentary mitsthathe knowsnothingabout"thestatue,whereit waserectedandhowit spoke"(FI-pito6Toutoivuv p Kai7-nSg0yy•'YTO,ig OOioTopjoapgev). Nevertheless,he speculatesthat TOO d&vptdvToqto6 TEY(YTaTO PEl mistaken. Gregoryis referringto Delphi.Nonnusis, however, AlthoughGregorynowheresays thatthe talkingstatuewas at Daphne,his referenceto theCastalianfount'ssilenceagain(closedbeforeJulian,reopenedby him,nowignored)andto Daphneherselfclearlyindicatesthathe is talkingabouttheApolloat however,sincethetempleandstaDaphne;cf. Bernardi,p. 358, n. 1. Gregoryis speakingmetaphorically, tue burnedto the groundbeforeJulianleft Antioch:Sozom.Hist. eccl. 5.20; TheodoretHist. eccl. 3.17; Amm.Marc.22.13.1. 57. I am not claiminghere that Daphnewas an oracularsite whose reputationrivalledDelphi or fororaclesonlywiththeprophecythat Didyma.Indeedit is possiblethatthe statuedevelopeda reputation Constantine reports.Thekeypoint,however,is that-as Sozomenindicates-Apollo'sstatuedidhavethis beforeGallus'arrivalin Antioch. reputation
68
ELIZABETH DEPALMA DIGESER
the same things are recorded about Daphne."58Robin Lane Fox took this passage to mean that Apollo "had lusted after his priestesses and inspired them by sexual contact."He assumed that Constantine was discussing, not "Daphne the place," but "Daphnethe person,"namely, the maiden who escaped Apollo's advances by turning into a laurel tree.59But Lane Fox's analysis conflicts with what Constantine says, since Daphne's story described her refusing Apollo's attentions, not serving him.60Constantine is more likely referring to something that he-and his audience-knew took place at Daphne,61namely the consultation of Apollo before Diocletian's persecution.62 Finally, a fragment of the late fourth-centuryhistorian Gelasios of Caesarea preserved in Theodoros Anagnostes' sixth-century Church History confirms both the date and venue proposed here for Constantine's oracle.63 T. D. Barnes has long thought that bishop Gelasios described a Daphnic oracle before the army purge (hinted at in Constantine's oration),64but the cleric's account has never been linked to the emperor's 324 report of the Pythianoracle. Nevertheless, the two sources are clearly describingthe same event. Gelasios reports that Galerius was one day "sacrificing to the daemons" in an antron or inner chamber,the same type of place in which Constantine's prophecy occurred.65During these rituals, a certain Theotecnus, v yevia0at, i'peta 58. i -roivuv'Epu)paia ipUOUXa, yeve teT6 Tbv 'aumilPv pQ~oKouava KaTaKXvO••u ) "E•tl 1v 0To An6Xso)vo4,6tda6rpaSn'TolT to) 0p t6bOzfj OE. (popo~oa,KaiTbvtpiinoa nEpiBy6 6Potq paKuooSV(o6nr' albOrijl T)iv yovfov sintsgEOM6Kr iXeirZonEtptinoUOa, o(potIirdov)d rEZoig XppLvost abrj1, str0t6zTrt Kai o66v oCpvbov 6t' jiv&aXTiovEg0uP1oi rCEpi Kazrd a6Tz olq iorTopoup?votq S7ntyivEzat, 1ota6Tn
r
r
Xarps•i, Ad(pvrlg(Greek text of the Oration is from Eusebius Werke,ed. I. A. Heikel, GCS 1 [Leipzig, 1902]). ri•j59. Lane Fox, Pagans and Christians (n. 23 above), 634. 60. Lib. Or. 11.94. Coming across this passage, R. P. C. Hanson-who thoughtthat Constantine'sspeech was apocryphal-declared it to be "inconceivable that Constantine the Great could really have referredto [Daphne] in a discourse," since "[t]he oracle was at no time famous and could not have rankedwith Delphi or Dodona." He thus takes the speech's reference to Daphne's fury to refer to the incident with Julian and dates the speech after 361 ("The Oratio ad Sanctos Attributedto the EmperorConstantine and the Oracle at Daphne,"JThS 24 [1973]: 505-11, at 507-8). While the date and venue of this speech are still hotly contested, most scholars now treat it as genuine; see, for example, Drake, "Suggestions of Date" (n. 38 above), 336-49; idem, "Policy and Belief in Constantine's Oration to the Saints," Studia Patristica 19 (1989): 43-51; T. D. Barnes, "Constantine's Good Friday Sermon,"JThS 27 (1976): 414-23; and idem, "Constantine'sSpeech to the Assembly of the Saints: Place and Date of Delivery,"JThS 52 (2001): 26-36. 61. Lane Fox, Pagans and Christians, 634, has argued that Antioch was the venue for this speech. 62. Already suspected by Barnes, "Sossianus,"252. 63. See Barnes, "Lactantius"(n. 19 above), 34, for date. 64. Barnes, "Sossianus,"252. In this article Barnes dated Gelasios' oracle to 302, the date he then supported for the persecution in the army.Barnes has since adopted 299 as the date for the army purge (see n. 1 above). Via Theodoros Anagnostes, Gelasios' account also found its way into the early ninth-centuryChronicle of Theophanes at AM 5794. See the comments of C. Mango and R. Scott, translators,in The "Chronicle" of Theophanes Confessor (Oxford, 1997), lxxv-lxxvi and at AM 5794, as well as E Winkelmann, "Untersuchungenzur Kirchengeschichte des Gelasios von Kaisareia,"SSBerT 3 (1965), 4-123. vv Matptptavb6roi 8ai65. Oe6)EKVO4 6 9V AVTtoX6i9 XptIotaviv i nidouhok 6)qrolOS iv T6 i &vtrpp SO6ev pootv UOSEv, 0S 610b XplOtpoo8at0povtKoO rv Krard XptoCta0v6jva6csT rtayYPhv XPrloo0ig ~iXpavcwsvE Tv KaO'filpsv vercsikaro. iKaiTor6n LcPdktora nbtoo0siq &vsppiRnt7oav pbg ToI6)r Kai S~•tylpbiv Mastpttavb6 (Gel. robq ouvrupdvvou4g nap6sppirloE rv Xprl-p6v ToO6aitovoq dA 6q6 t iya napdyyhula npopakX6p~svoq Caes. 3 [from Codex Baroccianusgr. 142] in Theodoros Anagnostes, Kirchengeschichte, ed. G. C. Hansen [Berlin, 1995], xli, 158). See also Gel. Caes. frag. 4 in C. de Boor, "Neue Fragmente des Papias, Hegisippus und Pierius, in bisher unbekanntenExcerpten aus der Kirchengeschichte des Philippus Sidetes," Texte und Untersuchungen5 (1889): 165-84. It is Maximianus Galerius and not Maximianus Herculius because the venue is Antioch (see n. 68 below for Theophanes' reference to Galerius at AM 5794). The aorist par-
THE GREAT PERSECUTION
69
"who preyed on Christiansin Antioch,"Gelasios says, slipped into the
antron with the emperor. "'Receiv[ing]' the prophecies,"Theotecnus, "via
Gelasios the daemonicoracle,"enjoinedGalerius"topersecuteChristians,"
continues. Galerius,"persuadedby" the oracle "mostof all,"not only "stirred up the persecution,"but the emperor also, by "presentingthe oracle of the daemon as some great directive,"incited the other emperorsto persecute as well, Gelasios concludes. Theotecnus' Antiochene venue combined with Galerius' presence suggests 299 as the most plausible date for this event since the Caesar is attested in Antioch only in the spring of 299.66 In the immediate environs of Antioch only one site-near the statue in Apollo's temple at Daphne-is at all associated with oracles in this period.67Moreover, the way in which the prophecy occurred,with Theotecnus hidden, "receiving" the prophecies, and conveying them via the oracle (&t6bXP1nClPo 6atclovtKou), harmonizes with the way in which the Daphnic Apollo prophesied for Julian. From a Christianperspective, Gelasios, I suggest, is identifying Theotecnus as the theurgistat Daphne and Apollo's statue as the "daemonic"oracle.68Gelasios' report, thus, supports the argumentthat in 299 an oracle from the propheticstatue of Daphnic Apollo preceded and led to persecution, first in the army and court and then more generally.69 EVIDENCE SUBSTANTIATING GELASIOS
Despite Gelasios' access as bishop of Caesareato whatever libraryhis forerunnerEusebius had assembled, his surviving fragments show that he was sometimes confused about events in the early part of the fourth century.70 Nevertheless, in the case of Galerius, Theotecnus, and the oracularstatue at ticiple i6to6gsabove would seem to indicate that even if the god gave several oracles, it was at one eventwhen Theotecnus slipped into the antron. 66. Lactant. Div. inst. 4.27.4; Barnes, New Empire (n. 32 above), 61-64. Although Barnes suggests that Antioch may have been Galerius' principal residence from 293-96, the Caesar was primarily on campaign in Egypt and against Persia in those years. 67. Theotecnus was affiliated with the prophetic statue of Zeus Philios in Antioch only after 311; see p. 72 and n. 84 below. 68. For the term "theurgist,"see n. 6 above; for their association with oracular statues in general, see p. 65 and n. 52 above. At AM 5794, Theophanes' account calls Theotecnus a y67; E~its OEO(To6tp, T~Kvpy61rzt nsts06pevoq6 FaekptoS MaS•tlttav6;9v z s06Etv Tro5s aigiooat oai kaLpfdivetv XplOtpo6o, 6noS6q p Kaer a XptortavC0v gyEipatCu zp av'cpqpb Oe6TEKvoXplPoLt6v, 8tayt6v, Oltop8F80(Kev), pejorative characterDe Mysteriis (AM 5794) ization against which lamblichus is at some pains to defend theurgists throughout (an observation that I owe to Olivier Dufault). Greek text of Theophanes is from Theophanis Chronographia, ed. C. de Boor (Leipzig, 1883). 69. Although Theophanes' account gives this incident a date of 301/2 (cf. Mango and Scott [n. 64 above] at AM 5794), his indication should not be credited. Since Theophanes gives two dates for the general edicts of persecution (Sth(yplhv in AM 5787 [294/5] and that in which churches were levelled and ••uyav Scripture burnt in 5795 [302/3]), in addition to a date of AM 5789 (296/7) for the army purge, it is clear that he is trying to integrate several accounts for this period. See Mango and Scott for Theophanes' other doublets and examples of incorrect dating, many of which resulted, no doubt, from his efforts to reconcile many different systems with the Alexandrian annus mundi he used as his chronological basis (pp. lxiilxiv). 70. For example, he claims in frag. 3 (de Boor) that after retirement Diocletian attempted to regain power with Maximian Herculius; cf. Barnes, "Lactantius,"34; for the contents of Eusebius' library see A. J. Carriker,"The Libraryof Eusebius of Caesarea"(Ph.D. diss., Columbia University, 1999).
70
ELIZABETHDEPALMADIGESER
Daphne, both Lactantius and Eusebius corroborateGelasios' account. Lactantius' Divinae institutiones, written in Nicomedia between 305 and 310,71 like Gelasios' report, also associates Galerius with an oracle, specifically a prophetic statue. In fact, Lactantius accuses Galerius of fraud, of issuing orders that would command "a statue to speak" (imaginem loqui).72By this feat (among others), Lactantiusclaims, many would be "influencedby him" (plurimi adlicientur ab eo) as he set out to persecute Christians (Div. inst. 7.17.5). This accusation occurs at the end of a monumental work, distinguished by its claim that many pagan prophetshad foretold Jesus' advent.73 Ostensibly discussing the end times,74Lactantiusdescribes the activities of Diocletian and Galerius, cloaked in the language of Daniel and Revelation.75According to Lactantius,an evil "king"will "rise out of Syria" (Div. inst. 7.17.2). This ruler will influence many "wise" people by commanding "fire to descend from heaven" and "an image to speak" (imaginem loqui), things "done under his order" (sub verbo eius, 7.17.5). Then, Lactantius continues, "he will try to raze the temple of God and persecute a righteous people" (7.17.6); in the process of doing so, the king "will cover" them "with the books of the prophets and thus will burn them."Precisely "fortytwo months will be given to him to desolate the earth"(7.17.8). But finally the righteouswill "call out to God,"who will "senda greatking from heaven" to "liberatethem" and "destroy all the impious" (7.17.11). 71. For the date, see E. Heck, Die dualistischen Zusatze und die Kaiseranreden bei Lactantius (Heidelberg, 1972), 37-46. 72. According to Lactantius, Galerius also commanded the "Sun to stand apartfrom its tracks"(solem a suis cursibus stare, Div. inst. 7.17.5), which may suggest that a statue of Apollo moved just like the one in Hierapolis described by Lucian. 73. See, e.g., Div. inst. 1.6 for Lactantius' intention to use Hermes Trismegistus and the Sibyl and 1.7 to use Apollo as prophets heralding Christ's advent and deeds. 74. See Digeser, Making of a Christian Empire (n. 4 above), 149-50. 75. Div. inst. 7.17.2-9, 11: "Alter rex orietur e Syria malo spiritu genitus, eversor ac perditor generis humani, qui reliquias illius prioris [i.e., Diocletian] mali cum ipso simul deleat. 3 Hic pugnabit adversus prophetam dei [Jesus] et vincet et interficiet eum et insepultum iacere patietur, sed post diem tertium reviviscet atque inspectantibus et mirantibuscunctis rapieturin caelum [probablymetaphoricallanguage for the persecution's inability to crush the Christiancommunity; see Rev. 11:7-12]. 4 Rex vero ille taeterrimus erit quidem et ipse, sed mendaciorumpropheta, et se ipsum constituet ac vocabit deum, se coli iubebit ut dei filium [references to the Jovian and Herculian emperors' claims to descend from divinity, Chron. min. 948]. Et dabiturei potestas, ut faciat signa et prodigia, quibus visis inretiat homines, ut adorent eum. 5 Iubebit ignem descendere a caelo et solem a suis cursibus stare [see n. 72 above] et imaginem loqui, et fient haec sub verbo eius: quibus miraculis etiam sapientium plurimi adlicientur ab eo. 6 Tunc eruere templum dei conabitur et iustum populum persequetur.... 7 quicumque crediderint atque accesserint ei, signabuntur ab eo tamquampecudes, qui autem recusaverint notam eius, aut in montes fugient aut conprehensi exquisitis cruciatibus necabuntur[cf. Rev. 13:16]. 8 Idem iustos homines obvolvet libris prophetarumatque ita cremabit. Et dabiturei desolare orbem terraemensibus quadragintaduobus. 9 Id erit tempus quo iustitia proicietur et innocentia odio erit.... 11 illi ... exclamabunt ad deum voce magna et auxilium caeleste inplorabunt, et exaudiet eos deus et mittet regem magnum de caelo, qui eos eripiat ac liberet omnes que inpios ferro ignique disperdat." Using an ostensibly apocalyptic and prophetic narrativeto comment on contemporaryevents would not have been as foreign to Lactantius' readers as it is to us. Indeed, Porphyry's Against the Christians illustratedhow the Book of Daniel-another ostensibly apocalyptic and propheticnarrative-was a veiled Jewish commentary on the reign of Antiochus IV. Since Porphyry's work aroused early fourth-centuryChristians vigorously to defend their Scriptures, it is not unlikely that Lactantius here has engaged in some ex eventu "prophecy"of his own. For those refuting Porphyry,see A. von Harnack,"Porphyrius,Gegen die Christen: 15 Bucher, Zeugnisse, Fragmente, und Referate,"Abhandlungender k6niglich Preussischen Akademie der Wissenschaften:Philosophisch-historische Klasse (1916): 29-31, 33-41.
THE GREAT PERSECUTION
71
Despite the thick Scripturalimagery, Lactantius'references to this "king" point his readers to Galerius, Diocletian's imperial colleague:76He has not only "come from Syria," the site of Galerius' famous victories against the Persians.77But also precisely forty-two months elapsed between the onset of the general persecution(23 Feb 303; Lact. De mort.pers. 12.1), for which Lactantius consistently blamed Galerius (De mort. pers. 11.3-8, 14), and the rise in the west of Constantine(25 July 306),78cast here as the heavenly king rescuing the persecuted. (Diocletian had retired in 305, leaving Galerius as Augustus in the East.79) The Divinae institutiones also accuses Galerius of commanding "fire to descend from heaven": this charge refers to a palace conflagrationthat Lactantiuselsewhere charged the Caesar with setting and blaming on Christian arsonists as he tried to intensify the general persecution (De mort. pers. 14).80 Regarding Galerius, Lactantius is clearly a hostile source, bent on claiming that this emperor engineered the persecution. Nevertheless, his allusive account harmonizes with that of Gelasios: the talking "image" once again suggests Daphne.81Moreover, since the force of the oracle allows the emperor-as in Gelasios' account-to influence others and initiate the persecution, the date is probably 299-the onset of the persecution from Lactantius' point of view.82 Where LactantiussupportsGelasios' story of Galerius and the oracle, Eusebius' Historia ecclesiastica, written in 306 and revised in the 320s,83corroboratesthe bishop's characterizationof Theotecnus as a man with a long history of anti-Christianactivism. This portrayalof Theotecnus occurs as Eusebius begins to describe the steps Maximin Daia (311-13) took toward renewing the persecution after Galerius' deathbed edict of tolerance (311): first, six months after Galerius died, Daia prohibitedChristians from meeting in cemeteries (Hist. eccl. 9.2); "next,"Eusebius charges, "throughsome
76. For his similarly apocalyptic treatmentof Diocletian, see Div. inst. 7.16 and Digeser, Making of a Christian Empire, 149-50. 77. Anon. Val. 2.2; Eutr. 25; Pan. Lat. 9.21.1; Sextus Aurelianus Victor 39.34-37. 78. Barnes, New Empire, 69. This is a date that Lactantius would certainly have known, whether he was still in Nicomedia in 306 or had joined Constantine shortly thereafter;see E. D. Digeser, "Lactantiusand the 'Edict of Milan': Does It Determine His Venue?"Studia Patristica 31 (1997): 287-95. 79. Lactant. De mort. pers. 17-19. 80. Although his argument-to my knowledge-has yet to be published, Oliver Nicholson was the first to suggest that Lactantius' apocalypse might be veiled commentary on events of his own day. Nicholson, however, thought the prophetic statue necessarily meant that the emperor in question was Maximin Daia, referringto the famous incident with Zeus Philios, described on p. 72 below. Nevertheless, Lactantius' reference to forty-two months of power effectively excludes Maximin Daia from consideration: Daia did not achieve the power to persecute until Galerius' death in early May 311 (De mort. pers. 35). Forty-two months after that brings us to November 314-well after Daia's 313 defeat at Licinius' hands. Likewise, forty-two months after Daia's accession to Caesar on 1 May 305 (Lactant.De mort.pers. 19.1) brings us to 1 Nov 306-a date with which nothing relevant is associated (Barnes, New Empire, 63-66). Moreover, Lactantius complains that Galerius set fire to Scripture(7.17.8), which was a feature of the first tetrarchy's edicts (Euseb. Hist. eccl. E 8.2.4), not Daia's (Euseb. Hist. eccl. 9.2-3). 81. If, when Lactantius says that Galerius "iubebit... solem a suis cursibus stare et imaginem loqui," he is describing two aspects of the same event, then the talking image belongs to a solar deity; see nn. 72 and 75 above. 82. See n. 2 above. 83. For the date, see R. W. Burgess, "The Dates and Editions of Eusebius' Chronici canones and Historia ecclesiastica," JThS 48 (1997): 471-504.
72
ELIZABETH DEPALMA DIGESER
mischievous men, he sends ambassadorsto himself againstus, having incited the citizens of Antioch to ask him to command that no Christian inhabit their town... and to contrive that others suggest the same thing" (9.2).84 Eusebius claims that "the originatorof all these things" was Theotecnus, "a creatureof Antioch" who not only "administeredthe affairs of the city,"but was also a bit of a rascal and a miracle worker. After this summary of the events leading to the Christians' exile from Antioch and other cities and after introducingTheotecnus, Eusebius then explains how the anti-Christian activist came to be involved in the exiles (9.3). "This man," Eusebius resumes, "having served in the army many times against us (and in various ways having zealously caused our being huntedout... as if we were some unholy thieves), having contrivedeverythingin orderto slanderus and make accusations against us, and having become the cause of death for countless people, at last sets up a certain statue of Zeus Philios with some magical tricks and sorceries,"after devising the requisite rites, initiations, and purifications for the cult. According to Eusebius, Theotecnus "also used to show off his knowledge of portentseven to the emperor,throughthings which he supposed were oracles,"an activity that the bishop does not directly connect with the statue of Zeus and its cult. Having done all this, Theotecnus then "arouses the daemon against the Christians,"saying that the "god ordered the Christians (as being inimical to him) to depart."These events, accordingly, encouraged "all the other officials in the cities under the same government"to "do the same" (9.4). "And,"Eusebius concludes, "as the tyrant by a rescriptdeclared himself well pleased with their measures, persecution was kindled anew against us." ReadingEusebiuscarefullyindicatesthatTheotecnus' anti-Christianactivities predated his involvement in the Antiochene exiles. Indeed, Eusebius accuses Theotecnus, not only of having "served in the army many times of Christians to be against us," but also of having caused scores (Lupoiotg) huntedout and killed (9.2-3). Since these activities preceded his erection of Zeus Philios and it is only through the statue's oracle that persecution resumed in Oriens (9.3), these deaths cannot be associated with Daia's per84. Euseb. Hist. eccl. 9.2: ETa 8td Ttvov novqp&vdvSpCovcat6tbEauTCKaO' l pACv rpOPEo3PEU'Eat, tOi6 V ttpM soCzOattnartpi{a AvrtoX•yv nokfita 7capoptli?oag !i trb~cp6aipogttva XptottavCovtfv abl'&)voiKSeV ta to 6 v fv Kai aU g p•yfioFn8Tops~ rap' at yTUXEsv6&t&•tat, Eztpou tbv 6tOropaLkaiv 6Staipd?aaO•ato rdvr i a6tfg AvXtoysiaq ftrp6ozeat OE6ZTcEvog, 8E1tv6 Kai y6rl77Kai novlpbOq 6vilp Kai rtov 6pxryb6 in' t 8 86KSt ta 3: 6' hoyst6eotyv &kX%6Tptpoq' KOta'd Tiyv rn6Xtv. oiv oztog Ka' ilpAivorpartpoa•vuloiaq retao1ra EKpuydUv Oplpeoat 8t&unou8ilg pjfpag voaiouqg reuodPCvogKai idvta Tp61tov tocg flLEsTpoutq sp o('6 Ztvdtg t rrE L KiC Kai Gtapoki xlnOtTlPVO;g t KaZrlyopitiQ KaO'?lro)v Oav0tvou 8fk too p•epCalXTvljpCivog, at ip qIoot; ln.vta Kai yorl0tsiTg at [note the ptupiot; yEyovo0g, TEklEtfovE8o'6v to AtbO •otNiou payyavoicttg rtoiv of aorist and followed the string perfect participles by present passive], EhEXdq;E &vdyvougaktt KaiacutE irtvoioaSg,p?ipt Kai PaotGtog Tilv zEpa(Xtiav St' jovf86KEL o10etg doakktpritoug 04ayioTouo Ka0appLobU Ot Kai 61 Kai ro KocaKEig ti Ka0' filovVil ro6 TOneyoipot Kart XptoXpalo)tOv Kpatobvvo v byfiCoteiKvuto. Kai TOV ttav~ d8aipovaKai toA6v bvil5hKE40oat (Potv iXspopiout tAVLPiTilV tn6ktvd&yprov lts) x6oosEq av a6ftC 4: Uf Xptatoavoi d6rshkdarto.Toc6tp obg fX0pobq rtpo6tpKtKaTd yvc0tplvrtpdoavt navtoEg oi 0otMoitiov TgfEt Tri fv 6pplrvrat yfppov rnotojo(aoatrpoaoptki; rdt; bxrb a•6tilv apXilv rtn6t; oiKo0vTCg Tilv 61poiav eovat tozto tyv Kar' fnapyiav f Yyp6v(v KOaiToOT' a6t OS tanpadpao0at troit Ipaototha ouvewpaK•it0v •irlK6ot; ov ivl Kai akt Tovtoit TOD St' &vtypacpilg opVtatc toirto03phrlK6otyv " •rnTpop(aty dto va rwVE6oaVTOt K(a0' pdvvou ai~t; Ek i• rapXifq 6 8toypo6q. io•gOv&ve0kopyoto
THE GREAT PERSECUTION
73
secution. As a y6rlg (probably Eusebius' characterizationof a theurgist) Theotecnus had long been associated with oracles in Antioch;85before he erected the statue to Zeus, this activity must have found him at Daphne. Consequently,Eusebius supportsGelasios' account of Theotecnus' theurgic role at Daphne prior to the purge of Christians from the army. Gelasios, Lactantius,and Eusebius thus add importantdetails to Constantine's account. Not only do they indicate that Galerius was the first to learn aboutthe Daphnic oracle, but they have also linked the prophecyto Antiochenes like Theotecnus who were hostile to Christiansand involved with divination. 86 Over time, Christianstended to suspect that people like theseincluding the emperorGalerius-had even engineered the persecution. Corroboratedby Lactantius and Eusebius, Gelasios thus confirms and supplements the argument,based on Constantine's report, that a prophecy of the Daphnic Apollo in 299 motivated the emperorsto persecute Christians,first within the army and court and then in the population at large. HARMONIZING Two ACCOUNTSOF 299
An oracle of Apollo is not, however, the prophetic crisis customarily associated with the purge of the army and court in 299. Rather, according to Lactantius'Divinae institutionesandDe mortibuspersecutorum,Diocletian's measures in 299 were triggered when Christians interfered with his ability to take the auspices. The incident, according to the latter text, occurred "while Diocletian was busy" in the regions of Oriens (in partibus orientis), or more precisely Antioch (De mort.pers. 10.1, 4).87 He and Galerius were "sacrificing cattle and looking in their entrails for what was going to happen" (De mort. pers. 10.1),88 when some imperial attendants,who "knew the Lord and were present at the sacrifice,"put what Lactantius calls "the immortal sign" on their foreheads. "At this," Lactantius claims, "the daemons were put to flight and the rites throwninto confusion."The haruspices were so "agitated"at the lack of the "usual marks,"he says, that they "repeated the sacrifice several times" to no avail (De mort. pers. 10.2). "Finally," Lactantius reports, the "masterof the haruspices" claimed-either "throughsuspicion" or "on the evidence of his own eyes"-that the sacrifices were "not yielding an answer"because "profanepersons were present"
85. See n. 68 above for Theophanes' use of the same term for Theotecnus. 86. Another was probably the vicarius, Sossianus Hierocles; cf. Barnes, "Sossianus,"246. Even though when he wrote this article, Barnes dated the army purge to 302, not 299, Hierocles was just as likely to have been an anti-Christianagitator in Antioch at this earlier date: Barnes, "Sossianus,"243-45; E. D. Digeser, "Porphyry,Julian,or Hierokles?The Anonymous Hellene in MakariosMagnls' Apokritikos,"JThS53 (2002): 466-502, at 486, 499-500. 87. Barnes, New Empire, 63, 55, for the Antioch setting. 88. Galerius' presence is indicated by the reference to the domini sacrificing in the parallel account of Div. inst. 4.27.4-5: "4 Cum enim quidam ministrorumnostri sacrificantibus dominis adsisterent, inposito frontibus signo deos illorum fugaverint, ne possent in visceribus hostiarum futura depingere. 5 Quod cum intellegerent haruspices, instigantibus isdem daemonibus quibus prosecant conquerentesprofanos homines sacris interesse egerunt principes suos in furorem, ut expugnarent dei templum seque vero sacrilegio contaminarent,quod gravissimis persequentiumpoenis expiaretur."
74
ELIZABETHDEPALMADIGESER
(profani homines interessent) at the ceremonies (De mort.pers. 10.3). In the Divinae institutiones, Lactantius describes the same incident, saying that "when some of our people who were ministers were standingby the emperors as they sacrificed, the emperors' gods were driven off; this happenedbecause these ministers placed the sign on their forehead so that [the gods] could not inscribe the things to come in the viscera of the victims." But in the next sentence Lactantiusadds an importantdetail: "Andwhen the haruspices understood(intellegerent) this, with the same daemons to whom they sacrifice inciting them (instigantibus isdem daemonibus quibus prosecant), complainingthatprofanepeople were takingpartin sacredthings, they drove their emperors into a rage" (Div. inst. 4.27.5). Accordingly, Diocletian immediately orderedhis courtiersto sacrificeor to be flogged. At the same time, Lactantius continues, he "sent letters to commanders"(datisque ad praepositos litteris) orderingtheir soldiers to sacrificeor face discharge(De mort. pers. 10.4). Lactantius' account of the events in 299 seems quite different from what I have reconstructed from Constantine and Gelasios. If the prophecy of Apollo that the latter two record occurred in 299, as seems likely, and led to Diocletian's purge of the army and court, can the two accounts be reconciled? The key to understandingwhat happened at Daphne in 299 is to realize that all of our sources emphasize different details. For example, in his De mortibus persecutorum, Lactantius has preferred to focus on the power of the Christiansign to drive away "daemons,"foreshadowing Constantine's effective use of this sign in his 312 battle against Maxentius (De mort. pers. 44). Constantine, conversely, is intent on linking Diocletian's decision to persecute with his ultimate demise. Articulatedin 324, Constantine's account helps to justify his recent defeat of Licinius whom he had accused of betrayingthe Christiansin his domain (Euseb. Vit. Const. 2.27-29). The second point to realize is that, despite being the shorter description, Lactantius' account in the Divinae institutiones actually follows Constantine's quite closely. What is likely to have happened in and near the Antioch palace in 299? From Lactantius' De mortibuspersecutorum it is clear that one day, when a mixed group of his courtiers (Christian and pagan) and his Caesar Galerius were in attendance, Diocletian slaughtered some cattle.89Lactantius also says here that the Christiancourtierscrossed themselves-perhaps surreptitiously-during the sacrifice and that the haruspices could not see any prophetic marks on the entrails (De mort.pers. 10.2-3). If some haruspices had seen the Christiancourtiersmake this sign, could their observationhave affected their ability to read the auspices-even after repeated efforts (De mort. pers. 10.3)? In the Divinae institutiones Lactantiuscalls these Christians "ministers"(Div. inst. 4.27.4). Given the status of these men, the diviners might have been reluctantinitially to identify them as the source of their anxiety, especially if the Christianministers were Diocletian's appointeesas is likely. What to do? In the Divinae institutiones, Lactantius says that 89. Lactant. De mort pers. 10.1-2; Div. inst. 4.27.4. For text see nn. 18 and 88 above.
THE GREAT PERSECUTION
75
"when the haruspicesunderstood"that the ministers' making the sign of the cross had "drivenoff" the emperors' gods, they complained about the "profane people" at the sacrifice and drove the emperors to persecute. I would suggest that the haruspices came to "understand"the failed auspices after the oracle of Apollo had complained about Christians interfering with his prophecies. Indeed Lactantius in this text says that the "same daemons to whom they sacrifice"incited them to drive the emperorstowardpersecution by complaining about the "profane"people. From Gelasios' account we know that Galerius was sacrificing when he received an oracle that "stirred up the persecution."And from Constantinewe know that an oracle of Apollo preceded the first efforts at persecution. Accordingly, I would argue that soon after the auspices had failed, Galerius and others went to Apollo's temple at Daphne-a site perhaps more famous for its statue's size and beauty than its oracles, but one long associated with Pythian prophecies. With Theotecnus,the anti-Christianattendantor theurgist,in the innerchamber (antron), Galeriuseither was simply sacrificingwhen the statue appeared to speak, or he specifically asked the god in the guise of his statue why the auspices had been mute. Constantinepreserves the resulting oracle, at least in part. Perhaps instead of blaming problems in divination on "the righteous" as the ChristianConstantineput it in 324 (Euseb. Vit. Const. 2.50),90 the oracle targeted "profane"people (profanos homines), Lactantius' term for those tied to the failed auspices (De mort.pers. 10.3; Div. inst. 4.27.4). News of the oracle at the palace had two results, I suggest: Diocletian-and others-would certainly have wanted to identify the profane or "righteous" people hamperingdivination (Euseb. Vit. Const. 2.51); and the chief haruspex, who may alreadyhave seen or at least suspected that the Christianministers had crossed themselves, now "understood"that Christians could be associated with problems in divination. In other words, emboldened by Apollo, Lactantius' "instigatingdaemon" (Div. inst. 4.27.5), the haruspices now blamed the Christians for the failed auspices (Lactant. De mort. pers. 10.3, Div. inst. 4.27.4). It is easy to see how Diocletian, always a keen "investigator of future events" (Lactant. De mort. pers. 10.1), would want to act quickly and forcefully. The prophecy, after all, in referring to divinations from the tripods in the plural (Euseb. Vit. Const. 2.50), called into question not only the recent auspices but all the oracles, including Delphi and Didyma, where priestesses conveyed Apollo's messages from this perch.91 And so, Diocletian wrote the first edicts "of blood and venom" (Euseb. Vit. Const. 2.51), calling for his Christiancourtiersto sacrifice or be whipped and for Christian soldiers to do likewise or face discharge (Lactant. De mort.pers. 10.4). As a man who trustedChristiansenough to make them his courtiers (Lactant. De mort. pers. 10.2, Div. inst. 4.27.4), Diocletian may have felt that these measures were enough. According to Lactantius, however, Galerius clearly thought otherwise. We do not have to accuse the Caesar of having manipulated the oracles to his own advantage. But clearly, 90. Fortextsee n. 13 above. 91. See n. 39 above.
76
ELIZABETHDEPALMADIGESER
he was deeply affected by what had happened to him at Daphne: for he continued to seek more repressive anti-Christian measures, and his zeal led Christians ultimately to believe he had manipulatedthe system to produce the outcome he desired (Lactant.De mort.pers. 10.6-11.6). Finally, in Februaryof 303, Diocletian, caught between his own inclinations and the pressure of his junior colleague, turnedto Didyma to resolve the issue (De mort.pers. 11.7). Constantinehad claimed that Apollo's priesthood was bemoaning the Christianisationof Roman society (Vit. Const. 2.50), and at Didyma their resentmentwas probably exacerbatedby the Christiansquatters that Polymnia Athanassiadihas identified in the temple precinct.92Thus it is not surprisingthat the oracle from Didyma gave the answerthatthe Caesar, at least, wanted to hear. Consequently, Diocletian immediately issued the edicts that targeted the Christianpopulation at large.93 As Constantineput it, "thepower of the Pythianoracles"(Vit. Const.2.54), both Daphne and Didyma, held sway throughoutthe reigns of Diocletian and his eastern successors, abating only with the death of Maximin Daia in 313. Nevertheless, both sites would pay for their involvement in the persecution. Not only did the emperor Licinius see to it that their priests and attendants-including Theotecnus-admitted under torture to having "invented" prophecies calling for persecution (Euseb. Hist. eccl. 9.11.5-6, Praep. evang. 4.2). But during the reign of ConstantiusI 1(337-61), Christians brought martyrs' relics into the temple precincts at both sites, in an apparenteffort to silence the oracles. Constantius' successor, Julian, had, of course, blamed his mantic difficulties at Daphne on Babylas' nearby relics. When Julian subsequently learned that relics had also been moved to churches under construction near Apollo at Didyma, the emperor "wrote to the governor of Caria,commandinghim" to burnthe finished structuresand to raze "the houses of prayer which were incomplete" (Sozom. Hist. eccl. 5.20).94 According to Glanville Downey, Babylas' burial at Daphne during Gallus' administration was the "first translation of a martyr's relics recorded in our extant sources."95If so, the burials at Didyma cannot have lagged far behind. It is no coincidence that the first two pagan sites that Christians attemptedto cleanse with their relics were those whose prophecies spurredon Diocletian's persecution.96 92. Athanassiadi, "Fateof Oracles," 272-74. 93. Lactant. De mort. pers. 12.1-15.7; Euseb. Hist. eccl. 8.2.4. s 94. Kai Tz pyv 8)6E vvtj cs otLsat 6t, 5K tyov tv pciptopaBa3av nuO68, Adspvr ou(idvtv 6 pat1etk86, 9ti EXv"y S pcEvoq ts plaptpTpwov EiKtipiouq otKouq Elvast tlo(Tiov toO vaob AtSuatioou An6XXo•vo;, npb TqijMtXilrtoiotiv, iypaWestf•lyes6vt Kapiaq, eiav h5posp6v?reKairpsnpeav isp&viXouat, 7cupiKazaEiRi ?jpcepyd9Fat ti iK avaoKadsat. cpkhia" 95. Downey, Antioch, 364.oiKo•hopolafaa, IdaOpov 96. From what we know, Constantineleft the site unmolested.According to Lane Fox (Pagans and Christians, 671) only six temples "areknown to have suffered in his reign."These are: Mambre("a site of great holiness in the Old Testament");a shrine of Aphrodite ("on the site of the Crucifixion and the Holy Sepulchre in Jerusalem");a temple at Aphaca ("an offensive Phoenician centre of sacred prostitution");and "[a]t Aigai, in Cilicia, [Christians] are said to have razed the shrine of Asclepius." At Didyma, Lane Fox observed, "Christiansseized a prophetof Apollo and had him tortured,as also at Antioch." It should be noted, however, that these last two examples do not involve the destruction of temples, but the punishment of temple personnel. Moreover, according to Eusebius, those punished in Antioch were those associated with Theotecnus' Zeus Philios. Euseb. Vit. Const. 3.25-26, 55-56, Praep. evang. 4.135C-36A.
THE GREAT PERSECUTION
77
Setting Constantine's oracle in its propercontext, then, clarifies a number of issues related to the so-called Great Persecution. First, it helps explain why the attitudes and behavior of Diocletian and Galerius were so different between 299 and 303. Given Diocletian's heavy dependence on divination, the Daphnic oracle may have convinced the emperor in 299 that Christians,whom he generally trusted,should be quickly punishedfor somehow compromising his relationship with the gods. Galerius, however, witnessed the Daphnic oracle firsthandand was much more profoundlyaffected by it than was his senior colleague. One does not have to accept Christian allegations that the Caesar engineered the whole affair to see that Galerius, who is not known for having Christiancourtiers, derived from this experience a strong conviction that Christianitymust be repressedif the properrelationship with the gods was to be preserved. Accordingly, although he was Diocletian's subordinate,first alone, then through advisors, and even after the first general edicts of 303, he lobbied his colleague to suppress Christians ever more intensely. Next, the account of the Daphnic oracle, together with the responses of Galerius and Christian authors to the events of 299, suggests that the period between the purge of the army and the general edicts should be considered part of what is now called the Great Persecution. Authorssuch as Eusebius and Lactantiusnot only saw the expulsion of Christiansfrom the army as the true beginning of their ordeal, but it is also clear that the oracle triggering this incident motivated Galerius' sustained effort to push repressive measures ever further.Finally, the story of the oracle affords a rare view into the social and religious tensions of the early fourth century. Clearly the Christians attending Diocletian's auspices are dramaticevidence for the extent to which Rome's eastern elite had become Christianizedby 299, a development that explains Christianefforts to blame the persecution on a wide-ranging pagan conspiracy. At the same time, it is easy to see how more traditionalpeople, Apollo's priests and attendantsin particular,could have feared the presence of Christians at their hallowed rites. Without stooping to accuse these traditionalistsof conspiracy,it is not hard to imagine that such anxieties might affect how they interpretedthe ambiguous signs by which they believed the gods communicated their intentions. This is the context in which the pagan and Christiantexts of this era must now be read and interpreted.The Christians,of course, ultimately won this contest and so the surviving accounts are theirs. That fact, however, should not blind us to the presence of competing voices and claims before the outcome was clear. 97 McGill University
97. I would like to thank the Social Science and Humanities Research Council of Canada and McGill University for funding this study. I am also grateful for the help of Hal Drake, Peter Digeser, Janet Fuchs, CourtnayKonshuh,and Olivier Dufault, as well as this journal's two anonymousreviewers. I presenteda version of this article on 25 May 2002 at the North American PatristicsSociety conference in Chicago, Illinois.
NOTES AND DISCUSSIONS IMPERSONALAND INTRANSITIVE EITIIHMAINEI From the fact of their derivation from orltcaivet,the verb tirtolrjai'vetand its related noun inrtorlpciaca would seem to imply significationin some fundamentalsense. But these two words have departedfrom the sense of their root word, and developed specialized technical meanings in two classes of ancient scientific texts: astrometeorological and medical texts. Astrometeorological texts are those which deal with predicting weather through the observation of the annual risings and settings of the fixed stars. This is closely related to the kind of practice we see in Hesiod, where times for planting, harvesting, and sailing are linked to specific annually occurring astronomicalphenomena, for example: rlXtd&cov At•ayeveovFytr•tTopEjevdov iPXso0' dtp'YroU,p6roto t8 BuoopevdaOv.
By the third century B.C.E.,and possibly earlier, such practices came to be codified in a class of text known as parapegmata,2 which laid out significant stellar phases (the annualrisings and settings of the fixed stars) for the whole year, and correlated these with weather changes, bird migrations, and more. Although only marginally familiar to many classicists, parapegmatawere written or quoted by a wide range of ancient authors, including Ptolemy, Aratus, Columella, Vergil, Ovid, and Pliny. In Greek astrometeorologicaltexts, 'ncttOlpa-wcit is one of the most common verbs even the most but to understand it and its related noun common), (maybe attempts have suffered from the to the root sense of significaattempts preserve cntorlglaoia tion.3 I will argue that the Greek verb cntorCpativet and the noun ointoipaia have specialized technical uses in the astrometeorologicalliterature,where the verb (both intransitiveand impersonalin Greek) means "thereis a change [in the weather],"and the plural noun means "changes [in the weather]."They thus have nothing to do with an idea of signification. As a parallel, a look at the medical use of the noun nttorltaoaiawill show that it too has departedfrom an idea of signification, implying instead the access of a fever, which, it should be noted, is not just the onset, but also the duration of the symptomatic phase of the disease. This parallels the astromete-
1. "At the rising of the Atlas-born Pleiades, / begin the harvest, and you should plow when they set"
(Hes.Op.383-84).
2. For the details of what parapegmataare and how they work, see, e.g., Evans 1998; for the third century as the terminus ante quem, see Lehoux 2000. 3. In particular, has found its way into the German literature as the word Episemasia (pl. tsatrlaoa' contrasted with Phasen (the annual appearances or disappearances of fixed stars) Episemasien), which is and functions as a very general term for the weather indications (Witterungsanzeichen,i.e., the "significations") of the astrometeorological texts (see esp. Rehm 1940; Boll 1909).
Permissionto reprinta note in this sectionmaybe obtainedonly fromthe author. 78
NOTESAND DISCUSSIONS
79
orological use by referringto an observable thing, not just the sign of an unobservable thing.4 ASTROMETEOROLOGICAL EIIIHMAINEI The astrometeorologicalliteraturein Greek contains repeateduses of the verbiitarlpativt in an unusual construction. The verb, primarily supposed to mean "indicate," "mark,"or "give signs" by LSJ, is used intransitively and without an obvious (pXcrat subject, as in Geminus, 1 Cancer:a' 4ItpFp' KopKrivo? Kaktkinnic• t Optvai Ka LSJ offers the following definition of &vatXX•itv" this use: Kai porti ttorlntaaivet.5 III.intr.,givesigns,appearas a symptomin a case,... ; ... of pubertyshowitself,... ; of weathersigns,indicatea changeof weather,... ; impers.,symptomsappear... Note that the definition as it pertains to weather signs implies that the verb is intransitive, but personal. But there is no obvious subject in the example from Geminus. Unless we supply a subject, such as "this stellar phase" or "this day," we cannot, strictly speaking, use LSJ's intransitivepersonal definition.6 Rehm treats EnrtorlTaivet as an impersonal by translating it as "Bei einer Phase gibt es ein Zeichen."7But a sign of what exactly? He argues that in the parapegmatic literatureEttomrLpaivet implies something about a change in the weather, which I think is correct, but he maintains the notion of signification as being always at the heart of the meaning of irrtarl1paivEt. Pfeiffer also sees signification as central: he translates ttOarlpativet as "liMiterwarten,zeigt an, deutet auf."8Pritchettand van der Waerdentranslate intorllptaivct with the verbless phrase "sign of weather."9 But the astrometeorologicaluses of nttorltlaivet are best understood as both impersonal and intransitive. There is no obvious subject stated, and implied subjects are, in the main, unsatisfactory.Look again at Geminus, 1 Cancer:a' llp't Ka•XHere there is Olptvat- Kai Kapicivog apXzTat dvaT~zXEtv rponai inirtt tlntoalC.taiv.10 no grammatical subject, although there is a candidate for a logical subject: the beginning of the rising of Cancer. But this interpretationdoes not work for passages such as 16 Leo: &v6i zT il' tflepqp &~ittorl vaivcst,11since there is no stellar E5664q, 4. The close connection between astrometeorology and medicine in antiquity may imply this parallel is no mere coincidence. See, e.g., Aetius of Amida's Tetrabiblos(excerpt published in Wachsmuth 1897), or the Quintilius parapegma(Boll 1910). 5. "On the first day: According to Callippus Cancer begins to rise; summer solstice, and EntorlaifvsEt" (Geminus, p. 210, 14-15). All Geminus citations are from Manitius 1898. 6. The example cited by LSJ, TheophrastusDe signis 10, does seem to be both intransitive and personal, and in this respect, Theophrastus' use of irntaolpaivetdoes not fit that of the astrometeorological literature. Theophrastususes i7rtorlpaivetintransitively but personally in De signis exactly where we would expect aorlpaivetwith a direct object. Indeed, he runs the two constructions in parallel: Kaili~v intizb r'iXtovv prl npooirl. i valpovorlaivet. TravIptgyvyrlat, ictalllcivet (Theophr.De signis 22). &aMp ....De ventis 38.225-29. Signification also seems to be present in Epicurus' use of the Compare Adamantius noun in the Letter to Pythocles. See Diog. Laert. 10.98.8; 115.9; 116.1. 7. Rehm 1940, col. 181. 8. See Pfeiffer 1916, 85-86. Pfeiffer furtherclaims that rntoarlpaivet was sometimes writtenjust as orTpaivet, but all but one of his examples are easily dismissed as scribal errors. His remaining example, from Lydus De mensibus 4.14 (see Lydus De ostentis, p. 296.1, in Wachsmuth 1897) cannot be an alternate writing for the intransitive ilntorlaivet since olpivet is being used transitively, with ppoXd as its direct object. 9. See van der Waerden 1984, 105-6. 10. Geminus, p. 210.14-15. 11. "On the sixteenth day: According to Eudoxus irtorlgaiv't" (Geminus, p. 212.23). See also p. 214.5; Ptol. Phaseis, p. 15.10, and passim (Heiberg 1907).
NOTES AND DISCUSSIONS
80
phase mentioned that could serve as logical subject, and the dative construction rules out taking ti' ljIgpqas an implied subject. If, however, we treat these uses as impersonal, we can renderthe meaning of Ent7orl•aivet as "thereis a change [in the 6 &i•p. weather],"more or less equivalent to CLtarCptlXXt The attemptto maintain the sense of signification, reading~intorltaivit as "there is a sign that the weather will change,"does not fit the general natureof these astrometeorological texts. An astrometeorological parapegma typically correlates days with stellar phases and weather predictions. The predictions are, almost without exception, concrete: "it will rain,""therewill be a south wind," and so on.12In no case do the entries in parapegmatatell the reader to watch for vague and unspecified signs of any kind. Rather,they state that particularphenomenawill occur on particular days. Texts from outside the corpus of astrometeorologicalliteraturepoint to such an 6a interpretationas well. Stobaeus reports that Anaximenes thought that T&U i"ntorlin This the middle of a sentence ijXtov appears yiyvE~oOat latoicg 68t z'ov gt6vov.13 longer section on Anaximenes' beliefs about the stars in general, and by saying that the lcntorllacoiathappened because of the sun, Stobaeus would seem to imply from this that Anaximenes did not think that they came about because of the fixed stars.14 here as signs of some kind, then the sentence Now, if we interpretthe irntorllaaoiat is difficult to understand.Keep in mind what was being observed in astrometeorology: the horizon was watched at either sunrise or sunset for the appearanceof a particularstar. It was not the sun that was observed, although information about its position or motion could perhaps have been abstractedfrom the observation. So the phenomenon seen was simply the appearanceof a starclose to the horizon at sunrise or sunset. If we take •'ntorlgaciiato imply signification then we have the strange situationof the sign (by definition,somethingmanifest,observable)happeningby means of something not observed (the sun). We get aroundthis problem, however, by transhere as "changes [in the weather]."So what Anaximenes is saylating Enrtitopacriat is that changes in the weather are due to the sun, not the stars.15 ing Furtherconfirmationof the centralityof the notion of "change"ratherthan that of "signification"can be found in the earliest preservedastrometeorologicalparapegma, P. Hibeh 27, dating from about 300 B.C.E.There we find the following: ['Efeit]
v5 'bptCOv to3, 1 86''fljtpa t3, ToiWAvo63jto? •tvoncoptv7, i1 srtttFrlJPaivtVt 7tpb0q Tirv svcipactv.16
K7y iorlpjpicTa
KCa 6Rto Tabt
opri,
12. The few instances where predictions are more vague are characterizedby the following two traits: (1) they take the form of either (a) "a storm usually happens,"e.g., jptlpq AlPoKpitC .....ve~Lot XE•pL~ptot rzT Rnod.6 (Geminus, p. 218.15-16), or rarely, (b) "wind tends to blow," e.g., fljpqg ArljloKpi' .T... .ntqn(lE (Geminus, p. 218.16), and (2) they are frequently attributedto Democritus, one of the earliest rtveiv named sources for the parapegmatictradition. 13. Stob. Flor. 1.24.1k. 14. My argument here does not claim that Anaximenes actually used the word i7torpeaoiat (in which case he would be the earliest authorknown to have used it in reference to weather). All that matters to the present discussion is that Stobaeus himself used the word. 15. Such an interpretationdoes turn up in LSJ's treatment of this noun (it reports that in the plural, tntorlpaoia can mean "changes in the weather"),but I am at a loss to explain why this meaning is absent in their entry for the verbal form of this word. 16. "[Epeiph]23: Autumnalequinox; the night is twelve hours, the day twelve hours; festival of Anubis; and the river irntorlppivetwith respect to its rising" (Grenfell and Hunt 1906, vol. 1, p. 149, col. 12).
NOTES AND DISCUSSIONS
81
Here we note that, unlike elsewhere in the parapegmaticliterature,'7toarlaivet is being used personally. Grenfell and Hunt offer the following comment: "This entry 'the river gives indication of rising,' . . . refers apparentlyto the flood reaching its full height, which it usually does early in October. Epeiph 23 ... being the day of the autumnequinox, was probably Sept. 27."" It is clear that the entry does refer either to the end of the rising of the river, or to the beginning of its receding. But rendering the idea that the river begins to recede as "the river gives indication of rising" loses this sense entirely. It is almost as strange as if we were to translate ivCLtpo ptyac rtvci as "the wind gives indication of calmness." The solution of this difficulty lies in treating ~ntolrioaivcthere as we have in the other parapegmata,as fundamentally implying change, rather than signification. Our understanding of 6 would thus be: "the river changes with re1noTapig ~intorlPav.Et Ipbg t~lVy tvd3pastv spect to its rising," that is, either "it finishes rising," or "it begins to recede." Combining this with the other evidence we have seen, the best translationof ~intorlIaivwtin the parapegmatawould thus be "thereis a change [in the weather]." ASTROMETEOROLOGICAL EnIIIHMAZIA We have already seen an example of 7~torjctaoia meaning "changes in the weather" in the Stobaeus Anaximenes fragment quoted above. In astrometeorological contexts, this use is common. Take, for example, Geminus 17.2: Kai InpizovPyv zily yiv 7tpoStaX7rl7rtov,i~t ai ytv6~svacti•lrtotFtaciat il4ppov Kcai tvcusdcrzovtcUpi . yivovrat, Ei 68 rkielov ogo 6tatEivouotv ("the changes in clouds and winds that {jUVo occur happen near the earth, and do not extend very high"). His point here is that weather phenomena happen only below certain altitudes, and he refers us to a story (also found in the Pseudo-Aristotelian Problems)18that is supposed to show that there is no wind above a certain height, since the ashes of sacrifices made on high mountains remain undisturbedover the course of the years. Now because they are earthly phenomena, srntcrlpaciatcannot refer to the signs of the winds and clouds, since the signs are well beyond the height of the mountains. Another example comes from Ptolemy Tetrabiblos2.94: Tb 0~Pv oTv Kipto6 06c&KaTIrlpptovKa0"';hou 1)pv 'o-Tt tot TilV itrptcptviv ~7ntorlpaCiav 13povt•C6Sg Xaka?66gS("The zodiacal sign of Aries is, on the whole, thunderousor hail bringing, because of the equinoctial change"). This passage parallels Aristotle's claim about the rising and setting of Orion being dangerousand uncertainbecause of the change (jIrca3o&'ri)of season.19 This being said, a very few passages in astrometeorologicaltexts do seem to lean back towards the root meaning of ~rtoirltaoia as signification,20or even perhaps "astralinfluence [on the weather],"21and in other astronomicaland astrological contexts, i~rtarlptaoia frequentlyimplies significationor significance.22These differences 17. Grenfell and Hunt 1906, vol. 1, p. 156, nn. 168-69. 18. [Pr.] 944bl 1-20. Compare also Mete. 340b36. 19. Mete. 361b30. 20. See, e.g., Ptol. Tetr.2.98. 21. AdamantiusDe ventis 48.6-10. I lean toward the idea of "influence"here, since Adamantius talks about T6btl tntorlpaoiag Ipyov in the context of the oupindOctaand carpnvotabetween the stars and the weather, in which he says that stellar phases shape (ruxno6ot)the weather (47.20-25). 22. See, e.g., Ptol. Alm. 6.11.
82
NOTESAND DISCUSSIONS
in meaning point to a not-quite-completecrystallization of the noun as technical vocabulary, such that it retains a degree of homonymy. UNUSUAL AND EXTENDED USES OF EIIIHMAINEI
with a dative object: one is a There are a handful of instances of F ntorTtpaivct 6'8act Democritus fragment from Ptolemy (ArlPOKpi'Tr KCiai hIrtoIrlaivwt atvtotgo),23 and a third inanother is the Miletus fragment456B ([Kai aXa•hdrt),24 ~int]orltpaivct . . . stance is in Wachsmuth's edition of Geminus: ptXEi (AtPOKpiZt rttoarllaivetv ~ av0t9S a ijSaCt1 , 6tEpatog zir od).25 The most ppov'rfKtai&oZpartIK1i6 "upL5" Pri "TherewillOtbe a in the is thatthesedativesarecomitative:
change plausiblereading weather,withhail,"thatis, "includinghail."26 Indeed,this is exactlyhowthe dative hdv &dyvitp is beingusedin the following:Kact Toiq dTpdvi'rl,6 paotX9E6q [AztptXf•q] less likely,is Anotherpossibility,albeitone I consider ipXovraq atbzobpovEt30et.27 thatthesearedativesof respect:"Therewill be a changein the weather[in particular]withrespectto the windsandrain." A second unusualuse of rntorlptaivet occurs in Geminus 17, where he employs the
he says: verbin an extendedsense.In a passagecriticizingthe parapegmatists, V tfI R e0P' eoi ? 6s0est tfYl flW.t~paq tpsig fI t aatpa •taisolCpve TrOXCKtSg rttofl _ Tpou, aoz6I' "T rtolpactoiav rpb pFpsCpov tcrodp0ov.28 rtpoXal3e •ciyV
o-
I translatethis as: Often[theparapegmatist] has markeda changein theweatherwiththerisingor setting of a starthreeor fourdays too late, and sometimeshe has anticipatedthe changeby fourdays.
Herethe subjectof hirnCilgtve is carriedoverfromtheprevious (theparapegmatist) sentences,and the verb itself seems to meanthathe has put down or markedan 23. Ptol. Phaseis, p. 63.22-23 (Heiberg 1907). 24. Diels and Rehm 1904, p. 110, left side, 4. Diels and Rehm's restorationof [Kcarint] does not seem unlikely. oarllive Xakdsrt on its own would be even more puzzling. Another instance of a dative construction occurs in a restoration by Rehm of a dative in the Miletus II parapegma: [bUt K]eai ?06tae irtTitlXt K(ta' (Rehm 1904, 756), but given the state of the text, this reading is ictorl[paivEt vo'o]t ET6K•rjpova highly uncertain. 25. Geminus, Sagittarius 16, in Wachsmuth 1897, and followed by Manitius 1898. The MSS all read ATpoKpi'T.. .rnttorLpasivtyv cptsXf3pov~TvKai &orpcapeiv Kai iSat ~1 vip( j cpspp6(popa•cEni dT noX'kdi, but the accusatives here would be even stranger than datives (indeed, they would be unique). See Rehm 1940, col. 181. 26. Rehm thought that these datives following gcntoCrptOaivt are instrumental, meaning that the sign itself is derived from the weather phenomenon in the dative (Rehm 1940, col. 181; see also col. 183). But only three predictions in the parapegmataare in this dative construction, and Rehm fails to show why so few occurrences of rain, wind, or hail were signs of particularphases. Secondly, we have already seen that the common absolute use of ictoaipaivt frequently has no stellar phase (nor anything else) attached to it, and thus, if we apply Rehm's interpretationconsistently, these instances would be vague to the point of meaninglessness. 27. "And if [in the month of April] there is lightning with wind, the king will kill his magistrates" (CCAG3:48). 28. Geminus, p. 188.18-21. Manitius inserts a (ztq) after noXdK~tq8I, but I see no real need for it. He translates the passage thus: "Oft hat jemand drei oder vier Tage zu spat an den Auf- und Untergang des betreffenden Gestirns eine Voraussage gekntipft, bisweilen eine solche auch vier Tage zu friih angesetzt." (See Geminus, p. 189). Aujac, eliminating the rtg, offers the following ratherfreely idiomatic translation: "Souvent aussi les indications donn6es valent avec 3 ou 4 jours de retard sur le lever ou le coucher de l'etoile; a d'autres moments, les indications sont en avance de 4 jours" (Aujac 1975, 87).
NOTES AND DISCUSSIONS
83
beside or with the stellar phase. This use is a simple extension of the tntorrlpacaia normal astrometeorologicaluse of the nominal form rttorllaoia. EI-EHMAZIAAs MEDICALTERMINOLOGY
occurs in the medical literature.Galen A different specialized use of mttorltacia defines ntirlolaaoiaas follows: KatO 86'tiopotoilv i6i7Srp5zpov Tbv Kptjqi'g tnapo4u1toio 8' ab 'TtobivE Elsewhere, Xp6vov 6voCd6srtv.29 Ec60aotv 'bv 'najjl0tiav •ov he says:attkaT•i, .v. .30 and: &v 6voot6voty 6-Tav zaiq&paipCzTivrapotujCwotv, •aqE~ntaorilpctag Ev TzaiqEiopl3oaig Ji t~ntorlCtaoiaq According to this use, tSv TrupEWZTv, 6vodolouatv.31 refers to the onset (and in some cases, the duration)of the appearthen, ~intorlapaoia ance of the symptoms in a cyclic fever,32and not typically to the symptoms (signs) themselves. Elsewhere, ntorllpaoiais contrastedwith veoGtq,remission,33and means the "manifestation"of the disease, by which translationI wish to imply not simply a symptom abstractedas such, but ratherthe period of the disease during which the patient is feverish. Indeed, Galen even refers to the "symptomsof the rtomrll.taoia" (T& zi~ oCpia).34 I choose "manifestation"ratherthan "access" sim?8t5rlotaicia of the because ambiguity of the English term access, which can refer just to ply the onset of symptoms.35For our present purposes, what is importantto note here is that the idea of "signification"is, as in the astrometeorologicalliterature,absent.36 29. "I call the onset of a paroxysm the absolute first moment of stability, and this is the same as what is usually called the 7intoparicia"(Galen De morborumtemporibus liber 5). 30. "when, during the beginnings of paroxysms, which are called irntorloiat . . ." (Galen In Hippocratis aphorismos commentarii 2.1). ." (Galen In Hippocratis prognos31. "during the onsets of fevers, which are called i7rtolatoiatt.. ticum commentaria 2.4). 32. I exclude non-cyclic fevers, since Galen seems to think that continuous fevers do not have Av p~vyp Xosot; nto7rlLpaaia: ainuaotupveroI;at rotaUTrtEilofoai a'X trtot yivovra•t Xpig To6 Tpapivat- Ka•t o i 1torltaoia yivErEat Ka' o668va Xp6vov,6&X'odrt Ecg 8R ToO; hiTKo6 T6b vavrtrteTaov, b0' X0Ws;g tniv 6 (De differentiisfebrium 1.11). i8iw; o6voXog ouve~fig nopEZrb diFansp, Kaeko6•pevo; 33. E.g., in Galen De typis 3. 34. See De differentiisfebrium 2.7, Synopsis librorum suorum de pulsibus 16, De praenotione ad posthumum 11. 35. That more than just onset is often implied in the ancient medical literature becomes clear in the contrast between intaoipaoaianddvsot;, as in Galen's De typis, where diseases are classified according to v oi pv Eiot the duration of the intorl.taaia versus that of the ivsot;: dTiv oi 6• Kai atpdTot, ZThe•0v 8se6cpot" oi Kai P0Ev io'r•teg, oi & KIVO6pevotE ardktv oispv nEiotvhnoi, oi 6~ a6v0ctot. inp&ot lv oUv EioGv oi ltlCKpevaPv ExovTEg tiiv i ntoraTpaoiav, placpdv•6 ziV vveotv (Galen De typis 3). Moreover, duration is imGalen plied in such statements as KKatn ("duringthe time of the TbvKratppv Tg itntorlpLaoiag irttorlpaoia," TCv ("therewill be discharges during Synopsis librorum suorum de pulsibus 17); KaCTappi)nt ittorltpaetosv cs v oiv Tbv irpoztovXp6vov zTi the gntorlpaoia," Archigenes Fragmenta inedita 68.6 Brescia); Krazdt intorl.patoaq("duringthe beginning of the intorlpaoia," Anonymi medici Peri lycanthropias 16 Ideler); il dKpCLi] tf; irto1lpaaia; ("the peak of the intorllaoia," Aetius of Amida, latricorum 78.41 Olivieri); ?v zail ("in the beginnings of the intorlpaaia," Aetius of Amida Iatricorum 83.25 Olivdpeal; rT&v irttoplpaactCv ieri). The best way to keep the sense of these examples clear is to interpretintorlttaoia as durative, i.e., taking place over some nontrivial length of time. 36. This holds true for virtually all the Galenic uses of rntorlltaoia(the exceptions occurring either in discussions of seasons and changes in the weather, as at De diebus decretoris 3.7, or in his quotations from and commentaries on Hippocratic works), and for post-Galenic medical writers in general (e.g., Paulus Aegineta, Stephanus of Athens, Aelius Aristides, Aetius, Erotianus, Dioscorides Padanius, Alexander Trallianus, and AdamantiusJudaeus). This is not, however, generally the case in the Hippocratic corpus, with the notable exception of Hippocrates De articulis 67.16. In the Hippocratic works, a number of different uses of both the noun irntsolpacaiaand the verb 9i1tol(1paivtare evident, e.g., (1) "indicate"(synonymous with orplaivet) (De morbispopularibus 1.9.5, 7.46.6, and possibly De septimestripartu 9.4); (2) "symptoms appear" (De morbis popularibus 4.39.5); or just (3) "appear"(e.g., De semine 21.8, 44.3; De articulis 30.14, 41.36; and passim). This last meaning, that of "appearance,"is related to several other uses of inrtorlpaivetin Greek.
84
NOTES AND DISCUSSIONS CONCLUSION
It is apparent,then, that nrtoraivEt has some hitherto unrecognized meanings in the astrometeorologicalliterature.Its impersonaland intransitiveuse in the parapegmatic tradition,meaning "thereis a change [in the weather],"is quite differentfrom its use in, for example, Theophrastus (intransitive, but personal: "x indicates a change in the weather").This loss of the sense of indication or signification is not, in the however, unusual. This is exactly what happened to the noun cnttorlpacia post-Hippocraticmedical literatureas well. As a final note, I would point out that in the Latin astrometeorologicaltexts, we sometimes see the verb significat occur exactly where we should expect to see irntorltpaivvtin a Greek text.37I would argue that reading significat as a rather literal "thereis a change [in the weather],"will allow us to make translationof &ntotrLaivEt, a good deal more sense of such passages in the Latin texts, as well.38 DARYN LEHOUX
University of King's College
37. See, for example, Columella Rust. 11.2.5: III Kal. Febr. Delphinus incipit occidere, item Fidicula occidere, significat. So also Rust. 11.2.94: VIIIIKal. Ian. brumale solstitium sic Chaldaei observant, significat. In these instances significat is clearly intransitive, although not necessarily impersonal. More clearly impersonal uses are at Rust. 11.2.94: XV Kal. Ian. ventorumconmutationemsignificat. We may be tempted to take the date as an implied subject, but that will not work in Rust. 11.2.34, since the verb would then have to be plural: VII. Id. Apr. et VI et V Austri et Africi, tempestatemsignificat. In these last two instances, as in most (but not all) of the instances of significat in Columella, some or all MSS have added direct objects in the accusative (tempestatem,pluviam) and modern editors have favored including these direct objects where possible as well. I would suggest, however, that if significat is a literal translation of rntlorlpaivEt, then tempestatemsignificat may actually be a corruptionof tempestas, significat. 38. I would like to thank Alexander Jones, Brad Inwood, and CP's anonymous referee for their helpful comments on earlier drafts of this paper, and the Social Sciences and Humanities Research Council of Canada for funding this research.
LITERATURE CITED Aujac, G., ed. 1975. Gdminos: "Introductionaux phinomenes." Paris. Boll, E 1909. Fixsterne. RE 6:2407-31. . 1910. Der Kalender der Quintilier und die Oberlieferung der Geoponica. In Griechische Kalender, vol. 2, ed. E Boll. Sitzungsberichteder HeidelbergerAkademie der Wissenschaften philosophisch-historische Klasse. Heidelberg. Brescia, C., ed. 1955. Archigenes: Frammentimedicinali di Archigene. Naples. Cumont, E, et al., eds. 1898-1953. Catalogus codicum astrologorum Graecorum (CCAG). Brussels. Diels, H., and A. Rehm. 1904. Parapegmenfragmenteaus Milet. Sitzungsberichteder kiniglich Preussischen Akademieder Wissenschaften,philosophisch-historische Klasse 23:92-111. Evans, J. 1998. The History and Practice of Ancient Astronomy. Oxford. Grenfell, B. P., and A. S. Hunt, eds. 1906. The Hibeh Papyri. London. Heiberg, J. L., ed. 1898. Ptolemy: 'Almagest."In Claudii Ptolemaei opera quae exstant omnia, vol. 1. Leipzig. . 1907. Ptolemy: "Phaseis." In Claudii Ptolemaei opera quae exstant omnia, vol. 2. Leipzig. Ideler, J. L., ed. 1842. Anonymi medici: "Peri lycanthropias."In Physici et medici Graeci minores. Berlin.
NOTES AND DISCUSSIONS
85
Lehoux, D. 2000. Parapegmata,or, Astrology, Weather, and Calendarsin the Ancient World. Ph.D. diss., University of Toronto. Manitius, C., ed. 1898. Geminus: Gemini elementa astronomiae. Leipzig. Olivieri, A., ed. 1935-50. Aetius of Amida: Aetii Amideni libri medicinales. Leipzig. Pfeiffer, E. 1916. Studien zum antiken Sternglauben. Leipzig. Rehm, A. 1904. Weiteres zu den milesischen Parapegmen. Sitzungsberichte der kiniglich Preussischen Akademie der Wissenschaften,philosophisch-historische Klasse 23:752-59. . 1940. Episemasiai. RE suppl. 7:175-98. van der Waerden,B. L. 1984. Greek Astronomical CalendarsI: The Parapegmaof Euctemon. Archive for History of Exact Sciences 29:101-14. Wachsmuth,C., ed. 1897. Lydus,Johannes Laurentis: loannes Laurentii Lydi liber de ostentis et calendaria graeca omnia. Leipzig.
BOOK REVIEWS The Sleep of Reason: Erotic Experience and Sexual Ethics in Ancient Greece and C. NUSSBAUM and JUHASIHVOLA. Rome. Edited by MARTHA Chicago and London: University of Chicago Press, 2002. Pp. viii + 457. This book has its origins in a conference held in 1997 at the Finnish Institute at Rome as part of a series aimed at nurturingthe internationalcontacts of Finnish ancient philosophers. Since then, the volume has expanded in scope, with the result that the final product is a widely ranging collection of essays on ancient sexuality, which maintains, however, a welcome focus on philosophy. After the editors' introduction,the volume begins with the first of two contributions by David Halperin, which are characterizedby the analytical clarity and cool self-awareness that make his work on ancient sexuality so valuable. In the first essay ("ForgettingFoucault") Halperin responds to some remarkablytenacious misreadings of Foucault's History of Sexuality.One centralpoint is especially worthemphasizing, as it applies not only to Foucault but also, more or less directly, to Halperin himself, to the writer of this review, and to others who have written on ancient sexuality. Foucault "is not attemptingto describe popular attitudes or private emotions, much less is he presumingto convey what actually went on in the minds of different historical subjects when they had sex"; rather,he "is speaking about discursive and institutional practices" (p. 26). Halperin's second essay ("The First Homosexuality?") takes on the question of ancient "lesbians"and "tribads,"raising some serious objections to the methods and conclusions of BernadetteBrooten's Between Women (1996), to date the only monographdedicated to female homoeroticism in antiquity. In particular,as Halperin shows, Brooten's use of the key concept "sexual orientation" is open to criticism. The first of two pieces by MarthaNussbaum, "Eros and Ethical Norms,"constitutes a lively engagement with Stoic thought on the dual potential of eros as source of both blessing and harm. Nussbaum begins effectively indeed, invoking the famous "up-and-down"position on Attic vase painting and asking: when the lover touches his beloved's chin with one hand and his genitals with the other, "does the left hand know what the right hand is doing?" (p. 55). Her second essay does the great service of bringing to our attentionthe much understudiedRoman philosopher Musonius Rufus and his striking convictions that women, too, should do philosophy and that marriage should be viewed as a partnership.In both instances, Nussbaum observes, "Musonius draws on views already currentin his culture, but moves considerably beyond them" (p. 300). Nussbaum's first contribution in particularshould be read together with A. W. Price's essay on "Plato, Zeno, and the Object of Love." Price carefully reviews the various responses by various ancient philosophers to a basic problem: how to "relat[e] the educative to the erotic satisfactorily" (p. 171). His discussion of the faPermissionto reprinta reviewin this sectionmaybe obtainedonly fromthe author. 86
BOOK REVIEWS
87
mous Stoic recommendationto retain an eromenos until the unusually advanced age of twenty-eight (an example of how "they strove to weave innovation within tradition") is particularlyinteresting (p. 191). Some readersof both Nussbaum's and Price's contributionswill be distractedand others stimulated by their explicit concern, somewhat more pronouncedin the case of Nussbaum's essays, with contemporaryethical applicabilityand their readiness to offer judgments from the perspective of, say, present-day feminism or liberal philosophy. In view of Musonius' failure to criticize the unequal treatmentof women in Roman law (as in the lex lulia de adulteriis), for example, Nussbaum observes that "feminists may forgive Musonius for his silence, but they should not emulate his example" (p. 306), while Price observes that "a modern liberal philosopher who looks back at [ancient philosophers'] attemptsto define a distinctively philosophical love is likely to feel that they were all misconceived" (for "love, like life, is something thathappens,"not always conformableto theoreticalmodels), but "whatwe may be readiest to admire in Plato and the Stoics is a sensitivity to reality that illumines their ingenuity in theory" (p. 192). In "Aristotleon Sex and Love," Juha Sihvola emphasizes that the philosopher has surprisingly little to say on the subject, and concludes that Aristotle saw "little difference between same-sex and heterosexual relationships"(p. 218): both are understood in asymmetrical terms, yet both contain the potentiality for "a development toward the most perfect form of friendship"(p. 218). Aristotle's idealization of marriage (Eth. Nic. 1148b28) goes far beyond the productionof legitimate children, and in the same text (1157a3-12) the philosopher representscertain kinds of pederastic relationships as being capable of leading to lasting friendships. Stephen Halliwell's piece on Old Comedy as "institutionalizedshamelessness" (p. 124) is a thoughtful discussion that takes as its startingpoint the phallic song in Acharnians 263-65 and the related question, "Should we (Athenians) laugh with or at Dicaeopolis?" (p. 122), and then focuses on the notion of sexual communism played with in the Ecclesiazusae. In the end, Halliwell argues, comedy reduces sexuality "to a level of immodesty that is ludicrous ratherthan truly threatening,"but this in turnentails that the audience "mustbe to some extent complicit in the shamelessness of the performance"(pp. 135-36). In short, "the erotics of comedy should count as a special case of Dionysiac voyeurism" (p. 136). There follows David Leitao's "The Legend of the Sacred Band," which ponders the possibility that the famous Theban band of soldiers consisting of erotic couples might never have existed. Such skepticism is healthy but comes with the usual risks: there is relatively little about classical antiquity whose historicity we could not find ways of challenging. But Leitao wisely does not push the historical question too far, and, whether or not one is convinced, the essay provides in any case a helpful discussion of the rhetoricaluses to which the Sacred Band was put by ancient writers. K. J. Dover's brief but stimulating discussion of Anthologia Palatina 5.207, an epigram by Asclepiades on two women from Samos, includes a highly memorable formulation:Aphrodite, ratherthan being (as one sometimes reads) the patroness of heterosexuallove, is "a goddess whose province was genital friction"(p. 225). Maarit Kaimio discusses some "good wives" in Attic tragedy (e.g., Deianeira, Alcestis, Evadne), and finds hints of the possibility of "sexual activity and even erotic enjoyment" on the part of wives (p. 97). Especially helpful is Kaimio's discussion of the
88
BOOK REVIEWS
usage of various words for the marriagebed. J. Samuel Houser takes on the question of Dio Chrysostom's alleged condemnation of sexual practices between males, showing among other things that the orator's target is hedonism: he condemns both homo- and heterosexual acts pursuedfor the sake merely of pleasure. Eva Cantarella considers the famous incident involving Cato the Youngerand Marcia, skillfully filling out the cultural and legal background to its surprising transferal of a married woman to a new husbandin orderto provide offspring to him and to the res publica. Simon Goldhill discusses the interesting connections among "erotic viewing, culture, and power" (p. 394) found in certain texts from the late second century C.E., above all Achilles Tatius, Clement of Alexandria, and Tertullian. David Konstan brings our attention to texts ranging from Homer to Catullus to Lucian in which the possibility of alternative styles of masculinity (softer, oriented more toward reciprocity than domination) is raised, although Konstan rightly observes that these alternative models hardly constituted a revolution: "the result was not a valorization of sexual reciprocity as such but ratherthe recognition of a possibility that surfaced as inexorably as it was repressed"(p. 369). The collection ends with an especially rewarding essay by ChristopherFaraone on ancient love magic-or rather,to use his more precise terminology, "eros magic" and "philia magic."This fascinating body of materialprovides some supportfor the theoretical models that have been reconstructedlargely on the basis of literarytexts: "masculine"and "feminine" styles, for example, are at times independentof physical sex. Thus we find male subordinatesusing not the usual "masculine"-stylecharms aimed at provoking uncontrollable lust (eros, pothos, himeros, oistros) but rather "feminine"-stylecharms aimed at arousing love or affection (philia, agape, storge), and seeking to arouse this affection not in potential sexual partnersbut in male social or political superiors, e.g., kings or governors; and we see female prostitutesusing "masculine"-styleeros spells in order to attractmen. On the other hand, anotherof Faraone's argumentscomplicates the widespread scholarly belief that Greekmisogyny rested on a conviction that women were naturallywild and promiscuous, hence in need of taming. Magic charmsand some ancient writings on animal behavior suggest a coexisting and competing model according to which females have "a traditional antipathyor coolness to sexual intercourse,whereas conversely it was thought that although men were 'naturally' passionate and sexually aggressive, they could be cooled and calmed by the philia spells of women" (p. 412). Faraone's material brings to mind furtherhints at competing models in antiquity, such as on the point of heat in particular.While Aristotle on several occasions describes the male as naturally hotterthanthe female, a Hippocratictext solemnly declaresthatthe warm pleasure stirredup in women's bodies by intercoursecomes to an end the moment their male partnerejaculates, much as a pot of boiling water stops boiling the moment a cup of cold water is thrown in (On the Generating Seed 4). As this overview makes clear, those who look for tight thematic unity in a collected volume will be somewhat disappointed. Seekers after unity will also observe a certain lack of interactionbetween the pieces: key passages from Dio Chrysostom are discussed in passing in Konstan's essay and form the central topic of Houser's, yet neither essay shows any awareness of the other, and the index confuses matters by giving "Dio Chrysostom" and "Dio of Prusa" as two distinct entries! Likewise
BOOK REVIEWS
89
Dover's discussion of the two Samian women in Asclepiades' poem uses the term "lesbian"without reference to Halperin's discussion in "The First Homosexuality?" But these are missed opportunities for editorial intervention more than anything else. The volume's strengths are clear. Above all, detailed discussion of the role of the sexual within ancient philosophical traditions was sorely needed, and it is to be hoped that this collection will pave the way for more. Furthermore,while The Sleep of Reason adds to the growing consensus on the broad outlines of ancient Greek and Roman sexual ideologies (masculinity is associated with a dominant style and the penetrativerole in sexual acts, regardless of the sex of the partner,while femininity and effeminacy are linked with a subordinatestyle and the penetratedrole), it also provides glimpses at the complexities, exceptions to rules, and unresolved tensions that will necessarily accompany any cultural system of this kind (e.g., Nussbaum, Konstan, Faraone). Next, while scholarship has rightly emphasized that the sexual and the conjugal are fundamentally independent spheres within ancient conceptual systems and thus deserve separate treatment,this may have led to a certain underemphasis on the fact that, after all, sexual relations were normatively a part of marriage: Kaimio's and Cantarella's contributions come as helpful reminders. Finally, although in each case the point has been made in general terms by others, Sihvola's and Houser's pieces decisively remove two major figures-Aristotle and Dio Chrysostom respectively-from the shrinkinglist of ancient writers who can legitimately be said to have condemned sexual relations among males per se. Craig Williams Brooklyn College and the Graduate Center City University of New York
Greek Literature and the Roman Empire: The Politics of Imitation. By TIMWHITMARSH. Oxford: Oxford University Press, 2001. Pp. xiv + 377. $74.00 (cloth). Early in Greek Literature and the Roman Empire, Tim Whitmarshdescribes his topic as a single "centralquestion,"namely,"howliteraryexperienceis constructedand thematizedin the texts of this period, and how 'the literary' is employed to construct Greek identity in relationship to the Greek past and the Roman present" (pp. 1-2). W.'s response to that question makes a learned and challenging book, full of brilliant readings. It displays deep affinities with a numberof recent works on imperial Greek literature,most prominentamong them, perhaps, Maud Gleason's Making Men and Simon Goldhill's Being Greek under Rome.1In this review, therefore, I concentrate on two broadareas of inquiry,ones of importancenot only to this book but to current ways of studying Greek sophistic literatureof the Empire more generally, the first
1. M. Gleason, Making Men: Sophists and Self-Representation in Ancient Rome (Princeton, 1995); cf. eadem, "The Semiotics of Gender:Physiognomy and Self-Fashioning," in Before Sexuality: The Construction of Erotic Experience in the Ancient Greek World,ed. D. M. Halperin, J. J. Winkler, and F I. Zeitlin (Princeton, 1990), 389-415; S. Goldhill, ed., Being Greek under Rome: Cultural Identity, The Second Sophistic and the Development of Empire (Cambridge, 2001); cf. idem, Who Needs Greek? (Cambridge, 2001), chap. 2, "Becoming Greek, with Lucian."
90
BOOK REVIEWS
being the view of literatureand social construction that emerges here, and the second being the new historical project in which this book participates.2 We might begin, however, with a review of the book's structureand contents.3In his first two chapters, W. draws out two central concerns of his texts, namely, imitation andpaideia (which he likens respectively to "the mimetic constructionof literature and of the self," p. 28), and relates these to imperial Greek literature's "belatedness"or "posteriority."These themes, he argues, are not innocent features of a literary culture, but rather reflect and participate in a (larger) project, that of "negotiat[ing]an identity discrete from Rome" (p. 2). The introduction,which elaborates this programin both literary-historicaland theoretical terms, thus concludes by aligning the book's aims with two scholarly traditions, one anthropological, which sees texts as "ludic forms" that "constitutenot secondary reflections upon social practice, but primary means through which social hierarchies are constituted" (p. 30), the other literary-critical,and more strictly speaking New Historical, which "emphasizes simultaneously the grounding of literary texts in political and socioeconomic materiality, and simultaneously the constitution of power and identity through 'literary' (or, at least, symbolic) modalities that require thoughtful and attentive unpacking" (p. 31). In the introduction, it is Judith Butler who bridges these traditions,by insisting upon the priorityof discourse in the constructionof the self (pp. 30-31). But these theoretical traditionsdo not always run parallel, nor do they only converge. It is to Butler's position in this hybrid scheme, and moments of divergence between these theoretical postulates, that I shall shortly turn. W. divides the book's second part into three chapters, each one focused on a different "paradigmaticmodel for relations between Greekpaideia and Roman power" (p. 34). The fifth, on Lucian, is perhaps the book's best, as W.'s formidable skills as a reader find a worthy subject in the virtuoso satirist. Chapters 3 and 4 take up related sets of what W. calls "oppositional"and "conciliatory"narratives.In particular, the third chapter considers the representations of Musonius Rufus, Dio, and Favorinus of themselves as exiles. In reading these texts, W. insists (pp. 137-38), it is crucialto understand the specificdynamicsof the culturein question,to site the of exile in the contextof the issues relevantto questionsraisedby the representation the authorin question.In the case of the textsdiscussedhere,the fundamental cultural andconceptualcontextwill be Greekculture'sself-perception as a traditionreanimating thepastas a meansof resistingthe threatfromRome. If the first claim-that on behalf of contextualization-can scarcely be faulted, the second strikes me as problematic:it is not at all clear to me that exile should be understoodas a form of "resistance,"or that the men in question saw themselves as exiled "fromRome,"or, indeed, that "resistance"as an ideology motivated such rep-
2. In a review of Being Greek under Rome forthcoming in Phoenix, I briefly consider two related problems in this field, namely, the view of ethnicity that underlies it and the periodization to which its participants generally subscribe. 3. As I started writing this text, Greek Literatureand the Roman Empire received an extended and sympathetic review by LarryKim, who provides a fuller summarythan that attemptedhere (BMCR2003.03.14). As Kim notes, much of Whitmarsh's argumentis "summary-defying,"in the best sense.
BOOK REVIEWS
91
resentations by the Greek elite or motivated the Greek elite more generally.4How would we recognize such resistance? Did Romans recognize it? Did Greeks practice it? More on this below. The fourth chapterturns to depictions of Greek advisors to Roman emperors and focuses first on Dio's Kingship Orations and then on the "mythicization"of Dio as free-speaking philosophical advisor, particularly in Philostratus. This chapter distills much of the promise and problems of W.'s work. For example, althoughW. cites with approvalmuch recent scholarshipon the politics of Latin literatureand Roman culture under the Empire, he does not take up Latin Roman analogs to such mythicization, among whose objects one might number both Cicero and Seneca.5 More seriously, though W. opens by suggesting that a betterunderstandingof the Kingship Orations might arise from a "moresolid" attemptto site them (pp. 187-88)-and he devotes an appendix to doing just that, arguingfor a "public,performativeframe"in "one or more of the highly sophisticated rhetorical centers of the urbanEast"-he closes the chapter by reading Philostratus' depiction of Dio as an example of what HaroldBloom called "kenosis,"an emptying out of a claim to priorityby a posterior writerclaiming to be a precursor(p. 236). In other words, what startedas a New Historical project with a properly historical concern for "political and socio-economic materiality,"that is, for context, ends as a purely literary inquiry into the anxiety of influence. And this is not the only example of such theoretical self-subversion.6 Centralto W.'s entire project is the claim that "literature"is an arena in which or, perhaps, a mechanism throughwhich "identities"are constructed.This understanding of subjectivity itself rests on two priorpropositions. First, althoughW. claims in his conclusion to have "sought to show that Greek cultural identity in the Roman period.., .was an intense concern, but... was never taken for granted: identities were made, contested, masked, ironized,dissimulated,or disavowed, but not received as self-evident," he soon asserts that "identity is never self-evident," indeed, that "identities do not 'exist,' except in so far as fictions have real and indelible effects upon peoples' lives" (pp. 295-96; my emphasis). W.'s wording to one side, I take his assertion regarding Greek identity under the Empire to be a sociohistorical illustrationof the more general, ontological claim about identity advanced in the second and third quotations. The second prior claim is that mentioned above and oft repeated in this book, that selves are, in Butler's terms, "discursively constituted" 4. A different version of chapter 3 appears as "'Greek is the World': Exile and Identity in the Second Sophistic" in Being Greek under Rome (n. 1 above), pp. 269-305. That version undertakesa deeper effort at contextualizing identity construction and seems to me the more successful. 5. On Cicero see R. A. Kaster, "Becoming 'Cicero,' " in Style and Tradition:Studies in Honor of Wendell Clausen, ed. P Knox and C. Foss (Stuttgart, 1998), 248-63; on Seneca, see T. Habinek, "Seneca's Renown: Gloria, Claritudo, and the Replication of the Roman Elite," CIAnt19 (2000): 264-303. 6. For example, with W.'s description of New Historicism quoted above and his placing of this project in that tradition, contrast his allowance that his book "is not centrally concerned with the material realities of literature, the circulation, ownership, performance, and reading of texts" (p. 1). Compare the assertion that "paideia ... was not simply a form of social practice (though, of course, it was that too): at a more abstract level, it was also a means of constructing and reifying idealized identities for Greek and Roman, a privileged space of complex cultural interaction (or 'contact zone') between Roman ideology and Greek identity, a foundation upon which both peoples constructed their own sense of their place in the world" (p. 16). I might characterize my concern as arising from a misgiving thatpaideia "at a more abstractlevel" can, in fact, properly be studied without consideration for paideia-as-social-practice, not least by someone with W.'s theoretical affiliations.
92
BOOKREVIEWS
(p. 31). I do not wish to take issue here with theories of social construction per se,7 beyond registering some concern that it seems unlikely that language should be so privileged in the constitution of the self, and even less likely that highly wrought literary texts, that is, those exemplifying paideia, written (in this case) exclusively by wealthy, powerful men in their maturity,should be given such a privileged role "in constructing Greek identity and exploring Roman power (pp. 16-17)."8 Rather, in keeping with W.'s New Historicist program,it might be appropriateto interrogatethe totalizing claims made on behalf of these very postmodern theories of language and the self.9 Such an inquiry might take place along two fronts. First, Butler's post-structuralismas it emerges in Bodies That Matter rests on a distinctly post-Saussurian view of language. Might it not be that cultures that subscribed to very different theories of language will have understoodand experienced socialization in very different ways? In his first chapter, W. contrasts Plutarch's esteem for mimesis in How a YoungMan Should Listen to Poetry with Plato's attackon mimesis in Republic 10.10Plutarch,W. notes, can allow that "good artcan imitate bad things" precisely because he believes students should learn to discern for themselves "what is advantageousand what deleterious in poetry."That established, Plutarchcan then allow that mimetic art's "excellence lies in its homoiotes [likeness, 18a], not in what it represents(18a-d)" (p. 51). W. traces the ancestry of Plutarch's argumentback to Aristotle, but it has Platonic antecedents, too. At Cratylus 432b-c, Socrates insists that images have their own "principle of correctness" (bp06tr) precisely because they have a differentontological status from that of the things they represent.He first illustrates and then extends this argument. The illustration consists in the famous question whethera chronologicallyposteriorduplicateof Cratyluswould be "another Cratylus or an image of Cratylus."Consider the question in its long formulation: Wouldtherebe two things-Cratylusandan imageof Cratylus--inthe followingcircumstances?Supposesome god didn'tjust representyour color and shapethe way paintersdo, butmadeall the innerpartslike yours,withthe samewarmthandsoftness, andputmotion,soul, andwisdomlike yoursintothem-in a word,supposehe madea duplicateof everythingyou haveandputit besideyou. Wouldtherethenbe twoCratyluses or Cratylusandan imageof Cratylus?[Trans.Reeve] I stress again, the question for Plato is an ontological one, and the example is not innocent: it is in this argumentprecisely a person, a self, a subject whose status is differentiatedfrom all things that might be labeled "mimetic." Crucially, Plato extends this argument back to language and in particular to names, whose representationalcapacities Socrates had comparedto, and would later distinguish from, those of objects of mimetic art. In that context, what Cratylus' concession (that there would be two Cratyluses) initially embraced was a further 7. On which see the now classic essay by J. R. Searle, The Construction of Social Reality (New York, 1995), and cf. E Collin, Social Reality (New York, 1997); and I. Hacking, The Social Construction of What?(Cambridge, Mass., 1999). 8. See also p. 29, on the preeminent place of the "realmof the literary imagination,"and p. 89, where he consciously glosses "what it is to be Greek" with "to be a Greek writer." 9. But cf. p. 31: "This book does not represent a 'new historical' 'approach' to ancient literature, in that I am aware of the degree to which new historical theory is circumscribedby the preoccupations of the modern West." 10. Readers of W. on mimesis will want to consult S. Halliwell, Texts and Modern Problems (Princeton, N.J. 2002).
The Aesthetics
of Mimesis: Ancient
BOOK REVIEWS
93
concession to Socrates' insistence that one must find a new "principle of correctness" that embraces both images and names (432d). These will be distinguished, of course, when Socrates and Cratylusconclude thatmimetic representationdiffers from language in being nonpropositional.All we must recognize, then, about their subsequent debate over the conventional or arbitrarynatureof signs is how different it is from "our"linguistics, for that portion of the debate closes, once again, with the allowance that it is best to learn about "things" (T tp6'ypaTa)from "themselves" rather than their "names" (8t' ai'TrCvrather than St' (439a-b). These 6voda•zov) "things"that are prior to representationsinclude people. The priority of people to representationsleads to the second of my avenues of inquiry regardingW.'s view of the self and language. In a world in which selves are prior, what are the effects of poetry on them? In the Republic, "poetic mimesis"of every kind has the same effect on the soul, stirringup "appetitesand pains and pleasures in the soul," "makingsuch feelings our rulers when they ought to be ruled ..." (406d). It is unnecessary here to explore the theory of the soul that lies underneaththis view of mimesis and its effects. That Plato could make such an argumentabout the soul illustrates at the very least how different an ancient view of the self and its relationship to language and mimesis might be, and how very different the theoretical postulates underpinning such views were.11 Of course, it may be that no one even of his contemporariesfollowed Plato in viewing language and selfhood thus, and it might be that the schismatic, schematized self that emerges from W.'s readings is very much real (cf. pp. 222 and 294). If so, we would want to know what made such a self possible, whether such an understanding of subjectivity was itself a response to "power."To address that question, we should want to think harderabout what power is, a problem to which I shall return. In this book, the imperial Greek self is ratherboth a productand productiveof a sophistic hermeneuticthatis almost Rortianin its kaleidoscopiccatholicity.At times, it seems as though all possible readings are potential at all times for all readersexcept powerful Romans, who never notice when they are being criticized (p. 70).12 Whether such "an imaginary,idealized pedagogical community"might have existed is, of course, one problem (p. 54). Anotheris the processual connection regularlybut not always drawn between this literary hermeneutic and the self, in a world (like ours) that had ready resources for divorcing the two. In "our"case, we might turnto theories of "rhetoric"(p. 190), and conclude a reading of Dio by speaking of his "rhetorical self-presentation" (p. 215). But a Greek might rather have spoken of ethopoeia, which W. translates "the art of characterization"(p. 188).13 A world
11. So, for example, I would argue that Marcus Aurelius' efforts "to become truly philosophical" cannot be considered apart from his view of the self, which was not, I think, a Tayloreandialogical one (p. 222). For Marcus, I suspect, a theory of the self was properly a theory of the soul; see, e.g., 2.2 ("OriTnozTeoTr6 s r% uIXi1 Kai TOiyfpOVtK6v) and esp. 5.26 (T6 -iyspovtKbv cai ~igt, capria a'c Ki r1uptKov T tVEUPdT1nOV l zpaxseag Ktv•olosx Kai Pi~)ouyKtlpvadoG, 6t &7Esptoou RIEpo; ntov EaoT dTIgsv fl capKli t-pe 0Itb •siag It is precisely those behaviors and parts ypaw~rE aroi nKepoptsCirr K&ai zt& t&EiosEg tKEiva;tv toiT;opiot;). of the body that might be schematized that Marcus divorces from the self/soul that he wishes to improve. 12. See, e.g., pp. 33-34, 54, 67-71, or 199-200. 13. Indeed, ithopoeia plays a much smaller role in the argument than one might have expected. No connection is drawnbetween ancient writings on characterization(or schematization) and the (modern) hunt for "identity,"even at those moments when W. refers to "literary"or "satirical"personae or to Dio's "playing the chameleon" (pp. 179, 216, 252).
94
BOOK REVIEWS
steeped in rhetorical training, in which ethopoeia "was central from its earliest days," might well have heard the Kingships as W. presents them, "as a performance of identity, not as an expression of innerpersonality"(p. 188, his emphasis). But that must leave open the possibility that Dio and his audience took as given the Platonicmimetic quality of self-representation,which is to say, that they placed the self on a differentontological plane altogetherfrom its representations,and so did not (rightly or wrongly) conflate a discourse about identity with its construction.14 What this book chiefly inspires are reflections on the future of New Historicist readings of Greek literatureof the early Empire. This will requirea reckoning with the second element of W.'s title, namely, the Roman Empire. For it is not individual Romans, or Roman practices, or Roman texts, or Roman ideas, that are presentin the pages of this book, but "Romanpower."The repeateduse of that simple phrase runs the risk not simply of fetishizing power, but of essentializing power's "Romanness."15There is first the very real problem, as Aristotle saw, that power inheres in social relations; what differentiates a community or society from a mere aggregate of individuals are the mechanisms by which power is regulated, legitimated, and institutionalized-in Cicero's terms, by a consensus about ius. To deplore "power" tout court is to deplore human sociability. Tenuredacademics may have that luxury; most people do not. More specifically, we might recall that the Athenianliteraturethat has historically been valorized purportedlyon aesthetic grounds was produced in and for a citizen body actively seeking to extend its domination over others (cf. pp. 41-44, 298). The curtailment of individual and community freedoms that followed upon the rise of Macedon-however that took place, and whatever form it took-coincided in the Hellenistic period with another of those diasporas that produced both new articulations and fractious contestations of Hellenicity.16Should we not ask whetherthe domestication of comedy and rise of philological scholarship were not also responses to changes in the politics of literaryproductionin the late fourth and early thirdcenturies B.C.E.(p. 9)?17 Insofar as that question is answered in the affirmative,the relation between "power" and literature will require correspondingly greater historicization than it here receives. What differences in theme and genre, in depictions and constructionsof subjectivity,are detectablebetween Greekliteratureof the Hellenistic period and that written under the Roman Empire?For that matter, what differences are detectable between Romans and Greeks when speaking to power-
14. Consider Lucian Pisc. 31, which W. quotes and translates (p. 261): a bravuraindictment of false philosophers who "imitate"the "obvious, common signs" of a philosophical disposition but whose "lives" W. writes "style,"but I wish (bioi) contradict their "outwardstylization" (my loose translationfor skhe^ma; here to highlight the distinction between "inner"life and "outward"style). Lucian seems to me to present us with an epistemic problem, that of seeing past insincere schematization to something within-a problem that mirrorsthe metaphysical distinction ultimately drawn by Socrates in distinguishing the referential capacities of different kinds of names from that of works of art. 15. This risk is present despite the attentionthat W. devotes to the ideological forces behind the "culturepower polarity" (pp. 17-20), as his interest in exposing it rapidly yields to a practice that more or less affirms it. 16. On this problem in the Roman period see now J.-L. Ferrary, "Rome et la g6ographie de l'hel16nisme:R6flexions sur 'hellbnes' et 'panhellbnes' dans les inscriptions d'6poque romaine,"in The Greek East in the Roman Context, ed. O. Salomies (Helsinki, 2001), 19-35. 17. See S. Lape, "DemocraticIdeology and the Poetics of Rape in MenandrianComedy,"CIAnt20 (2001): " 79-119, "The Ethics of Democracy in Menander's 'Dyskolos,' Helios 28 (2001): 141-72.
BOOK REVIEWS
95
which is to say, how shall we isolate what is Greek about imperial Greek literature?18Did literary strategies developed in response to monarchyrequiremodulation as new monarchs sprang up from new races? Or when old monarchs changed their race?19 As with Hellenicity, so with identity.Understandingsof the self must have evolved between Plato and Dio. How shall we analyze and describe those changes? W. suggests one importantroute in his reading of Favorinus' On Exile, of which text he also supplies a translation. He highlights Favorinus' critique of autochthony and contrasts it with the use made of autochthony by Demosthenes 60.4 (pp. 175-77), but the origin and context of his critique remainunexplored,as, curiously, do the implications of autochthonyfor notions of identity.As regardsthe latter,W. might profitably have returnedto his own reading of Musonius Rufus' rejoinderto Euripides' understandingof free speech. There W. argued that Euripides "presuppose[d]a context wherein the right to free speech defines the citizen body exclusively" (p. 144). More is at stake than this: for ideologies of autochthonynot only define communities, but situate individuals in communities and cultures prior, as it were, to socializing them. Not only do communities that subscribe to such ideologies not require theories of socialization, they are often hostile to them. It is little wonderthat the Socratic elenchus, which functioned to reveal the contingent natureof Atheniancultural identity, provoked such hostility. This inquiry into the Roman context of imperial Greek literaturereaches its greatest complexity and yet also yields the most promising fruit when we turn to the geographic, legal, utopian, and affective components of exile (pp. 133-80). Favorinus may end his critique of autochthonywith an appeal to cosmopolitanism (urging that W one should raiuav Piv Pily ac?ilv oioav, oiKEliv ? zpo'pbvrlZpa Kci y•lv nrivOv and nurse of all," 10.4 [pp. 175-76]), but "inhabitthe whole earth, as it is the mother he opens by reflectingon the contingentnatureof patriotismto "fatherlands"(10.1-2; p. 172). The conditions that enabled these ideas to exist in this particularconstellation are perhapsbest revealed by juxtaposing them on the one hand with the Delphic oracle's instructionto Dio, that he should take his exile "to the end ... of the earth" (Or. 13.9; pp. 161-62), and on the other with the rise underthe empire of a literature by exiles, in contradistinctionto a consolatory literature addressed to exiles. Seen thus, what is perhaps most remarkableabout imperial exiles is their immanence.20 The sundering of political identity from polis that made this discourse on exile possible finds its roots not simply in Stoic cosmopolitanismin its Hellenistic, utopian form, but in the new geographies of politics and power that arose in the late Roman Republic. For it was the (notionally) worldwide extension of Roman rule and the rise of Caesar that forced the reconfigurationof exile, from a physical to a metaphysical condition: "you ought nevertheless to reflect on this," wrote Cicero to Marcellus in 46, "that wherever you are, you will be in the power of him whom you 18. See, e.g, p. 66, n. 107 on Tac. Dial. 36-40, or p. 158, n. 95 on Plin. Pan. 47.1-2. 19. See, e.g., L. Koenen, "The Ptolemaic King as a Religious Figure," in Images and Ideologies: SelfDefinition in the Hellenistic World,ed. A. Bulloch et al. (Berkeley, 1993), 25-115. 20. Cf. p. 155 (author's emphases): "As a citizen of the world, Musonius is neither exclusively Roman nor exclusively Greek, but both and still more. The exile is a problematic, unsettling figure, one who points up the arbitraryand provisional nature of cultural identity."I concur, but stress the way in which it is not exiles "out there,"but exiles "among us" who, under the Empire, function in this "problematic,unsettling" way.
96
BOOK REVIEWS
flee." Even were Caesar willing to let Marcellus live in peace and quiet, "sundered from his fatherland and fortune,"Marcellus would still have to decide whether to live at Rome, or Mytilene, or Rhodes: "Forthe power of him whom we fear stretches so widely that it embraces the entire world" (Fam. 230.4). Imperial understandings of relegatio, in which the condemned must avoid both Rome and the emperorwhose transcendentstatus as pater patriae contingently endowed his location with the status of patria-develop directly out of the realities that Cicero and Marcellus confronted.21 The flowering of these ideas in Greek naturallytook a different form in the late Republic and early Empire, at least, than it did among Romans of that period, both because Greeks conceived of the affective bonds between individuals and their cities using paradigms for political life different from those Romans used, and because Greeks, lacking inherent claims to Roman citizenship, required different means to conceive their place-and the exile's lack of place-within the Empire. In jumping from Aristotle's view of humans as "animal[s] that belong in a polis" directly to Chrysostom,W. takes for granteda remarkabletransformationin the place of thepolis in the Greek political imagination (p. 139). The history of that transformationremains to be written, but it should be traced along two axes, at least. First, we might inquire after the place of the polis within the Empire as Greeks understood it, for even when the political identity of individuals was not explicitly raised in such theorizing, it will have been at issue.22 Second, at the level of practice, we might ask where the exile stood, if not within the Empire. Next to polis and cosmos lay another system of mapping identities, namely citizenship.23For what Musonius Rufus, Dio, and Favorinus speak to, in their figurationsof exile and fatherland,are the realities of the real geographic mobility of persons under the Empire. It was, for better or worse, the Roman government alone that could fix political identities in a fashion that transcendedthe polis and made that mobility possible, and so made imperial exile meaningful: it was, in short, the Empire that made the exile something other than a traveler from anothercity. It is, moreover, hard to believe that the fixing of identities throughcitizenship at an imperial level will not have recursively affected conceptions of political and cultural identity within communities at the local level. One thing at least it seems to have provoked was the rise of a new binarism within Greek mappings of the world, that of resident and nonresident of the Empire.24 The Romanness of Greek writings on exile is paradoxicallyperhapsbest revealed by a consolatory treatise on exile, that of Plutarch,whose addressee (probablyMenemachus of Sardis) suffered relegatio in the aftermathof political upheaval in his native city.25Like the authorswhom W. studies, Plutarchcites an abundanceof classical exempla-all those, in fact, cited by Musonius, Dio, and Favorinus, and more besides. But his addressee lives in a Roman world, among whose prominent exiles are Hannibal (606c), Camillus (605e), Cicero (605e), and Tiberius (602e). And 21. Some aspects of this history are elaborated in C. Ando, Imperial Ideology and Provincial Loyalty in the Roman Empire (Berkeley, 2000), 336-412. 22. For a provisional such essay, see C. Ando, "Was Rome a Polis?" CIAnt 18 (1999): 5-34. 23. On exile as a negotiation between polis and cosmos, see p. 138; on the citizenship of the authors he studies, see p. 300, the book's penultimate page. 24. On the census, see Ando, Imperial Ideology (n. 21 above), 351-85, and on residence as a criterionfor membership, and membership as a problem within provincial attitudes to the empire, see ibidem, 277-335. 25. C. P. Jones, Plutarch and Rome (Oxford, 1971), 117.
BOOK REVIEWS
97
within that Roman world, Plutarchargues, Menemachus has lost nothing more than a residence-which is all a patris really is (600e-f); what is more, in so doing he has been released from the tribulations of life in the contemporarypolis: embassies to Rome, the entertainingof governors, or, for that matter,worrying who the next governor will be (602c, f). Plutarch's appeal to Roman exempla provokes curiosity about the meaning of classical exempla in his essay. Like Musonius Rufus (frag. 9, p. 42 Hense, on which see W., p. 146), Plutarch attributesto Socrates a claim of citizenship in the world: "Socrates spoke better, saying that he was neither Athenian nor Greek, but 'Cosmian,' as one might say 'Rhodian' or 'Corinthian'" (600f). In point of fact, this anecdote appearsfirst in Cicero's Tusculans,whose paragraphson exile demonstrably influenced Favorinus (5.106-9). Socrates undoubtedly became the most common mouthpiece for this sentimentbecause of his paradigmaticstatus, but that status hinged in this period, at least, on his willingness to interrogate"the traditionalparochiality of Greek identity" (p. 139); in attributingthis sentiment to him, Epictetus, Plutarch,Musonius, and Cicero resituatedSocrates in a peculiarly Roman landscape. Indeed, Musonius' language-that the cosmos is the KotvilitaTpig tv0poj3ntv &dacvz-ov, "the common fatherland of all humans" (frag. 9, p. 42 Hense)-is distinctly nonclassical, and finds its closest Greek analog at Aelius Aristides 26.100, ratherthan in more distinctly Hellenistic, Stoic formulations such as PlutarchMoralia 329c, where the object is to unite the virtuous, which is to say, the Greek. And what of the place of classical exempla in Cicero? When Cicero cited Socrates, or his list of sixteen (Greek) philosophers who suffered exile, what sort of "self-positioning" was he undertaking(p. 151)? Was he, like Musonius, "construct[ing] a schematic opposition between Greek and Roman in terms of philosophical liberation and Roman power" (p. 151)? What did it mean that this strategy, that of appealing to classical, Greek exemplars, was available to Romans, for use in Latin? To answer that question, we might turn back to W.'s principal claim: "Where this book seeks to innovate is in advancing the proposition that authorsdo not write because they are Greek; they are Greek because they write" (p. 2, his emphasis). His claim representsan innovation, one need scarcely point out, because many have held that "to be Greek" required, in some easy and impossible way, no more than "to speak Greek."The political import in a Roman context of this common understanding of "being Greek" has received surprisingly little attention. To unpack that, we should need to spend a little time studying Roman understandingsof cultural identity. This much we can say now: that speaking Latin was not central to classical understandingsof being Roman, at least, not in the same way that speaking Greek was fundamentalto being Greek.26Indeed, so far as I know, reflection on the marginality of language to Roman identity develops only under the Empire, and chiefly in commentarieson the Aeneid, whose narrative,Servius came to understand,required one to ask why the Romans spoke Latin rather than Trojan (ad Aen. 4.16, and elsewhere). If the Romans did not conceive of language as centralto political or culturalidentity, and would therefore not have recognized Hellenism as a form of resistance, it 26. On this topic, see provisionally J. N. Adams, "'Romanitas' and the Latin language," CQ 53 (2003): 184-205.
98
BOOK REVIEWS
must have been very frustratingto resist by speaking Greek. And from the foundation of the Panhellenion, if not before, classicizing pretensions became a distinctly imperial way of being Roman. From W., whose readings of Greek literatureare so extensive and sophisticated, I would seek now a more deeply contextualized inquiry into Greek notions of culturalidentity, which, I suspect, cannot in the end be understood without considering contemporaneous alternatives. For when the currents in Greek thought on geography, politics, and identity are traced across a longer time span, from the second century B.C.E.to the fifth century C.E.,the ideology of Hellenism as resistance looks like a mirage: three hundredyears after Dio, residents of the eastern Empire were overwhelmingly calling themselves Roman at the very moment when the Empire became Greek. Clifford Ando University of Southern California
BOOKS RECEIVED [Not all workssubmittedcan be reviewed,butthosethatpertainto the subjectmatterof thisjournalare regularlylistedunder"BooksReceived."Offprintsfromperiodicalsandpartsof bookswill notbe listed Bookssubmittedarenotreturnable.] unlesstheyarepublished(sold)separately.
ADLER,EVE. Vergil's Empire: Political Thought in the "Aeneid."Lanham, Md.: Rowman and Littlefield Publishers Inc., 2003. Pp. 368. ALLEN,MICHAELJ. B., JOHNWARDEN,JAMESHANKINS,and WILLIAMBOWEN(eds.).
Marsilio Ficino: Platonic Theology.Vol. 3, Books IX-XI. Cambridge,Mass.: Harvard University Press, 2003. Pp. 362. $29.95.
BERS,VICTOR (trans.). Demosthenes, Speeches 50-59. Austin: University of Texas Press, 2003. Pp 237. $22.95 (paper); $45.00 (cloth).
G. R. (ed.). Metaphor, Allegory, and the Classical Tradition:Ancient BOYS-STONES, Thought and Modern Revisions. Oxford: Oxford University Press, 2003. Pp. 305.
BRAGUE, Ri'MI.The Wisdomof the World: The Human Experience of the Universe in Western Thought. Trans. TERESALAVENDERFAGAN. Chicago: University of
Chicago Press, 2003. Pp. 304. $35.00. Darius dans l'ombre d 'Alexandre.Paris: Fayard, 2003. Pp. 666. BRIANT,PIERRE. BYDEN,BORJE.Theodore Metochites' "Stoicheiosis Astronomike"and the Study of Natural Philosophy and Mathematics in Early Palaiologan Byzantium.G6teborg: Acta Universitatis Gothoburgensis, 2003. Pp. 547. CALAME,CLAUDE.Myth and History in Ancient Greece: The Symbolic Creation of a Colony. Trans. Daniel W. Berman. Princeton, N.J.: Princeton University Press,
2003. Pp. 200. $39.50. CONCILIO,CARMELA,MASSIMILIANOD'AIUTO, and SAiPA POLIZIO.La traditione metrica della tragedia greca. Naples: Universita Degli Studi Di Salerno, 2002.
Pp. 74. ANDREA CUCCHIARELLI, (ed. and trans.). La Vegilia di Venere:Pervigilium Veneris. Milan: Biblioteca Universale Rizzoli, 2003. Pp. 167. MARIACECILIA. D'ERCOLE, Importuosa Italiae Litora: Paysage et echanges dans 1'Adriatiquemeridionalea 1'dpoquearchat'que.Naples: CentreJean Bdrard,2002. Pp. 414. DEROW,PETER,and ROBERTPARKER(eds.). Herodotus and His World: Essays from
a Conference in Memory of George Forrest. Oxford: Oxford University Press, 2003. Pp. 376. DUFF, TIMOTHYE. The Greek and Roman Historians. London: Bristol Classical Press,
2003. Pp. 136. $14.00. Platonic Noise. Princeton, N.J.: Princeton University Press, 2003. EUBEN,J. PETER. Pp. 224. $18.85 (paper); $55.00 (cloth). FERRARI,G. R. F. City and Soul in Plato's "Republic."Sankt Augustin: Academia Verlag, 2003. Pp. 130.
99
100
BOOKSRECEIVED
WILLIAM W. Theophrastean Studies. Philosophie der Antike, 17. FORTENBAUGH, Franz Steiner Stuttgart: Verlag, 2003. Pp. 345. 472.00. KATHY L. The GACA, Making of Fornication: Eros, Ethics, and Political Reform in Greek Philosophy and Early Christianity.Berkeley-Los Angeles-London: University of California Press, 2003. Pp. 359. GARCIAFLEITAS,Luz MARIA, and GERMANSANTANAHENRfQUEZ.La imagen de
Egipto en los fragmentos de los historiadores griegos: Una primera aproximaci6n. Las Palmas de Gran Canaria: Universidad de las Palmas de Gran Canaria,
Servicio de Publicaciones, 2002. Pp. 124. GIBBONS,REGINALD,and CHARLESSEGAL(trans.). Sophocles:
'Antigone." Oxford:
Oxford University Press, 2003. Pp. 197. $25.00. GIBSON,ROY(ed.). Ovid: 'Ars Amatoria" Book 3. Introduction,Text, and Commentary. Cambridge Classical Texts and Commentaries, 40. Cambridge:Cambridge University Press, 2003. Pp. 446. $80.00. HALEY,EVANW. Baetica Felix: People and Prosperity in SouthernSpainfrom Caesar to SeptimiusSeverus. Austin:University of Texas Press, 2003. Pp. 297. $45.00. Hannibal's Dynasty: Power and Politics in the WesternMediterHoYos, DEXTER. ranean, 247-183 Be. London and New York:Routledge, 2003. Pp. 304. $80.00. HULT,KARIN(ed. and trans.). TheodoreMetochites on Ancient Authors and Philosophy: "Semeioseisgnomikai" 1-26 and 71. A Critical Edition with Introduction, Translation, Notes, and Indexes. G6teborg: Acta Universitatis Gothoburgensis,
2002. Pp. 360. BRAD(ed.). The Cambridge Companion to the Stoics. Cambridge:CamINWOOD, bridge University Press, 2003. Pp. 438. LLOYD,G. E. R. In the Grip of Disease: Studies in the Greek Imagination. Oxford: Oxford University Press, 2003. Pp. 220. $45.00. MAIRE,BRIGITTE (ed.). ConcordantiaeGargilianae. Hildesheim:Olms-Weidermann, 2002. Pp. 215. MCCARTHY,DANIELP. and AIDEN BREEN(eds.). The Ante-Nicene Christian Pasch
"De Ratione Paschali": The Paschal tract of Anatolius, Bishop of Laodicea. Dublin: Four Courts Press, 2003. Pp. 200. $65.00. CATHERINE. MORGAN, Early Greek States beyond the Polis. London and New York: 2003. Routledge, Pp. 326. $80.00. OIKONOMAKOS, KONSTANTINOS (ed.). Nikandrou Alexipharmaka. Athens: Akademia
Athenon, 2002. Pp. 104. PARKIN, TIMG. Old Age in the Roman World:A Cultural and Social History. Baltimore, Md.: Johns Hopkins University Press, 2003. Pp. 416. $55.00. KARINBLOMQVIST, PILTZ,ANDERS, JOHANNAAKUJARVI,VASSILIOSSABATAKAKIS, GEORGWALSER,and LARS NORDGREN(eds.). For Particular Reasons: Studies in
Honour of Jerker Blomqvist. Lund: Nordic Academic Press, 2003. Pp. 345. PHILOMEN. A New Short Guide to the Accentuation of Ancient Greek. PROBERT, London: Bristol Classical Press, 2003. Pp. 215. $20.00. ROSAFINO,PASQUALE.Studi sul colonato. Bari: Edipuglia, 2002. Pp 235. SANTANAHENRfQUEZ,GERMAN.Semdnticay lingiiistica: Aplicaciones m.del todo de la Sprachinhaltsforschung al griego antiguo. Las Palmas de Gran Canaria: Uni-
versidadde las Palmas de GranCanaria,Servicio de Publicaciones, 2000. Pp. 157.
BOOKsRECEIVED
101
Tradici6ncldsica y literaturaespahiola.Las Palmas SANTANA GERMAN. HENRfQUEZ, de Gran Canaria:Universidad de las Palmas de Gran Canaria, Servicio de Publicaciones, 2000. Pp. 220. CHRISTAN (trans. and ed.). Martial, Buch 8: Einleitung. Text, UbersetzSCHL)FFEL, ung, Kommentar.Stuttgart:Franz Steiner Verlag, 2002. Pp. 723. SAMUEL SCOLNICOV, (ed. and trans.). Plato's "Parmenides."Berkeley-Los AngelesLondon: University of California Press, 2003. Pp. 193. $49.95. SHAPIRO,ALAN, and PETERBURIAN (trans.). Aeschylus: The "Oresteia." Oxford:
Oxford University Press, 2003. Pp. 285. $27.00. SLINGS,S. R. (ed.). Platonis "Respublica."Oxford Classical Texts. Oxford: Oxford University Press, 2003. Pp. 428. EFROSSINI. Readers and Writersin Ovid's "Heroides": Trangressionsof SPENTZOU, Genre and Gender. Oxford: Oxford University Press, 2003. Pp. 231.
LEO.On Plato's "Symposium."Ed. and with Foreword by SETHBENARSTRAUSS, DETTE.Chicago: University of Chicago Press, 2001. Pp. 294.
V. Athens and Macedon: Attic Letter-Cuttersof 300 to 229 B.C. TRACY,STEPHEN Berkeley-Los Angeles-London: University of California Press, 2003. Pp. 205. TRAPP,MICHAEL (ed.). Greek and Latin Letters: An Anthology with Translation. Cambridge Greek and Latin Classics. Cambridge: Cambridge University Press, 2003. Pp. 348. $70.00 (cloth); $26.00 (paper). WATSON,LINDSAY,and PATRICIAA. WATSON(eds.). Martial: Select Epigrams. In-
troduction,Text, and Commentary.CambridgeGreek and Latin Texts. Cambridge: CambridgeUniversity Press, 2003. Pp. 374. $70.00 (cloth); $26.00 (paper). Simone Weil's "The 'Iliad' or the 'Poem of Force"'":A Critical EdiWEIL,SIMONE. tion. Ed. and trans. JAMESP. HOLOKA.New York: Peter Lang, 2003. Pp. 130.
$19.95.