Studies in Athenian Architecture, Sculpture, and Topography presented to Homer A Thompson (Hesperia Supplement: 20)

HESPERIA: SUPPLEMENT XX STUDIES IN ARCHITECTURE ATHENIAN SCULPTURE TOPOGRAPHY AND *~ p IAl PRESENTED TO HOMER...

Author: various authors

48 downloads 653 Views 34MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

HESPERIA: SUPPLEMENT XX

STUDIES

IN

ARCHITECTURE

ATHENIAN

SCULPTURE

TOPOGRAPHY

AND

*~ p

IAl

PRESENTED TO

HOMER A. THOMPSON

AMERICAN

SCHOOL OF CLASSICAL STUDIES AT ATHENS PRINCETON,

NEW JERSEY

1982

HESPERIA: SUPPLEMENT XX

STUDIES

IN

ATHENIAN

SCULPTURE AND

ARCHITECTUR

TOPOGRAPHY

PRESENTED TO

HOMER A. THOMPSON

AMERICAN

SCHOOL OF CLASSICAL STUDIES AT ATHENS PRINCETON, NEW JERSEY 1982

Libraryof CongressCatalogingin PublicationData Mainentry undertitle: Studies in Athenianarchitecture,sculpture,and topography (Hesperia.Supplement; 20) "Bibliography of HomerA. Thompson":p. vii-xii 1. Art, Greek-Greece-Athens. I. AmericanSchool of Classical Studies at Athens. II. Series: Hesperia (Princeton,N.J.). Supplement; 20. N5650.S8 709'.38'5 81-14994 ISBN0-87661-520-3 AACR2

ended

rv

HOMERA. THOMPSON ETH FEFONOTI iTENTEKAI EBAOMHKONTA XAPIXTHPIONY1TOTHE EN AGHNAIXAMEPIKANIKHE IXOAH1 KAA7XIKQN IXTOYAQN APETHX ENEKEN SEPTEMBER 7, 1981

These papers are offered by members of the American School of Classical Studies at Athens who have been students and colleagues of Homer A. Thompson. Contributions towardthe cost of publicationhave been generously providedby the following:

Anna S. Benjamin

JamesH. Oliver

WilliamR. and JaneC. Biers

JeromeJ. Pollitt

JudithP. Binder

HenryS. and RebeccaW. Robinson

Alan L. Boegehold

CarlA. and MaryC. Roebuck

OscarBroneer

RobertL. and LouiseC. Scranton

AlisonFrantz

T. Leslieand lone M. Shear

VirginiaR. Grace

EvelynLordSmithson

DorothyKent Hill

BrianA. Sparkes

EvelynB. Harrison

DorothyB. Thompson

RichardH. Howland

MargaretThompson

SaraA. Immerwahr

StephenV. Tracy

MabelL. Lang

EugeneVanderpool

MerleK. Langdon

EmilyT. Vermeule

JamesR. and MarianM. McCredie

MichaelB. Walbank

Malcolmand MargueriteMcGregor

PaulW. Wallace

The MeridenGravureCompany

SaulS. and GladysD. Weinberg

BenjaminD. and LucyS. Meritt

FrederickE. Winter

StephenG. and StellaG. Miller

NancyA. Winter

FordyceW. Mitchel

WilliamF. Wyatt

CharlesH. Morgan

R. E. Wycherley

TABLEOF CONTENTS A. THOMPSON OFHOMER BIBLIOGRAPHY ..................................... NANCY BooKIDIS: Attic TerracottaSculpture,Acropolis 30...................... JoHNMcK. CAMPII: Drought and Famine in the 4th CenturyB.C. ..... ......... B. DINSMOOR, JR.: The Asymmetry of the Pinakotheke-For the Last WILLIAM Time . ...................................................... FRANTZ:The Date of the PhaidrosBema in the Theater of Dionysos ...... ALISON A ClassicalMaiden from the Athenian Agora ..... EVELYN B. HARRISON: ........ The EarliestAthenian Grave ........... SARA A. IMMERWAHR: ................ An Athenian Stele in Princeton ........... FRANCESF. JONES: ................ The Ancient Image of Athena Polias .......... JoHNH. KROLL: ............... GERALDV. LALONDE:TopographicalNotes on Aristophanes........ ............ Lucy S. MERITT: Some Ionic ArchitecturalFragmentsfrom the Athenian Agora ... STELLAG. MILLER:A MiniatureAthena Promachos .......... ................ STEPHENG. MILLER:Kleonai, the Nemean Games, and the LamianWar .......... ISABELLE K. AND ANTONYE. RAUBITSCHEK: The Mission of Triptolemos ..... ...... BRUNILDES. RIDGWAY:Of Kouroi and Korai, Attic Variety ..................... T. LESLIESHEAR,JR.: The Demolished Temple at Eleusis...................... EVELYNL. SMITHSON: The PrehistoricKlepsydra.Some Notes ................... B. THOMPSON: A Dove for Dione .................................. DOROTHY MARGARETTHOMPSON: Reflections on the Athenian ImperialCoinage ..... ....... EUGENEVANDERPOOL: EHI UPOYXONTIKOACNCI, The SacredThreshingFloor at Eleusis................................................... CHARLESK. WILLIAMS,II: Zeus and Other Deities. Notes on Two ArchaisticPiers .. R. E. WYCHERLEY: Pausaniasand Praxiteles.................................

vii

1

9 18 34 40 54 63

65 77

82 93 100

109 118 128 141

155 163 172 175 182

BIBLIOGRAPHY OF HOMERA. THOMPSON 1932 "SyrianWheat in Hellenistic Egypt," ArchivfurPapyrusforschung 9, pp. 207-213 (with K. Kourouniotis) "The Pnyx in Athens," Hesperia1, pp. 90-217 1933 "Activities in the American Zone of the Athenian Agora, Summer of 1932," AJA 37, pp. 289-296 "TerracottaLamps," Hesperia2, pp. 195-215 (with K. Kourouniotis) "The Athenian Pnyx," AJA 37, pp. 652-656 1934 "Two Centuries of Hellenistic Pottery," Hesperia3, pp. 311-480 1935 "The Topographyof the West Side of the Agora," AJA 39, p. 114 1936 "Pnyx and Thesmophorion,"Hesperia5, pp. 151-200 1937 "Buildingson the West Side of the Agora," Hesperia6, pp. 1-226 1938 "AdditionalNote on the Identificationof the Pottery of the Salaminiansat Sounion," Hesperia7, pp. 75-76 "The Metal Works of Athens and the Hephaisteion," AJA 42, p. 123 (with N. Kyparisses) "A Sanctuary of Zeus and Athena Phratrios Newly Found in Athens," Hesperia7, pp. 612-625 1940 "A Golden Nike," HSCP, Suppl. I, AthenianStudiesPresentedto W. S. Ferguson,Cambridge, Mass., pp. 183-210 The Tholosof Athensand its Predecessors,Hesperia,Suppl.IV, Princeton (with Dorothy B. Thompson) "The Golden Nikai of Athens," AJA 44, pp. 109-110 1942 "The Pnyx in the Fourth Century," AJA 46, p. 123

viii

HOMERA. THOMPSON

1943 (with R. L. Scranton) "Stoas and City Walls on the Pynx," Hesperia12, pp. 269-283 1946 "The Influence of Basketryon Attic Geometric Pottery," AJA 50, p. 26 1947 "The Excavationof the Athenian Agora, 1940-46," Hesperia16, pp. 193-213 1948 "Excavationof the Athenian Agora, 1947," Hesperia17, pp. 149-196 "Greek and Roman Societies' Joint Meeting, Oxford, 3-10 August, 1948," Phoenix2, pp. 88-89 1949 "An Archaic Gravestone from the Athenian Agora," Hesperia,Suppl. VIII, CommemorativeStudiesin Honorof TheodoreLeslieShear, Baltimore,pp. 373-377 "The Athenian Agora: 1949 Season," Archaeology2, pp. 184-188 "Decouverte sur l'agorad'Athenes d'un edifice du Ve siecle av. J.-C. qui pourraitavoir appartenuau Poecile," resume of a letter in ComptesRendusde l'Academie et Belles-Lettres,p. 182 des Inscriptions "Excavationsin the Athenian Agora: 1948," Hesperia18, pp. 211-229 "Head of Nike from the Athenian Agora," Archaeology2, pp. 17-19 "The PedimentalSculptureof the Hephaisteion," Hesperia18, pp. 230-268 "Stoa of Attalos," Archaeology2, pp. 124-130 1950 "Agrippa'sConcert-Hallin the Athenian Agora," Archaeology3, pp. 155-157 "Excavationsin the Athenian Agora: 1949," Hesperia19, pp. 313-337 "The Odeion in the Athenian Agora," Hesperia19, pp. 31-141 1951 "Excavationsin the Athenian Agora: 1950," Hesperia20, pp. 45-60 "The 15th Campaignin the Athenian Agora: 1950," AA [JdI 66], cols. 141-151 1952 "The Altar of Pity in the Athenian Agora," Hesperia21, pp. 47-82 "Excavationsin the Athenian Agora: 1951," Hesperia21, pp. 83-113 1953 "The Athenian Agora, Excavationand Reconstruction,"Archaeology6, pp. 142-146 "Athens and the HellenisticPrinces," ProcPhilSoc97, pp. 254-261 "Excavationsin the Athenian Agora, 1952," AJA 57, pp. 21-25 "Excavationsin the Athenian Agora: 1952," Hesperia22, pp. 25-56

BIBLIOGRAPHY

ix

1954 "ClassicalCongress in Copenhagen," Archaeology7, pp. 249-250 "Excavationsin the Athenian Agora: 1953," Hesperia23, pp. 31-67 "Rebuilding the Stoa of Attalos, Progress Report, Spring 1954", Archaeology7, pp. 180-182 1955 "Activities in the Athenian Agora: 1954," Hesperia24, pp. 50-71 1956 "Activities in the Athenian Agora: 1955," Hesperia25, pp. 46-68 1957 "Activities in the Athenian Agora: 1956," Hesperia26, pp. 99-107 "The Athenian Agora. A Sketch of the Evolution of its Plan," Acta CongressusMadvigiani.Proceedingsof the SecondInternationalCongressof ClassicalStudiesI, Copenhagen, pp. 341-352 1958 "Activities in the Athenian Agora: 1957," Hesperia27, pp. 145-160 1959 "Activities in the Athenian Agora: 1958," Hesperia28, pp. 91-108 "The Athenian Agora, 1959," Archaeology12, pp. 284-285 "AthenianTwilight, A. D. 267-600," JRS 49, pp. 61-72 TheStoa of AttalosII in Athens, AthenianAgoraPictureBook, No. 2, Connecticut 1960 "Activities in the Athenian Agora: 1959," Hesperia29, pp. 327-368 "Odeion of Agrippaor Sanctuaryof Theseus?" RevueArcheologique,pp. 1-3 "The PanathenaicFestival," resume in Proceedingsof the ClassicalAssociation17, p. 26 1962 TheAthenianAgora:A Guideto the Excavationsand Museum,Athens "ItinerantTemples of Attica," Abstract of Paper read at General Meeting, 1962, AJA 66, p. 200 "The SculpturalAdornment of the Hephaisteion," AJA 66, p. 339-347 1963 Review of V. Scully, TheEarth,the Templeand the Gods, New Haven and London 1962, Art Bulletin45, pp. 277-280

x

HOMERA. THOMPSON

1964 "A ColossalMouldingin Athens," Xaptror-pIo Eus 'AaoraTa'o-ovK. 'OpX&v8o)'I, Athens,pp.314-323 "Some Consequencesof the Worshipof Heroes in Ancient Athens" (in modern Greek), Scientific Yearbookof the PhilosophicalSociety of the Universityof Athens,pp. 275-284

1965 "A Note on the Berlin Foundry Cup," Marsyas, Suppl. I, Essays in Honor of Karl Leh-

mann,New York,pp. 323-328 Review of A. N. Oikonomides, The TwoAgorasin AncientAthens, Chicago 1964, Archaeology 18, pp. 305-306

1966 "Activityin the AthenianAgora1960-1965,"Hesperia35, pp. 37-54 "TheAnnexto the Stoaof Zeus in the AthenianAgora,"Hesperia35, pp. 171-187 "ClassicalLands,"ProcPhilSoc 110, pp. 100-104 "SomeLibrariesin AncientAthens,"BrynMawrAlumnaeBulletin,pp.2-9 Review of A. N. Oikonomides, The Two Agoras in AncientAthens, Chicago 1964, JHS

86, p. 273 1967 Review of J. M. Cook and W. H. Plommer, The Sanctuaryof Hemitheaat Kastabos, Historians26, pp. 217London 1966, Journalof the Societyof Architectural

220 1968 "Activityin the AthenianAgora:1966-1967,"Hesperia38, pp. 36-72 1969 (withAlison Frantzand John Travlos)"The 'Templeof ApolloPythios'on Sikinos," AJA73, pp. 397-422 1971 "LucyTalcott(1889-1970),"Gnomon43, pp. 104-105 1972 Review of H. H. Busing, Die griechischeHalbsdule,Wiesbaden 1970, Art Bulletin54, pp.

537-539 "Sir John Beazley (1885-1970)," YearBook of the AmericanPhilosophicalSociety, pp.

115-121 (with R. E. Wycherley) The Agora of Athens:the History,Shape and Uses of an Ancient CityCenter,Princeton

BIBLIOGRAPHY

xi

1973 "Gisela M. A. Richter (1882-1972)," YearBook of the AmericanPhilosophicalSociety, pp. 144-150 Review of P. W. Lehmann, Samothrace,Volume3: The Hieron, Princeton 1969, Archaeology26, pp. 228-229 1974 "Some Recent Developments in the Excavation of the Athenian Agora," Transactions of the Royal Societyof Canada, ser. 4, 12, pp. 249-251 "WilliamBell Dinsmoor (1886-1973)," YearBook of the AmericanPhilosophicalSociety, pp. 156-163 1975 Review of J. A. Bungaard, The Excavationof the AthenianAcropolis,Copenhagen 1974, AJA 79, pp. 378-379 Review of P. Bernardet al., Fouillesd'AiKhanoumI, Memoiresde la DelegationArcheologiqueFrancaiseen AfghanistanXII, ArtibusAsiae 37, pp. 249-254 1977 "Dionysos among the Nymphs in Athens and in Rome," Journalof the WaltersArt Gallery36, pp. 73-84 1978 "A Golden Victory," A PortfoliohonoringHaroldHugo for his Contributionto Scholarly Printing,Essay No. 3, Meriden, Connecticut "Some Hero Shrines in Ancient Athens," AthensComes of Age:from Solon to Salamis, Papers of a Symposiumsponsored by the ArchaeologicalInstitute of America, Princetonchapter,and the Departmentof Art and Archaeology, Princeton University, Princeton, pp. 96-106 Review of Georges Vallet, Francoise Villard, and Paul Auberson, Megara Hyblaea, I, Le quartierde l'agoraarchadque,Paris 1976, ClassicalWorld72, pp. 122-123 Review of J. A. Bungaard,Parthenonand the MycenaeanCityon the Height, Copenhagen 1976, AJA 82, pp. 256-258 Review of R. E. Wycherley, The Stones of Athens, Princeton 1978, Archaeology31, pp. 63-65 1980 "In Pursuitof the Past: The AmericanRole 1879-1979," AJA 84, pp. 263-270 "John J. McCloy," ASCS Newsletter,Fall, p. 10 "Stone, Tile and Timber: Commerce in Building Materials in Classical Athens," Excpedition22, pp. 12-26 "The Tomb of ClytemnestraRevisited," FromAthensto Gordion:ThePapersof a Memorial Symposiumfor Rodney S. Young (UniversityMuseum Papers I), Philadelphia, pp. 3-15

xii

HOMERA. THOMPSON

1981

"AthensFacesAdversity,"Hesperia50, pp. 343-355 "TheLibrariesof AncientAthens.,"TheSt. John'sReview32, pp. 1-16 "The Pnyx in Models," Hesperia, Suppl. XIX, Studiesin Attic Epigraphy,History,and

Princeton,pp. 133-147 Topography, 1982

"Architecture as a Mediumof PublicRelationsamongthe Successorsof Alexander," Studiesin theHistoryof ArtX, Washington,pp. 173-189

ATTICTERRACOTTASCULPTURE:ACROPOLIS30 (PLATE1) T

HE SUBJECTfor this articlegrewout of a recentconversationwith Mr. Thompson which centeredon the functionof terracottasculpture.'The subjectsof the conversationwere the terracottastatuetterecentlydiscoveredin the AthenianAgora and the freestandingdedicationsfrom the Sanctuaryof Demeterand Kore at Corinth.2 The questionraisedwaswhethera city like Athenswitha readysourceof marblewould decoration.The have modeledstatuesin clay for any purposebut that of architectural questionimplieda contrastwith Corinthwheremarblewas lackingand whereclaywas and freestandingsculpture.Given this hypothesis,the folused for both architectural lowingAtticpieceraisesinterestingproblems. A fragmentary statuemodeledin Attic clay was found in the earlyexcavationsof the AthenianAcropolis,was publishedin 1906 by W. Deonna,3and subsequentlywas neglected.The statue preservesthe lower part of a drapedfigure roughlyfrom the one-half life-size or slightlylarger.Deonna thighs to the plinth.It is approximately identifiedit as a standingfigureand, undoubtedlyhavingthe koraiin mind, considered it to be freestanding. The statueis mendedfrommanyjoiningfragmentsand partiallyrestoredin plaster. Left side and backare most complete;the rightside breaksoff shortlyabove the hem. The left foot is intactexcept for the tips of the toes. The right is missingbut can be restored.Now 0.325 m. in height as preserved,the statue is estimatedto have been 0.70 to 0.75 m. in height.4 Withmore parallelsnow fromwhichto draw,we can see that Deonna'sidentification was incorrect.The figureis not standingbut seated.It is seated with feet drawn II wish to express my warm thanks to Mr. G. Dontas, Ephor of the Acropolis, for giving me permission to publish this piece, and my special gratitudeto Mrs. M. Brouskari,Director of the Kanellopoulos Museum, for greatlyfacilitatingmy study of it. 2T. L. Shear, Jr., "The Athenian Agora:Excavationsof 1972," Hesperia42, 1973, pp. 401-402, pl. 75: a, b. For previous finds, R. Nicholls, "ArchitecturalTerracottaSculpturefrom the Athenian Agora," Hesperia 39, 1970, pp. 115-138. For the Corinthianmaterial, R. S. Stroud, "The Sanctuaryof Demeter and Kore on Acrocorinth,PreliminaryReport II: 1964-1965," Hesperia37, 1968, p. 325, pl. 95:c, e; N. Bookidis, J. E. Fisher, "The Sanctuaryof Demeter and Kore on Acrocorinth, PreliminaryReport IV: 19691970," Hesperia41, 1972, p. 317, pl. 63:a-d. The Corinthianmaterialwill be treated more fully in a forthcoming study on technique. 3Acropolis30. W. Deonna, Les statuesde terre-cuite en Grdce,Paris 1906, pp. 48-51, with referencesto previouspublication.Deonna cites a small draperyfragment,0.18 m. by 0.13 m., which I did not see. 4In greaterdetail, dimensions and preservationare as follows: P. H., left front, 0.305-0.31; back, 0.24; right side, 0.06 m. The left leg breaks off for 0.14 m. above the knee, while the preserved edge of the upper torso begins 0.21 m. from the knee, breaking off shortly thereafter. The plaster restorationsare especiallyextensive on the left side near the back where the uppertorso breaksoff. The restoredcurvature visible in side and back views is probablyexcessive.

2

NANCY BOOKIDIS

backso that kneesand toe tips are on the same frontalplane.The feet are spreadapart andturnedslightlyoutwards.The left thigh,the outersurfaceof whichhas splitawayto exposethe underlyingcore, is not quitehorizontalbut risesvery slightlyfromthe knee towardthe hips.The beginningof the erect uppertorso is preservedat the backupper break.Alongthe outsideof the left thighthe bodywallappearsto thickenconsiderably at the pointwherethe outersurfaceof claybreaksoff.5Whilethe coarseclaycurvesup and in to block out the nearlyhorizontalthigh, the outer layer appearsto continue vertically.Althoughit is possiblethat this divergenceis due to somethingwhichonce restedon the lap, it is morelikelythatit relatesto the positionof the left arm.The arm may have been poisedabove the leg and slightlyto one side but connectedto it by a fall of drapery.Exceptingthe placementof the feet, the positioncan be paralleledin the LateArchaicmarbleseatedstatuefromPlateiaEleutherias,now in the NationalMuseum in Athens.6 The figurewearsa chitonwhichis visiblebetweenthe lowerlegs (PI. 1:a). A few surfaceripples,more apparentto the touch than to the eye, stretchfrom leg to leg to give the materialtexture.The remainingdetailsare not modeledbut only painted.A broadblackstripedefinesthe parypheof the chiton a little to the properrightof the centralaxis. A narrowerblackstripewith thin outer reservedline decoratesthe hem. Wherethe parypheand hem meet is a painted,doublezigzagfold. The remainderof the chitonis a lightred brown. The himationhangsdown the axis of the left leg in a series of modeled,stacked zigzagfolds.Theyendjust abovethe left anklein a tassle.On the properleft side three shallowlyincisedlines curve up and backfrom the calf towardsthe lap (P1.1:b). Too little is preservedof the left side behindthis pointor of the rightto knowwhetherthe remainingsurfacewas modeled.The backis not. The blackand reservedborderof the hem curvesdown over the ankleand chitonto encirclethe entire base of the statue.7 On the rightside the borderbeginsto curveupwardtowardthe missingrightankleas if to meet a similarfall of stackedfoldson the rightleg. The himationis a darkershadeof red brownthanthe chiton.8 The preservedanatomicalmodelingis confinedto the left foot and is quite skillfullydone.Tracesof red paintedstrapson the firsttwo toes andpossiblyover the instep suggestthat the feet were sandaled.The foot is slenderthroughthe instep (W. 0.048 m.), widening through the ball with a pronounced convex curve (W. 0.073 m.). The

toes are parallelto each other, includingthe little one, and the big toe may be very slightlylonger.They are long, slenderand bony.Thereis some modelingof the meta5Thisbreakis plasteredand cannot be fully examined. 6W. H. Schuchhardt,"SitzenderDionysos," AntP VI, 1967, pp. 7-20. 7Deonna, op. cit. (footnote 3 above), p. 50, thought that the hem of the chiton continued around the statue. It is clear, however, that above the right ankle the himationcovers the chiton. The borderis largely obliteratedaroundthe back but can be followed. 8MunsellSoil Color Charts, Baltimore 1973, slightly redder than lOR 3/2. The surfaces of both garments are worn, and it may be that the presentdifferencesin color were not originallyso obvious.

SCULPTURE: ACROPOLIS ATTICTERRACOTTA 30

3

tarsalbones.In profilethe foot makesa continuouslydescendingline fromthe ankleto nearthe missingtoe tips.The surfaceof the foot is well polished. Surroundingthe statue is a thin solid plinth ca. 0.01-0.02 m. in thicknesswith plainverticaledge (P1.L:c).9The plinthprojects0.01 m. beyondthe backof the statue, 0.03-0.033 m. beyondthe sides, and was probablyflush with the toes. It is 0.308 m. from side to side, 0.324 m. front to back.The surfaceof the plinthis roughand unpainted.In the preservedportionsthere are no nail holes for attachmentto another surface. With the exceptionof the solid foot, the statueis hollow (P1.L:d).The wallsare extremelythin (Th. 0.015-0.025m.). Althoughthe interiorsurfaceand floor are covered with plasterand can no longer be examined,it is unlikelythat the piece was thrownon a wheel, as Deonna thought.10 In plan the statue is not circularbut more with flattenedsides and frontface." It was undoubtedlybuiltup by horseshoe-shaped, hand by means of long stripsand wads of clay, the joints of which were smoothed thereafterand coveredon the exteriorwithfine clay.How the plinthwas handledwithin the bodycavitycannotnow be determined.Most likely,the bottomof the statuewas eitherentirelyopen, or largelyso, in orderto facilitatecirculationof air duringfiring. The two colors,red brownand black,combinedwith reservedgroundto give a threecolorsystem,wereappliedbeforefiring. The clay from which the statue was modeledis coarseat the core with a dense admixtureof darkinclusionsand has fireda rosy to purplishred.12The basicmodeling of the form is executedin this clay. Over it is a coat of fine clay withoutinclusions whichvariesin thicknessfroma thin skin to 0.005 m. It has firedto a graybuff.13The fine clay is employedfor the modelingof finer surfacedetails. As Deonna also remarked,the clayis Attic. The generalappearanceof the statue is extremelysimple. What little modeling there is, is concentratedon the frontface. Paintis used as a shortcut to modeling,as exemplifiedby the treatmentof the parypheas well as of the hem of bothgarments.On the properleft side the draperyrises to expose the entire foot, yet the foot is only modeledfrom the instep forward.It is as if the sculptororiginallyhad intendedto representthe himationdrapingover the foot, as is morecustomary,but then paintedin 90f the plinth the front half of the right side, most of the front, and the back left corner are missing. The plinth has been mounted on a plaster slab for better support, but the underside can no longer be examined. 10Deonna,op. cit. (footnote 3 above), p. 50. Because of the plastertechnicalobservationsare limited. This is regretablewith regard to the problem of interior struts. Terracottastatues were generally built without an armature,but not infrequentlyin the Archaic period a thin clay wall was built from front to back up the center of the statue in order to prevent the walls from collapsing.Such a wall occurs in the Zeus-Ganymedefrom Olympiaand might be expected in the Acropolispiece. "In the view from above (PI. 1:d) the statue looks oval but the restorationof the properleft back is misleading. 1210R 4/3, an approximatereadingsince the surfaceswere difficultto approach. 13Afresh chip in the knee is ca. 10YR 5-6/4.

4

NANCYBOOKIDIS

the hem at a higherlevel, therebycreatinga plainstretchof bodywallextendingdown to the plinth.The statue,moreover,is extremelycompact.The plinthis almostsquare. The sides of the statueare aboutflushwith the edge of the plinth.The feet are turned out to fill the frontcornersof the plinth. Large-scaleseated statuesare uncommonin terracotta,possiblybecausethey are more complexstructurally than standingfigures.Apartfrom the seated statueswhich restedon the ridgecovertiles of earlyEtruscanbuildings,"4 threeexamplesare known, andall threearefromthe GreekWest.The firstfromGrammichelein Sicily,0.73 m. in height,is thatof a very simplyexecutedwomanon a largeblockseat whichhas flaring ends.15The second,0.98 m. in height,fromthe samesite andalso female,is elaborately enthronedand attired,indeedtoo muchso for usefulcomparisonwith our piece.16The thirdwasfoundat Paestumandrepresentsa drapedandbeardedman, 0.905m. or more in height.Althoughthe arrangement of the draperyis differentfrom ours, the oblique placementof the lowerlegs is similar,as is also the generalsparenessin the renderingof details.The Paestumstatuehas been datedon stylisticgroundsto ca. 530B.C."7 Terracottasculptureas a categoryis ratherdifficultto date in absolute terms. Bronzeaffordsthe best comparisons,but large-scalebronzestatuesare rarein the 6th centuryB.C. Whileclayexhibitsthe generaltendenciesof stone workin this period,it tends to be simplerin execution.Such is clearlythe case when the treatmentof the The draperyof our statue is comparedwith examplesin marblefrom the Acropolis.18 amountof surfacedetailis muchless. Thereis little attemptto distinguishone garment from anotherand virtuallyno attemptto indicatethe anatomybeneaththe drapery. Nevertheless,sincestone sculptureis moreabundant,we must look to it for parallelsin the executionof specificdetails. As we notedin the beginning,the pose of our statueresemblesthat of the marble figure from Plateia Eleutheria,dated to ca. 530-525 B.C., more particularlywhen viewedin profile.It differsfrom the more commonpose of the statuefrom Grammichele, in whichlowerand upperlegs are at rightanglesto each other. A furtherlikeness betweenthe two statuescan be found in the way in which the feet of both are modeled.The contoursare similar,as is the treatmentof the toes. The terracottafeet 141.EdlundGantz, "The Seated Statue Akroteriafrom Poggio Civitate (Murlo)," Dialoghidi archeologia 6, 1972, pp. 167-235. 15p. Orsi, "D'una citta'greca a Terravecchiapresso Granmichelein provinciadi Catania," MonAnt7, 1897, cols. 217-220, pl. III. 16p. Orsi, "Anathematadi una citta'sicula-greca,"MonAnt18, 1908, cols. 136-145, pls. IV, V. 17p. C. Sestieri, "Statue fittile di Posidonia," BdA 1955, pp. 193-202. Feet and base are missing. Found near the north side of the "Neptune"Temple, the statue was tentativelyidentifiedas a cult statue. In view of the elaborateand unusual architecturalterracottasthat have been found at Paestum in recent years, one cannot overlook the possibilitythat the statue belonged on a roof. Cf. P. C. Sestieri, "Terrecotte Posidoniati,"BdA 1963, pp. 212-220. der Akropolis,Frank"8H.Schrader,E. Langlotz,W. H. Schuchhardt,Die archaischenMarmorbildwerke furt 1939, pp. 107-116, 207-212. The closest, if any may be so-called, is Acropolis655, p. 107, no. 57, figs. 63-64, a statuette identified as Cybele. The chiton has a double stacked fold, while the himation falls symmetricallyover both legs. It is dated to ca. 550 B.C.

30 SCULPTURE: ACROPOLIS ATTICTERRACOTTA

5

may have a sharperinclinationin profile.The stackedzigzagfolds of the himationcan be paralleledin a terracottastatue of a standingdrapedmale from Corinth,dated to about the same time as the marblepiece.19A date in the late third or early fourth quarterof the 6th centuryis also in keepingwith the colorscheme,the use of predominantlycoarseclay,andwiththe consistentlythinwalls,if Corinthianstandardsapply. We come to the question of the statue's function. It could have served as a free-

standingvotive, as Deonnahadsuggested,a freestandingcult image,or an architectural element.Seatedvotive statuesin marbleare not uncommonon the Acropolis.Thirteen fragmentsare listed by Langlotz.Ourstatue, however,shows a majordifferencewhen comparedwith those. The seat or stool on whichour figuresat is not represented.We understandthe pose fromthe positionof the legs, andwe must assumethat the seat is coveredby the continuoushem of the himation.20 Indeed,when comparedto all other seatedstatues,largeand small,in marble,limestoneand terracotta,our statueremains uniquein this respect.21 This omissionis disturbingin a freestandingdedicationwhich would have been subjectto closer scrutiny.It impliesthat what was requiredwas a simpleoutlineof the pose withoutthe accouterments.It is clearfrom the handlingof the foot, moreover,thatour sculptorwas not unskillful.He simplytook shortcuts. A technicalobservationmay also be revealing,althoughtoo little is as yet known aboutthe specificpracticesof terracottamodelingto be certainof its validity.Corinthian freestandingstatuesof this periodare made in one with a hollowbox plinthopen on the bottom.Examplesof such can be seen in the Warriorand the Athenagroupsfrom Olympia.22The thin, solid, slab plinth of our seated figure differs and is to be paralleled

in a specifickindof architectural sculpture,as we shallsee below. The shorthandiconography,then, combinedwith the structureof the base, argues againstDeonna'sinterpretation and againstthe possibilityof a veneratedcult image.23 We returnto our initialqueryandto the subjectof architectural decoration. Despitethe excellentparallelsprovidedby the seatedZeus and Herain the Introductionpedimentfromthe Acropolis,it is unlikelythatour statuestood in a pediment. In the Archaicperiod,groupcompositionsin clay were workedin units of severalfig'9R. S. Stroud, loc. cit. (footnote 2 above).

20Tosuggest that the figure is merely squattingis to disregardthe postion of the thighs and to call to mind bodily functionsnot befittinga sanctuary. 2ITogive a corpusof seated statues is beyond the limits of this article.Useful references to late Archaic examples are in B. S. Ridgway, The ArchaicStyle in GreekSculpture,Princeton 1977, pp. 121-139; J. Boardman,GreekSculpture,TheArchaicPeriod,London 1978; W. Fuchs, Die Skulpturder Griechen,Munich 1979, pp. 248-258. For figurines,F. Winter, Die TypenderJfigirlichenTerrakotten I, Berlin 1903. Even the limestone kourotrophosfrom MegaraHyblaia,which might be describedas overridinglyfrontal, is seated on a block seat. Cf. G. V. Gentili, "MegaraHyblaia,"NSc 8, 1954, p. 99, fig. 24. 22E.Kunze, "Terracottaplastik," OlBerV, Berlin 1956, pp. 114-127, pis. 74, 75; E. Kunze, "Terracottaplastik,"OlBerVI, Berlin 1958, pp. 169-188, pls. 77, 78. 23Beforeleaving the subject of freestandingsculpture,we should note a reference of Pausanias(i.2.5) to clay statues depictingAmphiktyonfeasting with Dionysos and other gods, which stood in an oikema in the Kerameikos.Although the citation is tantalizing,the statues, if such they were and not large reliefs, were undoubtedlynot Archaicand need not have been of Attic manufacture.

6

NANCYBOOKIDIS

ureswhichstood, again,on the hollowbox plinth.We have mentionedthe Warriorand Athena groupsfrom Olympia.Even more fitting an analogyis the Amazonomachy pedimentfrom Corinthwhichunites at least three figuresof nearlytwo-thirdslife-size on a singlebox plinthY14 A self-contained statueon a thin slabplinthwouldnot be out of placeas an acroteFew Archaicbuildingspreserveevidence rion but probablynot as an apex acroterion.Y5 for the workingsof the apex.Two whichdo, however,are the laterTempleof Aphaia at Aigina26 and the Templeof Apolloat Karthaia,Keos.27On both buildingsthe acroterion is carvedin one with, or fitted onto, a gabledplinthwhich, in turn, fits into a cuttingin the roof. At Karthaiathe cuttingin the roof beamalso slopesgentlyto both fromOlymsides. Gabledplinthsalso can be foundin clay, as on the Zeus-Ganymede at an angle latter The of the rises from Corinth. Nike plinth a pia,28andon fragmentary In otherwords,we shouldexpectthe plinthof an apex acroteof ca. eleven degrees.29 rionto slope. this worksout well. A terraWe are thus left witha corneracroterion.Structurally, Behind cotta sima from Kalydonprovidesus with a good exampleof the technique.30 base with sunkenfloorinto whichis set the the cornerlion-headspoutis a rectangular flat plinthof the acroterion.The Kalydonslab plinthwhich is still leaded into place preservesthe fourpawsof a sphinx. A quicksurveyof other such bases for acroteriashows that there is no fixed rule abouttheir size. Oursis unusualin that it is squareand thereforemassive.Its maxi24R.Stillwell, "A TerracottaGroup at Corinth," ClassicalStudiesPresentedto EdwardCapps,Princeton 1936, pp. 318-322. S. S. Weinberg, "TerracottaSculptureat Corinth," Hesperia26, 1957, pp. 306-307, no. 8. of ArchaicGreekAkroteria, 251regret that the dissertationof MarilynGoldberg, Typesand Distribution cf. K. Volkert, Das Akroter treatment, earlier an For me. to diss. BrynMawrCollege 1977, was unavailable 1932. Frankfurt Zeit, I, Archaische Baukunst, griechischen der besonders in derAntiken 26A.Furtwingler, Aegina,Munich 1906, pp. 274-295, pl. 47. The predecessorof this temple, from the first half of the century, shows a differentsystem. The acroterionis a large palmetteplaquewhich becomes the apex of the sima. Cf. E. L. Schwandner,"Der iltere Aphaiatempelauf Aegina," Neue Forschungenin griechischenHeiligtimern,TUbingen 1976, p. 111, fig. 8. A marble floral acroterion from the Athenian Acropolisis cut at the bottom to fit either onto the sima or onto a gabled base, T. Wiegand, Die archaische derAkropoliszu Athen,Cassel 1904, p. 182, fig. 191. Finally, a diagonalcutting in the peak Poros-architektur of the SiphnianTreasuryis mentioned but not describedfully. Cf. P. de la Coste-Messeliere,Ch. Picard, FdD IV, ii, Paris 1928, p. 165. 27AEXT, 1963, XpovLKa (19651,pp. 281-282, pl. 327:c. 28Kunze,OlBerV (footnote 22 above), pp. 103-114, pls. 57-62. The Zeus-Ganymedehas been identified both as a freestandinggroup and as an acroterion.Most recently as an acroterion, cf. A. Mallwitz, OlympiaundseineBauten,Munich 1972, p. 36. 29S. S. Weinberg, op. cit. (footnote 24 above), p. 312, no. 27.1, pl. 69, who did not note that the plinth actuallyslopes. If we are correct in assuming that apex acroteriain the Archaicperiod had gabled plinths, then compositions such as the Silene and Maenad from Olympia, originallyplaced at the apex, must be relegatedto the angles, cf. G. Treu, Olympia,III, Die BildwerkevonOlympiain Steinund Thon,Berlin 1897, pp. 35, 37-40, pls. VII.2, 3, VIII.1, 2. Athens 1951, pp. 73-74, figs. 52, 53. E. Dyggve, Das Laph30K.Rhomaios, KEapao rrjcKaXv&Cvoo, rion, Copenhagen1948, p. 173, figs. 177-180, pl. XXII:C.

ATTICTERRACOTTASCULPTURE:ACROPOLIS30

7

mum length, however, is no greater than that of several other acroteria and much smaller than the horse-and-ridergroup from the Athenian Treasuryat Delphi.3" Vitruvius (iii.5.12) recommended that angle acroteriaequal the height of the tympanum while central acroteriashould exceed it by one eighth. If such a canon existed in the Archaic period,32then we should expect to find a small building with a pedimental height of ca. 0.70 to 0.75 m. for our statue, or about the size of the Red Triton and Hydra pediments from the Acropolis. If there was no such relation, then the building could have been even larger.Given the date and materialof our statue, we should also associate it with a sima of the so-called black-figure"Wellensima" type in terracotta. Regrettablythe examples of this type from the Acropolis are extremely fragmentary.33 No direct attributionscan be made. If Schuchhardtis correct in associatingthe series of small marble simas with the small decorated poros pediments,34then our statue can belong with none of the latter. We are left with an unknown small structure of the second half of the 6th century.35 We have not discussed the identity of the figure, a problem perhapsas elusive as the building on which it stood. Although parallelsexist for seated men who wear chiton and himation, our statue is undoubtedlyfemale. This is suggested by the short draping of the himation which does not completely cover the upperlegs but falls open to expose the chiton. The difference is apparentwhen one compares the statue from Paestum or the Zeus from the Introduction pediment with any of the seated females from the Acropolis. Our statue could represent Athena who is often depicted sitting. Without some attribution,however, this can only be a supposition.More importantis the question of whether her stance is significant.The customaryposition for a seated figure is with legs and feet together or closely set. Some of the later Milesian statues show legs parted, presumably to allow more room for elaboration of the drapery.36The spread of our statue's legs is greater and is accentuated by the outward turn of the feet and by the 31Afew comparativemeasurementsare as follows: Kalydon, 0.164 x 0.245 m., Dyggve, op. cit. (footnote 30 above); Aigina, Temple of Aphaia, 0.25 x 0.33 m., Furtwdngler,op. cit. (footnote 26 above); MegarianTreasury,ca. 0.18 x 0.255 m. The best preservedof the acroteriafrom the AthenianTreasuryis at least 0.80 m. long, 0.30 m. wide and 0.95 m. high. Cf. P. de la Coste-Messeliere, FdD IV, iv, Paris 1957, pp. 182-187. 32Fora discussion of this question, P. Lehmann, Samothrace,3, TheHieronI, Princeton 1969, p. 351 and note 183, p. 386 and note 235. 33E.Buschor, Die Tonddcher derAkropolis,I, Die Simen, Leipzig 1929, pp. 16-19. 34W. H. Schuchhardt,"ArchaischeBauten auf der Akropolis von Athen," AA (JdI 78), 1963, cols. 797-822. It is interestingto note that none of the corner pieces appearsto preserve any evidence for the attachmentof an acroterion. 35Forthe most recent summaryof the problemsof the small buildings, S. Bancroft, ProblemsConcerning the ArchaicAcropolisat Athens, diss. Princeton University 1977, University Microfilms7920425, pp. 46-57. Discussions of the Acropolisbuildingsgenerallydo not take into account the terracottasimas published by Buschor, op. cit. (footnote 33 above). 36K.Tuchelt, Die archaischenSkulpturenvonDidyma,Berlin 1970. Among the later statues with spread feet are his K60, p. 90, pl. 59; K63, p. 92, pl. 61; L95, L96, L99, L100, pls. 85, 86.

8

NANCYBOOKIDIS

fainthorizontalripplesof the chiton.It calls to mind the stanceof a matureor elderly woman.It is tempting,moreover,to think that the pose is determinedby something held in the lap, suchas a child,and to view the statueas a representation of a kourotrophos. Kourotrophos is a fittingsubjectfor the Acropolis.Pausanias(i.22.3) mentionsa Sanctuaryof Ge Kourotrophos which stood near the entranceto the Acropolis,and, accordingto a late source,Erichthonios decreedthatall sacrificesmustbe precededby a sacrificeto Ge Kourotrophos.37 Thereare two problemswithsuch an identification. Suchconsciouscharacterization of maturityis unusualin Archaicart.It is furtherunlikelythatan Archaicbabyplaced on an Archaiclap wouldcause the legs to spreadand the draperyto pull.Ourprimary large-scaleexample,the limestonekourotrophos fromMegaraHyblaia,is interestingin this respect.38 Althoughshe nursestwo babies,her legs couldbe those of any simple, seatedfigure.The emphasisis on her breasts. If the explanationis not iconographic-andwe have not exhaustedthe possibilities-it could be stylistic.The.sculptorchose the compositionin orderto give an ima pressionof bulkandmassiveness.At the same time, he couldgive the superstructure broad,firmfoundationwhilekeepingthe thinbodywallswhichfiremoreeasily. We are still left with the questionof why a seatedfigureshouldhave been placed on the roof of a buildingon the Acropolis.We generallyassociateacroteriawith movement, strengthor action.Thus, we have wingedfiguresand animalsor scenes of rape It may be that the reasonfor and combat.We do not thinkof passiverepresentations. of the our subjectis to be soughtin the realmof Atheniancultsand in the topography Acropolis,a separatestudyin itself. The answer to our opening query may well be no, as Mr. Thompson suggested.

ArchaicAthenians,unlikeArchaicCorinthians,may have used clay only for architecturaldecoration.Havingansweredthatquestion,however,we must still finda placefor our statue. NANCY BooKIDIS EXCAVATIONS CORINTH

37Eitrem,s. v. Gaia, RE VII, 1910, cols. 467-479. L. Deubner, AttischeFeste, Berlin 1966, p. 27, for sacrifices. 38G.V. Gentili, op. cit. (footnote 21 above). Acropolis 655 (footnote 18 above) is a further case in point. The attribute,a lion(?), is simply placedon her lap.

PLATE

A~~~~~~~~~~~~~~~' -7

b. Left side

a. Front

.pg NAY

c. Back

B

I

above from ed. .* View ,, t

1

DROUGHTAND FAMINE IN THE 4TH CENTURYB.C.* XCAVATIONSIN THE AGORA have broughtto light much informationconcerningthe watersupplyof Athens, fromgrandiosefountainhouses to the humblest of wells cut into the soft gray-greenbedrockwhichunderliesmost of the city. The evidenceillustratesthe continuouseffortandingenuityrequiredof the Atheniansto provide the largepopulationof Greece'sgreatestcity with a reliablesupplyof water.As a resultof the excavationsin the Agoraand elsewhere,it is now possibleto reconstruct the long historyof the supply.When viewed againstthis background,certainperiods standout as beingespeciallynoteworthy;the late 8th century,for instance,whenAthens sufferedthe effectsof a severe drought,and the 6th century,when the Peiapparently sistratidsprovidedthe city with the famed Enneakrounosand other fountainhouses.1 Anotherperiodof interestis the 4th centuryB.C., especiallythe secondand thirdquarters, when the Athenianspaid particularly intense attentionto the problemof water supply,both publicand private.The archaeological evidencesuggeststhat duringthis periodthe city sufferedthe effectsof a prolongeddryspellor drought. E

PUBLICSUPPLY

In the Agora,a largefountainhouse was constructedin the southwestcorner.The buildingis in a pitiablestate of disrepairand its very identification as a fountainhouse dependson the great stone aqueductwhich broughtwater to the buildingfrom the east.2The channellies underneaththe road which runs along the south side of the Agoraand since its discoveryin the 1930'shas been tracedfor some 220 meters,from the fountainhouse eastwardto the limits of the excavatedarea of the archaeological zone. The source of the line has never been determined.As preserved,it was built throughoutwith large limestoneslabs for floor, walls, and cover, and it is the most substantialwaterline yet discoveredat Athens.3Of the fountainhouse it supplied,only *It is a pleasureto dedicate this to Homer A. Thompson, a mentor and friend for so many years. His long and varied career began with a study of the ancient grain trade, and I hope that the present piece, concernedas it is with both grainand the Agora, will be seen as an appropriatecontributionto this volume in his honor. 'For the 8th-centurydrought:Hesperia48, 1980, pp. 397-411; for the Peisistratidfountain houses: R. E. Wycherley, TheAthenianAgora, III, Literaryand Epigraphical Testimonia,Princeton 1957, pp. 137-142, and Doro Levi, "Enneakrounos,"ASAtene39-40, n.s. 23-24, 1961-1962, pp. 149-171. 2For the fountain and its aqueduct:H. A. Thompson and R. E. Wycherley, TheAthenianAgora, XIV, TheAgoraof Athens,Princeton1972, pp. 200-201, with earlierreferences. 3The inner dimensions of the aqueductare ca. 1.28 m. high by 0.50 m. wide, large enough for a man to walk through.Waterwas carriedin a channel in the floor ca. 0.30 m. wide by 0.16-0.30 m. deep.

10

McK.CAMPII JOHN

survive.These, however,are the foundationsand a few blocksof the superstructure enoughto give some idea of the basicplanof the building.A smallsquarecourtin the northwestcorner,pavedwith stone slabs and open to the sky, gave accessthrougha colonnadedporchof unflutedDoriccolumnsat the southand east to a largedrawbasin. The drawbasin was separatedfrom the porchby a low parapet,over whichwatercould be drawn.The parapetwas supportedat intervalsby unflutedcolumnssimilarto those used in the exteriorcolonnade.The drawbasinwas L-shaped,with a floor areaof just over 100 squaremeters.The foundationswere of soft yellowlimestone,the lowerpart of the superstructure, of hardgrayPiraeuslimestone.Detailsof constructionabovethe stylobateare obscure. Recent excavationsby Homer Thompsonwere carrieddown to bedrock,and a bowlwas foundin the floorpackingof the northwestcourt.It is of a smallblack-glazed typegenerallydatedto the middleof the 4th centuryB.C. or slightlylater.Fragmentary potteryfromunderthe floorin the same areaseems also to dateto the thirdquarterof the century.4 and the lackof any ancientreferenceto it, the southDespiteits poorpreservation west fountainhouse representsa significantadditionto the Athenianwatersupply.It lay in a prominentlocationnear to both the Agoraand a majorcrossroadswhichlies just to the southwest.Its dimensionsshowit to be the largestfountainyet discoveredin Athens. It was clearlyan importantpublicproject,perhapsto be associatedwith the Combinedwith its impressivesupply of eitherEuboulosor Lykourgos.5 administration aqueduct,the buildingis the principalhydraulicundertakingof its era, matchingin scopethe Mycenaeanfountainandthe Peisistratidsystemof earliertimes.6 Justwithinthe citywalls,set up againstthe innerside of the Dipylongate, thereis a fountainhouse whichfirstcameto lightin the 1870's.The entireplatformof the building is preserved,and numerousclamps,dowels,settinglines, and weatheringmarkson the uppersurfacepermita secure restorationof most of the building.7The fountain measured8.00 m. north-southby 11.25m. east-west.It hadan L-shapedbasinalongthe the buildingfrom the northand west, suppliedby a stone aqueductwhichapproached areain the southwestcornergave accessto the basinand was north.A marble-paved 4P 27562: H. 0.035 in., diam. 0.074 m. Complete, mended from several pieces. Small black-glazed bowl with thick incurvedwall and high ring foot. Thin black glaze; very worn, reddish brown clay. Similar to L. Talcott and B. Sparkes, The AthenianAgora, XII, Black and Plain Pottery,nos. 945-950, dated ca. 350-325 B.C. Also potterylots, Section K, nos. 346 and 347. 5G. L. Cawkwellhas suggested that Euboulos may have served as water commissioner: "Eubulus," JHS 83, 1963, pp. 47-67. Lykourgosseems likely on generalgrounds as the instigatorof numerous public buildingprojects:F. W. Mitchel, LykourganAthens,338-322, Lecturesin Memoryof Louise TqftSemple,2nd ser., Universityof Cincinnati1970, pp. 34-35, 48-49. 6For the Mycenaeanfountainon the Acropolis:0. Broneer, "A MycenaeanFountainon the Athenian Acropolis,"Hesperia8, 1939, pp. 317-433. See footnote 1 above for the Peisistratidhydraulicinstallations. 7G. von Alten, "Die Thoranlagenbei der HagiaTriadazu Athen," AthMitt3, 1878, pp. 37-39 and pl. IV; also G. Gruben, "Die Ausgrabungenim Kerameikos,"AA (JdI 79), 1964, p. 407 and idem, "Dipylon und Brunnenhaus,"AA (JdI 84), 1969, p. 39. Final publicationby ProfessorG. Gruben is in progress.

ANDFAMINE INTHE4THCENTURY B.C. DROUGHT

11

itself enteredfrom the south througha colonnadeof three columns.Wearin the floor suggestsplacesof easy access to the basin or the positionof spouts cut throughthe parapet.As a resultof recentexcavations,the fountainhousehas been datedto the third quarterof the 4th centuryB.C., accordingto the dateof the ceramicmaterial. sourcesrefer to other In additionto these two fountains,literaryand epigraphical hydraulicinstallationsin this periodas well. Most strikingof all is the monumental This was a largesupplyline apparently designedto bringwater "Acharnianaqueduct".8 to Athens from severalspringson the lower slopes of Mount Parnes,some 18 kilometersto the northof Athens.Fourinscriptionsattestits existence.These recordwater rights, collectinggalleries,and rightsof way for the conduitwhich were sold to the commissionersof the project.Largesums of moneyare recordedas havingbeen paidto propertyowners,and thereis reasonto believethat the oratorDeinarchoswas calledin ownerin a lawsuitover the aqueto arguethe commissioners'case againsta recalcitrant duct.By plottingthe findspotsof the four inscriptionsit has been possibleto show that the channelran from the slopesof Parnesalongthe line of the Kephissosrivervalley. Thoughit is not possibleto pointto any remainswhichcanwith certaintybe associated have takenthe formof a tunneldriventhrough with the aqueduct,it wouldpresumably when one rebedrock.It must have been an impressivefeat of engineering,particularly membersthat at this early periodthe aqueductwould have been a gravity-flowline throughoutits entirelength.' The careerof Deinarchosis confinedto the last thirdof the 4th century,and the letter formsof the inscriptionspointto the same datesIt has been suggestedthat the projectis to be associatedwith the surgeof buildingactivitycarriedout at the time of Lykourgos(338-326 B.C.,footnote5 above).The Athens'greatfinancialadministrator, largesums of moneypaidand the greatlengthof the systemboth indicatethat this was a majorpublicprojectof the firstorder,designedto increasethe watersupplyof the city in the secondhalfof the 4th centuryB.C. Finally,a decree of the year 333/2 B.C.honors one Pytheasof Alopeke for his performanceas watercommissioner.He was awardeda gold crownworth 1000 drachmas for repairinga fountainat the Amphiareionat Oroposand for buildinga new fountainin Athens or Piraeus.10 It is worthnotingbrieflythe prominenceof the water commissionersat this time. The positionwas one of the very few in the Athenian democracywhichwas electiveand not allotted(Aristotle,Ath. Pol. 43.1); the job was simplytoo importantto leave to chance.Pytheasand possiblyone other commissioner 8The Acharnianaqueducthas been brought to light by Eugene Vanderpool,a friend and teacher also being honored in 1981 on the occasion of his 75th birthday:"The AcharnianAqueduct,"Xapxr ptowvEL 'AvarTaratovK. 'OpXavov I, Athens 1965, pp. 166-175. On p. 174 he notes Athenian concern over water supplyat this period. 9The earliest pressureline that I am awareof from Athens dates to the second half of the 3rd century B.C.: Agora inv. no. A 2295, M. Lang, Waterworks in the AthenianAgora, Agora Picture Book No. 11, Princeton1968, no. 24. '01G II2, 338. The fountain was set up in front of a temple of Ammon, but otherwise the location is not specified.

12

JOHNMcK.CAMPII

are the only two knownto have been honored,both in the thirdquarterof the 4th centuryB.c."1 PRIVATESUPPLY

An interestingphenomenonoccursin the patternof privatewatersupplyat this same time. The bottle-shapedcisternreplacesthe well as the standardsourceof supply in privatehouses.As the nameimplies,these cisternstakethe formof a bottleor flask, startingwith a verticalroundshaft ca. 0.80-0.90 m. in diameter,openingout to 2.505.50 m. acrossat the bottom.They are anywherefrom threeto seven metersdeep, for the most partcut throughthe soft bedrockand lined with a thick coat of waterproof cement.Oftenthereare archedtunnelsleadingout of them, some intendedto connect one cisternto another,otherssimplyto provideincreasedcapacity.Such cisternswere designedto collectand store rainwateroff roofs, and with one or two exceptionsthey are foundin the courtyardsof privatehouses.They are a standardfeatureof Athenian domesticarchitecturethroughoutthe Hellenisticperiod,and no fewer than 140 have come to light in the residentialdistrictsaroundthe Agora;others are known from elsewherein the city. The natureof their construction(cut into bedrock)and the fact that they were apparentlycleanedout with some regularitymake precisedatingdifficult.A construcfor dates we are tion date or even a periodof use is consequentlyoften unobtainable; it ceased to be used. We a at the time into cistern thrown the material on dependent maysupposethata cisternwas in use immediatelybeforeits abandonment,but thereis no wayof determiningthe lengthof time it mayhave been in use. The earliestcisterns aroundthe Agoraseem to have been filledup aboutthe middleof the 4th centuryB.C., They become and this suggeststheiruse as earlyas the secondquarterof the century.12 commonthereafter,into the Hellenisticperiod. increasingly This new developmentis complementedby an interestingpatternamongthe wells. 'Some 28 wellsare knownto have been in use in the 4th centuryB.C.in the areaaround the Agora.Only 13 wells have been foundwhichwere in use in the 3rdcentury,however. The tremendousdecreasein the numberof wells in use betweenthe 4th and 3rd centuries,combinedwith the introductionof cisterns,suggestsa radicalchangein the "In additionto the decree in honor of Pytheas, a second decree (IG II2, 215) honors one Kephisodoros, son of Kallias,of Hagnousfor his performanceas water commissioner (? restored), in 346/5 B.C. The water supply was a matter of considerableimportancein earlier times as well, and prominent men con31), cerned themselves with the problemthroughoutthe 5th century:Themistokles (Plutarch, Themistokles Kimon (Plutarch,Kimon13), and Perikles (IG I2, 54). In additionto Pytheasand Kephisodorosin the 4th century, see also footnote 5 above. '2Fourcisternsin the Agora (F 19: 2, L 17: 6, Q 13-14: 1, and R 18: 2) were filled in the years around 350 B.C. Similarcisterns are reportedfrom Olynthos:D. M. Robinson, Excavationsat Olynthus,XIII, DoBaltimore1946, pp. 307-309, pls. 101 and 123. As only eight were discovmesticand PrivateArchitecture, cleared, it seems likely that the cistern was a relativelyrecent develop100 than houses ered in the more ment at the time the city was destroyedin 348 B.C.

DROUGHTAND FAMINEIN THE4THCENTURYB.C.

13

watertableceasedto be tappedto the privatewatersupplyof the city.The underground greatextent it had been for centuries,and effortswere made to collectand save rainThe mainimplicationof this developmentwouldseem waterfromthe roofsof houses."3 to be thatwaterwas increasingly difficultto find.This is not a suddenshift but rathera gradualone, startingin the yearsaroundthe middleof the 4th centuryB.C. andcontinuing throughoutthe rest of the century. Thusconsiderableactivityconcerningthe watersupplyof the city is apparentin the 4th centuryB.C.,particularly the thirdquarter.As noted, for the publicsupplywe have no fewer than four majorhydraulicinstallationsat this time: the southwestfountain house, the Dipylonfountainhouse, the Acharnianaqueduct,and the fountainbuiltby at Pytheas.These presenta concentratedexpansionof hydraulicresourcesunparalleled any periodin Athens with the possibleexceptionof the 6th century,under the Peisistratids.14 Largenew fountainswere built and waterwas broughtin from as far away as MountParnes.This publicconcernis complementedin the privatesectorby a gradual shift from wells to cisternsduringthis period.The watertable apparentlysank too low to be tappedeasily,and whateverrainfell was caughtfrom the roofs and carefully hoarded.15The likeliest explanationof these developments is that Athens experienced a droughtin the 4th centuryB.C. The archaeologicalevidence may be supplemented by literary and epigraphical

sourceswhichshed furtherlighton this droughtand its effect on the historyof Athens and the rest of Greece.Its originsmay well be noted in a passagefrom Demosthenes (vs. Polykles,61): 13Regularrainfall,of course, is requiredto keep the ground water at an accessible level. During a dry spell in 1977, the water level in an unused well in the village of Ancient Corinthdropped2.65 m. between Februaryand October.In dry areas or during a dry spell much of whatever rain does fall is lost as runoff. Cisterns are common today in many of the driest islands of the Aegean. The inferior quality of cistern water was recognized in antiquity: Hippokrates, rEpIa''pwv,

v8aaTv,

TOrnWV,

8.20, and Vitruvius, viii.6.14.

Note also from this period the huge cisterns in the mining districtof Laureion.The care with which the water was recycledand the highly refined and effective waterproofingof the cisterns indicate the pains taken to preservethe water so crucialfor washingthe ore. On the compositionof the mortarof the Laureion cisterns (almost 30%lead): K. Konophagosin IHpaKTrKa 49, 1974 [1976], pp. 251-261; K. Konophagos and H. Badeca, Annalesgeologiquesdes pays hellknique,Athens 1975, pp. 328-337, and K. Konophagos, Le Lauriumantique,Athens 1980, pp. 253-273. 14There is no evidence of a droughtin the 6th centuryat Athens. Ratherthe Peisistratidsytems (footnote 1 above) should be consideredan innovation, the first attempt to provide the city with a publicwater supply, analogous to similar efforts in Corinth, Megara, and Samos at about the same time. Once built, these Archaicestablishmentsremained the backboneof the public supply throughoutthe Classicalperiod until the great expansionof resources-under discussionhere-in the latterhalf of the 4th centuryB.C. "Increased populationcannot, it seems, be advanced as an argument for increased public efforts to obtain water. Both Gomme (The Populationof Athensin the Fifthand FourthCenturiesB.C., Oxford 1933, p. 26) and Ehrenberg(The GreekState, 2nd ed., New York 1969, p. 31) estimate a steadilydeclining population for Athens throughoutthe 4th century. Nor, on the other hand, can a decrease in populationbe used to explain the drop in the number of wells, for the wells were replacedin their function as private supply by numerouscisterns.

JOHN McK. CAMP II

14 r)

& U)

OVX 07r0o

dEVaVJ, EKEVO)TO)(TJ

a,Our,8E

Ttva Kap7rOv 79v11EyKEv,aXWa KatLTo' iwp W' 7raves

XAaxavov yEvEOat

UTTE, EK TWv OpEa7wv Ev a(, Kcr(0.

6v

EarExrEv,

my land not only produced no crops, but that year, as you all know, the water even dried up in the wells, so that not a vegetable grew in the garden.

The date, 361 B.C., correspondscloselyto whatwe have seen concerningthe beginning of the shift fromwells to cisterns,in the secondquarterof the 4th century.The effects of the droughtcontinuedto be felt, apparently,for in 357, accordingto Demosthenes (vs. Leptines,33), there was a universal shortage of grain (a&xxa Trp rEpvnt

LTrO8Eta4;

rapa 7Tau-tvavOp6ToL' 'yEVot4EvrE). These difficultiesmaywell have been the impetus

for a changein the natureof privatewatersupply.Activityin publicsupplyseems not to have been far behind;Demosthenes(Olynth.iii.29) refersto fountainsunderconstructionin the city in 349 B.C. Inscriptionsrefer to intermittentproblemswith grain supplyin the 330's,16and Athenssuffereda severe faminein the period330-320 B.C.17 The literarysourcespainta grimpictureof the city duringthe crisis: aEl To ) aTTEL OLKOVVTE( 8&E"'TpOV^vTOra

El~m )v 7s El TO) mexpLt) EXaC43aVOVKar' 6,8oXov a rquAEKToP roMv apTOVs KcatEMT T71s liaKpa' 0-oaq ra TaaX'tra, sa'aXLa ,.E7 ,~~~~~~~ra ,AETpOVtuVOL Ka KaTa7raTolVvoL. V1WOV Ot

VE

cb&"O,01 8'ETV,

Hapast

(Demosthenes, vs. Phormio,37) Those of you who dwelt in the city were having their barley meal measured out to them in the Odeion, and those who dwelt in Peiraeuswere receiving their loaves bit by bit in the dockyardand in the long stoa, having their meal measuredout to them a twelfth of a medimnos at a time and being nearlytrampledto death.

sourcesshed furtherlighton the shortagein the formof honorarydecrees Epigraphical for foreignerswho broughtgrainto the city. Here, too, specificreferenceis made to famine.18Individualswere honoredfor providinggrainat a low price and often the amountsare surprisingly small, as little as 3000 medimnoi.Togetherwith the literary showthatAthensreceivedgrainfromthe BlackSea, fromAsia sourcestheseinscriptions (335/4 B.C.) and IG II2, 342 (332/1 B.C.). Also Plutarch,Moralia851B. grecqueI, Paris 1925, "7Forthe extensive materialon the famine:A. Jarde, Les cerealesdans l'antiquite pp. 43-47; F. Heichelheim, s. v. Sitos, RE Suppl.VI, cols. 819-892; M. N. Tod, GreekHistoricalInscriptions II, Oxford 1948, no. 196 and commentary;M. Rostovtzeff, Social and EconomicHistoryof the Hellenistic WorldI, Oxford 1941, pp. 94-98, 168; J. Pec'irka,TheFormulafor the Grantof Enktesisin AtticInscriptions, Prague1966, pp. 70-72; S. Isagerand M. H. Hansen, Aspectsof AthenianSocietyin the FourthCenturyB.C., Odense 1975, pp. 17-19 and 200-208. Hitherto the famine has been seen as an economic problem rather than as the result of drought:H. Michel, The Economicsof AncientGreece,2nd ed., Cambridge1957, p. 275; M. Rostovtzeff, op. cit., pp. 97-98; W. W. Tarn, CAH VI, Cambridge1927, pp. 448-449; and W. L. Westermann,"New HistoricalDocuments in Greek and Roman History," AmericanHistoricalReview35, 1929, pp. 16-19. 18IGII2, 360, lines 8-9: Ev rjj Travoo-iL and IG II2, 398, line 11: arav~wz [orov yEvopEv7~1. For similar language in the literary sources, Aristotle, Oikonomika1352a: XquoviyEvogEvov Ev Tots aXXoL9 16G II2, 408

TO7rOL9

0fxo8pa.

INTHE4THCENTURY B.C. DROUGHT ANDFAMINE

15

Thoughthe city was never Minor,from Cyprus,from NorthAfrica,and from Sicily.19 in concentration of andthe specific the unusual honorary inscriptions self-sufficient grain, referencesto a shortagesuggestan especiallycrucialperiodof supplyin the 320's. It is of interestto note also that the faminewas not a localAthenianproblembut ratherseems to have been widespreadthroughoutGreece at this time. An important inscriptionfrom Cyrenerecordsthat huge amountsof grainwere suppliedto no fewer than41 cities in centralGreece, the Peloponnese,and the Aegeanislandsin the years around335-325 B.C.20 The openingphrasereadsas follows: ['IapIEvq lwoLaq Kak[XAa]8a [fHI[o-os cr701 ova

E&8KE a Iroxv

aTO8EL'a EyE'VETO Ev

Tat'TEka8

The priest Sosias the son of Kalliades.The city gave the following amounts of grainwhen there was a famine in Greece.

This is followedby a list of cities with the amountscontributedto each one. In all, 805,000medimnoiwere made availableby Cyreneto the Greek poleis. Such a widespreadfailureof the cropssuggestsa generalproblem,moreseverethanAthens'perennial failureto providefor herself.The likeliestexplanationis that not only Athens but also muchof the rest of Greecesufferedthe effectsof a droughtin the thirdquarterof the 4th centurywhichculminatedin the severefamineof the 320's. The evidence from several sites perhapsreflectsthe effect of this droughtelsewherein Greece.The systematicexcavationsof Corinthhave shed much light on the historyof thatimportantcity;thoughthe full storyof the watersupplyhas not yet been written,there are indicationsof concernfor the supplyin the second half of the 4th century.The SacredSpring,for instance,underwentseveralmodificationsin its long 19Themost complete document is IG 112,360, honoring Herakleides of Salamis in Cyprus for his benefactionsduring this period. Other decrees of a similar nature are IG II2, 363, 398, 400, 401, 407, 416, and possibly423 and 499; also Hesperia8, 1939, pp. 27-30, no. 7; Hesperia9, 1940, pp. 332-333, no. 39; Hesperia43, 1974, pp. 322-324, no. 3; and Hesperia49, 1980, pp. 251-255, no. 1. Many of the orationsof Demosthenes of this periodconcern the graintradeas well: xxxii, xxxiv.36 and 39, LVI.9, and xx.32-33. Also to be dated to this period (325/4 B.C.) is the projectedfoundationof an Athenian colony in the Adriatic,with the express intent of securing the grain route: IG II2, 1629, lines 217-220 (M. Tod, op. cit. [footnote 17 above], no. 200, pp. 284-289). Colonizationwas a standardresponse to famine, and it is interestingto note this late revival of the institution. The fact that most of the kernoi found in the Agora date to the 4th century (J. Pollitt, "Kernoifrom the Athenian Agora," Hesperia48, 1979, pp. 205-233, esp. pp. 226-227) might be taken as renewed interest in the cult of Demeter because of the droughtand subsequentfamine. Also related, perhaps,would be the exensive building programcarriedout at Eleusis and in the Eleusinion at Athens in the years around 329/8 B.C.as recordedin IG II2, 1672. 20G.Oliverio, Documentiantichidell'AfricaitalianaII, i, "La stele dei nuovi comandamentie dei cereali," Bergamo 1933, pp. 29-94. (Also Tod, op. cit. [footnote 17 above], pp. 273-276, no. 196). The date is set by grants of grain to Olympias,Alexander's mother, and to Kleopatra,his stepmother. See also Plutarch, Alexander,25, where it is recorded that after the siege of Gaza (332 B.C.) Alexander sent large amounts of the spoils home to both women. Accordingto the list, Athens received 100,000 medimnoi of grainfrom Cyrene, twice as much as any other city and 13 per cent of the total amount provided.

JOHNMcK.CAMPII

16

history.B. H. Hill describedthe natureof the supplyafterone of these alterationsas follows:"The amountof waterflowingfrom the lion head spouts must at best have As a resultof the most recent been very smallbut it was now very greatlyreduced."'" In excavationsin the area, this phaseis datedto the last quarterof the 4th century.22 addition,five wells went out of use in or aroundthe thirdquarterof the 4th century, and severalcisternsappearfor the firsttime.23A handsomebathsouthwestof the racewent out of use at aboutthis time.24Finallyone shouldnote the coursealso apparently unusualsupplysystem for the shops of the South Stoa, built duringthe latterpartof Thirty-oneof the shopswereprovidedwithwells.Thesewellsdid not the 4th century.25 tap the groundwaterdirectlybelow,however,but were fed by meansof a long tunnel which broughtwaterin from afar.These variousexamplesare not all equallysignifiand cant, perhaps,but, consideredagainstthe backgroundof the Cyreneinscription26 the evidencefromAthens, it seems likelythat Corinthalso experiencedsome difficulty withher watersupplyin the latterhalfof the 4th centuryB.C. Across the Gulf from Corinthlies the Perachorapeninsula,which was provided A handsomefountainhouse andseveralimmense withextensivehydraulicinstallations. cisternswereconstructedlate in the 4th century.27 A thirdsite, recentlyexcavated,deservesa word.The explorationof Halieisin the southernArgolidhas progressedfarenoughto show that the site and its environswere abandonedlate in the 4th centuryB.C.28 No compellingreasonhas yet been put forward to suggestthatthe site and it does not seem unreasonable to explainthis abandonment, as a resultof drought.The Cyreneinscriptionsuggeststhat the becameuninhabitable areaas a wholewas affected;Argos,Kythera,Troizene,and Ermionewereall provided with grain.With better sourcesof watersupplythese cities were able to last out the drought;perhapsHalieiswas not so fortunate.29 21B. H. Hill, Corinth,I, vi, Peirene,SacredSpring,Glauke,Princeton1964, p. 177.

K. Williams,II, Hesperia38, 1969, pp. 36-63. 23G.R. Edwards, Corinth,VIII, iii, CorinthianHellenisticPottery,Princeton 1975, pp. 198-212. Wells nos. 17, 20, 21, 24, and 43, cisterns nos. 22, 23, 37, 39, and 40. 24CentaurBath:C. K. Williams, II, Hesperia45, 1976, pp. 109-116 and idem, Hesperia46, 1977, pp. 51-52. 260. Broneer, Corinth,I, iv, TheSouthStoa, Princeton1954, pp. 59-65. 26Corinth(line 9) received 50,000 medimnoi of grain, the third largest amount after Athens and Olympias. 27R.A. Tomlinson, "Perachora:the Remains outside the Two Sanctuaries.Descriptionof the Waterworks of the UpperPlain," BSA 64, 1969, pp. 195-242. 28M.Jameson, Hesperia38, 1969, pp. 311-342, esp. pp. 328-329; W. Rudolph, Hesperia43, 1974, pp. 105-131, esp. pp. 130-131, and T. Boyd and W. Rudolph, Hesperia47, 1978, pp. 333-355. 29AtHalieis, right by the shore, the wells may have gone brackishif the water table fell at all. There are both wells and cisterns reportedfrom the 4th-centuryhouses at the site. For abandonmentas a gradual processas a result of drought, see Aristotle, Meteorologica,351b. Pausanias(ix.38.9) recordsthe abandonment of Aspledon in Boiotia because of a lack of water. Note also the Athenian colony planned at about this time (above, footnote 19). I should like to thank the following colleagues for readingearlier versions of this paperand making useful suggestionsfor its improvement:A. L. Boegehold, W. B. Dinsmoor, Jr., R. Townsend, S. Rotroff, and T. L. Shear, Jr. 22C.

DROUGHT AND FAMINE IN THE 4TH CENTURYB.C.

17

In conclusion,the archaeological evidence,takenwith the literaryand epigraphical sources,providesa coherentpicture.Greateffortsin the publicsectorand innovationin the privatesectorshow thatAthenianresponseto adverseclimaticconditionsin the 4th centurywas both imaginativeand extensive.Impressiveeffortswere made to securea reliablewatersupplyduringwhatseems to have been a periodof drought.At the same time, the evidencewouldseem to suggestthat this droughtwas the cause of a widespreadfaminewhichafflictednot only Athens but also much of the rest of Greece in the 2nd halfof the 4th centuryB.C. JoHNMcK.CAMP II ATHENIAN AGORA EXCAVATIONS

THE ASYMMETRYOF THE PINAKOTHEKE FOR THE LAST TIME?* rARIOUS REASONShave been proposedfor the asymmetryof the openingsin the windowwall of the Pinakotheke.The first one publishedwas in 1912 by GeorgeW. Elderkin,secretaryof the AmericanSchoolof ClassicalStudiesin Athens from 1908to 1910.1He pointedout thatthe west face of the east antaof the tristyle-inantiscolonnadealignedwith the east side of the east windowof the windowwall, and that the positionof the antawas fixed by a triglyphabove.2He believed,however,in He the now disprovedtheoryof a Classicalzigzagroad leadingup to the Propylaia.3 determinedthe positionand directionof his uppermostzigzagby theorizingthat it had to followa line fromwhichboth the windowsand the doorwaycould be seen through and finallypositedthat, of the colonnadeof the Pinakotheke,4 the intercolumniations from the aestheticviewpointalong "thatlast stretchof the zigzagroad,an asymmetric arrangementof door and windowswas absolutelynecessary."5G. Stevens, in 1946, thoroughlyacceptedElderkin'sidea.6He re-expoundedit andaddedan observationthat the open spacebetweenthe east antaand the easterncolumnis greaterthanthe actual spacings(he did not mentionthatthe widthof these spaceswas automatintercolumnar icallycontrolledby the triglyphfrieze).Enlarginguponthis observation,he went on to say that "the shiftingof the dooreastwardand the spreadingof the space [betweenthe east antaand easterncolumn]undoubtedlyeased the circulationfrom the centralportion of the Propylaeainto the 'PictureGallery'."W. B. Dinsmoor,in 1950, took into considerationthe cantedClassicalretainingwall whichalloweda straight,rampedroad some 22 metersin width.He believed,however,that only a portionof this rampwas V

*H. A. Thompson has a keen interest in, and an innate understandingof, ancient Greek architecture and its many problems. It is to be hoped that he will appreciatemy solution to one of the strangest of architecturalenigmas.The solution presentedhere appearsalso in tabulatedform at the end of this article. 1Problemsin PericleanBuildings,Princeton1912. 2Ibid.,p. 1. 'There are two preservedstretches of the almost straightretainingwall of the 11-meterwide Archaic rampwhich led up some 80 meters towardsthe early entranceto the Akropolis (see A. Keramopoullos,To 'ApX'EO, 1934-1935 [1936], p. HIEXapytKov, TO 'AcOKcXrlet1ov, at' 68ot at ava'yovoat 7Tpo' r Ha'poIratau, 87 and pi. I). These are now dated to 566 B.C. (E. Vanderpool,"The Date of the Pre-PersianCity-Wallof

Athens," 1OPOL.Tributeto B. D. Meritt,D. W. Bradeenand M. F. McGregor,edd., Locust Valley, New York 1974, pp. 156-160). Although the upperor eastern stretch was known, and dated as 6th centuryB.C. vonAthen, 2nd by W. Wrede (AttischeMauern,Athens 1933, p. 9, no. 20) and by W. Judeich (Topographie ed., Munich 1931, pp. 213-215), it was not recognizedby them as being partof a straightramp.The canted wall furthernorth retainedthe Classicalsuccessor to this straightramp, doubling its width. It should be concluded, therefore, that no zigzagapproachexisted after the middle of the 6th centuryB.C. 40p. cit. (footnote 1 above), p. 2. 5Ibid.,p. 5. 6"Architectural Studies," Hesperia15, 1946, pp. 87-89.

THELASTTIME? THEASYMMETRY OFTHEPINAKOTHEKE-FOR

19

executedand that it had to be supplementedby an unsymmetrical windingpathat the upperlevel.7He shunnedthe Elderkin-Stevens theoryfor the asymmetryof the window wallandconsideredthe off-centeringof the doorwayto be merelyfor easieraccessfrom the CentralBuilding,the same reasoningwhich he also used for the shorteningin width,in the planningstages,of the Pinakothekeitself (see below). In 1930,F. Francoevolvedthe imaginativeidea thata higher,widerdoorthanthe and withoutits attendantwinpresentone, centeredon its facing intercolumniation in a laterschemethe presentwindows,less high thanthe dows,was originallyplanned;8 aboutthe door.The doorwas then loweredto the doorway,were addedsymmetrically heightof the windowsto be, he said, in the linearspiritof Doric architecture,necessitating,for the sake of proportions,a reductionin its width.He suggestedthat the easternjambhadalreadybeen partiallyerectedby this time, andit was thereforetoo late to makeanotherchangein orderto allowsymmetry. J. A. Bundgaard, in 1957,suggestedthatthe openingscopiedthe theoreticalplacing of those of ArchaicBuildingB, whichhe consideredto have occupiedthe same spaceas the Pinakothekeand to have been dismantledto allowfor the erectionof the latter.9H. Riemannrefutedthistheoryon the groundthatsincethe central,interiorrowof columns in the Pinakotheke,thereis no reasonto believe of apsidalBuildingB wasnot reproduced thatthe hypothetically off-centeredlocationof the doorof BuildingB wascopiedeither.10 In 1964, C. Tiberitried to explainthe designof the Propylaiaand the reasonfor the off-centereddoor as evolvingfrom a complicatedgeometriccompositionof overlappingsquaresand triangles."For a soundrefutationof this idea, see the commentsof P. Hellstr6m.12

J. Travlos,in 1971, wrote:"The entranceto the Pinakothekewas shiftedoff axis ... so that dining-couchescould be installed." "When the door is off-centre [it allows]

for a normalarrangement of dining-couches.""3 For the refutationof this ideasee again P. Hellstr6m."4

Finally,in 1975,we have the explanationsof P. Hellstr6mfor the asymmetryof the windowwall.15He stressesa proportionate relationshipof the lengths of the sections 'Architecture of AncientGreece,London 1950, p. 198. 8"Le asimmetriedella Pinacotecadei Propyleisull'Acropolid'Atene," ASAtene13/14, 1930/31 [1933], pp. 9-25. 9Mnesicles,a GreekArchitectat Work,Copenhagen1957. "0Reviewof Bundgaard,Gnomon31, 1959, pp. 309-319. For my argument that Building B did not occupy this space at all, see W. B. Dinsmoor, Jr., ThePropylaiato the AthenianAkropolis,I, ThePredecessors, Princeton 1980, p. 2, note 10, and see also footnote 27 below. 1 Mnesicle,I'architetto dei Propilei,Rome 1964. 12"The Asymmetryof the Pinacotheca-Once More," OpusAth11, 1975, p. 87. "3Pictorial Dictionaryof AncientAthens,London 1971, p. 482. 14Op. cit. (footnote 12 above), pp. 87-89. In addition, the Pinakothekenever received a marble floor, as employed elsewhere in the building,since the face of the toichobatecourse againstwhich it would have abutted was not finished. Many of the toichobate blocks still retain lifting bosses. For the supportof dining-couchesone would expect a solid masonryfloor. l51bid.,pp. 87-92.

WILLIAM B. DINSMOOR, JR.

20

0

a

-

,-

:

:

?

?

-

-~~~~~~~~ ~~~~~~~~~0

I

co

~~~~~I

II @ ~ ~ Il6t

LI

~~21

(U

lj

ie~~~~

I

||N

*

;961t1

1

I

vz~wst r

-

szlz

o --|0~I 4

L

*x|ZZ

t--I1$**

slSS

<

8

l

l~~~~~~l 4

l

-

I

9

6

-

4~~~~~~~~U~

Hl

|I

14

Zz

-

21 U.

I

--

i

u l.," E

_,

_

=:=

8

0

-.2.

2~~~-

.

_.__~~~~~

11-

-

j/

|

9

1*

I

a,?

-

0 U)t giE

J

0

z

;

co

II I

'0

f~~~~~~~~~~~~~

o~~~~~~~~~ $-c

_

w 0~~~~~~~~~~~~~~~~~~'

LL. 0

-

0

22

WILLIAM B. DINSMOOR, JR.

of wallon eitherside of the windows,no two of whichare the samelength.Thereexists no vantagepoint, however,from whichthis phenomenoncould be appreciated in that one muststandwithinthe areaof the porchto see the wallin its entirety,withoutvisual obstructionby the colonnade,and thereis no placewithinthe porchthat is far enough removedfromthe wallto allowan all-encompassing view of it withinan angleof vision of 60?. Hellstrom'ssubtle proportionsof A:B as D:C for the wall segmentswould be totallylost throughperspectivedistortion.He does makevalidpoints,on the otherhand, when he iteratesthe often-madeobservationthat the blockdistributionin the wallwas carefullyworkedout, and when he observesthat this blockdistributionpreventedthe doorwayfrombeingcenteredbetweenthe windows,and thatthe innerwindowpilasters (i.e. those nearestthe door) "appear"to be centeredunderthe fourthand eighth triglyphsof the frieze, countingfrom the west. Hellstrbm,however,who unfortunately had to use Bohn'sand Bundgaard's measurements,has barelyscratchedthe surfaceof his idea.Now, by combininghis observationof the pilaster-triglyph arrangement andW. B. Dinsmoor'shighlybelievableconceptof the variousstages in the planningof the Propylaia,the reasonsfor the asymmetryof the windowwallseem to fallinto place. We must firstconsiderthe buildingin its entirety.Unlike other structuresof this and earlierperiods,which were relativelysimple, Mnesiklesboldlyplanneda monumental, multi-unitgatehouse composed of five distinct elements embodyingthree scales.The designhad to be workedout carefullyto createhardifferentarchitectural the various between mony parts.The subordinateWest Wings and East Halls of the CentralBuilding.It will be shown complexwere plannedto emphasizethe predominant how, after startingwith an initial simplifiedand symmetricalsketch, Mnesikleswas slowlyforcedinto his final scheme by the interrelatedmodularunits and proportions whichhe used, and it will be shownthat the developmentof his architectural design was evolvedcompletelyon the "drawingboard",not only for the planbut also for the elevations,since he had a keen three-dimensional graspof his project.In the planning had work to back and forth between stages,Mnesikles planand elevationsas he correlated the designof his friezesto the designof the plan, and vice versa.The planutilizedthe entirewidthof the Akropolisfromthe northernto the southernwalls,thereby effectuallyenclosingthe site exceptfor controlledpassagethroughthe gates.16 By workingin reversesequencefrom the presentdimensionsof the complex,and whichwere madefor the unfinishedcomthe preparations by takinginto consideration ponents,i.e. the East Halls, Dinsmoorwas able to reconstructwhat must have been sketchfor the project,a symmetricalbuildingwith its greatMnesikles'firstpreliminary est dimensions208 D.F. north-southand 96 D.F. east-west(Fig. 1).17 In this very first 16Therehas been considerablecontroversyas to how much "drawing-board"planningwas done by a ClassicalGreek architect(literallychiefmason). See, for instance, J. A. Bundgaard(footnote 9 above) and J. J. Coulton, GreekArchitectsat Work,London 1977. This article has been written, in part, to dispel a common belief that the architect'srole was considerablyless than it is today. 17Doricfeet, sometimes referredto as Attic feet, varied fractionallyfrom buildingto building, with a length of 0.32723 m. in the Propylaia.See now W. Koenigs, "Zum EntwurfdorischerHallen," IstMitt29,

THELASTTIME? THEASYMMETRY OF THEPINAKOTHEKE-FOR

23

in the sequence of schemes, as the planningevolved in the "drawing-board"stage, the various chambers of the Propylaiawere thought of in whole foot units, the rectangle enclosing the colonnades of the Central Building being 64 x 74 D.F., the East Halls being 44 x 72 D.F., as measured from the enclosing rectangle of the Central Building, and the West Wings 37 x 48 D.F. The Halls were to form a symmetricalcross with the CentralBuilding, the north-southaxes all aligning (Fig. 1). The modular size of the wall blocks throughout, except for the block lengths of the west and north walls of the Pinakotheke (3 53/64 D.F.) and of the gate wall of the CentralBuilding (4 D.F.), was determined by the unfinishedNortheast Hall. All except two of the blocks of the top course of the Hall's west wall are extant and the exact original location of each one is known. These blocks are slotted through from top to bottom to receive the ends of the never-erected roof rafters. At each fifth slot, set into the top of the epikranitiscourse below this uppermostcourse, there is a special rectangular anchorage cutting for the rafter which would have aligned with an interior column, and these are 18 D.F. on center. The five wall blocks which compose each of these intervals are therefore 3 3/5 D.F. (1.178 m.) in length, which became the basic unit for the building.18The center of the first, southern, cutting occurs 14 D.F. from the rectangle enclosing the colonnades of the Central Building, and since, as at the West Wings, the Northeast Hall would have had a hipped roof, in this case at the north, with hip rafters at 450 to the walls as seen in plan, the length of the Hall can readily be determined as having been planned with a length of 14 + 18 + 18 + 22 72 D.F., fitting snugly within the limit of the Akropoliswall. Mnesikles then discovered that his initiallyplannedwidth of 44 D.F. for the Halls needed to be increasedto 44 7/8 D.F. in order to allow his modular frieze units of 3 3/5 D.F. to work properly,with 12 metopes and 13 triglyphs.19It was therefore from these unfinished walls of the East Halls that the width, as well, of the wall blocks of the Propylaiawas automaticallydetermined, in that at the never-built north wall 11 normal blocks of 3 3/5 D.F. left 5 1/4 D.F. surplus, which had to be divided evenly at either end for the corner returns of the 1979, pp. 209-237, for a listing of buildingswith Doric feet of 0.325 to 0.32765 m. in length. The foot was divided into 16 dactyls, and dimensions were possible to as small an amount as 1/192 D.F., which equals 1/12 dactyl, or 1.7 mm., or more than 1/16 inch. Today, with modularwooden studs, brick, blocks, tiles, windows, doors, etc., dimensions as small as 1/32 inch (or 1/409 D.F.) occur on architecturaldrawings. "8Normallysuch intervals would be divided into four parts, to provide a canonicalexterior frieze of four triglyphsand metopes per interior inter-axialspacing.This would have produceda module of 4 1/2 D.F. here, resultingin a frieze about 82%as high as that of the CentralBuilding.In order to create better harmony between the Halls and the predominantCentral Building, the somewhat cumbersome unit of 3 3/5 D.F. was employed instead, providinga more modest frieze only two thirdsas high as that of the Central Building.This division of the interiorcolumn spacingsinto five parts, rather than four, precludedthe possibilityof an exterior colonnade beneath the frieze, since alternatecolumns would have occurred beneath a metope rather than a triglyph,and shows that a solid wall was planned instead (see W. B. Dinsmoor, op. cit. [footnote 7 above], p. 204 and fig. 76). 19Theincrease in width of the Halls by 7/16 D.F. on either side also diminished the dimension from their east wall to the outer face of the columns of the East Hexastyle of the CentralBuildingfrom 15 to 14 9/16 D.F., which equates to 4.765 m. vs. 4.757 m. as built.

24

JR. B. DINSMOOR, WILLIAM

frontand backwalls,i.e. 2 5/8 D.F. (0.859in.) each.The heightof the blocks,the only fixedat 1 1/2 D.F. (0.491m.). dimensionnow lacking,was arbitrarily of the variouspartsof the overAs more studywas given to the inter-relationship all plan,minormodificationsto the initialsketchwere necessitated,as invariablyhapdesign.The columnsof the West Wingswere designedto pens in modulararchitectural be two thirdsthe heightof those of the CentralBuildingof 26 11/12 D.F. (8.808in.), or 17 17/18 D.F. (5.872 in.). This was slightly modified to 17 7/8 D.F. (5.849 in.) so

that the soffitof the epistylewouldalignwith a walljoint. Fromthis heightof column for this size Mnesiklesarrivedat his columnspacingsby usingthe acceptedproportions of columnof 3:7, or 7 37/56 D.F. (2.507 in.), modifiedfor ease of constructionto 7 21/32 D.F. (2.5055in.). Then, as in the CentralBuilding,the lowercolumndiameter was determinedfrom the columnspacings,againusingthe proportionof 3:7, as 3 9/32 D.F. (1.074in.). The columnspacingsdeterminedthe triglyphspacingof the frieze,one half of 7 21/32 D.F. or an unwieldy3 53/64 D.F. (1.253 in.); this measurementalso becamethe new modulefor the wall blocksof the west and northwallsof the Pinakotheke, since these blocksnecessarilyhad to breakjoints uniformlywith the blocksof the friezewhichresteddirectlyuponthem. Of this 3 53/64 D.F., 1 1/2 D.F. (0.491in.) was allottedto the triglyphsandthe remaining2 21/64 D.F. (0.762in.) to the metopes. On the long west wallof the Pinakothekethese lengths,whichas we have seen evolved of the columnheightto thatof the CentralBuilding, ultimatelyfromthe 2:3 proportion fitted admirablywith the originallyproposedtotal length of 48 D.F. at floor level. At the frieze level, 13 triglyphsand 12 metopes,with even spacings,equal47 7/16 D.F. The remaining9/32 D.F. (0.092 in.) at eitherend was utilizedto enlargethe two end metopesnearesteachcornerby one dactyl(0.020m.) each and to providefor the slope of the wallsandthe projectionof the antae. Along the front colonnadeof the Pinakotheke,however,a slight adjustmentwas necessitated.In the initialsymmetricalsketch, the rectangleenclosingthe columnsof the CentralBuilding,74 D.F. deep, projectedequally15 D.F. beyondboth the northsouth wallsof the East Halls,and the west wall of the West Wingsprojectedwestward another22 D.F., giving a total widthto the Wings of 37 D.F. (Fig. 1:A). Mnesikles now discovered,still in the planningstages,that if he used his initiallyplanneddimension of 22 D.F., with his frieze and columnspacings,and also his columndiameters, rigidlyfixed, the west edge of his thirdcolumnfrom the west wouldfall 3/8 D.F. behind (i.e. to the east of) the west edge of the columnsof the west colonnadeof the CentralBuilding.This was intolerable,and so he had to increasethe westwardprojection of the Wingsto 22 3/8 D.F. to make the west edges of all the columnsalign. I have alreadymentionedthat the East Halls were increasedin width by 7/8 D.F. of whichhalf, or 7/16 D.F., was to be gainedby encroachinginto the area of the West Wings.As a result,his even figuresof 22 + 15 = 37 D.F. for the widthof the Pinakotheke necessarilybecame22 3/8 + 14 9/16 = 36 15/16 D.F. (Fig. 1:B).The designof the Wingnow seemedto be complete,exceptfor the windowwallitself. The roughopeningof the doorwaywas made interjacentin size betweenthose of the smallestand those of the intermediatedoorwaysof the gate wall of the Central

THEASYMMETRY OF THEPINAKOTHEKE-FOR THELASTTIME?

ATA0,6 SLOPE; 1

_

642

1642

i

_

_

20'

.

tr

7 12 No22

W

ofthe

frieze,

i2e.

2

anditanowesow

salmewt exasct

21/64

i

'

'

s2 DF,

5-

1

9 9i

while

the

triglyphsof th fiz

t

deind(3

small

tha the were deige

SLOPE

<

13t

64i.

1

6 12 265 2 _ ~~~~~~~~~~~~~5

1

W 1*~~~rWA

D

l7

3

~ -DJR4 ~ ~~~~~~~w~

-~

FIG.3.Windowwal as orgnal

s e

.

35

as 12DOR

51

5

-:

.

[![=]~~~~~~~~~5E

,

64

4-

6

-

6425

25

;

FEET 12C

8

oiue1

I antae

96

lik enframing

t the

tobe th sam wit

i.e56. 1 1/ DIg, Ic

windows

are

the

as th metopes

ftherotr

be shown

that the windowsand their antae aligned,in straightelevation,with the metopesand This last observation,concerningthe locationof the windows, triglyphs,respectively.20 201n1962/63, with an 8-meter ladder,I thoroughlymeasuredthe window wall, block by block, with no thought of correlationto the frieze. The course above the lintels is the only course that is accuratelymeasurable on the south face of the now sadly distorted wall. Its original length was 10.710 m., exclusive of open joints which now add another 0.141 m. The bottom of the lintel course on the north side is also accuratelymeasurable.It was originally10.711 m. long, while open joints now add another 0.128 m. The east-west dimensions between the bottom of the orthostates, in the line of the wall plane above, are 10.703 + 0.020 (projectionof orthostates) = 10.723 m. at the colonnade, and 10.706 + 0.020 = 10.726 m.

26

JR. B. DINSMOOR, WILLIAM

is the continuationof Hellstr6m'sincipientidea in connectionwith the placingof the inner-lyingantaeof the windows,which,he says, "indicatesthat the lay-outis actually contentionthat Mnesicproduct"and "tendsto disproveBundgaard's a drawing-board les workedwithoutusingaccuratedrawings.""2 As persuasivelyarguedby Dinsmoor, before the symmetricaldesign was abandonedthe over-allwidthof the West Wingswas plannedto be 1 1/2 D.F. greaterthan it now is, or 36 15/16 D.F. fromthe finishedwallplaneof the east wallto the west side of the westernanta. Of this, 1/32 D.F. (0.010 m.) was used by the projectionof the antabeyondthe west wall, and 2 5/8 D.F. (0.859m.) was used by the thicknessof the west wall. This left 34 9/32 D.F. for the exposedlength of the windowwall. For the windowsto have centeredexactlyunderthe thirdand eighthmetopesof the friezeover the colonnadeas modularlydesigned,workingfrom the west, the distanceto their center lines from the inner face of the west wall, at floor level, shouldhave been 7 15/16 D.F. (2.597 m.) and 27 5/64 D.F. (8.861 m.), leaving 7 13/64 D.F. (2.357 m.) for

the remainingdistanceto the east wall (Fig. 3).22 At sill level, the windowswere certainly planned to have widths equal to the metopes above, 2 21/64 D.F. (0.762 m.).23

With this proposedscheme, with one more triglyphat the east than was actuallyconstructed,an additional,or fourthcolumncould have been erectednear to where the easternanta now stands,and the windowswould have been utterlysymmetricalwith the colonnade(Fig. 3).24 Betweenthese windowswouldhave been centeredthe doorat the northwall; these dimensionsallow 10.704 + 0.020 = 10.724 m. at the windowwall, which is unmeasurableat the bottom. Since the east wall has an eastwardslope from its base to the course above the lintels of 0.042 m. (0.053 m. to the top of the wall), the west wall originallyhad a slope, also eastward,of 0.056 m. in this same height (0.069 m. to the top, althoughfurthersouth, in the line of the colonnade, this slope was only 0.050 in., a discrepancywhich means that the west wall must have been warpedslightly; see Fig. 5). The length of each course was now calculable.The doorway, although now tilted westward, has rigidly maintainedits originaltop and bottom widths, except for cracksin the sill; these were 2.290 m. and 2.338 m. (R. Bohn's 2.292 m. and 2.355 in.). Since I had the lengths of the jamb blocks, the size and exact location of the windowsas finallybuilt were accuratelydeterminable.On the south face of the wall the present dimensions, becauseof distortions,are 0.824 m. bottom and 0.830 m. top for the westernwindowand 0.820 m. bottom and 0.819 m. top for the eastern.These similarmeasurementsled Bohn, and Dinsmoor, to think that the windowsoriginallyhad verticaljambs, with the same widths top and bottom. In actuality,however, they were originallytapered,like the doorway,with dimensionsof 0.768 m. bottom and 0.737 m. top for the western and 0.812 m. bottom and 0.790 m. top for the eastern. The center lines of these windowsfell westwardof the center lines of the thirdand eighth metopes from the west, as finallybuilt, by only 0.018 m. and 0.0175 in., and even this slight discrepancywas an error on the partof the workmen (see footnote 29 below). It should be noted that Dinsmoor eventuallyacceptedmy findingsand restoration. 2"Hellstrbm,op. cit. (footnote 12 above), p. 91. 22Actuallybuilt as 2.590 m. and 8.885 in., with 1.839 + 0.491 m. (1 1/2 D.F.) = 2.330 m. left over. 230.768m. and 0.812 m. as built (see footnote 26 below for the increase in width of the eastern window). 24Thecenter of the western column, in order to have aligned with the center of the third triglyphof the modular frieze, would have been 5 41/48 D.F. (1.916 m.) from the east face of the west anta, the width of which was constructedas 2 5/6 D.F. (0.927 m.). The center of the fourth column, at the east, would have been slightly closer than this to the face of the east wall, 5 19/64 D.F. distant, from which would have to be subtracteda small amount for a terminatinganta, probably7/48 D.F. to match the inner returnof the western anta (Fig. 3).

THE ASYMMETRYOF THE PINAKOTHEKE-FOR THE LAST TIME?

__35

10v 17

SLOPE~~~~~j k SLOPE

7 16 19

;I

27

22821

49 j

264

2

1~225 I2 64

225 64

3 64 5 iL

2

SLOPE

140y1 I

16

ANTA 0

w:21

I~~~~~~~~~~~~~~~V

i

A

10

20 D.,JR-

3 W

-1980 E39

FIG. 4. Intermediatescheme for window wall after shortening (C on Figure 1)

way, whichwouldhave been centeredalso on the centralintercolumniation and on the corresponding triglyphof the friezeabove. Now, however, Mnesiklesmade one more change in his design of the Pinakotheke, thatof shorteningthe widthby 1 1/2 D.F. He did this by compressingthe entire west facade of the Propylaia,includingall the columns, eastwardtowardthe East Halls.25He thereby decreased the width of the passagewaysto the West Wings between

the end columnsand the greatantaeof the CentralBuilding,and also lost the fourth 25Theproof that he made this compressionlies in the returns of the entablatureon the flanks of the Central Building.At the east end, which was built as originallyplanned, the center of the third triglyph from the corner, on the flanks, lies centered over the great anta below, as it should. At the west end, however, the center of the correspondingthird triglyphlies most unhappily1 1/2 D.F. too far to the east of the center of the great anta below, since the entablatureand its returns had to move eastwardalong with the hexastyle colonnade below it, but the anta itself did not move from its originallyplannedposition.

28

WILLIAMB. DINSMOOR,JR. P50 SLOPE

94

,

782

490

783

490

759

492

764

490

11,586763 490

760

491

782

492

779

491

994

$<

u

s

*,0~~~~~,O15 1

+,0105

2,506

2,5065 ,069 SLOPE_,1

1

,135

j

_

053 SLOPE

1.234-

-2

2,549

MSIG l

(1,168) 1

1,190

1,189

1,191

1,185

1,220

1,177

1,182

1,178

1,177

1

I~~~~~~3

]2,184

737

76|,?

4,818

-

001'q 860

2,590 3,450

ANTA 0

1

4

> 2

,5

1,190

1

1,192

,8

,

,623

2,290

4 DF

1,55

0

1

735

1,3314

1,157

1

1,246

1,177

f

1,2 39

790!,904

|

812

|

57

11,448

2,338

-

6,295 10,724 ATANTA 10,723+,86+,01=11,593 5

1,839

II

M. WB.D.,JR.- 1980

FIG.5. Windowwall as built

columnof the colonnadesof the West Wings,whichhe replacedwith the now extant antae and anta walls (Fig. 1). Otherwisethere was no change either of his original relationshipof the columnsand of the door and windowsof the Pinakotheketo the frieze,or of the columnsof the Wingsto those of the CentralBuilding(Figs. 1[C], 4). The originallyplannedwiderspacingbetweencolumnsand greatantaeat the east end of the CentralBuildingremainedunaffected.One reasonfor this changewas that the triglyphfriezeof the WestWingsdid not terminateproperlyat the dividingwall,ending in a partialmetope (Fig. 3); with the reduction,the frieze could end with a more canonical1/2 triglyph,as at a re-entrantcorner(Fig. 4), and, indeed,this arrangement was employedat the matchingfrieze of the northwall. Overthe colonnade,however, an elongatedmetope was used instead (Fig. 5), reminiscentof the similartreatment whichmust have prevailedat the facadesof the predecessorto the Propylaia,the old Anothercontributoryfactorfor the reductionof 1 1/2 D.F. might have Propylon.26 26SeeW. B. Dinsmoor, Jr., op. cit. (footnote 10 above), p. 52 and p1. 15:B'.The elongated metope of the West Wings carriesupwardthe width of the anta wall below, thereby visually terminatingthe triglyph

THEASYMMETRY THELASTTIME? OFTHEPINAKOTHEKE-FOR

0

2

1

A

L

=

}

29

M.

W.B.D.,JR.-1980 FIG. 6. Entablature on westside of SouthwestWing

been relatedto his problemswith the SouthwestWing, the facadeof which had to balancethat of the Pinakotheke.As actuallybuilt, the constructionof the entablature on the west side of this stunted adjunctdefies credence.The epistyle consists of a single, narrowbeam, centeredover the westerncolumnof the frontcolonnade,rather than being composedof two canonicalbeams.The frieze above it, however, is composed of two beams, and the inner one, the antithema,projectsbeyond the narrow epistylealmostin its entirety,since continuoussupportfor it is lacking(Fig. 6). The only reasonfor such a RubeGoldbergconstructionmust have been thatMnesikleswas prohibitedfromplacingthis west side of his stuntedWingfurtherwest, into the Athena Nike precinct.In orderto supportthe northend of the epistylehe needed a column, frieze at the easternmost triglyph.Because of this, Mnesikles now quite naturallyenlarged the last two normal metopes at the east to the same size as the last two metopes at the west, i.e. to 2 25/64 D.F. instead of 2 21/64 D.F. With the change of the eastern metope sizes, and in order to maintainhis alignment of centers of windows to centers of metopes, he increased the distance between centers of windows from 19 9/64 D.F. to 19 15/64 D.F., i.e., to 6.294 m. (6.295 m. as actually built). The alterationin turn decreased the space from the center of the eastern window to the east wall by the same 3/32 D.F. to 5 39/64 D.F. (7 13/64 - 1 1/2 - 3/32). He now also increased the width of the eastern window to match the new enlargedmetope above (Fig. 4). The increasein width of the eastern metopes caused the first minor asymmetry in the window wall, in that the distances from the center of the metopes over the windows to the center of the triglyphover the door were no longer exactly equal, being off-centered on the triglyphby 3/64 D.F. (Fig. 4).

30

WILLIAM B. DINSMOOR, JR.

andhe keptthis columnas far to the west as he possiblycould;a slightlymoreeasterly locationwould have allowedhim to employa normal,double-beamedepistyle.If his westerncolumn, and the attendantdouble anta which balancesthe west wall of the Pinakotheke,hadbeen moved 1 1/2 D.F. furtherwest, the entablatureon the west side of the structurewouldhave had no supportat all, and at the symmetricalPinakotheke he wouldhave been unableto re-use the alreadyextant coursesat the bottomof the fan-shapedfoundationfor the supportof its west wall. With the westernlimit of his Wingsnow fixed, he might have moved the CentralBuilding(exceptfor its western colonnade)and the East Halls (includingtheir westerndividingwalls) 1 1/2 D.F. furthereast, but he must have been reluctantto do this since the relationshipof the position and orientationof the limitingMycenaeanwall to the dividingwall betweenthe SouthwestWingand SoutheastHallwouldhave forcedhim to reducethe depthof his alreadydrasticallyreducedSouthwestWing by approximately another2 1/3 D.F., or 0.76 m. (Fig. 1). He thereforemadethe best of an unpleasantsituation. Mnesiklesprobablythoughtthat his majorproblemshad now been solved and he startedactualconstruction.We knowthat workbeganat the southernend of the west facadeof the CentralBuildingsince this stretchof foundation,runningnorth to the centralpassageway,has two euthynterias,one placedon top of the other, while the northernstretchof this same foundation,north of the passageway,is composedof normalalternatingheader-and-stretcher coursesfor its entire height. As construction progressedfurthernorth,and the foundationof the dividingwall betweenthe Pinakotheke and NortheastHall was started,there was a periodof indecision.Exceptfor the firstsketchwhichhe must have made, in whichhe wishfullyentertaineda completely balanceddesign, Mnesiklessurelyrealizedfrom the outset that in plan he would be unableto buildhis SouthwestWingas a mirror-image of the Pinakotheke,thatit would have to be stuntedand mademore shallow.His distressover this is evidentin that he stoppedconstructionof the foundationof the east wall of the Pinakothekeat a point just beyondwhat was to be the correspondinglimit for the depth of the Southwest Wing (Fig. 2). Possiblyhe thoughtmomentarilyof stuntingthe Pinakothekeas well, to balancecompletelywith the revisedSouthwestWing.Up to this pointthe foundationis composedof coursed, re-used wall blocks, alternatingheaders and stretchersboth verticallyand horizontally.Then, after this interruption,he continuedthe foundation northto the Akropoliswall, evidentlywith the NortheastHall uppermostin his mind, Hall whichhe must have cherishedsince he had workedon it so the never-completed in the diligently planningstagesand had derivedhis size of wallblocksfor the building from its dimensions.This continuationis composedof a mishmashof wall blocks, geison blocks,blockswith socketscut for beams, etc. The coursesdo not aligneither with the earliersouthernsection or within the new individualcourses themselves.27 27Wall blocks and a lintel block from apsidalBuildingB had alreadybeen used in the foundationunder the West Hexastyleof the CentralBuilding.Now the uppermembers of the entablatureof BuildingB were used in this later northernextension of the dividing wall, in reverse sequence to that of demolition. See W. B. Dinsmoor, Jr., op. cit. (footnote 10 above), p. 2, note 10, for my objection to Building B having

THELASTTIME? OFTHEPINAKOTHEKE-FOR THEASYMMETRY

31

Then, eitherat this time or slightlylater,workon the Pinakothekeproceededand the east-westfoundationswere added; the foundationsunder the windowwall and the north wall, composedagain of coursed,re-used wall blocks, alternatingheadersand stretcherslike the earliersectionof foundationof the dividingwall,do not bondto, nor courseout with, the abuttingnorth-southfoundation,exceptat the euthynteria(Fig.2). At some time afterhe reducedhis Wingsby 1 1/2 D.F., Mnesiklesreturnedto the "drawingboard"and discovered,of course, that the changeproducedproblems,as all changesdo. The distancefromthe west wallto the centerof the west windowremained unchangedat 7 15/16D.F., but thatfromthe east wallto the centerof the east window was reducedfroman original7 13/64 D.F. to a much more lopsided5 39/64 D.F. (see footnote26 above). Regardless,the door could have remainedcenteredboth between in frontof it exceptfor one thing, the jointthe windowsand on the intercolumniation ing of the blocks (Fig. 4). In Classicalarchitecture,care was normallyexercisedwith joints, to terminatethe ends of each wallsof ashlarmasonry,with regularlyalternating coursewith some semblanceof symmetry,eitherwith blocksof equallengthor with a full blockat one end and a half blockat the other.Furthermore,againfor aesthetics,a balancedrelationshipof verticalwalljoints to openings,such as doorways,was desirable, and the ends of lintel blocksshouldfit into the patternof joints in a symmetrical manner.In the originalscheme,startingfromthe centerof the doorwayand proceeding of joints with blocksof 3 3/5 D.F. in eitherdirection,a fairlysymmetricalarrangement wouldhave been attained(Fig. 3). Butwith the shorteningof the wall,chaosdeveloped in regardto the lengthsof the end blocks, if the originaljointingsystem were to be maintained(Fig. 4). As a result,Mnesikleschangedhis jointingpatternand made the blocksat eitherend of each courseas equalin lengthas he could, althoughperfection was not possiblesince the side wallsof the Pinakothekeboth slopedin the same direction, to the east, creatinga trapezoidalelevationfor the windowwall. In orderto fit into the shiftedjointingsystem, the lintelfor the door, and thereforethe symmetrically dooritself, hadto be shiftedas well, althoughthe positionof the windows,keyedto the metopesof the frieze, was consideredto be immutable.He then had a choice of moving his door eitherto the east or to the west. He chose to move it to the east (by half of the decreasein lengthof the wall, or 3/4 D.F.) for betteraccess from the Central Building.28 By this time the less importantwindowsno longerfittedinto the schemeof jointingsincehe obstinatelymaintainedtheirrelationshipto the frieze,but he obviously consideredthis asymmetryof the windowsthe lesser of the evils (Fig. 5).29 One wonoccupied the site of the Pinakotheke.To this could be added the observation that had Building B stood here, the materialfrom the superstructure,in order to be utilized after the blocks from the walls below were re-used, would have had to be stockpiled at first on the Akropolis slope of 18 1/20 in front of the Pinakotheke. 28Asa result, the center line of the door, instead of falling under the center of the sixth triglyphfrom the west, as originallyplanned, was moved to lie under the eastern edge of that triglyph.As constructed, there was a discrepancyof 0.001 m. (Fig. 5). 291nFigure 4 the frieze is shown as it was redesigned by Mnesikles, still aligning perfectly with the columns. The columns and the east wall were placed accurately,and the eastern end of the frieze, above

32

WILLIAM B. DINSMOOR, JR.

dershow Mnesikles'choiceof priorityof the jointingsystemover a symmetricalsystem of openingswas regardedin his time. One also wondersif the thoughtever crossedhis mindto abandonthe relationshipof openingsto frieze,whichno longerworkedfor the door anyway,and to returnto a balancedsystem.He could have maintainedhis final jointingsystemup to the lintelcourse.If, above this, he had changedthe orderof the joints in the top threecourses,if he had moved his door 1 D.F. to the west insteadof 3/4 D.F. to the east, and if he had shiftedboth his windowsto the west, he couldhave createda totallysymmetricalscheme,not only of all the openingsto each otherbut also of theirlocationin the over-alllengthof wall (Fig. 7). TABULATEDSUMMARY 1. A first preliminarysketch of the Propylaiawas made, using whole Doric Foot (D.F.) dimensions (Fig. 1).

2. The lengthof the wall blocks was determinedduringthe design of the NortheastHall (specialanchorage cuttingsfor every fifth rafter) as 1/5 of 18 D.F. = 3 3/5 D.F. 3. The width of the Northeast Hall was increasedfrom the preliminary44 to 44 7/8 D.F. to accommodate a frieze of 12 metopes and 13 triglyphswith the determined modular unit of 3 3/5 D.F. The width of the Pinakothekewas decreasedby 7/16 D.F. because of the wideningof the NortheastHall. 4. The widthof the wall blocks was determinedby the length of the north wall of the Northeast Hall: i.e., 44 7/8 D.F. - 11 wall blocks of 3 3/5 D.F. = 5 1/4 D.F. . 2 = 2 5/8 D.F. at each corner. 5. The width of the Pinakothekewas increasedby 3/8 D.F. by moving its west wall and its columns westward so that the west face of its third column from the west would align with the west face of the columns of the CentralBuilding.The plannedwidth now became 37 - 7/16 + 3/8 = 36 15/16 D.F. (Fig. 1). 6. The windows and the window-antaeof the Pinakothekewere made to align with, and to be the same widths as, respectively, the metopes and triglyphsof its frieze; the door was to be centered. A fourth column was planned.The openings in the window wall would have balancedthe colonnade exactly (Fig. 3). the anta wall, should have been 3 5/64 D.F. (1.007 m.), which is exactly the dimension required for a normalmetope of 2 21/64 D.F. plus a half-triglyphof 3/4 D.F. (Fig. 4). During construction,however, the elongated metope which was used here instead was shortened by 0.013 m. to 0.994 m. (Fig. 5). This was an erroron the partof the workmen.With only minuscule variations,the remainderof the frieze was built as designed. Because of the initial mistake, the center lines of the windows actuallyfell 0.0175 and 0.018 m. too far to the west of the centers of the pertinentmetopes, and the centers of the Doric columns fell from 0.0105 to 0.015 m. too far to the west of the centers of the pertinenttriglyphsabove them (Fig. 5). For the variationsbetween plannedand actuallyconstructeddimensions (Figs. 4 and 5), see the following table: Equivalent Difference D.F. as Actual in meters in meters planned in meters 0.010 11.596 35 7/16 11.586 Frieze 0 0.010 1/32 0.010 WindowWall 0.001 2 5/8 0.859 0.860 (see the dimensions 0.007 7 15/16 2.597 2.590 at the bottom of 0.001 6.294 19 15/64 6.295 Figures4 and 5, in 0.003 5 39/64 1.836 1.839 consecutive order). 0.006 10 9/16 3.456 3.450 0.003 10.727 32 25/32 10.724 0.003 11.596 35 7/16 11.593

THEASYMMETRY OFTHEPINAKOTHEKE-FOR THELASTTIME?

0|

I

I5I

10

33

I

M. W.B.D.,JR.-1980

FIG.7. Windowwall as it could have been symmetricallybuilt 7. The width of the Pinakothekeand the depth of the western end of the CentralBuildingwere reduced by 1 1/2 D.F. by moving the entire facade of the Propylaia,includingthe columns of the Pinakotheke, eastward (the eastern end of the Central Building remained unchanged. Fig. 1). The planned fourth column of the Pinakothekewas now lost. The reasons for the change were 1. the north and south friezes of the Pinakothekedid not work out properlyin the greater length (Fig. 3). .2. Mnesikles placed his West Wings as far to the west as possible, but was limited (a) at the Pinakotheke [by his desire to utilize the pre-existing lower courses of the fan-shapedfoundation for his west wall], and (b) at the SouthwestWing [in that the western limit of his contractedwest side, and therefore the western limit of the emaciated epistyle which he employed on it, were surely fixed by mandate, and from the epistyle was determined the position of its northern support, the western column of the colonnade, thus preventing a furtherwestern location of the colonnade (Fig. 7); if, on the other hand, he had maintained the extra 1 1/2 D.F. by expandingthe West Wings towardsthe east, he would have lost 2 1/3 D.F. in the alreadyshallow depth of the SouthwestWing because of the angle of contact which the Wing made with the limiting Mycenaeanwall (Fig. 1)]. 8. Because of the reduction in width of the Pinakotheke by 1 1/2 D.F., Mnesikles shifted the vertical jointing system of the window wall to the west in order to provide a symmetricalarrangementof blocks of equal length at the ends of each course (Figs. 4 and 5). To fit into this shifted pattern of vertical joints, the door and its lintel were moved 3/4 D.F. to the east, thus destroyingthe symmetricalarrangement of the door to the frieze and to the windows, since the latter rigidly preserved their originally plannedrelationshipto the frieze. WILLIAMB. DINSMOOR,JR. AMERICAN SCHOOLOF CLASSICALSTUDIES AT ATHENS

THE DATE OF THE PHAIDROSBEMA IN THE THEATEROF DIONYSOS (PLATES

2 and3)

ONEOF THE MORECONSPICUOUSof the lesser antiquitiesof Athens is the long sculpturedfriezethatadornedthe stagefrontof the Theaterof Dionysos,its Dionysiacscenes punctuatedby crouchingsilenoi (P1.2:a, b).1 Only the westernhalf remains.As now visibleit representsthe last phaseof a long seriesof remodelings.For this periodthe frieze, almostcertainlymadefor the previous,Hadrianic,stage,was cut On the top step of the littleflightof stairsleadingfrom downto fit the new proportions. the orchestraup to the middleof the stage was carveda metricaldedicationby the archonPhaidros: (TOt T8E

'Fax8po'

ETEV~E, qLoXpyLE,

f3r'Aa

9ErqIpOV

Zco~Xov (P1.3:a,b) f{3t~o&Xopo' 'Amig8os&pXd'2

For you, loverof the sacredrites, this beautifulstagehas been built by Phaidros,son of Zoilos,archonof life-givingAthens. When the stage was first discoveredin the 1860's the excavatorsdated it tentativelyaboutA.D.300, seeingthe persecutionsof Diocletianas an opportunemomentfor the erectionof such a blatantlypaganmonument.3In 1910W. B. Dinsmoorestablished a relativechronologyfor (1) the destructionof the Monumentof Nikiasandthe Stoaof 'Homer Thompson's preeminencein mattersof classicalarchaeologyis universallyacknowledged.Less widely acclaimedis his tolerance in accordingequal rights to the classicaland post-classical,which has set an example extending far beyond the limits of the Athenian Agora. On behalf of the many scholarsof late antiquitywho have reapedthe profitsof his example this articleis gratefullyoffered. For a recent study of the frieze, including a convincing argument for its immediate source, cf. M. Sturgeon, AJA 81, 1977,pp. 31-53. 2IG 112,5021. Phaidros,son of Zoilos, is otherwise unknown except as the donor of a sundial now in the BritishMuseum (IG I12,5208). G. W. Bowersockhas kindly called my attention to the laboredarchaizingof these two clumsy hexameters:the long alphain Kako'vand the word "TEVfE, both common in Homer; the rare apxo' for apXwv, also Homeric, and the IonicizingOE71'pOV.Only one other example of 4X0opyLEis cited by Liddelland Scott and that in connection with Aphrodite,not Dionysos, in an epigramof Philodemos (1st centuryB.C.), Pal. Anth. x.21. Cf. A. S. F. Gow and D. L. Page, The GreekAnthology:The GarlandQfPhilipII, Cambridge 1968, p. 385. 'AOL'8-osmore properlyrefers to Attica than to Athens and perhapsjustifies the word 3Lto&.Topoc (which appearsin the OrphicHymns). 3'ApX'E0,1862, pp. 129ff., 131ff., 164-165, 209ff. It has often been thought that the heads, all of which are missing, were deliberatelyhacked off by Christianrebuilders.M. Sturgeon, however, has shown that while the backgroundbehind them was cut down to obtain a lower height for the stage, the heads remained in place, but now reaching up into the cornice (AJA 81, 1977, p. 44). Phaidros' dedication should dispel any doubt. Followingcurrentcustom, the term "bema" is used here for Phaidros'contribution as distinctfrom the originalstage.

THEDATEOFTHEPHAIDROSBEMAIN THETHEATEROF DIONYSOS

35

Eumenes on the south side of the Acropolis, (2) the constructionof the Beule Gate in front of the Propylaia, (3) the reconstructionof the stage in the Theater of Dionysos, and (4) the erection of buttresses againstthe back wall (then rebuilt) of the Stoa of Eumenes, in that order.4Evidence for exact dating, however, was lacking, so he cautiously ascribedthe whole sequence to a very short time within "late Roman or even Byzantine times." Since Dinsmoor wrote, excavation has disclosed the full impact of the attack on Athens by the Heruli in A.D. 267. This one raid left a great part of the lower city in ruins, including the buildingson the south side of the Acropolis, thus giving an almost certain date for the first step in Dinsmoor's sequence. In 1914 P. Graindor, with Dinsmoor's relative chronology in mind, attempted to pin down preciselythe date of the Beule Gate and the Phaidrosbema by introducinga new element: the large inscribed block which lies on the slope of the Acropolis just below the Beule Gate5 (P1.3:c). The inscriptionrecordsthe gift of a gate (or gates) by a certain Fl. SeptimiusMarcellinus,evidently a high official (Xa/,7rptTaTos).

(Z Z10TTi/uOta MapKEXXJiVO19o EK WeV shiv

ka/i(rpd-aTo)

TOV' 7rvACOvaq

Kat a1Troa'ywvo0ETv 7 FrTOA

On account of its finding place and appropriatenessthe block has been associated with the gate ever since the discovery of the latter,7but their actual physical relationship and date have remainedenigmatic.In view of the importanceattachedby Graindor to the inscriptionand its bearing on the Beule Gate and hence the bema, Graindor's evidence must be reviewed here briefly. He dated the inscriptionin the 4th century on the basis of the abbreviationdA',8 and so, by association, the Phaidros bema. Within the broad limits of the 4th century the threat of invasion by Alaric in 395 seemed to Graindorthe most suitable occasion for the erection of the Beule Gate, so, to fit the Phaidros bema into the whole scheme, he placed the archonshipof the otherwise unknown Phaidroswithin the last five years of the 4th century, a period which was unoc4AJA 14, 1910, pp. 459-484, with all the complexitiesof the argument. 5IG 112,5206. P. Graindor, BCH 38, 1914, pp. 272-295, esp. pp. 286-293. The block is now in two pieces with the middle section missing, but it was copied several times, while still intact, by early travelers, beginningwith Cyriacusof Ancona (1436). Cf. E. Bodnar, Cyriacus of Anconaand Athens, CollectionLatomus XLIII, Brussels 1960, pp. 176-177 for the relevant referencesand variantreadings. 6The earlier readingof OfkAain the first line has now been generallysupersededby o atut. (Bodnar, op. cit., p. 177). 7Eustratiades,aVEK8. fvXXc'8. II, p. 4, taken up by C. Wachsmuth (Die Stadt Athen im AlterthumI, Leipzig 1874, p. 704, note 3). 8An even more compelling reason is the appearanceof the double gentilicium,Flavius Septimius, Flavius being bestowed as a designationof rank from the reign of Constantineand continuingthrough the 4th and 5th centuries (cf. J. G. Keenan, "The Names Flavius and Aurelius as Status Designationsin Later Roman Egypt," ZPE 11, 1973, pp. 33-63, and A. Mocsy, "Der Name Flavius als Rangbezeichnungin der Spdtantike,"Akte der IV. internationalen Kongressesfiir griechischeund lateinischeEpigraphik,Vienna, 17-22 September, 1962, pp. 257-262, for a convenient summaryof the subject.I am gratefulto J. F. Gilliam for these references.

36

ALISONFRANTZ

cupiedin his archonlist.YLaterscholarsfollowedhim, and this, with few exceptions, has stoodas the accepteddate.'0 The basis of the argument,i.e., the assumptionthat the Beule Gate was built in directresponseto the threatof invasionin 395, deservescloserexamination.The fear arousedby Alaric'sproximitywas only the culminationof a mountingpreoccupation with dangerin Greeceas a whole thathad been presentsince the 3rdcentury.As early as the reignof Constantinethe citiesweretakingmeasuresto repairtheirfortfications," and the dangerwas intensifiedafterthe Battleof Adrianoplein 378, when bandsof victoriousGoths roved freely throughthe BalkanPeninsulaand threatenedthe cities of Greece, many of whose fortificationshad been seriouslyweakenedby the disastrous earthquakeof 375.12The Marcellinusinscription,as we have seen, could date anywhere

from the reignof Constantineon. But still unansweredis its relationshipto the Beule Gate. Was it partof the originalstructureor introducedat the time of some laterrepair?Or was it totallyunrelated?That cannot be determinedwithoutan exhaustive studyof the whole monument.Until that has been made the investigationof the date of the inscription.1 of the Phaidrosbemashouldgo on independently The dedicatoryinscriptionis of little help in findinga precisedatefor the construcof the whole tion of the bema, whichwould presumablybe that of the rehabilitation new forms with criteria, are unreliable in late notoriously antiquity theater.Letterforms over the centuriesbut few becomingobsolete.The lettersthemselvesin accumulating the Phaidrosinscriptioncan be matchedover a long periodbut the generalappearance, with the crampedspacingand shallowand irregularletters,wouldbe most at home in the 4th centuryor later.14 In defaultof concreteevidencethe literaryand historicalaspectsof the question take on some significance.One of the most importantbuildingsin the civic life of the Atheniansin any period,and the one most frequentlymentionedin the 4th century afterChrist,was the assemblyplace,alwaysreferredto by Eunapiusas "the theater" des archontesathenienssous l'empire,Brussels 1922, p. 270, no. 187. 9P. Graindor,Chronologie 10E.g.,Kirchner, IG 112,5021; J. Travlos, 'ApX'E4,1953/54, p. 309 and PictorialDictionaryof Athens, London 1971, p. 538. M. Bieber, in Historyof the Greekand Roman Theatre,2nd ed., 1961, p. 215, for unspecified reasons dated the remodeling of the theater ca. 270, although in her earlier edition she gave a date ca. 300. "A proconsul"Phosphorios"was honored by the Megarianswith a statue for having fortifiedthe city (IG VII, 96), and the Argives erected a statue of the same man, probablyfor a similar reason (IG IV, 1608). A persuasivecase for the date (318-320) and the identificationof "Phosphorios"with Aurelius ValeriusSymmachushas been made by G. Polarain La Paroladel Passato 29, 1974, pp. 261ff. I owe this referenceto the kindnessof T. D. Barnes. Zeit (DissertationesPannonicae,ser. 1, no. vonAchaiain spdtromischer 12E. Groag, Die Reichsbeamten 14), Budapest1946, p. 55; Zosimus, iv.31.3; iv.34.3. "3Travlos(PictorialDictionary[footnote 10 above], p. 483) sees the gate as part of the fortificationsof the time of Valerian(253-260). If that is the case the inscriptioncannot relate to the originalconstruction. "4Thedifficultyis well illustratedby the contrast between the Marcellinusand Phaidros inscriptions, both of the same century and, if Graindor'sreasoning is correct, carved within a decade or so of each other. Except for the classicalform of the omega in Marcellinus,its lettering is very similar to that of the dedicationsto Constantius11(337-361) over the east and north gates at Aphrodisias(P1.3:d).

THEDATEOF THEPHAIDROSBEMAIN THETHEATEROF DIONYSOS (To

37

OEaTpov).Thereseems to be no doubtthat this is not just a loose term for "audi-

torium"but thatone of the actualtheatersis meant.'5The Odeionof Perikles,adjacent to the Theaterof Dionysos,the Odeionof Agrippa,in the middleof the Agora,and the Odeionof HerodesAtticushad all been eithercompletelydestroyedor made unusable in 267. The Theaterof Dionysoshadsufferedsome damageduringthe raidbut was not beyondrepair.Moreover,the theateris knownfromliteraryandepigraphical sourcesto a from have sometimesservedas meetingplace Hellenistictimes and was perhapsthe recognizedseat of the Assemblyby the 1st centuryB.C.16 It wouldhave been naturalto followthe precedent. Two events describedby Eunapius, datable to within a few years, provide some il-

luminatinglandmarks.AlthoughEunapiusarrivedin Athens only in 362 and the relevant occurrencestook placebetweenabout340 and 346, he hada detailedaccountfrom a reliableeyewitness,Tuscianus,a pupilof the sophistJulian.On the death of Julian, ca. 340, a fiercestrugglebrokeout for the successionto his chair.As a result,Prohaeresius, the leadingcontender,was driveninto exile by the proconsul,who had been bribedby a jealousrival.With the permissionof the Emperor,Constans1,17Prohaeresius returnedto Athens where he was defended by a new proconsul.Prohaeresius himselfmade a triumphantdefense.He "lookedaroundthe theater,"and, as he continuedto speak,he had the satisfactionof seeing his enemies "tryingto slinkawayand lose themselvesin the crowdseated in the theater."At the end, "the proconsulwith his wholebodyguardescortedhim fromthe theater."'18 The impressiongiven in this accountis of a largebuilding,and this impressionis confirmedin the next episode. Shortlyafter Prohaeresius'returnfrom exile he was summonedto the Gallicprovincesby the Emperorandafterwardsent to Rome because he (Constans)"wasambitiousto showwhatgreatpeoplehe ruledover."Beforereturning to AthensProhaeresiuswas allowedto requesta gift of the Emperor:a numberof islandsfor the city. But the gift had to be confirmedby the Praetorian grain-producing Prefect,Anatolius,who had recentlyarrivedfrom Gaul.This was a momentousevent, describedby Eunapiusas "moreformidablethan the PersianExpedition.""Almostall the educatedmen of Greece"had come to Athens for the occasion,and "the theater wasfilled."AnatoliuswasPrefectof Italy,IllyricumandAfricain 345/6.19 The damageto the Theaterof Dionysosin 267 necessitatedfairlyextensiverepairs beforeit could be put backinto use. It is doubtfulthat these could be effectedin the '5A clear distinctionis made between "the theater"and the lawcourt,always8tKaoT-ptOV. '6K. Kourouniotes and H. A. Thompson, "The Pnyx in Athens," Hesperia 1, 1932, p. 138, with references. '7DuringConstans'reign Illyricumwas includedwith Italy and Africain a single prefectureand subject to the Emperorin the West. After his death it was transferredto the East as a separateprefecture.Cf. E. Groag, op. cit. (footnote 12 above), p. 30. '8Eunapius,v. soph. 488-492 (Loeb ed., pp. 489-511) for both passages. '9And again of Illyricumalone in 357 or 358 to 360. His visit to Greece must have been during his first prefecture.By the beginningof his second term of office ten or twelve years would have elapsed since the imperialgrantand Constanshad been dead for at least four years.

38

ALISONFRANTZ

still troubledremainingyears of the 3rd century.But all the evidenceshows that, althoughlittle or no majorconstructionwas undertakenfor a long time afterthe invasuch buildingsas were susceptibleof repair sion, a beginningwas madeat rehabilitating in the firstquarterof the 4th century.Providinga suitableassemblyplacewouldhave had a high priorityin such a program.Grantedthe premise,unprovenbut probable, thatthe Theaterof Dionysoshad alreadyresumedits role as the officialauditoriumfor highlyplacedspeakersand was the localefor the events describedabove, then a termiWith the generalcondition nus ante quem of 345/6 is establishedfor its reconstruction. post quem of about terminus a in 3rd century, the a date almost precluding of Athens 300 maybe safelyassumed.At whattime withinthose limitsit was carriedout remains uncertain. was new whenAnatoliusvisited Thereis nothingto indicatethatthe reconstruction Athens. Rather the contrary.In Philostratus'time sophists discoursedbefore large and the generalpracticeapparentlycontinuedin audiencesin the Odeionof Agrippa,20 the post-Herulianperiodeven though the Odeionand the other theatersexcept the Theaterof Dionysoswere beyondrepair.That some theaterwas availableis suggested in two passagesin Eunapius,one that "in accordancewith Romanlaw there had to be at Athens manyto lectureand manyto hear them";2'againin his descriptionof the houseof the sophistJulianwithits "theaterof polishedmarblemadeafterthe modelof a publictheaterbut smaller,of a size suitableto a house," followedby a referenceto the fact that on accountof the feuds betweencitizensand students"none of the sophists daredto go downinto the city and discoursein publicbut they confinedtheirutterancesto theirprivatelecturetheatersandtherediscoursedto theirstudents."22 Politicallyspeaking,improvementin Athens set in late in the reignof Diocletian, when the provinceof Achaiawas raisedin statusby being put undera corrector,who was presumablya senator.Constantinefurtherupgradedthe provinceby replacingthe of the Theaterof Dionysos On thatbasisthe reconstruction correctorwitha proconsul.23 The archaeological the first excavators. as about be as as 300, suggestedby early might evidence;whereverit is available,tends rathertowardthe reign of Constantinefor generalrecoveryin Athens,but any attemptto arriveat a close date betweenthe limits of ca. 300 and 345 couldonly be speculative.On the surface,Constantine'sreignseems more likelyto have provideda congenialatmospherefor such a paganmonumentthan those of his militantlyChristiansons, ConstantiusII and ConstansI. His interestin Athens"in wordand deed, throughouthis whole life" is illustratedby an annualdistributionof grainto the city andhis enthusiasticacceptanceof a statue"withan epigram" 20Philostratus,v. soph. 571 (Loeb ed., p. 192). 21Eunapius,487 (Loeb ed., p. 487). 22Eunapius,483 (Loeb ed., p. 468). 23A.H. M. Jones, TheLaterRomanEmpire,Norman, Oklahoma, 1966 reprint, p. 45 and note 12; pp. 106-107 and note 64. For epigraphicalauthorityfor the spelling Achala rather than Achaea, even when referringto the Roman province, cf. J. H. Oliver, "Achaia,Greece and Laconica," GRBS 21, 1980, p. 77, note 6.

THEDATEOF THEPHAIDROSBEMAIN THETHEATEROF DIONYSOS

39

erected by the Athenians in gratitude.24Restoration of the theater might conceivably have been among the benefactions of the proconsul Cervonius which earned the praise of Himerius,25or it might be attributedto some unknown benefactor. Further speculation would be unprofitable,and it is best to conclude on terrafirma with a terminusante quem of 345. ALISONFRANTZ Princeton,NJ 08540 24EmperorJulian, OrationesI.6. 25Himerius,OrationesXXXVIII (IV).

PLATE

2

xA_|t* lob-&

I-AL,

~

~

~

~

~

~

~

~

;

_~~~~~~~~~~~~~~~~~~~~~~~~~~~rLi

_ a. The theater from the Acropolis

.~~~~~~~~~~.

8

.

..

A,~4

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

PLATE

a. The center of the bema

A_~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~A

b. The dedicatoryinscription

c. The Marcellinusinscription

d. Inscriptionover North Gate, Aphrodisias

3

A CLASSICALMAIDEN FROMTHE ATHENIANAGORA (PLATES4-8)

HYPJHEN THE AMERICANEXCAVATIONSof the AthenianAgora began in 1931, HomerThompson,as one of the young AgoraFellows, took chargeof the sectiondirectlyto the northof the old excavationsby D6rpfeldon the west side of the Agora.On June 16, 1931, in a late Romandepositabove the southwestcornerof the Classicalbuildingwhichwas laterto be identifiedas the Stoa of Zeus (H7), he unearthedone of the most elegantsmallsculpturesever to be foundin the Agora(P1.4).1 As a headlesstorso, it did not claim a place among the outstandingsculpturalfinds reportedby the excavations'directorin Hesperia.Despiteits evidentbeauty,it was and remainsa puzzle.No obviouscontextor strongconnectionwith other worksfound in the excavationgave the key to its use and meaningthat mighthave inspiredthe publiare still cationwhichits qualityseemed to deserve.The problemsof its reconstruction unsolved.By now, however,we have learnedenoughabouta seriesof importantworks with whichit is relatedin style to make it worthwhileto presentthe statuettefor the considerationof scholars,in the hope that one of them, perhapsProfessorThompson himself,mayfindthe clues to its identityand the problemsof its reconstruction. DESCRIPrION

The little sculptureis madeof fine Parianmarble,carvedand finishedin exquisite detail.The preservedheightof the fragmentis 0.30 m.; that of the whole figurewould have been around0.52 m. The analysisof the drapedtype leads, as we shall see, to a makesit clearthat it is a 5th-century date around430-420 B.C., and the workmanship original.The figurewas evidentlypartof a two-figuregroup,such as we find frequently in vase paintingand relief in the secondhalf of the 5th centurybut rarelyin sculpture in the round.2A quiteyounggirl standswith her weighton the left leg. Her left hand restson her hip, whileher rightarmwas raisedas if to rest the handon the shoulderof anotherfigure.A rathernarrowstripof brokensurfaceon the properrightside shows that our figurewas joined to the neighboringone only from the shoulderto a little below the hip. The roundedoutline of the attachmentis not easy to read. It might representthe shoulderand loweredarmof a standingfigureor the raisedforearmof a 'Inv. S 65. The followingspecialabbreviationswill be used in this article: Despinis, EvAuoXr4= G. Despinis, Yv,43foXkjOIr /.LEET7) TOt) i'pyov Tov) 'AyopaKpITov, Athens 1971 Katalog der Skulpturen,II, Klassische Vierneisel-Schlbrb = B. Vierneisel-Schlbrb, Glyptothek-AMinchen, Skulpturen,Munich 1979 20n two-figuregroupssee especiallyH. Speier in RbmMitt47, 1932, pp. 1-94.

A CLASSICAL MAIDENFROMTHEATHENIANAGORA

41

seated one. Also difficultto explainis the indicationthat some partof the sculpture came close enoughto the rightthighof our figurein frontto have causeddifficultyin carvingthe folds in the lowerpartof the triangularoverfallof the himation.It is possible thatthe secondfigurewas seatedat an angleto the standingmaiden,so thatthe left knee of the seated figureoverlappedthe right thigh of the standingone. That would implyeithera largerscalefor the seatedfigureor a high thronewith a footstool.Let us leave this problemfor a momentandlook morecloselyat the survivingfragment. The head, arms,and legs belowthe knees are missing.The headwas attachedby a dowelfor whichthe hole (ca. 0.004 m. in diameter,0.013 m. in preserveddepth) survives nearthe properrightside of the neck; no smoothedjoint surfaceis preserved.A powderywhite substance,probablya cement, has left traceson the brokensurfaceof the neck to the right of the dowel hole as well as inside the hole. The left arm was attachedjust belowthe shoulderby a dowelof aboutthe same size (the joint surfaceis brokenhere too) and cementedwithouta dowelto the smoothjoint surfaceat the back of the hand,whichis partlypreserved.The presenceof cement on the brokensurface of the neck suggeststhat head and arm were originallyworkedin one piece with the statuebut werebrokenoff at some time andreplaced.3 The rightelbowand forearmwith partof the sleeve are brokenawaytogetherwith whateverthe arm restedon. Besidesthe breakwhere the other figurewas attachedin front, the whole front is somewhatbattered:the point of the right shoulder,both breastsand some of the draperyridges.The backis better preserved,with only chips gone from the higher ridge folds. The unbrokensurfacesare beautifullypreserved, unweatheredand showing tool and abrasivemarks. The whole was very carefully workedand finishedexceptwherethe proximityof the otherfigurepreventedit. Rows of holes madeby a stationarydrillappearin the deep furrowsbetweenthe legs in front and in the undercutting of the triangularoverfall.Smalldrillholes markthe ends of the furrowsof the himationwherethey wereoverlappedby the adjoiningfigure. DRESS

That the girl is very youngcan be seen from her smallbreastsand also from the draperyof her chiton.It has a very long kolposreachingalmostto the knees (visible only on the properleft side, wherethe himationis open, P1.4:d). This is frequentwith children,whose dressesare long in orderto allowroomfor growth.4The chitonhas no 3For the smooth joint surface without a dowel compare the cutting for a repairof the nose in a 5thcentury head of Parianmarblefound in the Kerameikos:Riemann, KerameikosII, Berlin 1940, pp. 87-88, no. 116, pl. 26; J. Frel, AAA 5, 1972, p. 74, no. 3, figs. 3, 4; A. Delivorrias, AttischeGiebelskulpturen und Akroteredes finften Jahrhunderts, TUbingen1974, pp. 164-166, pls. 56:b-d, 57. A resemblancein style to Agora S 429b, the head of a girl attributed to the Theseum, was noted by Frel and confirmed by Delivorrias. 4Athena wears a long kolpos on the white-groundcup in the British Museum which depicts the creation of Pandora(Anesidora). Hephaistosthere is shown as very young and Athena is probablymeant to be so too (AR V2, p. 869, no. 55, TarquiniaPainter). The long kolpos may also be worn by a grown person in a very rich dress, whose materialis long as well as wide. This is the case with the Basilinnaon the Par-

42

B. HARRISON EVELYN

overfall.A very full "sleeve"hangsdownfromthe rightarmandoverlapsthe himation witha deep loop in back(P1.4:b). It must have been buttonedall the waydownto the elbow.The left arm, by contrast,emergedbarefrom underthe himationover the left shoulder.Eitherthere were fewer buttonshere or the sleeve was pushedup onto the shoulder.The firstseems more likely,for there is no such mass of materialhere as on the rightside. If that is so, however,the neck-holemust have been off center,for the armholeis not deep as it is on the ValentiniAriadne,where the outer buttonsof the left sleeve have been unfastenedand the cloth fallsdownto revealthe flesh belowthe arm.The off-centerfasteningoccursin some figuresof children.5 Overher chitonthe girlwearsan amplehimation.One corner,drapedover the left shoulder,is gatheredinto fine foldswhichclingcloselyto the chitonin frontand follow the movementof the body.The tip is caughtunderthe front himationfolds crossing the left hip. In backthe himationhangsdown in straightfolds on the left side to the second corner,which is brokenaway.The upperpartis drawndiagonallyacrossthe backandunderthe rightarm,wherea loop of the chitonhidesthe upperfolds.In front the thirdcorneris foldedover to form a triangularapronwhose tip must have fallen above the left knee. The fourthcornerwill have hung verticallyfrom the left hand, whichholdsthe bunchedfoldsflat againstthe hip. Betweenthe frontand backedges a thin panelof chitonis revealedall the waydownon the left side (P1.4:d). The schemeof the draperyis uniqueand at the same time closelyrelatedto those of a series of importantworks.Althoughthis is a minor piece on a small scale, the amountof detailin the foldsequalsthat of muchlargerstatues,so that it can easilybe comparedwith them. The primaryancestoris the "KoreAlbani"(P1.5:a),6 but the influenceof the Nemesisof Rhamnous(P1.5:b)7is clearlyvisible.Closestof all in certain thenon frieze (M. Robertson, TheParthenonFrieze,London 1975, detail 4) as well as with the bride (formerly called bridesmaid)in the west pediment at Olympia(B. Ashmole and N. Yalouris, Olympia,London 1967, pls. 98, 101, 105). She is identifiedas the bride not only by her rich dress but by the fact that in the correctreconstructionshe is under the outstretchedarm of Apollo (see P. Grunauer,"Der Westgiebel des Zeustempels von Olympia," JdI 89, 1974, pp. 29-48, especially pp. 45-48). Theseus is her rescuer. For Theseus as rescuer of the bride, see B. B. Shefton, "Heraklesand Theseus on a Red-figuredLouterion," Hesperia31, 1962, pp. 341-344. Shefton, however, kept the traditionalidentificationof West H as Deidameia, explainingthat outside Athens Theseus might have been given less prominence. Kleidung,Berlin and Leipzig 1928, pl. 12. For full publicationand discus5See M. Bieber, Griechische sion of the AriadneValentini,see E. Bielefeld, AntP XVII,Berlin 1978, pp. 57-69, pls. 35-39. 6Helbig4IV, pp. 316-317, no. 3342 (Fuchs) with selected bibliography.Good picturesof most views in H. Schrader,Pheidias,Frankfurt-am-Main1924, figs. 21, 22, 62, 65. Back view in AntP XI, Berlin 1972, facing p. 20, fig. 2. In a study not yet publishedI proposeto identify the Kore Albani with the Sosandraof Kalamison the Athenian Acropolis.Cf. J. Frel, GettyMusJ1, 1974, p. 57 and note 19. Delivorrias ("Das Originalder sitzenden 'Aphrodite-Olympias',"AthMitt93, 1978, pp. 17-18) seems to accept the identification of the Kore Albani with the Sosandraof Kalamis,as well as its close affinitywith the seated Olympias type and the architecturalsculpturesof the Theseum but does not take the step of attributingthe latter to the school of Kalamis as I would do. Vierneisel-Schlbrb(pp. 114, 165-166) is against the identification with Sosandra.The text of the present article was alreadywritten when Vierneisel's impressive work appeared.I do not attempt here to argue in detail all the points of disagreementbetween us. So far as chronology is concerned, I hope that the style sequence outlined in this articlewill make my position clear. 7Despinis,Y-vii,.ok'q.

A CLASSICAL MAIDENFROMTHEATHENIANAGORA

43

forms,thoughtotallydifferentin personality,is a typeknownonly fromRomanportrait statues,whichDespinisidentifiesas Koreand attributesto Agorakritos(P1. 5:c).8 Since the evidence for its identificationis tenuous, I preferto call it "type of the Lateran A Grimanistatuettein Venice (P1.5:d), now generallyconsideredto repreAgrippina". sent Kore,was treatedby Hekleras belongingto the same type' but shouldbe separatThat in turn is very close to the VelletriAthena,with which ed from it as a variant.10 our statuettealso has affinities(P1.6:a, b).11 It will appearthat the Agora maidenis slightlyearlier,at leastin type, thanthe Athenaandthe Grimanistatuette. DEVELOPMENT OFTHETRIANGULAR OVERFALL

These close similaritiesmakeit possibleto studyas a formalseriesone of the characteristicelementsof the draperyof our figure,the apronliketriangularoverfallof the himation,in its linearpatternsand in its plasticrelationshipto the body.This shouldbe a help in dating. The triangularapronof the KoreAlbaniis dividedinto two mainsystemsof folds which are clearly separatedone from the other: a fan of diverging straight folds where the corner hangs down over the left leg and a group of catenaries over the abdomen. The edge forms broad zigzags at the ends of the straightfolds and then curves smoothly upward following the line of the catenaries until it disappearsunder the bloused chiton above the right hip. The folds along the upper part are not separatedfrom the catenariesby greaterdensity or projection.

In the LateranAgrippina,whichseems to be the best copyof its type, the cornerof the overfallhas swungmore towardthe center.The foldedupperedge is caughtunder the elbowratherthanhigherup as on the KoreAlbani,so that it risesgentlyinsteadof steeplytowardthe left side. The upperfolds are gatheredclose togetherand begin to projectas a separatemass.The straightdivergingfolds becomediagonal,and the catenariesare reducedto a few thin folds whichmediatebetweenthe two systemsso that the folds seem evenly spreadover the whole areaof the apron.A single straightcoun8Despinis, IvAiok 4, pp. 180-182, list of replicas in note 384. Add to his bibliography:M. Bieber, AncientCopies,New York 1977, p. 122, figs. 544-545; Vierneisel-Schlbrb,pp. 163-177, no. 15. 9A. Hekler, "Rbmische weibliche Gewandstatuen," MilnchenerArchdologische Studiendem Andenken Addf Furtwdnglers Gewidmet,Munich 1909, pp. 151-152, 225. 10SeeDespinis, Ivi4,3oX4, p. 191, where the Kore Grimaniis said to take a separateplace in the circle of works derived from the creation of Agorakritos.For full discussionof the Kore Grimani see R. KabusJahn, AntP XI, Berlin 1972, pp. 1-21. Her suggestion of an Attic origin should be correctedin favor of the demonstrationby L. Beschi that the Grimani statues must come from Crete ("Antichita'cretesi a Venezia," ASAtene,n.s. 34-35, 1972-73, pp. 494-502). See also below, footnote 11. "For the Velletri Athena, see E. B. Harrison,"Alkamenes'Sculpturesfor the Hephaisteion,"AJA 81, 1977, pp. 150-155, 164-174, list of replicaspp. 175-178. The date around 420 B.C. is based on style; it is not dependenton my identificationof the Velletri Athena with the Athena Hephaistiabut vice versa.Compare Kabus-Jahn,op. cit., p. 19: "Um 420 wirdman mit derAthenaunsereStatuette[Kore Grimanil ansetzen konnen."Vierneisel-Schldrb(pp. 138-140) defends the attributionof the Velletri Athena to Kresilas and suggests a date no later than 430 B.C. She accepts a date of 420 for the Grimani statuette, however (pp. 164-166).

44

EVELYNB. HARRISON

terfoldstrikesdownfromthe properright.The patternof the apronof the Agoramaidwith en is virtuallyidenticalwiththis. Onlythe foldsare somewhatfewer,in accordance the smallerscale, and the difficultyof workingthe lower part makes the folds here clumsyandheavywherethey are smoothand lightin the largerstatues. The VelletriAthenaswingsthe point of the apronstill farthertowardthe middle andspreadsthe fan of straightfolds untilit becomesthe dominantmotif.The bunchof curvedfolds occupiesthe remainingspace,leavingno roomfor catenaries.The straight counterfoldis still there, but very shallowand unemphatic.Broadspacesappearbetween the fan folds of the apronas well as betweenthe folds of the lowerhimation.In the Agorastatuette,as in the LateranAgrippina,the ridgefolds are close together,the valleysnarrow.In this, too, they showthemselvesearlierthanthe VelletriAthena. The Athena, dated around 420 B.C., marks a terminus ante quem for our statuette, and the Nemesis, datedin the late 430's,12a terminuspost quem.The LateranAgrippina

typeoughtnot to be laterthanour figure,since it is easierto thinkthatthe artistof the minorworkimitatedthe apronpatternof the majorone thanthe otherwayaround. This clearline of evolutionin the patternsof the himationwith triangularoverfall seems to place the Dresden "Zeus" (P1.6:c) after the VelletriAthena, i.e. after 420 of the wholehimationbringsthe pointof B.c.13Althoughthe somewhatcloserwrapping the trianglebackto the properleft side, the big, straightfan folds remainlike those of the Athena,and their spacingis even wider.The stanceand the close wrappingbring the foldsof the lowerhimationinto a diagonalpatterncontinuingthatof the overfall.14 Since the stance of the "Zeus"looks like a forerunnerof that of the Ares Borghese, one is temptedto returnto the old idea that the "Zeus"is a copy of an AsA fragmentaryfigurefrom a votive relief in the Agora (P1. klepiosby Alkamenes.15 '2Despinis, Edvgok, pp. 55-61. It is above all the close resemblanceof the Nemesis to figures from the Parthenonpedimentsthat suggests this date. We may add that the stance remainsclose to the Pheidian stance of the Athena Parthenosand does not show the Polykleitaninfluence which is alreadyvisible in the "Agrippina"and other Attic works of the 420's. zur griechiUntersuchungen '3This date is approximatelythat proposed by F. Hiller, Formgeschichtliche v. Chr., Mainz am Rhein 1971, pp. 65, 70. He does not give dates, schenStatuedes spdten5. Jahrhunderts but his sequence of monuments implies it. At the same time Despinis (1vp,4oXiv,pp. 133-145) dated the Dresden Zeus early in the decade 440-430 and identified it with the Hades by Agorakritosin the Temple of Athena Itonia in Koroneia.Also at about the same time P. Mingazziniproposeda date of 360-330 B.C. ("Lo Zeus di Dresda, lo Zeus di Cirene, lo Zeus di Faleri e lo Zeus di Fidia," ASAtene,n.s. 31-32, 196970, pp. 71-76), but no compelling stylistic argumentssupportedthis novel dating. Vierneisel-Schlbrb(pp. 148-149) agrees with Despinis' date and tentativelywith his attribution. "4TheVelletri Athena also shows this continuityof folds from the overfall to the lower himation. '5This was proposed by A. Furtwingler (Meisterwerkeder griechischenPlastik, Leipzig-Berlin1893, p. 84) on the basis of Pausanias'mention of an Asklepios by Alkamenes at Mantinea (viii.9.1). G. Treu (OlympiaIII, Berlin 1897, pp. 225-230 and FestschriftOtto Benndorf,Vienna 1898, pp. 99-110) argued against the identificationas Asklepios and restored the figure with a scepter. Despinis' identificationas Hades takes accountof the fact that the hair of the god, with its uncontrolledlocks covering neck and ears and closely framingthe face, does not belong to the classicalimage of Zeus. It seems suitable for an underworld god or a hero. The expression is too mild for Poseidon, who wears such hair on the Parthenon frieze. Opponentsof the identificationas Asklepios argue that the proud carriageof the head does not

A CLASSICALMAIDEN FROM THE ATHENIAN AGORA

45

6:d06 can be identifiedas Asklepiosby its very close correspondencein pose to the Asklepioson a votive relief in the Athens NationalMuseum (P1.6:e).17The Agora fragmenthas an overfallvery muchlike that of the Dresden"Zeus",as well as a wide stance which producessimilardiagonallystretchedfolds in the lower himation.The overfallof this figure is also strikinglylike that of a male himationfigurefrom the Nemesisbase.18 A secondrelieffragmentfrom the Agora (P1.7:a) showsa male figurewho is apparentlyleaningon a staffproppedunderhis left armpit.19 The slabseems too thin for a documentaryrelief; it must have been votive. This fragmenttoo is most easily interconform to the usual kindly image of Asklepios. This is rathersubjective, however, for we have no other preservedAsklepios types in the round so early as the time of Alkamenes. The Asklepios of the wagoner relief, Athens N.M. 1341 (see below, footnote 17) is hardlyless proud than the Dresden "Zeus". D. Mustilli (BullComm61, 1933, pp. 19-20) arguesthe suitabilityof the head for Asklepios. In defending his identificationas Hades, Despinis stresses the beneficent, fertility-givingaspects of Hades, but one would expect that kind of Hades to carrythe Horn of Plenty. A certainfatherlyqualityis lent to the statue by the shawllike way the himation covers the back of the right shoulder. This is frequent with older men on Attic gravestones but not among the gods, whom we do not imagine as being afraidof drafts. Asklepios is an exception; he wears his himation in this way in several votive reliefs from the Athenian Asklepieion. Standing:Athens N.M. 1331, EA 1222; I. N. Svoronos, Das AthenerNationalmuseum,Athens 1908-1937, pl. 36; Athens N.M. 1332, EA 1236, Svoronos, loc. cit., S. Karouzou, Collectionof Sculpture,Athens 1968, pl. 45:a; G. Neumann, Problemedes griechischenWeihreliefs,Tilbingen 1979, pl. 47:a. Seated: Athens N.M. 1330, U. Hausmann, Kunst und Heiltum,Potsdam 1948, pl. 9; Athens N.M. 1338, EA 1232. In Athens N.M. 1408, EA 1245, Zeus Meilichios (seated) has the himation over the back of the right shoulder. In this, as elsewhere, Zeus Meilichiossimply borrowsthe type of Asklepios. These argumentsdo not exclude the identificationof the Dresden "Zeus" as Hades, because we have so little evidence for his iconographyapartfrom that of Ploutos, but they make the identificationstrongly dependent on style. Neumann (op. cit., pp. 67-68) tentatively accepts Despinis' identificationprovided it can be supportedby stylistic arguments. It seems to me that the stylistic arguments for Alkamenes are stronger than those for Agorakritos.The oft noted resemblanceto Poseidon in the east frieze of the Parthenon links the "Zeus" with Alkamenes rather than with Agorakritos.The fragments of a high-relief frieze from the Agora are clearly related to Alkamenes and at the same time clearly descended from the atelier of the Parthenonfrieze (see Harrison, op. cit. [footnote 11 above], pp. 164-166). The affinitiesare especially clear with the slab on which Poseidon appears.The tightly wrappedhimation with transverse folds in the lower part, alreadyhinted at in the Poseidon, also seems to have interested Alkamenes (see ibid., pp. 275-276), whereaswe have no evidence of its use by Agorakritos. '6Inv. S 2050. Found June 1, 1959 in packingbeside a late Roman wall north of the south stoa terrace of the Eleusinion (U 20). P. H. 0.29 m.; p. W. 0.255 m.; Th. 0.09 in.; Th. of backgroundca. 0.06 m.; H. of plinth 0.035 m. Worked with the point on side and bottom, rough picked on back. Pentelic marble. It is interestingto have a fragmentof an Asklepios relief from the area of the Eleusinionwhich can be dated by its style to the time between the initial introductionof Asklepios into the Eleusinion in 420/19 B.C. and his establishmentin his permanentsanctuaryon the south slope of the Acropolis in 413/2 B.C. (IG II2, 4960; cf. Hausmann,op. cit., pp. 21-23). "Athens N.M. 1341. See especially Beschi, "Rilievi votivi attici ricomposti," ASAtene, n.s. 31-32, 1969-70, pp. 86-94.

'8L. Kjellberg,Studienzu den attischenReliefs, Uppsala1926, pl. 4, figs. 13-15 (original) and pl. 7, fig. 24 (Roman copy in Stockholm); B. Kallipolitis, "'H a3c'O-'q Tov aycaXAaTo' Tr Pajovova-i'a NEoucrrm," 'ApX'E4,1978 [1980], pl. 2:1A and pl. 26. '9Inv. S 621. Found February26, 1936 in demolition of the east foundationsof the Churchof Panagia Vlassarouin the central partof the Agora (K 10-11). Broken on all sides; rough-pickedback preserved.P. H. 0.21 m.; p. W. 0.235 m.; Th. 0.06 m.; Th. of background 0.043 m.

46

EVELYN B.HARRISON

pretedas representingAsklepios,who commonlyadoptssuch a leaningpose. Againwe have a triangularoverfall.Now, however,the materialseems very thin and clinging, withdelicateridgefolds.This style recallsthatof the reliefon the Choiseulstele in the Louvredated410/409B.C.20 and thatof the Nymphreliefof the Archandros(P1.7:b).21 The secondAgorafragmentshouldnot be earlierthan410; it seems laterthanthe previous fragmentand laterthanthe Rhamnousbase.22 It was the resemblanceof the Dresden"Zeus"to the votive statueof Lysikleides at Rhamnous(P1.7:c), itself relatedto the Rhamnousbase, that firstsuggestedthe attributionof the Dresden"Zeus"to Agorakritos.23 We have alreadyseen, however,that Alkamenesin the VelletriAthenaborrowedfrom the workof Agorakritosrepresented in the LateranAgrippina.It seems perfectlypossiblethat the sculptorsof the Nemesis base, made more than a decadeafter the statue, might have borrowedin turn from Alkamenes.To sum up, the argumentsin favorof Furtwingler'sdenominationof the "Zeus"as Asklepiosare (1) that the figurelooks like Asklepiosand seems to have of that god,24(2) that an Asklepiosby Alkamenesis influencedother representations recorded25 and none by Agorakritos,and (3) that the figureis a link betweenthe Veltwo figuresattributedon other groundsto Alletri Athenaand the Ares Borghese,26 kamenes. The factthat the "Zeus"seems to have held a scepterdoes not rule out his identificationas Asklepios.The SikyonianAsklepiosof Kalamisalso held a scepter.In describingthe latterPausaniasmakesno mentionof a snakeas partof the composition, thoughhe tells us that the god was broughtto Sikyonin the form of a snake.27If the youthfulAsklepiosof Kalamiswas standing,as seems quitepossible,it may have intro20IG 12, 304. Louvre. Depart. des antiquitiesgrecques et romaines. Cataloguesommairedes marbres de lart, Paris 1938 (Tel), III, pl. 168. U. Kron, photographique antiques,Paris 1922, no. 831. Encyclopedie Die zehnattischenPhylenheroen(AthMitt,Suppl.V, Berlin 1976), pl. 29. 21AthensN.M. 1329. Svoronos, op. cit. (footnote 15 above), pl. 44. BrBr,no. 439. 22Thereis also a relationshipto the himation figures of the grave lekythos, Athens N.M. 2584 (B. Berlin 1970, A 21, pls. 11, 12), but these seem zu den attischenMarmorlekythen, Schmaltz, Untersuchungen alreadysomewhat stiffened in comparisonwith our fragment. Schmaltz (p. 29) dates the lekythos before the Hekatompedontraditioof 397/6 B.C. (IG II2, 1392). p. 133. 23Despinis,EvA,8okrv, 24Besidesthe apparentreflections on votive reliefs mentioned above, we may note the considerable preponderanceof Asklepios heads over heads of other divinities in Curtius' census of Roman Umbildungen of the head of the Dresden "Zeus" (L. Curtius, Zeus und Hermes, RimMitt, Suppl. I, Munich 1931, pp. 20-37). It is not necessaryto agree with all Curtius' derivations in order to see the general picture. For criticismof Curtius,see Vierneisel-Schlbrb,p. 147. 25Footnote 15 above. Furtw-anglerderived the "Zeus" from the statue mentioned by Pausaniasat Mantinea,since it has generallybeen thought that no cult statue of Asklepios existed in Athens before the permanentinstallationof the god on the south slope of the Acropolis, but the question ought perhapsto be reopened. In the period 421-415 B.C., to which the statue can be dated by its style, Alkamenes was workingin Athens. It could well be that the statue in Mantineawas derived from one in Athens (set up in the Eleusinion, or simply made and kept in reserve until the site problemswere solved?) rather than the other way around.This is a largequestion, which cannot be gone into here. 126, Berlin 1977), pp. 33-37 and 59 (bibliography). 26SeeW. Schuchhardt,Alkamenes(BerlWinckProg Add Vierneisel-Schl6rb,pp. 178-187. 27Pausanias,11.10.3.

AGORA MAIDEN FROMTHEATHENIAN A CLASSICAL

47

We have seen that the ducedthe triangularoverfallinto the iconographyof the god-.28 Sosandraof Kalamis(the Kore Albani) standsat the:head of a series from which is overfallof our Agoramaiden. derivedthe triangular The boy of Lysikleides(P1.7:c)29and the Hera Borghese(P1.7:d)30belong to a stagebeyondthatof the Dresden"Zeus".Herethe smoothpartof the overfallclingsto the inside of the right thigh and the fan folds are confinedto the properleft. Many figuresfrom the Erechtheionfrieze show the himationclingingto the inside of the So does the Kore thighs,31as does the Apolloof a late 5th-centuryrelieffromBrauron.32 The leaningAphrodite34 is shownas fromthe Piraeusin the AthensNationalMuseum.33 belongingto an earlierstyle phasethanmost of its imitatorsby the fact thatthe overfall of the himationdoes not clingto the thighsbut formsa bridgeacrossthem. The linear patternof the folds in the overfallis still roundedlike thatof the KoreAlbani.The conventionaldate of the Aphroditearound430 B.C. or a little later is thus upheld.In the leaningNike fromthe Nike Parapet(P1.8:d) the overfallclingsto both thighs,and the upperfolds of the himationform a shell-likeprojectionaroundthe righthip. It seems withthe Erechtheionfrieze.35 likelythatthis partof the Parapetis contemporary OTHERDRAPERYMOTIFS

overfall,but in otherrespectsits inThe Nemesisof Rhamnouslacksthe triangular fluenceis clearlyfelt in the LateranAgrippinaand the Agoramaiden.Most notablein 28It is not clear to me why G. Mansuellisays (EAA I, pp. 720-721) that the Asklepios of Kalamismay representthe beginningof the traditionof a seatedAsklepios. 29AthensN.M. 199, 'EO'ApX1891, p. 55, pl. 6; Kjellberg,op. cit. (footnote 18 above), pl. 5, fig. 18; P. ZancaniMontuoro, BullComm61, 1933, p. 54, fig. 20. 30ZancaniMontuoro, op. cit., pp. 25-58, with illustrationof the principalreplicas,attributionto Agorakritos;W. Fuchs, Die Skulpturder Griechen,2nd ed., Munich 1979, pp. 205-207, fig. 222 (unchangedfrom first edition) with attributionto Kallimachos;Despinis, EvApok', pp. 156-158, attributionto Agorakritos; Bieber, op. cit. (footnote 8 above), pp. 47-49, pls. 29-31; Vierneisel-Schlbrb,pp. 166-167, 171, notes 11, 12, with dating420-410 B.C., cautiousabout attribution. 31E.g.AntP X, Berlin 1971, pls. 3, 4, 11, 12, 15, 16, 26. 32p. Themelis, Brauron,Athens 1971, p. 59; Neumann, op. cit. (footnote 15 above), p. 62 (with additional bibliography). 33AthensN.M. 176. S. Karusu, "Das 'M-adchenvom Pir-aus'und die originalstatuenin Venedig," AthMitt82, 1967, pp. 158-169, Beilagen 85, 86. 34Schrader,op. cit. (footnote 6 above), pp. 203-210, especially207, figs. 187, 188. Cf. Harrison,op. cit. (footnote 11 above), p. 417. 35Forthe Nike see R. Carpenter,Sculptureof the Nike TempleParapet,Cambridge,Mass. 1929, p. 49, no. 9, pl. 20:1; for the additionalfragmentfound in the Agora excavations, see E. Harrison,"New Sculpture from the Athenian Agora, 1959," Hesperia29, 1960, p. 376, pl. 83:a; also idem (footnote 11 above), p. 417 and p. 418, fig. 1. The style phase seems to be the same as that of the Hera Borghese.The question whether the Nike Parapetwas made between 421 and 413 or 409 and 404 has perhapsbeen wronglystated. The reliefs of the north flank, Carpenter'sMastersA and B, seem clearly to belong before 410, those of the south flank clearly after (in view of the resemblance to the style of the Erechtheion in the work of MasterE and elements foreshadowingthe 4th century in that of MasterF). The works of MastersC and D on the west face, though less obviously datable, also appearto belong to the time after 410. It seems logical, therefore, to assume that work startedon the north flank late in the Peace of Nikias and was interrupted, like that on the Erechtheion,in 413, to be resumed around410/9.

48

EVELYN B. HARRISON

the Laterantype is the enrichmentof the zigzagson the properleft side;36they are more numerousthan on the Kore Albani, and the biggerfolds are subdividedinto manysmallparallelfolds.The foldson the Nemesisare strikinglysimilar.37 On Nemesis, Lateranstatue,and Agoramaiden,the cornerof the himationwhich came forwardover the left shoulderis shorterthan on the Kore Albani,resultingin a smallermass of draperyover the shoulder.On the Nemesis, the tip hangsfree. On the Agrippinait is loopedup and tuckedunderthe crossfolds.On the Agoragirl it lies flat andthe crossfoldscoverit. A similartreatmentis used in the VelletriAthena. In the Nemesisas in the Agorastatuettethe himationwas pushedup onto the left shoulderratherthan halfcoveringthe upperarmas it does in the KoreAlbanior fully coveringit as in the Agrippina.The chitonis visiblein a very narrowstripon the side of the Nemesis, in a widerone on the Agoragirl. The Agrippinadid not show the chitonon the side. Becauseof the smallsize of the Agorafigureand becausethe edge of the backpartof the himationis brokenaway,it is not possibleto judge what the artistdid withthe zigzags,but they seem to have been partlysacrificedin orderto show a widerareaof chiton.The Grimanistatuettealso does this to some extent, but with heavier folds (P1.8:a).38

In the fine folds of the chiton, the Agorastatuetteborrowsfrom severalsources but remainsindividual.The absenceof the overfalldistinguishesit from all the socalled Kore types with which we have comparedthe himation,as well as from the Nemesis. The way in whichthe folds cling to the body betweenand below the small breasts,emphasizingthe gentlyslumpedpostureof the uppertorso, is very similarto thatof Figure23 in the east friezeof the Theseum(P1.8:b), and the basicpatternof all these folds is almostidenticalin the two works.Onlythe folds are muchmore numerous in the statuettein spiteof its smallersize. Wherethe himationcrossesin front,the chitonfolds breakagainstit in little hooks, as they do in the VelletriAthenaand in a This motifis not yet presentin the Nemesis,and high-relieffragmentfromthe Agora.39 thereis just a tentativebeginningin the LateranAgrippina. The backof the chiton (P1.4:b, c) gives an effect of richness,whichis not so apparentfrom the front.A big loop of cloth, subdividedinto two main catenaries,overlapsthe himationbelowthe rightarm.Belowthe napeof the neck are smallstackedcatenarieshatchedwith tiny verticalcrinklefolds. This effect is used in the front of the chitonof the KoreAlbani,whichmay also have inspiredthe loopsof loose cloth overlappingthe mantle. 36Sideview in Bieber, op. cit. (footnote 8 above), fig. 545. pl. 37. 37Sideview in Despinis, 1vji43oX'4, 38AntPXI, pls. 3b, 9b. 39Inv.S 679. Harrison, op. cit. (footnote 11 above), p. 165, fig. 20. I there suggested that this motif was an idea of Alkamenes. It appearsratherthat he only formalizedand made more emphatican existing motif. The Kore Grimani shows it still in a tentative stage, though the statuette seems otherwise to be contemporarywith the Velletri Athena. Vierneisel-Schlbrb(p. 166) rightlynotes that it is a linearizationof the plastickolpos motif of the Kore Albani.

A CLASSICAL MAIDEN FROMTHEATHENIAN AGORA

49

The gestureof the left handrestingon the hip and holding.the folds of the himation in placeis similarto that of the figurefrom the Nike Parapet(P1.8:d) whichalso wearsthe thin chitonwithoutoverfallor visible belt.40The only differenceis that the fingersof the Agoramaidenactuallygraspthe folds ratherthanjust restingon them. The comparisonof the two figuresshowsthe Nike as markedlylaterin style."1 The figurewho standsbehindAphroditeand Eros in the east frieze of the Nike She rests her righthandon Temple(P1.8:c) also wearsa thin chitonwithoutoverfall."2 a her hip.The folds have finenessthatis lackingin our piece, partlybecausethe Pentelic marbleof the Nike frieze permitsmore delicatecarvingthan the Parianmarbleof our statuetteand partlybecausethe Agorapiece is a little earlier.In the stringyquality of the fine folds and the lack of transparency, our figureis closest to the chiton-clad figuresof the Theseumfriezes.43These friezes,like our statuette,seem to be relatedto the frieze of the Nike Templebut clearlyearlier.A date around430-425 B.C. for the friezesof the Theseumwouldexplainthe evidentinfluenceof the Parthenonpediments on the seatedfiguresand yet allowtime betweenthese friezes and those of the Nike The groupof two girlsin the "Ephedrismos" Temple."" schemewhichprobablyserved as an akroterionfor the Theseum is extremelyclose in style to the Nike Temple frieze."5 We wouldexpectthe akroterionto be somewhatlaterthanthe friezes. THE SCULPTOR

This long look at the detailsof our statuetteand theirstylisticrelativeshas seemed worthwhilebecause,being an originalwork, it can be relatedwith greaterimmediacy thancan the copiesto the architectural sculpturesand votive documentaryreliefsof its 40Footnote35 above. 41Theprojectionof the upper himation folds and the clinging of the overfall to the thighs mentioned above are the principleelements. Cf. footnote 35 above for the date of the Nike. 42C. Blimel, Der Friesdes TempelsderAthenaNike, Berlin 1923, pl. 8, fig. 1. 43H.Koch, Studienzum Theseustempel in Athen, Berlin 1955, pls. 29, fig. 2; 33, fig. 21; 34, fig. 26; 39, fig. 16. C. Morgan, "The Sculpturesof the Hephaisteion,"Hesperia31, 1962, pls. 77, figs. 2, 7; 79, figs. 23, 26; 81, fig. 16. S. von Bockelberg,"Die Friese des Hephaisteion,"AntP XVIII, Berlin 1979, pls. 16, fig. 2; 19, fig. 7; 27, fig. 21; 28, 29, fig. 23; 30, fig. 26; 45, 46, fig. 16. 44TheNike Temple frieze is not exactly dated by external evidence. Ira Mark (Nike and the Cult of AthenaNike on the AthenianAcropolis,diss. New York University 1979, pp. 266-267) argues for a date shortlyafter 421 B.C. 45H.A. Thompson (The AthenianAgora, Guideto the Excavationand Museum,3rd ed., Athens 1976, pp. 195-196) suggests that the group belongs to one of the pediments.Previously ("The PedimentalSculpture of the Hephaisteion," Hesperia18, 1949, p. 235 and "The SculpturalAdornment of the Hephaisteion," AJA 66, 1962, p. 345) he had identifiedthe group "with certainty"as an akroterion.This was accepted by F. Eckstein in his publicationof the ConservatoriEphedrismosgroup (AntP VI, Berlin 1967, pp. 75-88), in which he also saw a central akroterion:"Kein Zweifel-die Gruppevon der PiazzaDante krinte einst, wie ihre geschwisterliche Gruppein Athen, den First eines Gebdudes"(p. 84). A. Delivorrias (op. cit. [footnote 3 above], pp. 33-40) arguedthat the Agora group belongs to a pedimentalscene, perhapsa Sack of Troy, in which a girl helps a wounded comrade. One cannot prove that the piece was not pedimental, for some of its weatheringprobablypostdates the destructionof the building to which it belonged, but I continue to believe that the subject, whose relationshipto the Ephedrismosterracottascannot be ignored, has eschatologicalimplicationsthat would make it suitablefor akroteria,just as abductiongroupsare.

50

EVELYNB. HARRISON

period,while, at the same time, as a carefullydetailedsculpturein the round,it helps us to have faith in the copies which it resembles.Especiallyinterestingis the close resemblancein the patternof the triangularoverfallof the himationto that of the LateranAgrippina.Becausethe latteris very immediatelyrelatedto the Nemesis of Rhamnousand to the Parthenonpediments,Despinis' attributionto Agorakritosis thoroughlyconvincing.46 matronlyfigure as Less persuasiveis his identificationof the broad-shouldered, the Demeter in the same cult Capitoline Kore, and his proposalto associateit with group.47 The fact thatthe designof the statuewas availableto be imitatedby the sculptor of our little figuresuggeststhat the originalof the Agrippinamay have been set up in Athens, perhapsactuallynear the Agora. The carvingof the chiton folds of the Lateranstatueso muchresemblesthatin copiesof the Nemesis, andeven, in places,of the Nemesisitself, thatone has the impressionthat the two originalswere madeof the same material,Parianmarble.Two Romancopieswhichhave contaminatedthe type by substitutingdifferentformsin the upperpartof the chiton, namelythose in Munich48 andin SyonHouse,49clearlyindicateby baringthe rightshoulderthatthe Romanladies they representare beingportrayedas Venus.50 All this suggeststhe tantalizingpossibilitythatwe have in the originalof the Lateran Agrippinathe cult statueof AphroditeOuraniain ParianmarblewhichPausaniassaw He calledit a workof Pheidias,just as he gave to Pheidiasthe statue nearthe Agora.51 No exact of the Motherof the Gods52which better authoritygives to Agorakritos.53 replicaof the Mothersurvives,and it is not possibleto form from the manyand various small reflectionsa sufficientlydetailedidea of the originalto date the figureon style, as we have tried to do with the Agrippina.Such literaryevidence as there is StevenLattimorehas recently suggeststhat a date around425 B.C.wouldbe suitable.54 46Loc.cit. (footnote 8 above). 47I continue to believe that we have the original of the Capitoline Demeter in the fragments of a statue in the Athenian Agora (Harrison, op. cit. [footnote 35 above], pp. 371-373, pl. 81:c). The scale is colossal. No fragmentsof a Kore of comparablesize have come to light. pp. 163-177, with bibliography. 48Vierneisel-Schldrb, 49F.Poulsen, Greekand RomanPortraitsin EnglishCountryHouses, Oxford 1923, pp. 16-17, fig. 13; C. Vermeule, "Notes on a New Edition of Michaelis,"AJA 59, 1955, pp. 147-148, pl. 45, fig. 27; VierneiselSchlorb, pp. 163-164. 60Besidesbaringthe right shoulder, these statues also introduceinto the overfall of the chiton a windblown effect which belongs to Aphroditein her connectionswith the sea and sailing.See below, p. 51. 5'Pausanias,i. 14.7. Testimonia, 62Pausanias,i.3.5. See R. E. Wycherley, TheAthenianAgora, III, Literaryand Epigraphical Princeton1957, pp. 150-160, for testimony on the Metroon in general. "3Pliny,N.H., 36.17. ReligionI, 3rd ed., Munich 1967, pp. 725-727, argues that the 54M.Nilsson, Geschichtedergriechischen cult was introducedinto Athens in the late Archaicperiod. One ancient source connects the establishment of the cult with a plague (Photios, s. v. g-rqpayVp v7) = Wycherley, Agora III, p. 155, no. 487). Julian, OEmOVand OEpaITELag-~vL8oq.If Orat.V.159b (Wycherley,Agora III, p. 154, no. 483) speaks of JAM3TvL' there is any truth in the tradition, it may refer to a plague in the late Archaic period. Photios says the Athenians made a Bouleuterion.These references would fit well with the date around 500 B.C. to which

AGORA MAIDEN FROMTHEATHENIAN A CLASSICAL

51

shownthatAphroditeOuraniaappearswiththe Motherof the Gods in a double-naiskos votive relief from Isthmia,where both goddesseshave the seated form familiarfrom If AphroditeOuraniawas closelyrelatedto the Mother other votives to the Mother.55 of the Gods in Corinth,she mightwell be so in Athenstoo.56It wouldnot be surprising to find her similarlydressedand with a similarlymatronlyphysique.Both goddesses weara richchitonwith full "sleeves"and a long overfalland have the himationdraped over the left shoulder. If the LateranAgrippinawent for a walk in a high wind she might look like the Parian-marble Aphroditein the Agora,whichDespinisassignsto the Schoolof Agorakritosbut not to the masterhimself.57The overfallis blownup over the rightbreast, and the third cornerof the himation(which forms the apronin the Agrippinaand relatedstatues) is pulled up over the crook of the left arm to hold the cloak more firmlyas it flapsin the wind.This statuecould well be a votive to AphroditeOurania whichemphasizesher aspectof Euploia,the bringerof victoryat sea. She retainsthe heavybodyof the originalOuraniaand addsthe wind-blowndraperylike an additional epithet.58

This wind-blownAphroditeshouldbe abouta decadelater than the "Agrippina". As has been pointedout elsewhere,it is close to MastersA and B of the Nike Parapet, around420-415 B.c.59Ourstatuetteis firmlyanchoredin the precedingphase,contemporarywith the Theseumfriezesand with the LateranAgrippina,around425 e.c. The most recentstudyof the Theseumfriezeshas suggestedthat its mastersbelongedto an islandschool.60So far as the draperyis concernedthis could be true, but the strong plasticrealismof the athletichumannudes of both friezes and of the horse bodies of the centaursin the west friezesuggestsinfluencefrom mainlandbronzework.61One of Plato (erroneously)assigns a purificationof Athens by Epimenides (Laws 642D). Cf. G. Huxley, "Nikias, Crete and the Plague," GRBS 10, 1969, pp. 235-239, on references to an early plague or plagues in Athens and the possibility of the revival of old oracles in the time of the PeloponnesianWar. It would be naturalto give new honors to the Mother after the plague of 430-426 B.C.Something similar may have happenedin the case of Apollo and Herakles Alexikakos, giving rise to the mistaken idea that their cults were founded in the time of the PeloponnesianWar. 55AJA84, 1980, p. 220. 56Forthe identity of Aphroditein the Homeric Hymn with the MountainMother, see Nilsson, op. cit. (footnote 54 above), pp. 522-523. For identificationof the Mother of the Gods with Aphrodite in a Roman dedicationfrom the Piraeus, see IG I12,4714 (time of Augustus). 57Harrison,op. cit. (footnote 35 above), pp. 373-376, pl. 82. Despinis, EvAuoxrj,pp. 188-189. W. Fuchs, op. cit. (footnote 30 above), p. 210, fig. 225. Fuchs repeats the erroneous statement of his first edition that the lower partof the figure is restored 10-15 cm. too short. Actuallythe fragmentsjoin continuously from top to bottom, though this is not readilyapparenton the surfaceof the restoredstatue. 58Astrikingexample of wind-blowndraperyas an attributeis to be found in the Oreithyia(Q) of the west pediment of the Parthenon. 59Harrison,loc. cit. (footnote 57 above). Despinis (loc. cit. [footnote 57 above]) agrees that the statue is close to MasterA. 60S.von Bockelberg,op. cit. (footnote 43 above), pp. 46-48. 61Thesequalitiesare well describedby Bockelberg.Cf. Harrison,review of AntP XVIII, AJA 85, 1981, pp. 232-234.

52

EVELYN B. HARRISON

these mainlandsculptors,probablyKalamis,62 may have been the designerof the metopes, and he or one of his followersmayhave returnedaround430 B.C. to take responsibility for the friezes, associatingwith himself some younger island marble-cutters, whose mastermay have been Agorakritos.The materialof the friezes was the same relativelycoarse-grained Parianmarblethat had been used for the metopes,and it is reasonableto assumethatthis materialhadalreadybeen acquiredin the earlierbuilding periodof the temple.One wouldwantcarverswho werefamiliarwiththis kindof stone, not men whose only experiencehad been with Attic marble.One of these, perhaps even the manwho carvedslabV of the east friezeof the Theseum,mayhave been the sculptorof our statuette.The figurehas somethingof Kalamisin its descentfrom the Kore Albaniand somethingof Agorakritosin its contemporary connections.The Theseum sculpturesseem to springfromthe workshopof Kalamisandat the same time the frieze of the Nike Templeis in manywaysa continuationof theirstyle. So long as we cannotdate the ends of the careersof Kalamisand of Agorakritos,it will remaindifficult to untanglethe skein of collaboration and mutualinfluencein the post-Parthenon period,but workssuch as the Agoramaidenhelpus to see whatwasgoingon.63 ANDUSEOFTHESTATUETTE IDENTIFICATION

The Agorastatuetteis unusualnot simplybecauseof the fine detailthat was lavished on its back, but becausethe three-quarters backview from the properleft side (P1.4:c) seems to have interestedthe sculptormore than any other. From this angle the somewhatcrudelycut channelsin the chiton on the left side disappearand the patternof the soft folds showsto advantage.At the same time the carefullyworkedout patternof the chitonfolds belowthe nape and over the rightshouldercan be enjoyed togetherwith the richloops of the chitonsleeve underthe rightarm.The frontof the figure,by contrast,is treatedratherlike the side planeof a figurein relief. Perhapsthis interestin the obliquebackview of the maidencompensatedfor a lack of interestin the backof the figurewithwhichshe was grouped.Thatwouldhave been the case if the secondfigurewas enthroned.Since a votive relief from Eleusis,earlier than our statuette,shows Kore as a very young girl clad in a chitonwith long kolpos an enthronedDemeter,64the identificationof our groupas Demeter and approaching Koreseems the most probable. Thoughits style is a purelymarblestyle, the minutelydetailedexecutionof our smallfigureassimilatesit to votivesin bronzeand makesit virtuallycertainthat its use 62Seefootnote 6 above. 63Somewherein this world Kallimachosmust have got his start. If he were in some sense a pupil of Kalamis, it would explain the fact that Dionysios of Halikarnassoscouples the names of the two sculptors as representinga gracefuland naturalstyle like that of Lysias (Isokrates3). 64G.Mylonas, Eleusisand the EleusinianMysteries,Princeton 1961, fig. 67, pp. 191-192. Mylonasinterprets the figure as Hekate, but A. Peschlow-Bindokat("Demeter und Persephonein der attischenKunst," athenienne,Paris 1965, p. 22) maintainthe JdI 87, 1972, p. 110) and H. Metzger (Recherchessur I'imagerie identificationas Kore.

A CLASSICALMAIDEN FROM THE ATHENIAN AGORA

53

was votive ratherthanarchitectural. We have no way of knowingwherethe groupwas set up nor when it sufferedthe damageindicatedby the ancientrepairs.The contextin whichit was foundlong postdatesthe Heruliandestructionof A.D.267, but the careful techniqueof the repair,as shownby the smoothedjoint surfacein the left hand, suggests an earlierdate. We may recallthat HomerThompsonalso found, about 100 meters to the northwestof where the statuettewas discovered,an inscribed4th-century base whichonce carriedbronzestatuesof a prominentAthenianand his wife dedicated to Demeter and Kore. This was re-used in a context which suggestedthat it had sufferedin the Sullansackof 86 B.c.65Perhapsour groupbelongedto the same, still unidentifiedsanctuaryof the Two Goddessesoutside the northwestcornerof the AthenianAgora. EVELYNB. HARRISON INSTITUTEOF FINE ARTS

1 East 78 Street New York, NY 10021 65H.A. Thompson and R. E. Wycherley, The AthenianAgora, XIV, The Agora of Athens, Princeton 1972, pp. 154-155, pl. 78:c, with earlierreferences.The statue of the wife was signed by Praxiteles.

PLATE

4

~~~~~~i:~~~~~~~~~~~~~~~~~~~~~~~~i

a. Statue of a maiden, Agora S 65, front. Agoraphoto

c. AgoraS 65, three-quartersback. Agoraphoto

b. Agora S 65, back. Agoraphoto

d. Agora S 65, left side. Agoraphoto

PLATE

a. "Kore Albani" (from cast in Hamburg). Museumphoto

N

c.

Agrippina",VaticanMuseum. Vaticanphoto "Lateran

b. Nemesis of Rhamnous,copy in Copenhagen, Ny CarlsbergGlyptothek. Museum photo

d. "Kore Grimani",Venice, Museo Archeologico.PhotoDA! Rome

5

PLATE

6

I

.

c. "Zeus", Dresden Museum. Museumphoto

a, b. Velletri Athena, front and side. Museo Nuovo. PhotoDAI Rome

-Yi

'; i

-

--

d. Votive relief, Agora S 2050. Agoraphoto

e. Votive to reliefho

Askepos Athens

,NM13.

PLATE

V.~

~

~

a. Votive relief, Agora S 621. Agoraphoto

b. Relief dedicatedby Archandrosto the Nymphs, Athens, N.M. 1329. PhotoDAI Athens

-s

<

c. Copy of Hera Borghese,

d. Statuettededicatedat Rhamnousby Lysikleides,

7

PLATE

8

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I.

$

I ft

-

t

11 ;

^.

b. Theseum east frieze, figure 23, detail. Agoraphotoby AlisonFrantz

a. "Kore Grimani", left side. PhotoDAI Rome

[

it

A~~~~~~~A

4v

c. Temple of Athena Nike, East Frieze, detail. From a cast (from Blumel)

~

~

:

d. Nike Parapet,West Face, LeaningNike. Photoby WilliamR. Biers

THE EARLIESTATHENIANGRAVE (PLATE9)

N THE SPRINGof 1935duringhis importantexcavationson the West Side of the Agora, Homer Thompsoncame upon a very early burialmade before the Agora becamethe civiccenterof ancientAthens.The gravewas locatedin a smallside chamber at the bottomof a deep shaftthree metersbelowthe surfaceof bedrockjust to the east of the porchof the HellenisticMetroon(5.60 m. belowits lowestcourseof poros withthe consequencethatthe burialwas neverdisturbedin the repeatedbuildingoperations in this area;Fig. 1). It containeda completemale skeletonin contractedposition on its side providedwith two crudehandmadebowlsas offerings,one at the head, the reportas "LateNeolithicbefore otherat the feet, and was publishedin the preliminary 3000 B.c."1Therewere, however,certainanomaliesaboutthis burial-the form of the graveat the bottomof a deepshaftwhichfilledup withwaterso thata well couldnot be MiddleHelladiccharacterof the potteryfromthe fill.2 excluded,andthe predominantly In preparingthe publicationof the prehistoricmaterialfrom the Agoraduringthe 1960's,I discoveredthat this burialin the Metroondraincuttingat I 9:2 was my most controversialdeposit.Becauseof the natureof the fill and the lack of clear parallels amongour NeolithicmaterialI finally,withoutmuch enthusiasm,reassignedit to the earlyMiddleHelladicperiodor the time of LernaIV,3 a redatingwhich has recently come underdiscreetcriticism.4 Since it is not only a questionof a thousandyears but of the uniquenessof the burialif it is indeedNeolithic,in comparisonwithquitea few MiddleHelladicgravesin Athens,5I proposehere to re-examinethe evidenceandbringit up to datein the lightof IT. Leslie Shear, Hesperia5, 1936, pp. 20-21; AJA 39, 1935, pp. 439-441, where he acknowledgesthe assistanceof Mrs. Leslie W. Kosmopoulosin identifyingthe Neolithic characterof the pots. In "Buildings on the West Side of the Agora," Hesperia6, 1937, p. 1, Homer Thompson refers to this early burial (without date) and shows its location on p. 120, fig. 64, section C-C with respect to the plan of the Metroon on p1.VI. Specialabbreviationsused in this articleare as follows: AgoraXIII = SaraA. Immerwahr,TheAthenianAgora,XIII, TheNeolithicand BronzeAges, Princeton1971 = John E. Coleman, Keos, I, Kephala,Princeton 1977 Keos I 2The excavatorapparentlyhad some doubts about its date, and discussed them with me-without prejudice, I might add-as I was working on the prehistoricmaterialfrom the Agora. It is my hope that this article, by restoringthe MetroonGrave to its rightfulimportance,will make amends for my previouserror of judgment. 3AgoraXIII, pp. 92-93, pls. 27-28, 71 and 78. (SIMA 49), Goteborg 1977, pp. 61 and 40. T. P. K. Dickinson, The Originsof MycenaeanCivilization date (Colin Renfrew, The EmerNeolithic Late a 116, note 9 with reference to Renfrew who maintained London 1972, pp. 70, 110B.C., Millennium Third in the Aegean the and Cyclades The of Civilization: gence 111, 195, and fig. 7:6). 5MariaA. Pantelidou,At ITpourropLKat 'A6Pvat, Athens 1975, pp. 49-54 and 247-249; S. A. Immerwahr, AJA 82, 1978, pp. 407-409 (review of Pantelidou).

55

THE EARLIESTATHENIAN GRAVE

C

A _

.4.0

*

D

E

^

.

F

G .

I

H

K

J

L

NM N

P

0

o~~~~~~~~~~~~ 0-

.

2

!

T

R

Y

X

VW

AGOPAEXCAVATIOH *400

~~~~503

A

ERIDANOS

4|

U

50

100

4

/

S

|

5z/

\

/~

_)

-

l

/

)

t--___

)"'t'0

/ r-9X -ol /lsL jt, 9 9 KO~~~~~LON03 /KOLONOS r --------10

/

h E2

,;

72

05R-

0r72

072/

d V J84,

t /

/

f)

_n

i__

__?

6

MIDDLE HELLADIC

7

075 ~~~077

K&1

5<

WELLS *

07

"''Kl,

-~s--701^

r-\ l _t NEOLITHIC

%:

BURIAL

t

1

g

$

PITS& BOTHROIt

eIt 4~~~~t~o

T

104.eD5

H

STRATIFIEDDEPOSITS

11

MYCENAEAN WELLS

/_-_-______EL_-------

|

A/'

'

2

------1-----

/.\t

L

! 14

) >'-

/

15ie.2

~

=,

__

*^2

CHAMBER TOMBS

131

1

PITS & BOTHROI * STRATIFIEDFILLS +

N43

1410 _ _

_

-

GRAVES

---Jr

--

-

/

ffff

.M/

1

400

!;X/

?009

17i,

S S:;S.;NL

-

_

_

_

_

_

_

_ _

_

_

_

\v_

_

_

20

ELSINION

I

0002 M215

t

I

6

I

ARE0>AGUS

AR21;4

1-221 &

23~~~~~~~~~~~~~~~~~~~~~~~~~~2 T24 4\#

-

ONU242

T2410

(U243 *

Ti424*

25~~~~~~~~~~~~~~~~~~~~~~~~-U4

T25=-= 51

________

*2124 MYCENAEAN SPRING-MCOjSE -

U251

k*

J

~~

4

T261~~ *up 27

570 S27 20

28

I

~\ A

fi

| B |C

>

g

\qv

DC| E

()

/

/

| FO| G | H | I

#

J

K

!

IL

/

!

T67

*

ssi

M IN

f

/

o0 P fQ

2

R

VRI S

28

tACrOPOLI3

g

I

~~~~2

\

f +? tX

2

~

*S277 -

810

/

rg

--

)

X

1

1

1

)

\

}

8

I

I

11 F?282*

?

~~~~~~~~

21 2000 N,

wo |

[17

_

AW

22

\

TC

NINA TRAVLOU 1963

(7

|

W

FIG.1. Neolithic, Middle Helladic,and MycenaeanWells in the Agora Excavations

X I

Y

3

56

SARA A. IMMERWAHR

more recentlydiscoveredcomparativematerial.We now knowmuch more aboutvery late or FinalNeolithicof about3500-3000B.C.,notablyfrom the Kephalacemeteryon Keos, andthe laterdepositsfromthe Kitsoscave nearLauriumandthe Franchthicave. In establishingthe date of the Agora burialthere are basicallyfour avenues of investigation:(1) the form of the graveand the attendantburial,includingthe skeletal analysis,(2) the potteryofferings,(3) the fillingof the shaft, and (4) the relationship of this single isolatedburialto the Neolithic and chronologically both topographically materialrecoveredfrom the wells on the northwestslope of the Acropolis,whichalso containedskeletalmaterial. parallels, The form of the grave still remainspuzzlingand withoutcontemporary althoughthe "doublegrave"at Corinthand the rock-cuttombs at Manikain Euboia have been cited.6These, however,are clearlyof the EarlyBronzeAge and not particularlyearly, the Corinthgrave havingsome distinctiveEarlyHelladicII potterywhile potteryof the Manikagraves otherpiecesresemblethe Anatolian-andCycladic-inspired whichforeshadowEarlyHelladicIII.7In otherwords,the formof grave,rock-cutwitha deep shaft, would be more in accordwith the later date assignedthe burialin Agora XIIIand presentsus with the paradoxof a gap of 1000or more yearsfrom FinalNeolithicto the gravesat CorinthandManika.AlthoughDickinsonand Renfrewimplythat there was a traditionfor such tombs on the mainlandand in the Cyclades,I have not close.8Late Neolithiclikewiseis unproducfoundtheirinterveningparallelsparticularly tive of real parallels:cave burialsat Franchthiand at Alepotrypain the Mani,9pit cist and builtgraveson gravesat Lerna,10and the Kephalacemeteryof proto-Cycladic Keos,11in short nothingto compareto our three-meterdeep shaft.12Thus, the grave form is uniquefor this periodand has its closest parallelsat Corinthand at Manikain Euboiatowardthe end of the thirdmillennium. The skeleton,althoughcrushed,was completeand in its originalposition,an adult male between30 and 35 yearslyingon his rightside with head towardthe openingof 6By both Dickinson and Renfrew (footnote 4 above). T. W. Heermanceand G. D. Lord, "Pre-Mycenaean Graves in Corinth," AJA, ser. 2, 1, 1897, pp. 313-332; Manikagraves: G. A. Papavasileiou, kept' Etoa &pXar~vr&a/ov,Athens 1910. TnOevEvl' 7See Jeremy B. Rutter, CeramicChangein the AegeanEarlyBronzeAge (OccasionalPaper 5, Institute of Archaeology,Universityof California),Los Angeles 1979, pp. 4-6 and Table 2. 8The rock-cut tombs at Phylakopihave short horizontaldromoi and elaboratechambers (cf. Christos Doumas, EarlyBronzeAge BurialHabits in the Cyclades[SIMA 481, G6teborg 1977, p. 49 and fig. 35), while the Early Helladic tombs at Zygouries are irregularoval chambers only a meter below the surface (CarlW. Blegen, Zygouries,Cambridge,Mass. 1928, pp. 42-55). 9Franchthi:Hesperia42, 1973, pp. 277-282; Alepotrypa:AAA 4, 1971, pp. 12-26, figs. 15-17, pp. 289-394, figs. 3-5. '0Hesperia27, 1958, pp. 136-137, pl. 37:a-c; 28, 1959, p. 205, pl. 41:a, b. "Keos I, pp. 103-105. 12Thepossibilityof its having been dug as a well with the side chamberadded and more crudely cut for a burialseemed real enough to the excavator (Notebook E, p. 904) because of the copious amount of water pouringin, but it is difficultto see how this could have been accomplished,and the relativelygood preservationof skeleton and pots would argue against this. Likewise Shear's suggestion that the Middle Helladicpeople converted the grave into a well seems unlikely. Possiblythe watertablehas changed.

THEEARLIEST ATHENIAN GRAVE

57

the shaftand knees drawnup in a positionmuch like that of the Late Neolithicburials at Lernaand Alepotrypa.It was sent to the AmericanMuseumof NaturalHistoryin New York, where it was studiedby J. L. Angel and presentedas a Type A 1 "Basic White"modifiedin a Dinaric-Mediterranean (TypeF 1) direction,consistentwithother Neolithicskullsfromthe Agora.13 The crucialargumentfor a Final Neolithic date comes, however, from the two simple and crude pots depositedas offeringsin the grave.14Althoughat the time of Neowritingthey did not seem to have much in commonwith the more characteristic the and the well such as Red biconical fabrics from Burnished lithic shapes deposits, jars, incisedscoops, and decoratedcoarseware,15an impartialre-examinationsuggests that they are not inconsistentwith our Neolithicceramicrepertory.They have more in common, however, with the potteryfrom the Kephalagraves, and indeed it is the excellentand full publicationof Keos I that has led me to re-examineour much scrappiermaterialfor otherparallels. I offerhere new photographs and a somewhatfullerdescriptionof these two pots, whichshouldsupersedethe catalogueentriesfor nos. 384 and 385 in AgoraXIII. P 6072 = AgoraXIII, no. 384

P1.9:b and d

Flaringopen bowl on high foot ring, very badly

At foot of skeleton. H. 0.095 m. (without handies), D. 0.102 to 0.112 m. Entire upper part of both handles restored in plaster as well as fragments from side walls. Fabriccoarse, but not particularlyheavy, reddish with quartziteinclusions. Orange-redslip showing traces of burnishingstrokes interior and exterior without producing a pattern or even a lustrous surface. Surface considerably coated with calcite accretionsfrom long burial.'6

shaped, the mouth an irregularoval from handle to handle, the ring foot varying in height from 0.02 to 0.025 m. The handles, although restored above rim, must have been similar to reconstruction, that is, a loop handle rising from plastic ridges attachedto side walls, these also very irregular, one pair vertical and close together, the other runningobliquely.

The Middle Helladicparallelscited in Agora XIII were based primarilyon the generalcrudenessof fabricand the risingloop handle,17but occurredon much smaller pots withoutburnishingand withoutattachedribs from the handles.These last two 13J. LawrenceAngel, "SkeletalMaterialfrom Attica," Hesperia14, 1945, pp. 279-363, here no. 2 (27 AA), pp. 291-292, fig. 1, pl. 41:2. Whether a later dating would be ruled out is hard to say, but six out of ten skulls at Kephalawere also Basic White "large-headed,deep-jawedand rugged" as was our Athenian (see Angel in Keos I, p. 137) and the Lerna Late Neolithic burialat JC-1 producedthe skull of a woman of much the same type (J. LawrenceAngel, Lerna, II, ThePeople, Princeton 1971, pp. 40-41, no. 242, pl. I). By Lerna IV and V new elements had been introducedwhich are visible even to the untrainedeye in some of the longer and high-headed Type D Nordic-Iranianskulls from the Middle Helladic cist-grave cemeteries (Lerna II, pp. 41-66 and 111). "Agora XIII, nos. 384 and 385, p. 93 and pls. 27 and 71. "Agora XIII, pp. 22-25, nos. 1-18, pls. 1, 2 (Red Burnishedjars); pp. 35-37, nos. 111-124, pl. 8 (scoops); pp. 38-42, nos. 130-170, pls. 9-11 (decoratedcoarse ware). 16I have profitedfrom looking again at these two pots and discussing their fabric and condition with Stephen P. Koob, Conservatorat the Agora. 17HettyGoldman, Excavationsat Eutresisin Boeotia, Cambridge,Mass. 1931, p. 177, fig. 245:5 and 6.

58

SARA A. IMMERWAHR

features,as well as the high foot ring, all occuron our NeolithicRed Burnishedbowls, The fabric,however,of some of our plain but usuallymuchmore carefullyexecuted.18 On the other hand, we is not dissimilar.19 red slip coarseand semicoarsewarewith a have no parallelfor the risingloop handlesas restoredon our bowl, nor do thereseem to be any amongthe Kephalabowls,but it wouldbe difficultto suggestan alternative restoration.A generalparallel,whichhas been cited, is the spreadingopen bowl on a ringbase witha tab on rim and a horizontallypiercedlug on the wallfromGrave20 at which,althoughit has tracesof crustedred on the exteriorand is considerKephala,20 ablymoreflaring,is as close a parallelas I have found. The secondbowlfromour graveis bettermadeand is perhapssomewhatearlier,to judgefromthe factthatit was mendedwhendepositedas a graveoffering. P 6073 = AgoraXIII, no. 385

P1.9:a and c

At head of skeleton. D. 0.175 m. Fragments missing from wall, restored in plaster. Somewhat better levigatedclay with quartziteinclusions,fired grayish tan; dark brown to black burnished slip interiorand exterior, much flakedand crackled.

Flaring hemisphericalbowl with flattened bottom and outturned lip. Double-scallopedtab rises from rim on one side; second opposite tab restored. Vase had been broken and mended before being deposited in grave; three pairs of drilled holes alongsidefracture. No Middle Helladicparallelscited.

Althoughat first glance the fabriclooks quite differentfrom our characteristic Neolithicwares, withoutthe reddishcolor of the Red Burnishedor the more solidly grayto blackof the GrayBurnished,a fragmentfromone of our Neolithicwells has an identicalcrackleddarkburnishedslip and belongedto a deep bowl with outturnedlip that couldwell have resembledthe one from the grave.21In generalshapeour bowl is closerto the Kephalaexamplecited above, but it lacksthe ring foot. Tab handlesare also frequentat that site, sometimesnotched or scallopedas on our example, and wherethe bowl is complete,as in Kephalano. 104, there is only one tab per rim (suggestingperhapsthatthe restorationof a secondhere is incorrect).22Tabhandlesmaybe a Cycladicfeature, as rathersimilarexamplesoccur at Saliagosin somewhatearlier Agorahandle,althoughotherforms Piercedlugs are the more characteristic material.23 at least two to now examplesof tabs, both of which and it is recognize possible occur, are smallfragmentsbrokenfromthe rim with only a smallamountof wall preserved.24 '8Cf. AgoraXIII, nos. 33-35, 55, pls. 4 and 68. 19Cf.AgoraXIII, nos. 127, 171, 193, 196, pls. 9, 12, and 13. 20KeosI, p. 65, no. 104, pls. 28 and 76, cited by Renfrew, op. cit. (footnote 4 above), p. 70. "Agora XIII, no. 78, pls. 6 and 69. 22KeosI, p. 14, pl. 29:P (= pl. 75:AX); 29:N (= pl. 75:AP); 38:P (= pl. 85:AT and AV), the last almost identicalto ours. at SaliagosnearAntiparos,Oxford 1968, fig. 58, nos. 6, 23J. D. Evans and Colin Renfrew, Excavations 9, 10 and pl. XXXI:a.For chronologicalrelationshipof Saliagosto Kephalasee Keos I, pp. 98-99. 24AgoraXIII, nos. 128 and 129, pls. 9 and 69. On p. 38 there was some doubt expressed as to how these small fragmentsshould be restored.

THEEARLIEST ATHENIAN GRAVE

59

Mendholes occurat Kephalain completeor nearlycompletepots, and probablythis is also the explanationfor the boredholes occurringin isolationon certainAgorasherds.25 While the grave offeringsnow speakunequivocallyfor a Final Neolithicdate for the burial,the fill of the shaftsuggesteda differentconclusionand was one of the main argumentsfor assigningthe graveto MiddleHelladic.Over one hundredfragmentsof Grayand YellowMinyan,some MiddleHelladicplainand Mattpaintedwere dominant in the pottery,as well as a few piecesof earlierEH II slippedand polished,but much was too badlywornand smallto be identifiedand some Neolithicmay have been present.26Therewas also a lot of bedrock,including"piecesthe size of a man'shead,"and relativelylittle potteryfrom a three-meterdeep shaft.The only explanationthat seems consistentwiththe graveofferingsis to concludethatthe shaftwas dug in orderto make the burialand was closed immediatelythereafterwith the bedrockand earthfrom the excavation,which may have includeda few Neolithic sherds. A gradualsubsidence must have led to the laterfilling-upof the hole with whateverrefusewas available,as in the case of our other MiddleHelladicbothroiand gullies.27Perhapsmore pertinent to the fill of our shaftis the existenceof a well-traveledroadalongthe West Side of the Agora, which, from the evidence of stratifieddepositsnear the Tholos, surely went backat least to EarlyHelladictimes.28The conditionof much of the smallwornscraps of unidentifiablepotteryfrom our shaft stronglysuggestsroad metal, as does a stone arrowhead or point (no. 386) whichshowsconsiderablesecondarywear.29 Althoughit now seems certainthat the Metroonburialmust be reassignedto the FinalNeolithicperiod,this redatingraisesa topographical problem,whichmay have a chronologicalsignificance.WhereasMiddle Helladicpotterywas plentifulthroughout the Agora area whereverbedrockwas reached,30the recognizableNeolithic material 25KeosI, p. 10 and pls. 79 (no. 170) and 80 (no. 136). Cf. AgoraXIII, nos. 24, 56 and 100. 26Inrestudyingthis deposit (September 1977), I carefullysorted, counted and bagged the recognizable fragmentsand put them in one tin; the residue, consisting of insignificantscrapsscarcelylargerthan 2 sq. cms., coarse, well worn, and without distinguishingfeatures, came to about one half tin. The excavator's notebook (E, p. 901) speaks of a "uniform filling with rocks the size of a man's head and oply a few sherds with the majoritycoming from sifting earth" and thus provides no recordof where within the threemeter depth the Gray Minyanand clearlyidentifiableMH materialcame from. "Agora XIII, pp. 51-52 (depositsat B 21:15, G 19:1, and R 21:4). 28AgoraXIII, pp. 52 and 113. Cf. plan, pl. 90 for location of ancient roads in Agora going back to Bronze Age and possibly Neolithic times. Inventoried sherds from West Side of Agora are nos. 241 and 243 (EH) and nos. 283, 292, and 346 (MH). 29No. 386 (Agora XIII, p. 93 and pl. 28) has recently been re-examined by Steven R. Diamant, who has identifiedthe materialas black flint ratherthan obsidianand the object as a rough point more like the scraperfrom Kephala (Keos I, p. 42, no. 90, pls. 25 and 70) than a tanged arrowheadlike those from Saliagos (op. cit. [footnote 23 above], fig. 66 and pl. XXXVII) or from MH deposits cited in Agora XIII. See also S. Diamant, "A Barbedand TangedObsidianPoint from Marathon,"JFA 4, 1977, pp. 381-386. 30The presence of Minyan and Mattpaintedwas everywhere noted by the excavators, and much is preserved in basement storage. The inventoried sherds referred to above, footnote 28, represent only a small portion of what was available, and several pieces found their way into the fill around Geometric graves near the Tholos (AgoraXIII, nos. 283, 292, 346).

60

SARA A. IMMERWAHR

came almostentirelyfrom the 20 wells at the northwestcornerof the Acropolisin the areaof the Klepsydra,withonly a few items in our catalogue-andthese often of a less destructiblenaturelike the marblestatuette(no. 219) fromthe areaof the EleusinionAt the time of preparingAgoraXIII, I found out of context in the Agora proper.31 envisagedNeolithic occupancyas restrictedto the slopes of the Acropolis,and the eccentricpositionof this burialon the West Side of the Agora,some distancefromthe Acropolis,was certainlya contributingargumentin rejectinga Neolithic date. The picture,however,is not quiteso clearcut today,for it is becomingapparentthat accidents of excavation,or preservationof remains,may have distortedit. In 1965 during tests in the area of the MiddleStoa a shallowwell was discoveredat M stratigraphic 14:1.32 Sinceit yieldedonly a half-dozenscrapsof potteryand since this authorwas not on hand,it was not includedin AgoraXIII,althoughits locationis markedon the plan, pl. 91, whereit was given a MiddleHelladicdesignation.Throughthe kindnessof John McK.CampII, who was curiousas to the date of the well, my attentionhas just now been calledto this depositand to its one significantpotteryfragment(P 27425), which Far from being an doubtlessled to an EarlyHelladicor MiddleHelladicdesignation.33 exampleof EH IIIto earlyMH incised"Adriaticware",the fragmentin questionrepresents our best preserved"scoop"fragmentof the FinalNeolithicperiodwith parallels among the materialfrom our wells and from Kephala.I publishit here with photographsanddrawings.34 P 27425. Fragmentof a scoop.

P1.9:e, f

From shallow well between Middle and South Stoas at M 14:1. P. H. 0.07 m., W. of handle 0.04 m. Pinkish buff clay with some grit; possible traces of red surfacingand perhapscream slip in incisions. Three joining fragmentsbroken all around, but clearly from the back of a scoop bowl with the

spring and beginningof the band handle, the end of which appears on the interior as a little excrescence. Broadincised lines (made by a blunt instrument as in other examples of this class): oblique parallel lines on handle, a zigzag line between double parallel lines top and bottom extending horizontally on bowl.

Althoughthe handleis less broadand much simplerin its decorationthan on the Seskloscoopor Kephalano. 98,35the generallayoutof the decorationon the bowlwith 310f the 218 items in our Neolithic potterycatalogue,only three came from the Agora proper (no. 48, Middle Stoa building fill; nos. 36 and 66, miscellaneous finds), the rest from the 20 wells. The marble statuette (no. 219) from the area of the Eleusinioncame from "demolition marbles",and the celt no. 222 (as well as other inventoried, but uncatalogued,stone objects) was a "miscellaneousfind." 32Hesperia35, 1966, p. 45 and pl. 16:a (where well is referred to as EH). Mention is made of the dressing-downof bedrock over a large area deep beneath the Middle Stoa, which certainly would have obliteratedmost prehistoricmaterialand which apparentlyall but destroyedour well at M 14:1. 33Parallelscited on the inventorycard are to fragmentsfrom the North Slope (Hazel D. Hansen, "The PrehistoricPottery on the North Slope of the Acropolis, 1937," Hesperia6, 1937, p. 545, fig. 4), clearlysocalled "Adriaticware" which belongs to an EH III to MH I horizon contemporarywith Lerna IV. The shapes are very differentfrom P 27425, and the incisions scratchierwith sharperwalls. 34Drawingand profileby ChristopherA. Pfaff, regularmember of the ASCS at Athens, 1980/81. 35Cf.AgoraXIII, pl. 8 (Sesklo scoop) and Keos I, pl. 82.

THE EARLIESTATHENIAN GRAVE

61

two parallellines running almost horizontallyfrom the bottom of the handle and with a pattern of oblique lines above-here a simple zigzag rather than the interlockingtriangles and meandroid squares of the Sesklo scoop-shows an awareness of the more highly decorated examples, present in a few scraps from the Agora wells.36Our closest parallelfor the new fragment is no. 120 from the well at U 24:2 which preserves the back of the bowl and the lower attachment of a broad band handle with two diagonal incisions.37Whether these scoops were all made locally, as Coleman suggests,38seems to me highly doubtful, given the complexity of the form, and I would favor a single center of export (Thessaly?) for the more highly decorated examples with local imitation for the cruderspecimens. Whatever the origin and use of these enigmatic vessels, the occurrence of one in an archaeologicalcontext in the central Agora extends the area of Final Neolithic occupancy from the slopes of the Acropolis toward the Kolonos Agoraios and the burial at I 9:2. Whether this has a chronologicalimplication suggesting a later spread from the Acropolis slopes is difficult to assess and can only be determined by the relationshipof the Metroon burialand its pottery to that of the 20 wells on the northwest slope. The shaft of our grave is dug in much the same way as the wells, some of which are much deeper,39but most within the three- to four-meter range with some even shallower. They too are circular, roughly cut with a diameter ranging from 0.75 m. to somewhat over a meter, and although some of the shallower ones showed no presence of water and were apparentlyabandonedwhen an intractableboulder was encountered, most were clearly wells with a good water supply. The fill in no case could be clearly divided into use and dump fill, but it seems likely that the more or less complete Red Burnishedjars40should be connected with the use of the wells. Although these wells were dug carefullyby the late Arthur W. Parsons over 40 years ago with the materialin most cases separatedaccordingto upper and lower levels, the dump did not seem to be stratified.41This fill also included animal bones (bos, capra,ovis, sus, cervus) and shells (cardiummurex) attesting to near-by habitation, and at least five wells yielded human skeletal material,includingthree skulls from one well, one each from two others.42This 36Nos.111, 116, 117 and 121 (cf. AgoraXIII, p. 12, note 64). 37Agora XIII, pl. 8, no. 120. 38See Keos I, pp. 16-17, Form C 1, and p. 101 for discussionof comparativematerial. 39Thedeepest Neolithic well at U 24:2 was 7.70 m., two others at S 27:4 and U 24:3 were over 5 meters deep. 40AgoraXIII, p. 2, nos. 1, 2, 3, 5. 41In1967 when I had finished the manuscriptof Agora XIII, I had come to the conclusion that it was impossibleto establisha chronologicalsequence for the wells, or individuallywithin their respectivedeposits, but today with more stratifieddeposits of Final Neolithic in the Kitsos cave, at Thorikos and in Euboia (see below, footnote 46) it may be possible to separateout earlier and later elements, and particularlyto establishthe relationshipof our best Red Burnishedbiconicaljars to coarserwares. 42Threeskulls (31, 32, and 33 AA) from the well at U 26:2 were publishedby Angel, op. cit. (footnote 13 above), pp. 293-294, nos. 3, 4, 5, pl. XLI; one skull (1 AA) from U 25:1, Angel, p. 291, no. 1, pl. XLI. A fifth (7 AA) from the well at S 27:4 was not included. A human humerus was found in each of two wells (S 27:5 and T 26:3).

62

SARAA. IMMERWAHR

is surely an indication that burials must have taken place in the area, probablynot in any of these shafts, which did not yield complete skeletons and seem clearly to have been cut as wells, but perhapsin the near-by caves of the North Slope. The association of burialsand domestic remains in, or in the vicinity of, caves is characteristicof Final Neolithic deposits at Kitsos, Franchthi, and Alepotrypa on the Greek mainland, and differs from the ordered separation of domestic and cemetery areas at Kephala on Keos.43 On the other hand the close relationship of the Agora Neolithic to Kephala is clearly attested in such pottery forms as the scoops, coarse-warepithoi with specialized lug handles and ornamental plastic bands, the head of a terracottafigurine of protoCycladictype (Agora XIII, no. 220), and numerous other details enumerated by Coleman.44There is not identity of culture, however, and our Red Burnished with painted patternsseems to have more in common with pottery from the Kitsos cave45than with most of the PatternBurnishedpotteryfrom Kephala.Parallelsalso exist in Euboia46and in the Red Monochrome ware from PalaiaKokkinianear Piraeus.47It is perhapssafe to conclude with Coleman that the Agora material covers a longer period of time. Where the Metroon burial fits into this sequence is not yet clear. Although its offerings have parallelswith the Kephalapottery, as well as with some of the pieces from the wells, it seems to me to belong to a slightly later horizon, as its location fartherfrom the Acropolis slopes and caves might indicate. Until Final Neolithic and the transition to Early Bronze on the mainlandhas been thoroughlydocumented in stratifiedsites like Kephala on Keos, a more precise dating seems impossible. We can now, however, surely reject a Middle Helladic date. New material has vindicated Shear's original proposal of Late Neolithic or "Subneolithic",and the Agora grave with its skeleton 27 AA becomes the earliest preservedburialof an Athenian. SARA A. IMMERWAHR AMERICANSCHOOLOF CLASSICALSTUDIES AT ATHENS

43KeosI, pp. 104-105. "Keos I, pp. 100-102. 45BCH93, 1969, p. 965, fig. 17; BCH 95, 1971, p. 711, fig. 22; BCH 96, 1972, p. 823, fig. 7. Cf. with these AgoraXIII, nos. 87-92. 461n his survey of the Karystosarea of southern Euboia Donald R. Keller of IndianaUniversity has found significantFinal Neolithic remains at the site of Plakariat the west end of the bay. Through his kindness I have seen this materialin the KarystosMuseum, and he has likewise comparedit with that in the Agora. In our opinion his Red Burnishedbowls have more in common with the Pattern Burnished material from Kephala, although a few parallelsdo exist with Athens. Clearly, however, all three sites belong to the same chronologicaland culturalhorizon. 1951 [19521, p. 104, figs. 12, 13: cf. Agora XIII, no. 125; p. 106, fig. 19: cf. Agora XIII, 4711paKTI-Ka, no. 211. I do not share Coleman's opinion that this site must be later than all the Agora Neolithic material (Keos I, p. 104).

PLATE

a. P 6073

b. P 6072

c. P 6073

d. P 6072

e, f. P 27425

Scale 1:2

9

AN ATHENIANSTELEIN PRINCETON (PLATE 10)

SMALLGRAVESTELE,longin the collectionof The ArtMuseum,PrincetonUniversity,is finallymakingits debut.It is a slabof Pentelicmarbleof the conventional, slightlytaperedshape,roughpickedon the backandshowingthe marksof a clawchiselon bothsidesandon the unsculptured areasof the frontface (P1.10).1The irregularbreakat the top risesdiagonallyfromthe left, takingpartof the reliefwithit, and then levels off horizontally.A hint, on the rightframe,of the usualinwardprojection(vestigialcapital) suggeststhatthishorizontalpartof the breakcoincideswiththe upperedgeof the recessed relief.Withthe topof the stelewentanyinscriptionthatidentifiedthe deceased.The crest was probablya simplepediment.The reliefshowsthe deceasedstandingon the left and her littlemaidat the right.The womanwearsa chitonthatis girdledacrossthe overfold andfurthersecuredby crossstraps;fallingbehindher is a lightcloakor veil thatshe holds in her left handandthatprobablycoveredher head.The maid,headin profileto the left, raisesa leaf-shapedfantowardher mistressandholdsa box againsther bodywithher left fall fromher shoulders. hand;her chitonandsomewhatlongerundergarment The stele is a productof an Athenianworkshopthat turnedout grave markers, recordsof decrees,and votive reliefsin the earlypartof the 4th centuryB.C. The style fallsinto placewith that of the Athenaon the well-knowntreasuryinscriptionof 398/7 B.C. andrelatedreliefs.2 In his notebook, on the page headed 1899/1900,ProfessorAllan Marquand,in chargeof the Departmentof Art and Archaeologyand of The Art Museumat PrinceA

ton University, made the following entry: Mrs. E. SandozpresentsAthenianstele. In the

Museum'scurrentinventorybook, begun in 1923 as a replacementof an old list, the stele is enteredbrieflyunderthe number204 with the additionalnote that it was purchased in Athens in 1850. The statementabout proveniencemust either dependon some recordno longerin the Museum'sfiles or reflectsomebody'smemoryat the time of entry (the roundnumberof 1850suggestsa recalledapproximation). ErnestSandoz,surveyorand civil engineerin Princeton,was the nephewof the geographerArnoldGuyot (when Guyot came to Princetonfrom Switzerlandin 1854 'Preserved height, 0.77 m. Width, 0.37 m. at bottom, diminishingto 0.355 m. at the level where the border to the spectator'sleft is broken. The thickness of the slab diminishes from 0.085 m. at the bottom to 0.075 m. at the highest preserved point of the right edge. The width of the border behind the maid diminishesfrom 0.04 to 0.033 m. in a rise of 0.32 m. The height of the unsculpturedlower face is 0.425 m. As the photographindicates, there are numerous scratchesand abrasionson the surface;the woman's left hand is chipped, leaving a surface that is relatively modern in its freshness, but not recent (a similar surface is on the lowest point of the break at the left edge near the woman's elbow). 2J. Svoronos, Das Athenischer Nationalmuseum,Athens 1908-1937, pl. CVII, no. 1367 and pp. 601604, for the inscriptionof 398/7; also R. E. Binnebbssel, Studienzu den attischenUrkundenreliefs des 5. und 4. Jahrhunderts, Meeuws 1932, no. 24.

64

FRANCES FOLLIN JONES

with his motherand threesisters,the nephewwas partof the familygroup).'Sandoz's secondwife wasHelen Crabbe,daughterof RearAdmiralThomasCrabbeanddonorof the stele. Her fatherwas the link with Athens. After a careerin the United States Navy, he died in Princetonin 1872 and was buriedin PrincetonCemetery;three volumes of archivalmaterial,given by Ernest Sandoz,are in the PrincetonUniversity Libraryand providea clue to the probableoccasionwhen Crabbeacquiredthe stele. The thirdof the three volumes, coveringthe years 1851-1855, containshandwritten copies of officialletters (mostly to the Secretaryof the Navy) by the then Captain Crabbewho hadbeen sent to the Mediterranean in commandof the U.S. SteamFrigate San Jacinto.The frigatearrivedat Cadizon March25, 1852, twenty-twodays out of Norfolk,Virginia,in whatCrabberefersto as "fairtime", andfinallyreturnedto Philadelphiaon July 5, 1853.The entirevoyagewas plaguedby mechanicaldifficultiesand mishaps,involvingseverallayoversfor repairs(one for more than two monthsat Trieste), leadingCrabbeto indicateto his superiorsthat the frigatewas a nationalembarrassment.The vessel was at Piraeusat least from August 1st to 18th, 1852, while en route from Constantinopleto Trieste.Crabbementionsthe ship's "beingassignedto convey the Hon. GeorgeP. Marshand familyto Triestewhen his [Marsh's]business has been concludedwith GrecianGovernment,"so one assumesthat the captainhad a bit of leisurewhile in port. The officialnatureof the correspondence precludedcomment on personaltripsashore,but surelythis wouldhave been the time when Captain Crabbeacquiredthe stele whichwe presenthere as a tributeto a formernavalperson whosedistinguishedcareerhas centereduponAthens. FRANCES FOLLIN JONES THE ART MUSEUM UNIVERSITY PRINCETON

Princeton,NJ 08544 3I am gratefulto Mrs. Robert Gunning of the HistoricalSociety of Princeton,to Mr. Earle Coleman of PrincetonUniversity Archives, and to Mr. CharlesE. Greene of Special Collections, Princeton University Library, for their help in supplying information and documentation about Mr. and Mrs. Sandoz and Thomas Crabbe.Crabbe'sthree volumes are in SpecialCollectionsunder the call number PB1184.27.

PLATE

10

ki:~~~~~~~~~~~~~~~~~~~~~~~~~~~'.

41

14;

IC,~~~~~~~~~~~~~~'

FRANCES F. Jo

FRANCES

F.

JONES:

,;.:AN' ATHENIAN STELE'iNPRINCETON

AN

t~

ATHENIAN

STELE

IN

PRINCETON

THE ANCIENTIMAGE OF ATHENA POLIAS (PLATE11) But the most holy object, that was so considered by all many years before the unification of the demes, is the image of Athena which is on what is now called the Acropolis, but in early days the Polis. A legend concerningit says that it fell from heaven.

T

HUS PAUSANIAS1on the old olive-woodimageof AthenaPoliasin the Erechtheion.This was the venerableimagewhichwas dressedin a newlywoven peplos and which,as if a every four yearsin the culminatingrite of the GreaterPanathenaia,2 protectivetalismanof the city, was taken aboardship with the Atheniansthemselves duringthe temporaryevacuationof Atticain 480 B.C.3Yet for all its primacyin the state religionof Athens, Pausaniassays nothingaboutthe image'sappearanceand thus has left us to reconstructit from a numberof scatteredreferencesand possiblereflections in the minorvisualarts.4 For more thana centurynow, therehas been notoriouslylittle scholarlyconsensus as to whichof this evidenceis necessarilyrelevant,much less as to how it shouldbe combined.Discussionhas polarizedaroundtwo opposingviews.Most earlierscholarship arguedthat the imagehad the form of an armed,fightingAthena,like the conventional, standingPalladiontype of Athenaor like the stridingAthenaspicturedon PanatheA. Furtwdngler, naic amphoras.5 however,proposedthat the image was unarmedand seated;'and in 1908this view was defendedby A. Frickenhaus,who calledattentionto epigraphicalevidence that the image wore a diadem and held a gold phiale in one 1i.26.6:

eo-rV ToV

TO 8E ayLaYTaToVEv KOLJNL7TOXXOLV7TpOTEpOP VOIU(TEV ETETLV J) ioVV7JX\ov aio% TJWv 87.uwv ' 'A07qva& aAya\A~a El V V aKpo0ToXEL, ToTE 8% 6voa~oA 8EEs aUTO EXEL E7ELJ) EK GroE ?r7.r oipavov . Translationof W. H. S. Jones, PausaniasI, Loeb ClassicalLibrary,Cambridge,Mass. and

London 1918, p. 137, with minor alterations. The following specialabbreviationswill be used in this article: Herington = C. J. Herington,AthenaParthenosand AthenaPolias, Manchester1955 Shear = T. L. Shear, Jr., Hesperia,Suppl. XVII, Kallias of Sphettosand the Revolt of Athens in 286 B.C., Princeton 1978 2Herington,pp. 17, 32-33, with references. On the frequencywith which the peplos was dedicated, now see Shear, p. 36, note 89. 3Plutarch, Themistokles,10 (quoting Kleidemos), which refers to the loss of TO yOPolYELOl) a7To T7g 6EOvj TOv ayaX,\LaToc and the subsequentsearch for it in the baggagebeing gatheredat the Peiraieusduring the evacuation.The gorgoneionwas of gold; see footnote 18 below. 4The literaryand most epigraphicalreferences are collected in 0. Jahn and A. Michaelis, Arx Athenarum, 3rd ed., Bonn 1901, pp. 68-69. 'E.g. 0. Jahn, De antiquissimisMinervaesimulacrisAtticis, Bonn 1866; L. R. Farnell, The Cults of the GreekStates, Oxford 1896, I, pp. 334-337; J. G. Fraser, Pausanias'sDescriptionof Greece,London 1898, II, p. 341; E. Petersen, Die Burgtempel derAthenaia,Berlin 1907, pp. 40-60. 6A. Furtwdngler,"Athene in der Kunst," in AusfihrlichesLexikongriechischenund ramischenMythologie, Leipzig 1884-1890, W. H. Roscher, ed., I, cols. 687-689.

66

JOHNH. KROLL

hand.'Notingthatthe only fourArchaicportrayals of Athenaholdinga phialeshowher in a seated posture(e.g., PI. 11:a), that two of these depicther with her helmet removed, and that a copiousnumberof Archaicterracottasfrom the Acropolisportraya seated,unarmedAthenacladin a peplos,8Frickenhausarguedthatall these representations were derivedfrom the ancientcult statue, whichthereforewas of a seatedgoddess, unhelmeted,andholdingout a phiale.But not all authoritieswereconvinced;and, the decisive detail of the gold phiale notwithstanding,the identificationof the cult statueas that of a fightingAthenaenjoyeda briefrevival9beforeit was finallylaid to rest by C. J. Heringtonin the most criticaland influentialreview of the problemto date.10AlthoughHeringtoncomes out on the side of Frickenhaus'Sitzbild,his endorsement is farfrom unqualified;for, as he says, the Frickenhausreconstruction is hazardously dependenton "argumentsfrom the minor arts of a periodwhen most of the minor artists ... were more likely to reproducethe living, or immortal, goddess as their

felt her thana statueonly a decadeor two old.""1 contemporaries The truthof these wordsis broughthome when one examinesthe most elaborate of Frickenhaus'four representationsof a seated Athena holdinga phiale (P1. 11:a). Here the seatedgoddessfacesan altarand templewhile a priestesspreparesa sacrifice. As in the other three representations, there is no statue base beneathAthena'sstool nor any other detailof style or iconographyto suggestthat the Athenais a statue.On thatshe is seatedat the altaroutsidethe templeand on the contrary,the circumstances a portablestool ratherthana throneimplythat it is Athenain personwho has come to partakeof the offerings;and, as any banqueterwould, she has sat down, removedher of the gods.12Once helmet, and extendedher cup, the phiale,the normaldrinking-cup it is recognizedthat a seatedAthenawith phialeis in effect a banquetingAthena, the associationbetween postureand vessel is easily understoodwithout referenceto a putativecult statue.As for the seated terracottaAthenasfrom the Acropolis,there is 7A. Frickenhaus, "Das Athenabild des alten Tempels in Athen," AthMitt33, 1908, pp. 17-32. Cf. idem, TirynsI, Athens 1912, p. 110, note 1. For the epigraphicalevidence, see footnote 18 below. 8SeatedAthena with phiale:black-figuredkalpis, ABV, p. 393, no. 20 = P1. 11:a, reproducedfrom E. Berlin 1840-58, IV, pl. 242; red-figuredsherd by Myson from the AcropGerhard,AuserleseneVasenbilder, olis, AR V2, p. 240, no. 42 = B. Graef and E. Langlotz, Die antikenVasenvonderAkropoliszu Athen, Berlin 1925-33, II, pl. 72; black-figuredlekythos, Athens N.M., no. P 1138 = Frickenhaus, op. cit., figs. 3, 4; terracottarelief from the Acropolis, ibid., fig. 1 = D. Brooke in S. Casson, Catalogueof the Acropolis Museum,Cambridge1921, II, pp. 419-420, nos. 1337, 1338. Seated terracottaAthenas: ibid., pp. 330-332, 355-369. 9E.g., M. Bieber, "Two Attic Black-figuredLekythoi in Buffalo," AJA 48, 1944, pp. 124-129; H. L. Lorimer, Homerand the Monuments,London 1950, pp. 445-449; D. von Bothmer, "A PanathenaicAmphora," BMMA 12, 1953, pp. 52-56. 0Herington,pp. 22-26. 'Ibid., p. 24. 12H.Luschey, s. v., 4LaX7J, RE, Suppl. VII, 1950, col. 1030, noting that in the visual arts the gods are never depicted drinkingfrom stemmed kylikes. Luschey suggests that the gods employed the phiale because of its sacramental,hence godly, character.

THEANCIENT IMAGEOFATHENA POLIAS

67

simplyno internalevidenceto decide their possiblerelevanceto the old Poliasstatue one way or the other.13Since the case for a seated imageof the Poliasrests solely on of seatedAthenawith phialeare relethe convictionthat they and the representations vant, it is hardto feel muchenthusiasmfor it. To judge from the literarytestimonia,one would expect, conversely,that the imagewas in fact standing.Aristophanesrefersto AthenaPoliasand the makingof her peplos at Birds, 827-828 and to Athena as fully armed and standing (iravoiroXtav

two lines later;but as Heringtondemonstratesin detail,these lines are best understoodas allusionsto the city goddessherselfand need not applyto a statue (or statues) of her.14Two other relevantpassages,however,are not easily dismissed. Athenagoras,Legatio,17.3, attributesthreecult statuesto the sculptorEndoios:the Artemis in Ephesos,the old olive-woodimageof Athena,and the SeatedAthena.15Since no locationfor the two Athenasis specified,they are surelyimageson the Athenian Acropolis,wherePausanias(i.26.4)saw a seatedAthenaby Endoios.The juxtaposition of the old olive-woodAthena with the seated one implies that the formerwas not seated.The secondpassageis fromStrabo,xii.1.41,a discussionof whetherthe Trojan statue of Athenaat Iliadvi.302-303 was a seated or standingfigure.Homerassuredly thoughtof it as seated since the women of Ilion placea votive peplos"on its knees" (Eft yoi'vao-tv). But some ancient commentatorsassumed that it was similarto or identicalwith the famousTrojanPalladion,an uprightstatue of Athena, and that the peplosmust thereforehave been placed"beside"the knees of the image.Straboargues for the seatedinterpretation, in partbecause"manyancientwoodenstatuesof Athena EOT7'K

EXovO-a)

'3Their relevance is emphaticallydenied by Brooke, op. cit. (footnote 8 above), pp. 330-332, but is accepted, althoughprovisionally,by R. A. Higgins, GreekTerracottas,London 1967, p. 72. 14Herington,pp. 24-26. 15To gv yap 4v 'E4ocx Tcar 'ApTE'WU80o Kat to trgs 'A6rjva& (.uaikov 8U 'Affqkav ta&O77 yap Wk ol .VcTTLKWTEpOVovTrw yapt) To aT7J' eXaia' to vakatoa v Kat re v Ka6-jgirrjv 'Ev8o0o0 ElyacoaTo UaO7JT7J AaLMkov. "Endoios, a disciple of Daedalus, made the statue of Artemis in Ephesus and the ancient olive

statue of Athene (or rather of Athela; for she is Athela, the unsuckled, as those ... the more mystical sense ... ) and the Seated Athena." Text and translationof W. R. Schoedel, Athenagoras,Legatioand De Oxford 1972, pp. 34-36. Resurrectione, Following the edition of J. Geffcken (Zwei griechischeApologeten,Leipzig-Berlin1907), Herington (p. 24, note 1, and pp. 69-70) argues that the text is too defective at this point to be admitted as evidence. But it is clear from Schoedel's more recent edition that the only textual problem lies within the parenthesized digression (which itself is largelyclarifiedby paragraph20.2 of the Legatio).Herington (p. 70) further objects that Athenagoras is "wildly inaccurate"in his attributionsof statues to name artists, although Heringtonis able to point to only one such inaccuracy.I have not been able to consult G. Botti, "Atenagora quale fonte per la storia dell'arte," Didaskaleion4, 1915, pp. 395-417; but Schoedel, who has, writes (op. cit., p. xx), "that Athenagoras'informationon the history of art, though not profound, is ... generally reliable."Certainlythere is nothing suspect about Athenagoras'attributionsto Endoios. On the authority of Mucianus, Pliny (N.H. xvi.213-214) also names Endoios as the sculptor of the Ephesian Artemis. And Pausanias(i.26.4) quotes from the inscriptionon the base of the seated Athena on the Acropolisthat Endoios was the maker (cf. A. E. Raubitschek,Dedicationsfrom the AthenianAkropolis,Cambridge,Mass. 1949, pp. 491-492). In supportof Athenagoras'attributionof the Polias to Endoios, see below.

JOHN H. KROLL

68

are seen to be seated, such as those in Phokaia,Massilia,Rome, Chios, and several The ancientxoanonof Athenain Athens is conspicuouslyabsentfrom otherplaces."16 this list; and, althoughHeringtoncountersthat the list "seems to be haphazardand the list is undeniablya learnedone and may be less does not claimto be complete,"17 thanwe canjudge.The Poliasimageon the Acropoliswas very well known haphazard in antiquityand was includedin both Plutarch'sand Philostratos'accountsof the most ancientcult statuesin Greece (see below). Had it been seated, and had Strabomentionedit, it wouldhave stoodfirstin his list andwouldhave strengthenedhis argument Thathe does not mentionit is less likelyto be becauseof any ignorance immeasurably. or carelessnesson Strabo'spartthan because,as we have seen from Athenagoras,it probablywas standing. Anotherproblemregardingthe appearanceof the Poliasconcernsthe locationof the image'sgold owl. Ourknowledgeof the owl, like thatof the phiale,comesfromthe inventoriesof the statue'spreciousornamentsthat were recordedby the Treasurersof Athenain severaltraditioinscriptionsof the late 370's and early360's. In the sections of these inscriptionsthat cataloguethe valuablesin the apXaqo' VEC0;, i.e., the Erechtheion, the statue's ornamentsare listed in unvaryingorder as "a diademthat the goddesswears,the earringsthat the goddesswears,a bandthat the goddesswearson her neck, five necklaces,a gold owl, a gold aegis, a gold gorgoneion,and a gold phiale that she holds in her hand."18As throughoutthese traditiocatalogues,each objectis Thus the diadem,earrings,and detailedonly as muchas it had to be for identification. neck bandat the beginningof the list are principallyidentifiedby their attachmentto the image (and hence distinguishedfrom miscellaneousvotives in the temple) rather thanby theirmetal,which,to judge fromancientjewelryin generaland in view of the importanceof the image,was almostcertainlygold. The owl, aegis, and gorgoneion,on the otherhand,weremoreconciselyand meaningfullydescribedas goldensince this far into the list of ornamentstherecouldbe no questionthat they (andthe five necklaces) belongedon the image.Finally,the phialeis said both to be of gold and held in the silver phialai goddess' hand specificallyto distinguishit from several, predominantly thatweredepositedin the cellaof the Erechtheionas votive offerings.19 It is clear that the ornamentsare systematicallycataloguedfrom the headband downwardsand that the owl, whichis recordedbetweenthe necklacesand the aegis, must have been perchedsomewhereaboutshoulderlevel. Frickenhaussuggestedthat the owl stoodon a pedestalnext to the imageor on the backof the throneon whichhis v6roxxa

8E

PWA/L, XtC , aX

Tw apXauov rTjq 'AOqva&, (oavWv Kacd7JuEva 8ELKVVTaL, Kava1TEp Ev (IwKaUa,

Maooakia,

'aL; ITEL'o0LV.

7Herington,p. 24, note 1.

18IG 112, 1424, lines 11-16; 1425, lines 307-312; 1426, lines 4-8; 1428, lines 142-146; 1429, lines

42-47; and (printedin IG 112,part2, fasc. II) 1424a, lines 362-366, in which alone the text is preservedin full: 0-TEAav7J, ,^v ,^ OEO' EXEL ITacTpa, oy~avf yxpvO, airyLs ypvOT,eOp7OVELOP

a ,^ OEO' EXEL O'X6OLf3O, OJXTVV- fLaX7J tJVO,

oS EXEL E

T TL

TpaX'XwtL Op/OL

TEVTE'

,^J El T7JLXELPLEXEL.

191nthe cella there were four of these silver phialaiand one made of gilded wood: IG 112,1424a, lines 354, 355, 356, 359, and 371.

THEANCIENT IMAGEOFATHENA POLIAS

69

Sitzbildsat.20Heringtonadds that the owl may equallyhave been affixedon the godBut these conjecturesmust now be rejectedin favorof a fourthpossidess' shoulder.21 bility,whichhas some concrete,if indirect,documentarysupport,namely,that the owl was held in the goddess'otherhand. In 1979T. L. Shear,Jr. publisheda recentlyexcavatedAttic inscriptionof 270/69 that honorsa certainKalliasof Sphettosfor his manyservicesto Athens.22From lines 55-70 of the decreewe learnthat Kalliaswas sent to the courtof PtolemyII in 279/8 and there persuadedthe king to donatea gift of ropes for escortingthe peplosat the GreaterPanathenaiain the followingyear. Somewhatsurprisingly,the text (line 65) refersto the festivalas the "Panathenaia for AthenaArchegetis"(Ta Hava0rvaux TEL 'ApXYYE'Td[8)rather than as the Panathenaiafor Athena Polias, as one would ordinari-

ly expect, inasmuchas this was the pre-eminentfestivalof AthenaPolias.As noted in Shear'scommentary,the implicationis that the epithetArchegetismust be a title of AthenaPolias.23'ApXiy'Ts, "FirstLeader"or "Founder",has alwaysbeen knownas one of Athena'smanyepithetsat Athens, but never is it attestedwith any distinctive civic or religiousassociationsthat might indicatewhetheror how Athena Archegetis shouldbe distinguishedfrom the city goddessin general.24 Thus beforethe publication of the new decreeAthenaArchegetiswas barelymore than a name. We can now see why:she and AthenaPoliaswereapparently one andthe same. Now accordingto the scholionon Aristophanes,Birds,516, there existed a statue of AthenaArchegetisand an owl was held in its hand:TI)' 8E 'ApXr'YETL80'A0rjva3To' ayaXApaykavKa EtxEVEv r XELPLThis notice, however, has never seemed a particularly meaningfulgloss on the Aristophanicphraseit is supposedto illuminate,whichsays merelythat "the daughterof Zeus has an owl."25One assumesthat the statueof Athena Archegetiscited by the scholiastwas an Athenianstatue;but even so, Athensmust have been full of representations of Athenawithan owl, and one mustwonderwhy the scholiastsingledout this particular statue,whichis otherwiseunmentionedin the sources. The answeris obvious,of course,if the statueis none other than the old imageof AthenaPoliason the Acropolis,the most ancient,the most authoritative,and probably the best knownimageof Athenainvolvingan owl. Furthermore,its owl, beingof gold and positionedat shoulderheight, was especiallyprominent.Since the epithetsPolias and Archegetisseem to have been interchangeable, and since there is every reasonto 20Frickenhaus,op. cit. (footnote 7 above), p. 24. 2'Herington,p. 23, note 3. 22Shear,footnote 1 above. 23Shear,p. 36, note 88. 24SeePlutarch,Alcibiades,2; Aristophanes,Lysistrata,644; scholion on Aristophanes, Birds, 516 (see below); IG II2, 674 (= TheAthenianAgora XV, Princeton 1974, no. 78), line 16 (the prytaneisof 273/2 sacrificeto Athena, Archegetisof the City, at the festival of the Chalkeia);and in the following dedicatory texts: IG II2, 3175, 3176, 3199, and 3474. Cf. Shear, p. 36, note 88. 25Birds,514-516: ... 6 ZEus yap o vvv f3aGLXEVIWV v sor o aLETOV etsEOTT7jKEV EITL so~K( EXWV (X@ W*V, KEaaXIX ^n OpVLV T71' '

8'

av9 OvyarT'qpykaaiX'. .. .

ean

CAv j3actLXEV';

70

JOHNH. KROLL

believe,therefore,thatthe scholiast'sstatueof AthenaArchegetisand the old imageof Athena Polias were identical,we may concludethat the image, which was probably standing,held a gold phialein one handandthe gold owl in the other. evidencebecauseI believe it enables I have dwelton this literaryand epigraphical us to recognizea representation of the Poliasimageon the reversesof some Athenian bronzecoins of the last thirdor last quarterof the 3rd centuryB.C. The coins were mintedin two varieties,the firstwiththe headof Zeus on the obverse(P1.11:1-4), the secondand morecommonvarietywith a headof Artemison the obverse,so identified by a quiverat her shoulder(P1. 11:6-12).26 That the standingAthena on the coins' reversesis explicitlydepictedin the form of a statueis clearboth from the rigidityof the goddess'pose and from the conspicuousdisplayof the objectsheld in her hands:a phialein her outstretchedrighthandand an owl in her left hand,whose upturnedpalm is held up at shoulderlevel so that the owl itself is abouteven with the image'shead. The imagewearsa Corinthianhelmet and is dressedin a peplos,whichblousesout in an overfoldjust below the waistbut otherwisehangsstraightdownwithouta breakat the knees. This last detail, togetherwith the formal,elevatedpostureof the bent left arm supportingthe owl, indicatesthat the image antedatesthe Classicalera of Greek statuary,when balanced,relaxedposes (includingan obligatorybent leg for standing figures)hadbecomethe norm.Yet despitethe archaicfeaturesof the statue'scomposi26Plate11:1-4, 6, 10-12 are reproducedfrom J. N. Svoronos, Les monnaiesdAthenes, Munich 192326, pl. 25:1-6, 8, 10; P1. 11:5 and 8 from F. S. Kleiner, "The EarliestAthenian New Style Bronze Coins," Hesperia44, 1975, pl. 75, nos. 344 (cf. p. 324) and 89 (cf. p. 306); P1. 11:7 (same reverse die as the corto Pausanias,London roded P1. 11:6) from F. W. Imhoof-Blumerand P. Gardner, NumismaticCommentary 1885-1887, pl. AA.ii (cf. pp. 134-135); and P1. 11:9 from B. V. Head, BritishMuseumCatalogueof Greek Coins:Attica,Megaris,Aegina, London 1888, pl. 15:3 (cf. p. 84). The approximatedate of the coins is deduced from their position (1) following coins of the type Svoronos, op. cit., pl. 24:10-17 (reverse of owl with amphora), all of which were overstruckon coins of Antigonos Gonatas, presumablyafter his death in 240; but (2) precedingcoins of the type Svoronos, op. cit., pl. 81:1-16 (reverse of Zeus holding thunderbolt), which Kleiner, op. cit., p. 328, has firmly dated to the very end of the 3rd century.The coins therefore were probablyminted during the period after Athens was freed from Macedoniancontrol in 229. If so, their iconographyis most plausiblyunderstoodas symbolic of Athens' newly won political autonomy, the reverse statue of the city goddess being particularly evocative in this respect, and the obverse heads of Zeus and Artemis possibly representingZeus Eleutherios and Artemis Soteira (cf. the heads of these two divinities and the accompanyinginscriptionsnaming them on coins of Syracuseduring the democracyof 344-317 restored by Timoleon: B. V. Head, On the Sequenceof the Coins of Syracuse,London 1874 [reprintedfrom NC, 18741, pp. 24-33, pl. Chronological 6:1, nos. 15, 16, pl. 7:8, no. 10). Prophetically(though tenuously, in view of the evidence then available, cf. Imhoof-Blumerand Gardner, op. cit., p. 134, and Head, BM Catalogue:Attica, p. 84), the statue on the coins' reverses was identified as the Athena Archegetis of the scholion on Birds, 516 by E. Beule, in his pioneeringLes monnaiesd'Athenes,Paris 1858, p. 387. It should be emphasizedthat the statue on these coins is the first statue of Athena or of any other deity to appearas an Atheniancoin type; and it is furthermorethe only statue in the numismaticiconography of Athens that can even be suspected of representingthe Polias image. There are three reverse types of an Athena holding a phiale on Athenian coins of the 2nd and 3rd centuries after Christ (Svoronos, op. cit., pls. 86:40-42, 87:11-14, and 87:33-37) but none of these have an owl and (like all the Athenas on Athenian coinage of the Roman period) are rendered in the fully developed Classicalstyle of the time of the Parthenonand later.

THEANCIENTIMAGEOF ATHENAPOLIAS

71

tion, the peplos,as shown on the earlierand more dependablyrenderedreverses(P1. 11:1-8) hangsmost naturalistically, with a roundedflounceof the overfoldand some relativelydeep and irregularhorizontaland diagonalfolds below it.27 Such naturalistic dresswouldof coursebe quiteincongruouson a typicalArchaicstatuemadeentirelyof stone, bronze,or any otherhardmaterial;but it is preciselywhatwe wouldexpectfor the old Poliasimagethatwas drapedin an actualpeplosof cloth. The only detail that might argue against identifyingthe statue on the coins with the

Poliasimage is the helmet, for in the inscribedinventoriesof the image'sornaments the goddess'headdressis listed as a UTE4C{ VT). If, however,the helmet were a normal helmetof bronze,as one wouldassumefrom the conventionalmaterialsof the image's woolen peplos and gold jewelry, it would hardlyhave been inventoriedamong the ornamentsof preciousmetal.Moreover,the wearingof a diademby no meansprecluded the wearingof a helmet. The stephaneneed not have been a large,showytiara;as D. B. Thompsonwrites, "The word appearsto be used of almostany ornamentthat binds the head, such as a fillet or diadem."28 And we see from Plate 1 :a and other representations of Athenawith her helmettemporarily removed29 that she often worea decorativehairbandbeneath her helmet. In all other respects-phiale, owl 'held at shoulderheight, naturalappearanceof the peplos,and the archaic,standingpose-the statuedepictedon the coins agreesso exactlywith the Poliasimageas it can be reconstructed from non-visual sources that it would seem almost perverse not to accept the

identificationand to regardthe helmet as the one new element of the image that the coins have to contribute.On the coins, the helmet is tiltedfar backat abouta ninetydegreeangle on the goddess'head, allowingher face to be fully exposed.The diadem conceivablywouldhave been visibleunderthe helmetat the forehead,above the ears, or in both places.30 27Theinternalchronology of the coins can be reconstructedthrough the progressivesimplificationof the reverse-dieengravingand the evidence of some probableoverstrikes.On the earliest specimens, with a head of Zeus on the obverse and a dotted borderaroundthe circumferenceof the reverse (P1. 11:1-5), the statue's draperyis renderedmost naturalistically;and P1. 11:1 even shows a slight bend in the right arm of the image. The realisticdraperyis continued into the next phase of the coinage (P1. 11:6-8), with Artemis head on the obverse and dotted border on the reverse. But in the final Artemis-headphase (P1. 11:9-12), the dotted border is omitted from reverses and the peplos tends to be renderedrathermore schematically, the overfold being indicatedas two pointed "tails".That such archaistic-lookingtails are the result of hasty die cutting and must not be regardedas faithful reflections of the peplos as it was actually seen on the image is especiallyclear from the cursive, linearrenderingof the entire statue on P1. 11:12. Most of, if not all, the coins from the third phase of the coinage appearto have been overstruckon earlierspecimens with the Zeus-headobverse. 28D.B. Thompson, "The Golden Nikai Reconsidered,"Hesperia13, 1944, p. 193. 29E.g.,the Pheidian "Athena Lemnia" (G. M. A. Richter, Sculptureand Sculptorsof the Greeks,4th rev. ed., New Haven 1970, figs. 654-658) and two vase paintingsconveniently illustratedin J. Boardman, AthenianRed Figure Vases, the ArchaicPeriod, London 1975, pls. 160 (amphora by the Berlin Painter, AR V2, p. 202, no. 77) and 185 (amphoraby the TyszkiewiczPainter, ARV2, p. 1643, no. 33bis). 30Cf.the amphoraby the Andokides Painter showing Achilles and Ajax gaming (Boardman,op. cit., pl. 2:1; AR V2, p. 4, no. 7). The helmets of both heroes are tilted fairlyfar back on their heads so that their headbandsare exposed over their ears.

JOHNH. KROLL

72

The representationson the coins show that the phiale, the owl, and the arms, whichwere specificallypositionedto supportthese gold objects,were not addedpiecemeal but were all elementsof a singleartisticwhole, as partsof whichthe helmet, the gold aegis, and the gold gorgoneionwere, in all likelihood,also created.The episodein that mentionsthe (temporary?)loss of the gorgoneionduringthe abandonPlutarch31 ment of AtticabeforeSalamisprovidesa terminusante quem of 480 B.C. for the manu550 is postquemof approximately ensemble.A terminus factureof this gold-ornamented indicatedby the generallyLate Archaicaspectof the image (note especiallythe slight, relaxedbend in the rightarmthat extendsthe phiale,shownon whatseems to be the earliestand most reliablecoin reverse,P1.11:1)and by the formof the helmet,which, becauseof the way it is worn, can only have been of the fully developedCorinthian by extendedcheek pieces and a type. This varietyof Corinthianhelmet, characterized cut-awayneckguardat the back,whichtogetherallowedthe helmet to be restedhorizontallyon the top of the head, did not evolve until aroundthe middle of the 6th century.32Since the resultingdatingof ca. 550-480 correspondswith the careerof Endoios,whose workinglife A. E. Raubitschekhas fixed between ca. 540 and 500,33 the coin representationslend no small credenceto Athenagoras'associationof the statuewiththatmastersculptor. Accordingto no fewerthanfive ancientauthorities,however,the imageof Athena Poliaswas vastlyolder than this. As quotedat the beginningof this paper,Pausanias of Atticaand that it was said informsthat it was veneratedlong beforethe synoikismos to have fallenfrom the sky. Othersvariouslyattributedits originto Kekrops,the first x.9.15); to Kekrops'offspring,ErechEvangelica, kingof Athens (Eusebios,Praeparatio thonios (Apollodoros,iii.14.6);or to the aboriginalinhabitantsof Attica,the autochthonoi (Plutarch, De daedalisPlataeensibus).MMoreover, Plutarch (ibid.) thought it re-

markablethat the Atheniansstill preservedthe imageto his day;and he lists it among the oldest cult statues of Greece, along with the originalwood image of Apollo on Delos that was given by Erysichthon(anotherson of Kekrops),the originalwooden Hera of Samos, Danaos' wooden image of Athena at Lindos, and the originalpear31Seefootnote 3 above. 32E.Kukahn, Der griechischeHelm, Marburg1936, pp. 45-47, pl. 4:1, 2; A. M. Snodgrass, Armsand Armourof the Greeks,Ithaca1967, pp. 93-94. 33Raubitschek,op. cit. (footnote 15 above), p. 495. 34Frag.158 in F. H. Sandbach,Plutarch'sMoraliaXV, Loeb ClassicalLibrary,Cambridge,Mass. and London 1969, pp. 293-297: 'H 8E TWcP4oavwv 7TrotqoLt apXatov EOLKEVElat LT Kat wakaLov, E"y Okwov 8E E'7rtT^7(0E OEWputv ayakaa, OvxL1wol' v7ro' 'Epvo-'X6oioo 'A7ri6XXwat EL' A7^jX0o^ 7rEO6E 7rp^7pWTOl AP P 'Hpac U Ka' P pvkE'', o ,utxpk i'i^ 'A6'qva^ot 8bOvWaTov01. 7^ Hao~tc~o' V7TOraW av'ToxO660cw Ea/Uot L

tvWov ELX0 s'8o,

cs 4firn KaAWuo',

OVIT(c) EKEXAP401

E"plyOl

EV(OO1/,

aA'

E7TL TEO/AZW

oa Pt'L. 'rfa V Aooq yxvo&i'a yacp tipv0oPTO OEOV' TOTE Kat yap 'AOr7v1 08o'. XtTLrV Er)KEJ' P ALP&p AavaS

8brwaco

tJ8E

r'HpacLEpOIvEtcra4LEiO ... XE'yETat8E HdL'pas o 7rpwTo'; 'Apyokl8o 'Hpac EVKEapop, /o(x*oat. TE~WO acyaXAt

EK TrW ITEp'l Titpvv~a

&'opcP olYxvxr

THEANCIENTIMAGEOF ATHENAPOLIAS

73

wood image of Argive Hera. Finally, Philostratos(Vita ApolloniiiII.14) cites the Polias in his list of the most ancient images of the gods, which includes the same xoanon of Delian Apollo, the image of Dionysos in the Marshes, and the Apollo of Amyklai. Nothing is known about the appearanceof the wooden Apollo on Delos or about the image of Dionysos in the Marshes, venerated at the site of the oldest festival of Dionysos in Athens, accordingto Thucydides, ii.15.4. But the fragment of Kallimachosquoted by Plutarch (footnote 34 above) informs that the oldest cult statues of Hera on Samos and of Athena at Lindos were, respectively, an aniconic plank (Jooa o-aptc) and a plain image (AXrTwAo,0.35 And Pausanias describes the pear-wood Argive Hera as a small seated statue (II.17.5-6)36and the AmyklaianApollo as an ancient and unskilful image about 40 feet high, having the form of a bronze pillar with feet, a helmeted head, and arms holding a spear and a bow (iii.9.2).37 Clearly, in the company of such comparanda as these, there must have been something conspicuously primitive about the Polias image as well. How is this abundant and reasonably consistent testimony about the prehistoric origin of the Polias to be reconciled with the Archaic statue depicted on the coins? Since we are not entitled to assume that the Erechtheion housed two images of the goddess,38both the prehistoricand the Archaic aspects of the Polias must have been combined in the same image. We have already accounted for its visible externals-the peplos that was renewed every four years and the helmet, the arms, and the gold ornaments and attributesthat are to be associated with Endoios. This leaves only one component that could antedate the 6th century:its body or core, which was hidden beneath the peplos and which may very well have gone back to the time of the Bronze Age kings of Athens, if not much earlier still. If the nucleus of the image was indeed as ancient as the sources insist, we may readily envisage it as a primitive, aniconic or quasi-iconicfetish of olive wood. Some rathermore concrete evidence to this effect is possibly to be found in Tertullian, who, in a defense of the Christians'alleged worshipof the cross, asks, "How much difference is there between the shaft of the cross and Pallas of Athens (Pallas Attica) or 35The unworked cravis of Samian Hera is mentioned also by Clement, Protrepticusiv.46.3. Cf. E. Buschor, "Heraionvon Samos," AthMitt55, 1930, pp. 4-5. On the originalAthena Lindia, S. Casson, The Techniqueof Early GreekSculpture,Oxford 1933, pp. 62-65; Lorimer, op. cit. (footnote 9 above), pp. 443-444. 36Cf.Lorimer, op. cit., p. 444, with references. 37TheApollo is illustratedon Spartancoins of the 3rd century after Christ: S. Grunauer-vonHoerschelmann, Die Minzprdgungder Lakedaimonier,Berlin 1978, p. 99, pl. 32:12, 13. Cf. L. Lacroix, Les reproductionsdes statuessur les monnaiesgrecques,Liege 1949, pp. 54-58, pl. 1:15, 16; Casson, op. cit. (footnote 35 above), pp. 56-57. The coins indicate that the columnar body was wooden and only sheathed with bronze. 38Ahypotheticalcase for two "ancient"images is hardlyworth considering.E.g., since the coins show the Panathenaicpeplos on Endoios' image, the sacrosanct,prehistoricimage, were it separate,would-incredibly-have been left undraped.The fact is that neither Pausanias,whose account of the contents of the cella of the Erechtheion is unusually thorough, nor any of the other literary or epigraphicaltestimonia collected by Jahn and Michaelis, loc. cit. (footnote 4 above) allow for more than one image.

74

JOHNH. KROLL

the Ceres [sc. Isis] of Pharos (CeresPharia), each of which is displayedas a rude stake and unshapedpiece of wood without effigy?"39That Tertullian'sPallas Attica should be understood as the Athena Polias on the Acropolis is probable enough.40By carefully pairingthe Attic Athena and the AlexandrianIsis, Tertullian has not chosen any two aniconic paganimages at random but comparesthe abstractform of the cross to images of the chief deities of the most prominent intellectualcenters of the pagan world. Apart from Pheidias' chryselephantineAthena Parthenos, the Polias is the only image of the goddess that could be meaningfullyreferred to as the Pallas of Attica, and it is, moreover, the only known wooden image of Athena in Attica that can even be suspected of having an essentially aniconic character.Granted that Tertullianmay never have visited Athens and known the image at first hand, the casualness of his reference and the fact that he is addressing pagans about their own objects of veneration suggest that he is alluding to what was common knowledge at the time.4' Yet for all that, Tertulliandoes not name the Polias, and his credibilityhere can be checked only insofar as we have independentgroundsfor thinkingthat the unadornedPolias may have been more or less as primitiveas his Pallas Attica. Consequently, while Tertullian'sstatement deserves at the very least to be taken seriously as possible complementaryevidence for the underlying natureof the statue, one can insist on neither its reliabilitynor its relevance. This is particularlyto be regretted when we turn to consider the image's face. The coins indicateonly that it had a face and that the face was to some degree naturalistic.If the olive-wood core was therefore genuinely aniconic, the face would have had to have been a late addition, ascribable,like the arms and gold ornament, to Endoios. As illustrated in a number of 5th-centuryAttic vase paintings, aniconic column or tree-trunk fetishes of Dionysos were regularlyanthropomorphizedby the additionof a mask and by cloakingthe wood column or log with a garment.42And from Hyperides, pro Euxenippo, a crucisstipitePallas Attica, et Ceres Pharia, quae sine 39Apologia,16.3.8: Et tamenquantodistinguitur a crucisstipite effigierudipalo et informilignoprostant?Cf. Tertullian,Ad NationesI.12.3: Quantodistingqitur PallasAtticaet CeresPharia,quaesine formarudipalo et solo staticulohigniinformisrepraesentatur? All the standardcommentariesnote that Tertullian'sCeres Phariais Isis Pharia,for whom see P. M. Fraser, PtolemaicAlexandria,Oxford 1972, I, p. 20; II, p. 54, note 125. All representationsof Isis Pharia portrayher in fully anthropomorphic,Hellenisticform (P. Bruneau,"Isis Pelagia'aDelos," BCH 85, 1961, pp. 435-446, and BCH 87, 1963, pp. 301-308). Tertullianis the only source that mentions a primitiveimage. 40A. Schneider (Le premierlivre Ad Nationes de Tertullien,Neuchatel 1968, p. 250) suggests alternatively that the Pallasmay allude to the Palladionimage housed near the Athenian lawcourtcalled Egii naXka8L'W (L. Ziehen, s. v., "Palladion,"RE XVIII, 1949, cols. 176-179). But if this image could be designated as a Palladion, it almost certainlywould have been anthropomorphic,unlike Tertullian's Pallas. On the form of PalladionAthenas, see G. Lippold, s.v., "Palladionin der Kunst," RE XVIII, 1949, cols. 189-201; dans la ceramiqueitaliote,Geneva 1975, pp. 87-97, with plates. J.-M. Moret, L'Ilioupersis 41T.D. Barnes (Tertullian,A Historicaland LiteraryStudy, Oxford 1971, pp. 107-108, 194-210, 213) emphasizes that Tertullian was a respectablylearned man, well educated and versed in classical pagan literature,and that his Apologywas directed specificallyto a cultured audience. Although evidence is lacking, Tertullianmay have traveledwidely throughthe Greek East (ibid., p. 198). 42Thematerial is collected and discussed in A. Frickenhaus, Lendenvasen,Winckelmannsprogramm LXXII, Berlin 1912. Cf. W. Wrede, "Der Maskengott,"AthMitt53, 1928, pp. 81-92, figs. 1-3. Summary by A. Pickard-Cambridge,TheDramaticFestivalsof Athens,2nd ed., Oxford 1968, pp. 30-34, figs. 17-22.

POLIAS THEANCIENT IMAGEOFATHENA

75

24-25, we learn of a magnificentface or mask that the Atheniansin the 320's prepared-alongwith "otherappropriate partsand muchexpensiveornament"-forembellishingthe apparently primitiveimageof Dione at Dodona.43 Thus,the additionof a fine would of have been maskto the olive-woodimage the Polias fullyin keepingwithestablishedpractice.On the otherhand,if we disregardTertullian'sreferenceto a featureless PallasAttica,it becomesequallypossiblethat the face of the Poliaswas carveddirectly out of the ancientolive-woodshaft.Thistoo couldhave been the workof Endoios.Or it couldhave been a featureof the prehistoricimage:in his surveyof pre-ArchaicGreek cult statues,S. Cassonconcludesthat "it is possibleto establishthe workinghypothesis that pre-HellenicCretanand some of the earliestHellenicfiguresof deities sharedthe peculiarityof havingplain or almost columnarbodies but realisticor, at least, partly detailedheads."44But howeverthis may be, withouta close-upview of the Polias,the questionof whenandhow she receivedher facemustobviouslybe left open. All the essentialsourcesneverthelesscan be reconciledand with the help of the coins may enable us to understandPausanias'puzzlingsilence about the statue'sappearance.Withthe possibleexceptionof the face, all the externalelementsof the Polias would have been regardedby the ancientsas ornamentation,distinctfrom the image per se. ThroughoutHyperides'discussionof the Athenians'embellishment(EntKOcrof the Dione statueat Dodona (footnote43 above), the A8osor 'yaXtAtaof the LELOv) goddessis consistentlydistinguishedfromits richand artisticKOcrUpOS.The same distinction is explicitlyappliedto the statue of Athena Polias in Plutarch'saccountof the Athenianfestivalof the Plynteriaas the time when the Praxiergidai removedthe KOYWu from the vwoqand veiled the latter from view.45For a serious antiquarianlike Pausanias,it was of course the ancient and true E'8oq,not its Kocrpsos, thatwas of consequence. Yet withoutlifting up the peplos there was nothingto be seen of the true image,exceptperhapsthe face.One must addto this thateven thoughEndoios'embellishmentsmust have been exquisitelycrafted,his remodelingof the Poliaswas lacking in the kind of artisticor iconographicnovelty that would have attractedPausanias' interest.The goddesswas given only her usualattributes-owl,helmet, aegis, and gorgoneion-and the portrayalof her holdingout a phialewas, to judgefromotherArchaic cult statueswith phialai,46 a rathercommonand undistinguished conception.Given the 43WVZ yap

6 ZEw 6 A8can'os

7rpoaTE'

Ev aj

tai

To

ayaX Tc

AUw0

ErLKO`OaLa Kat

W- OLOJ TE Ka'XXLO-TO Kat Ta9XXa7raJ'Ta Ta aKoXov~a, Kai KOO-A0u7roXVwv 7rpOOdT7rO7 TE 7roL7)OaA~Eu'oL KaEITOXvTEX7)1 co iTapacTKEvao-avTE' ... EITEKoOTA~'qOaTE To ebo' T Aviwr~ &gwc Kat VwIJ avhc Kat VEL1

Irq^- OEOV.

44Casson,op. cit. (footnote 35 above), p. 58. 45Alcibiades,34.1: 8p(Lt 8E%Ta opywa HlpaeEpyytbat0apy'qXu^0vo' E`Kn OOLIi'TOI''awopp)Ta, TO V TE KO-A/O V KaOEXO'VTE#Kat TO E80o' KaTaKaXlAIjaVTEc. If my interpretationof the image is correct, the passage implies that all Endoios' additions, including the arms and possibly a face, were so constructed as to be removable. On the Plynteriafurther:L. Deubner, AttischeFeste, Berlin 1932, pp. 17-22; Herington, pp. 29-30; D. M. Lewis, "Notes on Attic Inscriptions,"BSA 49, 1954, pp. 17-21. 46E.g.,the Piraeus bronze Apollo, whose phiale is not preserved (G. M. A. Richter, Kouroi, 2nd ed., London 1960, p. 136, figs. 478-480, no. l59bis); the Apollo Smintheus of AlexandriaTroas (Lacroix, op.

76

JOHNH. KROLL

thoroughgoingconventionalityof the statue's externalform, it was only naturalthat Pausaniaspassedover it and chose to commentinsteadon the image's unique and profoundhistoricalsignificanceas the religiousfocusof Athenssince time immemorial. Addendum

Line drawingsof severalcoins of the type discussedabove publishedby Beule, op. cit. (footnote26 above), p. 387, and by A. B. Cook, Zeus, III, i, Cambridge1940, p. 827, figs. 636, 637, show the reversestatue with the goddess'feet exposedand with pose commonto manyArone leg in advanceof the other in the "walking-standing" I a coins that conhave few unpublished seen chaickorai.Since submittingmy paper firmthese detailsof Beule'sand Cook'sdrawingsand clarifythat the feet are depicted, althoughas ratherformlessdots, belowthe peploson my Plate 11:11.Since the peplos clearlyflounceson the groundand hides ope or both feet on other specimens(P1. 11:5-7, 9), I conclude(1) that the feet of the imagewere sometimesexposedbut at othertimes covered,just as one wouldexpectof a statuethatwas repeatedlyundressed and dressedin a long woolengarment;and (2) that the feet and the legs to whichthey wereattachedmust be addedto the list of Endoios'embellishmentsfor the image. JOHNH. KROLL THE UNIVERSITYOF TEXAS AT AUSTIN

Department of Classics Austin, TX 78712

cit. [footnote 37 above], pp. 76-86, pl. 4:2-14); the Artemis statue depicted on Athenian coins of the 2nd century B.C. (M. Thompson, The New Style Silver Coinageof Athens, New York 1961, pls. 75, 76, nos. 709a-714c; Lacroix, op. cit., p. 205); the statue of Aphroditedepicted by the Meidias Painter in his scene of the rape of the daughtersof Leukippos (AR V2, p. 1313, no. 5); and the completed image of Samian Hera, which held a phiale in each hand (Lacroix, op. cit., pp. 206-216, pl. 17:6-10; Buschor, op. cit. [footnote 35 above], fig. 2). In the Classicalperiod, cult statues holding phialai seem to have become even more common; see B. Eckstein-Wolf,"Zur Darstellungspendener G6tter," A'IdI5, 1952, pp. 64-65, and Lacroix, op. cit., pls. 24:1, 26:1, and 28:7.

PLATE

Berlin a. Black-figuredkalpis (E. Gerhard,AuserleseneVasenbilder, 1940-58, IV, pl. 242)

*

b

~~j 3

2

..tj

7

X~

'

4

5

.

6

9ACt.

8

9

10

11

12

11

TOPOGRAPHICAL NOTESON ARISTOPHANES* 1. WASPS,844-847:HESTIA'SPIGSTY' B8.

TOVT TLEOcr;

CF.

B8.

XotpoKo/ELtvO'EV

Tora1.

EO' 1EpoOvXTVo-(ra'TEpELs;

C?.

aXXsva

OVK, Ls V0T9 &4l eEoTr &XX'Eta-ay'

ap01

&PX4EVO

E7

t

ETLTpt)

n

t T

Va

TLra.

aVvf(Xa%&,s E'YWTveaiV J3E7IT.

At line 838 Philokleonleft his do-it-yourselfdicasteryin quest of an indispensable pieceof courtroomfurniture,the 8pViaKT1o9, or barof the court.2Herea surprisedBdelykleongreetshis fatherwith"Whatis this?"as the old manreturnsdragginga contraption whichhe describesas a XOLPOKO,.tdoLv'Eor1a. Althoughit is apparentfromthe following lines thatthe old man'stermfor his latestpropis not lost on Bdelykleon,XOMPOKOItEioL The scholiastwrote:Xotpocuriosityto latercommentators. 'E.TIaK has been a persistent KOJUEdOV ETLV

E.TLK

ayyEtOV

8E, I TEI E7TLTs1

[Rogers: EOTIa'

EO-n (&ypE'6vI TpE4oVOI

TL KaVVcTor4,

O6TOV oi

XO'pOV'. E10r4EpEt 8E

XOPOLTPE'vrTat.

0OVTo aV7L 8pV4a'KT0V

Moderncriticshave acceptedthis explicationas faras X0tp0K0oEtiov,3 but they KUyKX1801. *For Homer A. Thompson, good teacherand friend. 'The texts of Aristophanesused in these notes are from the following editions: D. M. MacDowell, AristophanesWasps,Oxford 1971; K. J. Dover, Aristophanes Clouds,Oxford 1968. 2See lines 829-833. The word is more often used in the pluralas in lines 385-386 where Philokleon instructshis fellows in the chorus to bury him V'ro' rotl 8ppV xKTO. The barrieris often thought to have separatedpublic and court proper, but S. Dow ("Aristotle, the Kleroteria, and the Courts," HSCP 50, 1939, pp. 20-21) sees it as surroundingthe central propertiessuch as bema and ballot box. Cf. H. A. Thompson and R. E. Wycherley, TheAthenianAgora, XIV, TheAgoraof Athens,Princeton 1972 (= Agora XIV), pp. 57, 59, for physicalevidence of a dryphaktosor kinklidesin a later courtroom of the Agora. See also F. Salviat, "L'offrandedes Hermaisteset les TPYbAKTOIdeliens," BCH 87, 1963, pp. 259-264. survives elsewhere in early Greek writing only at Lysistrata,1073 where the Spartan 3XOtPOKOJUE'OV envoys arrive WOIrEp XOtPOKoIAktOV ITEpt TO; /A7)POYL0V EXOvTEg. For all the varied interpretationsof this line (see F. Blaydes, AristophanisLysistrata,Halle 1880, p. 286), it clearlyjests, as some critics have observed, at the aggravatedpriapicstate of the envoys, whose cloaks are so extended that the coryphaeus makes fun of them as covering pigpens, or, the posts to which pigs were tethered. If, as some have suggested, the humor does not stop there, I think that portionsof the scholia to both occurrencesof the word ... (XOtpOKOUE'OV ToPs xoLpov; TpE ovat ... irov oit X?opot TpEkovTatand raniraXo' v (t 8EO-AEViVTE' are clues to the baser meaning of the word at Lysistrata,1073: This XOLPOKO.ELOVis a place where X?opo? or Xomp&8a(see footnote 13 below) are nourished and reared. Blaydes (loc. cit.) perhapsalludes to some such view of Biset (O. Bisetus, 'AptcroToivovs KcoW/OJJatE48EKa ..., Aureliae Allobrogum 1607), but I could not get hold of the latter work. fko-lrEp XOLPOKO/ELOVseems independentof the precedingobservations that the Spartansare draggingtheir beards; the envoys are probablybent over with agony, and in Aristophaneslong beardsare a stereotypedattributeof Spartanmen (cf. Wasps,446). Translationof xotpOKO/UELOPhere as "sanitaryknapkin"(LSJ, s. v., and more recently, C. Ruck, "Euripides'Mother:Vegetables and the Phallos in Aristophanes,"Arion,n.s. 2, 1975, p. 46, note 58) is therefore contextually,if not syntactically,farfetched.Furthermore,the ending -Eov usuallyindicatesplace.

78

GERALD V. LALONDE

are dividedover the meaningof 'EOrTla.Is the pigpen'Eork'a becauseit was near a hearthin the houseor courtyard(Starkie,Rogers),or becauseporkers,at leastthe first?4 lings,weresacrificedto Hestiaas goddessof the household(Wilamowitz,MacDowell) Aristophanesmay have intendedone of these meanings,andwe may preferthe second butit is likelythat for keepingthe pigsout of the houseor the hearthout of the barnyard, which the poetwasplayingherewithmorethanjust domesticpieties.Sincethe trappings Philokleonsecuresfor his courtaredomesticmockeriesof the realthing,porridgebowls for votingurns (855), an amis for a waterclock(807, 858) anda pigstyfor the courtroom significancebeyond dock,one mightsuspectthat'EOTlaqhas some similartopographical cannotbe identifiedwitha hearth the comichero'shearthandhome. Although'EOrTiag or cultof Hestiain anyof the usualmeetingplacesof dicasticpanels,5Philokleon'suse of the wordwouldsurelyhave put the Athenianaudiencein mindof the civic worshipof bothbuildings Hestiaandthe hearthspar excellencein the PrytaneionandBouleuterion,6 whichservedat times as courtsof law. The Prytaneionhoused the chieflyceremonial courtwhereBasileusand Phylobasileisin the mannerof a coroner'sjurypresidedover cases of homicidein whichthe accusedwere unknownpersons,animalsor inanimate objects.7In the Bouleuterionthe Councilexercisedits jurisdictionover a varietyof offenseswith a limitedpunitivepowerof fines up to 500 drachmas.It couldrefermore seriouscases to the Ekklesiaor dicasteriesand, if the referredchargecarriedno fixed penalty, the Boule could recommendone to the receivingtribunal.8XotPoKOUEdoV 'E0o.iaq in its context suggestsespeciallythe CouncilHouse with its renowned8pvan orchestralareawith bemaand sacredhearthwhereofficialsmay kaKTrOsurrounding sacrificeof suckingpigs beforeeach meeting.9To Athenians, have madea purificatory 4W. J. M. Starkie, The Waspsof Aristophanes,London 1897, p. 284; B. B. Rogers, The Comediesof Aristophanes,London 1916, II, p. 129; Wilamowitz, SBBerl, 1911, p. 518, note 1; MacDowell, op. cit. (footnote 1 above), p. 244. For the sacrificeof pigs to Hestia, see Eupolis, I.335 (T. Kock, Comicorum AtticorumFragmenta,Leipzig 1880-1888, frag. 281); Athenaios, iii.96 B; iv.149 D; K. Keil, "Zum Corpus InscriptionumGraecarum,"RhM, n.s., 20, 1865, pp. 550, 556. 5The dicasts, however, seem to have met in more places than can now be clearlyidentified;see Agora XIV, pp. 71-72. 6For the hearth and statue of Hestia in the Prytaneion,see Pausanias,i.18.3-4; [Plutarch] Vitaedecem Testioratorum,847 d, e; Pollux, ix.40; R. E. Wycherley, TheAthenianAgora, III, Literaryand Epigraphical monia,Princeton1957, no. 570 and p. 174. For the &i-tra of the Bouleuterionand oaths and dedicationsto Hestia Boulaia,see Aischines, Ii.45 and the scholia which identify the Eom'a with the altarof Zeus Boulaios; Harpokration,s. v. Bovkaia. Regardingour identificationsof the Bouleuterionor Prytaneionin the mention of Hestia, cf. A. Boegehold's observations(in "Philokleon'sCourt," Hesperia36, 1967, pp. 113-114, note 10; the articleidentifiesPhilokleon's "real"court in the Wasps)that topographicalallusions to publicinstitutions could be made simplyby namingthe god (or man) whose statue stood on the site in question. 7See Andokides, 0.78; AthenaionPoliteia,57.4; Demosthenes, xxii.76; Pausanias,i.28.10-11; Plutarch, Solon, 19.3; Pollux, viii.120; I. Bekker, AnecdotaGraeca,Berlin 1814-1821, I, 311.15; D. M. MacDowell, AthenianHomicideLaw, Manchester1963, pp. 85-89. 8See P. J. Rhodes, TheAthenianBoule, Oxford 1972, pp. 144-207. 9The arrangementof 8p1OKTaOS and E'o-ra was roughly the same in Old Bouleuterionand New; see Equites,674-675; Xenophon, Hellenika11.3.50-56;Rhodes, op. cit., pp. 33-34 and Plans E and F; W. A. McDonald, ThePoliticalMeeting-Placesof the Greeks,Baltimore1934, pp. 131-138; AgoraXIV, p. 34. The sacrificeof piglets is attested only for meetings of the Assembly specificallyand for civil gatheringsin general, but Rhodes (p. 36) thinks, as some of the testimonia suggest, that the same ritualwas performed

NOTESONARISTOPHANES TOPOGRAPHICAL

79

jest was somewell, the full thrustof Aristophanes' who knewtheirjudicialtopography thing more than later critics have realized:Philokleonsteals a XOtpOKo0tdoV as his pretentionsto the hallowedchambersof Prytaneionor 8pv'faKTrOq and makesirreverent Bouleuterionby referringto the railingas a "pigstyof Hestia."10 As is often the case, Aristophanes'charactersare not contentwith one pun. Mention of Hestiabecomesthe cue for anotherpithyjoke in the next two lines with Bdelykleion'smockworrythat Philokleonhas plundereda shrineof the hearthgoddess,and the old man'sretortthat he has a redeemingpurpose-to squashsomeonea+' 'Eamraq apXo/.Evoq. As most critics note, line 846 makes a pun on the literal and proverbial meanings of &a' 'EUrkaxa~pXEorat.11First, Philokleon would begin his assizes auspi-

ciouslywiththe customaryfirstsacrificeto Hestia,12and it may be addedthat the impliv 'EriTag mentioned above and Philokleon's own comic savagery cations of XotpoKottEiO

suggestthathe sees defendantsas so manyswineand theirlegaldestruction(ULrpm co) as his sacrifice.3Secondly,with the meaningof a proverbderivedfrom the customof firsthonorsto Hestia,he intendsto squashsomeone "beginningfrom the utter beginning."14This is well-done irony, since with s1T~rp'qcorather than 8tKauR-co or some

such verb, Philokleonshows his real predilectionfor beginningwith the end. This sentimentthat punishment,especiallythe capitalsort, is the only kind of due process recursimmediatelyin the old man'sinsistenceon a damningvote withoutthe botherof Bdelykleon'so-av8es and ypajax (847-850), and a little fartheron in his premature judgmentthat the cheese-filchingLabesshould "die like a dog" (998). In this vein of absurddraconismAristophanesprobablyintendeda thirdentendreof &f' 'Easeos7apXo/.LEVOS E'LTpLtJIW

Ttva: Philokleon in his remorseless zeal for punishment would even

snatchhis preyfromthe supplianthearth.'5 before meetings of the Boule; see Aischines, I.23; Acharnians,44 and scholia; Ecclesiazusae,128 and scholia; Harpokration,Suda and Photios, s.v. Kaeapo-tov; Suda, s.v. EpwL-TtapXao; Pojlux, VIII.104; A. G. Woodhead, "IY.HFOPIA and the Council of 500," Historia16, 1967, p. 134. "0Itshould not surpriseus that Philokleon, elsewhere a dicast, should suddenly present himself in the judicial capacityof a councillor. In at least one other place in the WaspsAristophanesachieves a joke by allowinghis comic jurists to claim bouleuticduties in which the real dicastsapparentlyhad no share:at line 578 Philokleon lists among the delights of a dicast 7rTa8WVTotvV 8oKtaoAE'vwval8ota lapEo- 6Eao-6at. As far as we know, dokimasiaof candidatesfor citizenshipin the 5th centuryB.C. was the task of the Boule alone; see AthenaionPoliteia, 42.1-2. Cf. such satiric license on a grander scale in Aristophanes'use of Sokratesin the Cloudsas the mouthpieceof a varietyof contradictorynewfangleddoctrines. "See W. Suess, RE VIII A, 1913, cols. 1274-1275. 12Cf.Plato, Cratylus,401 D; Aristophanes,Aves, 865; Sophokles, fr. 658 (A. Nauck, Tragicorum grae2nd ed., Leipzig 1889, p. 287); Euripides,fr. 781, 35 (Nauck, p. 608). corum.fragmenta, "3Cf.lines 570-574 where Philokleon savors the image of a cringing defendant presentinghis crying daughterto the dicast as to a god who delights in the sacrificeof pigs: EL 8' av Xotpt8&ovxatpx OvyaTp o kOWVi/1E rn&'o-Oat There, as more notoriouslyat Acharnians, 769-796, Aristophanesgets a salacioustwist from the double meaning of Xotpttov (piggy/cunnus) and synechdoche of the part for the whole; cf. J. Henderson, TheMaculateMuse, New Haven 1975, pp. 60, 131-132. "Cf. Plato, Euthyphro,3 A: aTEXzVCO yyap jAm 8OKEJ ai' rea(tT' apXEo-Oat KaKOvpyELV TriV IfoxcV, CMXEtpwVa&tKEJV

"E.

'5Withthis meaningthe line may have echoed again the E'o-rTaof the Bouleuterionif that altarhad any of its later fame as a place of refuge; Andokides, I.44; ii.15; cf. especiallyXenophon, Hellenikaii.50-56, for

GERALD V. LALONDE

80

2. CLOUDS, 323-325: THE CLOUDS OVER PARNES lAd.

Pk 1 VVV 8EVpL lrpa J3XE'TE

77 V HTpr'ij9' a 1,p-o.

7Rovyjjai'rcs. IT.

En.

4EpE A aI

aTaw ut

irov; ^

qA

yap 6pc KaTLoVO-a'

&Eiov. ^

t

IraVv JroXkai xwpovio avTat f

KOL'XO) V KaLATraw 8ao-E'an,

t

avras ITkaytaL.

When this scene introducingthe Chorusof Cloudswas playedin the Theaterof Dionysos, Sokratesobviouslycould not call Strepsiades'attentionto Mount Parnes itself, for in that directionthe Acropolislooms over the theater.Dover has rightly advisedus not to infer from this that the playwas intendedfor productionin another theater.'6The dramaticscene could be imaginedas takingplaceat any spot in Athens with a view of Parnes,and, as Dover suggests,Aristophanesprobably"intendedthe actorto compromisewith theatricalconditionsby pointing,not too precisely,past the But why shouldAristophanessingle out Parnes? east or west end of the Acropolis."'7 He could easily have avoidedthe compromiseby havingSokratesdirect Strepsiades' attentionto an apparitionof the Cloudsover Hymettoswhich dominatesthe eastern skyline.A plausibleexplanationis that the poet, wishinghis fantasyto imitatenature, made Parnesthe placeof epiphanybecauseit was most commonlyassociatedwith the of cloudsand cloudyweather.In a fragmentof the lost earlierversionof his appearance Clouds,Aristophaneshintsfurtherthathe thoughtof Parnesas the prominenthauntof A the Cloudswho are said, "when angered,to go off to Parnes,by Lykabettos."'8 numberof scholars,ancientand modern,have remarkedon the specialrelationshipof Parnesand real clouds.Theophrastosprobablyspokefor the 5th centuryas well as his own when he noted that the Atheniansinterpretedthe gatheringof cloudsover Parnes C. C. Felton, who duringthe prevailingnorthwind as a forecastof stormyweather.19 was still regularly when Parnes the 19th in Athens months century during several spent visiblefrom the capital,observedthat he "seldomsaw the summitwithouta mass of delicate,silveryclouds restingon it ... ."20 Wachsmuthcited a study of the relative frequencyof morningclouds crowningParnes (143 days of the year) and Hymettos (103 days) and attributedthe differenceto the variationin altitudeof the two mounthe drama,of Kritiasand his henchmen draggingTheramenesfrom this altarwhile the councillorswatched in frightfrom the other side of the 8pV1KTOaI; cf. [Plutarch], Vitaedecemoratorumiv.3; Diodorus Siculus, xiv.4. "Dover, op. cit. (footnote 1 above), p. 143. 17Ibid. of AtticComedy,Leiden 1957, I, p. 682: 18J. M. Edmonds, TheFragments Photios. HIa'pv)- T' opo' 6rqxvKJs Es T-qV HIa'pV-q' op'yuait6oat 4pov8at Kaa'a Tr6 AvKaJ3qTTO'V 'Apto-o4&vv NE4AaLa Kal E7. 19Theophrastos,

de signis tempestatum, 47.

20 The Cloudsof

Aristophanes,Boston 1880, p. 136; Felton's commentaryis quaintlyromantic but still

useful in parts.

TOPOGRAPHICAL NOTESON ARISTOPHANES

81

tains.2'So, althoughclouds over Hymettosalso could be harbingersof rain,22since Parneswas the highermountain,on the prevalentnorthernroute of Athens' weather, andthus usuallythe firstandforemostgatheringplaceof clouds,it is not surprisingthat Aristophanesshould choose it over any other place for the first apparitionof his Chorus. GERALDV. LALONDE GRINNELL COLLEGE

Departmentof Classics Grinnell, IA 50112 21C. Wachsmuth,Die StadtAthen, Leipzig 1874, I, p. 111; cf. the scholiast's exaggerationof this point (F. Duebner, ScholiaGraecain Aristophanem, Paris 1877, pp. 97-98): (323) /3XEITEVVVJ KTX ... ...

[EIKOTW# 8E EITE 7TrOp?

T1 V

Ha'pv6a KaTE'pXEo-OataviTa. rvavat yap at rv KOpV4ta 8wa To v io# aEt 0-VVVE4ktlg ktvovrat.

22Theophrastos,de signistempestatum, 20, 24, 43.

O6pE'WV

SOMEIONICARCHITECTURALFRAGMENTS FROMTHE ATHENIANAGORA (PLATES12 and 13)

INCE 1929, when HomerThompsonand I arrivedat the Schoolat Athens on the same ship,we have been lookingat Greekmonumentsand discussingthem together, to my enormousprofitand pleasure.It seems fitting,therefore,thatthereshouldbe of over a half centuryof friendship,a offeredhere, as a wee token of my appreciation the AthenianAgorawhichwe have from adespota wordabouta few of the architecture discussedtogether.These remarksmust be consideredonly preliminary;a fuller acandwill be presentedwiththe several countof the capitaland basesis moreappropriate Agora. in the found and bases capitals Ionic other S

A much mutilatedbut still impressiveIonic capital(A 616, Fig. 1, P1. 12:a,c, e) was foundin the areaof the Odeionin 1936, brokeninto two pieceswhichhave been put together.The sides of the top two thirdsof the blockhave been cut downto make a circularshaft,0.397m. in diameterat the top, restingon the spiralsof the voluteand the necking.The top of the volute as cut downis roughlychiseledas is the bottomof the volutes which has been cut to make them level with the originalbottom of the necking,so that the bottomof the blockwouldrest firmlyon the whole lengthof the capital.The surfaceof the cylinderis more finelytooth chiseledbut still a roughfinish. In the centerof the top of the cylindera roundhole, 0.10 m. in diameterand 0.038 m. deep, was cut, presumablyat the same time that the cylinderwas cut. To what period or for what purposethis re-use is to be assignedis impossibleto determinewith any certainty.One mighthazarda guess that it was in the EarlyChristianperiodthat our capitalserved as a base plinth and bottom drum of a column, perhapsof a second storey.Ourconcernhere is withthe capitalin its originalstate. Enoughremainsof the originalform of this capitalof islandmarbleto drawattention to the fine qualityof the carvingand surfacefinishand the eleganceof the design as well as its unusualcharacterin whichthe eggs of the neckingare fullycarvedon one side but only blockedout, readyfor the detailto be delineatedin paint,on the other. Of the severalcapitalsknownfrom Athensand vicinityin whichone side is carvedand the other painted,one came to mind as strikinglysimilar;a look confirmedthe suspicion thatour Agoracapitaland the one lying (as recentlyas 1976,but not in late 1980) amongotherblocksbelowthe bastionof AthenaNike on the south side, nearthe onetime entrancegate to the Acropolis(Fig.2, P1.12:b,d, f), are a pair.Dimensions(Figs. 1, 2), kindof marble,surfacefinishand tooling,and detailsof design (P1.12:a-f) are the same, so that there can be no doubtthat both were madeto serve togetherin the

SOME IONICARCHITECTURALFRAGMENTSFROM THE ATHENIAN AGORA 83

same building.The capitalon the slope below the Acropolisserves to show how the Agoracapitalappearedbeforeit was cut down. The "Acropolis"capitalwas publishedby Mobiusin 1927 and dated440-420 B.C.1 In the half centurysince 1927manymore Ioniccapitalshave come to light throughout the Greekworldand have addedan enormousamountof new evidencefor our understandingof Ioniccapitalsat the same time that they have led to a realizationthat questions of originand developmentand of chronologyare farmore complexthanwas once thought.Detailed,comprehensive,and exhaustivestudiesof proportionsand detailsof form and ornamenthave been made by severalscholarsas they have presentednewly discoveredexamples.2This is not the time nor the placeto addto these valuablestudies with theirusefultablesof figures,but it may be of interestto makea few observations aboutsome of the characteristics of this pairof capitals. First,be it notedthatthe dowelhole nearthe centerof the top of eachcapitalwith a pry hole appropriately near (Figs. 1, 2) indicatesthat epistyleblockswere priedinto placeanddoweledabovethe capitals;these piecescappedthe columnsof a buildingand were not fromfreestandingvotive columns.The materialis clearlyof the highlycrystalline qualityregularlyrecognizedas islandmarble.'It shouldbe kept in mind that the in Athensis almostunknownafterthe Persianwars. use of islandmarblefor buildings The voluteson bothfacesare concaveratherthanconvex in relationto the bordering astragalbut are shallowrelativeto capitalsof the Pericleanperiod.Distinctiveis the factthatthe eye of the volute, the echinus,andthe cornerpalmettelinkingechinusand volute are treateddifferentlyon the two faces. On one the large eye characteristic of Attic Ionic is delicatelycarvedwith a rosette of eight double petals, its center set in separately;on the otherthe eye is flatand smooth.The echinusof ovolo profile,which is separatedfrom the lower borderof the volute by a plainsurfaceon line with the front of the volute (not directlyover the ovolo as commonlyon Attic capitals)is on one side fully carvedwith oval eggs and dartswhichextendthe full heightin between; on the otherside the shapeof the egg is blockedout on the flat surface,but no rounding of the egg or its borderhad been carved,althoughthe dartsare cut free. The five petalsof the half palmetteswhichfill the anglesare delineatedwith slightconvexityon whatwe may now call the "carved"side; on the otherthey are merelyblockedout. In 1"Attische Architekturstudien,"AthMitt52, 1927, pp. 171-173, Beil. XIX:2, 3 (= Mobius). 2E.g. W. Wrede, "Ein ionisches Kapitell in Athen," AthMitt55, 1930, pp. 191-200; F. Benoit, "Le chapiteauionique de Marseille," RA 43, 1954, pp. 17-43; W. Alzinger, "Ionische Kapitelle aus Ephesos I," OJh 46, 1961-1963, pp. 105-136; R. Martin, "Chapiteauxioniques de l'Asclepieion d'Athenes," BCH 68-69, 1944-1945, pp. 340-374, "Chapiteauxioniques de Thasos," BCH 96, 1972, pp. 303-323, BCH, Supply.I, Etudesdeliennes,Paris 1973, pp. 371-398, "Chapiteauionique d'Halicarnasse,"REA 61, 1959, pp. 65-76; G. Gruben, "Das archaischeDidymaion," JdI 78, 1963, pp. 115-142; D. Theodorescu, "Un chapiteau ionique de l'epoque archaiquetardiveet quelques problems concernantle style, 'aHistria,"Dacia 12, 1968, pp. 261-303; A. Bammer, "Beitragezur ephesischen Architektur," OJh 49, 1968-1971, pp. 1-22; "Beobachtungenzur ephesischenArchitektur,"AA (JdI 87), 1972, pp. 440-457. 3Pace Mobius, p. 171, who describedit as Attic.

84

LUCY SHOE MERITT

(D N-

0)

o

o

m

0.

0. _fli

/ - C'

/ ,/'

01

0

1\:

(0

-A0

FROMTHEATHENIANAGORA 85 FRAGMENTS SOMEIONICARCHITECTURAL

N.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

N~~~~~~~~~~~~~~~~O /~~~~~~~~~~~~~~~~~~~~~~O

Ii

'~~~

I

1

I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

'

~

~

~

~

~

~

~

'1C

~

~

~

~

~ ~

0

0

~~~~~~~~~~~~~~~~~~~~~~~ N.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a

~~~~~~

/

L~~~~~~~j~~~~~~~~~~~~~0

LJ

z

86

LUCYSHOEMERITT

otherwordswe have to do with one side completelycarvedin the Ioniantradition,the otherwhatwe mightcallhalfcarved,half painted.Examplesof capitalswith full carving on one side and on the other either no carvingat all or the volute indicatedbut the abacusand echinusleft plainare knownboth in Atticaand elsewhere,4but the treatment of the echinushere is unusual;anotherexamplewith such treatmentis known, also fromAttica,the pairof capitalsfromStavroandJeraka(see footnote4 above). The rosettein the eye has been discussedby Martin5andshownto be characteristic especiallyof Thasiancapitalsof the grouphe datesto the latterpartof the 6th and the early5th centurieswhichshow the shift fromCycladicto Anatolianinfluencein Thasos examplesat Neapolis,Histria,and and whichare to be associatedwith contemporary To be sure, the decoratedeye will become a featureof Ionic on the Halikarnassos. Acropolisof Athensin the last thirdof the 5th century,but the rosettesof A 616 and its companionare closer to those of the earliergroup.Two fragmentsof volutes with the volutes indicatedentirelyin painton a smoothflat surfacehave been found in the Agora(A 991 and A 1103),bothwiththe tracesof a rosettein the eye. They showthat the rosettewas sometimesused by Atheniansin their paintedcapitalsas well as in the betterknowncarvedcapitalsof the Acropolisbuildingsof the 5th century. In fact the Agoraexcavationshave addeda notablenumberof paintedcapitals, to the knowncorpus;some have carved some nearlycompleteand othersfragmentary, volutes with other detailpainted,othershave all decorationpainted.6They strengthen the impressionthat paintedratherthan carvedcapitalsare commonin Atticathroughout the 5th century,perhapsfavoredexceptfor the groupof buildingson the Athenian Acropolisand a few other worksof the architectsconcernedthere. What then of the groupof half-painted,half-carvedcapitals?Do they representa movementfrom a late carvedcapitaltowarda revivalof the earlier6th-century 6th-centuryIonian-influenced be Attic paintedtraditionto seen in the manypainted5th-centuryforms?Or are they moving from the 6th-centurypainted(largelyvotive) towardthe full carvingof the IlissosTemple,Propylaia,Nike Temple,andErechtheion? Let us look at what suggestionsthe bolsterof A 616 has to offer.The relatively form, only slightlycut in concavelyeitherhorizontallyor vertiearly,still well-rounded whichis familiarfromcapitalsof the latter to the decoration contrast cally,is in apparent capitals(footnote4 half of the 5th century,but whichappearsalso on the Stavro-Jeraka three flutes (here The four Archaic. bordering astragals above) usuallyconsideredlate 4E.g. Stavro and Jeraka,Mobius, pp. 167-171, fig. 3, pl. XXVII; Athens, Libraryof Hadrian-Monasteraki Museum, Wrede, op. cit. (footnote 2 above), pp. 191-200, figs. 1-4, Beil. LXII-LXIV; Delphi, Courtyardin front of Museum, found re-used in the Early ChristianBasilica and reported by Georges Daux, "Chroniquesdes fouilles en Grece en 1959," BCH 84, 1960, p. 756 (I owe permissionto mention it here to the kindnessof Pierre Amandry,Director of the French School); CorinthA-989 (unpublished,but noted here with the kind permission of Charles K. Williams); Chios, AntJ 39, 1959, p. 206, note 3, pl. XXVIII; Samos, AthMitt72, 1957, pp. 106-109, Beil. 108, pl. XV; Halikarnassos,REA 61, 1959, pp. 65, 68, pls. I, II. 5BCH 96, 1972, pp. 315-323. 6These will be treatedall together elsewhere.

SOMEIONICARCHITECTURAL FRAGMENTS FROMTHEATHENIANAGORA 87

shallow)in the centralarea of the bolstercomprisethe alternatebolsterdecorationof 5th-centuryAttic Ionic, the alternativeto the broad,flat center band (either in low relief or painted)which is the more common and is regularon capitalswith mostly painteddecoration.'Does this patterntie A 616 with the latergroup,or is it one of the first examplesof a style beginningmuch earlierbut to becomeincreasinglypopularin the latterpartof the 5th century?It is worthrememberingthat A 616 is linkedwith the Stavro-Jeraka capitalsnot only in this bolsterpatternbut also in the half-carvedechinus on one side and the islandmarble.The designof the bolsterappearsto be an attempt to combineelementsof the Ionian(bothAnatolianand Cycladic)bolster(in whichthe wholewidthis dividedinto flutes separatedby astragals)with the traditionalAttic restrictionto a bandin the center.A centralbandwas dividedinto two astragalsat Halikarnassos,clearlyAnatolian,and on a capitalat Delphi of coarse-grained marble,and into threeastragalsin two earlycapitalson Delos and lateron the capitalin the Athenian Libraryof Hadrian.8 Both the rosette carvedin the eye and the carvedegg and dart on the echinus alongwith the carvedpalmettein the angleswith the volute are distinctlyIoniancharacteristics,and the bolsterdesign is at least partiallyIonianin inspiration.The strong currentof Ionian ideas, dress, culture, and artisticstyle which begins to pour into Athens in mid-6thcenturyand continuesthroughoutthe four decadesof Peisistratean and Peisistratidperioduntil the last decadeof the centuryis widelyrecognizedin the field of sculpture.Payne9long ago realizedthat in the sculptureof the period two groupsexist: (1) the pieceswhichcan be consideredIonian,the workof Ionianartists workingin their home style whetherin Ionia (the pieces then imported)or as foreignersin Athens, and (2) the far more numerouspieceswhichcombinenative Attic and importedIonian elements to make a new truly Attic style from which the late Archaicand then earlyClassicalAttic styles grew.It is only reasonableto ask whether some similarsituationmay not have maintainedin architecture.Whole architectural membersfully carvedare not apt to have been importedas individualstatues could have been, but the fact that islandmarblewas importedmay well suggestthat Ionian stone cutterswent with it to carve the decorationon some of the marble.The many votive Ionic capitalsfrom the 6th centuryfound on the Acropolisat Athens10tend to have the decorativepatterns,Ionicthoughthey be in origin,painted;painteddecoration was naturalto the local Attic Doric architecture.But there must have been some demandin PeisistrateanAthens for architecturalmemberscarvedby Ioniansin the 'Variantsare a centralcarved astragalending at the top in a lotus blossom, M6bius, pp. 165-167, Beil. XVIII:5,6, and a lotus stem and blossom added to the centralband, ibid., Beil. XVIII:7,8. 8Halikarnassosand Delphi, see footnote 4 above; Delos, R. Martin, BCH, Suppl. I, EtudesdMliennes, 1973, p. 376, fig. 5 and p. 384, fig. 11; Athens, Libraryof Hadrian,Wrede, op. cit. (footnote 2 above), p. 196, figs. 2, 3, Beil. LXIII:2. 9H. Payne, ArchaicMarbleSculpturefromtheAcropolis,[London] 1936, pp. 55-63. 10AntDenkI, pls. 18, 29; R. Borrmann,"Stelen fUr Weihgeschenkeauf der Akropolis zu Athen," JdI 3, 1888, pp. 275-277; 0. Puchstein, Die ionischeSaule, Leipzig 1907; A. E. Raubitschek,BIABulg12, 1938, pp. 162-172 and numerous later studies.

88

LUCYSHOEMERITT

more elaboratestyle in which they were trainedand which presumablyhad as much appealto some Atheniansas Ioniansculpture.Can our pairof capitalsbe thoughtof as such Ionianwork?The bolsterdesignmightseem to give pause, if not also the halfThe egg and darthas been carvedside, but one paintedside is not unknownin Ionia.1" thoughtto be laterthanlate 6th century,but neitherthe egg bordernor the dartare as sharpor as sharplyseparatedfromthe egg as usualin the late 5th century.It shouldbe and memberof eitherGreekorderis the individuality realizedthat in no architectural firm independenceof the designermore freelyexpressedthan in the Ionic capital.No regulardevelopmentcharacterizesthese capitalsaroundthe Greek world. Except in certaingeneraltendenciesin proportionsand decorationand in certaingeographical predilections,there is in most capitalsa minglingof proportionsof form and of details of ornamentwhichfrequentlybafflesattemptsto date from this or that detail.Perhaps we have in this pairof capitalsnot the renewalof Ionianinfluencein Attic architecture in the latterpartof the 5th centurybut ratherthe work of an Ionianof that earlier periodof Ionianties a centuryearlier,an Ionianexperimentingwith detailssuggested by his associationwithAttictraditionsandcombiningthem withhis own fromIonia. Of possiblesignificancein our considerationof A 616 and its fellow capitalare piecesfoundin the excavationsof the AthenianAgora.Three threeotherarchitectural examplesof an Ionic base torus (A 829, A 1467, A 4543, Figs. 3-5, P1. 13:a-d) have been foundin contextsof no significance,from marblepiles in the easternpartof the area.12All three are of the same island marble, have the same profile and division into six horizontalchannels separatedby a single fillet, and have the same dimensions if al-

lowanceis madefor wear.They are clearlythe uppermemberof Ioniccolumnbasesof whichthe lowerelementmayhave been a plaincylinderas in AgoraA 197413(P1.13:e) or a cylinderenlivenedby horizontaldivisionsincisedas in AgoraA 3454 (P1.13:f)or this lattermay be the most by scotiaein the more Ionianpiece from the Acropolis;14 likely.The half-roundformof the torusand the characterof the channelingdividedby to those of late Archaicbases.15 singlefilletsare comparable The diameterof these tori at the top measures0.553 m. (A 1467), 0.546 m. (A 4543), and 0.552m. (A 829, whena and b are set together).Traceof the settingof the "Chios, Samos, and Halikarnassos,see footnote 4 above. "2A829 found in demolition marblepile in Section fQ, A 1467 in marbledump east of Stoa of Attalos, A 4543 in area east of Libraryof Pantainosnear Chapel of Aghia Matrona, included here with the kind permissionof T. Leslie Shear, Jr. 13L.S. Meritt, "The GeographicalDistributionof Greek and Roman Ionic Bases," Hesperia38, 1969, p. 188, pl. 49:c. 14L.T. Shoe, Profilesof GreekMouldings,Cambridge,Mass. 1936, p. 156, pl. LXXII:11; B. Wesenberg, Kapitelleund Basen, DUsseldorf1971, p. 119, no. 18, fig. 253. In fact the dimensions of the piece would permit it to be associatedwith our tori, which are very similar to the torus that is linked with this lower element of a base. 15E.g.Samos, Heraion, Second Dipteros, Profilesof GreekMouldings,pl. LXV:3; Athens, Acropolis, ibid., pl. LXV:5;Delphi, MassilioteTreasury(toichobate), FdD II, iii, 1, Paris 1923, p. 53, fig. 54, pl. XVI; Thermi, Wesenberg, op. cit., figs. 265, 266.

FRAGMENTS FROMTHEATHENIANAGORA 89 SOMEIONICARCHITECTURAL

? ,552 049

r

013

e

+526

-

HI 0~~~~~~~~~~~~~~~~

kilo,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~c

K-

'0515 1-

05

0

,10

M.

;-

'V~~~~~~M A 829B

A 829o-

50

5

W.B.D.,JR:-I1976

8

G.

3. Ionic base A 829

shaft above on A 829a gives a dimension of 0.526 for the diameter of the apophyge (if not an astragalbelow it) at the bottom of the shaft. If it is recalled that the diameter of the bottom of the echinus of capitalA 616 and its mate is 0.454 m., it will be seen that an associationof A 616 with the tori would give a difference of about seven centimeters between the top and bottom diameters of the Ionic column, an acceptablefigure. The empolion cuttings in the tori range from 0.066 to 0.07 m. square, that in the bottom of the capital0.073 m.; associationis permissable.Finally, the marble links the two capitals

90

LUCY SHOE MERITT

1042 MAX.

V/1~

~

3 1553

~

1066 SQ.

L' 0

10

,05

A 1467 0

150 W.B.D.JR.-1976

FIG. 4. Ionic base A 1467

and the three tori. It is impossible without further pieces and other evidence to prove the association of the capitals and tori, but a connection is both possible and even probableconsideringthe close similarityof the marble. If these five pieces are to be associated, the likelihood that they belong to a building of the late Archaic period is strengthened, and we have to look, as further excava-

SOMEIONICARCHITECTURALFRAGMENTSFROM THE ATHENIAN AGORA 91

H

048

MAX.

s--

,546 -0

070 SQ.

NL.

067 x,072

0

105

,10 7_ M.

A 4543 0

,5M W.B.D.,JR.-I976

FIG.5. Ionicbase A 4543

tion reveals more foundations, for one which would accommodate possibly a tetrastyle porch, since we seem to have evidence for three columns at least. If the date could be accepted, how very tempting to think of these pieces of strong Ionian influence as belonging to one of those buildings associated most particularlywith Peisistratos himself, namely his Enneakrounoswhich sits waiting for columns to fit three places in the

92

LUCYSHOEMERITT

foundationsof the buildingpreservedin the southeast corner of the Agora, dated to the third quarterof the 6th century B.C.16 Lucy SHOEMERITT UNIVERSITY OFTEXASAT AUSTIN

Departmentof Classics Austin, TX 78712 "6H.A. Thompson, AthenianAgora, XIV, TheAgoraof Athens,Princeton 1972, pp. 197-199, fig. 50; J. McK. CampII, The WaterSupplyof AncientAthens,diss. Princeton, 1977, chap. III, pp. 11-28. If, however, the four cylinders (Wesenberg, op. cit. [footnote 14 above], p. 119, note) on the Acropolisshould prove to belong with the Agora tori and the Acropolistorus prove to be from the same series as our tori, we must reckon with four columns. A recheck of the characterof the marble of the pieces now on the Acropolis must be made before any connection between the Agora and the Acropolispieces can be suggested.

PLATE

12

a. Ionic capitalAgora A 616, carved side

e

ji

7n_5

b. Ionic capitalunder the Nike Bastion, carvedside

Jo

t

r

~&Jf r

..

4

c. Ionic capitalAgora A 616, top

d. Ionic capitalunder Nike Bastion, half-carved,halfpaintedside

e. Ionic capitalAgora A 616, bottom

f. Ionic capitalunder Nike Bastion, bolster and carved side

PLATE

X

.~~~~~~~~~~~~~I-. .4

a. Ionic base torus Agora A 1467, top

c. Ionic base torus Agora A 829

b. Ionic base torus Agora A 1467, side view

d. Ionic base torus Agora A 4543

I.~~~~~~~~~~F

ot]~~~~

e. Ionic base Agora A 1974

f. Ionic base Agora A 3454

13

A MINIATUREATHENA PROMACHOS (PLATE 14)

HE GODDESSATHENA is surely most closely associatedwith the city which . derivedits name from her and whichviewedher as its own specialchampion.Reflectionsstemmingfroma tiny and very modestdepictionof this deity, recentlydiscovered in excavationsat Nemea, are offeredas a smalltoken of esteem and gratitudeto ProfessorThompson. The bronzefingerringillustratedon Plate 14:aandb whichbearsan intagliodevice representingAthenain fightingpose was discoveredin 1976 nearthe southeastcorner of the Templeof Zeus at Nemea.1It appearedin disturbedcontext of the first half of the 3rd centuryafter Christ,havingbeen churnedup from earlierlevels below.2The style of the engraving,as so often with bronzeintaglios,is of mediocrequality.Precise datingof this sort of ring is notoriouslydifficult,and shapehas to assumea disproportionaterole in chronologicaldetermination.On our ring the slenderhoop which supportsa curvingoval bezel (the curvaturevisible in Plate 14:b)3suggestsa date in the secondhalf of the 5th or possiblyinto the 4th centuryclose to a series of ringsfrom

T

Olynthos which provide, once again, an important terminusante quem. Our understand-

ing of the historyof Nemea leadsus to believe, however,that the site was abandoned duringthe firsthalf of the 4th centuryfollowingwidespreaddestructionin the late 5th century.5Probabilitywouldthereforesuggestthat our ring must date no laterthan the activityin the sanctuary.6 latterpartof the 5th century,a time of well-documented view to the right (as seen in the cast, P1.14:c) Athenais portrayedin three-quarter in time-honoredfightingpose. The spearis poisedat head level in her righthand, the lInventoryGJ 14. The bezel measures0.016 by 0.012 m. Most of the hoop is missing, andlthe metal is pitted. Unpublished. I am gratefulto my husband, Stephen G. Miller, for encouragingme to publishthis piece. As always, I have profitedfrom discussionwith him on problemswith my research. For specialabbreviationsused in this articlesee footnote 8. 21twas discovered on July 20, 1976 in Layer 31 with grid coordinates L/13-14/1, just off the Northwest corner of the "Nu Structure".For the area of its discovery see Stephen G. Miller, "Excavationsat Nemea, 1976," Hesperia46, 1977, pp. 4, 5, figs. 2, 3, respectively.For a discussion of activity of the 3rd century after Christ see ibid., p. 3. Note that Early Christiangraves of the 5th-6th centuries after Christ also pepperedthe area (ibid., p. 3, pls. 1:e, 2:a). See also plan, Hesperia47, 1978, p. 60, fig. 1 and aerialphotograph,pl. 10. 31t belongs to Boardman'sType II, consisting of rings dating from the middle to the end of the 5th centuryB.C.(J. Boardman,GreekGemsand FingerRings, New York 1970, pp. 212-214). 4See D. M. Robinson, Excavationsat Olynthus,X, Metal and Minor MiscellaneousFinds, Baltimore 1941, pp. 132-155. See esp. rings of Robinson's Type II, nos. 474-494. 5See excavation reports of Stephen G. Miller, Hesperia49, 1980, p. 186; Hesperia47, 1978, p. 65; Hesperia46, 1977, pp. 9-10, 21. 6See excavation reportsof Stephen G. Miller, Hesperia48, 1979, pp. 81-82 and Hesperia47, 1978 for a sacrificialdeposit; Hesperia46, 1977, pp. 19-20 for bronze casting.

94

STELLAG. MILLER

shield raisedobliquelyto be seen from the outside on her left arm. With regardto stance,the feet are plantedfirmlybut slightlyaparton the ground,and the shoulders are thrownbackin somewhatexaggeratedfashion.Detailsaboutthe head are indistinct but she evidentlywearsthe familiarhelmet. Her long girdedpeploshas no overfold, and she seems to wear no aegis. The peplosclings to the torso to reveal the breasts beneath,and on the lower body it is stretchedin wide obliqueridgesacrossthe legs from ankleto mid-thigh.An importantadditionto her costumeis the scarflikemantle thrownover the shoulders.This garmentis familiarfrom the mediaof Archaicsculpand frequentlyexploitedits decorativequaliture and vase paintingwhichappreciated on our ClassicalAthena,however,is an archaismof the sort which ties. Its appearance wasjust becomingpopularwithartistsin the late 5th century.The laterdevelopmentof this sort of mantleinvolvesthe splittingof the draperyends into swallowtails,a change bracketedbetweenthe late 5th centuryand its whose inceptioncan be chronologically on Panathenaicamphorasof 363/2.7With regardto our firstdocumentableappearance ring,we mayconcludethat the presenceof the archaisticmantlein its earlyformwithout the distinctiveswallowtailembellishment,while offeringnothingdefinitive,at least accordswell withthe proposedlate 5th-centurydatingof the ring. A greatdealhasbeen writtenon manyaspectsof AthenaPromachosandhervarious manifestations,origins,and survivalin latermonuments.8The high regardAthenaen7See J. D. Beazley, TheDevelopmentof AtticBlack-figure,2nd ed., Berkeley and Los Angeles 1964, p. 98 (= Development);E. B. Harrison, The AthenianAgora, X, Archaicand ArchaisticSculpture,Princeton 1965, p. 61; C. M. Havelock, "The ArchaisticAthena Promachosin EarlyHellenistic Coinages," AJA 84, 1980, p. 42. 8In general see C. J. Herington, AthenaParthenosand AthenaPolias, A Studyin the Religionof Periclean Athens,Manchester1955 (= AthenaParthenosand Polias); review of AthenaParthenosand Polias by E. B. Harrison, AJA 61, 1957, pp. 208-209; C. J. Herington, "Athena in Athenian Literatureand Cult," in Parthenosand Parthenon,Greeceand Rome, suppl. to ser. 2, vol. X, 1963, pp. 61-73(= "Athena"). On the iconographyof the Archaic Promachos in sculpturesee H. G. Niemeyer, Promachos,Untersuchungenzur Darstellungder bewaffnetenAthenain archaischerZeit, Waldsassen 1960 (= Niemeyer, Pround friihklassichenZeit," AntP III, Berlin machos); idem, "Attische Bronzestatuettender sp-atarchaischen 1964, pp. 7-31; C. Rolley, "Statuette d'Athena Promachos," RA 1968, pp. 35-48; H. Herdejurgen, "Bronzestatuetteder Athena. Bemerkungen zur Herkunft des archaischenPromachostypus,"AntK 12, 1969, pp. 102-110; H. Kenner, "Athena und die Gbtterwelt der Austria Romana," OJh 51, 1976-1977, esp. pp. 110-126; Havelock, loc. cit. (footnote 7 above); E. Simon, Die Gatterder Griechen,Munich 1969 (= Simon), esp. pp. 185-193; K. Schefold, Mythand Legendin EarlyGreekArt, London 1966, pp. 58-60; Kunst, Munich 1978 (= Gbtter-und Helidem, Gbtter-und Heldensagender Griechenin der spdtarchaischen pp. 38-43; L. Lacroix, Les reproductions JdI 52, 1937, auf Vasenbildern," "Statuen densagen),p. 14; idem, de statuessur les monnaiesgrecques:la statuairearchaiqueet classique,Liege 1949, pp. 116-129, 281-286; D. Le Lasseur, Les dressesarmeesdans l'artclassiquegrecet leursoriginesorientales,Paris 1919, pp. 56-71. On Panathenaicamphoras see J. D. Beazley, ABV, pp. 403-417; idem, Development(footnote 7 above), pp. 88-100; idem, "Panathenaica,"AJA 47, 1943, pp. 441-465; K. Peters, Studienzu den panatheBerlin 1942; G. von Brauchitsch,PanathenaischePreisamphoren,Leipzig 1910; J. naischenPreisamphoren, Boardman,AthenianBlackFigureVases, London 1974 (= Boardman,BlackFigure Vases), pp. 167-177; J. Frel, PanathenaicPrizeAmphoras,KerameikosBook No. 2, Athens 1973. On the PheidianPromachossee E. Mathiopoulos,Zur Typologieder GbttinAthenaim funftenJahrhunFrankfurt 1947, pp. 73-76, dert vor Christus,diss. Bonn 1968, pp. 7-47; E. Langlotz, Phidiasprobleme, 108-109; G. Becatti, ProblemiFidiaci,Milan 1951, pp. 161-167.

A MINIATURE ATHENA PROMACHOS

95

joyed as city goddessin the religiousandculturallife of Athensis well documentedand universallyrecognized.The factthatan imageof Athena,thoughby no meansthe exclusive preserveof the Athenians,9willin antiquity(as well as today)have evokedimmediate thoughtsof Athensis also not likelyto be disputed.WhatI shouldlike to propose, however,goes beyondthis to suggestthe possibilitythat a particular iconographic representationof the goddess,namelythe FightingAthena,mayhave hadspecial,widelyrecognizedsymbolicand indeedpoliticalovertonesof considerabledurationin ancientAthens. To examinethispossibility,we mustturnfirstto the imageryof the ArchaicAthena. of the developmentof 6th-centuryAthens is beset with probOur understanding lems and controversiesat every turn, hamperedas we are by the paucityof contemporaryliterarysourceson the one handand the fragmentary natureof the archaeological evidence on the other.10A documentedevent, however, importantfor our purposes, concernsthe reorganizationof the PanathenaicGames which accordingto historical traditionoccurredin 566/5, an event often associatedwith the name of Peisistratos himself.11As is well known, the practiceof awardingPanathenaicamphorasas prizes was institutedin conjunctionwith this reorganization, and it is these vases whichreproduce the distinctiveFightingAthena, a figurewhichwill gracePanathenaicsfrom this time forward.A representative exampleof the 6th centuryappearsin Plate 14:d.12 C. J. Heringtonhas discussedat lengththe characterof the Archaicor "Solonian" Athenaand her specialprotectiverelationshipwith the Athenians.13 The FightingAthena wouldappearto be the visualcounterpart of this Athenafromaroundthe time of its appearanceon the Panathenaics.Certainlythe subjectmatterwhich incorporatedthe bellicosegoddessdevelopedan enormouspopularityin Athens from shortlybeforethe middleof the centurythroughthe laterArchaicperiod.J. Boardmanhas suggestedthat Herakles,who is featuredso prominentlyin Archaicpedimentson the Acropolis,might have embodiedthe kindof politicalsymbolismwhichis generallyacceptedin the Periclean age with its manyallegoricalrepresentations.14 Certainlythe expansionand cultural 9Note that the earliest representationsof Athena (from the first half of the 7th century) include a terracottafrom Gortyn on Crete (Simon, figs. 169, 170) and a Cycladicrelief pithos from Tenos (ibid., fig. 165). See Niemeyer, Promachos,pp. 56-64. On the worshipof Athena outside Athens see L. R. Farnell, The Cultsof the GreekStates I, Chicago 1971, pp. 308-313 and references. '0For a recent statement on these issues see T. L. Shear, Jr., "Tyrantsand Buildingsin Archaic Athens," in AthensComesof Age: FromSolon to Salamis, AIA Symposium Papers, Princeton 1978, pp. 1-19; also W. H. Plommer, "The ArchaicAcropolis:Some Problems," JHS 80, 1960, pp. 127-159. "J. A. Davison, "Notes on the Panathenaea,"JHS 78, 1958, pp. 23-42 with references; idem, "Peisistratusand Homer," TAPA86, 1955, pp. 1-21. '2Fromthe Athenian Agora (P 24661); publishedby J. McK. CampII, Godsand Heroesin the Athenian Agora (Excavationsof the AthenianAgora, Picture Book No. 19), 1980, p. 6, fig. 8. On Panathenaicssee referencescited in footnote 8 above: Beazley, Peters, von Brauchitsch,and Frel. '3Herington("Athena," esp. pp. 62-63) stresses that it is Athena's closeness to Zeus which is being emphasized in the myth of the Birth of Athena whose iconographichistory in Athenian art starts at this period. See also below, footnote 15. 14J. Boardman, "Herakles, Peisistratosand Sons," RA 1972, pp. 57-72; see also idem, Black Figure Vases,p. 216; and Le Lasseur, op. cit. (footnote 8 above), p. 78.

96

STELLAG. MILLER

floweringin Athens in the secondhalf of the 6th centurycan be linkeddirectlyto the politicsand artisticpatronageencouragedby the Peisistratids.Althoughit defies proof, it is hardto believethatthe propagandistic valueof allegoricalsymbolism(subtlein the case of Herakles,blatentwith regardto the FightingAthena) will have escapedthe noticeof the Tyrants. A quickreviewof the numerousAtheniancontextsin whichthe FightingAthena appears,besides the seeminglyofficialmarkof the Panathenaics,will underscorethe pervasivenessof the motif.First,one thinksof the Atticvase paintingswhichdepictthe Birthof Athena,a motifwhichbeginsa historyof greatpopularityin Athenianartjust aboutthe time of the reorganization of the PanathenaicGames.15This Archaicversion is, of course,the one in whichthe tiny armedgoddessspringsin fightingpose out of her father'ssplit head. Attic vase paintingalso enthusiastically adoptsthe old myth of the Rapeof Cassandrawho seeks refugeat the Palladion,a statuewhich, interestingly enough,afterthe mid-6thcenturyalso takeson the pose and attributesof the Fighting Athenaof the Panathenaics.16 A mythof primeimportancein this periodis the Gigantomachy in whicha battling Athena (and, notably,her protegeHerakles)playsa centralrole. Significantly,it too startsto appearin Attic blackfigurein the 560's and rapidlybecomesone of the most popularAttic themes.17Numerousvase paintingsfeaturingthis motifwere dedicatedon the Acropolis,18 and it is likely that the iconographyof the FightingAthena in fact derivesfrom this myth. The theme was featuredprominentlyin the marblepediment usuallyassociatedwith the so-calledPeisistratidTempleon the Acropolis,19 as well as on severalArchaicreliefsdedicatedon the same spot,20and it is not impossiblethat the subjectmatterwas woven into the peplos presentedto Athena at the festival of the Panathenaia.21 Finally,the FightingAthena is knownthrougha groupof late Archaic bronzestatuetteswhichwere dedicatedon the Acropoliswherethey were discovered.22 Takentogether,the potentialfor symbolicinterpretation is very strong. 15Onthe Birthof Athena see Herington, "Athena," pp. 63-64 and see comments, footnote 13 above; F. Brommer, "Die Geburt der Athena," RGZM 8, 1961, pp. 66-83; K. Schefold, "Die Geburt der Athena," in Gbtter-und Heldensagen,pp. 12-20; Boardman, Black Figure Vases, p. 216. See also the whiteground pinax found on the Acropolis, B. Graef and E. Langlotz, Die antiken Vasenvon der Akropoliszu AthenI, iv, Berlin 1925, pl. 109, no. 2578. '6See Boardman,BlackFigureVases,p. 230. "7Boardman,Black Figure Vases, p. 220; Schefold, Gdtter-und Heldensagen,pp. 54-66; see F. Vian, figures dans lart grecet romain,Paris 1951; J. Dbrig and 0. Gigon, Der Kampf Repertoiredes gigantomachies der Gbtterund Titanen,Olten 1961. See Farnell, op. cit. (footnote 9 above), pp. 308-309. '8Graefand Langlotz, op. cit. (footnote 15 above), pl. 108, no. 2559; ibid., II, i, Berlin 1929, pl. 21, no. 327:a-c. "9Forrecent bibliographysee M. S. Brouskari, TheAcropolisMuseum, Athens 1974, pp. 76-78, figs. 144-150; B. S. Ridgway, TheArchaicStylein GreekSculpture,Princeton 1977, pp. 205-209; and for a good illustration,Schefold, Gbtter-undHeldensagen,figs. 71, 72. 20Brouskari,op. cit., fig. 142, no. 121; fig. 247, no. 120; fig. 59, no. 12992 (terracottaplaque). 2"Onthe peplos see Deubner, AttischeFeste, Berlin 1932, pp. 29-34 with references. See also C. A. Forbes, "HE'1rToa,"RE XIX, i, 1937, col. 561; and comments by Schefold, Gbtter-undHeldensagen,p. 55. 22SeeNiemeyer, Promachosand idem, "Attische Bronzestatuetten"(footnote 8 above).

A MINIATURE ATHENAPROMACHOS

97

The question whether a monumental statue inspired these many representations, what it will have looked like, and where it might have stood, has been much discussed. Most scholars agree that such a prototype did in fact exist and that it must have been located on the Acropolis itself.23The well-known group of Archaic Attic vase paintings (including some discovered on the Acropolis)-, which show the Fighting Athena in association with an obvious sacrificialscene including altar and gift-bearing adorants, give us additionalsubstantiationconcerningthis Athena.24 Whether, however, this hypothetical statue was an out-of-doors monument as maintainedby some scholars, or a cult statue from an early temple on the Acropolis as believed by others, cannot be ascertained.25Certainly,the nature of the remains, particularly of the early temples on the Acropolis, fragmentaryor hypothetical, continues to be highly controversial. But the theory, most eloquently stated by Herington, that such a Promachosserved as cult statue for an early ancestor of the present Parthenon is still an appealing,though not provable, suggestion.27At the very least, it has the advantage of supplyinga necessarycult image for a probabletemple. The Persian destructionof the Acropolis demolished many landmarks,and with the subsequent refurbishingwe see evidence of the dawning of a new era. The little Angelitos' Athena (Acropolis 140) has been called a "herald"of the new Severe Style which permeates the artistic spirit of the first decades of post-Persian Athens.28Whether or 23See,for example, Herington, AthenaParthenosand Polias, pp. 40-42 and Harrison, op. cit. (footnote 8 above), pp. 208-209; von Brauchitsch,op. cit. (footnote 8 above), pp. 167-180; Niemeyer, Promachos, pp. 13-15; Schefold, JdI 52, 1937, p. 38; Frel, op. cit. (footnote 8 above), p. 9; Simon, p. 192; Boardman, BlackFigureVases,p. 167. 24SeeSimon, pp. 192-193; Schefold, JdI 52, 1937, p. 40. Among the vases are a neck-amphoraby the Painterof Berlin 1686 (Berlin 1686, ABV, p. 296, no. 4); a Panathenaicamphoraby the Princeton Painter (New York, MetropolitanMuseum of Art 53.11.1, ABV, p. 298, no. 5) and an Attic band cup (private collection, Simon, fig. 176). 25See comments especially by the following: Niemeyer, Promachos,pp. 13-15; Herington, Athena Parthenosand Polias, pp. 40-65 (and review by Harrison, op. cit. [footnote 8 above]); Schefold, Gbtter-und Heldensagen,p. 14; Simon, pp. 192-193; A. E. Raubitschek,Dedicationsfrom the AthenianAkropolis,Cambridge, Mass. 1949, comments on no. 329. Niemeyer has argued persuasivelythat an early (perhaps the earliest) bronze Promachos statuette from the Acropolis (Athens N.M. 6450) may reflect the cult statue of the Eupatrids("Das Kultbild der Eupatriden?"FestschriftEugenvon Mercklin,Waldsassen1964, pp. 106-111). The figure with appropriately poised arms but with columnar body in closed-foot position is dated by him to around the second quarter of the 6th century (Simon, p. 189, would date it to the mid-6th century). The Panathenaicswhich from the start reproducethe stridingAthena may then have initiatedthe spread-footpose which, as has been noted by Harrison(op. cit.) and others, is so unlikely in large-scalesculptureat this early period. It is, however, perfectly in keeping with the nature of the two-dimensionalmedium of vase painting which shows the Promachosin what is essentiallya profilesilhouette. 26See J. Travlos, PictorialDictionaryof AncientAthens, London 1971, "Athena, Archaios Naos," pp. 143-147; "Hekatompedon,"pp. 258-260; and "Parthenon,"444-457. See also Plommer and Shear, opp. citt. (footnote 10 above). 27Notethat Herington's belief in the existence of the large Archaic temple below the Parthenonwas "seriously weakened" by Plommer's article (op. cit. [footnote 10 above]; cf. Herington, "Athena," pp. 61-62, note 1). 28B.S. Ridgway, TheSevereStyle in GreekSculpture,Princeton 1970, pp. 29-30, fig. 39; Brouskari,op. cit. (footnote 19 above), fig. 248, no. 140.

98

STELLAG. MILLER

not she held a swinging spear in her broken upraisedright arm is not certain, but B. S. Ridgwayhas argued cogently that her quiet pose sets her sufficientlyapartfrom the old striding-and-attackingformula to suggest that the spear was held quietly in front as a support.29In any case, the second quarter of the 5th century sees the old Attacking Athena receding in importanceto be replaced by the armed but peaceful Athena best exemplified by the Pheidianversion of the Promachos.30 Although the Athena Promachos by Pheidias is in many ways a shadowy figure to us, we do, of course, know a certain amount about this famous statue.31We learn from ancient testimonia that it was created sometime after the victory over the Persians. In additionwe know that it stood on the Acropolis, that it was made of bronze, and that it was exceedingly tall. If we can trust the evidence of certain coins of the Roman Imperial period, it stood between the Erechtheionand the Propylaia.Most important,however, is the manner in which it evidently held the weapons at ease, the shield in one hand, the spear probablyin the other with its butt resting on the ground. It was clearly an influentialmonument which inspired numerous copies throughout the rest of antiquity and helped set the tone for the revised iconographyof the warriorgoddess. The Fighting Athena survived in the Classicalage as an indispensableingredient in representationsof certain old archetypalmyths: in the Gigantomachyas an important participant,and in the Ilioupersisin the form of the Palladion. Indeed these myths are elevated to new heights in the formulaic glorification of the triumph of order over chaos, civilizationover barbarianism,by their inclusion in the sculpturalprogramof the Parthenon.32At the same time the conservative Panathenaicspreserved the old image of the fighting goddess, even changing it periodicallyto look more archaic.This image was quite purposely used to underscore the age and continuity of a venerable institution. But as a symbol of the new Athens the Fighting Athena was apparentlyfelt to be too aggressivea symbol. Thus we see, for example, the reworkingof the iconographyof the Birth of Athena from the old version, with the goddess emerging ready for battle from Zeus' head, to that which was so proudlyproclaimedon the East pediment of the Parthenonwhere (however she is restored) she appearsas the peaceful, armed goddess who promotes wisdom and artistryin war.33In terms of modern public relations, Athena's image and her "logo" have undergone a significantshift of emphasis. What, then, of the late 5th-centuryring discovered at Nemea? The circumstances of finding prevent a definitive statement concerning the nature of its context. The 29Ridgway,op. cit., p. 30, arguingagainst Langlotz (in H. Schrader,E. Langlotz, W.-H. Schuchhardt, derAkropolis,Frankfurt1939, pp. 48-49). Die archaischenMarmorbildwerke 30Cf.Herington("Athena," pp. 68-70) for his interpretationof the Periclean,AeschyleanAthena. 31Pausanias,i.28.2; Scholiast on Demosthenes, contraAndrotion,13; see Mathiopoulos, op. cit. (footnote 8 above), with references. 32TheGigantomachyon the East metopes (F. Brommer, Die Metopendes Parthenon,Mainz 1967, pp. 198-209; East metope 4 shows Athena), the Ilioupersison the much damagedNorth metopes (ibid., pp. 218-221). Mainz 1963, pp. 142-144. See also E. B. Harrison, der Parthenon-Giebel, 33F.Brommer.Die Skulpturen "Athenaand Athens in the East Pediment of the Parthenon,"AJA 71, 1967, pp. 30-32.

ATHENA PROMACHOS A MINIATURE

99

possibilityof accidentalloss by its ownercannot,of course,be ruledout, but the appropriatenessof such an objectas a dedicationto the goddess'fatherZeus and its findspot in the innermostsacredareaof the sanctuarymakethe idea of a votive most attractive. Equallyincapableof proofis a suggestionthat the ringmay have belongedto an Athenian. The close associationsof the motif with that city, however,and the long iconographictraditionwhichlinks it so closelywith Athenianmonumentsof an earlierdate, lend considerableweightto the suggestion. Panathenaicsalthough, the ring is independentof contemporary Iconographically, like them, it showsarchaisticmannerismsin its treatmentof the goddess.What,then, is its significance?I shouldlike to speculatethat the ownerof the ringwas a citizenof late 5th-centuryAthens with conservativepoliticalleanings.His politics,I would suggest, are revealedby the ringdevice whichis so clearlymodeled,howeverloosely, on the old Archaicimageryof the fightingcity goddess.34She served in this context as a symboland a reminderof the gloriousdays of that overtlyaggressivepowerwhichbelongedto Athensunderthe tyrants.OurhypotheticalAthenianmay, then, have sought to expresssuch idealsthroughhis personalsealingdevice.35It was thus an appropriate and piousact to dedicatethis significantpossessionat the venerablePanhellenicsanctuaryof Zeus at Nemea.36 STELLAG. MILLER STANFORD UNIVERSITY

Departmentof Classics Stanford,CA 94305 340n ring devices and their relationshipto statues see most recently comments by G. Horster, Statuen auf Gemmen,Bonn 1970, pp. 46-47. 350n ancient seal usage see Kubitschek, "Signum," RE, 2nd ser., II, 1923, cols. 2361-2455; and for engravedrings and gems employed as seals see G. S. Dontas, "'ApXaat a'EXkYvtJat' 8, 'a-ta o~payi8E," 'ApX'E0,1955 [19611,pp. 1-21; Boardman,op. cit. (footnote 3 above), pp. 235-238, 428-430. For the use of an Athena Promachos ring device see a Samian stamped amphora handle (V. R. Grace, "SamianAmphoras,"Hesperia40, 1971, p. 60, pl. 14, nos. 54-56); and for the same motyifa Seleucid bulla (M. Rostovtzeff, "SeleucidBabylonia:Bullae and Seals of Clay with Greek Inscriptions," YCS3, 1932, pp. 20-21, pl. 2, no. 5). Both are of the later archaistictype of Promachosdiscussed by Havelock, loc. cit. (footnote 7 above). 36Fingerrings, plainor with sealing devices, are not uncommon dedicationsto the gods (cf., for example, IG 112,1409, partof the Parthenoninventories).

PLATE

14

c. Cast of ring GJ 14 a, b. Bronze ring, Nemea inv. GJ 14

d. Panathenaicvase, 6th centuryB.C. (AthenianAgora p 24661)

KLEONAI,THE NEMEAN GAMES, AND THE LAMIAN WAR T ^>HE

DEATH OF ALEXANDER in 323 B.C. unleashed anti-Macedoniansentiments throughoutGreece, and especiallyin Athens, with the ill-fated Lamian War as the direct consequence. As the excavations of the Athenian Agora have shown vividly, this war was a disaster for Athens. Homer Thompson has given us an example of the interruption of construction begun under the Lykourgan program,1and the archaeological pictureof economic stagnationbecomes one of increasingdeprivationas Athens continued towardthe real hunger of the early 3rd century B.C.2In retrospect, the LamianWar can be characterizedas a falteringand self-deluded step on the road to self-destruction, but at the beginning of the war Athens could reasonablyhope for success.3As is true of every war, the causes of the Lamian War were varied, but the purpose can be defined, simply and obviously, as anti-Macedonian.It was this negative force of unification which helped Athens gain her allies in the opening stages and which is symbolized by the rathergrandiose title of "Hellenic War" which was used officiallyby the Athenians themselves.4The purpose of this paperis to suggest that Kleonai, a polis not previously associatedwith the Lamian War, may have played a role at certain stages of the conflict. First it is necessary to review the events which led to the actual hostilities. At the Olympic Games which culminated on August 4, 324 B.C.,5 Alexander's representative, Nikanor of Stageira, made an announcement which could not have been much of a surprisesince a crowd of more than 20,000 who were directly affected by the proclamation had gathered to hear it. This was the order for the restoration of exiles.6 This order, popularenough in many quarters,would have forced Athens and Aitolia to give up claims on Samos and Oiniadai, respectively. The proclamation accordingly gave 'H. A. Thompson and R. E. Wycherley, The AthenianAgora, XIV, The Agora of Athens, Princeton 1972, pp. 21-23, 60-61, 172. I would thank E. Badian and A. Boegehold for their helpful suggestions and criticisms during the preparationof this paper.The responsibilityfor errorsof fact or judgment which remain, despite their good offices, is mine. In referencesbelow the followingabbreviationwill be used: Badian= E. Badian,"Harpalus,"JHS 81, 1961, pp. 16-43. 2Stella G. Miller, "Menon's Cistern," Hesperia 43, 1974, p. 209, for references and bibliography.

3As has been pointed out recently by F. W. Mitchel, "LykourganAthens: 338-322," Lecturesin Memory of Louise TaftSemple,SecondSeries, Norman, Oklahoma1973, p. 212. 4IG JJ2, 448, lines 43-44; 505, line 17; 506, lines 9-10; cf. Plutarch,Phokion,23.1. 5S. G. Miller, "The Date of Olympic Festivals," AthMitt90, 1975, pp. 223, 230. Although I realize that the translationof ancient dates into modern calendardates is a risky business which does not receive universal approval,it seems to me valuable for present purposes. Such a policy quickly reveals relative chronologyand gives a general indicationof absolute chronology. 6DiodorusSiculus, xvii.109.1; xviii.8.207; cf. M. N. Tod, GreekHistoricalInscriptionsII, Oxford 1948, nos. 201 and 202, and p. 297, for additionalreferencesand bibliography.

KLEONAI,THENEMEANGAMES,AND THELAMIANWAR

101

Athens and Aitolia pause, and it has been suggested that they may alreadyat this time (i.e. after the Olympic Games of 324 B.C. but before the death of Alexander) have sought an alliance based upon a mutual distaste for the return of the exiles.7 It is of some interest that the leader (apXE9Ecvpo) of the Athenian delegation to the Olympic Games of 324 was Demosthenes.8 In the meantime, no more than a few weeks before the end of the archonshipof Antikles (325/4 B.C.), Harpalushad arrived in Athens.9 We need not recount here all the events of Harpalus'sojourn in Athens, but we should recall three results of it: 1) the Athenian treasury, already healthy thanks to the policies of Lykourgos,10was augmented by the monies confiscated from Harpalus;112) the number of mercenaries centered at Tainaronin southernmost Lakonia, and availableto the highest bidder, was increased;12and 3) Demosthenes (along with others including Demades) was forced into exile, probablyno earlier than March, 323 B.C.13The situation in Athens was thus in some disarray, with several of her leaders in exile, but considerable material resources were availableand anti-Macedoniansentiment had been heightened. The means and the motive were present, and the opportunitysoon came with the death of Alexander in Babylon on June 10-11, 323 B.C.14 We do not know how much time elapsed between Alexander's death and the first active move of the Athenians toward revolt (but see footnote 26 below). Early reports of his death, in the face of many previous false rumors, were regardedas premature,but confirmationfinally came.15This confirmation cannot have arrivedby the beginning of the new archonshipof Kephisodoroson July 12, 323 B.C.16but the first rumors had surely reached Athens in the month since Alexander's death. These may not have been sufficient to supportopen revolt, but they will have encouraged Athens to continue her policy of foot draggingvis-A-visthe restoration of the exiles and also the pursuit of alliance with the Aitolians. We can see the continuation of such a policy in the re-election of Leosthenes as strategos for the new 7For the properreadingof the heading of the alliance between Athens and Aitolia see F. W. Mitchel, "A Note on IG 12 370," Phoenix 18, 1964, pp. 13-17. For a summaryof the question of its date see L. Moretti, IscrizionistoricheellenisticheI, Florence 1967, no. 1, pp. 1-2. Further,see also below, footnotes 17 and 26. 8Deinarchos,i.81-82, 103. These may also have been the Olympicsat which Demosthenes defended the histories of Thebes and Olynthosagainstthe libels of one Lamachos,a pro-Macedonianencomiast; cf. Plutarch,Demosthenes,9.1-2. 9Badian,pp. 42-43, for the chronology of Harpalus'activities, and of Demosthenes' trial and exile. Although it makes no difference to the relative chronology, Badian'smodern dates ought, perhaps,to be revised by one day in accordwith B. D. Meritt, TheAthenianYear,Berkeleyand Los Angeles 1961, p. 133, where the year of Antikles is reckoned as ending on July 22, 324 B.C. Meritt's dates are those used throughoutthis paperunless otherwisespecified. "Mitchel, op. cit. (footnote 3 above), pp. 190, 210, et passim. "Diodorus Siculus, xviii.9.1-4.

'Badian, pp. 25-28. '3Badian,p. 43. 4D. M. Lewis, "Two Days," CR 19, 1969, p. 272. 'Diodorus Siculus, XVIII.9.4; Plutarch,Phokion,22.3-4. '6Meritt,op. cit. (footnote 9 above), p. 133.

STEPHENG. MILLER

102

year. Leostheneshad been oi-parrjqyoS EMrtn dpaP in 324/3 B.C.,17and his military leadershipwas to continueinto the waruntilhis deathduringthe seige of Lamiain the winterof 323/2 B.C.18 Perhapsduringhis firstyearas strategosin 324/3 B.C., or perhaps duringthe precedingyear,19Leostheneswas responsiblefor bringinga largenumberof This group,plus other mercenariesalreadythere and those mercenariesto Tainaron.20 of Harpalus(see above), were those with whom Leosthenesenteredinto negotiations, at firstwith the secret aid and consentof the AthenianBoule but, after the death of Alexanderhad been confirmed,with the open (OavEpcW) supportof the Demos.2"The resultwas the acquisitionof a substantialarmyfor Athens. The time was now the summerof 323 B.C., the archonshipof Kephisodoroshad allijust begun, and the Athenianswere readyto move towardopen anti-Macedonian ances. The first of these, for which considerablegroundworkhad been done already (above, footnote 7 and text ad loc.), was soon concludedwith the Aitolians,almost A generalidea of the time of the deathof Alexander.22 certainlyafterthe confirmation whichwas requiredfor this diplomaticoffensivecomes froma list of the alliesprovided "7G.Mathieu, "Notes sur Athenes 'a la veille de la guerre Lamiaque," RevPhil3, 1929, p. 162. It is possible that Leosthenes had been a general even earlier but I can find no proof of it. Mitchel, op. cit. (footnote 7 above), p. 16, paraphrasingDiodorus Siculus, xvii.110.1 and 111.3, says: "In the archonshipof Antikles (325/4) Leosthenes was selected by the boule and went to treat with the Aitolians, who were hostile to the king, concerningan alliance ... ." Even though Diodorus does list these activities of Leosthenes under the year of Antikles, I doubt that we can date them so confidentlyto 324/3 B.C. First of all, Diodorus freqentlygets ahead of himself. Just so he mentions, for example, the recall of the exiles which was announcedat the OlympicGames of 324/3 B.C. under his general heading for the archonshipof Chremes in 326/5 (xvII.109.1). Secondly, Diodorus represents Leosthenes' negotiations with the Aitolians as occurringafter his selection by the mercenariesat Tainaronas their leader, and this selection apparently av'ToKpatook some time (xvII.1 11.3: TO 8E TEXEvTa-LOV AEWOn-64 -v TOV'A0d vaio v ... IELoVVTO0TpaTqyov are set off Aitolians the with and later the mercenaries with Leosthenes of activities these Finally, Topa). by Diodorus againstAlexander's campaignagainst the Kossians. This campaigntook place after the death of Hephaistionin the winter of 324/3 (Plutarch,Alexander,72.3; Arrian,vii.15.1-3). Mitchel may be correct in placingthe first negotiationsof Leosthenes with the Aitolians before the death of Alexander but is surely wrong to place them earlier than the winter of 324/3 B.C. That is, the negotiations belong, at. the earliest, only a very few months before the death of the king. Indeed, Diodorus later (xviii.9.4) states that the secret negotiationswith the mercenarieswere taking place before the death of Alexander had been confirmed(i.e. after the rumorsof his death?) but that the Aitolian negotiationsfollowed the confirmation of Alexander'sdeath. So too Badian (p. 38) seems to place Leosthenes' secret negotiations with the mercenariesearlier (autumn324) than demanded by the text of Diodorus. It seems to me that Diodorus in xvii.111 is simply describingthe same situation which he discusses again in its properchronologicalposition at xviII.9. The problem is not with Diodorus' facts but with his organization, style of presentation, and tendency to redundancy. 18Diodorus Siculus, xvuii.13.5. '9Badian (pp. 26-27, cf. p. 38) places this repatriation in 325/4, but I can see no certain evidence for

this date. It appearsthat this activity of Leosthenes could have occurred during the early part of the archonshipof Hegesias;that is, duringthe early partof Leosthenes' first attested year as strategos. 20Pausanias, 1.25.5; vii.52.5.

2'DiodorusSiculus, xvii.111.3; cf. above, footnote 17. Does the shift by Diodorus from the Boule to the Demos imply that a formalvote was taken and a psephismapassedby the Ekklesia? 22DiodorusSiculus, xviii.10.5 and 11.1; cf. footnotes 7 and 17 above.

KLEONAI,THENEMEANGAMES,AND THELAMIANWAR

103

by Diodorus and from two Attic inscriptions.Diodorus' list is, ostensibly, in chronological order of the time when each ally joined with Athens, and I give it here:23 Aitolians Athamanians Thessalians, except Pelinna Leukadians Molossians (those subject to Aryptaios) Oitians, except Heraklea PhthiotianAchaians, except Thebes a few Illyrians Melians, except Lamia a few Thracians Dorians Karystosalone of the Euboians Lokrians Argives Phokians Sikyonians Ainianians Eleans Alyzaians Messenians Dolopians inhabitantsof Akte We have epigraphictexts which pertain to two of these alliances:the Phokian and the Sikyonian.24Assuming that Diodorus' list is more or less chronologicallycorrect in its sequence, these inscriptionsgive the general range for the period of the Athenian diplomaticpush. The Phokian text dates to Pyanepsion 19 (October 27), 323, the Sikyonian to Posideon 16 (December 23), 323.25 The quest for allies was thus well under way by late October and continued until late December, with the greatest successes enjoyed in the nearly two months between these two alliances.26This fits well with other evidence. For example, we hear that the Athenian ambassadors,including Hypereides,27were joined by the still exiled Demosthenes, who was of considerablehelp to them in the Peloponnesos. As a result of his efforts, Demosthenes was recalled,28but not in time to deliver the funeral oration for Leosthenes who had died during the winter of 323/2 B.C.29 Thus Athens' military and diplomatic efforts, begun secretly during the summer of 323, were bearing fruit during the autumn and early winter of that year. It is against this backgroundthat I would suggest that another decree of the Athenians should be set. This is the long known, but largely ignored, IG JJ2, 365.30 The 23DiodorusSiculus, xvIII.11.1-2;cf. the shorter list of Pausanias,i.25.4 which is arrangedgeographically. So, too, within subdivisions, is the list of Diodorus, but that appearsto reflect the concentrationof Athenian diplomaticeffort in successive regions. 24 IG 112, 367 (SEG XXI, 295) and 448 (SEG XXI, 297), respectively. 25Forthe dates, see Meritt, op. cit. (footnote 9 above), pp. 107-108, 133. 26Thischronology might imply that the first alliancewith the Aitolians had not been concluded until, perhaps,a month before the Phokian alliance at the earliest. The Aitolian alliance might thus be dated in mid- to late September (above, footnotes 17 and 22). This would accord well with the time necessary to have received confirmationof the death of Alexanderwhich can be estimated at about 90 days; cf. Badian, p. 42. 27Justin, xiii.5.10.

28Plutarch,Demosthenes,27.1-6. 29Diodorus Siculus, xviii.13.6.

30E.M.7178 + 7179; SEG XXI, 294. Thanks to the kindness and efficiencyof Mrs. Peppa-Delmousou and her staff, I have been able to examine the stone but understandthat it is to receive a full re-edition by

104

STEPHENG. MILLER

inscriptionis on a stele of Hymettian marble the right side and much of the central surface of which are badly damaged. The decree's heading is fairly well preserved, and the dating formula, although somewhat restored, is clear enough. It reveals that this law was passed on the 11th of Hekatombaionin the archonshipof Kephisodotos (July 23, 323/2). The law thus falls into that period about seven weeks after the death of Alexander when his death had been reportedbut not yet confirmed, when Leosthenes and the Athenians were beginning, but not yet openly, their preparationsfor revolt. It is thus chronologicallypossible that IG I12,365 had something to do with these early, as yet tentative preparations. Due to the damage of the stone's surface, the precise details of this law are not easy to determine. Following the name of the orator, in lines 7-9, we have: [7r]Ept [COI v o0 6' EIg Ta Xe[yEdI6 a[pKEOE9CpI

N[tI ea

K[aIL

A[alIrv

[pusd6

7r[p6EvoerI|

TrS '7rOXEcs.

Although this passage is fairly heavily restored, the remaining preserved parts of the text and the stoichedon requirementsmake the restorationssecure. Below this passage the text becomes very fragmentary,but it is clear that the Athenian archetheorosand the proxenos Lapyrisare, because of what they reportedto the Demos, to receive some recognition and payments.31Next the number of proxenoi increases,32and there is a shift from provisions concerning the past and present to provisions for the future. This begins to become clear on the second fragmentof the stone,33and is completely clear at the end of the main text.34After a slight uninscribedgap on the stone, there is a standard invitation extended to Lapyris, son of Kallias, from Kleonai, to deipnonin the prytaneion,and a provision that this law is to be written "on the stele on the acropolis on which was written the proxenia for Echembrotos of Kleonai, the ancestor of Lapyris." In the event, and probablyfor reasons of space, our law was not written on the stele of Echembrotos. This, too, survives and is of Pentelic marble with a thickness much greater than that of the Lapyrisstele.35Dating from the first quarterof the 4th century B.C., it appearsto document that Echembrotos' father had also been an Athenian proxenos at Kleonai.36Thus Lapyrisstands in a long line, extending over at least a century, of strong family ties with Athens. What was it that Lapyris and the archetheoros reported to the Athenians? We cannot know, but clearly it pleased the Athenians who not only rewardedthe reporters M. Walbank.Therefore, I offer here neither a text nor a photograph.I would, however, make two points about the stone. First, the two partsof the stele, presentedin IG as separatepieces fragmentstaduo marmoris Hymettii non contigua"), do join physicallyand have now been set together in the Epigraphical Museum. In accordwith IG I shall, however, refer to the two individualfragments.Secondly, the surface of the stone is so badly chipped that I could make only a very few minor readings in addition to those presentedin IG. These readingsmake no substantivechanges in the publishedtext. 31Line10: apKE6[Ewp--]; line 14: ap [ybvAptov]; line 15: XAa~.atPEP VITEp[--. 32Line

16:J[T1o0vt Wrpo9vov[c--.

6 irpo6vok-. 33Frag.b, line 5: Kaea Aa&WrptL 34Frag. b, lines 6-8:

TOvn 8E a1TO8EKUas

/uspl[ua

apyvptov.

35IG112,63; E.M. 6923. 6 d[aT'|p avrO--. 36Lines10-11: Ka6laITE[pl

I

Tci apKEI OE pco[LI os av aEt apKE6E[cop7

T7)

roll

KLEONAI,THENEMEANGAMES,AND THELAMIANWAR

105

but also made provisions for the future, almost certainlytied specificallyto the Nemean Games.37Remembering that at the time of this decree and of the Nemean Games in 32338 the rumors of Alexander's death must have been current, and remembering the actions taken by the Athenians as a result of those rumors, it is tempting to see a connection between the actions of Lapyrisand the archetheoros, and the earliest stages of the revolt against Macedon. The great Panhellenic festival at Nemea would have provided a very appropriateand convenient opportunityfor the Athenians to sound out potential allies. If this interpretationis correct, and if Lapyriswas helping the Athenians in these early and covert stages of the revolt, then we might expect that, once the Athenian actions became open, Kleonai, Lapyris' home state, would have joined with the Athenians. Yet the lists of allies in the Hellenic War do not include Kleonai (above, footnote 23). Perhapswe are to understandthe presence of Kleonai in the Argive entry into the alliance; certainly the smaller polis was frequently to be seen acting in concert with the larger,39but we cannot assume the presence of Kleonai in every Argive action. Nonetheless, some role by Kleonai in the Lamian War is implicit in the fact that, at the conclusion of the war, Antipaterwent to KleonaiA0This was after his victory at Krannon 37Themention in fragmentb, lines 3 and 4, of theoroi, as well as of future archetheoroiin lines 6-8 (above, footnote 34), would seem appropriateonly to a Panhellenicfestival, and since the Nemean Games have alreadybeen mentioned, and served in some way as the stimulus for this law, the future provisions seem surely to relate to those games. 38Thetime of the Games cannot be establishedwith great precision, but the careful investigationsof A. Boethius (Der argivischeKalendar,Uppsala1922, p. 68) indicatea date in the first half of July. 39Co-operationbetween Argos and Kleonai begins at least as early as ca. 460 (Diodorus Siculus, xi.65; cf. Strabo, viii.377 and Pausanias, ii.25.6), continues in 457 at Tanagra (Pausanias,i.29.7-9; cf. IG 12, 931/932), in 418 vs. the Spartans(Thucydidesv.67.2), and well into the 4th century against the Corinthians (between 387 and 364; Plutarch, Timoleon,4). For a possible brief period in the first half of the 5th century when co-operationmay have been lackingsee W. G. Forrest, "Themistoklesand Argos," CQ 10, 1960, pp. 230-231.

40Theevidence for Antipaterat Kleonai comes from two passagesin Plutarch(Demosthenes,28.4 and Phokion,29.1) both of which deal with the death of Hypereideswho, together with others who had sought refuge on Aigina, was sent to Antipaterat Kleonai, there to be executed and have his tongue cut out. [Plutarch],Decemoratorumvitae, 849 B-C, relates three variantson the place of the death of Hypereides: at Corinth, in Macedonia,and at Kleonai. Of these, Kleonai is by far the most likely place for Hypereides' death simply because it is the least obvious. One can imagine why it might have been thought in later times that Hypereideswas sent to Macedonia,or why someone might have thought of Kleonai as a mistake for Corinth, but why would Kleonai have ever entered the traditionunless it was actually the place where Hypereidesdied? F. Bdlte (RE, s.v. Kleonai 3) attempts to identify the site of Hypereides' death with a Kleonai in Phokis. Plutarch,de mulierumvirtutibus,244 D, relates a battle between the Phokiansand the Thessalians, and places this battle at a Kleonai which he regardsas so insignificantand unknown that he is obliged to define its location more preciselyas being in the territoryof Hyampolis (7cEpt KXECVana Tr 'Yacqtii-8o6o). When Plutarchmentions the rites which celebrate the battle and the events leading to it, he says simply that they are celebratedE5t vVv rTEpL'Ya'tiioXtv (244 B), and he elsewhere refers to the Phokianvictory as being'v e'YacqTi6Xt8 (244 E). He clearlyrealizes that his readerswill not know anythingabout the Phokian Kleonai and thus uses the more familiarlandmarkof Hyampolis.Aside from the improbabilitythat Antipaterwould have been in such an insignificantplace when Hypereideswas delivered to him, it is clear that Plutarchfelt that his readerswould know to which Kleonai he was referringin his mention of the events of 322 B.C. He felt no need to define further its location. B6lte's arguments are not convincing, and he is

106

STEPHENG. MILLER

on Metageitnion7 (August 6), 322,41and after the garrisoningof Mounychiaon Boedromion 20 (September 18).42 Antipaterwas cleaning up loose ends, including revolutionaries the last of whom was, appropriately,Demosthenes on Pyanepsion 16 (October13), 322.43A week earlier, on Pyanepsion.9,44Hypereideshad been handed over to Antipater at Kleonai (above, footnote 40). It must have been at roughly this time, in early October, that Antipater was arrangingthe affairs of the Peloponnesos which included the appointmentof Deinarchosof Corinthas ELLAEX'qT s HIEX0oro'v7zrr-ov.45 It is, then, possible that Antipater was using Kleonai as his headquartersfor his post-warclean-up operationsin the Peloponnesos, and his behavior in the month before his appearanceat Kleonai is instructive. Although he was then largely concerned with organizingthe situation in Athens, Antipaterstayed at Thebes, making his camp among the ruins of the city razed by Alexander.46That choice of headquarterswould appearto have been motivated by the fact that Boiotia had remained loyal during the war and therefore provided a safer, more secure base.47Just so, in the previous year, Antipater had used Lamia as a base because of her loyalty.48Thus a clear and reasonable pattern emerges, and we can see in Antipater's use of Kleonai the implication,which we have also discerned in other sources, that Kleonai had not joined the revolt. That fact, plus Kleonai's location on the main road into the Argolid and the southern Peloponnesos, controllingthe Tretos pass, made Kleonai a most suitable base for Antipater. Kleonai's loyalty to the Macedoniansmeans that, although a year earlier the Athenians must have hoped that Kleonai would join with them, the faithful proxenos Lapyris did not see his polis join the alliance. This also means that Kleonai had abandoned its long-establishedpolicy of following the Argive lead in internationalaffairs (above, footnote 39). Why did Kleonai strike out on this independent course which must have been at least a minor disappointmentto the Athenians? Motivations are never easy to determine but one possible explanationcan be offered. Control of the Nemean Games, originallyin the hands of Kleonai, had passed to Argos at some time before 388 B.C.49This shift in the control of the Games appearsto have occurred with at least the tacit consent of Kleonai for, as we have seen (above, simply wrong when he says: "Dass es [the site of Antipater'sstop after the LamianWar] nichtum K. im Sicherheit."I can find nothing to Peloponneshandelnkann, ergibtsich aus Plut. Phok. 26 mit volikommener suggest that the Kleonai in the cited text was anything other than the one in the Peloponnesos. The fact that Antipaterstopped for a time in Thebes before going on to Kleonai is hardlysurprising,even though Bblte seems to regardit as such. See the very appropriateremarksof K. J. Beloch, GriechischeGeschichte IV, 2nd ed., i, Berlinand Leipzig 1925, p. 78, note 1. 41Plutarch, Camillus,19.5. 42Plutarch,Phokion,28.1. 43Plutarch,Demosthenes,30.4. 44[Plutarch],Decemoratorumvitae, 849 C. 45Suidas,si'. AEtvapxo;; cf. Plutarch, Timoleon,21.2 and 24.4; Phokion,33.3. 46Plutarch,Phokion,26.3 and 27.1. 47Diodorus Siculus, xviii.11.3-5. 48DiodorusSiculus, xvnI.11.1 and 12.4. 49Xenophon,HellenikaIV.7.2.

KLEONAI,THENEMEANGAMES,AND THELAMIANWAR

107

footnote 39), co-operationbetween Kleonaiand Argos continuedafter that time. As recent excavationsat Nemea have shown, the Games themselveswere absent from Nemea, having been moved almost certainlyto Argos, from the time of a violent destructionin the late 5th centuryB.C.50 until the 330's when a new programof constructionshowsthat the Gameshad returnedto Nemea. This shift of the Games back to Nemea, probablyeffectedby MacedonianpolicyafterChaironeia,was not accompanied by a shift of controlbackto Kleonai.Rather,Argosretainedcontrolof the Games even thoughthey werenow at Nemea once again.5" What were the reactions of Kleonai and Argos to this Macedonianrelocation of the Games to Nemea in the 330's B.C.? We can deducea negativeArgivereactionfromthe

fact that in the firsthalf of the 3rd century,with Macedonianpowerin the Peloponnesos dissipated,the Argivesmoved the Games backto Argos.52On the other hand, we can suspect that Kleonai was very happy to have the Games at Nemea, and geography

revealsthe reason.With the Games a scant three miles to the west on the main road the latterwas clearlya main sourceof suppliesfor the tens whichpassedby Kleonai,53 of thousandsof visitorsto Nemea. Controlof the Gamesmay have carriedprestigebut proximityto the Gameswas profitable. Kleonai'sloyaltyto the Macedoniansduringthe LamianWaris thus explicablein termsof Macedonianpolicytowardthe NemeanGames.54It is very doubtfulthat Kleonai's entryinto the Athenianalliancewouldhave affectedthe outcomeof the war, but the Athenian diplomatic failure with Kleonai, despite early and tentative successes at the Nemean Games of 323 with the help of Lapyris of Kleonai, foreshadowed the

failureof the LamianWar, the end of any Athenianclaimto internationalleadership, and the beginningof economicproblemswhich have been given such definitionby excavationsin the AthenianAgora. Addendum A recent article by M. Pierartand J.-P. Thalmann, "Nouvellesinscriptionsargiennes

in the text I,"BCH, Suppl.IV, 1980,came to my attentiontoo late to be incorporated above.On pp. 261-269 of that articleis publisheda new inscriptionwhichshows con50S.G. Miller, "Excavationsat Nemea, 1979," Hesperia 49, 1980, p. 186. 51S.G. Miller, "Excavationsat Nemea, 1978," Hesperia48, 1979, pp. 79-80. 52W. Vollgraff,Mnemosyne44, 1916, pp. 65-69 and 221-232. 53Geographicalproximity is combined with lines of communication which were particularlystrong between Kleonai and Nemea. For example, Pausanias(ii.15.2) goes directly from the former to the latter on his way into the Argolid. Phlious, on the other hand, which lies roughly the same distance to the west of Nemea as does Kleonai to the east, was visited by Pausaniasfrom Sikyon ratherthan from Nemea. 54Wecan see the continuationof this policy during the two decades following the LamianWar in the presidencyof the Nemean Games by Cassanderin 315 B.C. (Diodorus Siculus, xix.64.1), the use of Nemea as the place for displayinga levy of troops from a league of Antigonos Monophthalmosin, probably,311 B.C. (D. J. Geagan, "Inscriptionsfrom Nemea," Hesperia37, 1968, pp. 381-384), and the use of Nemea, together with the other Panhelleniccenters, as a meeting place for the league of Demetrios Poliorketes (IG IV2, 1, line 68).

108

STEPHENG. MILLER

elusively that, in the period after the Lamian War (peut-etre en 318 ou peu apres, p. 268), Kleonai was incorporatedpoliticallyinto Argos. One is entitled to wonder if this development might not, in some sense, have been a result of Kleonai's role in the LamianWar. STEPHENG. MILLER UNIVERSITYOF CALIFORNIA,BERKELEY

Department of Classics Dwinelle Hall Berkeley, CA 94720

THE MISSIONOF TRIPTOLEMOS (PLATES15 AND 16)

STUDY of the Mission of Triptolemosseems to us an appropriateofferingto honor HomerA. Thompson;as the Attic Hero spreadthe knowledgeof agriculture to civilizethe world,so has our friend,the scholar,generouslygiven the fruitof his workon Attic monumentsto enlightenandinformour world. The connectionbetween artisticrepresentationsof Triptolemosand the political historyof Athens, hinted at by C. Dugas,1has been more fully developedby J. W. Day,2who reachedthe conclusionthat "duringthe fifth centuryTriptolemoswas employedas a symbolof Athens' civilizingmissionat the head of her imperialamphictyony." The missionof Triptolemoswas a favoritetopicin Athenseven beforethe time of Kimonand Perikles.Dugas,who was concernedalmostexclusivelywith representations on Attic vases, listed 15 black-figured examples;Grossmanand Bindokataddedfour andthree (twoof them fragments)respectively.3 A

La mission du Triptolemed'apr's l'imagerieAthenienne," MelRome62, 1950, pp. 7-31, esp. p. 14. TheGloryof Athens:ThePopularTradition as Reflectedin the Panathenaicusof AeliusAristides,Chicago 1980, pp. 15-39, esp. p. 24. J. P. Barronis going to include a chapteron the Eleusiniancult in his forthcoming book on "Athenian propaganda,"and M. Sakuraiis preparingan English version of her article on "The EleusinianCults and the Development of Athenian Democracy." 3B. Grossman, EleusinianGods and Heroesin GreekArt, diss. WashingtonUniversity 1959, pp. 67-77 and 81-92; A. P. Bindokat, "Demeter und Persephone in der attischenKunst des 6. bis 4. Jahrhunderts," JdI 87, 1972, pp. 78-157. A list of the black-figuredvases is given here in the order in which they will be mentioned in the first part of this essay: D refers to Dugas (see footnote 1 above), pp. 23-24, hos. 1-3 and p. 31, nos. 103, 104; G refers to Grossman, pp. 67-77 and 81-82; B refers to Bindokat. 1. Brussels A 130; CVA, Brussels 1 [Belgium 1], 7 [201:1;ABV, p. 308, no. 82 (Swing Painter); D 2; G AP 18. 2. GbttingenJ 14; ABV, p. 309, no. 83 (Swing Painter); D 1; G AP 20, 39. 3. Compiegne 975; CVA, Compiegne [France31, 10 [1081:4,7; ABV, p. 331, no. 13 (Priam Painter); D 12; G AP 17. 4. Beverly Hills (Dr. MarionPrinzmetal);ABV, p. 478, no. 2, Paralipomena,p. 217 (EdinburghPainter); D 13. 5. Vatican385; ABV, p. 374, no. 195 (LeagrosGroup); D 7; G AP 21. 6. LenormantCollection (formerly);D 5; G AP 16. 7. WUrzburg197; D 4; G AP 19. 8. Bologna, Mus. Civico 5; CVA, Bologna2 [Italy7], 14 [3131:4;D 11. 9. San Simeon 5503; D 9. 10. Hamburg;D 10. 11. Providence, Rhode IslandSchool of Design 25.083; CVA, Providence 1 [USA 2], 10 [63]:1a;D 3. 12. Budapest50.732; B V 53. 13. Munich 1539 (J 543); CVA, Munich 8 [Germany37], 398 [18161:4,402 [18201:2,412 [18301:2;D 8; G AP 37. 14. Acropolis675; ABV, p. 377, no. 236, Paralipomena,p. 171, no. 9 (near Chiusi Painter);D 6; G AP 23. 15. Athens, NationalMuseum 430 (CC 967); ABV, p. 587, no. 1 (BeldamPainter); D 104; G AP 40. 2

110

ISABELLE K. AND ANTONYE. RAUBITSCHEK

A groupingof these vases accordingto chronologyand iconographypermitsa of theirrepresentations. betterunderstanding The earliest(nos. 1 and 2) show Triptolemos sittingon a countrycart,4a beardedman holdingears of grainin his hand and givinginstructionsto mortalsaboutthe cultivationof grain.These vases are attributed to the Swing Painter and dated ca. 540-520 B.C.Both show the cartabove the ground, floatingby levitation.' On vases of the Leagros Period, the cart was rendered more majesticby transforming the stool into a throne-like seat and by adding occasionallywings to carry it over

landand sea. The wingsare probablyrelatedto the wingedcartof Dionysosmade for him by Hephaistos.6Dionysos sits on a winged cart on the black-figuredamphora in Compiegne (no. 3), while Triptolemos, on the other side of the same vase, has a wingless cart and a seat with arm rests. The only black-figuredrepresentationof Triptolemos

on a wingedcart is on an amphoraby the EdinburghPainter(no. 4; see H. R. W.. Smith, "FromFarthestWest," AJA 49, 1945, pp. 470-471, fig. 4:1); there the hero is not associatedwithDionysos.7 Triptolemosappearswith Demeter on a fragmentaryblack-figuredamphorain Reggio,8where,identifiedby name, he is a memberof a divineprocessionaccompanying Demeter;but here he is not on his mission.On the amphorasof the LeagrosPeriod (nos. 8-14), the attentivewomen on either side of Triptolemoscannot be securely identifiedas Demeterand Kore;for no. 10, see R. Pagenstecher,"Zu unteritalischen Terracotten,"AA [JdI23], 1917, p. 107, fig. 37. No. 13 has been assignedin the CVA to "nearthe ChiusiPainterandthe Antiopegroup." The Mission of Triptolemoson three black-figuredlekythoireflectsthe already establishedred-figuredsceneswithsnakesas an aid in drawingthe chariot(nos. 15-17). These vases belongto the 5th century;two of them show the hero alone (nos. 15 and 16) andthe third,the hero flankedby two unidentifiedwomen. The iconographyof these vases shows the policyof Peisistratosand his sons, who fosteredthe cultivationof grainand of the vine and who promotedthe cults of Dionysos and Demeter.9The City Eleusinionmay have been foundedalreadyin the age 16. Athens market;ABV, p. 518 (Haspels, ABL, no. 64: Theseus Painter);G AP 38. 17. Prague,NationalMuseum 1867; ABV, p. 708, no. 19 (EmporiumPainter);G AP 22. 4See H.-L. Lorimer, "The CountryCartof Ancient Greece," JHS 23, 1903, pp. 132-151. 5For illustrationsof no. 2, see A. B. Cook, Zeus I, Cambridge1914, p. 213, fig. 156; P. Jacobsthal, GattingerVasen,Berlin 1912, pl. 5. 6This is suggested by the red-figuredkylix in Berlin (2273) by the Ambrosios Painter (AR V2, p. 174, no. 31; see Cook, op. cit., p. 216, fig. 159), and the hypothesis is strengthenedby the appearanceof Hephaistos in such a cart on a kylix in Florence; see E. Simon, Gbtterder Griechen,Munich 1969, pp. 223224, fig. 209; M. Robertson, review of F. Brommer, Hephaistos,Mainz am Rhein, AJA 84, 1980, p. 104. 7Otheramphoraswith Dionysos on one side and Triptolemos on the other are nos. 5 (C. Albizzati, Vasiantichidipintidel Vaticano[n.p., n.d.], pl. 54, no. 385), 6 (Cook, op. cit. [footnote 5 above], p. 214, Munich 1932, p. 34, pl. 51, no. 197). figs. 157:aand b), 7 (E. Langlotz, GriechischeVasenin Wflrzburg, 8Reggio, Mus. Nazionale 4001; ABV, p. 147, no. 6 (manner of Exekias); H. Metzger, L'imagerieathenienne,Paris 1965, p. 8, pls. I and II. 9See A. Andrewes, TheGreekTyrants,London 1956, pp. 113-114.

THEMISSION OFTRIPTOLEMOS

111

of the tyrants.10 The associationof Demeter and Triptolemosis impliedby the scenes on the black-figured vases (nos. 1-17) which depictthe Mission.This missionis not mentionedin the Hymn to Demeterwhichis not only earlierbut also purelyEleusinian.11The elevationof Triptolemosas the hero who uses a "chariotof the gods" may belong to the time of the establishmentof the City Eleusinion,which should have takenplaceaccordingto the vases and the excavationsin the reignof Peisistratos.The poet Onomakritos,a friendof his son Hippias,is said to have composedpoems and oraclesaboutTriptolemoswhichhe attributedto Mousaiosto give them greaterauthority.12Thus the vase paintingsenrichour knowledgeand understanding of the establishment of the Eleusiniancult in the city of Athens by showingthat the storyof Triptolemos was addedat that time, possiblytakenfrom Argos, the home of Peisistratos'second wife, Timonassa.13 The transformation of Triptolemosfrom an instructorof the Athenianfarmersin the art of agricultureinto a hero, chargedby Demeter to spreadthe knowledgeof farmingthroughoutthe world,took place, accordingto the vases illustratingthis mission, between510 and 480 B.C., whenrenewedactivityis attestedboth in Eleusisand in the Eleusinionin Athens.14This propaganda effortwith its emphasison the Missionof Triptolemosbeyond the bordersof Attica may be connectedwith the claims of the newlyfoundedAtheniandemocracyto be the mothercity of all the Ionians.15 The skyphospaintedby Oltosand foundon the Akropolisis unfortunately so fragmentarythat the compositionof the three figuresis not certain:16 Demeter, holding ears of grain,sits on a throne;Triptolemosmust be in his chariotof which only the footrestis preserved;a secondfemalefigure,elaboratelydressedin embroideredchiton and mantle,is shownfrontally.Detailsnow missingfrom the skyphosby Oltosmay be seen on fragmentsof a contemporaryphialeby the EuergidesPainter,also from the Boardmanobserved that this paintershared "Epiktetos'admirationfor Akropolis.17 It seems that the Atheniansof the earlydemocracycontinuedthe interHipparchos."18 est of the tyrantsin the Eleusiniancult. On neithervase is the head of Triptolemos 10SeeH. A. Thompson, "Activities in the Athenian Agora: 1959," Hesperia29, 1960, pp. 334-338; H. A. Thompson and R. E. Wycherley, The AthenianAgora, XIV, TheAgora of Athens, Princeton 1972, pp. 152-153; R. E. Wycherley,Stonesof Athens,Princeton 1978, pp. 71-72. "1SeeN. Richardson, The HomericHymn to Demeter, Oxford 1974, pp. 194-196 and 301-302 with notes on lines 476-482. "See Herodotos, vii.6.3; Pausanias,i.14.2; 22.7; viii.37.5. "See Aristotle, AthenaionPoliteia,17; Pausanias,I.14.2. 14See IG I2, 5 (818 is anothercopy) = IG I3, 5; L. H. Jeffery, "The BoustrophedonSacralInscriptions from the Agora," Hesperia17, 1948, pp. 86-111; J. Travlos, PictorialDictionaryof Athens,New York 1971, pp. 198-199; Wycherley, op. cit. (footnote 10 above), pp. 71-72; K. Clinton, "IG 12 5, The Eleusinia, and the Eleusinians,"AJP 100, 1979, pp. 1-12, esp. note 13 on pp. 4-5. "See Herodotos, i.147; v.97; ix.97. '6E 13; Graef-Langlotz,Die antiken Vasenvon der Akropoliszu Athen, Berlin 1931, II, ii, p. 41, pl. 39, 449 a-d; ARV2, p. 66, no. 135. '7Graef-Langlotz,op. cit., p. 11, no. 147, pl. 6; ARV2, p. 89, no. 19. 'Athenian Red FigureVases,the ArchaicPeriod,London 1975, p. 60.

112

K. AND ANTONYE. RAUBITSCHEK ISABELLE

preserved;we cannot tell, therefore, whether he was alreadybeardless and long haired as he appearsin the canonicalscenes of the 5th century. After the Persian Wars representations of the Mission of Triptolemos on vases became frequent, perhaps as an acknowledgment of the aid given by the Eleusinian deities to the Athenians both at Marathon(Herodotos, viii.65) and at Salamis (Plutarch, Themistocles,15), perhapsin gratitudefor Demeter's gifts to the Athenians and, through the Athenians, to the world.19It was at that time, in 468 B.C., that Sophokles' Triptolemos was performed and favorablyreceived. Although only fragments are preserved (A. C. Pearson, FragmentsII, Cambridge1917, pp. 239-242), the play must have dealt with the Mission of Triptolemos since some of the fragments contain references to details seen on the vases: the dragonswhich pull the chariot (fragment596) and Demeter's charge to Triptolemos (fragment 597). The transmission of this message to certain unspecified nations, to which Dionysios of Halikarnassosrefers (Antiquitatumromanarum1.12.2), may have been part of Sophokles' play; an inscription of ca. 460 B.C. concerning the properconduct of the Mysteria refers in a fragmentarilypreserved passage20to "these cities" meaning the cities who share the cult of Demeter and benefit from her gifts. Later in the 5th century, the allies of the Athenians and the Greeks in general were requested and invited to contribute wheat and barley, the "fruits of the earth", to the sanctuaryin Eleusis, certainlyin return for the benefactionsreceived from Demeter and from the Athenians through the Mission of Triptolemos (IG 12, 76 = IG J3, 78). The strongest indicationof the importanceof the Mission of Triptolemosand of the benefactionsof Athens is given by the vases of the period between the Persianand Peloponnesian Wars. Twelve of the 245 vases attributed by Beazley to the Berlin Painter (AR V2, pp. 196-214) illustratethe Mission of Triptolemos.The earliest of these21shows the hero standingnext to his chariotwhich is now winged for the longer trip outside Attica. He holds the Eleusinian scepter and Demeter hands him ears of grain; Persephone and Hades appearon the reverse. Two fragmentaryvases from the same period probably echo the same scene, as suggested by the reconstructionoffered by G. Schwarz.22 The missionaryaspect of Triptolemos' travels was illustratedeven more clearly by showing him pouring a libation before his departure, evidently an adaptation of the "warrior'sdeparture", an invention of the Kleophrades Painter.23The scene first appeared on the Berlin Painter's hydria in Copenhagen (2696: ARV2, p. 210, no. 181) of his middle period. Triptolemos is seated in his winged chariot holding out a phiale 191nthe "EphebicOath" (M. N. Tod, A Selectionof GreekHistoricalInscriptionsII, Oxford 1948, no. 204; P. Siewert, "The EphebicOath in Fifth-CenturyAthens," JHS 97, 1977, pp. 102-111), which surely goes back to the PersianWars, the young men swore by the fruits of the earth, and in the Panegyricusof Isokrates(which repeatsdetails of the funeralorationswhich also go back to the PersianWars) the gifts of Demeter to the Atheniansand to the world are emphasized (28-29). 20IG12, 6 and 9 (SEG X, 6 and XXVI, 2), lines 36-43 = IG I3, lines 36-43; copies of this text were set up in Eleusis and in the Eleusinionin Athens. 68.101, Paralipomena,p. 344, no. 131bis. 21Karlsruhe 22"Zweieleusinische Szenen auf einem Kelchkraterdes BerlinerMalersin Athen," AA [JdI 861, 1971, pp. 178-182: ARV2, p. 211, no. 201 and p. 205, nos. 119 and 120. 23SeeE. Simon, OpferndeGitter, Berlin 1953, p. 71.

THEMISSIONOF TRIPTOLEMOS

113

towardsKore, who is about to pour a libation from an oinochoe in her right hand while she holds a flaming torch in her left. The same scene is repeated on a hydria in Graz from the painter's late period.24To this period belong also two stamnoi in the Louvre (G 371 = ARV2, p. 208, no. 158 and C 10798 = ARV2, p. 208, no. 159) with Demeter, Triptolemos and Kore in a multifigured scene, perhaps not independent of MakrQn'searlierskyphos.25 The invention of the multifiguredscene, includingTriptolemoswith Demeter and Kore, may be credited to Makron, who showed this group among six other figures on his famous skyphos in the British Museum.26Here all the charactersare labeled: Triptolemos, a beardless youth with long hair, myrtle crowned, with chiton and mantle, seated in a chariot equipped with wings and snakes to carry him over land and sea; behind him regal "Demetre" with embroideredchiton and mantle, a polos on her head, a flaming torch in her right hand, three ears of grain in her left. In front of Triptolemos is "Pherephatta"about to pour a libation from the oinochoe in her right hand into the phiale held out to her by Triptolemos, while in her left she holds a flaming torch. Others present include Triptolemos' parents "Eumolppos" and "Eleusis" as well as the gods Zeus, Poseidon and Amphitrite.Of the 354 vases attributedto Makron, this skyphos of ca. 480 B.C. is the only one which shows Triptolemosand which depicts him in a friezelike composition. One must, however, resist the temptation to claim that the vase painter was inspired by and copied a contemporarymural. This caution is also suggested by a vase painted by an unidentifiedmember of Brygos' workshopwhich was found in the same tomb at Capua as the Makron skyphos,27and which contains an eight-figure scene consisting of Triptolemos in a snakeless winged chariot, stretching out his phiale to a drapedwoman who holds a flower in her right hand. She stands in a building, indicatedby two Doric columns; in this building yet another woman is standing and Hades is sitting on a throne. Behind Triptolemosis another woman and behind her a warriorwith a phiale, a woman with two torches and a winged female holding both oinochoe and phiale. This ambitious composition, rather ineptly drawn, owes nothing either to Makron'svase paintingor to a hypotheticalmural in the Eleusinion.28 During the following period, when Athens had become rich and powerful, the Mission of Triptolemosgained in popularityto judge by its appearanceon vase paintings 24UniversititG 30: see G. Schwarz,"Eine Hydriades BerlinerMalersin Graz," OJh 50, 1972/73, pp. 125-133. On some vases, the BerlinPaintershows Triptolemosalone, but the wings on the chariotindicate that he is on his mission: Vienna 3726 = ARV2, p. 205, no. 113; Dresden 289 = ARV2, p. 201, no. 69 (both vases are of the middle period); RobinsonCollection = AR V2, p. 203, no. 97 (of his late period). 25B.Philippakihas observed a close connection in the manufactureof stamnoi in the workshopsof the BerlinPainterand of Makron(The AtticStamnos,Oxford 1967, pp. 44 and 151). 26E140 = ARV2, p. 459, no. 3; see Simon, op. cit. (footnote 6 above), pp. 111-112, fig. 105. 27J. D. Beazley, "The BrygosTomb at Capua,"AJA 49, 1945, pp. 153-158. This cup is now in Frankfurt's StddelscheKunstinstitutV/7 = AR V2, p. 386; see also M. Wegner, Der Brygosmaler,Berlin 1973, pp. 53-56. 28Afragmentarypelike in the Getty Museum in Malibu is probablyto be dated to the same ripe Archaic period.It shows Triptolemosand anotherfigure on one side, with a Dionysiac scene on the other. M. Robertson,who will publishthis vase, attributesit to the Pan Painter.

114

ISABELLE K. AND ANTONYE. RAUBITSCHEK

by distinguished artists working in the best workshops, for instance by the Altamura Painter and by his "younger brother", the Niobid Painter. Among the six Triptolemos pictures of the Altamura Painter there is one in Munich29which recalls a vase by the Berlin Painter (see footnote 21 above). Here the hero is standing behind his chariot while an Eleusinian goddess is facing him, holding an oinochoe in her right hand. Two of the eight Triptolemosvases attributedto the Niobid Painter have similar two-figure scenes,30 while the canonical three-figure scene, with Triptolemos between Demeter and Kore, appearson a number of vases painted by the two artists.31On other vases, the two painters include the three figures in multifiguredcompositions.32The Niobid Painter alone increased the number of figures in such a scene to eight on two kalyxkraters.The one in the Louvre (G 343 = AR V2, p. 600, no. 17), possibly in imitation of the AltamuraPainter's name vase in the British Museum (see above, footnote 32), shows the Mission on one side of the neck, while on the krater from Spina (Ferrara T13 = AR V2, p. 602, no. 24) it occurs on the lower of the two zones on side B. As on the Altamura Painter's krater in Lyons (see above, footnote 31), Demeter and Persephone stand in front of Triptolemos' chariot which is equipped with both wings and snakes; the other five figures are depicted as participantsin Eleusinian rites by the bunches of grain or by the torches which they hold in their hands. The scene on the reverse shows a Dionysiac thiasos, suggesting once again that Athens was the donor of both grain and wine.33 292383= AR V2, p. 591, no. 23; K. Vierneiselgenerouslysent us a picture. 30A neck amphorain Leiden (PC 76 = AR V2, p. 605, no. 59) and a Nolan amphora in the British Museum (E 274 = ARV2, p. 604, no. 53) on which the chariot and the oinochoe are omitted (CVA, BritishMuseum 3 [GreatBritain41, 13 [1781:2a-c). 31Mostpleasingis the scene on a kalyx-kraterby the AltamuraPainter in Lyons (E 120 = ARV2, p. 591, no. 24, illustratedby C. Dugas, "Le Peintre d'Altamuraau Musee de Lyon," JHS 71, 1951, p. 58, pls. 25, 26); more crowded are the four figures on another kalyx-kraterin Leningrad (639 = ARV2, p. attischeskievazy, Leningrad1967, pp. 146591, no. 15, illustratedby A. A. Peredolskaya,Krasnofigurnye 147, no. 170, pl. 112:1). The Niobid Painterhas two vases with three figures:a stamnos in Lugano (Paralipomena,p. 395, no. 41ter, illustratedin Munzen und MedaillenA.G., Auktion34, Basel 1967, no. 165, pl. 54) with a shape which is favored by the AltamuraPainter;see Philippaki,op. cit. (footnote 25 above), pp. 73 and 153. This stamnos shows Triptolemoswearingonly a mantle which leaves his right shoulder bare. The other vase is a kalpis in the MetropolitanMuseum of Art (41.162.98 = ARV2, p. 606, no. 80, illustrated in CVA, Fogg Museum, Gallatin Collection 1 [USA 81, 56 [4041:1); its representationreverts to earlier details in showing the dress of Triptolemosas a chiton with mantle and in having snakes added to the chariot, as on the British Museum skyphos by Makron (see footnote 26 above), although the vase is assigned to the late work of the Niobid Painter by T. B. L. Webster, Niobidenmaler,Leipzig 1935, p. 22, no. 50. 32TheAltamuraPainterused a crowdedsix-figurecompositiononly once, in a miniaturefrieze on one side of the neck of his name vase, the volute-kraterin London, BritishMuseum E 469 = ARV2, p. 589, no. 1. The Niobid Painteremployed five figures on the body of a bell-kraterin Perugia (AR V2, p. 603, no. 34), and possiblyon a kalyx-kraterfrom Lokroi in Reggio (ARV2, p. 603, no. 40, illustratedin NSc, 1917, p. 154, fig. 60) where two figuresto the right of Triptolemosmay be restoredto balancethe two on his left behind his chariot. 33D. Feytmans claims ("Une representationinusitee du depart de Triptoleme," AntCI 14, 1945, pp. 285-295) that on a pelike in Brussels (R 235 = ARV2, p. 1121, no. 11) the youth standing in a chariot drawnby horses is Triptolemos, although he has no attributesidentifyinghim. Since the women standing in front of the chariot hold, on one side, two torches, and, on the other, an oinochoe and a phiale, the

OFTRIPTOLEMOS THEMISSION

115

The continuedpopularityof the Missionis shownby the greatnumberof vases on which it was represented between 475 and 425 B.c.34 Five of these were decorated by

Polygnotos,a successorof the NiobidPainter,who was not so much influencedby wall paintingsas his masterand his own followershave been assumedto have been.35It is of the clear,however,thatPolygnotosimitatedthe NiobidPainterin his representations in turn, may have inspireda member These representations, Missionof Triptolemos.A6 of Polygnotos'group, the KleophonPainter,37to composethe pictureon the volutekraterin the StanfordUniversityMuseumof Art, 70.12.It seems, therefore,that there was an unbrokentraditionfrom the BerlinPainterto the KleophonPainter,a tradition probablyindependentof sculpture,relief,andwallor panelpainting. reliefshowingthe Missionof Triptolemoswhichwas foundin the The fragmentary on Eleusinionin the Agora(S 1013)seems to be laterthanmost of the representations vases, althoughit is proofthat the worshipof Triptolemoson his Missiontook place duringthe 5th centuryin the sanctuarywhichalso containeda "temple"of the hero.38 The greatrelief from Eleusisin the NationalMuseumof Athens is strictlyEleusinian (see above, footnote 11) and does not seem to have influencedthe representationof the Missionof Triptolemoson vases whichwerepaintedin Athens.39 The Stanfordkratershows on Side A (P1. 15) a uniquescene of the Missionof Triptolemosconsistingof five figures.On the left is Demeter (P1.15:b) sittingon the "agelastospetra"which was locatednext to the well Kallichoronin the sanctuaryin Eleusis;40she is identifiedby an inscriptionwhich is now difficultto read. In her left scene is surely Eleusinian, possibly echoing representationsof the Mission. Under one handle Herakles is seated, under the other Dionysos is enthroned. 34Dugas, op. cit. (footnote 1 above), counted 36 in the second quarterof the 5th century B.C. (pp. 26-28) and 16 in the third quarter (pp. 28-29). To these should be added the oinochoe in the National Museum, Athens 1545 with a two-figurelibationscene; see E. Buschorin FR III, p. 262. 35SeeP. E. Arias and B. B. Shefton, A Historyof 1000 Yearsof GreekVasePainting,New York 1962, p. 366; C. M. Robertson, "Attic Red-FigureVase-Painters,"JHS 85, 1965, p. 97. 36Noneof the chariotson the five vases has snakes, and they all float, except for no. 1. 1. Stamnos in Florence, 75748 = ARV2, p. 1028, no. 8; this may be the latest. 2. Kalyx-kraterin Duke University, Art Museum = Paralipomena,p. 442, no. 27bis. The four figures are identified by inscriptions;the one behind Kore is "Hekate". The scene on the back, three mantled youths, has no connection with the Mission. 3. Neck amphora in London, B.M. E 281 = AR V2, p. 1030, no. 36; here, the hero is clad only in a mantle as on the Niobid Painter'sstamnos in Lugano (see above, footnote 31). 4. Neck amphorain Cannes, privatecollection = AR V2, p. 1031, no. 37 = Paralipomena,p. 442. 5. Stamnos in Capua,7529 = AR V2, p. 1028, no. 7. Triptolemosis clad only in a mantle (see no. 3). The three drapedwomen on side B continue the Eleusinian scene since one carries an Eleusinian torch, another a scepter. 37See Arias and Shefton, op. cit. (footnote 35 above), p. 368, on the relationshipbetween the two painters. The argument for the identification of the painter of the Stanford vase will be presented elsewhere. 38See T. L. Shear, Hesperia8, 1939, pp. 207-211, fig. 9; Travlos, op. cit. (footnote 14 above), pp. 198-201, fig. 262, to which there is no reference in the text. 39SeeR. Lullies, GreekSculpture,London 1960, p. 80, figs. 172, 173; G. E. Mylonas, Eleusisand the EleusinianMysteries,Princeton 1961, pp. 192-193. 40SeeApollodoros, i.5. 1; G. E. Mylonas, TheHymn to Demeterand her Sanctuaryat Eleusis, St. Louis 1942, p. 69.

116

K. ANDANTONY E. RAUBITSCHEK ISABELLE

handshe holdsa torchandin her missingrighthandshe probablyheld out a phialeinto whichthe maidenwho is standingin frontof her is pouringa libationout of an oinochoe. This maiden, labeled "Parthenos",holds a torch in her left hand. The actual missionscene (P1.15:a)showsTriptolemos,Persephoneand a secondwoman(possibly Hekate).Triptolemoswearsa crownof earsof grain,and he is holdingothersin his left hand;his righthandis stretchedout holdinga phialefor the libation.His wingedchariot equippedwith two snakesis alreadyafloat.The young womanstandingin front of him is labeled"Kore";she may be the same personas the "Parthenos"standingin frontof Demeter,or the figurecalled"Parthenos"may referto the well calledPartheKorepoursthe libationfromthe oinochoe nion whichis an old namefor Kallichoron.41 in her righthandwhile her left hand holds the Eleusiniantorchreversedtowardsthe ground.Of the label for the figurebehindher only the letter E is preserved;it may have been Hekate;she standsquietlyholdingan Eleusinianscepterin her righthand. Evidently,the vase paintingshows two Eleusinianscenes, the arrivalof Demeter in Eleusisand her receptionby the "Parthenos"Kallidike,the daughterof Keleos (Hymn, line 145), and the departureof Triptolemos,sent off by Demeter in gratitudefor the receptionshe hadreceivedin Eleusis. The other side of the krater(B. P1.16:a)showstwo pairsof two figureseach. Dionysos (the firstfour lettersof the nameare still visible) standsquietlyin rightprofile; he holds a thyrsosin his righthandand a kantharosin his left. The youngwomanto the rightof the god standsfrontallyand holds an oinochoein her right hand and an Eleusiniantorchin her left; she is aboutto pourwine into the kantharosheld by Dionysos. Behindthe god in a separatescene standsa frontalPapasilenos,white haired, partiallybaldand bearded;he leanson a staffheld in his righthandand carriesa wineskin, decoratedwithan ivy garland,over his left shoulder;he is labeledPom(po)s.The youngwomanto the left of Papasilenosis namedbut her namecannotbe restoredwith she standsin right profileand playsthe lyre held in her left hand;her confidence;42 bent rightwristindicatesthe use of the plectrum.The use of the lyre, alongwith the dignifiedposes of all four figures, indicatesthat the scene is no orgy but a solemn religiousprocessionwithEleusinianovertones. of the benefactionsof Demeter The two sidesof the kraterpresentgoodillustrations andDionysoswhicharepraisedbyTeiresiasin the Bacchaiof Euripides(lines271-286). The scenes on the front and backare ingeniouslyconnectedby the figuresunder the handles.Runningawayfromthe Triptolemosscene on Side A and towardsthe lyre playeron Side B is a vigoroussatyr(P. -16:b),a thyrsosover his rightarmand a spotted animalskin over his left shoulder.More unexpectedis the Pan (so labeled)under the other handle (P1.15:b),who, also wearingan animalskin, looks backat Demeter on the "agelastospetra"while runningtowardsthe Dionysiacscene. Although no connectionis attestedbetweenPan and either Dionysosor Demeterfor Greek art of 41See Hymn to Demeter,line 99, and the comments by Richardson, op. cit. (footnote 11 above), pp. 326-327. 42SeeMUnzenund Medaillen,A.G. Auktion40, Basel 1969, no. 108, pp. 65-67, pls. 45, 46: (- - -)AIT(-

- -)A~VE.

THE MISSIONOF TRIPTOLEMOS

117

the 5th century B.C.,43 the appearanceof this god of open spaces on the same vase with

the deities of grainand of wine may be intendedto emphasizethe essentiallyagriculnote is againechoedwhen turalmilieuof the settingas a whole.The political-historical we rememberthat Pan saved the Atheniansin the PersianWarsby frighteningtheir enemies (Herodotos,vi.105,vii.65; Pausanias,i.28.4). The Stanfordvolute-kratercontainsthe most detailedrepresentationof the Athenian benefactionsto the world, a claim which was the very basis of a contemporary decreeregulatingthe offeringof the FirstFruits (Aparchai)in the sanctuaryof Eleusis and orderingsacrificesto the goddessesand to Triptolemos(lines 36-38).4 This claim of havingbenefitedall mankindis also stressedby Plato in the Menexenos(238 A-B) the dramaticdate of which belongs to the same period,before the death of Perikles. Accordingto Isokrates(Panegyricus, 28-29) the Athenianbenefactionswere a standard topic of the publicfuneralorations,the Epitaphioi,whichcontaineda patrioticversion of Athenianhistory.It was to this traditionthat the conservativeDadouchosKallias, the son of Hipponikos,referredwhen he deliveredhis speech in Sparta(Xenophon, Hellenikavi.3.6):Spartawas to makepeacewith Athens becausethe Spartanswere the firstbeneficiariesof the gifts bestowedby Triptolemos. Duringthe last quarterof the 5th centuryB.C.,when Athenianpowerandauthority over the allies beganto wane, the Missionof Triptolemoswas less often depictedon Attic vases.45In the 4th centuryB.C.,the vase paintingsof this scene correspondmore closely to the dedicatoryreliefs which show Triptolemosalong with many other Eleusiniandeities, includingDionysos;46the Missionitself has become mythicalhistory, and it had lost its politicalrelevanceafter the downfallof the AthenianEmpire.The religiousand culturalsignificanceof the Mission of Triptolemoscontinuedinto the Hellenisticand Romanperiodsjust as the culturalmissionof Athens was cherishedin latertimes.47 ISABELLE K. RAUBITSCHEK ANToNY E. RAUBITSCHEK STANFORD UNIVERSITY

of Art Department of Classics Department Stanford,CA 94305 430ur negative results in a search for such a connection were affirmedby H. Walter (per litt.) in June 1980; cf. idem, Pans Wiederkehr, Munich 1980. 44IG 12,76 = IG I3, 78; see R. Meiggs and D. Lewis, A Selectionof GreekHistoricalInscriptions, Oxford 1969, no. 73; a copy of this decree was to be set up in the Eleusinion in Athens, and a small fragment of this copy has been found. 45Dugas(op. cit. [footnote 1 above], pp. 29-30, nos. 88-91) cites only five vases of this period, and only one has been included in AR V2, p. 1315, no. 2a. 46Forexample, the kalyx-kraterby the Telos Painter, AR V2, p. 1427, no. 37, and the skyphos by the MarsyasPainter, AR V2, p. 1475, no. 8, both omit the libationscene. 47See A. Alfbldi, "FrugiferTriptolemos im ptolemdisch-r6mischenHerrscherkult,"Chiron9, 1979, pp. 553-606. G. Schwarzis preparinga monographon the representationsof Triptolemosin Hellenisticand Roman Art.

PLATE 15

i~~~~~~~~~~~~~~~~~~

_

10~

;

>

*

~~ 2~~~~~~

Lt~~~~~

_

^.

~~~~~~~~~~~~~~~~C4

PLATE 16

0 0.

m

pat~~~~~~~~~~~~~~~~~~~~~~~r

C',

OF KOUROIAND KORAI-ATTIC VARIETY (PLATE 17)

HAVELEARNEDa greatdealfromHomerA. Thompson,throughthe yearsof my acquaintance withhim, andtwo of his teachingsstandout withclarityin my mind:to payattentioneven to minutedetails,and to rise abovethem to view the greaterwhole. The ensuingspeculationand theoriesmay not alwaysbe correct,but as long as they are revisedwhenevernew andcontradictory evidencebecomesavailable,the attemptshould be made. I hope that these lines, stemmingfromjust such an approach,may not be amissas a modestoffering.The minutedetailsto be consideredhere are a set of swollen earsand a mantle.The largerpictureconcernsthe possibleinferenceson the meaning of kouroiandkoraiin ArchaicAthens.1 The so-calledRayetHead (P1.17:a)was foundin Athens, "nearthe gas-works"at the edge of the Kerameikoscemetery,in the early1870's,and is now one of the glories of the Archaiccollectionin the Ny CarlsbergGlyptothekin Copenhagen.Well known to scholars,the head is often illustratedin monographson Greeksculptureand is analyticallydescribedin Richter'sKouroi.2 There,however,no mentionis madeof the fact that the head has "cauliflower ears"as befit a boxer (P1.17:b, c).3 Otherdescriptions commenton the thick,fleshyrenderingof this anatomicaldetail,but treatit as a mannerismof its sculptor.4I was able to inspectthe RayetHeadin the springof 1979 and I

'Besides the standardabbreviations,the followingwill be used throughout: = G. M. A. Richter, TheArchaicGravestonesof Attica,London 1961 AGA der Akro= H. Schrader,E. Langlotz,W.-H. Schuchhardt,Die archaischenMarmorbildwerke AMA polis, Frankfurtam Main 1939 ArchaicStyle = B. S. Ridgway, TheArchaicStylein GreekSculpture,Princeton1977 = L. H. Jeffery, "The InscribedGravestones of Archaic Attica," BSA 57, 1962, pp. 115Jeffery 153 = G. M. A. Richter, Korai,ArchaicGreekMaidens,London 1968 Korai = G. M. A. Richter, Kouroi,ArchaicGreekYouths,London 1970 Kouroi OlympicGames= N. Yalouris et al., TheOlympicGames,Athens 1976 from theAthenianAkropolis,Cambridge,Mass. 1949 Raubitschek = A. E. Raubitschek,Dedications = G. Schmidt, "KopfRayet und Torso vom pir~Aischen Tor," AthMitt84, 1969, pp. 65-75 Schmidt = L. A. Schneider, ZursozialenBedeutungderarchaischenKorenstatuen,Hamburg1975 Schneider = F. Willemsen, "Archaische Grabmalbasenaus der athener Stadtmauer," AthMitt78, Willemsen 1963, pp. 104-153 2Copenhagen,Ny CarlsbergGlyptothekI.N. 418; F. Poulsen, Catalogueof AncientSculpturein the Ny CarlsbergGlyptotek,Copenhagen 1951, no. 11, pp. 28-29, dated in the third quarterof the 6th century. Kouroi,no. 138, figs. 409, 410. 3The term is defined in Webster's Dictionaryas "an ear deformed from injuryand excessive growth of reparativetissue, so as to suggest a cauliflower."Boxers, wrestlersand pankratiastswould suffer such injuries in practicingtheir sport, either under the blows of their opponent or when the cartilagewas broken by the pullingof the ear. Present-dayboxers and wrestlersoften displaysimilarear configuration(P1.17:d). formelhafteOhrist bei Boxerund Kopf 4See, e.g., Schmidt, p. 70: " ... dasfleischige,wenigdifferenzierte, gebildetwie bei dem Kouros-Kopf Rayetsehr dhnlich... "; p. 75: "Das Ohrist nichtmehrso kompaktfleischig

OF KOUROIAND KORAI-ATTICVARIETY

119

can verify that this is an iconographic,not a stylistic, trait. Specifically,the central orificeis quite small and the cartilagesurroundingit (the antihelix)appearsswollen, with creasesmarkingthe puffinessof the area, which is barelydistinguishedfrom the helix.5

This realisticrenderingis well attestedin 4th-centuryand Hellenisticsculpture;it was, however, generally surmised that it would not occur before the time of increased

in sportand, specifically,of decreasingidealizationin art. But the disprofessionalism covery of the so-called Boxer Stele in 1953 forced a revision of this dating: the relief

can be ascribedto the mid-6thcenturyB.C. on the bases of style and format,yet not only does the bearded man on the stele display cauliflowerears and a broken nose but

his raisedarm is clearlybound by the leatherthongs of the boxer.6The deceasedis intentionallycharacterizedas a mature athlete who survived many encounters, although it is perhapsunnecessaryto speculate that he might have met his end in one last match. This characterization,surprisingas it might seem at this early a date, is not alto-

getherout of placeon a gravestone.In the largecorpusof Attic stelai, severalArchaic examplesidentifythe deceasedas an athlete, while others depicthim as a warrior,a priestor a man of status.7But I did not knowof any similarindividualization in sculpture in the round,specificallywithinthe kourostype with its strictadherenceto a neutralformula.I have thereforemadean (admittedlysuperficial)examinationof the ears of kouroi,withoutbeingable to find a close parallelto those of the RayetHead.To be sure, severalmale figureshave thickand rathershapelessears, and it is impossiblein many cases to determinewhetherthe renderingis due to the early date or the small size of the sculpture,the relativelack of skill of the sculptoror the desire to depicta boxer'sear. In particular,it is not unusualto find that helix and antihelixare more or less on a level, thus suggestinga degree of swelling.My criterionin such cases has been the size of the openinginto the middleear:when normallylarge,I have assumed that no deformitywas meantand that the ear patternhadjust been simplified.In fact, some sphinxes and a few korai exhibit similarrenderings.The Rayet Head stands apart.8 ... " (in comparisonwith the Potter's Relief on the Akropolis). For a more general statement see J. Boardman,GreekSculpture,TheArchaicPeriod,London 1978, p. 83: "thickears". 5For terminologyand a diagramof the human ear see Kouroi,p. 17. 6BoxerStele: AGA, no. 31, fig. 92; Jeffery, p. 128, Stelai, no. 1. She, however, interpretsthe nose as possiblyhooked ratherthan definitelybroken. 70n characterizationon stelai see ArchaicStyle, p. 169; cf. also p. 167. For various examples see AGA, nos. 25-27 (athletes); 45, 46, 67 (warriors);70 (priestor man of status). Comments on the individualizationof grave stelai and the youth of the kouroi during the Archaic period are also made by B. Schmaltz, "Verwendungund Funktion attischer Grabmiler," MarbWinckPr, 1979, pp. 13-37; see especially pp. 35-36, where the Rayet head is mentioned as having swollen ears for characterization.My point of view as regardsthe kouroi is somewhat different. 80ne possible example of cauliflowerear would be even earlier than the Boxer Stele: the head from the Ptoion, Athens N.M. no. 15; Kouroi,no. 10, fig. 75 (ca. 580 B.C.).Only the left ear appearsdeformed, but this is not unusual in later times (see footnote 28 below). A break and an imperfectionin the stone cross the feature, however, and may have been responsiblefor the undetailedrendering.

120

SISMONDO RIDGWAY BRUNILDE

Becauseof its allegedfindspot,the CopenhagenHeadis likelyto have belongedto a funerarystatue, since many such monumentslined the road to Peiraieusnear the homonymousGate. Severalof them were incorporatedinto a stretch of the ThemistokleanWall in the vicinityand were excavatedby J. Threpsiadesin 1953.9Among torsoand partsof the legs of a kouroswhichG. Schmidthas them are the fragmentary connectedwith the RayetHead.10The latter,althoughlargerthan life-size,may be too small for the body,1"and the torso, at least in its presentstate, displaysno specific athletic connotation.The attributionof head to body remains uncertain.The fragmentarykouroshas also been tentativelyassociatedwith an inscribedbase for a Samian, or with anotherset up for the KarianTymnes,12but no joins exist. In turn, the Ny CarlsbergHead has been attributedto the as yet unidentifiedstatue which stood once on the so-calledBall PlayersBase in the Athens NationalMuseum, also found nearby.13Thatathleticcontests,includingwrestling,shouldbe depictedon the pedestal but appropriate, seems particularly of a figurecharacterized as a boxeror a pankratiast againcertaintyis lacking;moreover,the RayetHeadmay be too early (ca. 530?) for a basegenerallyplacedwithinthe last decadeof the 6th century. It has so far been assumedthatthe RayetHeadbelongedto a kouros-typestatue;I wonderwhetherthis was indeed the case. As late as the impressiveAristodikos(ca. 490) the standardformulafor the nude male gravestatueseems to have been retained, and I have arguedelsewherethat it may have borne little resemblance(in age, physA kouros, originallya iognomy or otherwise)to the deceased it commemorated.14 of Apollo, would have remaineda symbolof heroization,while greater representation sense of identitywas attainedby depictionson stelai. Towardthe end of the 6th century,with the widespreadinterestin the PanathenaicFestivaland the OlympicGames, 9Besidesthe general account in Jeffery, pp. 126-128, see also Willemsen, and ArchaicStyle, pp. 292295. The excavationreportby Threpsiadesappearedin HpaKTrLKa,1953 [19571,pp. 61-71. I0Schmidt,pl. 32. The torso is preservedapproximatelyfrom waist to mid-thighs,the nbn-joiningright leg preservesthe lower thigh and the knee, the left extends from mid-kneecapto ankle. Accordingto early reports, a foot had been found with the Rayet Head but is now lost; this information,however, may serve to discreditthe furtherassociationof a plinthwith a kouros' feet: Schmidt, p. 73, and W. Deyhle, "Meisterfragender archaischenPlastikAttikas," AthMitt84, 1969, pp. 1-57, especiallyp. 29. On the body fragments per se see U. Knigge, "ZumKourosvon pireiischenTor," AthMitt84, 1969, pp. 76-86, pls. 33-35. 11Cf.Schmidt, pl. 31, p. 69 and note 18, where the relative proportionsare said to be at least not incompatible. 12Cf.Schmidt, pp. 73-74 and notes 39, 41, 42; Knigge, op. cit. (footnote 10 above), pp. 85-86. For a discussionof the bases see Willemsen, pp. 123-129. 13SeeSchmidt, p. 73, note 37, and Willemsen, p. 135. Note that not only the Ball Players Base was found in that general area but also the Boxer Stele and the so-called MarathonomachosStele: Jeffery, p. 128, Stelai, no. [2]. Since the helmeted youth is usually taken to be a hoplitodromosor a pyrrhicdancer, the monuments from that vicinity would seem to be particularlyconnected with athletic representations. For the monument of an Olympicvictor see Willemsen, pp. 110-117; cf. also his pp. 129-136, for another renderingof the ball-playingscene on a different base. Jeffery, p. 123, no. [4], lists the Rayet Head as coming from a slightlydifferentarea, but see her statement on p. 116. 14Aristodikos:Kouroi, no. 165, figs. 492, 493. My comments are in ArchaicStyle, pp. 49-59, 77. Cf. also Jeffery,p. 150.

OF KOUROIAND KORAI-ATTIC VARIETY

121

victors' statues began to be erected-some in the increasinglypopularmedium of bronze,but some still in marble.In eithercase, theircompositionmay have been more animatedthan the traditionalkourospose, as suggestedby some bases and even some may The same transformation "narrative"monumentsfrom the AthenianAkropolis.15 have taken place among funerarystatues, and the Rayet Head could indicatethat it occurredas earlyas ca. 530. Althoughthe more symbolicand statickouroscontinued to markgravesuntilalmostthe end of the Archaicperiod,a new athletictypemay have existed side by side with it, eventuallyreplacingthe generickourosentirelyand giving rise to the long line of Severeathletesknownin both life- andunderlife-size. The RayetHeadhas not been convincinglyattributedto any of the knownArchaic mastersactive in Athens, althoughstylisticsimilaritieswith the Boxer Stele, the Potter's Reliefand the BallPlayersBasehave been pointedout. Endoiosand Aristoklesare known to have made monumentsfor burialsin the vicinityof the PeiraieusGate.16 More significantly,tombs for foreignersseem to have been gatheredin that particular area, as indicatedby the Karianand Samianepitaphsalreadymentioned.Perhapsthe of an athlete'sstatuewas promptedby the wishesof a familyfromAsia characterization Minor,wherewrestlingwas almostthe nationalsport.The literarysourcestell us that the Olympicrulesfor boxingwereestablishedby Onomastosof Smyrna,the firstOlympic winnerof the event in 688, and that Pythagorasof Samos,who won in 588, was the firstscientificboxer.17 Cauliflowerears, althoughwith less picturesquelanguage,are also mentionedby the ancientauthors.Plato refersto young oligarchs"withbrokenears," who imitated 15Pausanias(vi.18.7) seems to ascribethe first victors' statues to the late 6th century; cf. C. Mattusch, "The Berlin Foundry Cup," AJA 84, 1980, p. 443 and note 46. P. Levi, in his Pausanias:Guideto Greece II, London [Penguin Books] 1971, p. 337, note 152, comments that a 7th-centuryvictor's statue had already been mentioned by Pausaniasat Olympia, but that "Probablyit was a stiff, massive, primitive Pharaoh-likefigure, and Pausaniaswas instinctivelynot thinkingof it as an athletic portrait."This would imply that the later statues looked differentfrom the standardkouros. For an analysis of male statues found on the Athenian Akropolis see ArchaicStyle, p. 49, note 3. Raubitschek (pp. 80-82, no. 76) has suggested that the inscribed base E.M. 6379 held a dedication by Phayllosof Kroton, who mentions that he won the pentathlonat Delphi three times; the base supporteda marble statue but apparentlyone more complex than a standardkouros. Raubitschektentatively associates with it the so-called Blond Boy (Akr. 689). I would question the Blond Boy as a depiction of a pentathlete because of his elaborate hairstyle, unless he represented Theseus and thus only symbolicallyalluded to Phayllos' prowess. For other comments on victors' monuments see Raubitschek, p. 464. For a typical animatedgroup, perhapsTheseus wrestlingSkyron, see AMA, no. 410, pp. 281-283 and figs. 326, 327, pls. 155-157 (Akr. 145). "6Thestylistic groupingof the Rayet Head had alreadybeen attemptedby Poulsen (footnote 2 above). Boardman(footnote 4 above) seems to accept the connection with Endoios' workshop;cf. Deyhle (footnote 10 above), pp. 13-20 (Endoios); Schmidt;and Willemsen, p. 135. 171tis impossible to differentiatebetween a boxer's and a wrestler's ears; cf. footnote 3 above. For Onomastosof Smyrnasee Philostratos,de gymnastica,12; for Pythagorasof Samos see Diogenes Laertius, viii.47. These references are given by H. A. Harris, GreekAthletesand Athletics,London 1964, p. 98 and note 49 on p. 206. For an Archaicrelief with fat wrestlers from a Lycian tomb see E. Akurgal, Die Kunst Anatoliens,Berlin 1961, p. 135, fig. 86. See also B. Schrbder,Der Sportim Altertum,Berlin 1927, pl. 23, for coins of Aspendos showing wrestlers.

122

RIDGWAY BRUNILDESISMONDO

the Spartansin theirfashions(includingboxing)in orderto publicizetheir politicalinVariousgods and heroes clinations.18A boxer's physiqueis describedas otothladias.19 are cited as the mythicalpredecessorsof the Olympicathletes.Theseus, althoughsupposedlythe inventorof wrestling,is shownwith normalears, even when his opponent Apollo, another"inventor",was credsometimesclearlyexhibitsthe disfigurement.20 I have been unableto find a sure representation ited with outboxingAres at Olympia.21 of Apollowith cauliflowerears, but the Ares Ludovisitype, at least as reproducedby Romancopyists,definitelydisplaysthe deformity.22 Heraklesis shownwith swollenears as earlyas the Severe period.In the metope with the Nemean Lion, from the Temple of Zeus at Olympia,the hero is depicted youngand tiredafterhis firstlabor,apd with the obviouslyswollenears of the boxer; yet the subsequentmetopes,as far as we can tell in their presentstate of preservation, Perhapsthe Strugglewith the Lion was so close to a typical omit this characteristic.23 schemathat the Olympiamaster,although wrestlingbout in its traditionaliconographic electingto show a later moment in the event, wantedto alludeto the more familiar version.Conversely,the hero mighthave been portrayedprogressivelyless humanand more divine. Among the other sculpturesfrom the same building,the Lapithbeing bitten by a Centaur,on the West pediment,has cauliflowerears, while the Centaur The beinggrabbedby the ear exemplifiesthe tacklewhichcausesthatkindof damage.24 renderingis thereforewell attestedby ca. 460 B.C.in undoubtedGreekoriginals. For the laterperiodsour evidenceincreases,althoughsome examplesremainquestionable.On the Parthenonfrieze severalridersseem to have swollenears,25and the youthful head with a lion-skin helmet from the Tegea Temple has them beyond doubt.26The best knownexampleis perhapsthe bronzeBoxerin the TermeMuseum but in Rome;by the Romanperiodcauliflowerearsidentifyprimarilythe professional,27 18Plato: Gorgias,515e; Protagoras,342b 8; cf. the comments by E. R. Dodds, Plato, Gorgias,Oxford 1959, p. 357. Diogenes Laertius,v.67, speakingabout the philosopherLyko; Yalouris (OlympicGames, I9Otothladias: p. 222) cites it as a boxer's name. For the effects of the palaistraon ears: Philostratos,Heroicus,167.13. 2001ympicGames, pp. 82 and 202; fig. 108 illustratesa red-figuredcup by the Kodros Painter, from Vulci, in the BritishMuseum: London E 84, Beazley, AR V, p. 739, no. 4, ca. 430 B.C. Theseus, with normal ears, is fightingKerkyon,who has a swollen ear. 21 Olympic Games,pp. 35, 82, and 216. 22AresLudovisi:see S. Lattimore, "Ares and the Heads of Heroes," AJA 83, 1979, pp. 71-78, especially pl. 3, figs. 4, 5, pl. 6, fig. 13. 23Heraklesand Nemean Lion: B. Ashmole and N. Yalouris, Olympia,London 1967, fig. 147; contrast the renderingof the ear in fig. 145. Only the Herakles of the metope with the eighth labor, the Mares of Diomedes, may have the same swollen rendering:Olympia,fig. 177. 24Lapith,West Pediment, Figure Q: Olympia,fig. 88. Centaur,West Pediment, Figure D: Olympia,fig. 74. 25ParthenonFrieze: see, e.g., F. Brommer, Der Parthenonfries,Mainz 1977, pls. 131 (S XIII), 135 (S XVII). Some such ears may occur also on the Lapithsand Centaursof the South metopes but I cannot be sure. 26TegeanHead: A. F. Stewart, Skopasof Paros, Park Ridge, N.J. 1977, cat. no. 16, pl. 13:b and pl. 52, ear 2. 27TermeBoxer: OlympicGames,fig. 120; Roman boxers and practices,ibid., pp. 275-285 and fig. 153.

OF KOUROIAND KORAI- ATTIC VARIETY

123

a few Greek citizenscontinuedto have themselvesso portrayedduringthe Empire.28 The Conversely,the Etruscansdo not seem to have imitatedGreekart in this respect.29 Rayet Head, if not the first, may representone of the earliestexamplesof this renderingin Greek sculpturein the round, in the shift from the funerary-Apollo/anonto the moreindividualistic portrayalof athletes. ymous-kourosrepresentation The questionraisedabout the varyingmeaningof the kourostype in Attica can also be formulatedaboutthe Akropoliskorai,and althoughmy commentsmust remain moretentativethanin the case of athleticfigures,some pointsare perhaps considerably worthmaking. I had sharedin the generalsurmisethat these elegantfemalededications(P1.17:e) on the Atheniancitadelrepresentedgenericattendantsto Athena,whetherpriestesses, or arrephoroi.30 Specifically,I acceptedthe theorythatthey displayedcontempergastinai oraryfashionsof the Athenianaristocracy,regardlessof whetherthey depictedactual Atheniangirls or were simplypleasingimagesoften donatedby men. A monographic study has even affirmedthat these smilingcreatureswere the socialexpressionof the upperclassand of a city statewherewomenlent prestigeto theirhusbands.They would have been renderedas they appearedat festivals, the normal showcasefor female beauty,as sung by the lyricpoets.31I have now had occasionto re-examinethe issue and have come to questionthe diagonalmantleas an item of clothingworn by Athenianwomenduringthe 6th century. In chronological order,the factsseem to fall in this sequence: 1) The Athenianwomen used to weara dress fastenedwith long pins;when they turnedthese ornamentsinto weaponsto kill the sole survivorof the Battleof Aigina, they were madeto changefashionsand to adopt"Kariandress"as definedby Herodotos.32This statementhas usuallybeen takento meanthatwomenshiftedfrompeplosto a chiton/diagonal-himation combination.33 The Battle of Aigina is poorly dated but seems to have occurredstill in the 7th century.If this is the true natureof the change, it is not reflectedin any definitewayeitherin vase paintingor in sculpture.Perhapsthe 28E.B. Harrison, TheAthenianAgora, I, PortraitSculptures,Princeton 1953, no. 14 and note 2; no. 25, note 11. The renderingis limited to only one ear. 29Thisconclusion was reached by Sarah U. Wisseman in her Bryn Mawr College dissertation (1981): TheArchaeological Evidencefor EtruscanGames. 30ArchaicStyle, pp. 50, 108-112; see also the review by A. F. Stewart, ArtB 62, 1980, p. 486. Since I have discussed the costume of the korai at some length in my book, I shall limit myself to brief allusions here. 31Schneider,passim. See especially pp. 27-29 for the smile as a social expression of the upper class, typicalof the gods and therefore makingman "godlike".Appendix 1, on vase paintingswhere koraiappear as spectatorsto heroes' deeds, suggests that they function as indicationof the hero's aristocraticmilieu, not as literalparticipantsto the scene. 32Herodotos,v.87, where mention is made only of linen tunics, and his footnote, where the dress is called Karian. 33E.g., Korai, pp. 9-10. The change is dated to ca. 560 and attributedto both the lonian leanings of Peisistratosand to the anecdote mentioned by Herodotos, with a consequent lowering of the date of the Battleof Aigina.

124

RIDGWAY BRUNILDESISMONDO

change implied only the adoption of the chiton under a heavy, symmetrical mantle, such as we see on one of the earliest Athenian korai, Akr. 593.34 2) The earliest korai with diagonal himation on the Akropolis (ca. 560 B.C.) are imports or works by foreign masters: Akr. 619 and 677 are probablyCycladic; other, later statues are Chian (Akr. 675); and the activity of Lakonian (Gorgias) and possibly Ionian (Bion, Endoios) masters is attested by surviving signatures on dedications.35 That the diagonal mantle was at home in Asia Minor is not in question, although we cannot be sure whether it was worn by humans or by cult images (Artemis?). The literary sources conveniently collected by Schneider usually refer only to the trailing chiton, or use more generic terms. In addition, most of the poets quoted belong to Ionia or the islands. 3) By ca. 540 the wide distributionof the diagonal himation in all territoriesand colonies inhabited by Greeks-regardless of climate-is surprising,especially since the garment seems a latecomer within Archaic fashions, or at least it can be said not to coincide with the beginning of monumental sculpture.Perhaps the reason for the diffusion should be sought in the means of transmission:not in terms of the clothes per se but of the many terracotta statuettes reproducingdivinities, which reached even the Western Greeks as copiously as those in Greece proper.If these statuettes depicted cult images or Nymphs or other non-human beings of some kind, fidelity in reproducingthe costume may have been considered essential, although not paralleledby contemporary daily fashions.36 4) The earliest Attic kore from the Akropolis to display the diagonal mantle wears it with peculiar draping:the so-called Lyons Kore (ca. 540) shows no indication of fasteners for the himation, and the rendering cannot be coherently explained. Later Akropolis korai continue to show unusual variations:the himation appearsto be in two pieces, fastened along both shoulders (Akr. 605, 611, 678); or it is buttoned along the right arm but then again pinned once over the left shoulder (Akr. 600 and 673); or it hangs with balancingswags under both elbows, in unexplainablefashion (Akr. 685; P1. 17:e)0.3The diagonal border between the breasts is ornamented with an abundanceof frills that often defy understandingunless they were separately applied to the folded garment. 5) It may be coincidence due to the chance of the finds, but none of the funerary korai from Attica, both earlier and later than the Lyons Kore, wears the diagonal him34KoreAkr. 593: ArchaicStyle, pp. 104-105, fig. 29 and bibliographyon p. 118. For the date of the Battleof Aigina see, e.g., J. B. Bury, A Historyof Greece,3rd ed., London 1955, p. 204, "probablynear the middle of the seventh century." L. H. Jeffery (ArchaicGreece,London 1976, p. 84) would go as early as the second half of the 8th centuryB.C. 35Fora discussionof these koraisee ArchaicStyle, pp. 104, 106, 118; the mastersare mentioned on pp. 285-286 and in note 3. 36Theidea of a famous cult statue in Miletos as a possible prototype for the East Greek korai has alreadybeen advanced:B. Freyer-Schauenburg,Samos, XI, Bildwerkeder archaischenZeit und des strengen Stils, Bonn 1974, p. 47; ArchaicStyle, p. 98. 37LyonsKore: ArchaicStyle, pp. 105-106; the other renderingsof the himation:pp. 92, 94.

OF KOUROIAND KORAI- ATTIC VARIETY

125

ation. I once suggestedthat they are depictedwith indoorattire,38but the difference may be of anothernature.Korai from the Citadelwearingjust the chiton are also known, and may indicatethat not all female statues should be interpretedin similar manner;in particular,Akr. 683, the Red-slippersKore, was probablythe smallerfigure in a doublededication(by two men) whichmay thereforehave had additionalmeaning, such as thatof a humanvotarynext to a goddess.39 6) The inceptionof the fashion (diagonalhimation)has been connectedwith Peisistratos'ties with Ionia;40yet the real popularityof the mode seems to occur after Peisistratos'death, towardthe end of the 6th century.Accordingto statisticsworked half of all survivingdedicaout by A. Raubitschekand E. B. Harrison,approximately between500 and 480. were made 6th 5th centuries for the and tions to the Akropolis Of the 56 more or less preservedkoraicataloguedby Langlotz,most were made between 530 and 490, and Ionic influenceseems confinedto them. The dedicatorsof but amongthe 18 offeringson the Akropolisincludedsome membersof the aristocracy, and an Ionianwho used a Chiansculpa washerwoman womentherewere, remarkably, tor. Koraiwere set up by a potter,a fullerand a tanner.41 7) The periodaround530 coincidesalso with a markedincreasein the numberof extantPanathenaicamphoras.A studyby J. R. Brandtconcludesthat duringthe rule of Peisistratosthe PanathenaicFestivalseems to have undergonedecline and stagnation in the ratherthanexpansion,to judgefromthe numberof extantvases. Standardization patternof the amphorascoincideswith the rule of Peisistratos'sons, who may have used the expansionof the event as a politicalmeansto re-establishfriendshipwith the Koraiearlierthan 530 B.c. are thereforeless likely to have repalienatedaristocracy.42 and those whichfollowedshouldnot have resentedaristocraticergastinaior arrephoroi, variedin theirmeaning.43 vases beforethe last quarterof the 6th centurydo not depict 8) Attic black-figured a diagonalhimation,althoughthe garmenthad alreadybeen seen on the Akropolis. 38Archaic Style, p. 103. 39Archaic Style, p. 107 and note 28, fig. 19; see also index for additionalmentions and bibliography. 40Korai, pp. 9-10.

4'Raubitschek,pp. 464-467, especially p. 465; the washerwoman:no. 380 (dedicated a basin); the lonian woman: no. 3; the potter: no. 197; the fuller: no. 49; the tanner:no. 58. E. B. Harrison, TheAthenianAgora, XI, Archaicand ArchaisticSculpture,Princeton 1965, p. 8 and note 49. 42J. Rasmus Brandt, "ArchaeologiaPanathenaicaI: PanathenaicPrize-vases from the sixth century B.C.," Acta ad Archaeologiam et ArtiumHistoriamPertinentia,InstitutumRomanum Norvegiae 8, 1978, pp. 1-23, especiallyp. 19 and notes 6 and 11. 43Wehave no definite depiction of the Arrephoroi,even from later times, since the Erechtheion Karyatids,occasionallyidentifiedas such, hardlyqualify.A fragmentaryhydriaby the Kleophon Painterin the collection of the University of Tilbingen, E 112, supposedlyshows Aphroditebetween arrephoroi:Stewart, loc. cit. (footnote 30 above); AthMitt93, 1978, pl. 13:2. The young women, however, are not distinguished by objects carried on their heads (as described by Pausanias), and they wear mid-Sth-centurycostumes. The only comparabletrait may be the gesture of the first woman on the left, who lifts her (long) himation: perhapsa misrepresentationof the pose of the korai, or an allusion to the Charites and the Horai, more appropriatecompanionsto Aphrodite.

126

RIDGWAY BRUNILDESISMONDO

Conversely, the fashion is rendered in red-figurescenes, but most of them are, either openly or possibly, mythological in content. It is becoming increasinglyapparentthat even what we take to be representationsof daily life may have had heroic connotation for the contemporaryuser." It seems difficultto reconcile the tempo of fashions in vase painting and sculpture. Certainly nothing in the techniques made it inherently impossible for a black-figurepainter to depict a diagonal himation or particularlyappropriatefor a red-figuremaster to do so. 9) Figures of non-human nature are clearly shown with the diagonal himation: the korai flanking the chariot of Apollo on the East pediment of his temple at Delphi-a site which has otherwise yielded no freestandingkore; the maidens on either side of the central floral akroterionon the Temple of Aphaia at Aigina-a position usually reserved for creatures of the air or at least endowed with special motion; Nikai at a variety of sites, including the Akropolis. It should also be stressed that the diagonal ruffle becomes the distinctive trait of archaisticsculpture and that perirrhanterion-holders,idols and karyatidsare often depicted wearingsuch a mantle. Even when the costume actually shown is the peplos, the lower hem is treated as if it were part of a diagonal himation, a sign of venerability and antiquity. It seems remarkablethat such a widespread fashion should be abruptlydropped from sculpture everywhere at the inception of the Severe style, just when statues became personalized rather than generic, and that the few examples, either in the round or in relief, which retain the diagonal mantle betray their later date only because of archaisticfeatures.45 All this evidence is undoubtedlycircumstantialand I want to stress once again the extremely tentative nature of my suggestions. It is nevertheless worth considering the possibility that the diagonal himation originated in Asia Minor, perhaps as part of the dress of a famous cult image and maybe even under Anatolian influence. From there the motif, carrying the same religious connotation, was distributed throughout the Greek cities. On the Athenian Akropolisthe garment arrivedthrough the dedicationsof foreigners, carved by foreign masters. When the Athenians imitated it, they did so with some misunderstanding,since the fashion was not current in their city. The elaborate dress became the hallmarkof a "non-human"image, thus appropriateas a dedicationto a divinity that could then be offered by aristocracyand lower classes alike. The garment disappearedin the Severe period, when each statue became individuallycharacterized, but was retained in archaisticart through the centuries. If all the above is near the mark, to the question "Who are the Akropolis Korai?" we should answer:"Not Athenian aristocraticgirls, ergastinaior arrephoroi,but Nymphs 44See,e.g., J. Neils, "The Group of the Negro Alabastra,"AntK 23, 1980, pp. 13-23, esp. p. 23; idem, "The Loves of Theseus: An EarlyCup by Oltos," AJA 85, 1981, pp. 177-179. 451ffashionsin sculptureare taken as true reflectionof daily life, the change to the Severe peplos must coincide with a change in costume in Athens and everywhereelse. Patrioticreasons might have prompted the rejection of an eastern garment, but this explanationseems less valid outside Greece proper, and in any case it is not reflected in Attic red-figuredvases until some years later. On Archaisticfashions see Harrison,op. cit. (footnote 41 above), pp. 51-61; ArchaicStyle, chap. 11 and esp. pp. 313-316 and, on the Archaickorai, pp. 113-114.

OF KOUROIAND KORAI-ATTICVARIETY

127

or lesser deities in distinctive attire; they may have originally portrayeda major goddess, but their meaning became diluted into generality in the course of the 6th century in Athens." BRUNILDE SISMONDo RIDGWAY BRYNMAWRCOLLEGE

Departmentof Classicaland Near EasternArchaeology BrynMawr, PA 19010

PLATE

d. A modern cauliflower ear. Photo courtesy of Douglas Nigro

a. The Rayet Head, Copenhagen,Ny Carlsberg Glyptothek

b. Rayet Head, right ear

c. Rayet Head, left ear

e. Kore Akr. 685

17

THE DEMOLISHEDTEMPLEAT ELEUSIS (PLATES18-20) T

HE MUSEwho governsthe fortunesof excavatorsin the field has reservedit for the giants alone to explorethe remainsof buildingswhich were never built, to reconstructmonumentswhichstood on unknownsites, and to interpretthe historyof structuresdemolishedin antiquity.To ordinarymortalssuch feats may seem to require a magician'sartsof legerdemain;and yet when touchedby HomerThompson'sskillful hands, the stones of Athens have producedbefore our eyes the unbuiltstoas on the Pnyx, the migratoryTemple of Ares, and the SquarePeristyle,demolishedfor the conjuring,it constructionof the Stoaof Attalos.In view of this historyof archaeological to dedicateto him a story of two Attic buildings,one of seems peculiarlyappropriate in the 5th centuryB.C. to make way for the other whichwas never whichdisappeared constructed. The long and complexhistoryof the sanctuaryof Demeter and Kore at Eleusis finds its most graphicexpressionin the survivingremainsof the Telesterionitself. From a small Mycenaeanmegaronto the ceremonialhall of the Classicaland later periods,the temple doubledand redoubledits size in responseto the ever-growing popularityof the cult whose pilgrimsand pageantsit accommodatedthroughoutantiquity. Unique among the sanctuariesof Greece, the prosperityof the Great Mysteries can be readin the physicalexpansionof the templein whichthey were celebrated.The building'svicissitudeswere legion and have taxed the ingenuityof archaeologistsno less than the elusive secretof the Mysteriesthemselveshas tantalizedthe historianof religion.The ruins of the enormoushypostylehall, whichhoused the Mysteriesfrom Pericleanto Romantimes, forma familiarlandmarkfor everymodernvisitorto the site (P1.18:a).The tiers of rock-cutsteps and the northwesternwallquarrieddeep into the living rock of the acropolisso dominatethe site todaythat it is easy to overlookthe predecessorsof the ClassicalTelesterion;but beneathits floor, half a centuryof excavationhas revealedthe tangledremainsof five earlierstructureswhichservedthe cult continuouslyfromthe BronzeAge to the early5th centuryB.C. Into this pit of fragmentarywallsand re-usedfoundationsfew but the most dedicatedinitiatesof Eleusishave venturedto descend.The investigationsof earlierexcavatorshave served to elucidate the historyof the Telesterionin all of its successiveremodelingsand enlargementsso developmentare now clearlyunderthatthe generallines of the building'sarchitectural stood.1Thereremains,however,a shadowyepisodein the earlyhistoryof the greathall whichappearsas yet to have eludedexplanation. des Entwicklung 'The monumental publicationof the sanctuary,F. Noack, Eleusis,die baugeschichtliche Heiligtumes,Berlin 1927 (= Noack, Eleusis), has been considerablysupplementedby the excavationsof K.

THEDEMOLISHED ATELEUSIS TEMPLE

129

The Eleusinianepistatai,the overseersof the sanctuary,recordedin their annual inventoryfor the year 408/7, and again in 407/6, the sacredfunds and dedications belongingto the Two Goddesses,the annualincomeand expendituresof the cult, and in additiona considerableassortmentof architecturalblocks and buildingequipment whichwas storedin the sanctuary.2 Includedamongthe latterwere buildingblocksof Aiginetanporos and Pentelic marble,dressedblocks for the foundations,steps, and stylobateof a colonnade,columndrumsand at least one capital,new and unusedroof tiles in largenumbers,and an assortmentof lumber,some of whichwas storedaway from the elements under a tent. Togetherwith these were various builders'tools, ropes, scaffolding,and wagonswhichlend color to the pictureof a buildingsite where constructionwork is still in progress,or for some reasononly temporarilysuspended. The overseersalso cataloguedand passedon to the custodyof theirsuccessorsanother collectionof buildingmaterialsevidentlyremovedfrom an old buildingand storedin the sanctuaryagainstfuturere-use. This portionof the catalogueis introducedby the notation:"takendownfrom the temple,"and these materialswill presentlybe seen to yield some interestinginformationconcerningone stage of that building'shistory,even record.The thoughthey have perishedlong since and surviveonly in the epigraphical items in questionare listed in IG 12, 313, lines 103-110 for the year 408/7, and in IG 12, 314, lines 113-120 for the year 407/6. The betterpreservedversion of the text on the obverseof the stele is quotedhere. aIo ro

VEO KaOELpEzEva

XIRHHF

KEpaUO ;EVYE:

105

or0xv8vXo:

IIIII

rivE

p]a.:

ArI

xoivXa ErviTvXta: AAI EY 81VOL) KEKoXXEAE'Vo

0-EKLUKOL aro

110

TEl

I

oTOa'

AIIII

KaOELpEqEVOL

:

OvpOv

III :htKptT[Ep]a"

;EvyE:

:A

Taken down from the temple: 1750 pairsof roof tiles, 105 54 column drums, 16 Ionic bases, 21 wooden epistyles, Kourouniotes, AEXT,13, 1930/31 [1933], Hapapr'ux pp. 17-30; 14, 1931/32 [1935], Hap'pr., PP. 1-30; 15, 1933/35 [1938], HapapT.,PP. 1-48; idem, 'EEVO-LVLaKa , Athens 1937. Many of Noack's interpretations and chronologicalconclusions must be altered in the light of this work. See especially K. Kourouniotes and J. N. Travlos,

"TEXE0cpwLOV KaL vaos

Tr"jAb)ji)Tpoc,"

AEXT15, 1933/35 [1938], pp. 54-114 (

Kourouniotes-Travlos);Travlos, "To 'AVaKTopoV T7'q 'EXEvcivo-q,"'Apx'Eo, 1950-51, pp. 1-16 (= Travlos, "'AVaKTopoV");0. Rubensohn, "Das Weihehaus von Eleusis und sein Allerheiligstes," JdI 70, 1955, pp. 1-49; G. E. Mylonas, Eleusisand the EleusinianMysteries,Princeton 1961 (-Mylonas, Eleusis). [Plate 29, illustrating"The Sacred Threshing Floor at Eleusis," by E. Vanderpool, shows the model of the PeisistratidTelesterion.-Ed.] 2IG 12, 313, lines 70ff.; 314, lines 80ff.

T. LESLIE SHEAR, JR.

130

1 glued together from two, 18 rafters3taken down from the stoa, 110 3 pairsof doors, 10 roof timbers.4

Anotherdocumentfound at Eleusismentionsstill more blockswhichwere taken downfrom the temple.A decreeof the AthenianBoule and Demos datingto the year 421/0 (P1.18:b)authorizesthe constructionof a footbridgeover one of the Rheitoi,the two smalllakeswhichformedthe boundarybetweenthe plainof Athens and the Thriasianplainof Eleusis.The new bridgefor the sacredprocessionof the initiateswas to be constructedentirelyof re-used blocks, removed from the Archaictemple, which wereleft over aftersome workon the wallsof the sanctuaryhadbeen completed.5 STOIX:26

[HI]pE'7T EVtOEpoE'ypa/JtacTEVE. E8OXc vE TEL JoXEL KaL TOL8E.L0L

Ary7El EI7TpVTaVEVE,HpEMr9Euypa [A]AaTEVE,

5

[ahos;

HaTpOKX^;

EITEOTaTE,

e

EP(TOVTOU 7rapa TO [a]

ElTE rV

xlo'; XpoAtE[vI 'EXEvv[o61 EV TOYKa6ELPEJE[VI EAXrov OV EK TO VEO TOapXato, ho%' ES; TO TEtXO'; avaxLCtKOVTE';, ho' a' 10 v Ta%htEpa%O'pocuv hat hEpEat a OTEO'; yEfvpocrat

o'

[okaAXEcrTaTa. 7TXaTOq8E 7rItoVTOV [7]EVTEI7ro8a,

hiva

jAEhataxca

XavvovrTat, a&Xa Tois; incV

EL Oa

[81'EvEO

aE KaT

Tar htEpa. Xos

8&E

Ta% 15 [aKlakvkOaL &appoaqTO8 PPE[T] av Xo-vyyp(a)XkEL AEoAEUAEX LE';6 apXlTEKTOV. E]al6 8E ME ocil E[.]

o KaOoln

Prepisson of Euphereswas secretary.Resolved by the Council and the Demos; Aigeis held the prytany;Prepiswas secretary; Patrokleswas chairman;Theaios made the motion: to bridgeover 5 the Rheitos on the side towardthe city, makinguse of the blocks at Eleusis which had been taken down from the Archaictemple, are the raftersof a roof is indicatedby the use of the term in those portions of the 3That04-KiootK Erechtheion accounts dealing with the woodwork of the roof, see J. M. Paton and G. P. Stevens, The Erechtheum,Cambridge, Mass. 1927, pp. 314, 353, 366-369. Cf. A. T. Hodge, The Woodworkof Greek Roofs, Cambridge1960, p. 119. 4The technical meaning of hLKpLOTEpaq is not clear, nor is their function in the Telesterion. They are certainlyof wood and most likely part of the woodworkof the roof. In the specificationsfor the arsenalof Philon, the term is applied to vertical posts supportingthe shelves of the galleries, IG II2, 1668, lines 78-80; translatedas "posts" by K. Jeppesen, Paradeigmata,Aarhus 1958, p. 73, commentary,p. 84. Cf. the use of the term in the woodworkof ships, IG I12,1629, line 1156; 1631, line 339. 51GI2, 81, lines 5-9; for this interpretationof the work on the walls, lines 8-9, see SIG3, 86, note 5; for the Rheitoi, Pausanias,i.38.1; Hesychios, s.v. 'PELTOL.

THE DEMOLISHEDTEMPLEAT ELEUSIS

131

and which they left aside when they re-used them in the wall, in order that the priestessesmay carrythe sacredrelics as 10 safely as possible. Let them make the width five feet so that wagons may not drive across, but those going to the sacred rites may cross on foot. And furtherto cover over with blocks the 15 channels of the Rheitos as the architectDemomeles shall specify. But if they are not ....

Therecan be little doubtthat the phraseho VEOc,in an Eleusiniandocumentrefers to no other buildingbut the Telesterionitself, for it has been convincinglydemonstrated by Kourouniotesand Travlosthat the great hall of the Mysterieswas in fact the templeof Demeter.6In the 5th century,no other structurewithinthe sanctuarycould have been describedsimplyas "the temple"withoutfurtherqualification.It has long been recognizedalso that the buildingdescribedas ho VEoq ho a&pX'Cdo in IG 12, 81, line 8 and again more simplyas ho VEO' in IG 12, 313, line 103 (= 314, line 113) shouldbe identifiedas the Telesterionof the 6th centuryB.C., the constructionof which is commonlyattributedto the tyrantPeisistratos,althoughwith more chronological precisionit oughtpreferablyto be assignedto his sons.7Two entriesin the inventoryof 408/7 will shortlybe seen to confirmthis identificationof the building,for the overseers listed materialfrom a stoa, whichseems to have had some connectionwith the temple,and three pairsof doors.8 The survivingfoundationsof the ArchaicTelesterionwere found buriedbeneath the presenteasterncornerof the vast hypostylehallwhichsucceededit on the site in the second half of the 5th century.These remains,togetherwith cuttingsin the bedrock, have yieldedan accurateplanof the Archaictemple,almostexactlyone quarterthe size of the laterClassicalstructure.In additionto its foundations,the buildingsurvivestoday in numerousfragmentsof its superstructure (Pls. 19:b,20) whichcontributetheir evidence to the architectural reconstruction.9 The ArchaicTelesterionwas a large,square 6Kourouniotes-Travlos,pp. 54-114, and especially 59-61. The theory of Rubensohn (Die Mysterienheiligtumerin Eleusisund Samothrake,Berlin 1892, pp. 44ff.) that the temple of Demeter ought to have been a canonical Greek temple separate from the Telesterion, was followed by Noack (Eleusis, pp. 48, 218), who attemptedto locate the temple on the rock-cutterracejust north of the ClassicalTelesterion and referred to it as "Temple F" (ibid., pp. 85-88). This attempt was unsuccessful and the cuttings proved later to belong to a building of the Roman period, Kourouniotes-Travlos,pp. 72-75. Cf. also the discussion of G. E. Mylonas, TheHymnto Demeterand Her Sanctuaryat Eleusis,St. Louis 1942, pp. 28-63. 7See E. Cavaignac, Le tresor sacredd'Eleusisjusqu'en 404, Versailles 1908, pp. 38-39, followed by Noack, Eleusis,pp. 59-60, 68-70; Mylonas, Eleusis,p. 82; cf. also Dittenberger'scomment, SIG3, 86, note 4; Kourouniotes-Travlos,pp. 75ff. 81G 12, 313, lines 108-110. It may perhapsbe questioned whether the entries in line 110 belong with those immediately preceding since the line is separated from the others by a paragraphmark. But the identical entries in IG I2, 314, line 120 are not so separated.Here the mason placed his paragraphmark immediatelybefore line 113, where it had been omitted in the previous inventory; thus he set apartall the materialremoved from the temple. It is possible that the mark before IG I2, 313, line 110 is in fact an error. 9For detailed descriptionof the architecturalremains, Noack, Eleusis,pp. 48-68; Mylonas, Eleusis,pp. 78-88. The restored plan is conveniently reproduced in A. W. Lawrence, GreekArchitecture,Baltimore

132

T. LESLIESHEAR,JR.

hallapproached by threedoorwayson its southeasternside. Here a broadporticoof ten Doriccolumns10 crossedthe facadeof the buildingand projectedtwo intercolumniations to form a deep porch,of whichmanyof the originalporospavingslabs still remainin The threeotherinteriorwallswererangedwithbanksof nine stone steps,except place.1" in the westerncornerand extendinghalf way alongthe southwesternwall, wherethere of the stood a smallenclosedchamberknownas the Anaktoron,the sanctumsanctorum Eleusiniancult.12The building'sroof was carriedupon five parallelrows of interior columnsnumbering22 in all, andarrangedwithfive columnsin each rowexcepton the southwestside where two columnsalignedwith the front of the Anaktoron.Both the centeredrowof columnson the axis of the buildingand the survivingmarblesfromthe decorationof its roof showthatit was of the normalgabledtypewitha pedimentcrowning the facadeof the southeasternportico. The colonnadedporchof the ArchaicTelesterionwas the only structureof its kind builton the site beforethe greatprostoonof Philonwas addedto the Classicalbuilding in the late 4th century;and in fact it was the only buildingin the sanctuaryduringthe 6th and 5th centuriesto which the Greek word stoa could properlybe applied.The earlierArchaictemple of the beginningof the 6th centurywas a simple rectangular The later, so-called"Kimonian"reconstructurewith no evidence for a colonnade.13 structionwas designedto have its frontwallfoundedon the old stylobateof the Archaic Althougha stoa was projected porchand apparentlydispensedwith exteriorcolumns.14 and actuallybegunfor the ClassicalTelesterion,it was plainlynever built;and indeed we may suspectthat the columndrums,steps, and stylobateblockslistedin the invenThusthe toriesof 408/7 and407/6 are actuallymembradisjectaof this abortiveproject.15 materialslisted in the same inventorieswhichare said to have been takendown from the stoa could only have been removedfrom the porch of the ArchaicTelesterion. Includedamongthese, it will be remembered,were also three pairsof doors,while the existingfoundationsfor the southeastwall of the Archaictemple preservethe thresholds of three doorways.These were the only means of access to the hall of initiation since its other threewallswere linedwith steps and its westerncornerwas occupiedby the Anaktoron.Now the doorsof the inscriptioncan hardlyhave come from the Classi1957, p. 253, fig. 142; H. Berve, G. Gruben, M. Hirmer, GreekTemples,Theaters,and Shrines,New York 1962, p. 400, fig. 70. '?Kourouniotes-Travlos,p. 75. Noack (Eleusis,pp. 57-58) restored three possible schemes for the stoa with eight, nine, and ten columns, but he preferredthe facade with nine columns; followed by W. B. of AncientGreece,London 1950, p. 113. Dinsmoor, TheArchitecture 11Noack,Eleusis,p1.3. 12Theexistence of the enclosed chamber in the west corner and its identificationas the Anaktoron were first recognized and demonstratedby Travlos ("'Aa'aKTOpOI'," pp. 1-16, and especially p. 7, fig. 4). For the steps lining the walls, see Noack, Eleusis,pp. 95-97. I3Ibid.,pp. 16-23, and especiallyp. 19; Mylonas, Eleusis,pp. 67-70. 14Noack,Eleusis,pp. 99-101, fig. 47; Mylonas, Eleusis,pp. 111-113. 15Evidencefor the beginning of work on a stoa for the ClassicalTelesterion can be seen at the east corner of the buildingwhere the deep foundationsstep forward4.21 m. beneath the later foundationsfor the prostoonof Philon, Noack, Eleusis,p. 117, fig. 51; p1.39:d. -

ATELEUSIS THEDEMOLISHED TEMPLE

133

cal Telesterion;for it seems unlikelythat three of its six doors could have been removed and discardedso soon after their installationas to be cataloguedby the overseers of 408/7. On the otherhand,the so-called"Kimonian"phaseof the templeseems to have been designedwith only two entranceson its southeastfacadeand was, in any event, not sufficientlycompletedfor its doorsto have been hung in place.16Againone is led to supposethat the three pairsof doors mentionedin IG 12, 313 and 314 hung originallyin the doorwaysof the Peisistratidtemple.The two inventoriesand the decree (IG I2, 81) of the late 5th centurymay then be assumedin all probabilityto refer to buildingmaterialssalvagedfrom the Archaicstructure,the foundationsof which are preservedon the site today. and at what time in the historyof We may now inquireunderwhatcircumstances the buildingits roof and columns,its porchand doorswere dismantledand storedfor re-usein the sanctuary.Scholarshave explainedthe fate of the ArchaicTelesterionwith unanimityof opinion, and the conclusionhas been almost universally extraordinary acceptedthat the templemet its end in the Persiandestructionof 480 B.C. Indeed,no less a witnessthan Herodotosseemed to have madean explicitstatementto this effect when he recountedthe personalrevenge which Demeter exacted from the Persian troopsat Plataia."It seems to me a marvelousthing that thoughthe battlewas fought besidethe grove of Demeter,none of the Persiansdied in the precinctor even entered ground.But into it; and most of them fell roundaboutthe sanctuaryon unconsecrated I judge, if one may judge of the ways of heaven, that the goddessherselfrefusedto receive them because they had burnedher sanctuary,the Anaktoron,at Eleusis."17 Thatthe Persianarmydid, in fact, visit its wrathon the sanctuaryat Eleusiswith a fury nearlyequalto the sackof the AthenianAcropoliswas apparentto the earlyexcavators, who reportthe discoveryof burnedand brokendebrissimilarto that from the Acropolis.18The Archaicperiboloswall was apparentlybreachedat a pointjust southeastof the Telesterion,wherea sectionof the earlymud-brickwall was subsequentlyreplaced masonryfoundedon the originalpolygonalsocle (P1.19:a).19 with fine pseudo-isodomic Furthermore,other structuresassociatedwith the Archaicsanctuary,and in particular the SacredHouse, are known to have been destroyedat this time.20In view of the 16Travlosrestored three doors without comment in the "Kimonian" building as in its predecessor p. 13, fig. 4); followed by Lawrenceand Gruben, locc. citt. (footnote 9 above); but see ("'Ava'Kropo0V," Mylonas, Eleusis,p. 112, and Travlos' most recent version of the plan, ibid., fig. 26, C. On the incomplete state of the structure,see Noack, Eleusis,pp. 94-103. 17Herodotos,ix.65.2: OC'ux &' got O'KCV' 7rapa rTs Ar-,urqpos ro aXoosHEPcTE'C)V OV"TE EcTEXOC6 es cTOJ.

ro

TEqUIEVO- OVTE

OKEW 8E', Et Tt ITEpL TrVV OEL't

1p7)PffTWV

Eva7ro~ava'wv,

EuaXo/.tvwOv oi8E E'1s eOalq rC'V X [3E,37)AX(EI7rE-

7rEpL TE TO LpOV OL 7rITXELcTOL E TZ

8OKEELV86E,

-

6Eo% aVT-

cr-Eaq

OVK (E~KETO ETvp)caTava

a vaKropovO. [TO tpOId TO% 4 'EXEVOrtiVL

18Noack,Eleusis,p. 93. 19Thesection of the peribolosreconstructedin pseudo-isodomicmasonry is located between H 25 and H 29 on the plan, Mylonas, Eleusis,fig. 6. For discussion of the repairof the wall, see ibid., pp. 107-108; Noack, Eleusis,pp. 30-32, 90-92, and pl. 14 between C 6 and C 8. 20Forthe Sacred House, located to the south of the sanctuaryjust beyond the fortifications,see Kourouniotes, HlpaKrtKa, 1937, pp. 42-52; Mylonas, Eleusis,pp. 101-103, 192.

134

T. LESLIESHEAR, JR.

archaeological remainsand the statementby Herodotosthat the Persiansburnedthe Anaktoron,the innermostchamberof the Telesterion,all the evidence seemed to suggestthe totaldestructionof the Archaictemplein 480/79. Scholarshave maintained this date almost withoutexceptionand have gone on to describein vivid terms the holocaustwhichengulfedthe sanctuary.21 Closer examinationwill reveal a numberof facts about the ArchaicTelesterion which are at variancewith this pictureof utter destructionand may suggest that the buildingmorelikelysuccumbedto a ratherdifferentfate. In the firstplace,it shouldbe noted that much of its superstructure can be accountedfor in the late 5th century. Thereis no way of knowinghow muchmaterialwas re-usedfor repairingthe peribolos wall, but enoughwas left over aftercompletionof that projectso that the new bridge over the Rheitos was to be built entirelyof blocks salvagedfrom the old temple.22 Presumablyblocksof the most regularand useful shapeswouldhave been takenfirst, the wall blocks,the triglyphs,and architravesfrom the stoa. A numberof blocksfrom the exteriorDoric cornicewere re-usedin a short length of wall constructedon the polygonalsocle of the easternArchaicperiboloswall. It has been suggestedthat this may be partof the reconstruction es TELXOs of IG 12, 81, line 9;23 but whetheror not this is the wallmentionedin the inscription,its constructionis certainlyto be associated with the buildingof underground granariesto the east of the Archaicperiboloswall in the latterpartof the 5th century.Still other blocksfrom the Archaictemple,including anotherpiece of the Doric cornice(PI. 19:b),were employedin a greatsupplementary pierset againstone of the interiorpiersof the Classicalgranariesat some time before One interior they were filledwith earthand went out of use in the late 4th century.24 epicranitisblockand possiblysome of the wallblocksfromthe ArchaicTelesterionwere used as backersfor the fine pseudo-isodomic masonrywhichrepairedthe breachbroken by the Persiansin the south periboloswall.25Anothercorniceblock and a Doric anta capital(P1.20:a) were laterbuilt into the wallsof the little Romanbuilding,knownas while still other "TempleF", on the rock-cutterracenortheastof the Telesterion,26 21Ibid., pp. 88-90, 106-108; Noack, Eleusis,pp. 91f., 93; D. Philios, 'EXEVcTL', MvOrrTpuLa,'EpEL'lna

Kat

1884, pp. 75ff.; and cf. Cavaignac,op. Av r^s,Athens 1906, pp. 81, note 1, 90; idem, npaKTtKa, line 8. 81, note on IG 12, 7 86, note 4; SIG3, above), p. 39; cit. (footnote 221G12 1, lines 5-9 (p. 130 above). 23Suchwas the interpretationof Kourouniotesand Travlos, pp. 75-82. It is more likely that this wall was built when the low-lyingarea beyond the Archaicperiboloswas converted into undergroundgranaries. See Noack, Eleusis, pp. 189-193; Mylonas, Eleusis, pp. 125-127. The wall in question does not properly align with the polygonalsocle beneath it, and it abuts the northernmostpier of the granaries,Kourouniotes-Travlos,p. 76, figs. 11, 12. Cf. the plan Noack, Eleusis, p. 190, fig. 76, where the wall is indicatedby dotted lines runningnorthwardsat an oblique angle from the last pier in the right row. If this is one of the of which IG 12, 76, lines 10-12 authorizesthe construction,then the date of that inscription, three cnpo%, ca. 422 B.c.(?), providesa terminuspost quemfor the re-use of the cornice blocks from the Archaictemple. For a recent summaryof the discussionconcerningthe date of IG 12, 76, see R. Meiggs and D. M. Lewis, GreekHistoricalInscriptions,Oxford 1969, pp. 222-223. pp. 79-82, fig. 15. 24Kourouniotes-Travlos, 25Noack,Eleusis,p. 54. 26Ibid., pp. 86-87. These were shown to belong to the ArchaicTelesterion by Kourouniotesand Travlos (pp. 74-75); cf. Mylonas, Eleusis,p. 176.

Movo-JOV

ATELEUSIS TEMPLE THEDEMOLISHED

135

membersof the same seriesare to be seen in the west wallof the Archaicarchitectural long Romangranary,near the southwestcornerof the building.The well-knownsima blockof Parianmarble(P1.20:b), adornedwith its ornamentalram'shead, whichoriginally crownedthe southerncornerof the building,was found in the fill between the post-Persianrepairof the southernperibolosand the deep foundationfor the crepidoma The sima couldhave been buriedalreadyin the secondquarter of Philon'sprostoon.27 of the 5th century,when the originalfillingwas thrownin behindthe Archaicperibolos in orderto enlargethe terraceof the Telesterionabove;or the blockmighthave found its way into the fill whichwas replacedafterthe foundationsfor the prostoonhad been laidin the mid-4thcentury. In additionto all these piecesof the Archaictemplewhichsurvivetodayby virtue of their re-use in variouslater buildingsat Eleusis, there were extant in the late 5th centuryall the materialslistedby the overseersof 408/7 and 407/6, whichhad not been re-usedin otherconstructionup to that time. We have seen that these included54 column drumsand 16 Ioniccolumnbases,whichare to be assignedto the interiororderof the hall.28No less than 1750pairsof roof tiles had been salvagedfromthe buildingand survivedin the storeroomsas late as 407/6. It was possiblypartof this group,together with severalmarblesimas and antefixes(P1.20:c, d), whichwas found in the excavation of a small building,locatedon a rock-cutterracerisingtwo metersabove the SacredWay,just south of the Hellenisticexedra.29But by far the most strikingentriesin the inventoriesare the 22 woodenepistyles,the 18 woodenraftersfrom the stoa, and The epistylescame almostcertainlyfrom the interiorof the hall the ten roof timbers.30 and restedon the Ionic columns, 16 bases of whichwere still storedwith them, while the text statesspecificallythatthe rafterswere "takendownfromthe stoa." The foregoingsynopsisof existing or recordedarchitecturalmembersfrom the largeproportionof the building'ssuperArchaicTelesterionsuggeststhat a remarkably structuresurvivedthe Persiandestructionof the sanctuaryand was thoughtfit for reuse throughoutthe 5th centuryand later.Now, except for the sima and roof tiles of 27D. Philios, TIpaKrtKa, 1883, p. 63.

28Forthe attributionof these to the Archaictemple, see Noack, Eleusis, pp. 60-61; Mylonas, Eleusis, p. 82; W. B. Dinsmoor (op. cit. [footnote 10 above], p. 195, note 4) thought that they might have come from the post-Persianreconstructionof the Telesterion, no doubt because they survived to the end of the century, while the Archaic temple he had already consigned to destruction by Xerxes (p. 113). A. W. Lawrence(op. cit. [footnote 9 above], p. 252) assumed Dinsmoor's suggestion to be certain. 29Philios,'EXEVrq- (footnote 21 above), p. 86, note 1. It has been suggested to identify this building (Mylonas, Eleusis, fig. 4, no. 22) with one of the two treasuriesin the sanctuarywhich are mentioned in the overseers' accounts of 329/8, IG I12,1672, lines 300-302; see Noack, Eleusis, pp. 83-84; Mylonas, op. cit., p. 144. It is not impossible that some of the roof tiles from the Archaic temple were stored here during later periods. The fact that it would seem strange to us to store secondhand building materials in a treasurydoes not reflect the view of antiquity.We have only to recall that the treasurersof Athena for 369/8 recorded among the dedications in the west room of the Parthenon eight and one half boxes of rotten and useless arrows (IG 112, 1424a, lines 344-345: G-dipaKOL Pill Ka[l 7].upaKLOr T[o(]EVa'Tuwv rairpcivaxp-q~rwrw)). Their successors in 368/7 were perhapstidier housekeepers, for they seem to have reduced the useless arrowsin the Parthenonto one small box, IG I12,1425, lines 280-282. 30Seeabove, p. 129, lines 106, 108-110.

T. LESLIESHEAR,JR.

136

Parianmarble,the buildingwas entirelyconstructedof fine poros stone, of the kind whichis easilycalcinedon the surfaceand becomesextremelyfragileand friablewhen exposedto the heat of fire. It seems unlikelythat severelydamagedblockswouldhave been so carefullystored, recorded,and re-used, especiallyin the constructionof the bridgeacross the Rheitos. Far more surprisingis the very existence in the late 5th centuryof so manywoodenmembers,epistyles,rafters,beams,and doors, of a building whichhad been fired by the Persians.There is, after all, ratherlittle inflammable materialin a stone buildingapartfromthe doorsand the roof timbers;and it is scarcely crediblethat a fire, of such intensityas to destroyeven the chamberof the Anaktoron,31couldhave left the roof in such a state that it was worthsalvaging.It is well to recallthat the wallsof the AthenianAcropolisappearedscorchedby fire in the time of Herodotos(v.77.3), a generationafterthe Persiansack.Moreover,Pausanias,traveling in the 2nd centuryafterChrist,saw a numberof buildingswhichhad been burnedby the Persiansand were neverrepaired.Of one he writesin particular,"Onthe wayfrom Phaleronto Athensis a templeof Herathat has neitherdoorsnor roof; they say it was The roof was the one partof the building firedby Mardoniosthe son of Gobryas."32 whichwas certainto perishin any extensivefire;and yet woodenepistyles,rafters,and timbers,togetherwith a large quantityof roof tiles, seem to have survivedin sufficientlygood conditionto makeit worththe troubleof storingthem for 70 years. memberswill increaseone's doubtsthat An examinationof the extantarchitectural the ArchaicTelesterioncan have sufferedextensive damageat the handsof the Perfine state of preservation sians.These blockshave all in commontheir extraordinarily noteworthyin this upon whichthe excavatorshave especiallycommented.Particularly respectare the piecesof the marbleroof and the poroscorniceblocks.A considerable numberof marblesima tiles and antefixesof the cover tiles have survived,and far from showingsigns of damageby fire, manyretainto this day tracesof their original painteddecoration(P1.20:c, d). The palmettesof the antefixeshad 11 red petalsoutlined in blue, while the rakingsima was decoratedwith a bandof alternatinglotus and palmettessurmountedby a narrowbead-and-reelpattern.Againthe petalsof the blossoms and the beadswere paintedred and outlinedin blue.33Equallywithouttraceof burningis the cornersima (P1.20:b) of which the ornamentalram's head displaysa remarkablefreshness.The originalpaintwas partiallypreservedat the time of its discovery,red for the eyes and blue for the tightcurlsof hair.34 Moreinterestingstill is the seriesof poroscorniceblocksbuiltinto the west wallof If a buildingis deliberatelydestroyed,its cornice the granariesin the late 5th century.35 31Herodotos,

ix.65.2.

32Pausanias,i.1.5; and vII.5.5,x.35.2. 33SeeA. K. Orlandosin Noack, Eleusis,pp. 66-67; Mylonas, Eleusis,p. 80. 34Philios,HpaKTLKa, 1883, p. 63; Mylonas, Eleusis, p. 81. Orlandossaw the traces of blue as green, Noack, Eleusis,p. 64. 35Fordetailed discussion and interpretationof these cornice blocks, see Kourouniotes-Travlos,pp. 75-82. One block in the series was found in "Temple F", Noack, Eleusis, pp. 86-87. Another was built

THEDEMOLISHED ATELEUSIS TEMPLE

137

mightbe expectedto show clearersigns of damagethan some other partsof the superstructure.If its roof collapsesin a fire, we should expect many corniceblocks to be dislodgedfrom their positionshigh on the buildingand to shatterin their fall to the ground.But the evidenceof the preservedpiecesis precisely' the reverse.Almostall are completeblocks, some carvedwith one mutuleand one via and others with two mutules and two vias. The edges of the mutulesare crispand sharp;on some the guttae have scarcelyeven been chipped.The moldingson the faces of the blockshave naturallynot faredquiteso well as the more protectedundersides,but on manythe geison dripstill comes nearlyto a pointand the coronaretainsits smoothface undamaged(P1. 19:b).Patchesof ancientstuccostill cling to the less exposedsurfaces,the undercutof the geison drip, and the cornersof the vias. Some of these bits of stucco still show tracesof color, but much has been lost becauseof theirexposureto the weathersince the time of their discovery.There are no visible signs of damageby fire.36The comments of the excavatorsare expliciton the excellent conditionof the blocksand are worthquotingat some length.37"Thearchitectural membersused in the constructionof the littlewall, at any ratethe greaterpartof them, have not sufferedgreatdamage;and on some of them thereare quitewell preservedeven the very sensitive,delicateedges, on whichdependspartof the architectural functionof the poroscornices.On almostall the originalcolorwas preservedin lively tones at the time of theirdiscovery.This, too, is evidencethat these blocksdid not suffervery badlythroughcarelesshandlingor bad keepingduringthe intervalbetween their removalfrom the buildingto which they originallybelongedand theirconstructionin the little wall here in question."Inevitably it followslikewisethat they cannothave been heavilydamagedwhile in their original positionson the Archaictemple. All the availableevidence appearsto suggest that the building,whose wooden timberswere recordedin the inventories,and whose porosand marbleblockswe possess today, cannotin fact have been the same temple which Herodotossays that the Persiansburned.The carefuland detailedlists in the inventoriesof 408/7 and 407/6 reflectconsiderableadministrativelaborsuch as is not likely to have been expended upon ruinousdebris.Furthermore,the conditionof the building'ssurvivingmembers indicatesthat it did not meet its end in conflagration. On the contrary,the phraseology into the later pier, which was constructedin the undergroundgranariesbeside the northernmostpier in the east row and was removed by Kourouniotesin 1923, op. cit., p. 79, and note 1; fig. 15. The location of the pier is shown as a dotted rectangleon the plan, Noack, Eleusis, p. 190, fig. 76. Still another of the Archaic cornice blocks is still built into the west wall of the Roman granaries. 36Theblock of the interior epicranitis,now re-used as a backer in the post-Persianrepairof the peribolos wall, was described by Noack (Eleusis, p. 54) as heavily burned. This, like the cornice blocks, it should be noted, is a complete block; and while the upper curve of its hawksbeakis largely broken away, its other surfaces are not badly damaged.The slightly friable condition of the surface may well be due to weathering;and there are today no obvious traces of calcinationor discolorationwhich need be attributed to the action of fire. Noack (Eleusis, p. 93) likewise reported signs of what he took to be burning on the marbleroof tiles, but he admittedthat the roof did not appearto have fallen. 37Kourouniotes-Travlos, pp. 78-79.

138

T. LESLIESHEAR,JR.

of the inscriptionsis uniformand explicit.All the buildingmaterialssalvagedfrom the Archaictempleare describedas &ro'To VEO KaOELpELEva or aTro TEs -r0oisKaOEpEgEK To VEO To apXawo.38 The naturalinferencefrom these expresvot or KaOEpEgEvov sions is that the blockswere actuallytakendown from their positionson the building and committedto the chargeof the overseersfor storagein the sanctuary.The preservedinventoriesof 408/7 and407/6 presumablylistedonly those materialswhichhad not yet been re-usedin other construction.The simplestexplanationof the building's fate is systematicdemolition:it was deliberatelydismantledto make way for another building.Onlythis solutionwill accountfor the survivalof its woodentimbersand the care of the overseersin recordingthe salvagedmaterialin exact detail;only this can have promptedthe descriptionof this materialas Ka6EapEgJva in all the epigraphical the excellentconditionof the survivdocuments;and only this can explainsatisfactorily blocks. ing architectural It remainsto discover,if possible,at what time the old ArchaicTelesterionwas torn downand its successorbegun.The circumstancessurroundingthe constructionof the buildingwhich is generallycalled "Kimonian"may providea clue to the correct chronologyof that event. The Archaichall of initiation,of whichthe survivingblocks have been consideredabove,was succeededon the same site, at a date yet to be determined in the early 5th century,by a far more commodioushall apparentlydesigned doublethe numberof initiateswithinthe expresslyfor the purposeof accommodating The fixed featureof the site whichdeterminedthe planof the wholebuilding temple.39 was evidentlythe little chamberof the Anaktoron;for Travloshas demonstratedconvincinglythat this structureremainedthe same and inviolablethrougheach successive of the TelesterionfromearlyArchaicto late Romantimes.40The narrow reconstruction walledsanctuary,built aroundan uncutoutcroppingof naturalbedrock(P1.19:c,foreground),protecteda spot hallowedby the long traditionsof the cult and was plainlyof centralimportanceto the performanceof the Mysteries.The Anaktoronhad occupied the westerncornerof the Archaictemple,and althoughin its successorit was to retain preciselythe same physicalpositionon the site, the new hall was to have been laid out aboutthe Anaktoronso that it was centrallydisposedon the long southwestwall. The new Telesterionwas thus designedto have nearlytwicethe lengthof the Archaicbuilding, althoughthe width remainedexactly the same. The plan called for 21 interior columns,largerin diameterandmorewidelyspacedthanbefore,arrangedin threerows of seven columnseach. Also like its predecessor,the new buildingwas intendedto have tiersof steps, now seven in number,rangedaroundat least threesides of the hall. The frontwallwas pushedforwardto standon the stylobateof the earlierDoricportico, thus enlargingthe interiorof the hallitselfat the expenseof its exteriorfacade. 38GI12, 313, lines 103, 108-109 = 314, lines 113, 118-119; IG 12, 81, lines 7-8. 39Thearchitecturalremains are discussed in detail by Noack (Eleusis, pp. 93-106); Mylonas, Eleusis, pp. 111-113. For convenient reproductionof the restored plan, see Lawrenceand Gruben, locc. citt. (footnote 9 above). pp. 1-16. 40Travlos,"'Ava&K-opOv,"

THEDEMOLISHED TEMPLE ATELEUSIS

139

Evidencefor the reconstructionof this plan comes chieflyfrom the beddingsand foundationspreparedfor the interiorcolumns, and from the rock-cutsteps along the northeastwall. The smooth, rock-cutfloor in the north cornerof the existingTelesterionpreservescircularbeddingsfor three rows of three columnsof smallerdiameter and spacedslightlycloser togetherthan the later Classicaland Roman columns (P1. 18:a,foreground).These three rows of beddingsalign with three rows of foundations furtherto the southeast,where individualpiers for each column were set down on bedrock,deep beneaththe floor of the Archaictemple.41The patternof the interior columnscan thus be reconstructedwith accuracy.The existenceof the beddingsin the floorof the presentbuildingshows that masses of the living, blackrock of the hillside must have been quarriedawayat this time in orderto increasethe floor area of the initiationhall towardthe northwest.The same thing is suggestedby the new foundations belowthe floorof the Archaicbuilding,for these are composedof unworkedand irregularslabs of Eleusinianlimestone,which still show tracesof the wooden wedges used to prythem out of the hill.42It is a naturalinferencethatthey werequarriedin the northwestextensionof the building.Along the northeastwallof the Archaictemplethe nine originalrock-cutsteps were at this time reworkedso as to become seven wider steps in the new building.In the courseof quarryingoperationsfor the northwestextensionof the Telesterion,seven steps of similardimensionswere cut alongthe continuation of this wall; and these are visible today beneaththe later marblesteps of the Classicalbuildingjust wherethe northernmostportalpassesinto the initiationhall (P1. 19:d).43

This phasein the historyof the temple,the so-called"Kimonian"Telesterion,has been almostuniversallyattributedto a reconstructionof the buildingundertakenafter the Persianinvasion,which has been seen as the cause of its predecessor'sdemise. Now, its scantremainsbearwitnessto the new building'smost strikingfeature,that is the extraordinarily early stage at which the projectwas abandoned.It is obvious that partsof the buildingnevergot abovethe ground.Hereit is importantto emphasizethat no beddingwas ever preparedfor the foundationsof the southwestwall beyond the cornerof the Anaktoron.The rock-cuttrenchwhichservedas beddingfor the western cornerof the Peisistratidtempleand of the Anaktoroncan be seen in Plate 19:c (lower left). Clearly,however, no westwardextensionof the cuttingto carrythe wall of the later structurewas ever begun;and there is no indicationthat any remodelingof the Anaktoronhad taken place. As we have seen, the blacklimestoneof the Eleusinian acropoliswas cut downsufficientlyto carvethe stepsalongthe northeasternwall,where they are preservedunderthe marblesteps of the ClassicalTelesterion.But Noack44has 4"Theevidence for the interior columns is discussed by Noack (Eleusis, pp. 94, 99-102) and appears on the actual-stateplan, pl. 3. 42Cf.Mylonas, Eleusis,p. 112. 43Forthe rock-cutsteps, see Noack, Eleusis,pp. 94-98, figs. 44, 46. 44Loc. cit. Those parts of the building for which there is evidence that work was carried out are illustratedon the plan, ibid., p. 100, fig. 47.

140

T. LESLIE SHEAR,JR.

shownthat even the cuttingof these stepswas not fully carriedout at the northcorner of the building.As noted above, the spacingof the interiorcolumnswas determined, and rock-cutbeddingsand limestonefoundationswere preparedto carrythem. Not a has been identified;and in view of the incomplete singlefragmentof the superstructure state even of the foundations,it is hardto imaginethat manyblockswere ever fully workedand set in place.It should be equallyobvious that the doors, roof tiles, and woodentimbersof the inscriptionscannothave belongedto this phaseof the temple. The earlyand suddenabandonmentof this abortiveprojectto enlargethe Archaic Telesterionis strikinglyreminiscentof the fate of the OlderParthenonon the Athenian Acropolis.While this is not the place to indulgein furtherdiscussionof the controversiessurrounding the chronologyof thatcelebratedmonument,its conspicuousabanat an donment early stage of constructionis still most easily explainedby Xerxes' destructionof the Acropolisin 480 B.C. The evidenceat Eleusisnow suggestsa precisely analogouschainof events. If the constructionof the enlargedTelesterionwas cut short by the Persiansack, at that momentwhen the earlierArchaicbuildinghad been largely dismantledand the new buildingwas just beginningto be built, we can then more easilyunderstandboth the survivalinto the late 5th centuryof blocksand beamsfrom the Archaictempleand the unfinishedstate in whichits successorwas left abandoned. All the evidencehere reviewedsuggeststhe followingreconstructionof events in the sanctuaryat Eleusis.At some time duringthe decade490-480 B.C., the decisionwas made to replacethe old ArchaicTelesterionwith a buildingof twice its capacity.The Archaictemplewas carefullyand systematicallytaken apart,and the quarryingof the of foundationsfor its successorwas just begun when the Persian hill and preparation invasionof 480 B.C. causedthe workto stop abruptly.It is possiblyfor this reasonthat Herodotos45 explicitlyemphasizedthe Persiandestructionof the Anaktoron,the inner sanctumof the temple.This smallchamberwas very likelythe only partof the building standingabovegroundwhen the PersianssackedEleusis.In any event, the destruction of Eleusis had the same effect as the destructionof the AthenianAcropolis;for Demeter'stemple,like that of her sisterAthena,was allowedto lie in ruinsfor a generawasagainundertakenin the Pericleanperiod. tion beforeits reconstruction T. LESLIE SHEAR, UNIVERSITY PRINCETON

Departmentof Art and Archaeology Princeton,NJ 08544 45Herodotos,ix.65.2, footnote 17 above.

JR.

PLATE 18

~~~~~~~~~~~~00

ZY

~~~~~,~-

-Z-i~

~

Al

e

*~~~~~~

~

A

Cl~~~~~~~~~~~~~~~~R

-;V

-.

~

'

*er

4

N1,~~~~0 W

v<

-

~

\:

Cf

jU,

0t

- ~~~~C ~~ ~ I~~j ~~).'k:4%JtylV'

.

~

V-

I

PLATE 19

44i11

3

4,

W

"\

I

S

I V

Al

. :. TV, :.:

. a*..

X

*

S

~v

~~~~~~~ E

C

A j

,

s

j C i

'

M

'A~~~~~~~~~~~~~~C 1'~~~~~~~~~~~~~~~~~~I

IsL ,

'0

7E

4

.0

jg~~~~~~~

0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~FCZ

PLATE 20

'c~~~~~~~~~~~~~~

H

TIC }

,H'' 0

L

=

a;."~~~~~~~~~~~~~~~~~~~~~~~

_ 'S~~~~~~~~~~~~

_C

#tf

..,E~~~~s, a.

E,_

=

S

5

E

^

E

_.~~~~~~~~~~

l.

t

i

t )~ C~~~ofi

it

~~~~~~~~~~~~~

.D

=~~~~~I

;S s ads1 | - -----Is.,, S

M:~~~~~~

*

<

*'v_

5-

_

I

~~~~~~* 1

|

0

V

V

c

|

,0~~~~~~~

Ev~

THE PREHISTORICKLEPSYDRA:SOMENOTES (PLATES

21 AND 22)

of the Klepsydraandthe areaimmediatelyaroundit and the EEPEXPLORATION Court on the northwestslope of the Acropoliswas undertaken Paved adjoining initiallyby A. W. Parsonsas an adjunctto the campaignsof 1937-1940in the Athenian Agora.1These excavationsled to his enduringstudy in Hesperia12, 1943, which, conspring,mentionedonly in centratingon the Classicalandlaterhistoryof thatremarkable passingindicationsof earlierprehistoricinstallationstappingthe same source.2These earlierremainswere examinedand a selectionof finds from them publishedby Sara Immerwahrin her comprehensivestudy of the BronzeAge materialfrom the Agora Excavations.3Conspicuousamong them are Late Mycenaeandeposits from cuttings beneaththe ClassicalPavedCourt,whichshe attributedto "the very end of the MyceWellU 26:4,"4also naeanperiod,"andwhichformed"a transitionto the Submycenaean clearedby Parsonsto the east of the Court.It was the need to definemore preciselyits depositthatpromptedmy closerstudyof the Klepsydra relationto this "Submycenaean" have emerged,but the evidence, despitea wealthof new material.5Severalproblems findselsewhereand some additionalcleaningaroundthe cuttingsthemselves,is still too limitedfor definitiveanswers.I summarizethree here in the spiritof puzzles, in the hope thatothersmayfindthem intriguingandjoin in theirsolution,in the mannerthat HomerThompsoninvites others to wrestleProteanproblemsand hold them to question, if not alwaysto rest. These concern(1) the physicalformof the cuttingsand charsetting,and (3) the date of the latest acterof theirfilling,(2) theiruse andarchitectural of the Acropolisnorthwestslopes reoccupation regularactivityaroundthem priorto the in Archaictimes. To placethese questionsin their setting, it will be necessaryto look backagainat the historyof the PavedCourtandthe prehistoriccuttingsbeneathit.6 D

'Discussions with friends and colleagues have helped this paper.I am particularlygrateful to Dr. John McK. Camp II, Mr. William B. Dinsmoor, Jr., Professor Sara A. Immerwahr,and Dr. John Travlos. The photographsare the work of M. Alison Frantz (P1.21:a), Eugene Vanderpool,Jr. (P1.21:b), and Robert K. Vincent, Jr. (the rest); my deepest thanks to them all. 2A. W. Parsons, "Klepsydraand the Paved Court of the Pythion," Hesperia12, 1943, pp. 191-267. 3S. A. Immerwahr, TheAthenianAgora, XIII, TheNeolithicand BronzeAges (= AgoraXIII), Princeton 1971. 4lbid., p. 261: deposit T 26-27:2. 50n U 26:4 and the Klepsydrafinds, E. L. Smithson, "'Submycenaean'and LH IIIC Domestic Deposits in Athens," AJA 81, 1977, pp. 78-79; for reasons noted there, I found the term "Submycenaean" confusing and of no practicalapplicationto domestic deposits. J. B. Rutter ("A Plea for the Abandonment of the Term 'Submycenaean'," TempleUniversityAegeanSymposium3, Philadelphia1978, pp. 58-65) asks that it be droppedaltogether, a tidier and more rigorous proposalthat I would endorse; see fig. 1 for his dating of these and related deposits. 6Whatfollows is a summarythat of necessity skirts many difficultiesand omits much supportingmaterial; a full presentationof the finds from the Klepsydra,U 26:4, and related deposits from the Agora Excavationsis in preparation.

142

EVELYN LORD SMITHSON

0~~~~~~~~~~~~~~~~~~~~~~

0O

U1

V)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ IL~~~~~~~~~~~~~~~~~~~

lox)~~~~~~~~~~~~~~~~~~~~~,

01

THE PREHISTORICKLEPSYDRA:SOMENOTES

143

1. Thephysicalform of thecuttingsandthenatureof theirfilling Parsons'studypresentsthe mainoutlinesof the excavation,the form of the Paved Court,and its dating.The Court,whateverits functionalrelationto the SpringHouse, was builtat the same time with it in the secondquarterof the 5th centuryB.C. A drainage channel (Fig. 1:A) that carriedoverflowout beneaththe northwall of the Paved Court(Fig. 1:N) was cut at the same time or not long afterward.7 Followingthe collapse of a largepiece of Acropolisrock onto the centerof the Court,extensiverepairswere carriedout in the mid-Ist centuryafterChrist,includingthe additionof a north-south cross wall (Fig. 1:C) that bisectedthe Court, the filling in of the Classicaldrainage channel,and the openingof a new one (Fig. 1:B)a little to the west of it. Finally,the Wallin its coursedownfrom the Acropoliscrossedthe westLate RomanFortification the Court ern partof and the line of the drainagechannels.All these intrusionsleft their markon the earlydepositsin question.Parsonswas able to examinethe fill beneaththe Courtonly in a limitedareawherethe pavingslabs,eachweighingsomething over two thirdsof a ton, had been removedor displacedin the past, and hazardous diggingconditionssoon broughtexcavationto a halt. Three decades later, in 1969, while cleaningand repairingthe Paved Court, the Greek ArchaeologicalServicewas able to clarifyand extendthe areaprobedby Parsons.8A new planand section,madeat that time by Dr. John Travlos,are reproducedhere, throughhis kind generosity(Fig. 1). But in all, the area investigatedbeneath the Court covered a maximumarea of aboutseven meterseast-westby ca. 4.50 m. north-south,and only a small portionof thatcouldbe takento bedrock.Physicalimpedimentsvirtuallyprecludefurtherexploration,9so thatthe evidencefor the prehistoricinstallationis in hand,howeverfrustrating it mayseem. Two pits were found in bedrockbeneaththe Court, the precisephysicalrelation between them obscuredbeneaththe EarlyRomancross wall. The easternpit ("East Cutting",Fig. 1:E), with its neatlyparallelsides, is surelymanmade,the outlineof the northernpartnearlysquare,abouttwo metersto the side, withslightlyroundedcorners; the floor,about2.0 m. belowthe surfaceof bedrockat the north,slopedslightlyto the east. On its bottomlay at leasta halfmeterof undisturbedLateMycenaeanfill, a greenish graymud, diminishingtowardthe northandwest but risingnearlyto the full height of the cuttingat the southeastcorner,suggestingthat it had been dumpedor sweptin from that point.Above this was anotherlayer,sandyand gravelly,of undisturbedLate Mycenaeanfill, and on that a shallowfill, dumpedin to bed the Romancross wall, containeda sprinklingof Late Mycenaeansherds like those in the lower filling and presumablychurnedup from it. 0.50 m. above the floor, the south face of the cutting was steppedback, and on this "step",0.65 m. backfrom and parallelto its northern 7See p. 149 and footnote 35. 8G. Dontas, AEXr 25, 1970, B' [1972], pp. 26-28, pl. 39; K. Demacopoulou, "MvKMvauKd KEfaX7J," AEXr25, 1970, A' [1971], pp. 174-183. 9Dontas, op. cit., p. 28.

irTrMX&'v-

144

EVELYN LORDSMITHSON

edge, was a Late Mycenaeanwall runningeast-west(Fig. 1:M); it is made of neatly fitted,largeand smallstones, carefullyfinishedon its northernface, as if to be visible, Th. 1.20, p.H. 0.60, p.L. 1.55 m.10Above and behind the wall was compresseddug bedrockthat increasedin thicknesstowardthe south;in it wereLate Mycenaeansherds and pieces mendingto two splendidwide-mouthedpithoi.1"The cuttingwas traced, then, somethingover four metersto the south beforeexcavationhad to be abandoned beneathundisturbedpavingslabs.Any extensionof the wall outsidethe cuttingis unknown,andits functionalrelationship,if any, to the squarebasinis one of our problems. of the westernpit ("WestCutting",Fig. 1:W)is far from clear. The configuration The areaprobed,againrestrictedby intactpavingslabs, lay betweenthe northwall of the Court,the Romandrainagechannel,and the bouldercollapsedfrom the Acropolis cliffsaroundwhichit was diverted.'2This pit is considerablydeeperthan the East Cutting. Bedrockor compresseddug bedrocklay 3.40 m. below the floor of the Paved Court.Beneaththe crosswall,diggingwas suspendedat 3.50 m. among"boulders"and water;a crowbar,thrustdowninto the fill another1.50 mi., did not strikebottom.It is likelythat we are dealingwith a deep fissure,perhapsa continuationof the same natural reservoiragainstwhichthe ClassicalKlepsydrawas set, seven metersfurtherwest.'3 The surfaceoutlines of the western pit may have been regularizedsomewhat,as a roundedcorner(northeast?)found beneaththe cross wall suggests.Somethingover a meter of pure Late Mycenaeanfill was excavated,restingon bedrockor compressed dug bedrockto the west (cf. East Cutting,south of the Mycenaeanwall), and apparently still goingdownin placesunderthe crosswall.As in the EastCutting,a few Late Mycenaeansherds,displacedfromthese deposits,werefoundin the dumpedRomanfill abovethem. Apartfrom three small pieces in the West Cutting,two probablyof tiles and one debrisor quantitiesof claypackingof some sort, the depositswere free of architectural carbonizedmatter that might suggest destroyedor dilapidatedbuildingsnear by. A single piece of "miscellaneousclay", also from the West Cutting,most clearlyresembles in form and size the "enigmaticunbakedclay spools"from a house of developed LH IIICdate at Lefkandi.'4But grindersand pestles, among the most common and '0Ibid., pl. 39:y. 11Ibid., pl. 39:a, H. 0.82, D. mouth 0.62; pl. 39:,8 H. 0.98, D. mouth 0.70, D. base ca. 0.20 m. I thank Dr. Dontas for his kind permission to measure these, and to examine sherds recovered in 1969. They include nothing inconsistent with Parsons' finds and have not been included in any statistics. Parsons reached the step, and probablythe wall ("pieces of schist"), and noted "many pieces of pithos" in dug bedrock, some of which might fill gaps in the 1969jars; we did not try for joins. 12" ... boulder ... sinking ... beneath the ground level," Parsons, op. cit. (footnote 2 above), pp. 243-245, esp. p. 244. 13Jbid.,p. 212, where the Classicaldrawbasin is likened to a "bait-box"set into a naturalreservoir. Report,London 1968, p. 13, fig. 16, lower row, 14M.Popham and L. H. Sackett, Letkandi,Preliminary found with potteryof Phase lb (M. Pophamand E. Milburn, "The Late HelladicIIIC Pottery of Xeropolis (Lefkandi), A Summary," BSA 66, 1971, pp. 334-336). The Klepsydraobject is baked, however lightly, and has a stone, about half its thickness, encased at one end; L. 0.068, D. 0.035 m.

THEPREHISTORIC KLEPSYDRA: SOMENOTES

145

durablediscardsin fillingsdrawnfrom generalhouseholdrefuse, are lacking,and the assortmentof potteryhere is quite unsuitable.Finally,and most interestingof all the smallfinds, is the fragmentof a fine female head, found in the West Cuttingin 1969, 40 centimetersin height.15Otherwise,all finds suited in scale to a figurineapproaching from the two pits were pottery,so consistentin chronologicalrange,condition,fabrics, and shapes,as to leave little doubtthat the two, if not actuallypartof the same establishment,werefilledin andcoveredby materialdrawnfroma commonsource. Sherdswere few for the quantityof earthremoved:1100 from the East Cutting,'6 mostlysmalland few obviouslyfrom the same pot; 535 from the West Cutting,some fairlylargeor mendingto sizeablefragments,thoughno pot was anywherenear complete.17The assortmentis quite unlike depositsof ordinaryhouseholdrefuse," cemeteries,19or sanctuaries,20 all of which, beyondcertainfurnishingsfavoredby each, are characterizedby a plentifulnumberof open vessels, large cratersas well as smaller 15K.Demacopoulou (op. cit. [footnote 8 above], p. 182) considered the possibilitythat it came from a figure like those from the Citadel House at Mycenae but ultimately favored a sphinx. More recent finds show such large figurines to have been fairly common in late sanctuarycontexts; particularlystriking are the numbers associatedwith the bench sanctuariesagainst the circuit wall of the Lower Citadel at Tiryns (AA [JdI 93], 1978, pp. 461-464; AA [JdI 94], 1979, pp. 389-391). Such an identificationseems preferable. Her suggestion that the head and the filling in which it was found had been brought in from Mycenaean installationsfurther east along the slope is not supportedby the distributionof shapes in this filling (see next paragraph). 16Seeabove, footnote 11. 17Asingle lot of Klepsydrapotterycould not be found in 1957, when Parsons' pottery was transferred to permanentstorage. MargaretCrosby, who undertook that Herculeantask, noted "the barest possibility" that another lot of 330 sherds, ticketless, came from the cuttings, and saved it in the hope that positive evidence might emerge later; none has, but I think that provenience likely. Its characterand composition are like the other lots, though in the distributionof open shapes (17% of the total), painted kylikes and conical cups (FS 242) are more numerous.This lot has not been includedin any figures here. 18Ase.g. E. French, on early LH IIIC deposits at Tarsus ("A Reassessment of the MycenaeanPottery at Tarsus," AnatSt25, 1975, p. 56): "small proportionof closed vessels is typicalof settlement pottery and can be compared with the proportionsat Mycenae itself"; at that site 20% of closed vessels is high. At Korakou, see J. B. Rutter, The Late HelladicIIIB and IIIC Periodsat Korakouand Gonia in the Corinthia (University Microfilms International), 1974, e.g. East Alley Deposit, p. 97. The same is true of the dumped filling in the MycenaeanFountain at Athens, whatever its source; though figures are not recoverable, Broneerclearlystates ("A MycenaeanFountain on the Acropolis," Hesperia8, 1939) that among the fragmentarypottery in the dumped filling, "most common ... among larger vases is the large bowl or krater (p. 351),

...

of all shapes ...

is the skyphos (p. 362),

...

among the closed vases is the stirrup

vase" (p. 389). 19E.g.S. Jakovides, HEpar4, Athens 1969-1970; W. Kraiker and K. Kiibler, KerameikosI, Berlin 1939, pp. 1-88 (except Grave S 112), and note the "Submycenaean"pieces from the Grave Mound, pls. 38, 39, 48, 50. 20Tothe long-familiarexamples from Knossos (A. Evans, Palaceof Minosat Knossos II, London 1928, pp. 123-138, UndergroundSpring-Chamber;pp. 335-343, Shrine of the Double Axes) and Asine, House G (0. Frodin and A. Persson, Asine, Stockholm 1938, p. 298), may now be added the shrines on the Lower Citadel at Tiryns (AA [JdI 93], 1978, pp. 460-465, 495-497; AA [JdI 94], 1979, pp. 379-394). Preliminarynotices of the finds from the shrines at Mycenae (Antiquity44, 1970, pp. 270-280; AAA 3, 1970, pp. 72-80), on Kea (Hesperia41, 1972, pp. 400-401), and on Melos (JHS-ARfor 1977-78, No. 24, pp. 52-54; Antiquity52, 1978, pp. 7-15) also indicatean assortmentquite differentfrom that found here.

146

LORDSMITHSON EVELYN

drinkingvessels, a wide rangein size with the smallerin scale well represented,and a or an equivalent,lackinghere.In each componentof certainshapes,suchas stirrup-jugs to drawingand, especially,transsuited of the cuttings,pieces from largeclosed pots portingwatermade up nearlytwo thirds (63%E, 62%W) of the pottery;the largest componentin each cuttingappearsto be hydriai.Openshapes,almostentirelycups of some kind, amountedto less than a fifth (17%E, 14%W); cookingand light coarse ware, mostlysuitedto thoughnot designedfor waterwork,abouta tenth (10%E, 8% W). The slightdiscrepancyin pithosor vat fragments(5%E, 11%W) will have been Servicein the morethanequalizedby the pithoirecoveredby the GreekArchaeological East Cutting.Apartfrom the size of the fragments,the only other difference,possibly significant,is a slightlygreaternumberof pieces from small closed vases in the East Cutting,which, if not incidentalgarbage,mightsuggesta slightlydifferentorientation in waterinterests.Amongthese are an intactminiaturehydria(P1.21:a,b)21and a large fragmentof an amphoriskos(Fig. 6),22 both totally unsuitedto water collectionor transport.Thatboth were foundin the areaof the step or "bench"in frontof the Late MycenaeanWall, the hydriaat most ten centimetersabove, if not actuallyon it, may also proveof interest. With the The date of the finds in the cuttingshas been summarizedelsewhere.23 exceptionof a handfulof earlierpieces, all of the materialbelongsto LH IIIC, with some of the formsbeginningearlyin that periodbut of uncertainduration.The bulk, however,consists of AdvancedIIIC types, but nothingclearlyfrom the end of that period.J. B. Rutterhas assignedthe Klepsydramaterialto his "Phase4" of IIIC,leaving open the possibilitythat it may continue into "Phase5", his ultimatestage, in whichhe placesU 26:4.This seems acceptable.Exceptfor the presenceof a few onehandledconicalcups (FS 242), Klepsydramightequallysatisfyhis Phase3, thoughthe longevityhere of "Phase3-features" has yet to be established.The absenceof large closed depositsin Athens precludesvery detailedlocal refinementsfor the present. Phase 1 is surelyrepresentedin the North SlopeHouses, as he notes, but Phase2, as distinctfrom 1, and Phase 3, as distinctfrom 4, awaitverification.But that is another problem:"Phase4" or "Advanced,but not ultimate"LH IIICis sufficientfor the filling in the cuttings. setting 2. The use of the cuttingsand theirarchitectural The most obviousfunctionfor the East Cuttingis a broadopen basinfor drawing water, filled by seepage neatly channeledinto it along the shallowextension of the cuttingto the-south. Abundantwatercollectedafterexcavation,and judgingfrom the 16758, AgoraXIII, p. 262, no. 482, pl. 64 (= P1.21:a;cf. P1.21:b), p. 125; H. 0.099 m., D. 0.085 m. 22LotOA 327-1; H. est. ca. 0.105 m., D. ca. 0.10 m. Thick wavy line, cf. collar-amphoriskos,A. J. B. Wace, ChamberTombsat Mycenae (Archaeologia82, 1932), pl. XI, 11, Tomb 502, last burial, there as more commonly with bands edging the panel. Shape, cf. KerameikosI, pl. 17, inv. no. 424, Gr. S 2, with wide mouth, depressedshoulder, and greatestdiameterlow set. 23Smithson,op. cit. (footnote 5 above), pp. 78-79; Rutter, op. cit. (footnote 5 above), p. 65, fig. 1. 21p

KLEPSYDRA: SOMENOTES THEPREHISTORIC

147

lie of the fill, at times actuallyrushed in from the southeastcorner.The wall conthough structionis not so tightas to have impededflow undernormalcircumstances,24 the line may have becomecloggedat some stage duringuse or after the abandonment "periodof the installation,divertingthe flow to the corner.There is no characteristic of broken or lost water of in i.e. the jars, the sort of-use filling" basin, largefragments familiarfrom laterwells. But with a maximumdepth of 0.50 m., the basin is shallow enoughthatlargerfragmentsmighthave been retrievedas they fell in or piledup so as to obstructbailing.It was a practiceof the Mycenaeansgenerally,unliketheirDarkAge and Classicalsuccessors,to keep their wells clean, whetherthroughperiodicclean-out Also pointingto such bucketsis the fact that or a preferencefor non-ceramicbuckets.25 most of the sherdscome from largeto very largewaterpots, mainlyin the rangeof H. ca. 0.40 m., a size moresuitedto transportthanbuckets,whichin wells of laterperiods are in the medium-sizedrange, H. ca. 0.25-0.30 m. (as the standardProtogeometric oinochoe, trefoil lippedto facilitatethe transferof liquidsto a transportcontainer). Pieces in the basin were mostly too small to repayremoval;other bits, fallen to the groundoutside the basin, may have washedin after its partialor completeabandonment. The size of the fragments,then, is not againstuse as a basin, and the kind of potteryis very muchin its favor.26 Broaddrawbasinsof this sort are unusualin Mycenaeantimes. Wells, like U 26:4 to the east, are the commonsource,the smallermouth (D. ca. 1.0 M.)27 insuringcool water, more easily protectedagainstcontaminantsand accidentsinvolving animals, children,or the infirm.Basinsbelongmoreproperlywith piped-inwater,28hardlynecessary in this location.A comparableinstallation,furtherafieldand somewhatearlierin adjoiningthe Caravanserai its initialconstruction,is the UndergroundSpring-Chamber at Knossos.29It is smaller, ca. 1.15 x 0.75 in., but the depth, 0.45 m., is about the same as the East basin.Like the basin,it was fed by seepage,there oozing up through the pebbledfloor. That in its last phaseat least, choked with debris, the basinwas a 24Butsee J. C. Wright, "MycenaeanPalatialTerraces," AthMitt95, 1980, p. 81; "weepers" may have been advisable. 25Woodenor metal. On lead buckets, see AgoraXIII, p. 259, no. 471. 26Thegreat number of largejars noted at Lefkandiwas attributedby the excavators to the need for storage on a waterlesshill (Pophamand Sackett, op. cit. [footnote 14 above], p. 17). Though storage could be a factor here, I think transportmore likely, especiallysince the smallness of the pieces implies that most jars were taken elsewhere. 27A number of Mycenaeanwells, unlike their successors, have shafts that are square or rectangularin section: Agora S 13:2 (1.0 x 0.65 m.; LH IIIA 1), S 27:7 (1.10-1.15 m. to the side, Agora XIII, pp. 259261; EarlyLH IIIC), U 26:4 (Depth, 14.0 m., the lower two thirdsroughly square, 0.90-1.0 m. to the side; Late LH IIIC). Cf. the Rhyton Well at Mycenae, with square coping, ca. 1.0 m. to the side, set againsta circularshaft (BSA 24, 1919/20-1920/21, p. 200). 28As,e.g., the aqueductand plasteredreservoirat Pylos, Area 102, only slightly largerthan this basin (C. W. Blegen and M. Rawson, Palace of Nestorat Pylos I, Princeton 1966, pp. 329-336); perhapscoincidentally the floor of the East Cutting also has a slight slope. Cf. the "foot-bath" of the Caravanseraiat Knossos (1.52 x 1.38 x D. 0.45 m.), Evans, op. cit. (footnote 20 above), pp. 116-121. 29Evans,op. cit., pp. 123-138; surfacedimensions scaled from fig. 60, p. 125.

148

LORDSMITHSON EVELYN

shrine invites furtherspeculation,for one would like somehow to accommodateour "votive"hydria,and perhapsthe terracottahead. The "bench",too, calls to mind a but since the bench shrine, and the small size of the "room"would have parallels,30 and the basinare mutuallyexclusiveone wouldhave to choose betweenthem. But the mere fact of these features-votive hydria,terracottahead from a largefigurine,and a benchor basin-does not makea shrine,and there are no otherfigurinesor sanctuary Whatis undeniable,I think, is that the equipmentto reinforcesuch an identification. we are to associatethe fragmentswith unless basin, a water as used was East Cutting activitysolely aroundthe West Cutting,a possibility.In any case, this is no ordinary Mycenaeanwaterinstallation,and as one thinksof the laterPavedCourtand ponders its purpose,one wondersif some primordialholinesspromptedits monumentalsetting. No continuityis attestedby the survivingremainson the slope, but rediscoveryby chance,in the Archaicperiodif not before,is not ruledout. I wonder,then, if the West Cuttingwere not the businessend of the establishment,and that the East Cuttingwas not somethingspecial,the preserveof water nymphswho guaranteedthe continued of the springs. productivity So little is knownof the Late Mycenaeanwall that, beyonda few bare facts, the field is free for speculation.The sherdsfrom it are of the same date as those in the cuttings.It runsparallelto the cliffsbelowthe Sanctuaryof Apollo,at a distanceof only five meters from their sheer face. It is sturdybut hardlycomparablein scale to the Cyclopeanblocks of the Acropoliscircuit;it cannot be pressed,alone, to fortifythe northwestcornerof the citadel.31The unfinishedinner face suggestsa terraceor retainingwall, as does the dug bedrockfillingbehind it, which would come most convenientlyfrom the constructionof the East basin, a circumstancethat would support their havingbeen partsof the same project.That the dug bedrockappearsto spread over aroundthe West Cuttingsuggeststhat the constructionextendedover to thatarea as well. The wall and terracemay well have supportedconstructionabove them, perhapsa stairwayleadingto the ApolloTerrace,and from there to the citadelabove; it storagechambersbeneathor adjoiningthe staircase,conmay also have incorporated ceivablythe sourceof the pithoibehindthe wall.The constructionand use of the Klepsydrainstallationis laterthan the MycenaeanFountainand is its logicalsuccessor.The difficultto maintain,and very restrictFountainis, at best, a cumbersomearrangement, ed anddifficultof access.The Klepsydra,as a replacement,offersan abundantsupplyof water,andwitha stairway,easy accessfor those on the citadel,as well as on the slopes, providingwaterfor as long as it seemed necessaryor advisableto live in the immediate shelterof the citadel.How long thatneed continued,we mayconsidernext. 30Tiryns,Lower Citadel, Room 110: 3.0 x 1.28 m. (AA [JdI 93], 1978, p. 463), includingthe depth of the bench at the rear;Shrine of the Double Axes at Knossos: 1.50 m. square (Evans, op. cit. [footnote 20 above], p. 336). 31Anyspeculationmust take into account the "PelargikonProblem", which goes far beyond the limits of this paper.I do believe, though, that the Klepsydraspring, along with U 26:4 twelve meters to the east, cannot have been outside it.

KLEPSYDRA: SOMENOTES THEPREHISTORIC

149

3. The latestpre-Archaicactivityaroundthe cuttings

Evidencewhichhas bearingon the latestprehistoricactivityin this areacomes not from the cuttingsthemselves but from the filling dumpedin to close the overflow channelof the ClassicalKlepsydra(Fig. 1:A). This channelappearsto beginin the West Cutting,32pass under the north wall of the Paved Court, and continuestraightsome nine meters to the northbefore it was shearedoff by late construction;W. 0.60-0.75 m., D. ca. 1.0-1.50 m. Fewerthan400 sherdswere recoveredby Parsons,a densitynot appreciablyless than in the cuttings.The channelwas closed in the mid-Ist century afterChristwith fillingdepositedin three layers.Sherdsin the lowertwo layersare in a handfulof Classical,the majoritylate Hellenisticand EarlyRosimilarproportions: the rest, a little over a quarter,were LH IIIC, man to mid-ist centuryafter Christ;33 and five except that from each layercame one piece of indisputableProtogeometric,34 or six othersthat mightbe. The proportionsin the top layerhad altered:Classicaland Hellenisticwere negligible;Romanhad droppedto 26%and includeda fair numberof mid-3rdcentury(afterChrist)sherdsthat intrudedwith the Late RomanFortification Wall; early sherds now comprised 69% of the total, again LH IIIC, but also eight that are certainly Protogeometric, and as many as 15 others that might be. Excluding the black glaze in the mud at the bottom which may come from the construction of the channel,35the filling suggests reverse stratigraphy:the heavily Roman component of the lower levels drawn from debris contemporarywith the closing, the light admixture of

early earlymaterialcomingfromdisturbedlevels thatlay beneathit. The predominantly materialin the top layerwould representdeeperpenetrationinto less disturbedearly levels (cf. upperlevels in the cuttings).A selection,from these earlypieces appearson Plates21 and 22 andFigures2-5, 7-11.36 32Parsonssuggested (op. cit. [footnote 2 above], p. 223) that the connection was with the East Cutting, near its northeast corner. Dr. Travlos reports, in conversation, that subsequent cleaning shows that the channel entered the Court beneath the cross wall, where it seems, rather, to have been linked to the fissure of the West Cutting. 33Dr. BarbaraJohnson very kindly re-examined the channel lots, OA 314-316, for their Hellenistic and Roman content in the summer of 1978. 34LotOA 316-1, P1. 21:c. Lip from a high-footed cup of standardlarge size, D. rim ca. 0.10 m. Cf. HeidelbergGrave B, G 82 e, f (KerameikosI, pl. 37), miniaturesfrom a very early Protogeometricgrave. Lot OA 315-3, P1. 21:c. From shoulder of large closed pot; D. circles, ca. 0.08 m. Three arcs from compass-drawncircles, carelessly executed, as if from inexperience. Dull dark brown glaze, streaking to thin orange;earliest Protogeometric. 35Eightblack-glazedpieces came from the mud, none giving positive support to a Kimonian date. Among them are two skyphoi, nearly complete, with closest parallelsaround 525 B.C. They are shoddy, like mass-producedsanctuarypots; their source and how they were preserved to enter the drain nearly whole are unclear.The only other datable piece is from a Pheidias mug, ribbed and stamped, that cannot be earlier than the beginningof the last quarterof the 5th century. I thank Dr. John McK. Camp II for assistancewith these pieces. 36CLOSEDSHAPES. Lips (Fig. 2), Hydriai/Neck-amphoras: OA 314-32, IIIC;the second band inside disappearsbefore EPG. The glazed neck of OA 314-13, if from a hydria, is exceptionalin either 111Cor EPG, and may be a contaminationfrom other belly-handledforms, amphoriskoior amphoras,where the neck is always glazed; it looks ahead to developed PG, when glazed necks will be mandatoryon hydriai as well,

EVELYN LORD SMITHSON

150

D.-0.Q.4........

D. c a0.I12~.

D. cao .IO...

D.co 0.1-1

OA 314 -= ~ 314-9O OA 314 OAN4-32 1-l/OA ~~OA

D.co 0.18

....

-12 I 314

OA314-1

I OA314-34

314 -13

~~~~~~~~~~~~~~~~~~~~~~~~OA

D.ca0.13

FIG.2. Lips and a base fragment

OA 314-18 7

OA 314-7

FIG.3. Lip and neck

FIG.5. Belly fragment

7

D.co 0.13-0.14_

-~~~~~~~~~-~ ,

327-1l ~ ~~~~~~~~~~OA

OA 314-14

FIG.4. Shoulderfragment

FIG.6. Amphoriskos

FIGS.2-6. Late HelladicIIIC/EarlyProtogeometricclosed shapes. Scale 1:3

THE PREHISTORICKLEPSYDRA:SOME NOTES

151

2 X Xa 0i0n D.co0.14

iccao52

\

w

~~~~~~~~~~~~OA

Oco .40.I4O6.

OA314-IA4*

D.caoIO,,----- UcO010. .

314-39

O 0314-2

Q~~~~~A

314 -1

FIG.8. Lips:skyphosand cups OA315-1

FIG.

7. Ring-stemmedkylix D.co 0

16

_

.......

.

OA 314-6

OA 314-5

FIG. 9. Lips:bowls

D.coO.047 OA 314-4

D.co0.038-004 OA314-3

FIG. 10. Bases

D.co 0.045 OA 315-2

D.co 0.055 OA 314-37

D.co0.048 OA 314-38

D.cO.Qm OA314-36-

FIG.1l-. Bases

FIGS.7-11. Late HelladicIIICopen shapes. Scale 1:3

0.4XTD08 35. C34i

152

LORDSMITHSON EVELYN

The early potteryin the channel, both LH IIIC and Protogeometric,is exactly comparableto the potteryin the cuttings:mostly small pieces but a few of fair size, from largeclosed pots, most of them hydriai.It is likely that the same predominantly sourcethatfilledthe cuttingsalso providedthe materialfor the channel.If, as Margaret Crosbysuggested,37the channel filling was actuallydrawnfrom beneath the Paved Court,it will have been displacedin beddingthe crosswall, settlingthe boulder,and in cuttingthe new drainagechannelaroundit; the few largerpiecesareconsonantwith the West Cutting.If at some latertime thoroughexplorationbeneaththe Courtshouldbe we shouldthen have to look elseundertakenand fail to produceany Protogeometric, waterjars and at least two cups were in where.But in any case, EarlyProtogeometric of the the Acropolis. on northwest slopes servicehigh The LH IIICin the channelbelongsmostlyto the Advancedphaseof the cuttings. 20%,go beyondat least one Ambiguousor certainlylaterpieces, possiblyapproaching there is nothingof a developedProtogeostylisticphase into EarlyProtogeometric;38 metricstage.39Again,we feel sorelythe lackin Athensof stratifiedlargecloseddeposits connectingthese phases,but we may question,nonetheless,whetherthe channelis, in fact, a mixed deposit composedof chance rubbishfrom several discreteperiods,or whetherit may not be made up primarilyof types in currentuse, wheneverthey may have been introducedinto the Athenianrepertory.Well depositson the slopes and in the lowerAgoraarea,now fairlynumerous,are eitherclearly"AdvancedIIIC"in date, thoughboth containsome slightlyearlierpieces, presumed or "EarlyProtogeometric", "intrusive"and to have entered as incidentalcontents of the filling earth. Grave groups,too, have tendedto reinforcethis dichotomy,being furnishedexclusivelywith I had elsewheresuspecteda gap in IIICtypes, or with those of EarlyProtogeometric. time, a missingphase that would supplya convincingmixtureof old and new, along but there are no intermediarypieces to mark this as the beginningof a trend. OA 314-7 (Fig. 3, P1. 22:c), EPG hydria;the broad band and wavy line (or "zigzagcapping"),common on PG light-groundoinochoai, may be a refinementof the wavy line at mid-neck, the only "decorativeinnovation"of Rutter's "Phase 5" (Rutter, op. cit. [footnote 5 above], p. 60). Base: OA 314-34 (Fig. 2), IIIC/EPG. Shoulders(Fig. 4, Pls. 21:d, 22:c): antitheticalpendantspirals (OA 314-30) and streamers (OA 314-31), IIIC; nedklace,OA 31428 and OA 314-29 (EPG?); OA 314-14, EPG neck-amphora(as KerameikosI, pl. 29, inv. no. 522, Gr. A); OA 314-26 (handle-ringat left) and OA 314-27, EPG. Bellies (Fig. 5, Pls. 21:d, 22:c): OA 314-18, OA 31416 and OA 314-22, IIIC;EPG, OA 314-15 and OA 314-25 (the glaze at the top of the fragmentis from the ring of a vertical handle without finial; there are no finials in this deposit). The wavy line is optional on bellies of hydriaiin IIICand EPG, the clear belly becoming mandatoryonly in developed PG. kylix:OA 315-1 (as P 17328, U 26:4). Skyphoi:OA OPENSHAPES(Figs. 7-11). All IIIC. Ring-stemmed 314-1 and OA 314-2, and another with reserved band inside the lip. Cups:OA 314-39 and OA 314-40. Bowls(?):OA 314-5 and OA 314-6. Bases: OA 314-35 through OA 314-38, OA 315-2; bases like OA 314-3 and OA 314-4, though a little low, occur in EPG deposits. EPG high-footedskyphos:OA 314-41 (PI. 21:e), D. rim est. ca. 0.23 in., light-ground,"circle-and-panel"type, as K. Kubler, KerameikosIV, Berlin 1943, pl. 23, inv. no. 2012, Gr. PG 34, a little later than this piece; the base was probablyhigher than any preserved in this deposit. 37Seeabove, footnote 17. 38As,e.g., KerameikosGraves PG A, 1, 13, 4; Agora Wells J 14:2, N 12:3. 39As,e.g., KerameikosGraves PG 15, 34; Agora Wells A 20:5, K 12:1.

THEPREHISTORIC KLEPSYDRA: SOMENOTES

153

with transitionaltypes that wouldprovidea simplelink betweenthem. But gravesand wells tend to be highlyselective, representingthe tastes of individualfamiliesand the cast-offsfrom individualworkshops.A communalsource, as we assume Klepsydrato If we assume, for the moment, that have been, need not have been so discriminating. in the channelis not intrusivein a substantiallyearlierfilling,40but the Protogeometric ratherthat it and some of the IIICpieceswere in concurrentuse, then we may suggest a furtherpossibility-thatthere is no gap, and that Protogeometriceruptedquite suddenly, in the cornerof a largelybankruptindustry.Like red figure,it may have been the inventionof one potter,one painter,who substituteda compassfor spirals,on the shapesand in the decorativesyntaxcurrentin IIIC.Such experimentsare known (P1. 22:a),41and it need not requirean intermediatechronological phaseto producethe new style. Our pictureof the last years of IIIC in Athens may not be entirelyfair, and to some extent it may be of our own creation,since we have tendedto emphasizeits poor and drabaspectsand to assignwhatis good and experimentalto a later"EarliestProtogeometric"phase.It is hardto imaginethata totallybankruptindustrycouldhave been the first to progressto the Protogeometricstyle. But there is more going on in latest IIICthan the routineoutputsuggests.Even on traditionalforms there is some excepThe stylish spiralson hydriafragmentsin the dumpedfillingof tional draftsmanship. the Fountain42 are not unique;they can be duplicatedin U 26:4 and other "Phase5" wells in the Agora.Even the wavyline in the handsof giftedpaintersfairlyleapsalong with assuranceand grace.Thereis experimentation, too, in the decorationof these traditionalpots, in the configuration of bands,wherethey are to be placed,andwhatareas will ultimatelybe preferredfor patterneddecoration.The "two-pairsystem",i.e. a pair of pendantspiralsin front, a pairat the rearflankingthe handle,with dropsseparating them at the sides, is a creationof this period (P1.22:b);43substitutingcompass-drawn circlesor semicircles,it becomesa mainstayof Protogeometricdecorationon oinochai and hydriai.And there are many other signs, both in shapes and decoration,that a smallpartof the industry,at least, is layingfoundationsfor the future. It is a painterlike these, an accomplished draftsman,with a good eye and an experimentalframeof mind, that I wouldexpectto have turnedto the compass.He and his associateswill have progressedovernightto trimand ambitiousproductsin whatwe call the EarlyProtogeometricstyle, some stayingcloselywithinthe traditionalframework, others, inspiredby exotic productsor immigrantpotters, in the bold exuberanceof 40A single Protogeometricsherd, perhapscoincidentally,came from the disturbedupper levels of the wash deposits in the Sanctuaryof Demeter and Kore, a "Phase 5" deposit at Corinth, Rutter, op. cit. (footnote 5 above), p. 60. 41p 8141: kraterwall, monochrome inside; the central core is drawnwith a compass; max. dim. 0.081 m. From disturbedfilling in a cemetery containingLH IIIC and Protogeometricburials, Kolonos Agoraios, B-C 9-10:1. 42Broneer,op. cit. (footnote 18 above), p. 398, fig. 79. 43P 17315, U 26:4: H. 0.407, D. 0.315 m.; four others, substantiallycomplete, imaginativelydecorated, came from the same deposit.

154

LORDSMITHSON EVELYN

Desborough's"WildStyle", progressinghand-in-handwith experimentin shapes.At the same time, otherswill have continuedfor a generationor more to turn out a drab products.The difference and limitedrepertory,presumablythe cheaper,mass-produced betweenprogressiveand routinemay have been quite pronouncedat this time. Only amplestratifieddepositswill answerthis questionfully, but for the time, at least, we mighttry consideringthe sequencecomplete,and not require,as I have in the past, a step-by-stepand uniformprogressionof the entire industryfrom the old to the new style. The alternative,a gap, would requirethe Klepsydrato have lain fallow for a period,to be revisitedonlyjust brieflyduringsome crisisin the earliestyearsof Protobeforethe Acropolisslopesand the protectionit still mayhave affordedare geometric,44 abandoneduntilthe Peisistratidperiod. EVELYNLORD SMITHSON OFNEW YORKAT BUFFALO STATEUNIVERSITY

Departmentof Classics Clemens Hall Buffalo,NY 14260 14Protogeometric is rare on the slopes. A few pieces came from the dumped filling in the Fountain,

Broneer, op. cit. (footnote 18 above), pp. 403, 404, fig. 85:h, j, k, 1, n. Only one other lot from Section OA contained any; the field ticket was lost during the War, and the provenience cannot be fixed more precisely. 620 of 650 sherds were Protogeometric,but the selection is quite unlike that in the channel. Several phases are represented,and closed and open shapes are in equal proportion(42%each); some purelygrave types are present, and a handful of later sherds included pieces from Middle Geometric pyxides, smoke darkened as if from pyres. The lot has nothing to do with the Klepsydraand comes more likely from disturbedgraves, the closest evidence for which is now 160 m. northwest on the Areopagus.A very few pieces of Advanced IIICor EarlyProtogeometricdate are recordedfrom the Acropolisitself. See B. Graef, Die antiken Vasenvon der Akropoliszu Athen I, Berlin 1909, pl. 7:212 (krater), 213 (shoulder, large closed pot); pl. 8:237 (amphoriskos),239 (and 240, not illustrated,lekythoi); only the kraterfragment, pl. 9:273, has compass-drawncircles.

PLATE

V'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'

a. P 16758. Miniaturehydria,side view b. p 16758. Miniaturehydria,top view

t

_

316- I ~~~~OA

''-

OA 314-31

il~~~~~~~~~~~~~~4

OA315-3 -I

OA 314-30

I

-an

c. Protogeometricsherds from channel A OA314-22 OA 314 -16

d. Late HelladicIIICbody sherds

OA314-41 e. Protogeometrichigh-footedskyphosfrom channel A

21

PLATE

22

' . . .......

.

j

.\'

t

fI

;/

:

b. Late HelladicIIIChydria,well U 26:4. Scale 1:5

.>1 OA:314 s7OA3t4-26O3

0 A314-29

OA314-7O

-27

OA 314-42

OA 314-15

c. Late HelladicIIICand Protogeometricpotteryfrom KlepsydraEast Cuttingand channel A

CA 314-25

A DOVE FOR DIONE (PLATES23-27)

N THEMUSEUMOF IOANNINA,one of the most charmingin all Greece, a room is dedicatedto objectsfrom the sanctuaryof Zeus at Dodona.In the middleof this room a case presentstwo smallboys in bronze,one a stockylittle ballplayer,the other a slim youngsterof six or seven years tenderlyholdinga flutteringbird. This latter figureimmediatelyarrestsattention.'He seems strangelymodernin the worldof ancient Dodona, among the warriors,gods, goddesses,arms and jewelrythat surround him. Nevertheless,we recognizehim as Greek, a traditionalfigureand yet a unique piece. His typewas createdthroughhundredsof yearsof tradition,reachingbackto the beginningsof the sanctuaryof Dodona, but his essence is classical,even philosophical. Canwe understandhim? Herodotos(ii.55) tells the localstoryof the foundingof the oracle:how a priestess, a iTEXEMa gLEvkaw(black dove) from Egypt, established it under an oak tree. She served

the shrine of the primevalgoddess, the EarthMother,who was originallyassociated withdoves2-inevitably,as all visitorsto Greeceand Asia Minorknowhow theirrepetitive call still throbs in the summer shade.' Later her shrine was taken over by the goddessDione, wife of Zeus, whose priestessesthemselvesstill were called "doves", Peliades.Strabo(vii.7.10)has a common-senseexplanationof this story: As for the myths that are told about the oak tree and the doves ... they are in partmore appropriateto poetry. 'lIoanninaMuseum, inv. no. 1371. J. P. Vokotopoulou, 'O8,qjy' MovO-Elov 'Iwavvivwv, Athens 1973, p. 66, pls. 22-23 (Pls. 23, 24). Measurements:H. 16 cm., H. of head 2.7 cm., H. of face 2.2 cm., chin to navel 4.2 cm., width at armpits3.3 cm., base of neck to genitals 4.8 cm. The surface is in fair condition, showing ancient rectangularpatcheson chest and abdomen, modern mending on right thigh and back and details elsewhere. The incised areas of the hair, bird's feathers, fingers and toes are sharply reproduced from the wax model without much reworking.No base was found. This figure was excavated in 1965 by the Greek ArchaeologicalService near the theater. I am deeply indebted to Professor S. I. Dakaris, Ephor of Epeiros, and to the Director of the Museum, Dr. Andreou and to Mrs. Andreou for their permissionto publish this piece and their generous cooperationin arrangingfor study and photography.The photographs in the Ioannina Museum are by R. K. Vincent who gave generously of his time and talent. I must also acknowledge,with much appreciation,help from the AmericanCouncil of LearnedSocieties by a Grant-inAid, in 1980, for the travel and photographyrelated to the preparationof this article. Finally, I was also assisted by the honorandhimself, my husband,who helped me measure and study the bronze, unknowing (I trust) that he was enhancingthe qualityof his own Festschrift. 2Full bibliographyin S. I. Dakaris, ArchaeologicalGuide to Dodona, Ioannina 1971, p. 107; bronze figure, p. 105, pl. 35:1. The figure was found buried in a disturbeddeposit with two bronze coins; one of Kassiopeia,dating 342 B.C.,the other of the Molossians, dating 168 B.C. These imply that the bronze suffered in the Roman destructionof Dodona, 167 B.C. 3D'Arcy W. Thompson, A Glossaryof GreekBirds, Oxford 1895, pp. 129-134; on the doves of Dodona, pp. 133-134. This is the "fan-tailed"type of dove; see R. Peterson, A Field Guideto the Birds of Britainand Europe,London 1971, p. 80.

BURRTHOMPSON DOROTHY

156 and (fragment1):

In the language of the Molossians (the people of Epeiros) old women are called "peliai".Perhapsthe much talked of Peliades were not birds, but three old women who busied themselves aroundthe sanctuary. (translationby H. L. Jones)

A lovely bronze of one of these priestessesof the early 5th centurysurvivesfrom Dodonawith her sacreddove confidentlyperchedon her hand.4The blood of a dove was employedeven in Athens for ritualcleansingof the sanctuaryof AphroditePandemos,5probablybecause it was not forgottenthat Aphroditewas the daughterof Dione according,at least, to the Epeirotetradition.Though earlierfiguresshow the dove in the handof the goddessor priestess,by the late 5th to early4th centurychildren are frequentlyshown handlingdoves as pets with no apparentintentionof sacrifice. We all know two famous reliefs of little girls, one with two doves and another holdingout a dove to her babybrother.6In the 4th centurymanyother such charming grave stelai may still hold a faint referenceto Aphroditeor to the GreatMotherherself.7 Parentsmust have gained a certainconfidencein seeing their child on happy terms with the goddess' own birds.When the child tired of the pet, perhapshe was inducedto learnreverencefor the deityby offeringher his birdas a votive. Ponderingthese things,we feel more deeplythe emotionalcontentof this bronze statuette;the boy's wistfulglancemay well mean that the child is about to offer the birdto Dione or to Aphroditeherself.ProfessorDakaris,in a sensitivedescriptionof the figure,8thoughtthat the gestureof the righthand (P1.24:d) was "tryingto attract the bird'sinterestor pull a stringattachedto the bird's leg." Perhaps.Or also it is possiblethat the thumband fingerof the righthandheld the stem of a missingbunch of grapesthat excites the dove. Or again, the boy may have extendeda prospective mate in the righthand,held only by the wings.9But the implicationof the gesturemay remainforeverbeyond our grasp,though its intent createsa charmingcross play of glances. Shouldthe child seem merelya genre figurein his home, considerfor a moment his peculiarhair dressing.Little boys on the Attic Choes vases,10who are sketched 4For Mycenaeandoves of the Earth Mother, see Dakaris, op. cit. (footnote 2 above), pl. 38:1 and 3. Earlyfigurines, R. A. Higgins, Catalogueof Terracottasin the Departmentof Greekand RomanAntiquitiesin the BritishMuseumI, London 1954, pls. 52, no. 336; 164, no. 1201 and J. N. Coldstream, Knossos, The Sanctuaryof Demeter,BSA, supply.vol. VIII, London 1973, pls. 59, nos. 213, 214. In bronze, H. Schrader, Phidias,Frankfurt1924, p. 152, figs. 129, 130. 5L. Deubner, AttischeFeste, Hildesheim 1962, p. 215. 61) G. M. A. Richter, Catalogueof GreekSculpture,MetropolitanMuseum of Art, Cambridge,Mass. NationalMuseum,IllustratedGuide, Athens 1954, pl. LX, no. 73, pp. 73-74; 2) S. Papaspyridi-Karouzou, 3845. 1980, pp. 58, 60, no. 7See below, footnote 25. 8Qp.cit. (footnote 2 above), p. 105. 9E.g., footnote 6 above (2) and H. Diepolder, Die attischenGrabreliefs,Darmstadt1965, pl. 33:2. 10G.van Hoorn, Choesand Anthesteria,Leiden 1951, passim.

A DOVEFORDIONE

157

playingwith animals, wear their hair short but loose and curly. Our Boy's hair is strange;it couldfall to the shoulders,but it is meticulouslyboundtightto the head (P1. 24:c, e). Behindthe ears, the hair is parted;the front ends are brushedforwardand braidedinto a plait reachingfrom partto forehead;likewise,at the back, the hair is brusheddownwardto form a similarbraidfrom partto nape.The braidsare thin, not thick as on most children who wear this central plait (o-KopfLo'?)l11The hair in such

plaitsoften was reservedfor dedicationby the child to one of the divinitieswho watch over children, the Kourotrophoi,Asklepios and Aphrodite,Demeter and Artemis. Originally,the plait marksa votary,a child who will ultimatelygive his lock or who himself has been offered to a divinity for service throughouthis childhood.'2That Dione receivedsuch childrenseems clearwhen we look also at the bronzeBallPlayer'3 who standsbesideour Boy in the Museum(P1.25). This chubbyfellowholdsa leatherballand moves to throwit. He too is naked,and he too wears a strange,even fantastic,hairdress.His abundantlocks are plaitedand twistedinto a complexerectionnot unlikethose fancycoiffuresof Trajaniccourtladies or bewiggedwomenfrom the Fayum.Theirclosestearlyparallelsare Boiotianfigurines who carryfillets and basketsas offeringsto the gods at the Kabeirion.Manyof these boys carrycocksor doves andwearimmenseandcomplicatedhairdresses.14 The custom of protectingthe sensitivehead has been commonthroughoutthe ages. Since the hair is a seat of personalpower,it must be cautiouslytreated(as witness, the tale of Samson)."5Ourboys fromDodonashowthatDione was happyto receivesuchvotaries. It is probablethat these bronzesrefer to living boys who were, accordingto the long tradition,vowed to the goddessat birthand dedicateduntil puberty.Both bronzes show boys well below that age-either becausethe parentsremovedthem at an early time or possiblybecause the child died in the sanctuary.The unusuallysentimental mood of our bronzemay suggestthat the parentsdedicatedthe statuemore in sorrow than in pride.They chose a distinguishedsculptorwho caughtthe mood and preserved it for us. Canwe guess at his artistictradition? Smallbronzesof the late Classicalperiodare surprisingly rare;most are heroicor athletic,not portraits.OurBoy is certainlyearlieras well as olderin age than the chubby BallPlayer.The legs of the BallPlayerare widelyspread;he moves at a diagonalto expressthe impendingthrustof the ball, a centrifugalpose that swingsoutward.The RE, s.v. Haaropferund Haartracht,col. 2125 (Bremer). It seems impossible to define the Greek terms with precision.See D. B. Thompson, Troy,supplementarymonograph3, pp. 40-44. 12V. von Gonzenbach, Untersuchungen zu den Knabenweihenim Isiscultder rimischenKaiserzeit,Bonn 1957, pp. 27-29; p. 31 for the Hellenistic plait in Greece. 13loanninaMuseum, inv. no. 1410, Vokotopoulou, op. cit. (footnote 1 above), p. 66, pl. 21; Dakaris, op. cit. (footnote 2 above), p. 105, pl. 35:2. I am much indebted to Dr. Andreou for permitting me to publish these photographsby R. K. Vincent (P1.25). "4B.Schmaltz, Terrakotten aus dem Kabirenheiligtum bei Theben,Berlin 1974, pl. 11, no. 152; pl. 12, nos. 153, 159, 167. "5A. Gotsmich, "Volkstiimliche Ausschauungen in der griechischen Kunst," InternationalKongress, BerichteVI, Berlin 1939, pp. 436-438, pl. 42:c and f.

158

BURRTHOMPSON DOROTHY

Boywiththe Dove, in contrast,standsquietly,to concentrateon his charge.The movement of his head and armscreatesa centripetalcomposition.His left relaxedleg is set forwardto stabilizethe stance.The axis of the body,seen best fromthe back,runsin a very faintcurvefromnapeto buttocksto rightheel. The bodysways,but very slightly; it also leans a triflebackwardso that the left shoulderand the abdomenare advanced. The movementof the arms raises the left shoulderand lowers the left hip, but the chiasmosis too delicateto be calledPraxitelean.It is a quiet pose, balanced,but not lazy. It is, fundamentally,the pose of Lysippos'Agias (datedsomewherebetween340 and 334 B.C.). It even more closely resemblesthe pose of the very youthfulbronze athleterecentlyacquiredby the GettyMuseum(P1.26).16Allowingfor the differencein age of the subjects,the balancedpose, the profilesof the bodies and the restrained modelingcertainlyrelateour bronzeto the Lysippiantradition.The soft musculatureof the childis keenlyobservedbut not exaggerated.The stomachis not plumpbut swells slightlyout belowthe navel to the immaturegenitals.The thighsare slender,the lower for the subject'sage to legs are a little short, and the head is realisticallyproportioned aboutone fifth of the height.The childhas come into its own as a subjectfor realistic a noveltyperhapsinitiatedby Lysipposhimselfwhen in the 340's he underportraiture, took portraitsof the child Alexander,"a pueritiaeiusortus"(Pliny,xxxiv.63), thus no doubtsettinga fashionamongthe well-to-dowho wereproudof theirsons. Let us look our Boy in the face for a moment.It is a roundchildishface with a high foreheadunder tightly combed hair (P1. 24:c). The long, narroweyes, under droopingupperlids, have delicatelyprickedeyeballs.The nose is aquiline,not stubby like the nose of a babyEros.The largeears are set ratherhigh.The mouth, a thin line with deep cornersand a wide lower lip, shows barelya trace of a smile. In many of these details,as well as in the treatmentof hairand thin plait,the headresemblesthat which also has a wistfulair. Our Boy is not of an Eros often consideredLysippian,17 offeringhis pet affection,not teasingit with roughposErosbut a seriousbird-watcher sessiveness.In this expressionwe alreadyread characteror ethos. The sculptormust have felt a realaffectionfor children,perhapsfor this childin particular. It is this lively expressionof ethos that placesour bronzein the lateryearsof the 4th century,just at the time when Aristotlehad expressedthe growinginterest in naturalismin art and in reasonablenessin raisingchildren.He believed (Politicsvii.17) that childrenwere trainableby modelsthat fit the idealsof their society.Perhapschildren's pets were chosen also with purpose:cocks to encouragea pugnaciousspiritin boys, doves to developthe gentlercharacterof girlsandof the sons of "gentlemen". These figuresof votaries,in bronzeand also in marble,were dedicatedin many sanctuaries,but particularly,as we have noted, in those of the various deities, the 160n the Lysippianpose, F. P. Johnson, Lysippos,Durham 1927, pp. 128-131, pl. 20; E. Sjdqvist, Lysippus(Semple Lectures), Cincinnati1966, pp. 13-14, pl. 3. J. Frel, The GettyBronze(preliminarytext), p. 12; I am most gratefulto Dr. Frel for supplyingthe photograph. 17Forour head type, cf. Johnson, op. cit., pp. 114-115, pl. 18, an Eros head in Copenhagenwith close hair, heavy-liddedeyes, a thin nose, a straightmouth.

A DOVEFORDIONE

159

Kourotrophoi,that protectedchildren.In particular,the precinctof Artemisat Brauron which was destroyedby floodingat the end of the 4th centuryhas producedmany examplesof our period.18We must look carefullyat these child-votariesfor a fuller of our Boy. understanding The marblestatuesfromBrauronpresentan engagingpictureof the little devotees who lived, like the AthenianArrephoroi,in the sanctuary,as in a convent, to serve their goddess.All are aristocrats:the girls know how to wear their long dresses, the boys how to bearthemselvesin nudity.One girl, whose head is aboutone fifth of her height,carriesa pet hare, another,a triflemore advancedin style, holds a dove in the overfoldof her dressand looks downat it, as deeplyconcernedas our Boy (P1.24:a).19 Her sparefeaturesare also very like his, as are the eyelids of the "BlindChild" (PI. 27:a)0.2A little boy also carryinga dove has a similarseriousface.2"Closestof all the Brauronvotariesto our bronzeis the statueof a little boy claspingtwo doves closelyto his chest and listeningwith delightto theircooing (P1.27:b).22 a date Our Boy with the dove has been datedin the "earlyHellenisticperiod,"23 entirelyplausiblein the widersense, since realisticfiguresof childrenwere most popularjust at that time. Manyportraitsof children,however,were createdwell beforethe The historyof this type is clear, particularly from the long end of the 4th century.24 serieson Attic gravereliefs.Of all the pets withwhichthese childrenplay,the birdand the dog are by far the most popular.They are renderedin manyposes, often together, in delightfulcross movementsbetweenthe eagerdog and the timidbird.By the 320's Otherclose examples the childrenon gravereliefshave attainedrealisticproportions.25 are seen in a groupof marblestatuesfoundin Athensnearthe Ilissos.26One of these is reminiscentof the statue of the girl with the dove from Brauron;it could be by the Both girls standquietlylike our Boy. Both wear ampledresseswith a same sculptor.27 crinkledsurface,a treatmentknownalso in majorsculptureof the late 4th century.The 18Thematerialfrom Brauronis in process of publication.For preliminaryreports, see P. G. Themelis, Brauron,Guideto the Site and Museum,Athens 1971, bibliography,p. 35. 19Themelis,op. cit., pl. 71:b. Photo R. K. Vincent. 20Themelis,op. cit. (footnote 18 above), pp. 22-23 and p. 26. See S. Karouzou, 'ApX'E0,1957 [1961], pp. 68-83. 21Themelis,op. cit. (footnote 18 above), pl. 71:c. 22Brauron,inv. no. 1239. Photo Emil, through the kindness of Lilly Kahil with generous permissionof Dr. V. Petrakos. 23Dakaris,op. cit. (footnote 2 above), p. 105. 24E.g.,Diepolder, op. cit. (footnote 9 above), pls. 33, 38, 39:2, all of the later 4th century. 25A.Conze (Die attischenGrabreliefs,Berlin 1893-1922) gives the full development. Note particularly vol. I, no. 108 (for heavy eyelids), no. 290 (for elaborate hair); vol. II, nos. 938, 969, 977, 1100 (for doves). 26I. Svoronos, "Ao-KXq71rtaKa ArvrqAE'aKal KIovoXarpdta 4V'AO'4vas,"'ApX'E0,1917, pl. I:a (Athens, N.M. inv. no. 695), pl. II:f8(inv. no. 696). M. Bieber, TheSculptureof the HellenisticAge, New York 1955, fig. 542. 27Svoronos,op. cit., pl. II:-y(inv. no. 693); cf. Themelis, op. cit. (footnote 18 above), pl. 71:b. The style of the Ilissos figure seems a little more advancedthan that of the Brauronianand may run to the end of the century.I owe ProfessorEvelyn Harrisonmuch enlightenmenton this group of sculpture.

160

DOROTHY BURRTHOMPSON

lowest date for these statuesmust be close to 317 B.C.when Demetriosof Phaleron passedsumptuarylaws that ended exuberantfunerarysculpturein Attica.28The subsequentmiserablerule of DemetriosPoliorketes(317-307) producedno sculpturalrevival. A depositof clayfigurinesin the AthenianAgoraillustratesthe later,freerstyle, our lowerlimit.Twenty-onefiguresof boys wearingMacedoniandress,chlamysand kausia, show us the taste of the period,307 to 287 B.c.29These figurineshave chubbyfaces, thickcurlsto the shoulder,slimmerproportions,and they show more animatedmovements than those of the funerarysculpture.In style they are obviouslymore advanced thanthe elegantBoy fromDodona.His date, therefore,so faras we can now tell, must fall in the late 30's or early20's of the 4th century. Not only do the close parallelsin Brauronand Athensindicatethat the artistof our bronze statuetteworkedin Attica, but more evidence is at hand. On the South and West of the AthenianAgora,the AmericanSchoolof ClassicalStudieshas uncovered the remainsof modest houses in a wide industrialarea. When, in 1932, I excavated there myself, the best preservedancientremainson the rockysurfacewere channels and drainsbuilt to organizethe overflowof abundantwaterthat pouredout in springs fromthe rocksabove.Certaindrains,whichrannearseveralof the dwellingsof artisans in clay, marble,and bronze,were carefullybuiltwith stone wallsdeep set into bedrock and packedwith earthfilling,includingwastersfromthe near-byshops.The date of the potteryfrom this fillinglies in the late 4th and early 3rd centurieswhen a need for waterwas badlyfelt.30The qualityof the fragmentsof certainterracottafigurinesfrom this fillingis remarkablyhigh, even at times sculpturalin style. Among them a mold turnedup that startlesus by its resemblanceto the DodonaBoy (P1.24:f).31A modern clay cast from this mold shows the head and upperpartof a nakedboy (P1.27:c, d). The head lookssharplydowna trifleto its left. The armsare cut off at the elbowssince the forearmswould have been addedlater.The child's face with its long eyes under droopinglids, its slim nose and wide mouth are almost identicalwith those of the bronze.The bronzeis regularlya little largerthan the mold;a consistentdifferencein Variations measurementsis more thaneight percentbetweenbronzeand clayfigures.32 occuron the mold, such as the bunchesof curlsof laterstyle whichwere addedover 28Forthe style, see 0. Palagia,Euphranor,Leiden 1980, p. 31, fig. 27. 29StellaG. Miller, "Menon's Cistern," Hesperia43, 1974, pp. 194-245; chronology, pp. 209-210. For zur politischenGeschichteAthensim 3. Jahrhundertv. Chr. ( Vesthe period, see C. Habicht, Untersuchungen tigia 30), Munich 1979, pp. 22-33. I am most grateful to Professor Habicht for giving me references on this period. 30Miller,op. cit., pp. 209-210; see particularlypl. 37. R. S. Young, "Sepulturaeintraurbem,"Hesperia vonAthen, 2nd ed., Munich 1931, p. 86. 20, 1951, pp. 113-114 and W. Judeich, Topographie 31Agora T 153 from Area H 16 in the city grid on the slope of the Areopagus. Measurements:P. H. 9.5 cm., H. head 2.7 cm., navel to base of neck 3.5 cm., width across body at armpits2.8 cm. These measurements should be comparedwith those in footnote 1 above. They are also identicalwith those of Agora T 2960/2, a much damagedcopy of the same type, to be publishedin the Agora volume on terracottasof the Hellenisticperiod. 32In general, R. V. Nicholls, "Type, Group and Series," BSA 47, 1952, pp. 217-226; B. Neutsch, Koroplastik,JdI-EHXVII, Berlin 1952, pp. 10-12, well illustrated. Studieszur vortanagrdisch-attischen

A DOVEFORDIONE

161

the ears and the less open positionof the arms.Moreover,fragmentsof other related piecesalso have been foundin the Agorawhichhelp us developthe storyof this Attic type.These will be discussedin the forthcomingCatalogueof the Agoraterracottas. Whateverthe historyof the clay descendantsof the bronze Boy, they definitely identifythe sculptoras an Athenian.He too must have workednearthe Agora,beside the sculptorsof marblegravereliefs and of dedicatorystatuesof children.Possiblyhe copied one of them, or, more likely as a bronzeworker,he createdthe models from whichthe sculptorsworked,which, in turn, were boughtor borrowedby coroplaststo maketheirown models.33 How did this bronze statuettecome to Dodona? Fortunately,oratorsrecountin They tell how detailthe storyof Athenianrelationswith Dodonafrom 343 to 325 B.C.34 Athens, in fear of Macedonianexpansion,sought advice from the oracle of Zeus at Dodona.In 343 B.C. an Athenianembassywas advisedto bewareof her commanders, and, above all, to keep the city unified.The who were too readyfor self-advancement, Atheniansneverthelesscontinuedto squabbleover policy, althoughthey expressed theirrespectfor the oraclesby obeyingZeus' commandto refurbishthe statueof Dione and to makerichofferingsin Dodona.In 330 B.c.,Olympias,the motherof Alexander, who had fled to her old home Molossia(in Epeiros),enteredthe scene. First, she gave the Atheniansa silverphialefor the statueof Hygieiathat had been erectedat Oropos by a generouscitizen, Euxenippos,trierarchin 333/2 B.C. He was later accusedof improprietyin acceptingthe gift.35Olympiasin angryletterswarnedthe Atheniansto stay out of Molossia,her land, she claimed,aftershe had drivenawayher daughterKleopatra.She said that the Athenianshad no right to build at Dodona or to bringgifts there.36Demosthenes, however, succeeded in defendingEuxenippos'conduct, and embassiescontinuedduringthis period(ca. 330-325 B.C.).37 Usually such delegations, as we see on inscriptionsrelatingto Delphi and Oropos,38were composedof distinguishedcitizens-like Lykourgosand Demades.On such an embassyat this perioda rich aristocrat(perhapsEuxenipposhimself?)must have carriedthe votive statuetteto Dione. Perhapshe had a specialrequestof the goddess-that she rememberthe attractivelittleboy who broughther a dove to sacrificeon his behalf. This statuetteof the Boywith the Dove for Dione was thrownout fromher temple eitherwhen it was looted by the Aitoliansin 219 B.C. or, more likely, in 167 when the 33Theprocess of copying metalworkin clay is studied by W. Zuchner, "Von Toreuten und T6pfern," JdI 65/66, 1950/51, pp. 175-205. 34P. R. Franke,Alt-Epirus unddasKonigtum derMolossier,Kallmunz1955, pp. 55-67; N. G. L. Hammond, Epirus,Oxford 1967, pp. 543-546, 583. 3Hypereides, 111.13,19; 32, on the characterof Euxenippos.A. N. Oikonomides (Neov 'A@rjva'v 1, 1955, pp. 57-59) identifies Euxenippos as a trierarch in 333/2 B.C. F. W. Mitchel, LykourganAthens (Semple Lectures), Cincinnati1970, pp. 24-25. 36Hypereides,ni.24-26. G. Bartolini, Iperide,Padua 1977, pp. 70-71. A. N. Oikonomides, 'YlrEapdbov AloyoLA"Evevh7rrou, Athens 1958. 37Demosthenes,xix.298; xvII.253; Deinarchos, 0.78. 38E.g., SIG3, I, 296 (330-326 B.C.), 298 (329 B.C.).

162

BURRTHOMPSON DOROTHY

It lay buriedfor some two thousandyearsuntil Dione Romansburnedthe sanctuary.39 graciouslypermittedits resurrectionin 1965.Now, in a strangeworld,the Boy brings his dove as one of the manyofferingspresentedwith gratitudeand affectionto a philhellenewho, like Dione, loves smallchildren,birdsanda fine bit of sculpture. BURRTHOMPSON DOROTHY Princeton,NJ 08540 39Seefootnote 2 above for the context.

PLATE 23

C)

C)

9A~~~~~A E: 0$

0

:

>~ 0)

N 0

*

0t a)

:? E

0)0

0

09

PLATE 24

iA

'r)

04V

CZ

PLATE 25

_

0

E >0)

0

._

s

0

C I-

a

PLATE 26

I

s

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a

oor,

a

Add a)

0:

a-

.

0

~

H

27

PLATE

.X~'d/

Sb

U

C

C)

C-

~> ~ ~

C

U~~~~t

H~~

0

REFLECTIONSON THE ATHENIAN IMPERIALCOINAGE (PLATE28)

of ancientart and artifactsrepresentedin AMONG THE VARIOUSCATEGORIES X' \.this volume, the coins are uniquein that they have somethingto say about all threeaspectsof the chosen theme:Athenianarchitecture,sculpture,and topography. It is especiallyappropriate, too, thatcoins be includedin a tributeto the eminentarchaeologist whose name is indeliblylinked with the AthenianAgora, for they compriseda high proportionof the excavationfinds, coming from the groundin such abundance that they might be said to have litteredthe site. Althoughoften lackingin aesthetic appeal,these smallrecordsof the city's economy,like everythingelse from the Agora, were importantto HomerThompson,who took a keen interestin the analysisof the numismaticmaterialand its ultimatepublication.This articleis an inadequateexpression of admirationand appreciation from one who has had the rewardingexperienceof workingwith him.1 A Roman numismatistrecentlydescribedthe Atheniancoinage as "amongthe most boringever producedby man"-a veritablenumismaticdesert2 and it must be admittedthat there is some truthin the judgment.For centuriesthe mint of Athens put out a silver currencyof such probitythat it found universalacceptancebut one characterized by a rigidretentionof traditionaltypes:Athena's head on the obverse, Athena'sowl on the reverse.Eventuallyas the trend in coinageintroducedlargerand thinnerflans, Athens adoptedthe new formatbut with scant alterationof the basic types. Athena'shead was now that of the Parthenosstatue, her owl rested on a Panathenaicamphora,and the abbreviatedethnic of the reverse was supplementedby names, dates and control marks.The style was new but the repertoireof types remainedconventional. Undoubtedlythe standardized simplicityof designwasthe resultof officialpolicyand not of engravingincompetence.Individualdies of Archaicand Classicaldate are of high quality,attestinga skill whichcouldhave dealtwith far more elaboratemotifs.Even in the Hellenisticperiodwhenthe generallevel of artisticachievementwas lower,thereare 'The only substantialcompilation of Athenian Imperial material and hence the basic work for the presentdiscussion is J. N. Svoronos and B. Pick, Les monnaiesd'Athenes,Munich 1923-1926 (hereafterSv. followed by the plate and coin numbers). Publicationof the coins from the Agora, now being undertaken by John H. Kroll, will add hundredsand perhapsthousandsof specimens to the existing record. I am indebted to G. Dembski, 0. Morkholm, M. Oeconomides, M. Price, H.-D. Schultz and T. Volk for casts of coins in the cabinetsof Vienna, Copenhagen,Athens, London, Berlinand Cambridge(England). A scholarin this countryhas kindlysuppliedthe photographof the coin of Bizya,which is in his collection. 2T. V. Buttrey, "The Athenian Currency Law of 375/4 B.C.," GreekNumismaticsand Archaeology. Essaysin Honorof MargaretThompson,Wetteren 1979, p. 33.

164

THOMPSON MARGARET

The die-cuttersof Athensweresurelyableto reproexamplesof superiorworkmanship.3 ducemorecomplicatedandimaginativetypes.Thattheyfailedto do so musthavebeen a decisionof the state, awareof the advantagesfor a majortradingcenterin havinga coinage whichwas easilyrecognizableand hence readilyacceptable.The "owls"of Athens traveledwidelyandwerewelcomeeverywhere.If one deplorestheirstereotypedmonotony, one mustagreethatit madegoodsense froma commercialpointof view.4 Only in the bronzecoinage,primarilyfor local circulation,was there a degree of innovation,but as a late development.On some of the Hellenisticbronzes,a fulminatingZeus or a fightingAthenareplacedthe owl on the reverses;a few issues showed unusualtypes such as a Gorgoneion,a Sphinxor a seated Dionysus.These, however, whichdominatedthe were mainlysupplementsto the Athenahead-owlrepresentations coinageas a whole. Withthe periodof Romanhegemonythe situationchanged,and it is with this late Atheniancoinagethat we are concerned.As Rome extendedher empireto embrace Greeceand most of Asia Minor,she imposedcurrencyrestrictions.The hithertoindependentGreek cities and states, with few exceptions,were no longerfree to strikein gold and silver.They were allowedto issue bronzecoins and these, knownto numismatistsas GreekImperials,were hybridpieces.5Obeisanceto Caesarwas obligatory,or at least politicallyexpedient.The obversesof the coins carriedthe portraitand name of the rulingRomanemperor.For the reverses, the cities were apparentlyat libertyto choose theirown typesand these can often be interpretedas expressionsof civic pride. The presentmightofferlittle morethansubsistenceunderthe shadowof Rome but the pasthadbeen glorious,andit wasthe pastthatthe best coin typesevokedwiththeiremphasison the buildings,worksof art,andlegendsfor whichthe citieshadbeen famous. This was the standardpattern.At Athens, however,there were notabledeviations fromthe norm.Frombeginningto end the AthenianImperialcoinagecarriedthe head of Athenaas its basic obversetype. Rome's rulerswere never represented.That this was approvedpolicyat the Imperiallevel is almostcertainand since the coinagebegan it wouldseem to have been his personaladmirationfor Athens at the time of Hadrian,6 thatdictatedthis concessionto Athenianpride. 3Consideringthe difficultiesof cutting an ancient die with rather primitivetools and a lack of sophisticated magnifyingdevices, one marvels at the anonymous engraver of the early 2nd century B.C. who adorned the neckguardof Athena's helmet with a tiny biga and charioteer, a personal signature which formed no part of the originaldesign and would have passed unnoticed by the average Athenian. See M. Thompson, TheNew StyleSilverCoinageof Athens,New York 1961, pl. 1:1-3 and frontispiece. 4Nothingbetter illustratesthe range and influence of Atheniancoinage than the extent to which it was imitated in the Near and Middle East once the Athenian mint had curtailedor cut off exports. Only the regnalcoinages of Philip II, Alexanderand Lysimachuswere copied on a comparablescale. 5Even among museum curatorsthere is no agreementas to where these coins belong: with the Greek material,or the Roman, or as a separatecategory. 6The latest discussion of the chronology is that of J. Kroll, "The Eleusis Hoard of Greek Imperial Coins and Some Deposits from the Athenian Agora," Hesperia42, 1973, pp. 312-333. He dates the inception of the coinage to the reign of Hadrianand its terminationto the Herulianinvasion of A.D. 267.

COINAGE REFLECTIONS ON THEATHENIANIMPERIAL

165

It is more difficultto understandthe limitedrepertoireof distinctivereversetypes. and architectural The coins of other cities providea rich legacyof topographical representations.7FromPergamumwe have at least six differenttemplesand a detailedrenderingof the great altarof Zeus; from Tarsusa long series of temples,shrines, city walls, and gates; from Corinthnot only eight temples but representationsof Acrocorinth,the harborof Cenchreae,Peirene,the Agora,the shrineof Melicertes,and the tomb of Lais.These were greatcities but so was Athens, and the Atheniancoinageis almosta travestyby comparison.Of all the splendidcivic monumentswhichmighthave been featured,only two are reproducedon the coins:the theaterof Dionysusand the Acropolis. The workmanship is indifferent, and the types themselves are scarcely

modelsof artisticdesign. No greatercontrastcould be presentedthan the juxtapositionof the two Athenian issues (P1.28:1, 2) and one from the obscureThraciancity of Bizya (P1.28:3), whose recordof the topography of the site. bronzecoinageof late datehas left us a remarkable Withimaginationandskill the northerndie-cutterhas reproducedthe entirecivic center surroundedby massivewallsand towers.Withinto the left are two stoas, one above or behindthe other, then a smalladjacenttemple,severalstatuesin the open arealeading to a sizeablebuildingat the right,whichmayrepresentthe publicbaths.The city gate is depictedwith notableattentionto detail,its entranceflankedby nicheswithinwhichwe see a horse and rider, perhapsthe emperor,and a group of three standingfigures, perhapsthe Graces.Above are seven smallerniches,each witha statue,and acrossthe top of the gate a quadrigadrivenby Nike with stridingfiguresto left and right. It is impossibleto believe that the die-cuttersof Athens were incapableof producinga similarlygraphicview of the Acropolisor at least one with a better sense of proportion.As it is, the detailedrenderingof the rocksand masonryof the foundation togetherwith the oversizedstairwaytake up so muchspacethat there is scantroom for the magnificentbuildingson the hill. The Propylaeaand the Erechtheum8 are sketchily depicted;of the Parthenonthere is no trace. Furthermore,the type is standardized. Even when we have what seems to be a view from a differentdirection,showingthe stairwayto the left insteadof the right, it is reallyonly a mirrorimage of the original design,for it is still the Erechtheumand not the Parthenonthat we see. The Acropolis type persiststhroughseveralstages of the coinage, but it is as though one die-cutter simplycopiedthe workof a predecessorwith no thoughtof innovation. It seems strangeindeedthat the most renownedof Greek cities gave so little numismaticprominenceto her splendidbuildingsand monuments.One almostwondersif the dearthof architectural typeswas not in itself an expressionof civic pride.Did Athens feel that her greattemples,stoas and sanctuarieswere so well knownthat no one neededa numismaticreminderof the gloriouspast?Whocan say? 7These have recentlybeen collected by M. J. Price and B. L. Trell, Coinsand TheirCities,London 1977. 8For the identificationof the building see Price and Trell (op. cit.), p. 77. The two Athenian coins illustratedhere (P1. 28:1, 2) are from the British Museum; enlarged reverses are shown as figs. 130 and 133 in Price and Trell.

166

THOMPSON MARGARET

Architecture,of course,was not the whole story.The AthenianImperialcoin types are predominantlyanimate,and in the beginningat least the choice is imaginative. Some of the earliestcoins reproducecelebratedevents in Athenianlegendand history. On the divine level Athena and Poseidonare shown in their strugglefor the land of Attica.9Three exploitsof the Athenianhero Theseusare depicted:the battlewith the bull.10 Minotaur,raisingthe stone at Troezen,and the subjugationof the Marathonian Finally,on the human level, there are the famous victoriesof two men of Athens: Themistocleson a galley at Salamisand Miltiadesleadinga Persiancaptivetowarda These are the most interestingof the animate'types, but they are trophyat Marathon.11 overshadowedby the numerousbanalrenderingsof variousgods,andagoddessesof the enough,on Athena. GreekPantheonwithemphasis,understandably SincePausaniashas given us an invaluablerecordof the templesatrdcult statuesto be seen in and aroundAthensat the time of his visit in the 2nd centuryafterChrist,it is inevitablethat attemptshave been made to link the coin typeswith the literaryeviand the more recent study by dence. The pioneer commentaryby Imhoof-Gardner Bothdrawobviousparallelsbut bothare someLacroixare basicworkson the subject.12 what guardedwith respectto firm identifications.In essence, the questionis whether the coin types faithfullycopied particularstatues or were merely inspiredby certain sculpturalprototypeswith individualengraversadaptingtheirmodelsas they saw fit. In the finalpublicationof the Greekcoins of the Agora,a discussionof Imperialtypeswill surelybe included.Any detailedanalysisat this pointwouldbe premature,and beyond the author'scapacity,but a few wordsof cautionmaynot be amiss. city or rulerwas a matterof great Clearlythe gold and silvercoinageof a particular of basictypeswas madeat the highthat the choice importance,and it is to be assumed est level. One can scarcelybelievethat anyoneotherthan Alexanderthe Greatdecided the formatof his internationalcurrency,or that it was a mint masteror some other minorofficialwho decreedthe longevityof the "owls"of Athens or the "Pegasi"of message,an associationwith the ruling Corinth.When the coins carrya propagandistic been DemetriusPoliorceteshimselfwho It have authorityis even more implicit. must chose to commemoratehis navalvictoryat CyprioteSalamisby a beautifulseries of tetradrachms showingNike on a galley, and Lysimachusand Ptolemy who chose to their relationshipwithAlexanderby placinghis deifiedheadon theirsilver. underscore Oncethe decisionas to typeshad been madeand perhapsconfirmedby approvalof a particulardesign, it seems that the die-cuttershad a certainamountof freedomin workingout the details.Even the stereotypedmoney of Alexanderand the Successors showsa degreeof deviation:Nike is seen in a stifflyfrontalpositionor advancinggracefully left; she occasionallycarriesa long palm branchinsteadof the customarystylis; 9Sv. 89, 1-18. ?0Sv.95, 16-36; 96, 1-19 and 30-36. "Sv. 97, 1-35. on Pausanias(reprintedfrom JHS 6-8, '2F. Imhoof-Blumerand P. Gardner,A NumismaticCommentary sur les monnaiesgrecques,Liege 1949. de statues Les reproductions Lacroix, L. 1885-1887), Chicago 1964.

IMPERIAL REFLECTIONS COINAGE ONTHEATHENIAN

167

the helmet of Athena, normallydecoratedwith a coiled serpent,is sometimesadorned with a griffinor a sphinx.These are mattersof smallmoment;the form of the legend was of greatersignificancebut here, too, there was no rigidconformity.Tetradrachms whiledrachms,with the same control of Side are inscribedBAUIAE01 AAE-zXNAPOY marksand hence contemporaneous,read simplyAAE-ANAPOY.This might be explainedby the smallerflansof the fractionalpieces13were it not that coins of the same size at other mints reveal a similarinconsistency.At Sardesa staterreversewith the title and one withoutit sharean obversedie. To what extent the bronzecoinagewas regulatedis anothermatter.Since it was primarilyfor localcirculationand hence less importantas an instrumentof state policy, the decisionon types may well have been made on a lowerlevel, perhapsat the mint itself. On the large AthenianImperialcoins the obverse type is alwaysthe head of Athena,14but details of the representationseem to have been left to the individual engraver.The great majorityof dies show the goddess facing right, wearinga Corinthianhelmet, and a borderof dots surroundsthe design. On some dies she has "an Attic helmet and althoughthis type is most frequentlyfound on earlyissues, it recurs occasionallyat a laterperiod.In the long bucraniumseriesthereare two coins sharinga reversedie (P1.28:4, 5);15for the obversesone die-cutterhas chosen an Attic helmet and the other the Corinthianversion. A few obverse dies show Athena facing left insteadof right,16and quite clearlythese are the workof the same hand.The wreath, whichreplacesthe borderof dots on variousobverses,has been takenas an indication of specialcommemorativeissues17but here again it seems ratherto have been the personalpreferenceof an individualengraver.In all probabilitythe same man cut the wreathedobversesused with the diversetypes of Athenaand Poseidon,Themistocles on a galley,Miltiadeswith captive,and Theseus battlingthe Minotaur(P1.28:9-13).18 At a laterstage of productionthe obverseof a coin with the runningAthenatype (Sv. 85, 20) is wreathedbut anothercoin from the same reverse die (Sv. 85, 19) has a dottedobverse.Occasionallythere is an isolatedwreathedreverse:one of Athenain a chariot(Sv. 88, 18) and one with the goddessstanding(Sv. 83, 24-26 from the same die). Since other reversesof the same type and the same periodare unwreathed,the additionmust surelybe thatof the individualdie-cutter. It has been suggestedthat the reversetypescommemoratemajorcivic festivalsand that the coins were issued on these occasions.19 A particulartype may, of course,have "It should be noted, however, that die-cutters at other mints were able to place both name and title on their drachms. "4Thesmaller denominationssometimes use a different type such as the head of Heracles or Theseus or a bucranium. '5Sv. 99, 20 and 21. 16Sv. 99, 14 and 22, e.g. 7J. P. Shear, "AthenianImperialCoinage," Hesperia5, 1936, pp. 296-298. '8Stylisticcomparisonsbased on photographsdistributedthrough Svoronos' plates present difficulties. Bringingthe coins themselves together would surely produceadditionalexamples of a single hand at work. The coins illustratedhere are Sv. 89, 3 and 5; 96, 30; 97, 2 and 34. "9J.P. Shear, op. cit. (footnote 17 above), pp. 296-316.

168

MARGARET THOMPSON

been chosen becauseof an especiallylavish celebrationof a particularfestival, but in generalit wouldbe difficultto drawthe line betweena type inspiredby a festivaland one inspiredby a cult statueassociatedwith the festival.Wasit the GreaterDionysiaor simplya statueof Dionysusthat promptedthe representation of the seatedgod on the coins? Certainlyit is unlikelythat there was any direct temporalconnectionbetween the appearanceof the coin type and the celebrationof the festival.The patternof dielinkagearguesagainstit. Examplesof differentreversetypessharingan obversedie are fairlyfrequent.Even withinthe small collectionof the AmericanNumismaticSociety andwithoutprolongedsearch,the followingconnectionswerenoted: Theseus and Minotaur(as Sv. 96, 34) Miltiadesand captive (as Sv. 97, 32) Themistocles on galley (as Sv. 97, 9) Bucranium(as Sv. 99, 5) Apollo.(as Sv. 93, 8) Athena Parthenos (die not in Svoronos; type of Sv. 82) Athena "Promachos"(P1.28:6) Demeter in chariot (P1.28:7) Bucranium(P1.28:8)

Dionysus(Sv. 92, 17)20 Acropolis (Sv. 98, 22)

Carefuldie comparisonof the Agoracoins would undoubtedlysupplymany additional examples,but even these few linksindicatethat varioustypeswere being issuedat the same time and not spacedto coincidewith the dates of the festivalswith which they mightbe thoughtto be connected.Nor does thereseem to be any ostensiblecorrelation among these contemporaryemissions.It is as though the engraverhad a list of approvedtypes from whichhe was free to make a selection, and this indeed may have been the case. Whateverthe procedurefor choosingthe reverse type, its reproduction,too, was not rigidlycontrolled.Of all types, those depictingAthenaare the most numerousand most varied.Pausaniasrefersto 17 statuesof the goddessand an additionalfour temSvoronos'platesincludean even greaternumples or shrinesdevotedto her worship.21 identifiedwith particular statuesor as differentttypes".The Atheber of representations nianengraverwith Athenaas his subjecthad an abundanceof sculpturalmodelsin and about the city, and he must have been influencedto some extent by these familiar monuments.It would be strangeindeed if that were not the case, but there is always a particular statueand, if so, how accurateis the problem:is he consciouslyportraying the reproduction? OThe die identityderives from Svoronos' plates; there are no comparablecoins in the ANS collection. 2ITheseare listed by Imhoof-Blumer,op. cit. (footnote 12 above), pp. 125-126.

IMPERIAL REFLECTIONS ONTHEATHENIAN COINAGE

169

Minorvariationsin adjunctsymbolsare numerous.The type identifiedas Athena Parthenos(Sv. 82 and 83, 1-14) is frequentlyshown with a coiled serpentin the left field;on a few dies the serpentis placedto the rightanda facingbucraniumoccupiesthe spaceto the left; one die replacesthe serpentwith an owl whileothersare withoutsymbol. Svoronos'AthenaMedici(?) may hold her pateraover an altar,over a serpentor over an emptyfield (Sv. 86, 30 and 40-42). These are insignificant differenceswhichdo not affectthe basictypebut one wondershow muchfurtherartisticlicenseextended. It is evident that Svoronosacceptedcertainvariationsin the type itself as compatiblewith a singlemodel.His AthenaParthenosstandsfacing,head to the left, holding a Nike in her outstretchedrighthandwhile her left supportsan uprightspearand a shieldseen side view at her feet. Also identifiedwith the Parthenosis a version(Sv. 83, 15-19) in whichthe shieldis seen face on, obscuringthe lowerpartof Athena'sbody, and the spearis tuckedinto her left armas she holds the shield.Tentativelyassociated with the same statueare variantsof the firstand dominanttypewith Athenaholdingan owl or a paterainsteadof a Nike (Sv. 83, 38 and 40; 87, 13 and 14). One mightsuggest that some of Svoronos' "autre types de lAthena Nikephoros"(83, 20-28) are similar adaptationsof the Parthenoswith Athena'sarmraisedas she holds the spearwhile her shieldmay be shownbehindher or on her armclose to her body. Numerousreverseshave an advancingAthena.Some are describedas the goddess of the east pedimentof the Parthenon(Sv. 85, 8-37 and possibly38-40). If such diverse renderingsas nos. 36-40 are to be construedas adaptationsof the basictype, one mayask whetherotherversionsof the runninggoddesson the same platemay not also derivefromthe same model.The differencesare not great. Assumingthe die-cuttershad an existing statue or sculpturalgroup in mind, it would seem as though the dominanceof a particularrepresentationmight indicatea more or less faithfulcopy, but this is not necessarilythe case. An interestingseries of reversespertainto the Theseus saga. Since Pausaniasmentionsthree monumentson the Acropoliscommemorating his exploitsand since the coins deal with the same three themes,22there shouldbe, primaface, some connectionbetweensculpturaland numismatic representations. All reversesshowingthe hero raisingthe stone at Troezen to uncoverthe proofsof his identityare consistent;23 in all probability the coins furnishan accuratereproductionof the sculpturaltype. The other two exploits are diverselydepicted.On one reversedie (Sv. 95, 23 and 24) Theseusis leadingthe Marathonian bull towarda seatedfigure;all other dies show only Theseusstandingbehindthe bull with upraisedclub.Eliminationof the seatedfigure,if it formedpartof the Acropolisgroup, is understandable in view of the difficultyof compressingan elaboratemotif into a flan 22Pausanias,i.24.1 and i.27.8-10; P1.28:14-19 (= Sv. 95, 17, 23 and 30; 96, 5, 17 and 36). The provenances of Sv. 95, 16 and 17 have been transposed;our no. 15 is the Cambridgecoin (= Sv. 95, 17). Our no. 18 (Sv. 96, 17) is not in the Berlincollection. 23Representationson coins of Troezen are also consistent with the Athenian type. Cf. A CatalogueQf the GreekCoins in the BritishMuseum,Peloponnesus(hereafter BMC followed by the district), pl. 31:5 and 9.

170

MARGARET THOMPSON

of smalldimensions,but we cannotbe sure thatone imaginativeengraverdid not addit to make the story more explicit.24 Thereare three numismaticversionsof the struggle with the Minotaur.Earliercoins show Theseusand the monsterconfrontingeach other as the battle begins; later dies portraya fallen Minotaurwith Theseus first towering over him with upraisedclub and then kneelingon the prostratefigurewith club tucked into his arm.Versionsone and two are linkedby an obversedie (P1.28:17, 18; Sv. 96, 36 and 17); the strikingchange in representationoccurs within the lifetime of this obverse.It is the thirdversion, however,whichis dominantduringthe laterstages of the coinagedespitethe fact that it lacksthe vitalityof the previousrenderings.Which,. if any, of the numismatictypes is to be relatedto the sculpturalgroupof Pausaniasis a questionthe coinsalonecannotanswer.25 24Thereis also the possibility, here and elsewhere, of a pictorialrather than a sculpturalprototype. Evelyn Harrison,who has kindly read this articleand offered helpful comments, has called my attention to the similaritybetween this rare coin type and the representationon a cup by the Penthesilea Painterwhich she illustratesin "The Iconographyof the EponymousHeroes on the Parthenonand in the Agora," Essays Thompson(footnote 2 above), pl. 6. As she says in a letter, although we have no literaryreference to a paintingof Theseus bringinghome the bull, it is probablethat there was one somewhere in Athens. Without doubt coin types were at times inspiredby paintings.Notable examples include a bronze of Abydus depicting the Hero and Leander legend (BMC Troas,pl. 3:2) and a Heracles medallion published by M. Thompson, "A Greek ImperialMedallionfrom Pautalia,"ANSMN 22, 1977, pp. 29-36. On occasion the decorationon a vase would seem to have provided the model. The fighting-Athena type on coins of Ptolemy I and other Hellenistic rulers has recently been discussed by C. Havelock, "The ArchaisticAthena Promachosin EarlyHellenisticCoinages," AJA 84, 1980, pp. 41-50. Citing (her note 3) L. Lacroix and E. B. Harrisonas against A. B. Brett and others, she rejects the identificationof the coin type as the Athena Alkis statue of Pella and connects it with the fighting Athena of the Panathenaicvases. In this she is almost certainlyright, but I think she goes too far in readingsubtle propagandisticmessages into the choice of the type by Ptolemy and others. For one thing, propagandato be effective needs to be disseminatedwidely. Ptolemy's Athenas did not circulateoutside the country, and indeed Ptolemy's own fiscal policies prevented their moving about freely. Forty-one hoards with coins of Ptolemy I are recorded in An Inventoryof GreekCoin Hoards,edited by M. Thompson, 0. Morkholmand C. M. Kraay,New York 1973, p. 392, with no. 226 omitted since the presence of the tetradrachmof Ptolemy I is uncertain. Of these, 24 were found outside Egypt, and in all but four the type of coinage can be identified. Exactly two deposits contained Athena tetradrachms:one example of a Sidonian issue unearthed at near-by Byblus (IGCH 1515) and more in a hoard from Chiliomodi near Corinth (IGCH 85) where the Ptolemaic garrison of 308-306 B.C. explains their presence. Even less plausibleis the argument that Seleucus I copied Ptolemy's type and used it for the same propagandisticreasons. The Athena of Seleucus appearsnot on his silver or gold, which did travel, but on bronze issues, which could have had little more than local impact since bronze coins rarely circulatedfar from their place of emission. There seems to me nothing particularlysignificantabout the choice of the Athena Promachos type for various Hellenistic coinages. If a king wanted-to stress his military prowess, at home or abroad, an obvious device was the warringAthena, whose representationon Panathenaicamphorasmust have been familiarto his die-cutters. 25A fragmentarysculpturalrepresentationof Theseus and the Minotauris discussed briefly by S. KaMuseum.Collectionof Sculpture,Athens 1968, pp. 43-44. Evelyn Harrison, rouzou, NationalArchaeological who has helpfully checked this out for me, tells me it is a Roman Group in which the Minotaurwas a fountain with water flowing from his mouth. The torsos were found in the area of the "ValerianWall" between the Libraryof Hadrianand the Acropolis;where the figures originallystood is not known. Even if this is a copy of the statues Pausaniassaw on the Acropolis, which is quite uncertain, a detailed recon-

IMPERIAL COINAGE REFLECTIONS ONTHEATHENIAN

171

Thereis no need to laborthe point.The AthenianImperialcoins are not uniquein theirambiguity;the same phenomenoncan be observedin the GreekImperialcoinages of Asia Minor.One outstandingexampleis that of the famoustempleof the Ephesian Artemis.This was a populartype, appearingat a numberof cities as well as at Ephesus itself.Thereis no doubtas to whatis beingrepresentedfor the distinctivestatueof the of the facade.The temple goddess is alwaysshown in the centralintercolumniation itself, however, may be distyle, tetrastyle,hexastyleor octastyle.Only at Ephesusis there a fair degreeof correlationbetweenthe numberof columnsand the size of the coin flan but even here there are puzzlingdiscrepancies.At least one die for a coin of largedimensionshas only six columnswhile the engraverof a much smallerbronzeof Nero was able to squeezenine columnsonto his die in an awkwarddesignshowingthe buildingin three-quarterview.26Under Hadrianthere are bronzeswith the octastyle templeand also cistophoroiof only slightlysmallerscale with a hexastyleor tetrastyle Yet one wouldexpect the Artemisionto be renderedwith greateraccuracy building.27 on silvercoinsthanon localbronze. It is clearthat the die-cuttersof the Imperialperiodat Athens and elsewherehad considerablefreedomin composingtheirnumismaticdesigns.Thatthe greatmajorityof these derive from, or were inspiredby, existing monumentsis indisputable.When a particularrepresentationis entirelyconsistentthroughouta series of coins, it is likely that we possess a faithfulcopy. When there are variantrenderingsof what is almost certainlythe same buildingor statue, it is hazardousto assume, without supporting evidence, that any one of the coin types suppliesan absolutelyaccuratereplicaof the basic model. This is not to denigratethe value of the numismaticcontributionto our knowledgeof ancientedificesand worksof art, but like all other evidencewe need to use it with caution,rememberingthat the engraverswere not photographers but artists and adaptingtheirprototypes. exercisingartisticlicensein interpreting MARGARET THOMPSON AMERICAN NUMISMATICSOCIETY

Broadwayat 156th Street New York, NY 10032 struction of the group seems impossible on the basis of the surviving evidence. From what remains, an associationwith the numismatictype of our P1.28:18 is a possibilitybut little more than that. 26SyllogeNummorumGraecorum Deutschland: Sammiungv. Aulock,nos. 7863 and 7879. 27BMClonia, pl. 13:7 and W. E. Metcalf, TheCistophorolof Hadrian,New York 1980, pls. 3, 4.

PLATE

28

Far

r i

l

e

16

15

,40

e

*

e

I

~~

8

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~S-

0

11

+Wmm

2

1

+9~~~~N

71

~~~~~ Pt

AGRTTOPO:RFETOSO

7

17

H

TEINIPRA

-

ONG

EHI HPOYXONTIKOAfINflI THE SACRED THRESHINGFLOORAT ELEUSIS (PLATE29) ayE

/mot V1qOV TE

Beav

KaLtf3At

' a'er

TEVXOVTWVwTas S7,uoq v1TaL 7oktOL alIT? TE TEltXo,

KaXXlX6pOV Ka0lflTEpOEV,

EI

VC0. TfpOVXOVTl KOXw

Let all the people build me a great temple and an altar below it beneath the citadel and its sheer wall, above the Kallichoronwell, on a projectingspur of hill.

~JU ITH THESEWORDSin the HomericHymnto Demeter(lines 270-272) the

goddess herself establishesher sanctuaryat Eleusis. A few lines fartheron (296-298) the poet tells us how Keleos, the king of Eleusis,calledthe peopletogether and badethem buildfor Demetera templeand an altaron a projectingspurof hill. Whatis the significanceof the locationchosenfor the temple"on a projectingspur The of hill"?Hithertoit has generallybeen explainedonly in termsof localtopography. commentatorson the passage,whetherthey be editors,archaeologists,or historiansof religion,are all agreedthat the descriptionas a whole appliesperfectlyto the site of Eleusis,and they merelyseek to identifythe landmarksthat are mentioned,the temple and altar,the acropolisandits wall,the Kallichoronwell, and the spurof hill.1 OnlyYves Bequignonhas soughta deeperreason.In an articleentitled"Demeter, dgesse acropolitaine"2 he passes in review a number of sanctuaries of Demeter known

eitherfromactualremainsor fromliteraryreferencesand findsthat they are locatedby preference,not on the summitof a hill but on the flank, sometimeson a hillockor knoll. "Whyare these sanctuariesgenerallyplacedon heights?"he asks in concluding. "Thereis no need for a long commentarysince the HomericHymngives us the elementsof the answer.It indicatesin severalpassages(lines 60, 75, 442) that Demeteris the daughterof Rhea and that she arrivesfrom Crete (line 123). In Crete Rhea is honoredas ,urjrqp opedr, and she has retainedcertainattributesof the GreatMother, goddessof the mountainsbut not the peaks,and in the same way she has exercisedin her turnan influenceon Demeter." This is not in itselfa very convincingexplanation,and in any case the whole theory the Cretanoriginof the Eleusinianworshiphas been shownby Mylonasto be withof out basiseitherin archaeology,historyor linguistics.3 'The most recent discussions are by G. E. Mylonas, Eleusisand the EleusinianMysteries,Princeton 1961, pp. 33-49, and N. J. Richardson, TheHomericHymnto Demeter,Oxford 1974, commentaryon lines 270-272, and AppendixI. 2RA, 1958, II, pp. 149-177. 'ApTco PwovKEpaguolrovoXov,Athens 1953, pp. 3"'H 7TpoE'XEvaq FEpaq rns sXEVGULVLaK^qskavpdaq," 42-53.

THESACRED THRESHING FLOOR ATELEUSIS

173

I should like to suggest anotherexplanationwhich seems more appropriatefor Demeterespeciallyin the aspectin whichshe was worshippedat Eleusis.The Eleusinian Demeter is the GrainGoddess, and the whole story of Demeter and Kore is an allegoryof the death and rebirthof grain.The GreaterMysteriestook place in Septemberat the season when the autumnrainswere expectedwhichwouldbringnature backto life again.Offeringsof the firstfruitsof the harvestwere madegenerallyto the two Goddesses,and the grainwas kept in largeundergroundstoreroomsin the sanctuary. Stalksof wheat are amongDemeter'sattributes,and the climaxof the Mysteries was the revelationto the initiatedof a reapedear of wheat. Furthermore,Demeterwas the goddessof the threshingfloor as we learnfrom a vivid similein Homer: Even as the wind carrieschaff about the sacred threshingfloors of men that are winnowing, when fair haired Demeter amid driving blasts of wind separates the grain from the chaff, and heaps of chaff grow white.4

Now in the sanctuaryat Eleusistherewas a SacredThreshingFloor.This fact has been knownsince 1883 when the big inscriptionwith the Eleusinianaccountsof 329/8 B.C. was firstpublishedand in whichwe findmentionof Tr'v ciAXTr'v 1Epav.5This entryhas receivedvery little attention,and the commentatorswhen they mentionit at all usually take this SacredThreshingFloor to be identicalwith the ThreshingFloor of Triptolemos knownfromPausanias,i.38.6.6But the ThreshingFloorof Triptolemoswas in the RharianPlain, and the RharianPlain, thoughprobablynear Eleusis, was certainlyneither in the town nor in the sanctuary.The SacredThreshingFloor, on the other hand, appearsto have been in the sanctuary,if we mayjudge from the context in whichit is mentionedin the inscription. And here, perhaps,we have the answerto our questionof why a projectingspurof hill was designatedas the site of the sanctuary.Sucha site wouldhave been suitablefor a threshingfloor as threshingfloors are normallyplacedon projectingspurs of hills wherethey will catchthe breezeso necessaryfor the winnowingprocess.Demeterthus ordersthe Eleusiniansto build her temple on a site suitablefor threshingand winnowingthe grain,an essentialact in the annualcycle. The SacredThreshingFloormust have been locatedat the outerend of the terrace in frontof the Telesterion.This artificialterrace,supportedin historicaltimes by a high retainingwall, representsthe "projectingspurof hill" of the HomericHymnand is the successorof the naturalspur of hill that we may postulatefor the originalsanctuary. The situationis shownmost clearlyin the model of the sanctuaryas it was in the time 4Iliad v.499-504. Translationby A. T. Murray (Loeb Classical Library).Cf. Theocritus, vIi.155, and OrphicaXL.5. 5GI112, 1672, line 233. 'Eo'ApX,1883, p. 122, line 20. 6So J. G. Frazer, Pausanias'sDescriptionof Greece II, London 1898, p. 512 and H. Hitzig and H. Bluemner, PausaniaeGraeciaeDescriptioI, Berlin 1896, p. 357.

174

EUGENE VANDERPOOL

of Peisistratosin the 6th centurybeforeChrist7(P1.29). The positionof the two altars on the terraceis not fixed, and they could be shiftedif necessaryto accommodatethe threshingfloor. EUGENEVANDERPOOL AMERICAN OFCLASSICAL STUDIESAT ATHENS SCHOOL

7John Travlos has directed the making of two models of Eleusis, one showing the sanctuaryas it was in the 6th century before Christ, the other as it was in the 2nd century after Christ. Both models are on 1973 [19751,pp. 217-218 and pls. 231-236). exhibit in the museum at Eleusis (HpaKTLKa,

PLATE

$1,

,

V

.

"

We

3n

-,

l W t~s'

Moelofth

Mother

Tletein

Tt

EUGEE VNDEPOO:

t

t

M

i

lesi

,

te

-_

*0Xfi Aavo ie

f

eiisraos

Cutey

f

oh

a

TE

SCRDTESNGLORA

EUGENEVANDERPOOL: THE SACREDTHRESHING.

EUI

AELUi

A

29

ZEUS AND OTHERDEITIES: NOTESON TWO ARCHAISTICPIERS (PLATES30-32)

NOTEis offeredin honorof ProfessorHomerA. Thompsonin [>HE FOLLOWING recognitionof the obvious:he, like Atheniancultureitself, has had greatinfluence and effectin placeswell beyondthe geographical limitsof Attica. During the excavationsof 1974 a large rectangularblock decoratedwith a single archaisticfigure on each of three sides but plain and undecoratedon its fourth was recoveredin the southwestcornerof theCorinthianforum (P1.30).1 Althoughbuilt into a 10th-centuryByzantinewall not far, apparently,from its originalemplacement,it cannotbe identifiedwithany specificfoundationpreservedin the area. In 1979a secondrectangularbase, resemblingthe first,was foundbetween30 and 35 metersto its south.This new base also has a singlefigureon each of three sides and is plainon its fourth.The secondblockis much more mutilatedthan the first, missing the lowest thirdof its shaft.Its damagedconditionmightbe expected,however,for it was founddiscardedin a pit with modernfill. It probablyhad passedthrougha number of vicissitudesbeforeit was buriedin the 18thor 19thcenturyafterChrist. When the first block was recovered,it was consideredto have been designedto supporta statue,probablythatof Zeus Chthoniosmentionedby Pausaniasin his tourof the Corinthianforum.The existenceof this second base, which is describedin detail below,demandsa re-examination of the hypothesis. T

Rectangularshaft with three archaisticfigures. S-1979-6. Pres. H. 0.97 m. Width of front panel below apophyge, 0.562 m.; width of side panel below apophyge,0.535 m. Upper half of four-sided, white marble block, the front and sides each decoratedwith a single figure, back finished but without figure. Top horizontal surface preserves much of original roughly dressed surface, includingdowel cutting at 0.23 m. in from back face. Pour channel from back face to dowel cutting. Figure 1 (P1. 31:a): standing female on left side of block, helmet pushed back on top of head, shield on left arm, trace of bracelet on wrist. Figure faces front of pier, head in profile,

body in three-quarterprofile. Hair, drawn to

back of neck, is gatheredbehindshoulderby ribbon;two stiffcurlsfall frombehindrightear onto right breast, one from left side to left breast.Traceof pendantpyramidalearringon jaw line, of typewornby DemeterandKoreon the 1974 block.Chitonis buttonedfour times along upperarm and is covered by himation drapedfrom left shoulderto underrightarm. Upper hem of himationfalls across chest in zigzagfolds. Figure2 (P1.32:a):standingfemaleon front of block, facingleft, head in profile,body in three-quarter view. Figurewearspolos, carries cornucopiaat left armwith rightarmextended

IS-74-27. C. K. Williams, 11, "Corinth, 1974: Forum Southwest," Hesperia44, 1975, pp. 23-25, 29, PIS.9-10.

176

CHARLESKAUFMANWILLIAMS, II

in downwardangle. Hairstylesimilar to that worn by helmetedFigure 1, except that this figure has two curls fallingonto each breast. Chiton with heavy seam around neck, sewn seam along upperarm. Himationis worn over rightshoulderand underleft arm, its top hem endingin gravity-defying zigzagfolds.It fallsin two sets of stackedfolds, in contrastto the singleset of stackedfoldson the himatiaof the Demeterand Kore depictedon the companion base.

Figure3 (PI. 32:b): standingmale on right side of block,thyrsosin righthand.Figurefaces front of pier, head in profile,body in threequarterview.The god is bearded,crownedwith stephanethatcatcheshis long backlock;lockis doubled up with end projectinghorizontally frombackof headover stephane.Twocurlsfall from behindleft ear onto chest. Himationcarried from left shoulderto pass, presumably, under right arm, its upperhem falling from necklinein longruffle.No traceof chiton.

One fact evidentin the base that was found in 1974 is the resemblanceof its one flankfigureto the other.Althoughthese are motherand daughter,the fact is admitted hand. Hair, dress, attributein the appropriate only by the inclusionof the appropriate postureand gesturesare basicallythe same, althoughin mirrorimage.Withthe finding of the 1979 base the principleis clearer.Figure2 is patternedafter or similarto the Demeterof the firstblock,but with the handon hip now holdingthe cornucopia.Note that the horn is superimposedupon the arm, not tuckedinto it. The rightarm is lowered, probablybecausethe figurecarriesa phiale,echoingthat held by Zeus Chthonios. The positionof the phialeis probablythat at the moment of libation.OtherwiseDemeter,Kore, and Figure2 are similar,executedwith varyingtexturesin hair,cloth and variationsin folds. Figure1 is executedon the same principle,but becauseshe (Athena?), apparently,has to carrysomethingin her right hand, probablyspear, phialeor oinochoe, the right arm crosses the body in a descendingdiagonal.'Although this gesturedefiesthe principleestablishedby the otherfiguresthatthe armnearthe viewer shouldfall to the figure'sside, it allowsFigure1 to carryan attributeor objectthatwill relateher to the generaltheme beingpresentedon these two blocks. Figure1 is providedwith the hairarrangementthat is used by Figure2, Demeter and Kore, except that Figure 1 wears over it a helmet rather than a polos.3 Also, she alonehas her upperarmcoveredby a buttoned,ratherthana seamed,chiton.4 2For general discussion and illustrationsof Athena and her attributes,iconographyand areas of activity, see D. le Lasseur, Les dresses armeesdans lart classiquegrec et originesorientales,Paris 1919; A. B. Cook, Zeus, A Studyin AncientReligionIII, Cambridge 1940 (= Cook, Zeus); L. R. Farnell, The Cultsof the GreekStates I, Oxford 1896 (= Farnell), pp. 258-352. 3Athena wears a variety of helmets during her tenure as protectressof Athens. The Corinth relief is badly damaged around the helmet, although the crest is clearly preserved. She does not wear the high crested helmet of the Promachos,nor a variationthereof. One sees preservedon the Corinth relief at 0.06 m. in front of the crest a visor ridge and behind the crest the extension of the helmet which protects the back of the neck. This is the Archaicform of helmet. It is worn, also, by the archaisticAthena from Herculaneum in the Naples Museum. See E. B. Harrison, TheAthenianAgora, XI, Archaicand ArchaisticSculpture, Princeton 1965, pl. 63:d. The Corinth Athena definitely is not wearinga Corinthianhelmet, the type favored by Athena on Corinthiancoins and by her in Attica aroundthe end of the 5th century and later. 4The combination of curls falling onto the breast, buttoned chiton, and himation draped from one shoulder to pass under the opposite arm, as well as the wearing of aegis, appearsnot to be restricted to Athena in any one of her aspectsnor with any overpoweringlogic. She can be born wearingher aegis; see F. Mainz 1963, pp. 122-125. Note the variationsof dress when derParthenon-Giebel, Brommer, Die Skulpturen

ZEUSANDOTHER DEITIES: NOTESONTWOARCHAISTIC PIERS

177

Figure3 is executedfollowingthe basiclayoutthat is establishedby the Kore on the 1974 base, at least in that partof the relief which is preservedto us. Certainfacts suggest,however,that he may be a more exact copy of a prototypethan are any of his companions.Unfortunately,Figure 3 is preservedonly in his upper half, while two probableparallelsfrom Athens preservelargelywhat the Corinthfigure lacks.5This beingas it is, the followingobservationscan still be made. Figure3 is not given a kalathosor polos, althougha polos would have filled the expanseof plainbackground abovehis head, somethingthatwas achievedby use of the polos for all the figuresof the first block and on Figure 2 of the second block. His krobylosis executedin flattenedrelief and is more stylizedthan the same lock on the Zeus Chthonios.His two curls, whichfall from behindhis ear onto his left chest, are straightas the thyrsoshe carries,havingnone of the curve that is seen in the locks of the four females.The stackedfolds of the upperhem of his himationare like nothing else to be foundon eitherof the Corinthianbases. Moreover,the forwardthrustof his chest and the three fine folds of his himationacrosschest, waist, and hip, suggesting the projectionof the hip of the weight-bearingleg, recall clearlyan Attic archaistic prototype.The forwardthrustof his upperbody is not found in the postureof his Corinthiancompanions.Figure3 (Dionysos?),then, is at variance,even if only slightly, with the otherfive andseems to be drawnfroma separateline of inspiration. The Zeus on the front face of the blockfound in 1974 and Figure2 on the front face of the blockfound in 1979 each carriesa cornucopia.The male figurefaces right, his femalecounterpart faces left. Becauseof the hieraticcompositionof the two blocks, similarin size and in design, as well as balancedin arrangement,one cannotimagine the two except in close relationship,with the cornucopia-carrying figuresfacing each other.The two basescan only have been executedas partof a singleproject. Whathad been suggestedinitiallyconcerningthe identificationof the male figure decoratingthe front of the first block now is reinforcedby the findingof the second. The beardedfigurewho holds the cornucopia,placedin predominanceamonga range of deitieson plinths,seems logicallyidentifiedas Zeus, and is, in fact, probablydepicted as Zeus in his aspectas guardianof fieldsand crops.In Hesiod,Zeus is invokedas Zeus Chthoniosto insurea good harvest.6At Delos he is offeredsacrificesas Zeus Chthonios and as Zeus Bouleus,alongwith Ge Chthonia,Demeter,Kore, Dionysos,and Semele.7 she accepts Erechthoniosfrom Ge; Cook, Zeus, p. 182, fig. 93, pis. 22-24. In Olympiashe is not always consistent in her dress while helping Herakles.She wears an aegis in the Stymphalosmetope but doffs it in the Hesperidesand Augeias metopes; B. Ashmole and N. Yalouris, Olympia,TheSculpturesof the Templeqf Zeus, London 1967, cf. pl. 153 with pls. 188 and 202. In matters of dress, the Athena on the Corinth block found in 1979 is paralleledat Corinthin an archaisticstatue of laterdate, perhapsaroundA.D. 225 (PI. 31:b). See 0. Broneer, Corinth,X, TheOdeum,Cambridge,Mass. 1932, no. 1368, pp. 117-123, figs. 111, 112. The statue is one of a pair, i.e., with no. 1348, pp. 123-124, fig. 114. This is a left, not a right hand holding drapery.The two may be caryatids,with the himationfallingfrom upperright to lower left on no. 1368; the statue in mirrorimage, representedby fragmentno. 1348, would have her himation fall from left to right. The two probablywere designed to be symmetricalaroundan architecturalaxis. 5See below, p. 181, and footnote 20. 6Hesiod, Erga, 465-476. 7SIG3, 1024; F. Sokolowski, Lois sacreesdes citesgrecques,Paris 1969, pp. 185-186, no. 96.

178

II CHARLES KAUFMAN WILLIAMS,

In the base from CorinthZeus might also be identifiedbest as Chthonianwith Ge Chthoniaas his counterpart on the secondbase.They are accompaniedby Demeterand Koreon the firstblock,by Dionysosand Athenaon the second. Athenais the one figureof the six who does not have an obviousrole amongthe agrariandeities. Her importanceas giver of the olive tree is limited in large part to Attica.8She does have other connectionswith produce,however, and she is closely coupledwith Dionysos, especiallyin the AthenianfestivalEKtppa.9 At Epidaurosthe Kissaiaimpliesconnectionsbetweenthe cult of Dionysosand Athena.'0Her agricultural festivals are the HIXvvTr'pa, fkao-oXoxpa and the Hpoxapto-nIpta." It is probable,

becauseof the companythat Athenakeeps amongthe deities of these two piers, that she is being celebratedhere for her partin the productionof cropsof the fields, less probablyas the giverof the olive. Disregardingat the momentthe functionof the two sculpturedblocks,both must have had a base and a crownmolding.A crowningelement is attestedfor both blocks by the roughlyfinishedtop surfaceand by the existenceof a cuttingfor a dowelat the centerof the top surfaceof eachof the sculpturedblocks.The newlyfoundblockhas its cuttingwell preserved,the dowel being 0.12 by 0.10 m. and 0.08 m. deep. A pour channelconnectsit withthe backedge of the block.Neitherbase nor crowningmember that mightgo with the sculpturedblockshas yet been identified.The formof the moldings is restorable,however,if only in approximation. A good parallelfor the designof the monumentwithmoldingsis a 1st-centuryaltar in the collectionof the Museo Nazionalein Rome."2This is a rectangularaltarwith a singlefigureon each of threefaces;the backface is plain.Also, the Rome altarhas its figures,in this case Charites,on individualplinths.The base moldingsof the altarare half round,cavettoand cyma reversa.The top moldingsare astragaland ovolo. Such sets are found on monumentsof the 1st centuryafter Christin Corinth,as, for example,on the podiumof the Bema, the podiumof the BabbiusMonumentand as the toichobatefor the interiorof TempleF.'3This combinationof moldings,however,does not appearat Corinthon bases designedto supportstatues;a cyma reversaor cyma rectais preferred,manytimes withoutthe half roundbelow.'4In fact, no fragmentof moldingwith half round,cavetto,and cymareversais to be foundamongthe marbles, for a statue base of any sort. A inventoriedor on the site, that might be appropriate 8Farnell,p. 293; Cook, Zeus, pp. 749-764. 9Farnell,pp. 291-292, 390-392. 10Pausanias,i.29.1; relationto Dionysos is suggested by Farnell, p. 292. "Farnell, pp. 291-292. 12B.Candida, Altarie cippinel Museo NazionaleRomano, Rome 1979, pp. 111-114, nos. 47 a-c, pl. XXXVII, dated in the second half of the 1st centuryB.C. 13R.L. Scranton, Corinth,I, iii, Monumentsin the LowerAgoraand Northof the ArchaicTemple,Princeton 1951. For the Bema, see pp. 91-109, figs. 55, 56, pl. 44; for the BabbiusMonument, pp. 17-32, figs. 9, 11; for Temple F, pp. 57-63, figs. 38, 39, pl. 23:2, 3. "4Foran idea of the rangein the moldingsof altarbases, see P. M. Fraser, RhodianFuneraryMonuments, Oxford 1977, pls. 32-91:h. For a largestatue base, 1.40 m. high, with one of the customarysets of moldings, 1926-1950, Princeton1966, p. 97, no. 226, pl. 21. see J. H. Kent, Corinth,VIII, iii, TheInscriptions

DEITIES: NOTESONTWOARCHAISTIC ZEUSANDOTHER PIERS

179

conservativerestorationof moldingsfor the two sculpturedblocks might be a base plinthwithcymareversaor cymarecta.The crownprobablywas an ovolo withcavetto. Whicheverof the two types of base moldingsis restoredto the sculpturedblocks, one cannotrestorethe whole as less than two meters tall. Such a height, along with considerationof the generalproportions,makes unlikelythe identificationof the two monumentsas companionaltars. What purpose,then, did the sculpturedshafts serve? In the initialpublicationof the 1974relief, the suggestionwas put forththat the blockfeaturingthe relief of Zeus The findingof a supportedthe statue of Zeus Chthoniosmentionedby Pausanias.15 second block, its front face decoratedwith a representationof Ge Chthonia,raises seriousquestionsaboutthe validityof the initialtheory,since, accordingto Pausanias, the statue of Zeus Chthonioswas one of three statuesof Zeus set up, apparently,in close proximityone to another.It is possiblethat the two blocksnow underdiscussion supportedtwo of the three statues,but it is not as probablethat both wouldhave been decoratedwith a numberof deities all specificallyrelatedto the fertilityof the fields when only one of the statuesof Zeus was to be associatedwith this realm. If the shaftsare sculpturedbases for freestandingstatues, they shouldbe restored to some positionclose to the wallof a buildingor againsta column,for the backsof the piersare not decoratedwith relieffiguresas are the sides and fronts.A temptingposition is a placealong the front, or north, wall of the Long RectangularBuildingwhich definesthe forumbetweenthe South Stoa and the West Shops.'6This structurehas a large,raisedterracein frontof it uponwhichsuch monumentscould have been set. It is here, precisely,where the first blockwas found. No tracesof foundationsfor bases are preservedon the terrace.This is quiteunderstandable, however,for the terracehas been the victim of continuousbothrosdiggingfrom the EarlyChristianperioddown into the 11thand 12thcenturiesafterChrist. An alternativefunctionfor the two blocks is their possibleuse as piers in some buildingof the Romanforum.If the blocksare to be restoredto one of the podium templesat the west end of the forum,then one mightenvisionthe piersin the portico of TempleF or G, if not TempleD at the northwestcorner.These temples,however, are restoredwith columns;one may be portrayedon Corinthiancoins with four columns, not piers,acrossits front.'7The use of the carvedblockson the facadeof one of the temples, even with the additionof a plinthbetween stylobateand base molding, wouldgive extremelysquatproportionsto the facade;the templewouldappearlow and ill proportioned besideany of its columnedneighbors. The Dionysionis the one buildingat the west end of the forum into which the piers in questionmight be fitted withoutignoringthe archaeological evidence.18It is 15Pausanias,ii.2.8. Williams, op. cit. (footnote 1 above), pp. 25, 29. "6C. K. Williams, II, "Corinth, 1975: Forum Southwest," Hesperia45, 1976, pp. 127-135. 17Scranton,op. cit. (footnote 13 above), pp. 68-69, pl. 26. 181bid.,pp. 85-91; Scranton dates the Dionysion "among the earliest structures on the line of the central terrace ... " (pp. 124-132). But pottery is not an aid in the search for the precise date of construction. Scrantontries to relate the stylobate level of the Dionysion with the leveling of the forum itself

180

II WILLIAMS, KAUFMAN CHARLES

sitedat the westernend of the line of low buildingswhichdividesthe upperlevel of the forumfromthe lower. The Dionysion is not a canonicaltemple; ratherit has a central apsidalroom room.The threeroomsare unitedby a flankedon eitherside by a smaller,rectangular commonporticothat opens to the north.It has in frontof it the foundationsof a rectangularaltar.The portico,to whichone mightassignthe two piersunderdiscussion,is designedwitha singlestep, a columnat eitherend and four piersbetweenthe columns. The easternand westernpiersare compositein plan, but the two centralpiers,whose positionand shape are attested by trimmedbeddingson the stylobate,are 0.85 m. square.Eachbed has two dowel cuttingsand pourchannelsfor leading.One thus can restorea rectangularplinthabout0.80 m. squareto the stylobate,a base moldingon the plinthwith a sculpturedpier on that, then a crowningmoldingand epistyle.With the use of the rectangularplinthat the bottomof the pier one can achievea clearance underthe architraveof 2.50 m. or more. The smalldiameterof the end columns,attestedby the beddingson the stylobate, is no more than0.40 m. at base. Sucha smallcolumndemandsthe restorationof a low roof at the ends of the building,with its epistyleno higherthan 3 metersfrom stylobate.By placingthe sculpturedpiersat the centerof this porticoone is able to restorea continuousepistyleto the facadeat the approximatelevel of the epistyleof the shops immediatelyeast of the Dionysion,withwhichthe Dionysionis aligned.19 In the firstpublicationof the 1974blockit was suggestedthat the date of execution mighthave been as earlyas the Augustanperiod.This was consideredpossiblebecause of the completeabsenceof the drill, as well as on pointsof style. The-block found in (pp. 125-126). Tests in 1980 in the northeastcorner of the forum tentatively point to a date in the Flavian period for the marble paving of the forum; the Dionysion could, however, possibly have predated the paving. What does suggest that the building is later than Julio-Claudianis its plan. Cellas with apses or niches are attested in Rome in the Augustan period. See B. Tamm, Auditoriumand Palatium,Lund 1963; also J. C. Balty, AntCl33, 1964, p. 575. See Tamm, p. 160, no. 10, Temple of MarsUltor after A.D. 2, and, possibly, p. 150, no. 6, Temple of Venus Genetrix. The Corinth Dionysion is a different concept in plan, however, being slightly more than a semicircle, not a rectangle with niche or apse. For the difference in temple plan, comparethe Dionysion with Temple F at Corinth, the latter dated at its earliest by P. Gros to the Augustan period. It seems, however, more unlikely now than when Scranton dated it that it can be Augustan. See P. Gros, Aurea Templa,Recherchessur architecturereligieusede Rome a I'epoquedAuguste, Rome 1976, pp. 124-143. Temples with cellas that have curved back walls closer in style to the Corinth building are found in the East, but are dated around A.D. 50 or later. See D. Krenckerand W. Zschietzschmann,RdmischeTemV, Berlin 1938: kieinerBezirkat Hossn Soleiman (exedra), pl. pel in Syrien,DenkmdlerantikerArchitektur 38; Rahle, pl. 95; Burkush, pl. 100; for chronology see pp. 271-275. The closest parallelto the Corinthian Dionysion both in size and plan is the inner room of the temple at Rahle. Regardlessof whether or not one restores to the Dionysion the sculpturedblocks in question, the facade with central element of heavy piers and small flankingcolumns is difficult to parallelin any early Roman public architecture.Note, also, that the Dionysion faces north, suggesting that the building may have been plannedafter the east end of the forum had taken its shape, and thus the Dionysion sited as it was because no space was availablefor it on the terrace with the temples facing east. There are serious difficultiesin maintainingan Augustanor even a Claudiandate for the constructionof the Dionysion. 19Scranton,op. cit. (footnote 13 above), pp. 114-115.

DEITIES: NOTESONTWOARCHAISTIC PIERS ZEUSANDOTHER

181

1979 suppliesnew materialby which one can examine the figures stylistically.The Dionysos of the 1979 block is closely paralleledby two figuresin Athens and one in Rome.20The Corinthfigureis the stiffestof the group,perhapsthe result of copying from a cartoonor from a desireto enhancethe antiquityof the subject.His stiff beard is broughtto a pointin frontof him; his krobyloshas been flattenedand shapedinto a triangle.Whenthis krobylosis comparedwith thatwornby the Zeus of the 1974block, one is inclinedto see the addedstiffnessof the Dionysosmore as a personalpreference of the sculptorthanas a chronological criterion. Two pointsmay help to clarifythe problemof chronology,however.The first is a deductionthat can be madefrom the architectural remains.If the two sets of archaistic figureswere set up on the terracein front of the NeronianLong RectangularBuilding, their date of erectionand dedicationmust be Neronianor later, or, if one acceptsthe possibilitythat the blockswere designedas piersin the Dionysion,then, too, logic has it that the blocks were commissionedin the time of Nero or later, probablyin the Flavianperiod.The second point is a deductionmade from the historicalfacts.21The intellectualclimatemightnot have been rightuntil the time of Nero for the erectionof two large monumentscelebratingthe glory of the Greek past of ColoniaLaus Julia Corinthiensis.We do know, however,that by the time of Hadrianit was prestigiousto have "becomethoroughlyhellenized,even as your own city [Corinth]has."22Although archaisingis a style that appearsin one aspector anotherthroughoutGreekart, the use of it on two largepiersas partof the designof a templeor on a pairof monumentsin frontof a buildingdedicatedby a sacerdosof Corinthsuggestspridein declaringa preRoman pedigreefor some officialCorinthiancult or institutionin the Roman forum ratherthan pridein the Romanroots of the population.23 From the logic of these arguments, then, one mightbest placethe two piersin the Flavianperiodand not earlier. CHARLESKAUFMANWILLIAMS,II CORINTHEXCAVATIONS

20ForZeus of the Four Gods Base in the Athenian AcropolisMuseum, see Harrison, op. cit. (footnote 3 above), pl. 64:c. For Aigeus of the tripod base in the Athenian Agora, see ibid., pp. 79-81, no. 128, pl. 30, right. For the Roman example, see W. Fuchs, Die Vorbilder der neuattischenReliefs, JdI-EHXX, Berlin 1959, p. 46; also E. D. Van Buren, "Praxias,"MAAR3, 1919, pl. 72:2, which shows the complete figure of Zeus, close in pose to the Corinth Figure 3, but not exactly alike in details. 21E.L. Bowie, "Greeks and their Past in the Second Sophistic," Studiesin AncientSociety,M. I. Finley, ed., London 1974, pp. 166-209. I thank most warmly Mr. A. Spawforth,Assistant Director of the British School at Athens, for this reference and for sharingwith me his considerableknowledge in this area. 22Dio Chrysostom, The Thirty-seventh Discourse:The CorinthianOration,Loeb ClassicalLibrary,Cambridge, Mass. 1946 (trans. H. L. Crosby), p. 26. 23Foran example of a totally Roman treatmentof the subject of prosperity,especiallyagrarian,note a rectangularaltar, 0.50 m. high, from Naples (storeroom inv. no. 147827). It has a single figure on each of its four faces, each making a libationwith a phiale; two of the figures also hold cornucopias.The representations are of Bona Dea, Vesta, Mercuryand Genius, who is Roman even to his resemblanceto Augustus. See Adolf Greifenhagen, "Bona Dea," RbmMitt52, 1937, p. 228, pls. 51, 52.

PLATE 30

F7-

ilk~i' M~L.

.*

)Q

i__av~~~~~~~~~~~~~~~~ Ir~~~~~~~~~~~~~~~~~~~~~~

0Y

PLATE 31

-Jil-

-x2~~~~~~~~~~~~~I

co

Room--

,0

~~~~~

Nl

-~~~~~~~~~"

_< -

_r

D_>

@

i;=

Q~~~~~

-

2~~~~~~ E~~~~~~

PLATE 32

a.-,

Wrap~~~~~~~~~~~~~~~~~~~~a WI

.s~~~~~~T

t~~~~~~~~~~~~~~ki

1?

|

~~~~~~~~~~~~~~~~~ -~~~~~~~~~~~~~

_

PAUSANIASAND PRAXITELES to this volume it will give me specialpleasureif I may at the N CONTRIBUTING same time pay tributeto anotherold friend.HomerThompsonand I have walked the streets of Athens with Pausaniasfor over half a century,found him a congenial companionin spite of faultsand eccentricities,and become constantlymore awareof and uniquevalue as a guide, and of the dangersand pitfallswhich his trustworthiness one is aptto encounterif one abandonshim to followa pathof one's own. a Withsome trepidationI take an examplefrom sculptureratherthan topography; seriesof quotationswill introducethe subject. Laterthey set up otherstatuesin the Heraion,includinga Hermesof stone, carryingthe infantDionysos;this is a work (techne, a specimenof the art) of Praxitelesa bronzeAphroditeis a work(ergon) of Kleonof Sikyon. v.17.3. Pausanias,

This followsthe mentionof other statuesin the Heraionat Olympia,which Pausanias calls "very old".

It has sometimesbeen said that by using the word technePausaniasis sayingonly that the Hermesis in the style or mannerof Praxiteles;but an examinationof other instances'showsthathe uses technesimplyas a synonymor stylisticvariationfor ergon. The groupwas foundin the Heraionon the 8th of May 1877and afterone or two initialexpressionsof doubtwasgenerallyacclaimedas a masterpieceby Praxiteles. Casts give no conceptionof the soft, glossy, flesh-like,seeminglyelastic -surfaceof the original,whichappearsto glowwithdivinelife. Lookingat the original,it seems impossibleto conceivethat Praxitelesor any man ever attainedto a greatermasteryover stone thanis exhibitedin this astonishing work. v.17.3 (London J. G. Frazeron Pausanias, andNew York1898). This is the only case in whichwe possessan undisputedoriginal,straight fromthe handof one of the greatestmastersof antiquity. E. A. Gardner,Handbookof QreekSculpture, 2nded., London1915,p. 388.

Berlinand Leipzig1927, pp. 37-48) mainBildhauerarbeit, {CatlBliimel(Griechische the Hermes was not made by Praxiteles that evidence tained on dietailedtechnical himselPtutwas.iaRomancopy. 'Notably .143;,v17.1-2.

PAUSANIAS ANDPRAXITELES

183

AJA 35, 1931 publishedarticlesby six scholars-on the subject,includingBluimel and, for example,the following: It is not a questionwhetherPraxiteleshimselfhada handin it; but whether the Athenianmarble-cutter who copiedthe bronzeoriginallived in JulioClaudianor Hadrianic times. RhysCarpenter,p. 261. If the Hermesis a copyit is one of the best we have;if it is an original,then Praxitelesis a disappointing sculptorwho inventedsome most remarkable technicaltricks. StanleyCasson,p. 268. The stylisticandtechnicalchargesagainstthe Hermesas a workby Praxiteles cannotbe substantiated. The evidencein favor of the identification is too strongto be affectedby a few unusualfeaturesfor all of whichexplanations or parallelscanbe found. GiselaM. A. Richter,p. 290. The parallelsfor the draperyof the Hermesfoundon worksof the fourth centuryB.C.are closerthanthose on Romanworksand thereforethe drapery supportsthe view thatthe Hermesis a Greekoriginal. ValentinMuller,p. 295.

0. Antonsson (The PraxitelesMarble Groupin Olympia,Stockholm 1937) put for-

wardthe theory,basedon technicalevidencefrom the groupitself and representations in minorarts, that Praxitelesmade the groupas a Pan carryingDionysosaccompanied by a nymph,and afterdamagethis was reworkedin imperialRomantimes. This theory has not been receivedwith muchfavour,thoughit is commonlyacceptedthat thereare signsof more superficialreworking. CarlBluimel(DerHermeseinesPraxiteles,Baden-Baden1948) changedto the view thatthe groupis a Hellenisticoriginal,perhapsby a HellenisticPraxiteles;RhysCarpenter in AJA 58, 1954, pp. 4-6 dismissesthis latter-dayPraxitelesas a fictionand insists on a Romanimperialdate~. The styleandtechniqueof the draperyof the Hermesconfirmthe suspicions arousedby;other features-the hair, the polish, the child, the sandal,the pedestal-thatthe statuecannothavebeen carvedin the fourthcentury.The difficultyof findinggood later parallelsdoes not alter the impossibilityof assigningit to a date duringwhichwe know that sculptorswere still employinga universalstyleandtechnique. The manwho carvedthe Hermeswas a first-classsculptor.He may have copieda work by Praxiteles,his name may have been Praxiteles,or Pausaniasmayhavegot his factswrong.It wouldhelpif we couldfix the dateof the statueat all accurately,but for this a detailedstudy-of'thtechnique of

184

R. E. WYCHERLEY later periods would be necessary.However the base and those statues which furnishsome parallelsto the Hermes indicatea date c. 100 B.C. Sheila Adam, The Techniqueof Greek Sculpturein the Archaicand ClassicalPeriods, London 1966, p. 128. It is a fine Hellenisticcopy. This sad but importanttruth is argued irrefutably by Sheila Adam. Peter Levi on Pausanias,v.17.3 (New York 1971). Whatever grounds have been or may be advanced against accepting the Hermes as an original by Praxiteles, they will be inconclusive so long as no unquestionableoriginalof his time, equal in qualityand as well preserved, is availablefor comparison.... Even those of us who dislike the Hermes must admit that it has remarkable vitality. A. W. Lawrence, Greekand RomanSculpture, London 1972, p. 185. One might say that the Aberdeen head gives the impression that if pressed by a finger its flesh would give, but the flesh of the Hermes appearsfrozen It is to me inconceivable that the Hermes was carved by a sculptor of the ability of the one who carved the Aberdeen head or indeed by any reputable sculptorof the mid-fourthcentury B.C. R. M. Cook, "The Aberdeen Head and the Frank Hermes of Olympia,"Festschrift.fuir Brommer,Mainzam Rhein 1977, p. 77. Few of those who consult the well-compiled and authoritatively written handbookson Greek art have any suspicion that the proud edifice of Greek sculptural history is reared on a quagmire of uncertainty, ambiguity and baseless conjecture.It could not be otherwise. The ancient statuarywhich has survived into modern times is largely anonymous; it carries no label to tell us what it is or whence it came. Rhys Carpenter,GreekSculpture,Chicago 1960, p. vii.

These excerptsrepresentfairlywell the vicissitudesof the Hermes since his disthe first half-centuryhe was placed coveryjust over a centuryago. For approximately on the highest of pinnacles.In the second he has been toppled.Not only has Praxiteleanauthorshipbeen questionedand denied, but simultaneously,in accordancewith sculpturaltastes,his artisticqualityhas been decried.He has not even mid-20th-century been set securelyon a moremodestpedestal;he seems at times to lie flat on his face in Rhys Carpenter'squagmire,as once he lay in the disintegratedbrickworkof the Heraion.

ANDPRAXITELES PAUSANIAS

185

But perhapsRhys Carpenteris too severe. The historyof Greek sculptureis based on foundationssolid enoughas far as they go: firstlythe informationgiven us by reliable ancientwriters,of whom Pausaniasis by far the most important;secondlythe sculpturewhich happensto survive.The troubleis that besides being very far from complete, these two are largelyseparate,with only sporadicpoints of contact.The "quagmire"has been createdby those who tramplearoundon the soft insecureground in between. For manyyearsthe Hermeswas consideredthe firmestof these pointsof contact. At the same time he was exceptionalin that he carrieda clearlabelof origin,affixedby an authoritativeand trustworthyhand. But now a series of criticshave made it their businessto tear it off and to substitutenot one but half a dozen labels of their own devising.ApparentlyFrazerand the earlierwriterswere deluded.Pausanias,alongwith thousandsof others,was duped,not havingthe wit to see that the Hermeswas a substitute-an unacknowledged substitute-lackingthe quality,the true Praxiteleantechne, whichhe saw in the Satyr,the ArtemisBrauronia,the Leto and manyotherworks. PeterLevi considersthe mattersettled.And indeedone mightbe gladto settle for the next best thing,a trulyfine Hellenisticcopy (thoughhow, if the originalis lost, one can tell that this is a fine copy,as distinctfrom a fine piece of sculpture,I fail to see). In fact, the studentwho, when told that the Hermesis not a genuinework of Praxiteles' hands, very properlyasks, "Then preciselywhat is it?" is confrontedby the followingpropositions(more,for all I know): The Hermesis a Hellenisticcopy (fine or not so fine or poor-aestheticjudgments varywidely). It is a Hellenisticoriginal,possiblyby a latter-dayPraxiteles(the laterBluimel). It is a copyof Romanimperialdate. The originaltoo was of marble. The originalwas of bronze. The workwe have differsin importantrespectsfromthe workof Praxiteles,though the extent of the differenceis not agreed. It was made by Praxitelesas a Pan but later after damagewas convertedinto a Hermes(Antonsson). Ourstudentmay well feel that afterpartingcompanyfrom his faithfulguide he is flounderingin the "quagmire".He may have doubtsaboutthe methodswhichproduce such bewilderingresults. If I now take a closer look at SheilaAdam'sdiscussionof the Hermesin the appendixto Technique (pp. 183-184 above), let it not be thoughtthat I am beingparticularlycriticalof her-quite the contrary.One learnsa greatdeal fromher excellentbook; her expositionof the case againstPraxitelesis thoroughand duly cautious.But I cannot accepther uncompromising conclusionquotedabove, in particular her curt dismissalof our prime authorityPausanias.Her statementthat we know that in Praxiteles'day sculptorsemployeda universalstyle and technique,I cannotunderstand.

186

R. E. WYCHERLEY

Mrs. Adam says merely that "suspicions"are arousedby variousfeatures;and afterfull discussionshe admitsthatfor all these one is in difficultybecauseof a "lackof It couldnot be otherwise.We have only a fractionof suitablematerialfor comparison." "classical"5th- and 4th-centuryGreek sculpture;and of that only a fractioncan be datedabsolutely,objectively.And what has survivedis scrappy,fortuitousand highly nearlyall the very best is lost. Of the individualmasterIn particular, unrepresentative. pieces of the great sculptors,of the choice worksselected by Pausaniasfor mention, only one here and thereremains. Pausaniasmentionsover 160 sculptors,most being of the Classicalperiod.Of how statue?Of not one do we have the kindof manydo we have one single authenticated solid oeuvrewhich historiansof art in other periodsvery rightlyrequirebefore they presumeto attemptstylisticanalysis,or attributionand rejection.If we had one hundredthof what Pausaniassaw, or just two or three statuesfrom the handof each of a that would make!The majorsculptors, score of these sculptors,what a transformation above all, were men of supremegeniusand technicalvirtuosity.Whatexper'Praxiteles iments, whatspecialstylisticvariations,whattechnicalinnovations,whetherephemeral or more permanent,are lost to us throughthe vanishingof their workwe can never know. In his excellent essay on the frieze of the Nike balustradesRhys Carpenterhas shown how even in the workof a closely knit, homogeneousgroupone finds striking variations.Severalof the sculptorsare first-rateartists.One shows odditiesamounting to faults,and it has even been suggestedthat his workis a late replacement;Carpenter for our presentsubject,thatit is doubtful will not have this, but he tells us, significantly whetherany expert,withoutknowledgeof provenance,wouldplacethis workcorrectly in periodand school. As CharlesMorganhas remindedus with referenceto Paioniosand Alkamenes,3 we also have to bear in mind that some artistslive to a great age, still plyingtheir crafts,and in the courseof a long workinglife remarkabledevelopmentsand changes take place in their style. Morganhas shown convincinglythat there is no compelling need, on groundsof date or style, to say, as many have done, that Pausaniasmustbe wrongwhen he gives the pedimentsat Olympiato Paioniosand Alkamenes.As promising youngmen they may have been given this assignment,undera suprememaster, and with a groupof subordinates,each with his own artisticpersonality,to share the carving.

On all counts,when discussingthe style and techniqueof the lost masterswe must makeampleallowancefor the meagreness,sometimessheer absence,of clearevidence andfor the consequentuncertainties,and recognizethatwhatwe say is tentative,provisional,indecisive.We are constantlyengagedin a kindof skiamachia,in both senses of the word,fightingin the darkandfightingwithshadows. TheSculptureof the Nike TempleParapet,Cambridge,Mass. 1929. 3"Pheidiasand Olympia,"Hesperia21, 1952, 308-312. 2

PAUSANIAS ANDPRAXITELES

187

Mrs. Adam continues, "This (lack of comparablematerial)is not true for the drapery."This is not the impressionwhichone receivesas one readsher account.What I have said appliesto the drapery-in this too one can expect endless variation.Mrs. Adamcites the ApolloPatroos,probablyby Euphranor,from the Athenianagora.'But that, unlike the Hermes,is a greatcult statue, standingformallyrobed in full dignity. Mrs.Adamsearchesfor parallelexamplesof 4th-centurydraperyand producessix; and it is curiousthattwo othersare the Mausolosand Artemisiafromthe Mausoleum.Now her most powerfulally in takingthe Hermesfrom Praxitelesis probablyRhys Carpenter; and RhysCarpenterinsists5that the "Mausolosand Artemisia"too must be downdated by two centuries,largelybecauseof the "unclassical"style of the drapery.Our studentmay againfeel disconcertedand bewilderedwhen he sees that a very important workwhichhe thoughtwas firmlyanchoredin a historicalcontext is on high authority castadrift;and thattwo leadingproponentsof the theorythatthe Hermesiis a late work contradictone anotherflatly,aboutjust the kindof evidenceon which,theirwhole case depends. In fact we find very frequentdisagreementaboutthe interpretation and use,of the technicalevidence.Againand againMrs.Adamfindsit necessaryto correcther distinguished predecessor,StanleyCasson.On her first page she disagreeswith him completelyabout the use of tools in the early Archaicperiod.One might expect that the marksleft on the stone by varioustools would provideprecise,objectivecriteriaand lead to agreement.It is alasfarotherwise.The use of the "runningdrill"offersa crucial example.Pausanias(i.26.7) tells us that Kallimachos,who was called"katatexitechnic" becauseof the elaboratenessof his art and who workedin the latter part of the 5th century,"firstdrilledstones."This is not literallytrue, but there may be somethingin it. It may be that Kallimachosinvented the use of the "runningdrill". Of modern authorities,some have detectedtracesof the runningdrillalreadyin Parthenonsculpture. Mrs. Adam (p. 66) tells us, "The new techniquemust have been introduced between370 and 350."How can we know,when from the handof Kallimachoshimself we do not have a singleauthenticated work? Othercriteriaare inevitablyless precise.Mrs. Adamwarnsus that the terms "linear" and "plastic",much used by Rhys Carpenterand others in referenceto drapery styles, are "so vague that they are open to misinterpretation." Some criteriaagainare largelysubjectiveand personal.Not that subjectiveimpressionsshould be wholly despised-"transcendingany reasoningconcerningtools, supports,tree-trunksand whatnots," says Miss Richter,6"standsour 'subjective'reactionto the Hermes."Casson7 condemnsthis as "archaeological fundamentalism", which makes the searchfor evidence futile.One mayreplythat, confusedand drivento despairby the conflictingideas producedby 50 years'searchfor evidence, some of us are constrainedto fall backon 4H. A. Thompson, 'Apx'E, 1959, pp. 30-34. ' GreekSculpture,Chicago 1960, pp. 213-215. 6Quoted by Casson, "The Hermes of Praxiteles," AJA 35, 1931, p. 267. 7Ibid., p. 268.

188

R. E. WYCHERLEY

our subjectivereactions-and our Pausanias.Pausaniashimself was no doubt largely guidedby personalimpressionsin acceptingattributionsand occasionallymakingthem; of course he had the advantageover us in that his impressionswere drawnfrom the whichhe hadseen. hundredsof masterworks We may be thankfulthat anotherstatue long regardedas a 4th-centurymasterpiece, the Demeterof Knidos,has been rescuedfrom the fate of the Hermes.She too to the Hellenisticperiod;the draperyin particular was downgradedby Rhys Carpenter8 he regardedas "completelyunclassical".Now BernardAshmole9has effectivelyreinstatedher (I trust). I can make no pretenceof doingthe same for the Hermes.It is beyondme, as an earnestbut inexpertstudentof sculpture,to treatin detailthe varioussubtlecriteriaof style and technique.My properconcernis with Pausanias,the meaningand importance of whathe says and how we shouldtreathis evidence.He tells us clearlythat the Hermes is the workof Praxiteles;and from this we may safelyinfer that it was presented and acceptedas such at Olympia.Whatthe rejectorsseldomif ever tell us is that they are askingus to believethatthe Hermesis not merelya copybut a fake.Pausaniaswas awarethat a masterpiecemightbe removedand a copy or substituteprovided.He tells us what happenedin such circumstancesat Thespiai (ix.27.3).

Pausaniasis a reliableauthor.Frazerdemonstratedhis reliabilityin archaeological matterseightyyearsago and predictedthatthis wouldbe confirmedby furtherresearch. His predictionhas been wonderfullyfulfilled.'We normallytake it for grantedthat a plainstatementby Pausaniasis correct;if we did not we shouldhave difficultyin proceedingwith the studyof the topographyof the Greek sites, or the worksof the great Greek sculptors.However much he is ignored, however often he is brusquelydismissed,he will not go away.His testimonystandsandcarriesgreatweight.Unlikethose who contradicthim, and one another (occasionallythemselves), he walkedon firm ground,and in the light of day. To challengeand disprovea statementby him one needs to be sure of one's ground,to have clear,solid, ancientevidenceand agreement aboutit. Of coursehe goes wrongfromtime to time; but seriousprovenerrorsare few. Moreoften he does not tell us as muchas we wouldwish, or his accountis not entirely clearand precise. Pausaniasis not very cleveror subtle.He has no thesisto argueand prove,no axe to grind.But he is a man of good sense, of good judgmentand good taste, notablyin Lucian,the gift of facileselfmattersof art, thoughhe had not, like his contemporary expression.Frazerin his introductionclearlydemonstratedthe qualityof his treatment of the sculptors,but I shouldlike to add a few furthernotes. To his credithe shows great appreciationof the art of the earlier periods; in this he is closer to modern taste

than to Frazerand the Victorians.The worksassociatedwith the legendaryname of Daidalos,he says, are &To rcOaTEpa,but have a touch of the divine (II.4.5). He partic8Op.cit. (footnote 5 above), p. 173. 9"SolviturDisputando," Festschrftfir FrankBrommer,Mainz am Rhein 1977, pp. 13-20.

ANDPRAXITELES PAUSANIAS

189

ularlyadmiredcertainsculptorsof the firsthalf of the 5th century.Pythagorasof Rhegion, he says (vi.4.4), was a good sculptorif ever there was one. Onatasof Aigina wouldmeanvery little to us but for Pausanias,who singlesout half a dozen remarkable worksfor mention,sayingthatone is a marvelfor size and techne,and that the sculptor is inferiorto none of the followersof Daidalosand the Attic school (v.25.13;viII.42.7). Pausaniassays nothingof the Aiginetanpedimentswhichfor us representthe best of early5th-centurysculpture.In fact he mentionsonly seven sets of pedimentalsculpture and gives creditto the sculptorsof only three. These three includethe Olympia pediments,whichhe describesin some detail,but not the Parthenon.There he merely notes the subjectsof the two pediments,and is silent aboutthe metopesand the frieze, on Pheidias'greatchryselephantine cult statue. concentrating We put it downto good tastein Pausaniasthathe considersbest worthseeingof all the worksof Pheidiasnot the Zeus of Olympianor the Parthenosnor yet the colossal Promachosbut a much more modest and apparentlymore subtle bronzeAthena, the Lemnia.Lucianagreeswith him, and is more articulatein describingher beauty (Eikones, 4), noting in particularTnoarraXovof the cheeks. Now arraXovmeans soft and yieldingto the touch;it wouldnot be used of a formwhich,howeverexquisite,looked rigidor "frozen".So, we mayinfer, Pheidiastoo, waybackin the 5th century,had this trick of the trade at his disposal,and could use, it on suitablesubjects.The Lemnia madethe same impressionon Lucianas the Hermeson Frazerand the Aberdeenhead on ProfessorCook. Pausanias'treatmentof Alkamenesis peculiarand illustratesboth his value and his shortcomings.We have alreadylooked at the Olympiapediment;it is in connection with this that he tells us that Alkameneswas second only to Pheidiasin his skill in makingstatues (v.10.8). The Aphroditein the Gardens,he says (X.19.2),was amongst the thingsbest worthseeing at Athens;once againit is Lucianin the Eikones(6) who gives detailsof the beautyof this work. In ii.30.2 Pausaniastells us that Alkamenes createdthe sculpturaltype of the three-bodiedHekate, makingthe statue of the goddess on the bastionnear the Templeof WinglessVictory.Curiouslyhe does not mention this shrineand statue in his accountof the Acropolis.He mentionsthe Hermes Propylaioswhich stood near by, but he does not name the sculptor;it is generally inferredfrom the inscriptionon a fine HermesPropylaiosat Pergamon,assumedto be a copy, that the sculptorof this too was Alkamenes,but becauseof Pausanias'silence we cannotbe quitesure. In the templeof Hephaistos,Pausaniastells us, the god stood with Athenabesidehim. Almostinevitablythis Hephaistosis usuallyidentifiedwith the very fine statueof the god by Alkamenes,seen by Ciceroat Athens,10and naturallywe 1ODenaturedeorum,1.30. I cannot accept Miss Evelyn Harrison'stheory that the "Theseion" was occupied not by Athena and Hephaistosbut by Eukleia ("Alkamenes' Sculpturesfor the Hephaisteion,"AJA 88, 1977, pp. 139, 421-426.). The impossibilityof findinganother site for the HephaisteionI leave to those who have repeatedlyworked over the area. I am more concerned that Pausanias,though he might have spoken more clearly about the position of the Hephaisteion, places the temple of Eukleia very clearly southeast of the agora (i.14.5; for his use of arrwTE'pw see especiallyii.3.2-6), and Miss Harrisonimputes

190

R. E. WYCHERLEY

tend to assumethat the same sculptormade the Athena.A wordfrom Pausaniascould he does not namethe sculptor. have settledthe matter,but unfortunately is the groupof motherand childfoundon the Acropolis,idenMost problematical tifiedby nearlyeveryoneas the Prokneand Itys seen by Pausaniasand said by him to be a dedicationof Alkamenes(X.24.3).We naturallyassume that this is the famous sculptor,but we cannotbe certain.And did he makehis own dedication?Is it worthyof the second greatest?Opinionsvary enormously.Some have said that it is comparable with the Parthenonsculptureor the Caryatids,othersthatit is very faultyin designand unevenin execution.Some say it is the workof a pupil,or a copy.It certainlydoes not providea safe aphormefor studyof the style of Alkamenes."1 Pausaniasshowsus thatAlkameneswas a very greatsculptor,who in variousways carriedon the workof Pheidiasand, like him, addedto the acceptedideas about the gods. But at several points the incompletenessor unclearnessof his accountmakes troublefor us. There is no sufficientreason,however,to believe that anythinghe says is untrue;and it is exceedinglyodd to takefromAlkamenesthe fine Olympiapediment, definitelyascribedto him by Pausanias,and to foist on him the very dubiousProkne. Pausaniasis less than enthusiasticabout the 4th-centurysculptors.He recognizes the importanceof Praxitelesand mentionsover twentyworks (only Pheidiasis better represented),and he must have been thoroughlyfamiliarwith the sculptor'soeuvreand techne;but he awardsno special mark of commendation.He saw an Aphroditeby Praxitelesat Thespiai(Ix.27.5),alongwith the Eros (andone mightinferfrom1.1.3that he had been to Knidos);but yet againit is left to Lucianto revealthe seductivecharm of a PraxiteleanAphrodite(Eikones,6). To returnto Hermesin conclusion,whatwe have beforeus is not simplya dispute betweenmoderntechnicalexperts.The oppositionis ratherbetweena plainstatement by Pausaniasand half a dozen divergentmoderntheories, of which all but one are boundto be mistaken.Of courseI acceptthat it may be Pausaniaswho is wrong.But the case againsthim is certainly"not proven".I am not saying "case dismissed",in spiteof the deficiencyof evidenceand lackof agreementamongthe witnesses.But it is wrongto set aside Pausanias'evidence, here or elsewhere.He may not have knowna raspfrom a runningdrill, thoughwe are not entitledto make this assumption,but he had a colossaladvantageover us who have never stood before an ancientstatue and to him a flagranterror-the transpositionof an importantmonument to a wholly wrong site-such as is not found in all his descriptionof Athens or as far as I know anywhereelse. To what I wrote in "The Temple of Hephaistos," JHS 79, 1959, pp. 153-156, I would add that the east frieze of the "Theseion" is peculiarlyappropriateas an introductionto Hephaistosand Athena, showing "the sons of Hephaistos taming a land untamed" (Aeschylus, Eumenides,13-14; H. A. Thompson, "The SculpturalAdornmentof the Ilephaisteion," AJA 66, 1962, pp. 341-344.) "1PaceW. Schuchhardt,Alkamenes,Berlin 1977, pp. 9-22. A possibilitywhich occurs to me is that Alkamenes began the group, made a dreadfulmistake in carvingso that the boy appearsto be partlycarved out of his mother's leg, handed over the work to an assistant to finish (the back is said to be inferior), then, unable to dispose of the group otherwise but unwillingto scrap it, dedicatedit himself, unsigned, to Athena, with a vow to do better next time.

PAUSANIAS ANDPRAXITELES

191

been able to say, withknowledge,with true confidence,"Thiswas madeby the handof Kalamis" -or Myron,or Pheidias,or Alkamenes,or Polykleitos,or Skopas,or Lysippos, or-if we takefromhim the Hermes-Praxiteles. Whatin the end is our studentto do? He may find some reassurancein whatwas said to me by one in whose knowledgeand soundjudgmentI place the utmost reliance-reaffirminghis belief that the Hermesis by Praxiteles,he addedthat the question is not afterall so very important,since the qualityof the statue is so outstanding that even if it were a copy it couldwithoutmore ado standin for the original.My own lastword-stay with Pausanias. Heald Green, Cheadle Cheshire, England

R. E. WYCHERLEY