Antonaccio, Carla M. Contesting The Past - Hero Cult, Tomb Cult, And Epic In Early Greece.pdf

Contesting the Past: Hero Cult, Tomb Cult, and Epic in Early Greece Author(s): Carla M. Antonaccio Source: American Journal of Archaeology, Vol. 98, No. 3 (Jul., 1994), pp. 389-410 Published by: Archaeological Institute of America Stable URL: http://www.jstor.org/stable/506436 . Accessed: 08/05/2013 11:32 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access to American Journal of Archaeology.

http://www.jstor.org

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

Contesting the Past: Hero Cult, Tomb Cult, and Epic in Early Greece CARLA M. ANTONACCIO Abstract Greek hero cult has been extensively discussed by both archaeologists and philologists. This paper considers two current hypotheses: one links the development of hero cult in the eighth century B.C. with the circulation of Homeric poetry; the other views hero cult as a transformation of ancestral veneration in the context of the emergent polis. A review of the archaeological evidence for the Iron Age and Early Archaic period suggests that the earliest hero cult in the archaeological record emerged at Sparta during the eighth century. The small number of early hero cults, and their location and distribution, do not lend support to the theory of Homeric influence. Venerationof ancestors, on the other hand, was practiced widely in the Greek world throughout the Iron Age; it did not disappear with the emergence of the polis and hero cult. Rather than a single, unified concept, ancestral and hero cult articulated different versions of the past. Conflicting or competing concepts, both ritual and epic, serve to debate the past within and between communities. In Greece, as elsewhere, the debate helps to mediate social change within a frameworkof culturally determined rules.* Over a century ago, when Erwin Rohde published Psyche: The Cult of Souls and Belief in Immortality among the Greeks,' Homer was the chief authority

* This paper originated in the panel "Immortal Mortals," coorganized by Deborah Lyons and myself at the 1990 Annual Meeting of the American Philological Association, and from the first and last chapters of my doctoral dissertation, The Archaeologyof Early Greek "HeroCult" (Princeton Univ. 1987). My thanks to the panel participants (Deborah Boedeker, Alan Shapiro, and Rebecca Sinos) and audience on that occasion, and to Elizabeth Bobrick, Douglas Charles, Joseph Day, Susan Downey, Marilyn Katz, Ian Morris, Gregory Nagy, Susan Sherratt, and an anonymous AJAreviewer. The following abbreviationsare used below: Antonaccio C. Antonaccio, "The Archaeology of 1993 Ancestors," in C. Dougherty and L. Kurke eds., CulturalPoeticsin Archaic Greece(Cambridge 1993) 46-70. Antonaccio C. Antonaccio, An Archaeologyof An1994 cestors,Tomband Hero Cult in Early Greece(Lanham, Md. 1994). Farnell L. Farnell, GreekHero-Cultsand Ideasof (Oxford 1921). Immortality

for early Greek attitudes and practices. Archaeology offered only limited evidence for long-term patterns in Greek burial or cult. Since Rohde, most accounts of hero cult have attempted to reconcile literary pictures of cult in different periods, and, especially recently, to draw upon archaeology for ritual practice and the history of its development. The past few decades have witnessed a focus of archaeological research on the Greek Iron Age (or "Dark Age," from the end of the Mycenaean period in the 12th century to the late eighth century B.C.), a crucial, formative period for later Greek society. The place of epic poetry has changed; although epic poetry purports to describe the second millennium B.C., it is necessary to contextualize the poems in the Iron Age, the period in which they attained most of their final form. The pace of this work has intensified in the last decade, but even with the development of new approaches to Greek pre- and protohistory, virtually all attempts to describe Iron Age Greek society use epic poetry to explicate archaeological data and to provide detail for cross-cultural comparisons.2 Moreover, archaeologists and philologists have continued to find it difficult to harmonize the literary views of heroes and hero cult,

Hagg and Nordquist

R. Higg and G.C. Nordquist eds., Celebrations of Deathand Divinityin the Bronze Age Argolid (SkrAth4, 40, Stockholm 1990). Morris I. Morris, "TombCult and the 'Greek Renaissance': The Past in the Present in the 8th Century B.C.," Antiquity62 (1988) 750-61. G. Nagy, The Best of the Achaeans: Nagy Conceptsof the Hero in ArchaicGreek Poetry(Baltimore 1979). Rohde E. Rohde, Psyche:The Cultof Soulsand Belief in Immortality amongtheGreeks,8 trans. W.B. Hillis (New York 1925). 1 Originally published as Psyche:Seelencultund UnsterderGriechen(Freiburg 1898). blichkeitsglaube 2 Recent examples include I. Morris, "The Use and Abuse of Homer," ClAnt 5 (1986) 82-138; Morris, "Gift and Commodity in Archaic Greece,"Man 21 (1986) 1-17; and J. Whitley,"SocialDiversityin DarkAge Greece,"BSA 86 (1991) 341-65.

389 AmericanJournal of Archaeology98 (1994) 389-410

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

390

CARLAM. ANTONACCIO

and the sometimesdivergentevidenceprovidedby archaeology.3 Whileepic praise(Kk•og)is deliberatelytimeless,4 actual practice operates within a frameworkof space, time, and action,some of whichat least archaeologycan detect.Still,if we acceptthatthereis an oral traditionthat has roots in the BronzeAge, and that epic poetry is Panhellenicin its significance,Homer and his criticsare the place to begin any discussionof hero cult and a consideration of Greek attitudes toward the past. Both Homer and Hesiod already speak of past, more powerfulgenerationsof men, of honoring heroes and the importance of their tombs. Of course, Homer's heroes are living men in the poems, the rulersand warriors,divinelydescendPacotXig, ed but not themselvesdivinenor yet immortal.The referencesin the poemsto the heroes'honor (TL•tl), whichin cult is expressedas ritualaction,embrace theirprerogativesand possessions(see infra).Aside fromepic, the earliestwrittenreferenceto a cult of heroesostensiblydatesto the laterseventhcentury B.C., when the Athenian lawgiver Drakon prescribedthat gods and local heroes should be honored together according to ancestral custom. Referenceto customindicateswell-established practice,and the specificationof localheroesa multiplicity of them already,though we do not know how earlythiswasthe case.5 As founders(evenif mythical)of laterGreekfamilies, and of communitiesthatin turn identifiedheroic tombsand veneratedheroic relics,heroes are also ancestors.The writtensourcesemphasizethe importanceof heroictombsand otherrelics,and an 3 A. Snodgrass, The Dark Age of Greece (Edinburgh 1971), followed by V. Desborough, The GreekDark Ages Greece(Lon(London 1972) and N. Coldstream, Geometric don 1977). American excavations at Nichoria in Messenia produced a major study (W McDonald et al., NichoriaIII: DarkAgeand ByzantineOccupation(Minneapolis 1983), and a spate of work followed, with R. Hagg et al. eds., TheGreek Renaissanceof the8th CenturyB.C.: Traditionand Innovation (Stockholm 1983); A. Snodgrass, An Archaeology of Greece: ThePresentStateand FutureScopeof a Discipline(Sather Classical Lectures 53, Berkeley 1987); I. Morris, Burial and AncientSociety:The Rise of the GreekCity-State(Cambridge 1987); R. Hagg et al. eds., EarlyGreekCultPractice(Stockholm 1988); C. Morgan, Athletesand Oracles:The Transformation of Olympiaand Delphi in the Eighth CenturyB.C. (Cambridge 1990); and J. Whitley,Styleand Societyin Dark Age Greece(Cambridge 1991). New research continues to appear. 4 Nagy 117, 119, passim. 5 Porph. Abst.4.2: Ooi~gTLtivaKLLIpwCg it mpLX0)LoVg i~v

[AJA98

essentialconservatismand emphasison continuity generally govern scholarshipon Greek religion. Thus, the well-knownemphasison the importance of a hero'stomb and of his bones and other relics wasprojectedintothe BronzeAgeand the protohistorical period, and the origins of hero cult have been identifiedin BronzeAge burialpractices.For example, at Mycenae'sGrave Circle A Nilsson claimedcontinuityof worship,or atleastof memory, from Late Helladic to the historicalperiod. The concept of the hero as a semidivineor divinized figurewasextended to Mycenaeanrulers,both for theirown timeand after.6 Rohdeheld thathero cultand a beliefin an effective afterlifeare both absentfrom Homer,and that this omissionstands isolatedfrom the indigenous ancient beliefs of the Greeks originatingin the Bronze Age. He thought that the attitudetoward the afterlifein Homericpoetryrepresentedthe beliefs of the Ionian Greeks,and not the "popular beliefs"of allthe Greeks.Forhim,thisexplainedthe Homericconceptionof the dead as "withoutclear self-consciousness" so that it "neitherdesires nor wills anything. It has no influence on the upper world, and consequentlyno longer receives any shareof the worshipof the living."'ForRohde,the greatimportanceof heroesand heroworshipin the historicalperiodswasa revivaland amplificationof ancientand nativeancestorworship,not lateinventions.8A hero'spowerrestedin his bones and their burialspot, and worshipwas maintainedafter the Mycenaeanperiod"perhapsfor a long timeonlyby a few,in those placeswhere there remaineda cult attachedto a grave."9Withthe rise of the polis, the

Kotvl6

'a L Kat& 61'vacLtL,Wv iOpaPiotg TPXaIg i~ETeLoVg. Cf. R. TEXkdOVog v KaPJT•g New Gods:ThePolitics Athenian

irO'ivotgv6Cgotg

eJi'tUi~a KL

ReGarland, Introducing of ligion (Ithaca 1992) 45-46; ur a'd pLt might specify any practice with origins predating the 460s. See Nagy passim for the literary development. 6 MMR2 584-614, GGR3378-81, contra G. Mylonas, "Homeric and Mycenaean Burial Customs," AJA 52 (1948) 56-81; Mylonas, "The Cult of the Dead in Helladic Times," in G. Mylonas and D. Raymond eds., StudiesPresentedto DavidM. Robinson1 (St. Louis 1951) 64-105; Mylonas, "BurialCustoms,"in A. Wace et al., A Companion to Homer(London 1962) 478-88; see the papers in Hagg and Nordquist. 7 Rohde 24 and in general ch. 1. 8 Rohde 25: "the cult of Heroes everywhere has the same features as the cult of ancestors . .. the remains of a true cult of ancestors provided the model and were the real starting-pointfor the later belief and cult of Heroes." 9 Rohde 121, 123; emphasis in original.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1994]

CONTESTING THE PAST

ancestral worship of noble families broadened into hero worship. If original ancestral identities had been lost, heroes' names were fabricated, or their cults simply died out.1' Rohde's reconstruction of hero cult and its origin in ancestor cult depended on the importance of heroes' bones in literary sources. Although cult is absent, epic attitudes to burial and the stress on the hero's body ultimately culminated in a vigorous commerce in heroic bones and assorted other relics carried on by some Archaic and Classical poleis, as well as hero cults at Panhellenic sanctuaries." With Fustel de Coulanges, Rohde assumed the importance of ancestral graves to the Greeks based on a pastiche of written references and notions of ancestor cult derived from comparative anthropology. These ideas have been influential in the interpretation or use of excavated evidence as it became available over the past century.12 In 1921, Lewis Farnell criticized Rohde for what he termed "the chief defect" in his work: "he does not distinguish clearly between 'tendance' and cultus."'3To distinguish ancestral cult, he preferred the term tendance over worship for this behavior. Farnell agreed with Rohde that ancestors overlap with heroes, but he identified ancestors as purely local, while heroes could be located in more than one place. The actual worship of the dead, Farnell believed, can be evidenced from the eighth century B.C. onward; therefore, the festivals and observations that reflect "affection" and not awe were traces of a pre-Homeric attitude and mentality. Pointing out that hero cults are often not found in a hero's home territory, he inferred that they were not ancestral in origin. Although he allowed the possibility that some epic heroes received local cult before they were taken up by epic, his general view was that "much hero cult was directly engendered by the powerful influence of the Homeric and other epics ... one may discern that the old epic poetry not only suggested many a name to forgotten graves, [but] occasionally also imposed laws on the ritual."'4 10 Rohde 121, 123.

" E Pfister,Der im AltertumI-II. ReligionsReliquienkult und Vorarbeiten 5 (GieBen 1909-1912) Versuche geschichtliche is still standard; see also E.T Thompson, TheRelicsof the Heroesin AncientGreece(Diss. Univ. of Washington 1985) and Antonaccio 1994; E. Vandiver, Heroes in Herodotus (Leiden 1991) 34-38 with references on Orestes' relics. 12 Cf. Antonaccio 1993. 13 Farnell2; cf. 343 and 6. Rohde and Farnellwere both

especially concerned to refute the prevalent theory of the day, that heroes were faded or decayed gods who had

391

Farnell's category of heroes encompassed seven groups, covered in individual chapters and an appendix of ancient written sources.'5 The figures ranged from Herakles, both god and hero, to Titus Quinctus Flamininus, at the end of historical Greek autonomy. Criticism of Rohde continued with R.K. Hack, who took up the influence of epic in hero cult in an article published in 1929. Drawing upon Nilsson's work on Minoan and Mycenaean religion, he affirmed not only a Mycenaean worship of the dead but its continuation into hero worship in a later period.16 The only apparent exception to this development from Mycenaean to historic times, the attitude of Homer, Hack found not to stand outside the mainstream at all. Hack pointed out the various exceptions to Rohde's rule that Homer knew of no worship of the dead or of heroes. He detailed ancestor and hero cult among the Ionians, the existence of which for him precluded the notion that Homer's background could be responsible for the omission of any references to cult. He found instead internal reasons for the lack of hero worship. Since Homer is recounting the ideal, heroic past, as indicated by his use of various archaisms, hero cult is suppressed, because the heroes are shown as contemporaries and equals, and cult between them would be inappropriate. Hack concluded that hero cults are not Bronze Age survivals, but part of a continuous tradition to which the poems of Homer contributed. Meanwhile, more evidence for the Iron Age had become available. T.H. Price continued this line of argument, differing with Rohde in identifying evidence for cult within Homer: the sacrifice made by Odysseus in the underworld; references to the continued life and honor held by the Dioskouroi under the earth; the treatment of Erechtheus in the Iliad; and the topographical importance of ancient tombs in Homer. Price found fault with the assumption that hero cult is a Mycenaean tradition that somehow continued uninterrupted." She also recognized that purportedly earlier hero cults, those of degraded to the status of elevated humans. Rohde 117 n. 7 and Farnellch. 11. 14 Farnell 283-84. 15 Farnell 403-26. 16 R.K. Hack, "Homer and

the Cult of Heroes," TAPA 60 (1929) 57-74, 59. 17 T.H. Price, "Hero-Cult and Homer," Historia 22 (1973) 129-44: "There is no evidence of continuity,and no such cult was instituted inside a Mycenaean tomb before the 8th century B.C., according to the available archaeological evidence" (emphasis in original).

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

392

CARLAM. ANTONACCIO

[AJA98

Fig. 1. Prosymna. Finds from tombs VIII and IX. (Photo courtesy American School of Classical Studies at Athens) Pelops and Odysseus, for example, are not connected with Mycenaean tombs.'8 In her critique, she identified several points of confusion in terminology and concept in previous scholarship: hero cult and the cult of the dead were confounded, and the practice of hero cult not distinguished from a mere belief in heroes. She defined cult of the dead as "the burial rites and ceremonies after it" while hero cult is "a continuously repeated ritual over a long period of time."'9 Her own criterion for a hero cult, however, would seem to be compromised in regarding tomb cult as hero cult.20 The first thorough consideration of archaeological evidence for post-Bronze Age cult in Mycenaean tombs was Carl Blegen's 1937 study of Late Helladic tombs at Prosymna, site of the Argive Heraion.2' Nearly a third of the chamber tombs excavated un-

der his direction contained Late Geometric and Archaic pottery, and bronzes including phialai, bowls, and pins (figs. 1-2). Blegen interpreted this material as votive offerings to heroic ancestors, possibly by actual descendants of the Mycenaean families who had built the tombs. In support of continuity, which was unattested by an unbroken sequence of burials or of offerings, Blegen had to appeal to Protogeometric finds in Mycenaean tombs at Dendra and Thebes. Blegen's view followed that of Rohde, Farnell, and Nilsson: the Mycenaeans must have been Homer's epic heroes, and therefore Mycenaean graves were centers of heroic power. John Cook followed Blegen's work in the Argolid with his publication of a shrine founded in the eighth century at Mycenae, which he assigned to Agamemnon (fig. 3).22 Prior to this two other hero

18 See Price (supra n. 17) 131 and Cook's work, infra n. 22. The same observation about the citing of hero shrines is made by A. Snodgrass, "Poet and Painter in Eighth-Century Greece,"PCPS 204 (1979) 118-30, 12324, without reference to Price'sarticle. 19Price (supra n. 17) 129. 20 Price (supra n. 17) 143. On Erechtheus:II. 2.546-51; Dioskouroi: Od. 11.298ff., where they live on under the earth, but cf. 11. 2.243, where both are dead and there is no mention of continued life anywhere. Tombs as landmarks, e.g., 11. 2.604 (Aipytos), 2.793 (Aisyetes), 10.414,

11.371, 20.232 (Ilos); but cf. that of Myrine, an Amazon, not a hero. Discussion in Price (supra n. 17) 137-40 and Antonaccio 1994, ch. 3. 21 C. Blegen, "Post-MycenaeanDeposits in ChamberTombs,"ArchEph1937, 377-90. 22 J.M. Cook, "The Agamemnoneion," BSA 48 (1953) 30-68; Cook, "The Cult of Agamemnon at Mycenae,"in (Athens 1953) 112-18; A. FUpa Avroviov Kepaqy6sroAAov Foley, The Argolid800-600 B.C. An Archaeological Survey (SIMA80, Goteborg 1988) 151-53; and infra n. 51.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

CONTESTING THE PAST

1994]

393

~

-r?i?.~'-~' -i-il~i~iii--li:I-_i~iii -iiii?-ii:-:??: C5:"''''"?: ii::ii?;: ~:i-i-iiiii ?"'::'-Iii~i~-; ?..:..:._ ~:-I:-~,:~,~ :-~ ili: _%`-~:40 _-~:::ii::::,i~- i?-~:..~., i.i-~_-?.?:Xi~?. iii::i

~;f: :-:i_;~i:i`i-i~i;.

i ~ili isi?l?-: i ~i: .:

i , :;?;~?,

::_w??.

?,

.

_~: ~ ii912i~:~ai

.i

: ~i

iiiii

i i i i~-

:_

:

-:

:~--:i-i: b~e:;~

:

_i

:- -:- .:i-1;:~:

: :

gI

i(I:

- 1

~

-:lA"hri:11-1- _: :~:.- -

:-ii:-:-?-ii:-::i:i~iii~iR11,-

"'E---6i?i''i~i~i~l~~;:l~. ~. . . . . . . IFi i-i~ii-i..-i.i-:::--:?ir:-6-i'ii~~::l-;:~:_:- :i

?i-

:i _

i:

j: i:_ : ij: i_ :il i

?E:-iii-iiiii:-ii

U15:~~~-:::~~~

ii

i -:.'i

i-

i

8'4?~ ~

Il~i i i i :i ??f?

:i-l i ?i

- i-

:

f%:

1%

_

Fig. 2. Prosymna. Finds from tomb IX (nos. 1203-1205) and tomb XLII (nos. 911-12). (Photo courtesy American School of ClassicalStudies at Athens)

shrines had been archaeologicallyinvestigated: one for Odysseus on Ithaca, and another for Menelaos at Sparta (figs. 4-5).23 Two points emerged:

the

shrine at Mycenae (like the other two) was situated well away from the chamber and tholos tombs, contrary to Pausanias's testimony that the graves of

Agamemnon and his followers were shown inside the walls.24 Second, Cook, following Farnell, theo-

rized that this and other hero cults were the direct result of the circulationof the Homeric poems, since the shrine was founded in the eighth century B.C. He did not, therefore, feel obliged to argue continuous memory; in fact, the opposite. Homer's description of Agamemnon's murder outside the walls provided an explanation for the location of the shrine. To Cook's mind, the cults were begun "by

Fig. 3. Mycenae. Shrine of Agamemnon. (Photo C. Antonaccio)

23 See infra n. 50.

24 PaUS. 2.16.3-7.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

CARLAM. ANTONACCIO

394

[AJA98

Fig. 4. Ithaca. Polis Bay. (Photo C. Antonaccio) people who preserved no continuity of memory-and little enough of blood-some centuries after the occupants had passed into oblivion."'25 Nicolas Coldstream'sJHS article in 1976 also emphasized the influence of epic in the development of hero cult as proposed by Farnell in 1921, and combined it with the most comprehensive collection of excavated evidence to date.26 Coldstream collected the votives deposited in Mycenaean tombs all

over Greece during what he defined as the "Age of Homer" (750-650 B.C.) (fig. 6).27 For Coldstream, hero cult was being practiced if two conditions applied: a significant hiatus existed between the last use of the tomb and a votive deposit, and deposited objects were not associated with a later structure or grave.28 Furthermore, in areas where tholoi or chamber tombs remained a tradition after the Bronze Age (as in Thessaly and Crete), Coldstream

Fig. 5. Sparta. Menelaion. (Photo Deutsches Archiologisches Institut, Athens, neg. Sparta 195)

25 26

Cook, in Frpag (supra n. 22) 115.

N. Coldstream, "Hero-cults in the Age of Homer," JHS 96 (1976) 8-17. 27 Coldstream (supra n. 26) 10: "it remains true that no

offerings anywhere-to my knowledge-are earlier than the third quarterof the eighth century." 28 Coldstream (supra n. 26) 9.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

CONTESTING THE PAST

1994]

395

pt

B

4

.4 e

Fig. 6. Distribution of hero cults, 750-650 B.C., according to N. Coldstream. (After N. Coldstream, JHS 96 [1976] 12, fig. 1) found no later votives in the Mycenaean tombs during the century under consideration. For the areas where votives do occur in Mycenaean tombs, Coldstream pointed to the change in burial from multiple-use chamber and tholos tombs to single graves. Only burials that employed unfamiliar practices would prompt any show of respect, but not until Homer had primed interest.29 Coldstream believed the currency of epic would account for the occurrence of votives in Dorian areas, where an alien, immigrant population might otherwise be expected to show no particular reverence for Mycenaean predecessors. While Coldstream believed that tomb cults were honors of private and ordinary Greeks to unidentified heroes, he saw in the shrines of Menelaos at

Sparta and Agamemnon at Mycenae early Dorian attempts to co-opt epic heroes."0 In support of this view he cited "heroic burials" that he believed showed Homer's influence, such as that of Amphidamas of Chalkis (Hes. Op. 654-59), and the early seventh-century B.C. West Gate burials at Eretria, warrior cremations in bronze urns, accompanied by weapons.31 Finally, he proposed that hero cults that he believed preceded his "Age of Homer," such as those of Erechtheus and Akademos at Athens and Odysseus on Ithaca, depended on racial continuity. In response to Coldstream, Price broadened her earlier work, identifying 10 cults that antedate Homer (dating Homer to 750 B.C.).32 Price looked to the Indo-Europeans for the ultimate origin of ancestral veneration, and identified saga and the

29 Coldstream (supra n. 26) 14. 30 Coldstream (supra n. 26) 15-16. 31 On the West Gate burials, C. Berard, Eretria III: L'Hir6ona' la Portede l'Ouest(Bern 1970); and Berard, "R6cupererla mort du prince: heroisation et formation de la cite," in G. Gnoli and J.-P Vernant eds., La mort:lesmorts dans les socitics anciennes (Cambridge 1982) 89-105; cf.

Price (supra n. 17) 143 concerning "heroicburials":"Most probably certain elements went both ways: from common burial practices to hero-cult and from hero-cult to special important burials."Cf. Antonaccio 1994, ch. 5. 32 TH. Price, "Hero Cult in the 'Age of Homer' and Earlier,"in G. Bowersock et al. eds., Arktouros:Hellenic StudiesPresentedto B. Knox(Berlin 1979) 219-28.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

396

CARLAM. ANTONACCIO

[AJA98

LIV

"CID

i.4

Fig. 7. Distribution of tomb cult at Bronze Age tombs, 1050-600 B.C. (C. Antonaccio) desire to define identity as incentives for hero cult.33 Price criticized Coldstream for using too low a lower limit for his evidence, for conflating hero cult and the tomb cults, and for including propitiation and accidental intrusion into Mycenaean tombs. She observed that these cults, ostensibly engendered by epic, did not seem to be directed to epic heroes.34 Bronze Age tombs are locations where hero cults could be expected to develop, given the written evidence, and since Coldstream's summary, it has widely been accepted that the later material found at them consists of votive offerings to heroes. In the last few years the tomb evidence has been the subject of continual comment, and the term tomb cult often employed to differentiate it from hero cult at formal shrines. Tomb cult is used of family vis-

its to graves in the Classical period by Humphreys; Farnell called this tendance.35In general, the later material in and around Mycenaean tombs includes offerings of pottery and occasionally small bronzes and figurines, evidence of activities that took place at the tombs, including ritual eating and drinking (fig. 7; see infra). Snodgrass, Whitley, and Morris have all discussed this phenomenon in recent years, deemphasizing Homer's influence on these practices and appealing more to regional historical and social factors (infra).36 Yet the objects of tomb cult are still often assumed to be heroes. In fact, the canonical concept of the rise of hero cult during the eighth century is due to the later material in Mycenaean tombs, rather than the founding of hero shrines per se.

33 Price (supra n. 32) 228: "The exact time of intersection between mythology and actual practice or worship is still to be discovered." Snodgrass also responded to Coldstream: A. Snodgrass, Archaeology and the Rise of the Greek State(Cambridge 1977) esp. 30-31. 34 Price (supra n. 32) 220. 35 S. Humphreys, "FamilyTombsand Tomb Cult in Ancient Athens: Tradition or Traditionalism?"JHS 100

(1980) 96-126; and R. Garland, The GreekWayof Death (London 1985); see also Antonaccio 1994. Morrisuses the term, as does S. Alcock, "TombCult and the Post-Classical Polis,"AJA95 (1991) 447-67. 36 Supra n. 35. Snodgrass (infra n. 40). J. Whitley, "Early States and Hero Cults: A Re-appraisal,"JHS 108 (1988) 173-82 does not make a clear distinction between hero cult and tomb cult.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

CONTESTING THE PAST

1994]

Although archaeologists were now focused on the Iron Age and on new data and approaches, epic remained crucial. Both Snodgrass and Morris have drawn on the work of Gregory Nagy, who has argued that hero cult was not a feature of later Greek

397

mentary aspects of the hero's due. KXkog,not generally used of present time in epic, requires memory (the oral tradition) while TLuLitis the communal ac-

knowledgment of a hero's status.41The

fjltog or

3t6ktg, observer-participants, offers this recognition,

religion instigated by epic, but "a highly evolved

and sacrifice and feasting play a key role in this

transformation of the worship of ancestors within the social context of the city-state.'"3 This transformation in practice mirrors one in poetry, which moved from stories about ancestors to the epic celebration of heroes; while hero cults are local, epic is Panhellenic.38Drawing in turn on the work K•kog of Snodgrass, Nagy emphasized the emergence of epic in the eighth century along with the polis and Panhellenism.39The apparent absence of explicit hero cult in the Homeric poems remarked upon by Rohde is, then, due to the importance of heroes beyond their home locale, and does not imply that the ideology underlying the cults is also absent. In this model, the local and Panhellenic aspects of heroes gradually fuse, so that epic heroes, too, eventually receive cult as local figures.40 Nagy suggested that celebration of heroes may be

semantic field. In the context of ritual feasting, t•Lti is expressed by the yipag, an "honorific portion" of the sacrifice.42 At the end of the eighth century,

framed in terms of ckXog, the glory or reknown con-

ferred by epic poetry, and

uTqi,

honoring heroes

with cult. Epic K~ktogand cultic TLrtiare two comple-

37 Nagy 115, building on Rohde. 38 See also Rohde, and Nagy 166-67, 284, etc., makes epic a determining influence on practice. The erroneous notion of a primitive and continuous ancestral cult for the Greeks, on which the transmission of property was based, goes back to N. Fustel de Coulanges, TheAncientCity(Baltimore 1980), originally published in 1865 (see S. Humphreys' and A. Momigliano's introduction to the 1980 edition). 39 Nagy 166-67 citing Snodgrass; see also Rohde, ch. 1.

On Panhellenism,see below.

40 Referred to in turn by Morris 752-54 and A. Snodgrass, "The Archaeology of the Hero," Aion. Annali del seminariodi studidel mondoclassico,sezionedi archeologiae storiaantica 10 (1988) 19-26, esp. 20-21. 41 Nagy 118 ?1 n. 2 and 148-50 in general on also Ty!il; Nagy, GreekMythologyand Poetics(Ithaca 1990) 132-38. Kkog as recompense for death: Nagy 119; cf. C. Segal, AntCl52 (1983) 24"Kleosand Its Ironies in the Odyssey," 47. Odysseus refers, exceptionally,to his own Hom. K:xog: and Od. 9.19-20; Penelope to hers, 19.127. On Penelope see now M. Katz, Penelope'sRenown,Meaningand K:xog, in theOdyssey(Princeton 1991). K)Xogis used Indeterminacy by Pindar of the living athlete: Nagy 1990 (infra n. 53) 150-51, but TLI"iis regularly translated as "honor."See in general 146-98, but cf. L. Kurke, TheTrafficin Praise(Ithaca 1991) 16-19 and passim. The physical counterpart of KXEog,rather than cult, is the oflta: see M. Lynn-George, Epos:Word,Narrativeand theIliad (AtlanticHighlands, N.J. and Poetics 1988), esp. 153-229, and Nagy, GreekMythology

therefore, a simultaneous celebration of heroes in epic and cult emerges, at the time both Homer and archaeologically visible hero cult are now generally placed. This elegant formulation is, however, not exclusive to heroes and hero cult.43 By a wider definition, It[ti within the poems refers to the communal recognition offered to individuals ascribed a given level of status, and yicpag refers to the exact

measure of this recognition.44 So far, the two main theories on the origins of hero cult and the role of Homer can be summarized as follows: 1) Homer's omission of hero cult and his attitudes toward the dead are anomalous; there is continuity between a native ancestor cult and the hero cult that develops from it over time; 2) Homer inspires hero cult among mainland Greeks (Dorians), who have no connection with the Mycenaeans

A (supra) 202-22. Heroic stature expands in the Odyssey. bard like Demodokos is also called a hero; Odysseus, howand she is ever, promises Nausikaa 0C 6g ~; eXeoC0t41iv not a "heroine"(Od.8.467). Historically,founders of cities may become heroes, though few are warriors. See Nagy 149, and GreekMythologyand Poetics133, noticing where heroes get TtLil 1gOE6vwhen they function as "priest"or "king";on founders, see I. Malkin,GreekReligionand Colonization(Leiden 1987). 42 Nagy 132, though see W. Burkert, GreekReligion (Cambridge, Mass. 1985) 199-203 on chthonic cults. also describes Briseis, the prize captive in 43 E.g., cremation is the ypca' II. book 1;y'pag pBav6vTwv; to command is for Nestor the yFpcCgyep6vorvy (Il. 4.323), all part of a whole system of heroic prerogatives. See especially on the ypctg OCtv6vTwvR. Garland, "Gerasthanonton:An Investigation into the Claims of the Homeric Dead," BICS 29 Simi(1982) 69-80; and Nagy 132 on Briseis as a ygpag. are varied in nature and kind. As James larly, TCt•i McGlew has recently pointed out, TCL[iwithin epic is the prerogative of the living; he stresses that connected "with material objects of value, its quantity is naturally limited. One man's gain of T[Uil implies another's loss" ("Royal Power and the AchaeanAssemblyat Iliad 2.84-393," ClAnt 8 [1989] 283-95, esp. 286-87, ns. 7-8). 44 Nagy 1990 (supra n. 41) 137. As Nagy points out to me, however, characters sometimes speak from the audience's viewpoint, which is exemplified by the speech of Sarpedon to Glaukos in Il. 12.310-21.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

398

CARLAM. ANTONACCIO

[AJA98

and very different burial practices;this leads to both the tomb cults and hero cult, the latter an attempt to co-opt the heroes of epic. The mutual reliance of archaeologists and philologists on each other's data, however, whether using Homer to determine hero cult or archaeology to explicate Homer, leads to circularity.A contextual approach to the Iron Age produces two results: first, a diachronic perspective that indicates change, tension, and different patterns in space and time; and second, a context within which to locate symbols and consider how their meanings change.45The term cult identifies a pattern of ritual behavior in connection with specific objects, within a framework of spatial and temporal coordinates. Ritual behavior would include (but not necessarily be restricted to) prayer, sacrifice, votive offerings, competitions, processions, and construction of monuments. Some degree both of recurrence in place and repetition over time of ritual action is necessary for cult to be enacted, to be practiced.46Such factors distinguish a cult of a god or hero from occasional rituals, such as apotropaic gestures. Of course, the location and even specific object of ritual may vary and yet a cult may be spoken of: multiple cults of the gods in different aspects and locations, and heroes in multiple locations. It is in this sense that cult is used in the term hero cult.47Ritual action

and cult as so defined are in part at least archaeologically visible;the practiceof ancestor or hero cult leaves a recoverable trace.48A complete archaeological record has not been and cannot be recovered, however, nor do all activities (much less ideologies) leave a recoverable trace. Archaeology is critical for pre- and protohistory. Arguing backward from the practices of the Classical period leads to unwarranted conclusions about these periods, and the written record is biased and incomplete, even for the period that produced it. When dealing with the written record in particular, we should keep in mind what Morriscalls the "Nuer paradox," in which a society says one thing and does another.49Hero cult is a prime example of a case in which the written evidence and ritual behavior as recovered by archaeology positively collide. First, despite the prominence of heroes in the written record, securely identified early (eighth century) hero shrines are few in number and peculiar in pattern. Shrines to Helen and Menelaos, and perhaps one to Agamemnon together with Kassandra or Alexandra, are both located in Laconia. At Mycenae, the shrine of Agamemnon has already been mentioned; on Ithaca, a shrine for Odysseus has been claimed in the cave at Polis Bay.50Only Laconia, however, has epigraphical evidence from the Archaic period as-

45Archaeological theory has moved beyond New (or Processual) Archaeology and is currently struggling to bring history back into archaeology: for a critique, see T. Patterson, "History and the Post-Processual Archaeologies,"Man 24 (1989) 555-66, and recently I. Hodder, "Interpretive Archaeology and Its Role," AmerAnt56 (1991) 7-18 with references, for a "contextual"approach, and a similar emphasis on hermeneutics for classical archaeology by G. Gibbon, "ClassicalArchaeology and Anthropological Archaeology: A Coming Rapprochement?" in N. Wilkie and W.Coulson eds., Contributions toAegeanArchaeology(Minneapolis 1985) 283-94. 46 On ritual in burial practices, see the remarks of I. Morris,Death-Ritualand SocialStructurein ClassicalAntiquity (Cambridge 1992) 8-15, and cf. C. Bell, Ritual Theory,Ritual Practice(Oxford 1992). 47 See Antonaccio 1993 and cf. Price (supra n. 32) 220: in tomb cults, "many deposits contain vases of one period, therefore how can one talk about instituted continuous cult?" 48 On this see, e.g., C. Renfrew, TheArchaeology of Cult: The Sanctuaryat Phylakopi(BSA Suppl. 18, London 1985) for the Aegean. 49 Morris(supra n. 46) 7-9, esp. 6: "The social structure we are born into and socialised within is a set of assumptions about what we should say, do, and even think in given situations, but it does not determine our behaviour. Everything we do is informed by learned social structure, but the structure itself is only transmitted through time and space by real people as they repeat what they them-

selves have learned, or react against it." It should be said that Farnelldid admit change over time in the category of

the hero.

50 For the Menelaion, shrine of Helen and Menelaios, at Sparta, see Droop et al., "I Laconia: I. Excavationsat Sparta, 1909. ?6. The Menelaion," BSA 15 (1908-1909) 108-57; H. Catling, "Excavationsat the Menelaion, 19731975," LakonikaiSpoudai2 (1975) 258-69; "New Excavations at the Menelaion, Sparta," in U. Jantzen ed., Neue Forschungenin griechischerHeiligtiimern(Tuibingen 1976) 77-90; "Excavationsat the Menelaion 1976-1977," LakonikaiSpoudai3 (1977) 408-15; "Excavationsat the Menelaion, Sparta, 1973-76," AR 1977, 24-42; "Study at the Menelaion 1982-1983," LakonikaiSpoudai7 (1983) 23-30; "Sparta:A Mycenaean Palace and a Shrine to Menelaus and Helen," CurrentArchaeology130 (1992) 429-31; H. Catling and H. Cavanagh, "Two Inscribed Bronzes from the Menelaion, Sparta," Kadmos 15 (1976) 145-57; W. Cavanagh and R.R. Laxton, "Lead Figurines from the Menelaion and Seriation," BSA 79 (1984) 23-36; R. Catling, "Excavationsat the Menelaion: 1985," Lakonikai Spoudai10 (1986) 205-16; R. Catling, "AVotiveDeposit of Seventh-Century Pottery from the Menelaion," and R. Tomlinson, "The Menelaion and Spartan Architecture," in J. Sanders ed., $IAOAAK?2N. LakonianStudiesin Honour of Hector Catling(London 1992) 57-75 and 247-56, respectively. For Amyklai, see G. Salapata, "Lakonian Plaques and Their Relation to the Stone Reliefs,"in Akten des XIII internazionalenKongressfiir klassischeArchdologie 1988 (Mainz 1990) 525; Salapata, "Pausanias3.19.6: The

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1994]

CONTESTING THE PAST

399

signing its shrines to their respective heroes. The dedication of the shrine at Mycenae to Agamemnon during the Iron Age has been strongly challenged: though the offerings begin at the end of the eighth century, the two graffiti naming Agamemnon as the recipient are at least fourth century in date, and an early fifth-century destruction of the shrine took place in between.51 The location of the Ithacan cave and its offerings of bronze tripods and cauldrons fit the epic tradition very well, but this shrine provides no evidence for an Iron Age dedication to Odysseus.52 Although their foundations date from the Iron Age, then, two of these shrines are not securely connected with heroes from their inception. Nor are they founded by leading poleis, although the archaeological record for this type of sanctuary is very lacunar: little is known of arrangements for cults of Erechtheus at Athens, for example.53 Even more important, however, this handful of shrines is not located at Mycenaean tombs, as to be expected from accounts that emphasize the importance of physical relics in hero cult, especially their bones,

which locate the hero in a particular place or places.54 The early evidence for hero cult at sanctuaries of Olympian deities is not at all clear, either, though Greek athletic contests are said to originate in funeral games for heroes. For example, Alfred Mallwitz has shown that cult at the Pelopeion in Olympia belongs to the Archaic period, and heroes at other Panhellenic sanctuaries have proven to be similarly absent in the Iron Age.55 The preceding evaluation is made on a narrow criterion for hero cult: its identification by inscription or other written source. Types of offerings alone cannot securely identify any cult, Olympian or other, nor can location; as we have already seen, early hero shrines are not located at tombs, though our written sources lead us to expect this. It will seem contradictory to downplay the testimony of written sources, and then rely on them for such information, but we will never break out of circularity if we assume that any later activity at a Mycenaean tomb must be hero cult. If a cult was founded in the Iron Age and attracted inscribed dedications in the

Sanctuary of Alexandra at Amyklai,"AJA95 (1991) 331; and the reports in Ergon 1956, 100-104; 1957, 12-13; 1957, 548-51; 1960, 167-73. Mycenae: Cook (supra n. 22) and see infra n. 51. Ithaca: S. Benton, "Excavationsin Ithaca, III. The Cave at Polis, I," BSA 35 (1934-1935) 45-73; Benton, "AVotive Offering to Odysseus,"Antiquity 10 (1936) 350; Benton, "Excavationsin Ithaca, III. The Cave at Polis, II," BSA 39 (1938-1939) 1-51; on the pottery, see W. Coulson, "The 'Protogeometric' from Polis Reconsidered,"BSA 86 (1991) 42-64. 51 C. Morgan and T Whitelaw, "Pots and Politics: Ceramic Evidence for the Rise of the Greek State,"AJA95 (1991) 79-108, esp. 89 challenge the dedication to Agamemnon in the Archaicperiod as I did in my dissertation in 1987 (see also C. Antonaccio, "Tombs, Terraces, and the EarlyArgive Heraion," Hesperia61 [1992] 85-105 with references). E de Polignac'sdoubts in La naissancede la citegrecque(Paris 1984) 130-31, n. 12 are based on the presence of female terracottas among the finds, but these do not provide an identification or contradict one. 52 Hom. Od.book 13 describes the cave where Odysseus hides his Phaiakian treasures, which include 13 tripods, with Athena's help; the entire case rests on a single graffito from the Hellenistic period, whereas there is other earlier epigraphical evidence for Athena, Hera, and the Nymphs; see Antonaccio 1993. 53 Nagy 7, with notes; Nagy 1990 (supra n. 41) 36-82; Nagy, Pindar'sHomer (Baltimore 1990) 52-115; see also Antonaccio 1993, 61-62. Panhellenism as evidenced by interregional sanctuaries like Olympia will now have to be considered in light of Morgan (supra n. 3) and Morgan, "The Origins of Pan-Hellenism" in N. Marinatos and R. Higg eds., Greek Sanctuaries,New Approaches(London 1993) 18-44, who challenges the early (i.e., eighth century) importance of these centers far beyond their regions. On the other hand, her work makes clear that elite inter-

action and competition took place long before the rise of the polis. 54 Snodgrass (supra n. 18) 124 (also that hero cult could proliferate to several locations). See my article, "Marking Time: The Bronze Age in the Cultic Topography of Early Greece," in S. Alcock and R. Osborne eds., Placing the Gods: GreekSanctuariesin Space (Oxford, forthcoming), though the earlier work ofA. Nock, "The Cult of Heroes," HThR 37 (1944) 141-74 and W. Fergusson, "The Attic Orgeones," HThR 37 (1944) 61-130 indicated that hero cult did not necessarily require a tomb. 55 See Nagy (supra n. 53) 119-29 on Panhellenic games, especially the Olympics. On Pelops see A. Mallwitz, OlympiaundseineBauten(Munich 1972) 133-37; and Mallwitz, "Cult and Competition Locations in Olympia,"in W. Raschkeed., TheArchaeology of theOlympics(Madison 1988) 79-109, as well as H.-V.Hermann, "Pelopsin Olympia,"in N. (Athens 1980) Zrg&e:T6o! Et Mvlv?) KovroA•ovro; that a 59-74. Recent excavation demonstrating prehistoric tumulus did in fact reside beneath the Classical shrine does not alter this conclusion: see H. Catling, AR 1987-1988, 24; E. French, AR 1989-1990, 30, and AR 1990-1991, 24; and H. Kyrieleis,"Neue Ausgrabungen in Olympia,"in H. Kyrieleisand W Coulson eds., Symposium on theOlympic Games(Athens 1992) 19-24. Palaimon/Melikertes at Isthmia: E. Gebhard, "The Evolution of a Pan-hellenic Sanctuary:FromArchaeology to History at Isthmia," in Marinatosand Higg (supra n. 53) 154-77, esp. 170-72; Archemoros at Nemea: S. Miller et al., Nemea.A Guideto the Museumand Site (Berkeley 1990); Neoptolemos/Pyrrhos at Delphi: J. Pouilloux, FdD II: Topographie et architecture. La regionnord du sanctuaire(Paris 1960) 49-60; J. Fontenrose, "The Cult and Myth of Pyrrhos at Delphi," CSCA4 (1960) 191-261; and E Stihler, "Die Lesche der Knidier--ein Heroon des Neoptolemos?"Boreas12 (1989) 15-16.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

400

CARLAM. ANTONACCIO

[AJA98

00

OF

t2tcA

co

.0Colo ?~a ?~P

..,:

,,

?c

Fig. 8. Reuse of Bronze Age tombs, 1050-600 B.C. (C. Antonaccio) Archaic period, as at the Menelaion, then its attribution to a particular hero from the beginning is probably acceptable. Otherwise we overlook the fact that Greek ritual was a dynamic, changing category, which underwent important periods of active generation and which in turn created change. The recipients of tomb cult are anonymous; if they were named by participants, whether as generic hero or specific character, there is no record in any instance. Descriptions of heroic relics and their recovery are very different (see infra), and so the practice, interestingly, does not appear in the written sources on cult. The locations of the few bonafide hero shrines away from tombs, the anonymity of tomb cults, and their lack of continuity and regularity indicate a different and separate phenomenon

from hero cult. Another factor in tomb cult is that Bronze Age tombs also receive new burials in the Iron Age and later. Such burials occur throughout the Argolid and at other sites across Greece (fig. 8).56 This is especially significant because it suggests a connection with funerary practice, and calls into question the interpretation of the votive deposits as offerings to heroes. Tomb cult is a type of ancestor cult, which, however, in returning to Bronze Age tombs creates ancestors by the adoption of ancient dead unrelated by linear descent and unacknowledged for centuries.57 Although the Greeks did speak of ancestors using a variety of terms, there is less written evidence for a cult of ancestors than for hero cult.58In the historical period, the Greeks practiced what Humphreys

56 See Antonaccio 1993 and 1994, passim for the Argolid, Attica,and Laconia. Some of the burials were noted by Snodgrass and Coldstream but their significance was not brought out; cf. Snodgrass 1971 and Coldstream (supra n. 3) 14. 57 For archaeological visibility,Morris (supra n. 3) 97109; Morris, "The Archaeology of Ancestors: The Saxe/Goldstein Hypothesis Revisited," CambridgeArchaeologicalJournal 1 (1991) 147-69; and Antonaccio 1993. I am indebted to Ian Morris for sharing his manuscript

before its appearance in print. I hope to address elsewhere the issues he raises there. 58 Terms include yovig, jTp6yovot, urpon;d~opEg, uptsee Antonaccio 1993. A festival ToMdropEg, GPLTuorTcvpELg; called Genesia, an annual celebration of the dead at Athens (and possibly elsewhere), was perhaps connected with the genos:see E Jacoby,"FENEIIA:A Forgotten Festivalof the Dead," CQ 36-39 (1944) 65-75; and D. Kurtz and J. Boardman, Greek Burial Customs (London

with references.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1971) 147-48

1994]

CONTESTINGTHE PAST

401

calls "memorialism"rather than ancestor worship; they were motivated (at least when they write down their thoughts on the subject in the Archaic period and after) by a desire to be remembered by family and passersby. The Greeks of historical periods, however, did not routinely immortalize or divinize their family dead. Moreover, for most Greeks genealogies did not articulate the generations of the dead or structure the relations of the living.59Yet, as discussed already, several scholars have proposed that ancestors were local and familial, the founders of a clan, for example, and that their descendants maintained a cult at their graves. Famous figures could receive worship as both ancestors (family)and heroes (community), either in the same or different places. Ancestor cult was suggested as the original model; when some ancestors became heroes, the practice became more general, or some originally unheroic ancestors were elevated to heroic status after the cult of heroes had taken hold.60 Heroes, then, could be considered as ancestors, especially when claimed by members of an elite; Nagy, too, observed that heroes could be the completely unhistorical ancestors of a kinship group, like a genos,and he proposed that hero cult grew from worship of ancestors; hero cult was a "revival of a continuous heritage."61For Farnell, the earliest clear-cut evidence for ancestor cult was only sixth century in date, and he denied that the Greeks generally worshipped the dead.62Since the early work of Rohde, Farnell, and others, ancestors all but dropped out of the discussion until the last 10 years. For this the extension of hero cult to Mycenaean tombs with the increase of archaeological evidence is largely responsible. Testing the practice of hero cult and the assumption that hero cult is based on ancestor cult depends

on archaeology. The locations, use, and reuse of graves, and the ritualsconnected with them provide the archaeologically visible data for the acknowledgment of ancestors. Obviously, archaeology cannot recover oral (and many written) accounts of kinship or genealogy, nor such practices as prayer, or some types of offerings and sacrifices that leave few or no detectable traces. It can, however, recover burial patterns and the frequency and intensity of some ritual behavior. The Bronze Age Greeks mostly practiced group burial in chamber and tholos tombs and in tumuli, perhaps based on the extended family. The procedures that were followed in reopening a tomb for multiple use during the Bronze Age long ago led George Mylonas to conclude that there was no contemporary hero or ancestor cult; he pointed out that the remains of earlier burials were often accorded no respect at all when the tomb was reused, and tombs were regularly robbed as well.63What Mylonas and others saw as disrespect and evidence for looting, however, is at least in part due to the practice of secondary burial, where the remains of earlier interments were collected and placed in pits in the floor of the tomb, or swept aside.64 The Mycenaeans, though, also showed respect for the locations of earlier burials: a familiar example is the incorporation of Grave Circle A within the fortificationsof Mycenae, and much earlier tombs were also reused, even by the Mycenaeans. It is becoming clearer that such treatment of monuments in the Bronze Age, too, was an attempt to legitimate the present using the past.65 From the Bronze Age onward, engagement with dead kin is limited, in burial custom and memory both, ordinarily not extending back beyond the third generation. In burial practices as known from several areas of the Greek world, this generally

59 S. Humphreys,"DeathandTime,"in S. Humphreys and H. King eds., Mortalityand Immortality: TheAnthropology and Archaeologyof Death (London 1981) 261-83, esp. 269-70 referring to M. Friedman's work. See also Antonaccio 1993, 47-48.

to 200 B.C. (Farnell 353). 63 Mylonas (supra n. 6) argued forcefully against either ancestor or hero cult among the Mycenaeans;see now, on Bronze Age burial practices, Higg and Nordquist. On the familial basis for the use of chamber tombs, see W. Cavanagh, "Citiesand Synoecism,"in J. Rich and A. Wallace-Hadrilleds., Cityand Countryin theAncientWorld(London 1991) 97-118; C. Mee and W Cavanagh,"The Spatial Distribution of Mycenaean Tombs,"BSA 85 (1990) 22543. 64 B. Wells, "Death at Dendra: On Mortuary Practices in a Mycenaean Community," in HFigg and Nordquist 125-40. 65 Antonaccio 1994, ch. 2; Mee and Cavanagh (supra n. 63) 242; and also E. Protonotariou-Deilaki, "Burial Customs and FuneraryRites in the PrehistoricArgolid,"in Higg and Nordquist 69-83; E. French, "'Dynamis'in the Archaeological Record at Mycenae,"in M. Mackenzieand C. Roueme eds., ImagesofAuthority(PCPS Suppl. 16, 1989) 122-30.

60

61

Farnell 343-44.

See Nagy 115-16, and esp. E Bourriot, Recherches sur la naturedugenos.Etuded'histoire socialeathenienne-periodes archaiqueet classique(Diss. Universite de Paris 1976) and D. Roussel, Tribuet cite (Annaleslittirairesde l'Universitede Besangon193, Paris 1976); add now E. Kearns, TheHeroesof Attica(BICSSuppl. 57, London 1989). 62 Farnell 355. Although Farnell devoted a chapter to "The Cults of Ancestors," he drew most of his evidence from later Athenian practicesand by conflating sources on other Greek communities from all periods. For example, the will of one Epikteta, which specifies sacrifices at a hero6n to the family's"heroicspirits,"is called "one of our chief documents concerning the Greek worship of the dead" but it comes from the island of Thera and dates only

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

402

CARLAM. ANTONACCIO

means that a chamber tomb or group of graves is used seriatim for just a few generations. Although a single cemetery might see centuries of use, graves and tombs are continually covered over or cut through.66 In memory, too, descent usually does not extend beyond the father's father. This is apparent in the long term as well: there is little if any trace of regular, long-term veneration at tombs from the Bronze Age to the Classical period.67 This excludes for the moment mythic, epic, or civic ideologies of descent. There were, of course, families (yevfl) in the historical period that did use ancestors as defining points in genealogies, and who claimed descent from heroes. The point is that this is not true of all Greeks, only a few families.68 Yet, all Greeks had ancestors, in the acknowledged ties with past members of their oikos. To be fully understood, the tomb cult data should be coordinated with patterns of burial practice. Regional differences should be considered in detail, to bring out local patterns or distinguish widespread trends. Space prohibits such discussion here; the data are available elsewhere, and some conclusions can be drawn.69 Tomb cult is a sporadic, though recurring, feature, and therefore we confront again the terminology of cult. No single grave, or even single cemetery, is continuously venerated or maintained from the Bronze Age throughout the Iron

66 See

Humphreys (supra n. 35). 67This is a highly simplified summary of the evidence: see Morris (supra n. 3) 90-91; Bourriot (supra n. 61) 1,178; Roussel (supra n. 61); and Humphreys (supra n. 35); see also W. Cavanagh and C. Mee, "The Location of Mycenaean Chamber Tombs in the Argolid,"in Higg and Nordquist 55-64; Cavanagh (supra n. 63) for the Bronze Age. Extensive genealogies do exist in the Greek world for certain gene in the historical record and in epic, but they are exceptional; see Bourriot (supra n. 61) for discussion. 68 This is the crucial result of Bourriot's work (supra n. 61). I cannot agree with Nagy's insistence (supra n. 53) 152-58 on early Greece as a "tribal"society,"detribalized" by the polis. As he puts it, "inherited ideologies and practices concerning ancestors-a key determinant of aristocratic individuality-were drastically curtailed. For one, the inherited ideologies about ancestors as encoded in genealogical traditions became differentiated into mythological genealogies of heroes and historical genealogies of immediate ancestors .., with this differentiation of inherited ideologies came a parallel differentiation of inherited practices: the institutional worship of ancestors became differentiated into two separate but related practices, the worship of heroes and the cult of immediate ancestors" (153; cf. 144). This assumes a continuity and uniformity in

[AJA98

Age. In tomb cult, most often visits are made just once, perhaps twice.70 Such excursions are very different from hero cult, which once established continues to be practiced for long periods of time. The few more durable instances may mean that these particular tomb cults qualify for cult in the sense of an established practice, rather than an occasional ritual, or they may be an instance where hero cult is in fact located at a Bronze Age tomb. In view of the permanent anonymity and lack of monumentalization, it is more likely that duration is a regional characteristic in these cases, though exceptions occur. It is important to emphasize that tomb cult is not confined to the later eighth century; it occurs at least as early as the 10th, and continues beyond the emergence of the polis.71 Furthermore, tomb cult in the Iron Age is part of the regional complexes of contemporary funerary rituals. This can be illustrated with an example from Mycenae: a chamber tomb built at the edge of Grave Circle B, outside the citadel, contained a circular structure built of fieldstones and used for offerings (and possibly graveside meals). Very similar structures are also known from chamber tombs at Prosymna and Argos, but chiefly from Iron Age cemeteries in Naxos and Asine, in the Protogeometric and Late Geometric periods. Although the platforms known are from the Protogeometric and Late Geometric periods,

Greek society, while current work on the Iron Age has brought out the ruptures and instabilityof the period, see esp. Morris (supra n. 3) and Whitley (supra ns. 2 and 3). The development of ancestor cult and of hero cult are important precisely because the Greeks' traditions concerning ancestors are created and recreated throughout their history. 69 See Antonaccio 1994. 70 Only at the Menidhi tholos and tomb 1 at Thorikos in Attica do cults that begin in the late Iron Age continue over a long period of time. Menidhi: P Wolters,"Vasenaus Menidi II,"JdI 14 (1899) 103-35; and R. Hdigg,"Giftsto the Heroes in Geometric and Archaic Greece," in T. Linders et al. eds., Giftsto the Gods(Boreas15, Uppsala 1987) 93-99; Thorikos: M. Devillers,An Archaicand EarlyClassical Depositfrom a MycenaeanTombat Thorikos(Miscellanea Graeca8, Ghent 1988). There are other cases of long-term reuse (e.g., Medon in Phokis) but these are not certainly cult. See Antonaccio 1994, ch. 2 for details, as well as Antonaccio 1993. 71 Morris 750 acknowledges earlier cases of tomb cult, but emphasizes the eighth century.On the other hand, the earlier cults (shown on 759: table 1 and discussed on 753) are not the tomb cults discussed here, but very questionable cases and not at Mycenaean tombs.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1994]

CONTESTING THE PAST

burials can be found throughout the Iron Age, and even earlier.72 The veneration of ancestors in the Iron Age and Early Archaic period, with graveside meals at the circular platforms and offerings of pottery and small terracottas and bronzes, takes place in areas that also see the reuse of tombs, both Bronze Age and contemporary.7" The reuse of Bronze Age tombs for burials and tomb cult both, however, may not be the exclusive preserve of the elite. Placing of both elaborate and simple burials (including those of children) and the very modest types of offerings in most instances do not point to worship of noble ancestors by those of lower status.74 Mycenaean graves, resting places of the ancestors of the people later burying in the same area, may have been adopted by more than one group within a community in competing or complementary strategies in which they sought status by claiming priority and a connection with the past. The distribution of reused tombs and circular platforms is most extensive among the communities of the Argolid, but such features are also known in Iron Age Messenia, Euboia, and Naxos. By contrast, Attica provides relatively few instances of Iron Age reuse of Bronze Age tombs, rather more deposits and no circular platforms.75 The oscillation between reuse and visits from the end of the Bronze Age, the range of offerings and burials, and regional variations all point to

72 Mycenae: G. Mylonas, O rat4oLKgKKAdog B"' yov

(Athens 1972-1973) 18, pl. 5a-b; I. PapadeMvwrv0yv metriou, "AvaoxUa@lEvM vYLg," Prakt 1955, 218-23 •lKillustrated in Antonaccio for the circular structure (also 1993, fig. 4). The circular structure might have a Bronze Age precedent in the circular "altar"over Grave IV in Grave Circle A, and an apparently similar structure reported in a tumulus at Argos: Protonotariou-Deilaki(supra n. 65) 82, figs. 29-30. 73 On the Argolid, see, e.g., R. Higg, "FuneraryMeals in the Geometric Necropolis at Asine?"in R. Higg et al. 1983 (supra n. 3) 189-94. Instances of contemporary reuse are clear at Lefkandi and Argos, for example. Cf. R. Higg, Die Graberder Argolis 1: Lage und Formder Graber (Boreas 7.1, Uppsala 1974) esp. 157-59. At Lefkandi (Toumba necropolis) and Nichoria (settlement), circular platforms are found in different contexts. For Nichoria, see McDonald et al. (supra n. 3) and recently K. Fagerstr6m, "Finds, Function, and Plan: A Contribution to the Interpretation of Iron Age Nichoria in Messenia," OpAth 17 (1988) 33-70 with references; on the PG Lefkandi building, see now M. Popham et al., Lefkandi11.2:TheBurial Buildingat Toumba (London 1993) with references to the Toumba cemetery, and my article "Homer and Lefkandi,"

403

a range of users and practitioners, and contesting claims to the past. Moreover, the distribution of such practices among early Greek communities, like artifact styles and other cultural practices, indicates the use of these rituals in a widespread network of communication, though local meanings must not be ignored (see infra). Although a continuous concern with the earlier dead emerges, a continuous tradition, first of ancestors and then of heroes, or an unbroken veneration of heroic tombs does not, nor does tomb cult come to an end with the establishment of the polis, and the rise of hero cult at separate shrines.76 The modulation and small-scale contrast with the cults of epic heroes, from whom Iron Age and Archaic elites claimed descent as part of an ideology confirming social or political realities, are stable and long-term. Distinctly local, hero cult also defines community identity, while at the same time it participates in both ritual and poetry, which come to be Panhellenic.77 The lack of shrines constructed by later Greeks at Mycenaean tombs is consistent with a familiarity with these monumental relics of the past and a local meaning apart from hero cult. Such meanings were unstable and variable: Mycenaean tombs were reused as tombs, transformed into furnaces and mills, trash pits, and shelters, and tomb cult is not necessarily the rule for Bronze Age remains in later periods. Even as late as the Roman period, a tholos

in 0. Anderson and M. Dickie eds., Homer'sWorld:Fact, Fictionand Tradition(forthcoming). 74 E.g., a rich MG burial at Berbati: G. Sdiflund,Excavationsat Berbati1936-1937 (Uppsala 1965) 35-37, 81-90, figs. 17-18, 55-75; cf. Antonaccio 1993, fig. 2. A seventhcentury burial in T 533 at Mycenae: A. Wace, Chamber Tombs at Mycenae(Archaeologia 82, London 1932) 114, 117 figs. 47-49, pl. 56. 75 Naxos: V. Lambrinoudakis,"Venerationof Ancestors in Geometric Naxos," in R. Higg et al. 1988 (supra n. 3) 235-46; see Antonaccio 1993 and supra n. 73. 76 Cf. even Farnell 344: "the facts of modern anthropological study convince us that the question of priority in regard to these two motives of cult [hero and ancestor] is an ideal one; both are found operative simultaneously in early and late periods." On later tomb cult: Alcock (supra n. 35). 77 Explored by S. Sherratt, "'Reading the Texts': Archaeology and the Homeric Question,"Antiquity64 (1990) 807-24. If anything, local cults like that of Helen and Menelaos in Laconia led the way for hero cult at Panhellenic sanctuaries, whose origins would have been as regional centers of communication and competition.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

404

CARLAM. ANTONACCIO

Fig. 9. Tiryns. Roman oil press in Mycenaean tholos tomb. (Photo C. Antonaccio) tomb might serve as a mill or become a shrine with the addition of an imperial monument (figs. 9-10).78 This cannot be ascribed to the Greeks' ignorance of their ancestors' stories, which is the assumption behind the hypothesis of Homeric influence, or unfamiliarity with the past, implied by the concept of an eighth-century "Greek renaissance." Rather, the monuments were not always significant to later Greeks in ways that we would predict from written sources alone. We may now consider the relationship of tomb cult to written accounts in which the bones or other relics of heroes figure. The bones in particular confer a military or political advantage on the polis or sanctuary that possesses them, and on those who

78 See Antonaccio 1994 and supra n. 51, contra Morris 758. A striking instance of continuities with the present: AR 1991-1992, 16 reports that intact chamber tombs south of Mycenae have been recently used as dwellings by migrant fruit pickers, and collapsed tombs as trash pits. For Orchomenos, see A. Schachten, Cults of Boiotia 1: Acheloosto Hera (BICSSuppl. 381, London 1981) 208 with references. 79 On this see esp. Pfister (supra n. 11) 196-211, 42528, 507-508; and Antonaccio 1993 for references; and Vandiver (supra n. 11) 36 n. I with reference to G. Huxley, "Bones for Orestes,"GRBS20 (1979) 145-48 (and cf. H. Parke, "The Bones of Pelops and the Siege of Troy,"

[AJA98

recover them. The interest in relics does not seem to be the rationale behind tomb cult. I have suggested elsewhere that such bodily relics were often actually fossils. The findspots of relics, when recorded, do not include actual Bronze Age tombs.79 As seen above, it is difficult to find evidence for the early identification of Mycenaean graves specifically with any sort of hero, and hero cults are not founded at actual graves of any kind.80 The cult of heroes as depicted in written sources involves only purported graves, which do not seem to be the ones involved in tomb cult. The accounts of tombs from which relics were collected do not describe tholos and chamber tombs; furthermore, Mycenaean tombs visited in tomb cult had frequently collapsed, sealing off Bronze Age levels. Tomb cult, therefore, did not supply these relics, nor is tomb cult another name for relic cult, although the practice of tomb cult demonstrates that authentic Bronze Age remains clearly were known and available. Similarly, hero cult was not predicated on actual tombs, although the practice of tomb cult makes plain that Mycenaean tombs were known in this period. Yet hero cult, the cult of relics, and genealogies that traced members of the Greek elites to mythical forebears all fabricated links with the past. While tomb cult used remains of the Bronze Age, hero cult and the cult of relics compete with tomb cult, or complement it, providing another aspect to the uses of the past. The traffic in relics requires the interstate arena to provide a space in which to enact it.81 In addition to the competition for the past revealed in the archaeological record for ritual, epic constructions of time suggest additional tensions, among epic traditions and between ritual and epic. Susan Sherratt suggests that in the wake of the Bronze Age collapse, the early Iron Age saw the most active creation of epic, in which cremation, for example, became the heroic burial idiom (whereas inhumation had been standard Late Bronze Age

48 [1933] 153-62). Hermathena 80 As pointed out by A. Snodgrass, "Les origines du culte des h6ros dans la Grace antique,"in Gnoli and Vernant (supra n. 31) 107-19, 115-16. Snodgrass 1987 (supra n. 3) 160-61, however,argues that the institutionof "hero cult"at chamber tombs "positivelyexcludesfamiliaritywith Homer-or, at least, identificationof the object of the cult with a 'Homeric hero'." See also 164: "'Homeric' cults seem never to be located at genuine graves of the heroic age." See Sherratt (supra n. 77) on cremation in epic as opposed to inhumation in the Bronze Age. 81 See Antonaccio 1993.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

THE PAST CONTESTING

1994] :7

405

'IF'.... :

: ii2i :) •;

7

....

....

:~~

~

k':::? : :::

M77

1

Fig. 10. Orchomenos(Boiotia).Romanmonumentbasein Mycenaeantholostomb.(PhotoC. Antonaccio) The final phase in the eighth century saw practice).82 the integration of episodes and the end of composition, giving way to recitation. Sherratt only mentions Hesiod briefly, though his poetry uses epic language and form. While Hesiod would seem to belong in Sherratt's final phase of consolidation, with the Five Ages in Worksand Days(lines 106-201) he is still contributingto the constructionof the past, and providing a detailed frameworkfor it unique in the Greek epic tradition.83This passage has been the subject of numerous interpretations, but none has reallyaccounted for the lack of unity in Greek poetic traditions. Hesiod says that a Gold generation was the first created by the gods of mortal men (tpEp6oCWv They lived like the gods under Kronos, avOpCdCrw). and disappeared when the earth covered them after which they became baliK•EX1Uv•v),

(KaT& yakLI

82

But cf. Nagy (supra n. 41) 85-87. Vernantconstructsa systemof ages with complementary concepts of hubrisand dike:J.-P Vernant, Mythand Thoughtamong the Greeks(London 1983). Other recent considerations: C. Querbach, "Hesiod's Myth of the Four Races,"CJ 81 (1985) 1-12; J. Rudhardt,"Le mythe hesiodique des races et celui de Promethderecherche des struc83

[toveg;oOe0ol,

E'LXO6vLoL, by the will of Zeus. The

Silver generation that followed was much worse; they were covered by Zeus who became angry because they ignored the gods. Yet they, too, are <j3XEKEg

Jd'EKapeg, OVTIT(]vAvOpd•9nwV,OUOX6OVLOL

second after the Gold generation but still deserving of

ttl."84

Each of these generations appears to fit

into conceptions of divinity that, rather than Hesiod's invention, the Greeks may already have had in his time. Hesiod's violent and dark third generation was of Bronze. Its members went down to Hades anonymous, and unlike the other races, without a continuing existence and explicitly without TLtil. This is because they failed, according to Nagy, in their attempts to win KXEog,and so are ignored in epic. We know their names from myth, but they are not celebrated in epic because their deeds did not merit it.85

tures et des significations,"Revue europdenne des sciences sociales19 (1981) 245-81. M.-C. Leclerc, "Le mythe des races une fiction aux sentiers qui bifurquent,"Kernos6 (1993) 207-24 appeared too late for consideration here. 84 "Second"may refer to temporal place, not hierarchy. 85 Cf. Farnell 157 ?9.4.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

406

CARLAM. ANTONACCIO

Immediately after this generation, Zeus created the fourth, which is specified by no particular metal: it falls between the third of Bronze and the fifth of Iron, which Hesiod specifies as his own. This fourth generation is the OELovyEvog, the divine or godly generation of heroes, called i~lL0eot, the previous race on the earth, as Hesiod says. He states explicitly that this generation fought at Thebes and Troy, so

they are the figures of epic poetry as well. While they perished in war, the fate of at least some was to

live without cares in the Blessed Isles as fortunate, happy heroes (6XPLtot •poeg).86 As often observed, this fourth generation appears as an intrusion into the metallic sequence, which declines with each step except for the heroic interlude. Fontenrose suggested that the Bronze generation did not really end and the Iron generation is actually a part of it. Although the language is different for the introduction of the Iron generation, it will not, however, allow the two to be merged. Hesiod clearly states that war destroyed the heroes, some at Thebes, some at Troy (lines 161-63). The language used concerning the fate of those without an afterlife echoes that used of the other generations. The tattered lines between this and the intro-

duction of the final Iron generation make certainty about this section impossible, but Hesiod does call the Iron generation the fifth genos (lines 174-75).87 West suggests that "Greek traditions about men of

the past were almost wholly concerned with those who fought at Thebes and Troy and with people linked to them by a network of genealogies. They had to be accommodated in any survey of man's

86 There is no space here to thrash out all the possible dimensions of the word i]pwg; on the term "hero," see Farnell 15-16; cf. J. supra, and Rohde 615 s.v. pipwg; O'Brien, The Transformationof Hera (Lanham, Md. 1993). Nagy (supra n. 41) 15 on West's point that hemitheoirefers to parentage, rather than to divinity; M.L. West, Hesiod's Worksand Days (Oxford 1978) 191, on line 160, but see W

on Hesiod,Worksand Days,vv. 1Verdenius, A Commentary

382 (Mnemosyne Suppl. 86, Leiden 1985) 99, who points out that heroes in the epic tradition are often not of divine descent: "The original meaning of the word seems to be 'almost gods' (cf. LtOvL#g). . . apart from the immortality of the gods there was not a sharp dividing-line between gods and men, but only a gradual difference . . . Just as the gods in their actions and feelings may sink to the human level, prominent men may conversely rise to the level of the gods. . . . The anthropomorphism of the gods is complemented by the theomorphism of the heroes." eetov perhaps refers, in this case, to their being like the gods, in the way that the first race lived like gods. "Work, Justice, and Hesiod's Five 87 J. Fontenrose, Ages," CP 69 (1974) 1-16, esp. 8 against decline; 10 on the Heroic to Iron transition. See also J. Gwyn Griffiths, "Ar-

[AJA98

past. The position they occupy in Hesiod follows from the view that they were the people who preceded us ([line] 160), coupled with an unwillingness to identify them with the Bronze race-perhaps because the epics showed them as users of iron." 88He goes on to note that the Bronze, Heroic, and Iron races fairly accurately follow Greek concepts of history, whereas the Gold and Silver races seem alien to Greek tradition. Although the East provides parallels, West acknowledges that Hesiod is earlier than any of the Eastern texts he cites. West concludes that Hesiod's view stands apart from the usual one, which did not isolate the Age of Heroes from the rest of the past, or the present.89 Nagy remarks that generations one and four, the Golden and the Heroic, mirror each other in theme and diction, and points out that the cycle of ages comes full circle through the fifth, the present.90 Hesiod interrupted his narrative to wish that either he had died or not come into being until a later time, rather than live in the fifth age in which a total breakdown of social conventions takes place. This amounts to a desire for a Golden Age, for an end of cyclical (rather than linear) time.91 For Hesiod's own generation, the present "incorporates all the oppositions of the past and the hereafter." This is part of Nagy's larger analysis of the myth, in which he argues that Hesiod's diction provides evidence that the first and second generations depict the ritual, cultic aspect of heroes, while the third and fourth are the epic aspect.92 Archaeologists have tried to find a place for tomb cult in this structure. It is interesting to note that

chaeology and Hesiod's Five Ages,"Journal of the History of Ideas 17 (1956) 109-19, 113 with references. 88 West (supra n. 86) 174; to this he contrasts the Catalogue of Women, where the Heroic is like the Gold Age. See also Griffiths (supra n. 87) 116 summarizing the Eastern comparanda. He concludes that Hesiod must have had access to folklore traditions of human origins, probably dispersed from Egypt and Mesopotamia. 89 West (supra n. 86) 176. See also J. Gwyn Griffiths, "Did Hesiod Invent the 'Golden Age'?" Journal of the History of Ideas 19 (1958) 92-93. 90 Nagy 169 with note ?30 n. 1. 91 Nagy 169, emphasis his. Cf. Fontenrose (supra n. 87) 10. 92 Nagy 155. Nagy and Vernant (supra n. 83) both mention the cyclical character of the ages, but Vernant's emphasis is on the four metallic phases with the heroes intervening to complete the conceptualization of hubris and dike as constructed in pairs of metallic ages. Querbach (supra n. 83) pairs Gold and Silver, Heroic and Bronze, with regard to their creation and fate after death. He admits, however, that the Bronze race is nameless.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1994]

CONTESTING THE PAST

Hesiod's five ages received much attention in the earlier work of archaeologists. Griffiths, for example, discusses the archaeological basis of the Hesiodic scheme: "That the magical and religious associations of gold have played an important part in the idea of the Golden Age is quite clear; but it seems highly probable that the first meaning was a more literal one, and that Hesiod's scheme purports to show the ages of man in relation to the metals discovered by him. The order of his scheme is chronological. What is more, the sequence gold, silver, bronze and iron is one which the archaeological record attests in many areas."93He admits that of the four metals, Hesiod mentions the use of only bronze and iron (lines 150-51), but for Griffiths, if the choice of metals had only symbolized progressive decline, he would not have accurately mentioned the actual use of bronze and iron in the correct historical order. He then cites archaeological evidence for the use of gold and silver without knowledge of smelting and before bronze and iron. Griffiths includes copper, missing from Hesiod but important in the development of metallurgy, under the Greek word Xchic6g. He concludes that Hesiod's sequence blends myth and history, leaving the latter visible.94 Snodgrass has suggested that tomb cult was directed to the Silver generation, the 3no;X06vtoL, who are second after the Golden generation, but also deserving of Z1Lti. This suggestion seems due to their location "under the earth"; similarly, West thought Hesiod specified the Golden race as nJtLX06vtoL because they were concerned with mankind, and could expect veneration in return. He, too, suggested that the Silver race as a{(OX06vtoL was

3 Griffiths(supra n. 87) 112. Griffiths(supra n. 87) 119. Cf. Griffiths(supra n. 89)

94

for further discussion. 95 West (supra n. 86) 181; Snodgrass (supra n. 3) 165. See also Griffiths (supra n. 87) 111; burials of the men of the Silver Age who did not worship the gods were regarded as the blessed dead, "as though the term originated from the discovery of richly equipped interments not associated with recognizable symbols of religion, nor sites of temples in the Greek fashion, nor any abiding memories." 96 West (supra n. 86) 186; cf. Farnell 12-13; Rohde 6779.

97 West (supra n. 86) 182; see also Verdenius (supra

n. 86) 86, 92-93; he identifies JoTox66vtotwith "local heroes" who do not go to Hades like the Bronze men but "stay just under the surface of the earth. They are one stage further away from the gods than the golden race, which stays on the surface of the earth ([line] 123) Hesiod does not call the silver race ipmwegbecause he....reserves this term for the fourth race ([line] 159). The latter could not be identified with the silver race because of their

407

identified with specific graves. The failure of Hesiod's system to achieve the same status as Homer's would apparently explain why this view was shortlived and limited.95 Although both Snodgrass and West explain in;TLXOOVLO and ToX6(0vtoL as, respecwho range the earth and generations tively, past those who are buried and fixed in space,96elsewhere West cites Rohde's remark that the term jTyLXOoVLot only serves to distinguish the Golden generation or gods themselves; in Homer from the njTovpavtoL, it designates the generation of men, not some original race. 'YjToXOovtotare then a further refinement in this scale of distinctions.97 The nonmetallic, fourth generation of epic heroes in Hesiod's system is situated between the anonymous and unreachable Gold, Silver, and Bronze generations of the cycle and the fifth, the present Iron. Each of these had been covered up, and succeeding generations were not descended from them. If located in terms of absolute chronology, the heroes lived at the cusp of the historical Iron Age: the Bronze race, Hesiod's anonymous, dark, and violent generation, did not know iron, nor did they eat bread, or have an afterlife, but were destined for Hades without a transformed existence. The absence of ironworking places these figures in the Late Helladic period, the historical Mycenaeans. In a diachronic reading of Hesiod's system, the heroes of epic were separated from this Bronze Age. If so, the Late Helladic period is anonymous, though it leads into the final phase of war and imminent catastrophe, that which is celebrated in song.98

warlike spirit, but the local heroes could, because many of them were anonymous. . . . The fact that Hesiod leaves their function undefined shows that he simply included the local heroes in his scheme [line 141]." As Querbach states (supra n. 83) 3, the Silver race is honored but not divine, but he also says they do not have any continuing existence, which seems contradicted by their function as OVlTOiV ~iXaCKECg JTXovUTo 6OTat (Hes. Op. 121). 98 See Farnell 13, and Rohde 75. The old notion of H. Chadwick, The Heroic Age (Cambridge 1912), that construction and celebration of a heroic age springs up during an ensuing dark period, may thus be of use here; see Antonaccio 1994 ch. 5 and Sherratt (supra n. 77) 814. Fontenrose (supra n. 87) 9 says that the heroic age is the Bronze Age "from another point of view, romanticized and glorified. As Heriod's bronze age preserves a genuine, if not entirely accurate, memory of the later historical Bronze Age, roughly 1400-1000 B.C., so the age of heroes carries the legendary and epic tradition of that time. Hesiod did not realize that the two genea were really two representations of a single period."

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

408

CARLA M. ANTONACCIO

[AJA 98

Sherratt characterizes this latter phase as one of "social and political fluidity and change when new family or social groups emerge jostling for power and eager to establish their credentials, and when legitimation and self-propaganda of individuals or small groups become particularly crucial issues."99 As for the offerings at chamber tombs, she thinks that "the growth of local 'hero' (or ancestor) cult, focussed on tombs of the evidently distant past, marks a transition away from legitimation grounded in the present. Heroes in general were settling into the past. From now on their value to the interests of family, community and wider groupings lay in possession and conservation of the heritage of tradition they already provided. The scene was set for the spontaneous transubstantiation of kleos (epic glory) into a ready-formed body of polydynamic myth."'"00Aswe have seen, the identification of tomb cult with hero cult is not secure, and if anything the eighth century represents an intensification of the competition over the past, though its significance as contested ground may have been agreed (infra). While it is not legitimate to seek a completely consistent historical consciousness within this myth, there is a faded memory behind the description of the Heroic Age that is preserved in epic. The tombs of the Mycenaeans that are the object of tomb cult beginning in the early Iron Age lie just beyond epic's reach, and seem to belong to Hesiod's Bronze hidden by Zeus. As Rohde noted, "immeVWVUVLot, diately after the disappearance of the Heroes the poet begins the age in which he himself must live. Where the reach of poetry ends, there is an end of all further tradition; there follows a blank, and to all appearances the present age immediately begins. That explains why the Heroic Age is the last before the fifth, to which the poet himself belongs, and why it does not, for example, precede the (undated) Bronze Age."'o' By contrast, the epic heroes, for all their identification with Late Helladic places and features of material culture, function in a Dark Age matrix: the ideology of epic poetry is current although the setting of the Heroic Age is in the past. The poems do not present a fossilized society, but they embody one that evolved gradually and did not

arise without undergoing long gestation. Hesiod's wish to belong to another time expresses the tensions inherent in the process; the heroes immediately prior are the pivot, and the early historical Iron Age now appears to fit this very well, as recent finds at Lefkandi and elsewhere suggest.102 The stresses in these competing versions of the past, mythic as well as ritual, are illuminated by the work of Arjun Appadurai. The past is a "scarce resource," not mere grist for contemporary ideological mills (whether a genos's heroic genealogy or polis cult). Instead, cultures have rules that govern the past's debatability.'03 There are four dimensions to these norms: first, authority, or communal agreement on credibility; second, continuity, or agreement about the connection with the source of authority; third, depth, the differing valuation given to depth of time in the past; and finally, interdependence, the degree to which competing versions of the past are minimally credible.104 A shared past may exist (Panhellenic, say), but different groups in a community may each hold a separate past, or place a different emphasis upon some aspect of the shared past. The rules within which the past is debated also mediate change, as occurs at the beginning and end of the Iron Age in particular (cf. civic kinship and ideologies of descent in the polis).'05Whereas Ernest Bloch identified a pragmatic past, within which originate challenges to the unchanging ritual past dominating the present, Appadurai suggests a past that negotiates between ritual eternity and the present (cf. KXEog and uLtL'): "the past is an intrinsically alternative mode of discourse to those other cultural modes of communication which can, and often do, assume an eternal present. Such norms, therefore, constitute an aspect of culture in which concessions to change are built in, and division and debate are recognised. As a result, such norms permit new forms of action, at the same time as they allow cultures to regulate social change."'06 The norms that seem to be reflected in our written sources especially do not necessarily determine action; ritual behavior actually generates symbolism. Morris makes a similar observation: "Social structure, as a set of internalised but constantly

99 Sherratt (supra n. 77) 815. On the contested topic of the end of composition, see G. Nagy, "Homeric Questions," TAPA 122 (1992) 17-60 with references. 100Sherratt (supra n. 77) 816 with references. 101Rohde 75. 102 Cf. Querbach (supra n. 83); on Lefkandi, supra n. 73. 103 A. Appadurai, "The Past as a Scarce Resource," Man 16 (1981) 201-19, esp. 201, contra M. Bloch, "The Past

and the Present in the Present," Man 12 (1977) 278-92 (and see also Morris). 104 Appadurai (supra n. 103) 203. 105Appadurai (supra n. 103) 216-17. 106 Appadurai (supra n. 103) 218. Note, however, that Bloch's main point is to link the appearance of the past in discourse about the present as a correlate of hierarchy, not problematic for Indian society but of interest for the Greek Iron Age as it advances and develops.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

1994]

CONTESTING THE PAST

renegotiated roles and rules, is an artefact of this knowledge."'07 It is Appadurai's emphasis on cultural norms or rules that is useful for the Greek case. Morris has argued that tomb cult had several meanings to different actors in different regions, as well as within a single community or region: "Each cult probably had its own associations and special significance"in which he sees the "conflict of ideologies in the 8th c."'18In the end, though, he is unable to account for both the widespread distribution of tomb cult and its local variations, or to integrate it with literary constructions of the past: Hesiod's, which he explicitly cites. Instead, Morris glosses apparent contradictions in the preserved epic system(s) of the past, which he sees as contemporary with the late eighthcentury developments attending the emergence of the polis, by claiming that our own understanding is at fault: we underestimate "the subtlety of early Greek poetic thought." This solution consists in simply denying the contradictions generated by different versions of the past. Yet it is true that we cannot "arbitrarilyexclude the tomb cults as being 'archaeological' rather than 'literary"'and Morris is right to point to the eighth century as a time of crisis and change, when tomb cults became particularly important and their frequency peaked.'10The focus on the Mycenaeans fulfills Appadurai's criterion of authority, but the present is the key; epic and cult therefore function for contemporary purposes. Tombs, rather than avenues to supernatural relics, allowed direct access to this authority,but by way of an active creation of ancestors."10 The focus on Bronze Age tombs begins with the very close of the Bronze Age, and intensifies in the eighth century, when hero cult is also instituted, connected to habitation sites as well: it should be noted that some Dark Age cemeteries were also located in former habitation areas."' Appadurai's

107Morris(supran. 46) 7-9 withreferences.

10sMorris 758.

109Morris754-55, referringto Vernant(supran. 83) 64;he alsoseemsto havein mindRudhardt's introductory remarks(supran. 83). Cf.alsoQuerbach(supran. 83) 6, on Hesiod'spoetictradition:"Acertainamountof anomaly,bothin structureand content,canbe attributedto the immediateneedsof a particularsituation." 110Morris 750-51; cf. 758: "WhetherHeroes or a Silver

race,theirbonesbecamea sourceof powerand an arena

for conflict." "' See my forthcoming article (supra n. 54). 112B. Qviller, "The Dynamics of the Homeric Society," SymOslo56 (1981) 109-55; R. Drews, Basileus:TheEvidence for Kingshipin GeometricGreece(New Haven 1983), and esp. E. van der Vliet, "'Big-Man', Tyrant, Chief: The

409

model applied to the Iron Age choice of Late Helladic tombs and other remains acknowledges their authority, their connection with actors in the present, values their place in time, and accepts limits for the range of variabilityin accounts of the past. We can acknowledge tensions within a community and begin to deal with what we can see in the archaeological record: not a uniform or continuous evolution but a dynamic and variable process of rupture, appropriation, and conflict both synchronicallyand diachronically. The cusp of pre- and protohistory provides a communally acknowledged source of authority for emergent Greek communities. If Hesiod's adaptation of a metallic sequence is from an Eastern source, the status of Eastern cultures in the Bronze and Iron Ages as sources of prestige with their goods, motifs, and forms may help to account for the choice of this framework for the past in Works and Days.

By providing us with authority,continuity,depth, and interdependence as a cross-culturalframework for using the past, Appadurai's work also helps to explain an apparent anomaly in Greek systems of thought and practice. Genealogy dominated Greek thought in the Archaic period, but this is contradicted by the archaeological record for sporadic tendance of tombs, which as already discussed shows a limited concern in practice. In Greece, changing structures of authority, from basileia("kingship")to tyranny and aristocracy, are for the most part achieved statuses, rather than inherited. These figures of authority are all "big men," persons whose position depends on the abilityto attractand keep followers through personal talent, feasting, and gift-giving, right through the historical period."2 The status of "big man," which rather than "king"or even "chief" best fits the basileusof early Greece, is what emerged after the end of Bronze Age power structures based on the palaces."3Since

Anomalous StartingPoint of the State in ClassicalGreece," in M.A. van Bakel et al. eds., PrivatePolitics:A Multi-DisciplinaryApproachto 'Big-Man'Systems(Leiden 1986) 117-26; cf. Y. Ferguson, "Chiefdomsto City-States:The Greek Experience," in T. Earle ed., Chiefdoms: PowerEconomy,and Ideology(Cambridge 1991) 169-92. 113 See Antonaccio 1993 contra J.-P Vernant, The Origins of GreekThought(London 1982) 38-48. Whitley (supra n. 3) 184-86, and (supra n. 2) 352 has contested the appropriateness of the "big man" model for all of Greece in the Iron Age, and sees it as a phenomenon of the early Iron Age only. His argument seems to me too determined by the specific case of Melanesia and its settlement patterns, and he is not aware of van der Vliet's work (supra n. 112), which points out the commonalities of Greek authority through time.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions

410

CONTESTINGTHE PAST C.M.ANTONACCIO,

birth alone does not secure position, and competition for authority is the rule, tomb cult, the cult of ancestors, may constitute a legitimating device for the elite that allowed leaders to claim links with the past that did not exist in real kinship, descent, or other continuity. In fact, Appadurai's allowance for a mediating past makes it possible for various groups among the Mycenaeans' successors, on the cusp of new social realities at the end of both the Mycenaean period and the Iron Age, to appropriate them as their ancestors in a period of change. In the polis, hero cult creates a civic kinship that may serve

individuals or the needs of the state to foster a new group identity.114But tomb cult, distinct from hero cult, was practiced at different locations, using different forms, and established other links with the past. When hero cult emerges, perhaps first at Sparta in the late eighth century, heroes indeed

114 In Athens, e.g.: E. Kearns, "Change and Continuity in Religious Structures after Cleisthenes," in P Cartledge

and ED. Harvey eds., Crux: Essays Presented to G.E.M. de

Ste. Croix(Exeter 1985) 188-207 and supra n. 61. 15 There is no room here to pursue several other issues: hero and tomb cult in ethnjor the Classicaland Hellenistic periods, the connection of tombs with possession

were settling into the past, as Morris and Sherratt suggest.

The richness and variety of the evidence for the Greeks' concern with their own past make many readings possible. No single version should be sought; ritual and text, archaeology and philology, reveal differences that should be acknowledged, not reconciled. The ambiguities and multiple stories of the Greeks are keys to understanding how the past functioned for them: a source of authority,a fertile field for the ever-shifting definitions of power, identity, and authenticity.115

DEPARTMENT WESLEYAN

OF CLASSICAL

STUDIES

UNIVERSITY

MIDDLETOWN,

CONNECTICUT

06459

INTERNET [email protected]

of land, or the cults of heroic city-founders, especially in the West. See Whitley (supra n. 36) and de Polignac(supra n. 51) on activities at Mycenaean tombs in Attica and the Argolid and their possible relation to the possession of territory, as well as Morris (supra n. 57); Alcock (supra n. 35) on post-Classical tomb cult; and Malkin (supra n. 41) on founders.

This content downloaded from 152.3.102.242 on Wed, 8 May 2013 11:32:05 AM All use subject to JSTOR Terms and Conditions