INSCRIPTIONS FROM PANAKTON

d i n i n g i n t h e s a n c t ua r y o f d e m e t e r a n d k o r e Hesperia Th e J o u r nal of the Amer ic a n Sc ho ol of Cl assi c al S t udie s at Athens Vo l u m e 9 0 2021 Copyright © American School of Classical Studies at Athens, originally published in Hesperia 90 (2021), pp. 281–337. This offprint is supplied for personal, non-commercial use only, and reflects the definitive electronic version of the article, found at <https://www.jstor.org/stable/10.2972/hesperia.90.2.0281>. 1 hesperia Jennifer Sacher, Editor Editorial Advisory Board Carla M. Antonaccio, Duke University Effie F. Athanassopoulos, University of Nebraska-Lincoln Angelos Chaniotis, Institute for Advanced Study Jack L. Davis, University of Cincinnati A. A. Donohue, Bryn Mawr College Jan Driessen, Université Catholique de Louvain Marian H. Feldman, Johns Hopkins University Gloria Ferrari Pinney, Harvard University Thomas W. Gallant, University of California, San Diego Sharon E. J. Gerstel, University of California, Los Angeles Guy M. Hedreen, Williams College Carol C. Mattusch, George Mason University Alexander Mazarakis Ainian, University of Thessaly at Volos Lisa C. Nevett, University of Michigan John H. Oakley, The College of William and Mary Josiah Ober, Stanford University John K. Papadopoulos, University of California, Los Angeles Jeremy B. Rutter, Dartmouth College Monika Trümper, Freie Universität Berlin Hesperia is published quarterly by the American School of Classical Studies at Athens. Founded in 1932 to publish the work of the American School, the journal now welcomes submissions from all scholars working in the fields of Greek archaeology, art, epigraphy, history, materials science, ethnography, and literature, from earliest prehistoric times onward. Hesperia is a refereed journal, indexed in Abstracts in Anthropology, L’Année philologique, Art Index, Arts and Humanities Citation Index, Avery Index to Architectural Periodicals, Current Contents, IBZ: Internationale Bibliographie der geistes- und sozialwissenschaftlichen Zeitschriftenliteratur, European Reference Index for the Humanities (ERIH), Periodicals Contents Index, Russian Academy of Sciences Bibliographies, and TOCS-IN. The journal is also a member of CrossRef. Hesperia Supplements The Hesperia Supplement series (ISSN 1064-1173) presents book-length studies in the fields of Greek archaeology, art, language, and history. Founded in 1937, the series was originally designed to accommodate extended essays too long for inclusion in Hesperia. Since that date the Supplements have established a strong identity of their own, featuring single-author monographs, excavation reports, and edited collections on topics of interest to researchers in classics, archaeology, art history, and Hellenic studies. Back issues of Hesperia and Hesperia Supplements are electronically archived in JSTOR (www.jstor.org), where all but the most recent titles may be found. For order information and a complete list of titles, see the ASCSA website (www.ascsa.edu.gr). The American School of Classical Studies at Athens, established in 1881, is a research and teaching institution dedicated to the advanced study of the archaeology, art, history, philosophy, language, and literature of Greece and the Greek world. he s per ia 90 (2021) Pages 281–337 INSCRIPTIONS FROM PANAKTON A B S T R AC T Six inscriptions of the 4th and 3rd centuries b.c. recovered during surface survey and excavation at Panakton (the paleokastro above Prasino/Kavasala) are published here in full, four of them for the first time. The earliest, an arsenal inventory, preserves an archon date of 343/2 b.c.; three are ephebic texts of the Lykourgan era; one is a dedication by soldiers of the garrison in the second half of the 3rd century; one is a fragmentary heading. These inscriptions, the first found on this site, prove beyond doubt that this was the Athenian fortress of Panakton, and they provide new evidence for armaments, the ephebeia, and the history of Panakton among Attic garrison forts. In his 1978 article on “Roads and Forts in Northwestern Attica,” Eugene Vanderpool presented evidence and arguments challenging a half century of general consensus about the identification of Panakton.1 In the face of scholarship assigning that name to the impressive fortress of Gyphtokastro (correctly identified now as Eleutherai; Fig. 1), Vanderpool defended an older view that identified Panakton with the ruins on the summit overlooking the 1. Vanderpool 1978. On the debate over the identification of Panakton and the new consensus, see Ober 1985, pp. 224–225; Munn 1988, p. 365; 1989, pp. 233–234; 2010, pp. 191–192; Eretria XI, p. 156, n. 322. The former consensus that Panakton is to be identified with Gyphtokastro has been maintained by Clinton (I.Eleusis, vol. 2, pp. 255–256), but this view does not take into account the full weight of the evidence presented here, nor the identification of the fortifications of Gyphtokastro/Eleutherai as Boiotian, as argued by Cooper (2000) and now more thoroughly substantiated by Fachard (2013) and by Fachard et al. (2020). The inscriptions presented here (with the exception of 3; see n. 2, below) were discovered in the course of excavations at Panakton begun in 1991 under the auspices of Angeliki Andreiomenou, then director of the Ninth Ephoreia of Prehistoric and Classical Antiquities (now the Ephorate of Antiquities of Boiotia), with the cooperation and support of Vassilis Aravantinos, and continued in 1992 with the cooperation of Charis Koilakou of the First Ephoreia of Byzantine Antiquities (now the Ephorate of Antiquities of the City of Athens). I am grateful to these individuals for making the first excavations at Panakton possible. Preliminary reports of the findings of these © American School of Classical Studies at Athens excavations were published in Boeotia Antiqua (Munn 1996). A full account of the late medieval remains that represent most of the surface features visible at Panakton was published later in Hesperia (Gerstel et al. 2003). A full report of the prehistoric, Classical, and Hellenistic findings from these first excavations is in preparation. Excavations were carried out under the co-direction of myself and Vassilis Aravantinos and with the assistance of Sharon Gerstel, Mary Lou Zimmerman Munn, Patrick Thomas, and Carl Lipo. Funding for the survey and the excavations was provided by the Tressider Fund of Stanford University, by the David and Lucile Packard Foundation, by the generosity 282 mar k munn southwestern edge of the Skourta plain from above the village of Prasino (formerly Kavasala; Panakton is so located in Fig. 1). In the course of his argument, Vanderpool pointed out that “[a] stele base in the ruins of a chapel near the center of the Kavasala fort holds out the hope that inscriptions may one day be found: they are so far almost entirely lacking in the whole northwest frontier area.” As often, Vanderpool’s insight was providential. In 1988, in the third season of surface survey in and around the Skourta plain, my co-director on the survey, Mary Lou Zimmerman Munn, and I brought our student team and visitors up to the summit above Prasino to introduce them to the extensive ruins that, following Vanderpool, we were convinced were the remains of the Attic garrison fortress of Panakton. Pointing to the stele base that Vanderpool had indicated, I summarized his argument and relayed his prediction that more inscriptions were likely to be found on the site. Looking to the rubble pile at my feet, I picked up a fragment that glistened whiter than the rest of the limestone debris and said, “And here is one now!” It was a moment that seemed too good to be true—but it was true. The weathered stele fragment, the first inscription from the site, proved to be part of an Athenian ephebic decree of the tribe of Robert and Linda Rosenberg, and by the Institute for Aegean Prehistory. I thank all of these sponsors, and I thank Angeliki Andreiomenou and Vassilis Aravantinos for their permission to publish these finds. In the work that has led to the publication of these inscriptions I have benefited greatly from the advice of John Traill, Angelos Matthaiou, John Friend, and especially Stephen Tracy, who has been able to examine squeezes of these inscriptions and share his observations with me. I am particularly appreciative of the thorough critique that the first draft of this article received from two anonymous referees for Hesperia, which led me to some important corrections and reconsiderations. I am also appreciative of the meticulous care taken by the editorial staff of Hesperia, Jennifer Sacher, Sarah Figueira, and Karen Figure 1. Panakton and the fortified sites and major settlements located along the Attic-Boiotian-Megarian frontier. M. Munn Donohue, in seeing to many small but significant improvements to this article. All inscriptions are presently stored in the Archaeological Museum of Thebes. Photographs in Figures 3 and 12–14 were taken by John Dean and Ellen Burchenal for Panakton excavation records; the remaining photographs and drawings are by the author. All translations are my own. All dates are b.c. unless otherwise indicated. insc r ip t ions f r om panakton 283 Figure 2. Plan of Panakton, with the medieval features labeled. Contour intervals = 1 m. Gerstel et al. 2003, p. 152, fig. 5. Courtesy American School of Classical Studies at Athens 2. The discovery of this inscription was announced in Munn 1988 (p. 366), and in the spring 1991 newsletter of the American School of Classical Studies at Athens (Munn and Munn 1991), with a picture taken on the day of discovery and accompanied by a representative drawing of the stone, details of which are corrected here after readings were checked by John Traill. Inscription 3 was a discovery of the Skourta Plain Survey Project, the results of which are summarized in Munn and Munn 1990. Leontis (3 in the present catalogue).2 Excavation on the site of Panakton followed in 1991 and 1992, yielding five more fragmentary inscriptions and elements, such as stele bases and rock-cut niches, indicating the high probability that more inscriptional material remains to be found on the site. Survey and excavation have shown that the post-prehistoric occupation of Panakton belongs to two principal phases: a Classical and Early Hellenistic phase extending from the mid-5th century until the end of the 3rd century b.c., and a late medieval phase extending from the early 14th to the early 15th century a.d. Aside from the remains of walls and towers flanking the classical gateway and sporadic traces of the classical circuit wall, the building remains that are visible on the surface of the site and revealed just below the surface in all excavated areas belong to the late medieval phase (Fig. 2). All but one of the inscriptions and related finds reported here were found where they had been incorporated into late medieval structures, 284 mar k munn or in the ruins of those structures. The exception is 2, a broken base found amid debris from the destruction of the fortress in the Hellenistic period. The catalogue below presents, in approximate chronological order, all of the stone inscriptions recovered to date.3 THE INSCRIP TIONS 1 Inventory of an arsenal Fig. 3 Panakton 1992-300. Fragment of a white (“Pentelic”) marble stele recovered on July 28, 1992, from the ruins of the cross-wall joining the two rooms of late medieval House II in 20 m square K-9 (see Fig. 2). Broken at top and bottom, with rough-picked back and smooth-picked sides. H. 0.260; W. 0.283; Th. 0.050 m. L.H. 0.004–0.005 m. Previously published in Munn 1996, p. 52; SEG XLVI 185. Commentary in Couvenhes 2007, pp. 529–531, 538–539; Rihll 2007, pp. 51–52; Munn 2010, p. 196. 343/2 b.c. 5 10 15 Non-stoich. ca. 34–40 Unknown number of lines missing ca. 9–12 ca. 17–20 [- - - - - - - - - - - - - - - - - -]ς 𐅆𐅅ΗΗΗ̣Η̣[- - - - - - - - -] ca. 12–17 [- - - - - - - - - - - - - -]τ̣ι̣κ̣οι 𐅃Ι, σταθμὸν σ[τατῆρες - -] ca. 6–9 ca. 10–14 [- - - - - - - - - - -] σταθμὸν στατῆρες ΙΙΙΙ Λ̣[- - - - - -] ca. 5–8 [- - - - - - αὐ]λ̣οὶ φύσαις δύο, σταθμὸν στα[τῆρες - -] ca. 6–8 [- - - - - - -]ς̣ ΙΙΙ, σταθμὸν στατῆρες 𐅃ΙΙΙΙ· σφῦ[ρα χαλκ][ευτι]κὴ [μ]ία, σταθμὸν στατῆρες ΙΙΙ· φ[ια]λ̣ίδ[ια? δια]βεβρωμένα Ι· τοξεύματα ἀκίδας οὐκ ἔχ[οντα] ἐπτερωμένα 𐅅ΗΗ𐅄· ἕτερα τοξεύματα ἄπτε[ρα] ἄνευ ἀκίδων Χ𐅅ΗΗ𐅄· ἀκίδες σκυθικαὶ ΧΗΗ[- -] σώρακοι τρεῖς κενοὶ οὐχ ὑγιεῖς ἄνευ ἐπιθημ̣[ά]των· σχασ⟨τ⟩η⟨ρ⟩ία καταπάλτου· οἰστοὶ καταπάλ[του] ἄπτεροι 𐅄ΔΔ𐅃Ι· ἕτεραι ἀκίδας οὐκ ἔχουσαι Δ𐅃Ι[- -] ἀκίδες τούτ⟨ω⟩ν ΔΙΙΙ· κιβώτιον, ἐν ὧι ΔΙ οἰστοί, οὐ[χ ὑ]γιές· σαύνια ΗΔΔΙΙΙΙ· ἀκόντια ΔΔ𐅃ΙΙΙ, λογχία οὐ[κ ἔ]χοντα ΙΙΙΙ, ⟨ὧ⟩ν οἱ παλαιοὶ 𐅃Ι· τούτων εἷς δεδεμένο[ς], ἕτερος ἐρρωγώς. ταῦτα π[ά]ντα παρέδωκα Δημοκ̣[λ]ε̣ῖ Κλεοκρίτου Ἀχαρνεῖ ἐπὶ Πυθοδότου ἄρχοντ[ος]. vacat Tran s l at ion [- - - - - - - - - - - - - - -] 5,900 [+ ? - - - - - - - - - - - -]TIKOI?: 6, weight in s[taters- - - - - - - - - - -] weight in staters: 4 [+ ? - - - - - - - - - - pi]pes for bellows: two, weight in sta[ters- - - - - - - -]s: 3, weight in staters: 9; ham[mer for smithin]g: one, weight in staters: 3; sm[all] sa[ucers rust]ed through: 1; arrows not ha[ving] points, feathered: 750; other arrows unfeather[ed], without points: 1,750; Scythian points 1200 [+ ?]; three empty chests not in sound condition without lid[s]; a trigger mechanism of a catapult; bolts for catapul[t], unfeathered: 76; 3. The presentation of texts follows the Leiden convention as discussed by Dow (1969) and Pritchett (1955) on the representation of dotted letters. insc r ip t ions f r om panakton Figure 3. Inscription 1 (1992-300). Scale 1:2 285 others not having points: 16[+ ?]; points for these: 13; a box no[t in so]und condition, in which are 11 bolts; light javelins: 124; javelins: 28; ones with[ou]t heads: 4; 6 of these are old, one of these is tied together, another is broken. All these items I handed over to Demok[l]es, son of Kleokritos, in the archonship of Pythodotos. E p ig rap h ic al N ot e s Lettering is compact and clear but with inconsistencies that make it appear somewhat hasty. Cross-strokes of epsilons vary in length and intersect the vertical at different points; strokes of sigmas are at variable angles and can cross each other irregularly; diagonal strokes of kappas are at variable angles and lengths and intersect the vertical at various points; strokes of mus and nus are at variable angles and are often asymmetrical; omicrons and thetas vary in size and are sometimes more angular than circular; rhos are variable like omicrons, often with attenuated vertical strokes. Line 1: At left, sigma has three strokes showing, in form like the second sigma in line 11; third heta has only two vertical strokes preserved; fourth heta has lower half of left vertical and crossbar preserved. Line 2: At left, the bottoms of two vertical strokes are preserved; the third letter, previously read as a possible sigma, preserves a trace of a vertical stroke with a long diagonal stroke like the kappa in line 14; after the 286 mar k munn omicron a letter that appears to have been pi, a mistake for the numeral, has been mostly erased leaving a vertical iota; possible readings: -ρικοι, -τικοι. Line 3: At right, after the numeral the diagonal stroke of an alpha or lambda is visible. Line 4: At left, most of the right diagonal of the lambda shows. Line 5: At left, the tip of a stroke possibly of a sigma survives. Line 6: At left, the bottoms of the vertical and the diagonal of the kappa are visible at the break. At right, after the gap a right stroke of lambda is visible. Line 7: At left, bottom loop of first beta is preserved. Line 8: At left, the vertical and lower and middle cross-strokes of epsilon are clear, placed high in the letter space like the epsilon in the eighth position in this line. A mark previously read as the angled left stroke of an alpha is a break in the stone.4 Line 9: In first numeral, chi is oriented close to + while chi in second numeral is closer to x. At right, there is room for two additional hetas. Line 10: At right, left stroke of mu is visible. Line 11: For the eighth character the cutter has incised an epsilon instead of a tau; the tenth letter appears as an omicron instead of a rho. Line 12: At right, there is room for up to three more vertical strokes. Line 13: For the eleventh character the cutter has incised an omicron instead of an omega. Line 15: For the tenth character the cutter has incised an omicron instead of an omega. Line 16: At right, trace of the intersecting strokes of kappa is discernible. Line 17: At left, the top of the vertical and top cross-stroke of epsilon is visible. Te xt ual C omm entary Weights listed in the surviving part of this inscription are in staters, a common weight unit not to be confused with the coin equivalent of a didrachm also referred to as a stater. A weight talent was divided into 30 weight staters, and the weight stater was divided into various fractional units, starting with two mnai to the stater, the mna being composed of 100 weight drachms.5 As Lang has shown, from at least the late 5th century until the late 2nd century, the weight stater in Athenian usage was in the vicinity of 900 grams.6 By this measure, the heaviest weight recorded—nine staters for three items in line 5, giving an average weight of three staters per item, equal to that of the hammer listed in line 6—is about 2.7 kilograms. Line 1: The exceptionally large number, between 5,900 and 6,000, is too large for a weight in staters as exemplified in the following five lines (it would amount to almost 200 talents). It seems most appropriate for a quantity of missile weaponry of some sort, as enumerated in somewhat smaller quantities in lines 7–16. Lead sling bullets, [μολυβδίδε]ς, might be recorded here by number or by weight in drachmas (nearly a full talent); see IG II2 1488 (end of 4th century), line 6, where μολυβδίδες are mentioned with the enumeration of other projectiles; IG II2 468 (306/5 b.c.), line 10, where the melting of lead, possibly as much as 50 talents, is mentioned 4. I appreciate the care of an anonymous reviewer for challenging the reading previously published (Munn 1996; SEG XLVI 185), which is corrected here after reexamination of squeezes. 5. Agora X, pp. 2–3. 6. Agora X, pp. 15–20. insc r ip t ions f r om panakton 7. For the type of Scythian arrowheads represented at Panakton, see Olynthus X, types G III and G IV, pp. 406–411, nos. 2103–2134, pls. CXXV, CXXVI (Olynthos); Miller 1977, pp. 8–10, pl. 5:f (Nemea); Baitinger 2001, types II D 1–3, pp. 25–30, nos. 378–486, pls. 11, 12 (Olympia); Schmitt 2007, types C III a–c, pp. 484– 486, nos. 407–420, pl. 98 (Kalapodi). Snodgrass (1967, pp. 82, 99–100) describes the diffusion of Scythian archery and this type of arrowhead from the Late Archaic through the Classical period. The Scythian origin of the type is demonstrated by finds from Scythian tombs; see Anokhin et al. 1971, pp. 131, 450, figs. 34, 132; Parzinger 2004, pp. 38–45, fig. 4. 287 in a context of defensive weaponry for Athens; cf. the inventory of naval stores, IG II2 1627 (330/29 b.c.), column B, lines 322–325, and inventories of following years, where six talents and three minas of μολυβδίδων σταθμὸν καὶ μολύβδου in a large wicker chest are listed among both building materials and catapult parts and accessories. Also possible is another class of arrowheads, [κρητικαὶ ἀκίδε]ς, for which see IG II2 1424a (369/8 b.c.), column III, line 383, where 650 of these are counted in the Chalkotheke. Line 2: Possibly [- - στρατιω]τικοί, or [- - καταπάλ]τικοι, or [- - χαλκευ]τικοί, or [- - βαρβα]ρικοί. Line 3: At right and continuing to line 4, there is insufficient space for an item to be listed with its weight, so an item or items must have simply been counted. In view of the following items, this was possibly ἄ̣[κμων - -], anvil; cf. the four iron anvils and two hammers listed in a context that includes chests of arrows in the opisthodomos of the Parthenon, IG II2 1424a (369/8 b.c.), column I, line 119, and an example from Delos, I.Délos 442, face B, line 168. Other possibilities include ἀ̣[κόνη - -], whetstone (cf. IG II2 1628, column D, lines 522–523), or λ̣[αβίδες - -], tongs (cf. I.Délos 442, face B, line 168, where the meaning is more likely “handles,” but see LSJ, s.v. λαβίς). Line 5: Possibly [λαβίδε]ς. Lines 5–6: On the weight of this smithing hammer, approximately 2.7 kilograms, see above. The hieropoioi on Delos record a hammer among stone masons’ tools weighing 24 mnas (= 12 staters), approximately 11 kilograms (a sledgehammer), IG XI.2 158, face A, lines 79–80. Line 6: The restoration φ[ια]λ̣ίδ[ια δια]-, suggested by Chaniotis, SEG XLVI 185, seems probable despite the fact that the plural is counted as one. Lines 6–7: On the poor state of metal vessels frequently noted in inventories, cf. the broken (κατεαγώς, -υῖα -ός, -ότες -υῖαι -ότα), defective, or pierced (τετρυπημένη) condition of many bronze vessels in 4th-century inventories from the Acropolis (for example, IG II2 1469, 1471) and Eleusis (for example, IG II2 1541, 1542 = I.Eleusis 140, 149), and note the rustedthrough condition (διαβεβρωμένα) of vessels including phialai recorded in several Delian inventories of the mid-2nd century, such as I.Délos 1439. Lines 7–14: Stores of projectiles (τοξεύματα, βέλη καταπάλτων), regularly in σώρακοι, are found in the 4th-century inventories of the items kept in the Parthenon and in the Chalkotheke (IG II2 120, lines 37–38) and in the inventories of the naval stores, sometimes described as in poor or unusable condition, σαπρὰ ἄχρηστα (IG II2 1424a, column III, line 345), or τὰ μὲν ὑ[γιῆ τὰ δ’ού]χ ὑγιῆ (IG II2 1469, face B, column 1, lines 79–80). Projectiles are often described as incomplete, or stored as component parts: for example, [ἀκίδ]ας βελῶν τοξικῶν (IG II2 1487, face B, column 2, line 100), and βέλη καταπαλτῶν ἀνηκίδωτα και ἀπτέρωτα ΗΗΗΗ𐅄𐅃𐅛 καὶ ἠκιδωμένα 𐅄Ι𐅛 σχίζαι ⟨εἰς⟩ βέλη καταπαλτῶν 𐅛 ΔΔΔΔ𐅃Ι[Ι] (IG II2 1627, column B, lines 337–341). Line 9: Ἀκίδες σκυθικαί were a common form of small bronze socketed arrowhead with three narrow vanes, more suited to penetrate hard targets than broad-headed leaf-shaped arrowheads.7 Several examples have been found on the site (Fig. 4). 288 mar k munn Figure 4. Bronze socketed arrowheads from Panakton. Scale 1:1 Line 10: Σώρακοι were large containers, generally wicker chests, suitable for transporting produce or commodities. In addition to storing large numbers of projectiles, they could contain small shields (Aen. Tact. 30.2: ἀσπίδια), or catapults, presumably disassembled (IG II2 120, line 37, unless σώρακοι καταπαλτῶν means σώρακοι [βελῶν] καταπαλτῶν, as most commentators assume; see, for example, Marsden 1969, pp. 65–66). Line 11: The cutter has incised ΣΧΑΣΕΗΟΙΑ for ΣΧΑΣΤΗΡΙΑ. Was this device perhaps novel and not familiar to the cutter of this inscription? Although it was an essential mechanism in the engineering of catapults for a half century prior to this inscription, this is the earliest attestation of the term σχαστηρία (trigger release), otherwise first attested in the metaphorical usage of the pseudo-Aristotelian De Mundo 398b15, referring to devices created by μηχανοτέχναι, or in Philo Mechanicus, Belopoeica 51 and 53, over a century later than this inscription. The storage of component parts of catapults is attested in the inventories of dedications of the 330s to the end of the century, where catapults are often listed as defective or missing parts (for example, IG II2 1467, lines 50–55: [οὐχ] ὑγιεῖς ούδ’ ἐντελε[ῖς]), or as disassembled components (IG II2 1627, column B, lines 328–336), indicating that they must often have been scavenged for spare parts. Lines 12–13: The ἀκίδες for the οἰστοὶ καταπάλτου were most likely tanged iron points with elongated pyramidal points, several examples of which have been found on the site (Fig. 5).8 Line 14: The σαύνιον (also accented σαυνίον; LSJ, s.v.) was a short javelin, comparable in length to a long sword (Diod. Sic. 5.30.4), sometimes said to be solid iron (Diod. Sic. 5.34.5), described by Hesychius and Photius as an ἀκόντιον βαρβαρικόν, and used by mounted Indians (Arr. Indica 16.10; Diod. Sic. 17.88.5) and Persian horsemen (Strabo 15.3.18; Diod. Sic. 17.20.3) in hunting and in battle. Commonly mentioned among projectile weapons from the time of Alexander onward, the σαύνιον was well enough known to Cratinus (fr. 490 KA; see Henderson 1975, pp. 123–124) and his 5th-century audiences for it to serve as a deprecatory reference to a man’s phallus, evidently because it was shorter than a standard ἀκόντιον. Lines 15–16: Following the enumeration in line 14 of ἀκόντια, a neuter plural noun, the relative clause introduced by ⟨ὧ⟩ν qualifies a masculine plural 8. As opposed to the Scythian arrowheads described above, which fit arrow shafts less than 0.005 m in diameter, these iron tanged points will have fitted shafts close to 0.010 m in diameter. The type has been generally identified as an arrowhead, but the difference in material, heavier shaft, and generally later distribution of datable finds makes it likely that these were points for early bolt-shooting catapults. See Olynthus X, type E, pp. 392–397, nos. 1976–2022, pls. CXXIII, CXXIV, where these examples from Olynthos are all considered to be remains from the siege of 348 b.c.; for examples from Hellenistic levels at Kalapodi, see Schmitt 2007, types B VI c, d, pp. 479–480, nos. 368–387, pls. 95, 96. insc r ip t ions f r om panakton 289 Figure 5. Iron tanged projectile points from Panakton. Scale 1:1 antecedent; one would expect ⟨ὧ⟩ν τὰ παλαιὰ 𐅃Ι· τούτων ἓν δεδεμένο[ν], ἕτερον ἐρρωγός. Has the scribe simply been careless and forgotten that the clause is governed by a neuter plural noun? Or has the scribe, or the cutter, transposed a clause that properly should follow the 11 οἰστοί in line 13, the nearest masculine plural antecedent? I find the former possibility to be the more likely (and translate accordingly), as it seems more probable that a small number of old javelins (ἀκόντια) would be repaired or in need of repair, being more substantial arms than the more numerous catapult bolts.9 Lines 16–17: Δημοκλῆς Κλεοκρίτου Ἀχαρνεύς (PAA 315810; AO 23) might be the same as Δημοκλῆς Ἀχαρνεύς (PAA 315800; LGPN II, 20; AO 22), a member of the genos of the Salaminioi, whose son Aristarchos was an oath taker on behalf of the Salaminioi of the Hepta Phylai as recorded in 363/2 (Agora XIX, no. L4a = Rhodes and Osborne 2003, no. 37, line 77). Ἀρίσταρχος Δημοκλέους Ἀχαρνεύς (PAA 164280; LGPN II, 19; AO 21) was likely the same as Aristarchos archon of the Salaminioi in 363/2 or in a closely following year (Agora XIX, no. L4a, line 82; see Lambert 1999, p. 110), and therefore probably a fairly senior individual. The present Δημοκλῆς Κλεοκρίτου was therefore more likely a grandson of Demokles father of Aristarchos, if Kleokritos was a brother of Aristarchos, or a kinsman, and in either case very likely a member of this prominent genos. Kleokritos the father of Demokles (PAA 576842; AO 6) is not otherwise known. H istor ic al C ommen tary 9. I owe the observation of this grammatical incongruity to the careful attention of an anonymous reviewer. 10. See the corpora compiled by Clinton for Eleusis (I.Eleusis) and by Petrakos for Rhamnous (I.Rhamnous). The arsenal inscription from Panakton is unique among the border forts of Attica—in fact, so far as I am aware, among forts and civic garrisons throughout the Greek world. There are no comparable inventories of arms from Eleusis, Rhamnous, Phyle, or Sounion. In light of the substantial number of inscriptions known from Eleusis and Rhamnous,10 the discovery of such an inventory so far only at Panakton is suggestive of the importance of this fortress as the central forward base for military operations along 290 mar k munn the border of Attica and Boiotia. The only comparable arsenal inventories are the records of the naval stores of the Piraeus, especially those of the later 4th century in which catapult parts and projectiles are listed as being stored ἐν τῶι οἰκήματ[ι] τῶι [με]γάλωι τῶι πρὸς τ[αῖ]ς [πύ]λαις (in the large building next to the gates, IG II2 1627, column B, lines 280–281, of 330/29, and inscriptions of immediately following years). Like Panakton on the land frontier, the arsenal and naval yards of Piraeus were the central base for Athenian military operations at sea. This inventory has the format of a regular paradosis, transmitting responsibility from one officeholder, presumably the ἐπιμελητής τῆς φρουρᾶς or τοῦ φρουρίου (superintendent of the garrison/fort)11 to his successor in the manner best exemplified by the inventories of the treasurers of Athena,12 and in a secular context, by the naval inventories of the ἐπιμεληταὶ τῶν νεωρίων (superintendent of the shipyards).13 As the sole example of the paradosis of an arsenal, it is not clear if the present document represents part of a series like these other examples of paradosis, or if it is a unique specimen of accountability in the sphere of military stores, recorded under special circumstances. Unless further examples are discovered, for the historical reasons noted below I prefer to regard this example as a special case of an inventory that was not normally recorded on stone.14 In either event, it is a manifestation of the mid-4th-century reforms in military organization and authority that brought about, among other changes, the designation of one of the 10 annually elected generals as the στρατηγὸς ἐπὶ τὴν χώραν (general in charge of the countryside).15 Projectiles, catapults, and many other weapons are listed in the inventories of the items stored in the Parthenon and Chalkotheke on the Acropolis, several of which are cited above, but there are important distinctions to be drawn between those stores and the arsenals of Piraeus and Panakton.16 Acropolis inventories list many items of personal armament (shields, helmets, body armor, spears, swords), none of which is attested in the naval inventories or at Panakton. While the arms kept on the Acropolis might have been put to use in time of need, as were the monetary treasures of Athena, 11. Or the ἐπιμελητής ἐν τῶι φρουρίωι, cf. Reinmuth 15, left side, lines 7–9 (I.Oropos 353; Friend T15); or τοῦ φρουρίου, cf. IG II2 1312, lines 7–8 (I.Rhamnous 13). 12. See the transmission of the inventory of the Hekatompedon by the treasurers of Athena, IG I3 329, of 418/17; see, in general, Harris 1995. 13. First attested in IG I3 236, lines 5 and 6, of 410–404, but represented under other titles earlier; see Jordan 1972, pp. 25–46; Rhodes 1972, pp. 117–120. 14. Contemporary with this inscription is the naval inventory IG II2 1622, which exceptionally includes a passage, lines 379–579, recording debts extracted from officials in default of their responsibilities for naval equipment over the previous 30 years; see Gabrielsen 1994, pp. 150–151. This suggests that at this particular time, in the archonships of Euboulos, Lykiskos, Pythodotos, and Sosigenes (345/4– 342/1), there was an unusual urgency to account publicly for military stores. 15. On the selection of the στρατηγὸς ἐπὶ τὴν χώραν (sometimes ἐπὶ τῆι χώραι in inscriptions) see Ath. Pol. 61.1; the earliest surviving epigraphic reference to this military command (ὁ στρατηγὸς ὁ ἐπὶ τὴ[ν φυλ]ακὴ[ν τῆς χ]ώρας κεχειροτονημένος) is in the “Orgas” inscription, IG II3.1 292 (IG II2 204), lines 19–20, of 352/1 (Ober 1985, p. 89; Munn 1993, p. 190). This assignment of responsibility for territorial defense was roughly contemporary with the creation of naval symmories to facilitate the financing of the fleet ( Jordan 1972, pp. 73–87; Gabrielsen 1994, pp. 182–193), and with them the designation of a στρατηγὸς ἐπὶ τὰς συμμορίας (Ath. Pol. 61.1; Rhodes 1981, pp. 679–682). Note also that IG II2 1622, the naval inventory contemporary to the Panakton arsenal inscription (see n. 14, above), also attests a new organization of hegemones of the symmories; see Jordan 1972, p. 81. 16. Ober (1985, p. 95) refers misleadingly to “Athenian military inventory lists” when he describes items recorded by the treasurers of Athena stored on the Acropolis. 291 insc r ip t ions f r om panakton those items were dedications to Athena, most of them probably prizes of war.17 The records of Piraeus and Panakton were records of state property, under the care of public officials, and comparable in that respect to the walls of the city and of Piraeus and of the forts built and maintained under the Athenian democracy for purposes deemed essential by the Athenian demos. It is significant, therefore, that the arsenal of Panakton consisted of projectile weaponry and the tools needed to keep them in operational repair. Defense of fortifications could require large quantities of projectiles, and where there was no industrial and commercial center within the fortifications such as there was at Athens and Piraeus, these would have to be stockpiled before the moment of need arose. Garrison troops could bring their personal gear with them when manning the fortress, but they would be ill-equipped if they relied for sling bullets and arrows on what they had personally carried with them.18 This was even more of a concern in the era of mechanized warfare; the widening deployment of catapults, in use by Athenians for at least a generation when this inventory was recorded, required special provision by the authorities responsible for defense.19 The arsenal inscription has historical interest in several respects. It is dated to 343/2, the year in which Demosthenes delivered his speech On the False Embassy (19), accusing Aischines of misconduct and corruption in the recent negotiations with Philip of Macedon.20 Three years earlier the Athenians had concluded a treaty with Philip, by which the Athenians withdrew their support from the Phokians, who had been at war with Philip and with the Thebans. The treaty had backfired badly for the Athenians, who had been expecting concessions from Philip that never materialized. Near the end of his long diatribe against Aischines for his role in those negotiations, Demosthenes summarizes the situation that eventually emerged: Right away [as soon as the treaty was made], instead of seeing Thespiai and Plataia resettled, you heard that Orchomenos and Koroneia were enslaved; instead of the Thebans being humbled and stripped of their arrogance and pride, the walls of your own 17. The inspection of the Chalkotheke by members of the boule and all military officers alongside the treasurers of Athena (IG II2 120, ca. 354/3) suggests that use of armaments was envisaged, possibly another example of midcentury concern for a full accounting of military stores (see n. 14, above). Couvenhes (2007, pp. 526–528) discusses this text as evidence for the provision of arms at public expense. On the use of the Chalkotheke for the storage both of weapons and of processional furnishings, see La Follette 1986, p. 78, n. 14. On weapons as prizes of war among diverse other votive offerings kept in the Parthenon, see Harris 1995, pp. 108–110. 18. Less than a decade after the date of this inscription, Panakton and the other garrisons of Attica were regularly manned by the newly organized corps of ephebes, whose training included ὁπλομαχεῖν καὶ τοξεύειν καὶ ἀκοντί­ ζειν καὶ καταπάλτην ἀφιέναι (to fight with heavy weapons, to use the bow and the javelin, and to discharge catapults); see Arist. Ath. Pol. 42.3. All of the requisite weaponry was provided by the state to the ephebes, including their heavy weapons, the ἀσπίς and δόρυ, and, notably, all of the projectiles that they would have to train with. A century later, individual benefactors were publicly thanked for providing projectile weapons (βέλη) and other provisions and ameliorations to the fortresses of Attica at their personal expense; cf. IG II2 1281 at Sounion; I.Rhamnous 26 (Oetjen 2014, no. 34) at Rhamnous. 19. On the development and diffusion of catapult technology in the 4th century, see the discussions in Marsden 1969, pp. 65–73, and Rihll 2007, pp. 46–73; see further in n. 23, below. Ober (1992) discusses the development of catapult technology as indicated by the design of towers that were built, in the second quarter of the 4th century and later, to house catapults. 20. The dating of this speech to the archonship of Pythodotos is given by Dionysius of Halicarnassus (Letter to Ammaeus 1.10); see MacDowell 2000, p. 22: “probably before the end of 343.” 292 mar k munn allies, the Phokians, were torn down—torn down by the very Thebans whom Aischines, in his speech, said would be dispersed into scattered villages. . . . And instead of recovering Oropos, we are marching out under arms to defend Drymos and the land around Panakton, something which, as long as the Phokians were safe, we never had to do!21 This inscription confirms what Demosthenes declares: there had been relative quiet on the northern border with Thebes for some years, but now the Athenians had mobilized to man the garrison and patrol the surrounding countryside in force.22 Our text indicates that the mobilization took place in the previous year, the archonship of Lykiskos (344/3), when the (unnamed) composer of this inventory was appointed to his post, presumably as ἐπιμελητής τῆς φρουρᾶς. Our text records the substance of his accounts rendered after his term of office was over, when the duty was handed on to Demokles of Acharnai in the archonship of Pythodotos. The inventory also attests to a period of relative neglect, as the armaments are described mostly in a state of disrepair. From the enumeration of stored armaments and equipment we learn that Panakton had been garrisoned and armed in some strength not too many years earlier, within the period when catapults were being deployed by the Athenians. This could, in my view, have been any time from the 380s on.23 I consider the most likely circumstance for the first arming of Panakton with catapults to be the Boiotian War of 378–375, when Athenians in alliance with Thebans were confronting Spartan-led armies passing the western frontiers of Attica, and the two sides challenged each other in the fields of the Asopos valley some 10 km to the north of Panakton. Although no military action involving Panakton is directly attested at this time, in the aftermath of the Spartan invasion under Sphodrias it must have been kept under heavy guard. Panakton lay along the direct route from Athens to Thebes, which was used by thousands of Athenian and mercenary troops each year, for three years. When a treaty was negotiated with the 21. Dem. 19.325–326: καὶ γάρ τοι παραχρῆμα, ἀντὶ μὲν τοῦ Θεσπιὰς καὶ Πλαταιὰς ἰδεῖν οἰκιζομένας, Ὀρχομενὸν καὶ Κορώνειαν ἠκούσατ’ ἠνδραποδισμένας, ἀντὶ δὲ τοῦ τὰς Θήβας ταπεινὰς γενέσθαι καὶ περιαιρεθῆναι τὴν ὕβριν καὶ τὸ φρόνημα [αὐτῶν], τὰ τῶν συμμάχων τῶν ὑμετέρων [Φωκέων] τείχη κατεσκάπτετο· Θηβαῖοι δ’ ἦσαν οἱ κατασκάπτοντες, οἱ διοικισθέντες ὑπ’ Αἰσχίνου τῷ λόγῳ. . . . ἀντὶ δὲ τοῦ τὸν Ὠρωπὸν ὑμῖν ἀποδοθῆναι, περὶ Δρυμοῦ καὶ τῆς πρὸς Πανάκτῳ χώρας μεθ’ ὅπλων ἐξερχόμεθα, ὅ, τέως ἦσαν Φωκεῖς σῷοι, οὐδεπώποτ’ ἐποιήσαμεν. 22. Demosthenes 54, Against Konon, likely also refers to this same period. The private suit emerged from a dispute that began at Panakton: “I went out on military service two years ago, when we were assigned garrison duty at Panakton” (Dem. 54.3: Ἐξῆλθον ἔτος τουτὶ τρίτον εἰς Πάνακτον φρουρᾶς ἡμῖν προγραφείσης). In the speech we learn that a general and taxiarchs were present in the garrison. On Drymos and its relation to Panakton, see Munn 2010, pp. 195–197. 23. Most commentators place the acquisition of catapults by the Athenians in the late 370s or 360s, when they become epigraphically attested. The earliest epigraphical evidence for catapults at Athens is in IG II2 1422, an inventory of the Chalkotheke dated to 371/0 on the basis of a restored archon’s name. Marsden (1969, pp. 65–67) accepts this date. Challenging it, Cole (1981) suggests that an archon’s name and a date in the later 360s are more probable; see also Rihll 2007, pp. 52–54. Regardless of this epigraphic evidence for catapults as votive gifts, I see no reason to suppose that it would take the Athenians 20 or more years after the deployment of this new technology on Sicily by Dionysios I of Syracuse in the early 390s to learn of its usefulness. Experienced craftsmen could well have brought their skills to Athens within a few years of the invention of catapults in the arsenals of Dionysios. insc r ip t ions f r om panakton 293 Spartans in 375, one of the factors that motivated the Athenians to make peace, according to Xenophon, was the cost of guarding the countryside (φυλακαὶ τῆς χώρας).24 After 375 the active theater of war moved away from the Attic-Boiotian frontier, but the growing hostility of the Thebans and the potential for raids across the frontier required sustained vigilance on the part of the Athenians. Aischines, in his speech rebutting the charges of Demosthenes, refers to the beginning of his career of military service in the late 370s as an armed scout, a peripolos of the countryside, a service that would be very familiar to his audience in 343.25 By then the defense of the countryside, φυλακὴ τῆς χώρας, had become a mandatory subject for discussion in every meeting of the ekklesia kyria, and one of the 10 annually elected generals was designated στρατηγὸς ἐπὶ τὴν χώραν.26 Panakton, at the forefront of Athenian territory facing Boiotia, could be considered a peripolion, a regular base for the peripoloi patrolling the frontiers in this period.27 It was in this interval that Xenophon attests to debates in the assembly about how many of the watchposts (φυλακαί) in the countryside were well situated, how many garrison troops (φρουροί) were sufficient, where reinforcements were needed, and where troops should be redeployed.28 These were the conditions, between the late 370s and the mid-340s, in which care for the military stores at Panakton would have become desultory. With no immediate threat of an invasion in force, armaments like catapults (once deemed essential there) might have been packed off for use elsewhere, either within Attica or on one of the campaigns undertaken by the Athenians abroad, leaving empty cases and a random trigger mechanism in the storeroom. Panakton was probably kept under the guard of a peripolarchos, to command the armed scouts along this sector of the Attic frontier, until the crisis alluded to by Demosthenes demanded a more substantial military call-up, and a general took up command at Panakton, with a curator of the arsenal appointed to render and publish a report on the state of armaments available there. 24. Xen. Hell. 6.2.1: οἱ δ’ Ἀθηναῖοι . . . ἀποκναιόμενοι καὶ χρημάτων εἰσφοραῖς καὶ ληστείαις ἐξ Αἰγίνης καὶ φυλακαῖς τῆς χώρας, ἐπεθύμησαν παύσασθαι τοῦ πολέμου (The Athenians . . . being worn out by monetary contributions and by piratical raids from Aigina and by measures for the defense of the countryside, became eager to put a stop to the war). On the relationship of Panakton to the events of the Boiotian War, see Munn 1993, pp. 148–170. 25. Aeschin. 2.167. For a discussion of peripoloi as armed patrols of the countryside, see Xen. Vect. 4.47, 52, with Pélékidis 1962, pp. 35–47; Gauthier 1976, pp. 191–192; Ober 1985, pp. 91–93; Couvenhes 2011, pp. 296– 301; see also n. 27, below. 26. The mandatory discussion of the φυλακὴ τῆς χώρας is noted in Ath. Pol. 43.4; see also Xen. Mem. 3.6.10, and the discussions in Ober 1985, pp. 87–90; Munn 1993, pp. 187– 195; Couvenhes 1999, pp. 193–202. The topic is epigraphically attested as a category of public deliberation in inscriptions of the 330s and 320s, starting with IG II2 435, line 13; see Rhodes 1972, pp. 231–235. On the appointment of the στρατηγὸς ἐπὶ τὴν χώραν, see n. 15, above. 27. Peripolarchoi, commanders of the peripoloi, are attested during the Peloponnesian War in Thuc. 8.92.2; cf. IG I3 93, line 42, of 415. In the 4th century they occur in the “Orgas” inscription, IG II3.1 292 (IG II2 204), lines 20–21, of 351/2, and in honorary inscriptions from Rhamnous (I.Rhamnous 92–95, and 96 = IG II2 2968) and from Eleusis (I.Eleusis 80 [= IG II2 1193] and 81), from the second half of the 4th century. On peripolia as garrison posts in the countryside, see Thuc. 6.45.1, 7.48.5 on the territory of Syracuse. 28. Xen. Mem. 3.6.10–11. This passage and other 4th-century sources referring to the defense of the countryside are discussed by Ober (1985, pp. 77–80) and Munn (1993, pp. 2–3, 187–195). 294 2 mar k munn Dedication to the Dioskouroi by the ephebes of Hippothontis Figs. 6–10 Panakton 1991-350. Rectangular base of blue-gray (“Hymettian”) marble recovered on July 24, 1991, from a debris spill in 20 m square L-11 (see Fig. 2), just inside the gate of the Classical and Hellenistic fortress. The stone was face down, with its right edge slightly more deeply buried (Fig. 6). Water seepage has left accretions over much of the face and caused heavy erosion of the left half of the face. Most of the right side of the base is broken away but a portion of the original right side survives. The sides and top are smooth-picked. The hammer-dressed back has vertical striations 0.04–0.07 m apart. The underside has a band of anathyrosis roughly 0.05 m wide on all edges. A rectangular cutting is centered on the top surface: W. 0.215, L. 0.175, D. 0.047 m. The left edge of the cutting is 0.175 m from the left edge of the stone; the right edge of the top of the stone is broken away; the front of the cutting is 0.100 m from the face of the stone; the back edge is 0.095 m from the back face. H. 0.390; W. 0.575; Th. 0.370 m. L.H. in heading 0.014 m, below heading 0.006–0.007 m. No trace of an incised wreath survives. Notices in PAA 1995–2012, sub nominibus; Eretria XI, p. 156, n. 322; Humphreys 2004–2009, p. 84, n. 3; Hansen 2006, pp. 21 (n. 15), 34 (n. 82); Tracy 2016, p. 265, n. 10. The foregoing have commented on this text based on my original conjectured restoration of the archon’s name as Nikokrates, which is revised in this editio princeps. Friend had access to the revised text and an intermediate draft of the commentary while preparing his collection (where this inscription appears as T20). Figure 6. Inscription 2 (1991-350), as uncovered face down in the Panakton gateway 295 insc r ip t ions f r om panakton 332–321 b.c. Non-stoich. Heading [Ἱπποθω]ντ̣ί̣δος [ἔφηβοι οἱ ἐπὶ] Κ̣[- - - - - - - - - - -] [ἄρχο]ντος Δι⟨ὸ⟩ς v Κ̣[ο]ύροις ἀ[νέ]θε[σαν] 10–11 Wreath 1 [στρατηγὸν] [­ ­ ­ ­ ­ ­]ο̣[­] [­ ­ ­ ­ ­ ­ ­]ο̣[­] Wreath 2 σωφρονισ{σ}τὴν Ἰσοκράτην Παλληνέα Wreath 3 κοσμητὴν Κτησικλῆν Κόπρειον Wreath 4 ταξίαρχ̣ο̣[ν] Πολυσ[- - - -] Πειραιέα [Wreath 5] [λοχαγὸν?] [- - - - - - - - - -] [- - - - - - - - -] Column 3 [Wreath 6] [λοχαγ]ὸ̣ν̣ [­ ­ ­ ­ ­ ­ ­ ­ ­] [­ ­ ­ ­ ­ ­ ­ ­ ­] [Column 1] [- - - traces - - - - - -] [­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­] [- - - - - - - - - - - - -] [­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­] [- - - - - - - - - - - - -] [- - - - - - - - - - - - -] [- - - - - - - - - - - - -] Ἀριστο[- - - - - - - - - - - -] Πυθοκρ[- - - - - - - - - - -] Wreath 7 Wreath 8 Γέρων Ἐλ[- - - - - - - - - -] λοχα̣γὸν λοχα̣γὸν Δημ⟨ο⟩κράτ[ης - - - - - - - -] [.]Ε̣[. . .]Ο̣[. . . .]Ṇ Ξενόφαντο[ν] 5 Θεόξενος Ἐρο[ιάδης] Ο̣ Ι ̣ Ι ̣ [- - - - - - -] Ἐλευσίνιον̣ Μνησίθειος Κόπρ̣[ειος] Κινησίας Δεκε[λειεύς] Column 2 Εὐφάνης Πειρα[ιεύς] [- - - - - - - -] Ἐ[λε]υ[σί]ν̣ιος Δα̣μόλας Ἐλευσί[νιος] [. .]Ο̣[.]Ι[̣ .]Ο̣[.]ς̣ [Ἀ]ζηνιεύς 10 Ἀντιφῶν Πειρα̣ι̣ε̣[ύς] Μν̣η̣[σι]π̣τ̣ό̣λ̣εμος Ἀχερδούσιος Εὐφρόνιος Οἰναῖ⟨ο⟩ς [. .]α̣θι̣[.]ς Πειραιεύς Ναυχαρίδης Ἐ̣λ̣[­ ­ ­ ­ ­ ­ ­] 5 [. . .]Ο̣νας Δεκελειεύς vacat [- - - - - - - - Ἐ]λ̣[ευσ]ίνιος [- - - - - - Ἐλ]ε̣[υ]σ̣ίνιος vacat Ep ig rap h ic al N ot e s 29. IG II2 1189 is illustrated by Reinmuth (no. 3, pl. IV), by Mitchel (1984, p. 114), and by Clinton (I.Eleusis 84, pl. 36). 30. Tracy 1995, pp. 112–116, on the Cutter of IG II2 337. Regarding inscription 2, Tracy writes (pers. comm.), “I am of the opinion that it is indeed the work of the Cutter of IG II(2) 337. The shapes of the letters and the small idiosyncrasies (seriflike thickenings, the curving of some strokes, placement of omega) match exactly.” Letters in the heading are deeply cut but eroded; verticals do not respect a consistent baseline; there are serifs at ends of all strokes; the outer strokes of sigmas bow slightly. The letters in the heading are strikingly similar to those in the heading of IG II2 1189 (Reinmuth 3; I.Eleusis 84; Friend T3), an ephebic dedication of Hippothontis at Eleusis.29 The serifs are similar; the vertical of two upsilons descends noticeably below the baseline of other letters; most outer strokes of sigmas bow slightly; omicrons are generally set high in the letter space and sometimes show a central dot. Tracy has identified IG II2 1189 as the work of the Cutter of IG II2 337, and based on his examination of squeezes of 2 he has confirmed that this is also the work of the Cutter of IG II2 337.30 The smaller letters below the heading are more lightly incised, although omicrons and thetas are slightly more deeply cut (also a feature of IG II2 1189), making it possible to see the traces of some of these circular letters in the highly eroded left side of the stone. The spacing of the names of the honorands allows room for them to have 296 mar k munn Figure 7. Inscription 2. Scale 1:5 been surrounded by wreaths; these must have been painted, as no trace of an incised wreath survives. Specific characteristics of the work of the Cutter of IG II2 337 as described by Tracy that are found in this inscription include delta, which “has a squat appearance because it is quite wide and does not extend to the bottom of the letter space” (see the first letter in line 9 of column 3 in Fig. 9); epsilon, where “top and bottom horizontals are longer than the vertical and tend to be placed slightly in from the ends of the vertical. The center horizontal is shorter” (this describes most epsilons; see epsilons in lines 8, 9, 10, and 11 of column 3 in Fig. 9); mu, where the “outer strokes of this letter slant rather sharply, with the result that the central V (which reaches to the baseline) is unusually narrow, insc r ip t ions f r om panakton 297 Figure 8. Inscription 2, detail of upper right edge 5 10 Figure 9. Inscription 2, detail of column 3, with drawing almost awkward in appearance” (see the mu in line 1 of wreath 3 in Fig. 8, and the third letter in line 9 of column 3 in Fig. 9); omicron, which is “quite round, small, and placed somewhat up in the space” (see the omicrons in column 3 lines 5, 6, 9, and 11 in Fig. 9); pi, where “the horizontal usually bows downward perceptibly; it begins in the first vertical and either ends at the second or extends just a bit beyond it. This is one of this cutter’s most idiosyncratic letters” (the bowing described by Tracy is not evident here, nor in the illustration he provides, but the rightward extension of the horizontal is evident in several examples here; see pi in line 3 of wreath 3 and the eighth letter in line 10 of column 3 in Fig. 9); sigma, where “top and bottom strokes always slant; the lower half is often slightly wider and extends into the interline” (true of all sigmas, notable in the heading and clearest in column 3, lines 5–9 and 11 in Fig. 9); upsilon, which “is made with three strokes. The vertical is normally just slightly more than half the 298 mar k munn height of the letter. The V is large and symmetrical” (evident in all cases, clear in the second letters in lines 8 and 11 in column 3 in Fig. 9); omega, which “is a relatively short letter which is placed up from the baseline in the middle of the letter space. Short horizontals extend to right and left, one usually slightly longer than the other. . . . In contrast to the dominant tendency of his contemporaries, this cutter does not extend the horizontals much into the letter” (see the sixth letter of line 10 in column 3 in Fig. 9). The identification of the cutter of this inscription with that of IG II2 337, and significantly of IG II2 1189, has implications for the dating of this inscription that are discussed below. Heading line 1: At left, bottom of the vertical and the bottom angle of the nu are preserved; vertical strokes of the tau and iota are preserved for half or more of their heights. At right, the end of a single diagonal stroke with its serif is clear (see Fig. 8). The angle of this diagonal stroke is 45°, distinctly steeper than the 20° angle of the lower stroke of sigma, and not as steep as the 55° angle shown by delta and alpha, which would be the same for lambda; as comparison to the letters below the heading shows, this letter is either a kappa or a chi. Heading line 2: After the break at left the preserved letters are clear as far as the second sigma, after which there is clearly a blank space; the second preserved omicron has been cut as theta; after the vacat there are highly eroded depressions indicating traces of the vertical and the lower diagonal of a kappa at the edge before the second break. The spacing of Δι⟨ὸ⟩ς v Κ̣[ο]ύροις is remarkable and to my knowledge unparalleled. After the second break the vertical and traces of the right diagonal and its serif of the upsilon are visible; most of the left diagonal and the start of the crossbar of the alpha are preserved. After the third break most of the theta and epsilon are clear (although the stone has been damaged in handling since the original reading and portions of the epsilon are now gone; compare Figs. 7 and 8). Wreath 1: Indeterminate traces in the first line indicate that this wreath was placed slightly higher than wreaths 2–4 (and presumably 5). Faint traces of two omicrons or thetas can be discerned in the second and third lines. There are other marks on the stone but it is impossible to be confident that they are letter strokes. The first line can be restored as [στρατηγὸν] on the basis of the frequent pattern of ephebic dedications honoring one or both of the generals overseeing their service in Piraeus and in the garrisons (ἐπὶ τῆι χώραι) along with their sophronistes, kosmetes, and the ephebic officers, the taxiarchos and lochagoi (and sometimes didaskaloi).31 Wreath 2: Letters of the left half are faint, but all letters can be discerned. In the first line, the cutter has incised an extra sigma. Wreath 3: All letters are clear. Wreath 4: Visibility of letters deteriorate rapidly toward the break at right. In the first line, partial traces of chi and omicron can be discerned at the right. In the second line, nothing can be discerned after the sigma. In the third line there are faint traces of half of the alpha at the right end. Wreath 5: This is assumed to have existed on the missing right side to balance the distribution of wreaths. Wreath 6: The spacing of the legible names of wreaths 7 and 8 allows an equivalent space for a wreath preceding them. A faint trace of an 31. Ephebes record honors to one or more strategoi along with other ephebic officers in IG II3.4 329 (Reinmuth 4; Friend T4); IG II3.4 337 (I.Eleusis 86; Reinmuth 5; Friend T6); IG II3.4 330 (Reinmuth 7; Friend T5); Meritt 1940, no. 8 (Reinmuth 9; Friend T9); Pritchett 1949 (Reinmuth 12; Friend T19); I.Oropos 353 (Reinmuth 15; Friend T15); Petrakos 2004 (Friend T8); and probably in IG II3.4 342 (Reinmuth 10; Friend T14). insc r ip t ions f r om panakton 299 omicron and what may be a vertical stroke to its right can be discerned. The alignment of these letter traces with the first lines of wreaths 7 and 8 makes the restoration [λοχαγ]ὸ̣ν̣ probable. Less likely is the possibility that these traces belong to the ephebic roster, which more probably began below, where faint traces of what may be letter strokes begin to appear in alignment with column 2. Wreath 7: Omicrons or thetas are the clearest letters in all three lines. In the first line, all letters are faint but discernible. In the second line, the second letter could be epsilon, pi, or rho, or possibly tau; in the sixth letter space an omicron or theta is clear; in the eleventh letter space there are traces that could be nu. In the third line, the first letter could be omicron or theta; it is followed by a clear vertical that could be iota, or eta if the trace of a following vertical is not part of the third letter. Ο̣ἰ̣ν̣[αῖον] is a possible restoration. Wreath 8: Except where the face is damaged at the right, all letters are clearly discernible. As in wreath 7, the letters of the first line are more widely spaced than the letters in lines two and three. Roster of ephebes: Legible names and surviving traces make it clear that the last two columns contained 19 names (seven in column 2, 12 in column 3); the face is well enough preserved below the last legible names to be certain that there were no more names below the last legible traces. The two columns, in the center and at the right, are evenly spaced to allow for a third column at the left (column 1) occupying the same lateral space as the two legible columns. It is most probable that column 1 began below wreath 6 in alignment with column 2, and probably continued no lower than column 2. In that case, the complete roster listed 26 ephebes, not counting officers. Column 1: Faint traces of possible letter strokes are discernible starting in line with the first line of column 2; no letters can be confidently read. Column 2: The surface is sufficiently well preserved below wreath 8 to be certain that there is no inscription above the first legible text in this column. In line 1, the last three letters are clearest, but all undotted letters are clearly discernible. In line 2 the circular letters could be omicron or theta, and the intermediate vertical stroke is uncertain. In line 3, the first three strokes of the mu are discernible, after which three vertical strokes are discernible, most likely traces of ΝΗ; the vertical and cross-stroke of the letter read as Π are discernible; there is a trace of the juncture of the vertical and cross-stroke of the tau, of the top right quarter of the omicron, and of the left diagonal of the lambda, after which the reading is clear. In line 4 there are traces of two uncertain letters followed by a barely discernible alpha; the theta is distinct, followed by a trace of a vertical; after a space where the stone has been chipped, the letters ΣΠΕ are faintly discernible, after which all is clear. In line 5, a circular letter is faintly discernible in the fourth letter space, following which letters become gradually more distinct. In line 6, the lambda is discernible, and the last five letters are clear. In line 7 the stone has been damaged in handling since the original reading; now what can be discerned is [Ἐλ]ε̣[υσίνι]ος. Column 3: Almost all letters before the break at right are clearly legible. In line 3 at right, most of the left diagonal of lambda is preserved. In line 4, the omicron has been cut as theta. In line 6 at right, part of the 300 mar k munn vertical of the rho is preserved. In line 7 at left, damage to the stone has destroyed part of the kappa and most of the iota, though sufficient traces remain to be certain of the readings. In line 9 at left, the apex of the first alpha is preserved, and the rest of the letters surviving on this line are distinct. In line 10 at right, most of the diagonals of the alpha are preserved, followed by a trace of a vertical and thereafter the top horizontal stroke of an epsilon. In line 11, the final omicron has been cut as theta. Line 12 is cut by a different hand; the letters are more widely spaced and do not follow a consistent baseline, and the final sigma is more widely splayed than those above. The reading of this previously unattested name is clear; only the last two letters, giving the demotic, are faint. Pr os op o g rap hy A preliminary reading of this inscription was shared with John Traill, and names from that reading have been entered into his Persons of Ancient Athens (PAA). They are recorded there under the archonship of Nikokrates, according to my original conjectured restoration of the heading. The present restoration takes into account corrected measurements from the stone and consideration of further evidence (see below). Here I list individuals in alphabetical order with references to PAA, LGPN where applicable, and the current numbering in AO.32 Ἀντιφῶν Πειρα̣ι̣ε̣[ύς] (column 3, line 10): ephebe, PAA 138600; AO 58. Ἀριστο[- - -] (column 3, line 1): ephebe, PAA 167015. Γέρων Ἐλ[- - -] (either Ἐλευσίνιος or Ἐλαιούσιος; column 3, line 3): ephebe, PAA 273845; AO 3. Δα̣μόλας Ἐλευσί[νιος] (column 3, line 9): ephebe, PAA 273845. Notable for his Doric name not otherwise attested at Athens, but attested at Epidauros (IG IV2.1 194) and at Dodona (Lhôte 2006, no. 58, face B) in the Classical era, and more widely in Hellenistic and Imperial times (Rhodes, Sparta, Epidauros; see LGPN I, 117; LGPN IIIA, 112). Δημ⟨ο⟩κράτ[ης] (or possibly the rare Δημ⟨ο⟩κρατ[ίδης]; column 3, line 4): ephebe, PAA 316527; AO 78. Εὐφάνης Πειρα[ιεύς] (column 3, line 8): ephebe, PAA 449207; AO 22. Εὐφρόνιος Οἰναῖ⟨ο⟩ς (column 3, line 11): ephebe, PAA 451191; AO 24. Possibly the same as PAA 451190; LGPN II, 20; AO 23, [Ἐ]υφρό[νι]ος Μελ[α]νώπ[ο]υ [Οἰ]ναῖος, IG II2 2393, line 12 (second half of 4th century). Μελάνωπος Οἰν[αῖ(ος)], possibly the ephebe’s father, was a syntrierarch in 322, IG II2 1632, lines 100, 132–133, 312; see APF 9792. Θεόξενος Ἐρο[ιάδης] (column 3, line 5): ephebe, PAA 509182; AO 17. Ἰσοκράτης Παλληνεύς (wreath 2): sophronistes, PAA 542186; AO 19. Possibly the same as PAA 542185; AO 18, a thesmothetes on Samos ca. 350, IG XII.6.1 262, column XI, line 362. Also possibly the father of [- κ]ράτης Ἰσοκράτο[υς] Παλληνεύς, bouleutes in 281/0, Agora XV 72, column 6, line 229, PAA 542180; LGPN II, 17; AO 20. Byrne (AO 18) speculates that the sophronistes of this inscription might be identified with [. . .6. . .]την Ἀ[. . . 7. . . .]υ Παλλην[έα], one of two didaskaloi honored by the ephebes of Leontis enrolled under Nikokrates, 333/2, Meritt 1940, no. 8, column I, lines 34–35 (Reinmuth 9; Friend T9). Reinmuth suggests that the name 32. The reference numbers in the online lexicon AO will change as new names are added. Individuals will in any case be identifiable by the references given here. insc r ip t ions f r om panakton 301 in this last inscription could be identified with Λε[ωκ]ράτ[η]ς Παλλ(ηνεύς), father of an epistates of building accounts in the second half of the 33. For examples of ephebic or garrison dedications on hermaic bases from Athens, see IG II3.4 335 (Reinmuth 6; Friend T12), and Meritt 1940, no. 8 (Reinmuth 9; Friend T9); from Eleusis, see IG II3.4 337 (I.Eleusis 86; Clinton 1988); from Rhamnous, see IG II3.4 336 (I.Rhamnous 98; Reinmuth 13; Friend T10), IG II3.4 342 (I.Rhamnous 102; Reinmuth 10; Friend T14), and I.Rhamnous 95, 96, 107, and 108; see also Petrakos 1999, pp. 323– 324, on the popularity of dedications to Hermes at Rhamnous. 34. In her survey of the type, Evelyn Harrison notes (Agora XI, p. 125): “The representation in herm form of other divinities than Hermes seems to have begun as early as the 4th century, but it apparently started with those for whom the form had some special appropriateness, for Aphrodite and Pan because of the connection of the herm with fertility, or for Apollo Agyieus and Hekate because of their tradition as aniconic gate protectors.” 4th century, IG II2 1669, line 3. Κινησίας Δεκε[λειεύς] (column 3, line 7): ephebe, PAA 569990; AO 3. A rare name, best attested as that of the dithyrambic poet and public figure, a contemporary of Plato, Lysias, and Aristophanes (PAA 569985; LGPN II, 2; AO 2). Κτησικλῆς Κόπρειος (wreath 3): kosmetes, PAA 587170; AO 26. Μνησίθειος Κόπρ̣[ειος] (column 3, line 6): ephebe, PAA 656325; AO 34. Possibly the same as PAA 656329; LGPN II, 30; AO 35, father of Μνησίθεος Μνησι[θέ]ου Κόπρειος, PAA 656330; LGPN II, 31; AO 36, the younger being katapaltaphetes for the ephebes in 266/5, IG II3.1 91, line 27, and column III, lines 127–128 (IG II2 665; SEG XXXVIII 78). Μν̣η̣[σι]π̣τ̣ό̣λ̣εμος Ἀχερδούσιος (column 2, line 3): ephebe. Possibly the same as, or son of, Μνησιπτόλεμος Ἀχερδούσιος, PAA 657400; LGPN II, 3; AO 4, in IG II2 5861, grave stele of the second half of the 4th century. Ναυχαρίδης Ἐ̣λ̣[­ ­ ­] (either Ἐλευσίνιος or Ἐλαιούσιος, column 3, line 12): ephebe, PAA 702760; AO 1. A name previously unattested at Athens or elsewhere. Ξενόφαντο[ς] Ἐλευσίνιος (wreath 8): lochagos, PAA 733663; AO 10. Πολυσ[- - -] Πειραιεύς (probably Πολύστρατος, less likely Πολυσθένης, wreath 4): taxiarch, PAA 777305. Possibly the same as Πολύστρατος Πολυκλέο Πειραιεύς, PAA 781730; LGPN II, 29; AO 30, bouleutes 303/2, Agora XV 62, line 248. Πυθοκρ[- - -] (possibly Πυθόκριτος or Πυθοκράτης, column 3, line 2): ephebe, PAA 795152; AO 16. Possibly related to fellow Hippothontid tribesmen Πυθόκριτος Ἁμαξαντεύς, PAA 795160; LGPN II, 8; AO 12, lessor of a mine workshop, IG II2 2749, 4th century; or Πυθοκράτης Σμικύθου Ἐλευσίνιος, PAA 795115; LGPN II, 1; AO 1, known from a grave stele of the end of the 3rd century, SEG XXI 849. [. .]α̣θι̣[.]ς Πειραιεύς (column 2, line 4): ephebe, might be [Γν]άθι[ο]ς, a name well attested in the 4th century, PAA 279010–279080; LGPN II, 1–15; or possibly [Ἀγ]αθί[α]ς, a name popular in the Imperial period, attested for a bouleutes in the 2nd century b.c., PAA 102640; AO 1; IG II3.1 1328, line 95. [. . .]Ο̣νας Δεκελειεύς (column 2, line 5): ephebe, might be [Δημ]ονᾶς, a rare name probably attested for the 4th century, PAA 317440; LGPN II, 1; AO 1; IG II2 2414, line 11. Com men tary The D edic at ory M on u m en t A base for a herm or hermaic statue was a common form of dedication for the public display of honors by soldiers of a garrison, and specifically by ephebes of the Lykourgan era.33 The images erected on such bases were most often of Hermes himself, either in the form of hermaic pillars (the conventional herm with anthropomorphic head and phallus) or as hip-herms (anthropomorphic from the hips up, usually clothed), although both types could be used to depict other deities.34 Examples of hermaic bases with ephebic dedications in Athens bear dedications to “the hero,” who must 302 mar k munn be the eponym either of the tribe or of the relevant ephebic age class.35 In these examples, and in the case of most ephebic hermaic monuments where the deity to whom the monument is dedicated is not named (or the name is not preserved), it is impossible to say with certainty what image the hermaic monument depicted. The preservation of one such ephebic dedication from Rhamnous with its associated statue, a hip-herm of a youthful Hermes, and fragments of several similar hip-herms suggest that Hermes was most often the deity depicted.36 As the patron deity of gymnasia and as an archetype of athletic youth, Hermes would be especially appropriate for ephebic dedications.37 In the present case, the dedication of the hermaic monument to the Dioskouroi is, to my knowledge, unparalleled.38 While a divine image dedicated in the name of a deity is nearly always an image of the deity in question, there are instances of an image of one deity dedicated to another.39 That must be the case here; the single hermaic monument is not likely to represent the two Dioskouroi, but could appropriately depict Hermes, as Hermes and the Dioskouroi share many traits in common. The youthful Dioskouroi and the youthful Hermes, as he is regularly depicted by the later 4th century, are alike paradigms of young athletes, hunters, and protectors of travelers.40 Moreover, the evidence that this monument likely stood originally at or near the main gate of Panakton, near where it was found, further reinforces the likelihood that it represented Hermes. Herms were typically found beside doorways and gates, as Thucydides reports, and Hermes was often invoked as Propylaios.41 A hermaic base from Eleusis with ephebic dedication to an unnamed deity (likely to be another dyad, Demeter and Kore) was similarly found in the vicinity of the gate to the fortress and sanctuary of Demeter and Kore, although displaced from its 35. Habicht 1961, no. 2 (IG II3.4 335; Reinmuth 6; Friend T12), from the vicinity of the propylon of the Pompeion in the Kerameikos, is dedicated ἥρωι Μουνίχωι, line 5. Meritt 1940, no. 8 (Reinmuth 9; Friend T9), found in the Athenian Agora, is dedicated [τῶι ἥ]ρωι. A similar dedication [τῶι ἥρωι] is reasonably restored in IG II3.4 334 (Reinmuth 8; Friend T7). 36. The monument is a circular base with a dedication by the ephebes of Erechtheis enrolled in the year of Nikokrates, I.Rhamnous 98 (IG II3.4 336; Friend T10; earlier editions: IG II2 3105; Reinmuth 13; SEG XXXIX 185). The hip-herm is Athens, National Archaeological Museum 313 (illustrated in Kaltsas 2002, p. 253, no. 528); their discovery along with fragments of four more hip-herms below the retaining wall of the Sanctuary of Nemesis and along the road leading down to the fortress and the sea is reported by Stais (1891, pp. 56–62), recounted and illustrated by Petrakos (1976, p. 52, nos. 7–11; 1999, pp. 283–286, 324). The monument and its associated hipherm are discussed and illustrated by Palagia and Lewis (1989); they suggest that “the base and herms fell off the terrace of the sanctuary” (p. 338) and were originally therefore dedications to Nemesis. Petrakos (1999, p. 283), supposes that these monuments originally stood along the road where they were found. 37. On the common association of Hermes with gymnasia and with ephebes, especially from the 4th century onward, see Delorme 1960, pp. 339– 340, 364–365; Agora XI, pp. 124–129. 38. The closest parallel to such a hermaic dedication to the Dioskouroi is the depiction on a mid-5th-century red-figure skyphos of an aniconic pillar inscribed with the names of the two Dioskouroi, flanked by two young athletes with strigils; see Gaifman 2012, pp. 246–248, figs. 6.2, 6.3. 39. Dedications of images of one deity to a different deity are not common, but not without widespread parallels, as a review of the dedications described by Pausanias in major sanctuaries reveals; e.g., a statue of Apollo in front of the Parthenon (1.24.8); a statue of Artemis Leukophryene elsewhere on the Acropolis (1.26.4); a wooden Hermes in the sanctuary of Athena Polias (1.27.1). Further literary and archaeological examples are cited by Rouse (1902, pp. 391–393). 40. The Dioskouroi, Hermes, and Herakles are named together by Pind. Nem. 10.49–54, as champions of athletes and wardens (ταμίαι) of games at Sparta. On this association, and especially on the Dioskouroi as the paradigms of Spartan ephebes, see Kennell 1995, pp. 138–142. 41. Thuc. 6.27.1: ἐν ἰδίοις προθύροις καὶ ἐν ἱεροῖς. Identified elsewhere as Propylaios, Hermes by this name is preeminently associated with the Hermes at the Propylaia of the Acropolis; see Paus. 1.22.8; Agora XI, pp. 113–114, 122–124. insc r ip t ions f r om panakton 303 original location.42 At Rhamnous, Petrakos reports that herms and herm bases, several bearing dedications to Hermes, were the most common votive monument, being found in the sanctuary of Nemesis, at significant places in the fortress, and in private houses; he cites as examples one beside the gate to the acropolis and others erected by ephebes at the south gate, the main gate into the fortress.43 As the guardian of a doorway or passage, a herm need not be an accessory to a specifically demarcated cult place. The dedication of the herm to the Dioskouroi, however, suggests that there was a shrine or sanctuary of the Dioskouroi at Panakton, perhaps in the vicinity of the gate. The cult of the Dioskouroi was known in Attica, and worship of these youthful heroes and protectors of travelers would have been appropriate for ephebes at one of their garrison stations.44 Observance by ephebes of the Dioskouroi—or the Anakes, as they were also known in Attica—may relate to the use of the sanctuary of the Dioskouroi in Athens, the Anakeion, as a mustering point for ephebes. The Anakeion, an enclosure that lay below the east foot of the Acropolis, was large enough for a gathering of many men under arms and was a suitable mustering point for cavalry in the city.45 It lay immediately below the sanctuary of Aglauros, where the ephebic oath was administered.46 The association of the ephebes with the heroized sons of Zeus at the starting point of their tour of the sanctuaries of the city would have encouraged the ephebes to regard the Dioskouroi as models for ideal behavior.47 It is therefore appropriate to find them here commemorated at one of the terminal points of their two years of ephebic service. The E ph ebic In sc r ip t ion There is no doubt that this is an ephebic dedication of the Lykourgan era; it identifies honorands as ephebic officers—the sophronistes and the kosmetes along with cadet military officers, the taxiarchos and lochagoi—and contains several likely prosopographic links appropriate to the period.48 There are, 42. Orlandos (1954, pp. 9–10) describes the discovery of this base; Clinton (I.Eleusis, vol. 2, p. 96) deduces that “[t]his dedication was apparently set up in what was the most conspicuous place in Eleusis, the entrance to the sanctuary of the Two Goddesses, i.e., παρὰ τὰ προπύλαια (I.Eleusis 99, lines 25–26) or ἐν τῆι αὐλῆι (I.Eleusis 196, line 28).” 43. Petrakos 1999, p. 323, fig. 92; see I.Rhamnous 74, 100, 107, 108. 44. Testimonia to the cult of the Dioskouroi at various demes in Attica are gathered by Kearns (1989, p. 148), who notes that Pausanias (1.31.1) identifies them as the μεγάλοι θεοί in their cult at Kephale, and cites IG II2 1006, line 29, a decree of the Athenians honoring the ephebes in the year 122/1, as evidence that the ephebes participated in a procession for the Dioskouroi as μεγάλοι θεοί. 45. Thucydides (8.93.1) says that hoplites who marched from Piraeus to the city to demonstrate opposition to the oligarchy of the Four Hundred stationed themselves in the Anakeion. Andokides (1.45) reports that the Anakeion was the mustering ground for the cavalry in Athens. 46. Pausanias (1.18.1–2) locates the ἱερόν of the Dioskouroi (i.e., the Anakeion) below the τέμενος of Aglauros. On the enclosure of Aglauros as the place where ephebes took their oath, see Dem. 19.303 (On the False Embassy) and Plut. Alc. 15.4; on the oath itself, see Rhodes and Osborne 2003, no. 88; on the location of the enclosure of Aglauros in front of the cave on the east slope of the Acropolis, see Dontas 1983. 47. Aristotle (Ath. Pol. 42.3) reports that a circuit of τὰ ἱερά of the city accompanied the induction of young men into the ephebeia. Steinbock (2011) appraises the importance of this introduction of the ephebes to the gods and heroes of Attica as a means of contextualizing the values represented in the ephebic oath. On the Dioskouroi as paradigms of ephebic virtue, see n. 40, above. 48. See Ath. Pol. 42.2–5 on the institutions of the ephebeia, established or reformed by the law of Epikrates (Harp., s.v. Ἐπικράτης) dated to 335/4 (see Eretria XI, pp. 381–382; Chankowski 2014, pp. 19–26). On the conventions of the ephebeia as an institution of the Lykourgan era, see Pélékidis 1962, pp. 83–152; Reinmuth, pp. 123–138; Burckhardt 1996, pp. 26–75; Chankowski 2010, pp. 114–134; Friend. Chankowski 304 mar k munn however, some atypical aspects of the inscription that require consideration. The most anomalous feature is the fact that the sophronistes of this text, Ἰσοκράτης Παλληνεύς, comes from the tribe Antiochis and not from the tribe Hippothontis, whose ephebes make this dedication. This is contrary to the practice described in the Aristotelian Athenaion Politeia (42.2–3) and attested in all 10 of the 4th-century ephebic texts that preserve the relevant evidence, where the sophronistes is appointed from among the members of the same tribe as the ephebes.49 In the present case, the kosmetes, Κτησικλῆς Κόπρειος, does in fact come from the tribe Hippothontis. The kosmetes, who had charge of all ephebes from all tribes at any given time, was inevitably from the same tribe as one of the tribal cohorts. Perhaps the coincidence that this kosmetes was from Hippothontis accounts for the deviation from standard practice in this case; this hardly seems likely, however, since the duties of the sophronistes (receiving the stipend from the state for the ephebes of his tribe and managing their food purchases, as well as taking care of “all other things” that pertain to that tribe, according to Ath. Pol. 42.3) were so much more intimately tied to the young men in his charge than those of the kosmetes. The sophronistes was one of the tribesmen nominated specifically by fathers of the ephebes of that tribe for his recognized qualities, and therefore it is fitting that, when honors were awarded to the ephebes of a particular tribe for their excellence in service, their sophronistes was regularly named and honored at the same time. The present circumstance is therefore highly unusual.50 The second remarkable feature of this ephebic dedication is the comparatively small number of ephebes in the roster of dedicants. Allowing for the likely alignment of column 1 with column 2, there were 26 ephebes named on the roster. If we assume that five ephebic officers (one ταξίαρχος and four λοχαγοί) should be added to the roster, the total would be 31.51 If the ephebic officers were also named in the roster, as was sometimes the case, then the total number of ephebes would be just 26.52 Even in the more (2014, pp. 19–26, following Eretria XI, pp. 367–389, and Papazarkadas 2011, pp. 44–45, 112–117), convincingly suggests that revenues from the newly acquired territory of the Oropia provided the resources that enabled the Athenians to underwrite the expenses of this reorganized institution in 335/4. Friend (pp. 48–53) persuasively argues that a leading motive for mobilizing young men in a two-year program of military training and service specifically in 335/4 was the instability and lawlessness that prevailed along the Boiotian border after the destruction of Thebes in 335. He also argues (pp. 8–37) that the ephebeia was a new institution in 335/4; I favor the understanding of it as a reorganization of existing practices (see Pélékidis 1962, pp. 7–72; Reinmuth, pp. 123–138; Gauthier 1976, pp. 190–195; and the review of the question by Chankowski 2014, pp. 15–57). 49. The sophronistes can be identified by name with his ephebic tribe in IG II2 1156 (Reinmuth 2; Friend T2: Kekropis), IG II3.4 337 (Reinmuth 5; Friend T6; I.Eleusis 86: Kekropis), IG II3.4 335 (Reinmuth 6; Friend T12: Aiantis), Meritt 1940, no. 8 (Reinmuth 9; Friend T9: Leontis), Pritchett 1949 (Reinmuth 12; Friend T19: Oineis), IG II3.4, 336 (Reinmuth 13; Friend T10: Erechtheis), I.Oropos 353 (Reinmuth 15; Friend T15: Leontis), IG II2 1159 (Reinmuth 19: Pandionis), IG II2 1181 (I.Rhamnous 99; Friend T16: Aigeis), and the unpublished inscription from Rhamnous described by Petrakos (2004; Friend T8: Leontis). 50. This irregularity has prompted Humphreys to suggest that the inscription is post-Lykourgan (Humphreys 2004–2009, p. 84 [n. 3], 89), although the only post-Lykourgan ephebic texts of the 4th century that preserve this information, IG II2 478 (Reinmuth 17) and IG II2 1159 (Reinmuth 19), conform to the standard pattern by naming sophronistai from the same tribe as the ephebes. 51. Although contested by Sekunda (1992, pp. 327–330), the consensus among commentators is that the taxiarchoi and lochagoi named in ephebic texts were themselves ephebes, as argued by Roussel (1941, pp. 222–226) and Mitchel (1961, pp. 350–357), and accepted by Reinmuth (p. 22), Hansen (2006, pp. 302–304), and Friend (pp. 96–97). 52. Ephebic officers are named both separately and in the full roster of ephebes in the Kekropid dedication at Eleusis (IG II3.4 337; I.Eleusis 86; Reinmuth 5; Friend T6), where the officers are both receiving and giving honors; also in the Leontid dedication from Oropos (I.Oropos 353; Reinmuth 15; Friend T15). insc r ip t ions f r om panakton 305 likely case that there were 31 ephebes from Hippothontis on this dedication, this would rank among the smallest ephebic rosters attested; if the total were 26, it would be markedly below the lowest numbers attested for a tribal ephebic cohort.53 A small ephebic cohort might indicate that this was a dedication made early in the years of the ephebic institution established or reformed by the law of Epikrates in 335/4. Or there might be special circumstances that affected the tribe Hippothontis or the ephebes generally at the time this dedication was made.54 These circumstances might find an explanation if the date of the inscription can be established, and so the first challenge is to identify the archon named in the heading.55 The heading conforms to a standard formula of ephebic dedication of the Lykourgan era, [tribal name] ἔφηβοι οἱ ἐπὶ [archon’s name] ἄρχοντος ἀνέθεσαν, enabling us to suggest some likely restorations of the original text.56 Assuming that the archon’s name filled out the width of the base in the first line, we can calculate the number of letters in the archon’s name based on the spacing of letters in the heading.57 The center-to-center spacing of letters in the preserved portions of the heading is an average of 0.0163 m for ΝΤΙΔΟΣ in line 1, and in line 2 the spacing is 0.0153 m for ΝΤΟΣΔΙΟΣ, 0.0150 m for ΥΡΟΙΣΑ, and 0.0164 m when this is extended to ΥΡΟΙΣΑΝΕΘΕ. If we set aside the unusually extended letter spacing in 53. Dedications by the ephebes of Pandionis enrolled in 333/2 name an estimated 30–32 ephebes (IG II3.4 334; Reinmuth 8; Friend T7) and an estimated 28–31 in 332/1 (IG II3.4 342; Reinmuth 10; I.Rhamnous 102; Friend T14). The ephebes of Leontis enrolled in 333/2 numbered 38 (Meritt 1940, no. 8; Reinmuth 9; Friend T9; also in the unpublished text from Rhamnous, Petrakos 2004; Friend T8). Estimates for other ephebic rosters range from 42 up to 65; see Rhodes and Osborne 2003, p. 453, and Hansen 2006, pp. 34–35 (“On the whole, early lists are shorter than later lists”). 54. Unless evidence emerges to show that ephebic cohorts were divided into smaller groups at the time when honors were awarded by or to them, the possibility that the cohort listed on the present inscription is only part of that year’s full ephebic cohort of Hippothontis is unlikely. The ephebes of Leontis enrolled under Nikokrates (333/2) made a dedication to their tribal hero in Athens (Meritt 1940, no. 8; Reinmuth 9; Friend T9) and a dedication at Rhamnous (Petrakos 2004; Friend T8) with identical rosters of 38 ephebes (Petrakos suggests that the dedication in Athens was made in their first year of service, the dedication at Rhamnous at the end of their second year). The ephebes of Leontis enrolled sometime between 332 and 326 erected a monument at the Amphiaraion of Oropos, therefore in their second year of service, with evidently their full cohort of 62 ephebes (I.Oropos 353; Reinmuth 15; Friend T15). Two rosters of ephebes of Erechtheis enrolled under Nikokrates (333/2), one on a victory monument at Rhamnous (IG II3.4 336; Reinmuth 13; I.Rhamnous 98; Friend T10) and one from an unknown location (IG II2 2401; Friend T11), appear to match each other in numbers (see Palagia and Lewis 1989, pp. 333–337). 55. This inscription is to be dated to the early months of the year of the second archonship after the archon named in the heading; see Chankowski 2014, pp. 57–63. Ephebes were identified by the archon of their year of enrollment, and they served the second of their two years of ephebic training in the garrison forts and in the countryside. The ephebes of Hippothontis in this inscription from Panakton were ephebes who had completed their garrison service, and the honors recorded here were therefore awarded at the end of their second year, that is, in the first months of the second archonship after the archonship in which they were enrolled. Building on observations made by Clinton (1988, pp. 28–30), Chankowski (2014, pp. 67–73) discusses the implications of these circumstances for dating ephebic monuments. 56. Headings following this formula are IG II2 1189 (Reinmuth 3, disregarding the first line erroneously restored by Reinmuth, see Mitchel 1984; I.Eleusis 84; Friend T3) and IG II3.4 329 (Reinmuth 4; Friend T4), both honoring enrollments in the archonship of Ktesikles, and IG II3.4 335 (Reinmuth 6; Friend T12), from the archonship of Nikokrates. Other preserved headings of ephebic dedications follow an expanded version of the same formula, typically: οἱ ἔφηβοι οἱ τῆς [tribal name] οἱ ἐπὶ [archon’s name] ἄρχοντος ἀνέθεσαν, or a variation of this word order, sometimes with the addition of the σωφρονιστής as dedicant along with the ephebes; see IG II3.4 337 (I.Eleusis 86; Reinmuth 5; Friend T6), IG II3.4 334 (Reinmuth 8; Friend T7), Meritt 1940, no. 8 (Reinmuth 9; Friend T9), and IG II3.4 336 (I.Rhamnous 98; Reinmuth 13; Friend T10), all recording enrollments in the archonship of Nikokrates. 57. Headings that do not fill the full width of a base are normally justified to the center of a face, and so do not begin at the left margin. For examples, see I.Eleusis 74 and 88 (pls. 32, 40); I.Rhamnous 120 and 133. Wherever monuments are sufficiently preserved we see that ephebic headings normally fill the first line on the stone (cf. Reinmuth 6, pl. VI), even when the second line does not (cf. Reinmuth 14, pl. XVI). 306 mar k munn the second half of Διοσκούροις in line 2, this establishes a range of centerto-center letter spacing from 0.0150 to 0.0164 m, or an average of around 0.0158 m.58 Applying the evidence of letter spacing to the formula for the language of this heading, we can limit the possible restorations of the archon’s name with a high degree of probability. The single letter stroke preserved in the first line before the break at right is key to the restoration of the archon’s name. As noted above, that letter must be either kappa or chi. Possible archons of the era of the Lykourgan ephebeia with these letters in their names include Ktesikles (334/3), Nikokrates (333/2), Niketes (332/1), Kephisophon (329/8), Euthykritos (328/7), Chremes (326/4), Antikles (325/4), and Kephisodoros (323/2). If the letter in question were the first letter of the archon’s name (Ἱπποθωντίδος ἔφηβοι οἱ ἐπὶ Κ/Χ-), then the center-to-center letter spacing up to that point would average 0.0162 m, which is within the attested range. If the kappa or chi were the third letter of the archon’s name (Ἱπποθωντίδος ἔφηβοι οἱ ἐπὶ Νικ-), then the center-to-center letter spacing up to that point would average 0.0148 m, somewhat below the attested range; if it were the fifth letter (Euthykritos, Antikles), the spacing would be 0.0138 m, significantly below the attested range. These last two possibilities should be excluded from consideration, and the previously mentioned possibilities (Nikokrates, Niketes), should be considered less than likely. Calculating spacing from the letter in question to the original right edge of the stone as determined by the stone’s surviving width closer to the base, the genitive of the archon’s name would require a center-to-center spacing of 0.0372 m for Νικήτου (much too wide), 0.0233 m for Χρέμητος (also much too wide), 0.0206 m for Νικοκράτους (too wide), 0.0169 m for both Κτησικλέους and Κηφισοδώρου (acceptably close to the attested range), and 0.0155 m for Κηφισοφῶντος (within the attested range). Kappa is therefore certainly the letter in question, with three possible names restorable based on spacing: with an average center-to-center letter spacing of 0.0162 m before the kappa and either 0.0155 or 0.0169 m after the kappa, they are Ktesikles (334/3), Kephisophon (329/8), and Kephisodoros (323/2) (Fig. 10). One of these three should therefore be the archon of the enrollment year of this class of ephebes of Hippothontis.59 The likelihood of each possibility should be considered and compared. Ktesikles (334/3) If the enrollment year was that of Ktesikles, then these ephebes of Hippothontis will have been the first class of their tribe to serve under the newly reformed ephebeia. As attested by the dedication of the ephebes of Antiochis or Erechtheis enrolled in the archonship of Ktesikles,60 ephebes of this class will have served under Konon son of Timotheos of Anaphlystos as general in the Akte in their first year, and under Sophilos son of 58. The estimated spacing of [ΙΠΠΟΘΩ]ΝΤΙΔΟΣ is 0.0164 m, and that of [ΑΡΧΟ]ΝΤΟΣΔΙΟΣ is 0.0154 m, conforming to this range and this overall average for letter spacing. 59. It should be noted that there is no archon’s name later in the 4th century beginning with a kappa or a chi that would come close to a suitable fit if restored here. As to archons before 335/4, the names of Kephisodoros (366/5), Charikleides (363/2), Kallimedes (360/59), Kephisodotos (358/7), Kallistratos (355/4), and Kallimachos (349/8) could fit, but all are implausibly early for this dedication; Chairondas (338/7) is close enough in time, but his name is too short to be a good fit. 60. IG II3.4 329 (IG II2 2970; Mitchel 1964, pp. 349–350; Reinmuth 4; Friend T4). insc r ip t ions f r om panakton 307 Ktesikles Kephisophon Kephisodoros Figure 10. Possible restorations of the heading for inscription 2 Aristoteles of Phyle as general of the countryside in their second year.61 In that case, wreath 1 should read [στρατηγὸν / Σώφιλ]ο̣[ν / Φυλάσι]ο̣[ν]. Entering service a year after the destruction of Thebes by Alexander of Macedon, and in the year that saw Alexander’s first battles in Asia Minor, this class of ephebes will have seen wall-building work on Mounichia and general repairs to the walls of Piraeus either underway or recently completed.62 Also in this period, the fortress at Phyle was under construction or refurbishment,63 and likely work was underway at Panakton as well, where cohorts of the new ephebic corps were taking up their station for the first time. As an end-of-service dedication, this inscription will have been set up after the beginning of the archonship of Niketes (332/1), and as the first such end-of-service dedication of ephebes at Panakton (the first, at least, for the tribe Hippothontis), it would be appropriately placed in a prominent position near the main gate of the fortress. An enrollment in the archonship of Ktesikles would make this dedication a companion to IG II2 1189 (Reinmuth 3, revised by Mitchel [1984], and again by Clinton in I.Eleusis 84; Friend T3), a dedication by the same class of ephebes of Hippothontis and their sophronistes at Eleusis. Unfortunately, none of the names of the ephebes or of their sophronistes are preserved on the fragmentary stone from Eleusis, so a decisive identification eludes us. But the evident certainty that the Panakton dedication was inscribed by the same hand as that of the dedication by the same tribe at Eleusis (as noted above) makes this identification highly likely. If the Panakton dedication belongs to the same ephebic class as the one from Eleusis, then the sophronistes honored at Eleusis would be Ἰσοκράτης Παλληνεύς (a 61. Tracy (2016) gathers the epigraphic evidence for Sophilos’s distinguished career, highlighted by his service as στρατηγὸς επὶ τῆι χώραι in 334/3 and 333/2. 62. IG II3.1 429 (IG II2 244; Maier 1959, no. 10; Schwenk 1985, no. 3). 63. IG II3.1 429, line 10. 308 mar k munn possible restoration, along with his patronymic, in lines 8–9 in the Eleusis monument).64 Also in favor of the identification of the archon in the Panakton dedication as Ktesikles is the comparatively small number of ephebes on the roster. As noted above, their number can be compared to others within the first two years of ephebic enrollment, when it may be assumed that the newly instituted practice had not yet gained the currency that it would have within a few years.65 This early date could also account for the irregularity in the appointment of the sophronistes, again on the principle that the newly instituted practice had not become fully normalized in the manner described by the Aristotelian Athenaion Politeia. An identification of the Panakton dedication as that of the ephebes enrolled in the archonship of Ktesikles entails another consideration affecting the identity of the ephebic officers of that year. Following the discovery and publication by Mitsos of EM 13354 and its discussion by Reinmuth,66 the identification of the archon named at the head of the second decree on that stone has been the subject of much dispute. Mitchel, following Lewis’s observations, provided strong arguments for restoring the name of Ktesikles as archon.67 The inscription preserves the name of Autolykos of Thorikos, who is honored for his service as kosmetes (according to the plausible restoration). If Ktesikles is the correct restoration to both EM 13354 and to the present inscription from Panakton, which names Κτησικλῆς Κόπρειος as kosmetes, then we have two different kosmetai attested for the same enrollment class of ephebes (although honored by different tribes). Perhaps this irregularity is acceptable, given the evident irregularity in the assignment of a sophronistes in the present inscription. On the other hand, the evidence presented here could suggest that the restoration of Ktesikles’ name in EM 13354 is not correct, and that another name should be restored there. Chankowski has recently published carefully reasoned arguments for regarding the restoration of Ktesikles’ name in EM 13354 as uncertain, and has advanced reasons for considering the names of Kephisodoros (358/7), Agathokles (357/6), Apollodoros (350/49), and Lysimachos (339/8) as equally plausible restorations (with the consequent implication that the office of kosmetes existed prior to the reforms of Epikrates in 335/4).68 In short, whatever the resolution of the questions surrounding EM 13354 may be, there is no serious objection on account of that inscription to restoring Ktesikles’ name as archon in the present inscription from Panakton. Kephisophon (329/8) If the archon’s name is Kephisophon, then we may note that this class of ephebes was enrolled in the year in which Philon was the general responsible for the Drymos, that is, the Panakton command, according to the inscription that records the tithe in grain that he delivered from that territory to the treasurers of the Two Goddesses at Eleusis.69 In that year the first quadrennial games of Amphiaraos were held near Oropos, games in which the second-year class of ephebes would have participated.70 In that year, of course, this class of ephebes would not yet have been on the frontier, but would have been performing their service in basic training in Piraeus, either on the Akte or in the fort on Mounichia.71 For the following year of 64. Such a circumstance would favor the restoration of IG II2 1189 proposed by Clinton (I.Eleusis 84), where a longer heading allows room for the patronymic and demotic of the sophronistes to be listed. 65. See n. 53, above. 66. Mitsos 1965; Reinmuth 1. 67. Lewis 1973, p. 254; Mitchel 1975; see Friend T1. 68. Chankowski 2014, pp. 26–53. 69. IG II2 1672 (I.Eleusis 177), lines 271–272. On Drymos and Panakton, see Dem. 19.326, quoted above, with nn. 21, 22. 70. See Tracy 1995, p. 13, nn. 46, 47, on the games of Amphiaraos. On the participation of ephebes in the games at the Amphiaraion of Oropos, see I.Oropos 353 (Reinmuth 15; Friend T15); Humphreys 2004–2009. 71. Arist. Ath. Pol. 42.3. insc r ip t ions f r om panakton 309 garrison service at Panakton for these ephebes, in 328/7, we have no record of the στρατηγὸς ἐπὶ τὴν χώραν, the general whose name will have appeared in wreath 1. These years, in the aftermath of Antipater’s crushing of the Spartan uprising in 330, appear to have been quiet and uneventful, as far as military matters and the defensive needs of Attica were concerned. The mood of quiet resolve and the memory of resistance against Macedon in these years is reflected in Demosthenes’ forensic speech On the Crown, as well as in Lykourgos’s Against Leokrates, where ephebic duty to defend the homeland is invoked.72 Against that background, the honors awarded and recorded here—according to this dating, at the beginning of the archon year 327/6—will have been routine. Still, there are aspects of this inscription that do not conform to what we would expect of a routine end-of-service dedication. Ephebic inscriptions from other tribes enrolled in years close to 329/8 have rosters of 55 to 65 ephebes, so it would be unusual to have such a short roster of 26 ephebes and five ephebic officers, or 31 in all.73 In addition, it would be more difficult to account for the fact that the sophronistes and the ephebes in this inscription come from different tribes at a time when the institutions described in Ath. Pol. 42.2 should be fully normative.74 There is no evident circumstance about ephebic service in the years between enrollment in 329 and commemoration in 327 that could account for the unusual features of this ephebic text. Kephisodoros (323/2) The year of Kephisodoros was a momentous year for Athens, for its στρατηγὸς ἐπὶ τὴν χώραν, Phokion, and certainly for the ephebes who began their two years of service then.75 Near the beginning of their first year, in midsummer of 323, news of Alexander’s death in Babylon reached Athens. Within a few months, the Athenians led by Leosthenes had mobilized an army, invaded Boiotia, and by the onset of winter were besieging Antipater in Lamia. At some point in the following summer, a Macedonian force commanded by Mikion landed at Rhamnous and began devastating the countryside. Phokion, who had at his command three of the 10 tribal regiments as well as the second-year ephebes, engaged, defeated, and killed Mikion and put an end to this incursion.76 The ephebes enrolled under Kephisodoros presumably remained for their first year on duty in Piraeus, 72. Both speeches were delivered in 330. Invoking the ephebic oath: Lycurg. Leoc. 76–77; on Lykourgos’s development of this theme, see Steinbock 2011. 73. Numbers of ephebic tribal cohorts are displayed in a table by Hansen (2006, p. 34; see also Rhodes and Osborne 2003, p. 453). See n. 53, above. 74. The 10 inscriptions conforming to the normative practice of selecting a sophronistes from the same tribe as the ephebes are listed in n. 49, above. Among them, IG II2 1181 (I.Rhamnous 99; Friend T16) and Pritchett 1949 (Reinmuth 12; Friend T19) date to the two enrollment years immediately preceding that of Kephisophon. 75. Phokion’s appointment as στρατηγὸς ἐπὶ τὴν χώραν is an inference from his role in driving off the Macedonians who landed at Rhamnous (Plut. Phoc. 25). See Munn 1993, pp. 190–194, for the argument that Phokion must often have been elected to the post of στρατηγὸς ἐπὶ τὴν χώραν over the course of his long career. 76. Diodorus Siculus (18.10.2) reports the division of Athenian forces for expeditionary action and homeland defense: στρατεύσασθαι δὲ πάντας Ἀθηναίους τοὺς μέχρι ἐτῶν τεσσαράκοντα καὶ τρεῖς μὲν φυλὰς τὴν Ἀττικὴν παραφυλάττειν, τὰς δ’ ἑπτὰ πρὸς ὑπερορίους στρατείας ἑτοίμους εἶναι ([it was decreed that] all Athenians up to the age of forty should perform military service, and three tribes should guard Attica while seven tribes should be ready for expeditionary service abroad). Second-year ephebic cohorts must have been among those whose duty it was τὴν Ἀττικὴν παραφυλάττειν; they were probably assigned to hold fortified positions while soldiers in the 22- to 40-year age classes engaged the Macedonians in the field. 310 mar k munn where vigilance against Macedonian naval activities was required. By midsummer the war in Thessaly began to go badly, and when, at the end of summer, this class of ephebes was transitioning to garrison duty on the frontiers of Attica, the Macedonian army under Antipater had arrived in Boiotia and was preparing to enter Attica.77 If the archon’s name is Kephisodoros (323/2), then these ephebes of Hippothontis were poised to transition to garrison service on the frontiers at the very time when the Athenians accepted Antipater’s terms: (1) surrender the leading anti-Macedonian politicians, Demosthenes and Hypereides chief among them; (2) cede the territory of Oropos and the sanctuary of Amphiaraos to the Boiotians; (3) accept a Macedonian garrison in the Mounichia fort of Piraeus; and (4) limit Athenian citizenship to those who possessed property valued at or above 2,000 drachmas.78 This last measure drastically reduced the number of fully enfranchised Athenians. Plutarch reports that over 12,000 Athenians lost their citizenship.79 Diodorus Siculus gives the number as more than 22,000, and states that about 9,000 Athenians were thereby “designated masters of the city and the countryside, and conducted the government in accordance with the laws of Solon.”80 Whatever the actual number of disenfranchised was, the aim of this purge of the citizen rolls was clearly to leave all effective authority in the hands of propertied Athenians who were more likely to be compliant with Macedonian interests (assured, of course, by the Macedonian garrison in Mounichia), while ostensibly leaving the laws of Athens otherwise in place.81 What happened to the ephebeia? All commentators admit that there is no direct evidence as to whether it survived these radical changes or not. Habicht comments: “If the corps of ephebes continued to exist at all, it must have lost members and potential recruits in the same proportion as the citizenry lost citizens with full rights, namely about four-sevenths.”82 The ephebic corps, in other words, would fall to something under half the size of what it had been. When we compare the number of ephebes in this inscription, most likely 31 in all, including ephebic officers, to the attested tribal divisions of 55–65 just a few years earlier, we find a difference that would correspond to about half of what had been the norm. If this ephebic text is the record of a second-year class that began its service in 323 and 77. Ephebes entered service and transitioned from first- to second-year duties at the beginning of Boedromion; see Pélékidis 1962, pp. 110–111; Chankowski 2014, pp. 57–67. On the course of the Lamian war, the date of the decisive battle of Krannon (7th of Metageitneion, 322), the surrender of Athens and the arrival of the Macedonian garrison in Piraeus (22nd of Boedromion, 322), see Ferguson 1911, pp. 16–20; Tracy 1995, pp. 23–29; Habicht 1997, pp. 36–41. 78. Diod. Sic. 18.18.1–6; Plut. Phoc. 27–28.1; Plut. Dem. 28–29; Ferguson 1911, pp. 19–20; Habicht 1997, p. 40. 79. Plut. Phoc. 27.4. 80. Diod. Sic. 18.18.5: οἱ δὲ τὴν ὡρισμένην τίμησιν ἔχοντες περὶ ἐννακισχιλίους ἀπεδείχθησαν κύριοι τῆς τε πόλεως καὶ χώρας καὶ κατὰ τοὺς Σόλωνος νόμους ἐπολιτεύοντο (Those possessing the specified property evaluation, being about 9,000 in number, were designated masters of the city and of the countryside and conducted the government in accordance with the laws of Solon). 81. Ferguson and Habicht (n. 78, above) favor the figure reported by Plutarch. Hansen (1985, pp. 28–36; 2006, pp. 19–60), having given close consideration of these and other population figures, gives credence to the uncorrected numbers reported by Diodorus. 82. Habicht 1997, p. 45. Ferguson (1911, p. 22), followed by Pélékidis (1962, p. 157), gives the same figure for the hypothetical reduction of the ephebic corps. insc r ip t ions f r om panakton 311 completed its year of garrison duty in the summer of 321, then it apparently reflects a reduction in the size of the ephebic class nearly proportional to that of the citizen body as a whole in 322/1.83 The foregoing reasoning could account for the comparatively small cohort of ephebes from Hippothontis listed in this inscription.84 The irregular circumstances that must in any event have affected the ephebic classes that were in service during and at the end of the Lamian War could also account for the irregularity of having a sophronistes from a different tribe overseeing the Hippothontid ephebes at the end of their second year of service. An original sophronistes from Hippothontis might have been someone who was disenfranchised in 322/1, or he may have been a casualty of war in that year. The sophronistes named in this dedication, Ἰσοκράτης Παλληνεύς, might be identified with the man of that name who is otherwise attested as one of the nine archons (a thesmothetes) on Samos in an inscription of the mid-4th century. If he were the same man, by 322/1 he would have been back in Athens as a consequence of the Athenian surrender to Antipater.85 If he were among those expelled at this time, as a man of some standing, Isokrates of Pallene would have been suitable for a post of responsibility.86 Conclusions Three epigraphically plausible possibilities for the date of this inscription are thus under consideration, and while definitive proof is lacking, the relative likelihood of each of these possibilities can be compared and a most probable dating may be suggested. The inscription presents certain peculiarities—the anomalous appointment of the sophronistes and the comparatively small size of the ephebic cohort—that strongly tend to favor either an early date, when the institutions of the newly reformed ephebeia were not yet entirely normalized, or a late date, at a point in time when those institutions were severely disrupted. The possibility that this inscription is to be associated with ephebes enrolled in the archonship of Kephisophon (329/8), when the institutions of the Lykourgan ephebeia 83. Pélékidis (1962, p. 157) makes the hypothetical calculation that after 323/2, if the ephebeia survived, the total number of ephebes from all 10 tribes would be reduced from 650–700 each year to around 300. The 31 ephebes of Hippothontis, if the present inscription is to be dated to this time, conforms well to one-tenth of that estimate. Following Hansen’s acceptance of the even higher proportion of disenfrachisements, the less severe drop in numbers of this ephebic cohort could be accounted for by the presumption that ephebes who came from families whose properties would entitle them to retain citizenship were disproportionately represented in the ephebate. I follow Hansen in supposing that ephebes were enlisted regardless of wealth, and with him I also suppose that total enrollment of all eligible young men was never actually achieved; see Hansen 2006, pp. 35–38. 84. It is also possible to speculate that the small cohort of ephebes recorded here were the survivors of fighting along the borders in advance of the arrival of Antipater, assuming that as soon as they were deployed to the frontier garrisons they became engaged in fighting, presumably with raiders anticipating the advance of the main Macedonian army. It is hard to imagine, however, that such raw recruits would have been expected to face the army that had just destroyed the Athenian and allied forces at the battle of Krannon. 85. In 322 the Exiles Decree promulgated by Alexander in 324 was finally imposed upon the Athenians, and they were forced to receive back at Athens all of the citizen cleruchs who had been settled on Samos for the previous 40 years, at least all of those who still qualified for Athenian citizenship under the new regime. See Habicht 1997, pp. 31–33, 41–42, with references to the relevant textual and epigraphic sources. 86. See the references to Ἰσοκράτης Παλληνεύς on pp. 300–301, above, and note the possibility that his son went on to become a bouleutes at Athens a generation later. 312 mar k munn were flourishing, seems decidedly less probable than either the early or the late dating. The late dating, which would make this the final record of ephebes who were enrolled under Kephisodoros (323/2) and performed their second year of service at Panakton in 322/1 at the very time when the Lamian War devolved into a catastrophe for the Athenians, has a number of circumstantial features that make this possibility worth consideration. But the circumstances are so exceptional that we may well wonder if what otherwise appears to be a conventional memorial to their service is indeed likely to be so. Would an ephebic class constituted under the influence of this fundamentally democratic institution, and under the influence of a patriotic fervor that must have prevailed as the Athenians resolved to pit themselves against the Macedonians, have survived, even in truncated form, the trauma of defeat, surrender, occupation, political disruption, and social dislocation that befell the Athenians at the end of the summer of 322? The ephebes and their officers, if this were a dedication by the survivors of either the citizenship purge or the fighting in 322, would have been hard put to show the esprit de corps that would have been fostered in their first year together. An argument could be made that the oligarchic regime imposed by the Macedonians, sometimes referred to as the oligarchy of Phokion, might accommodate a revised form of the ephebeia as a training ground for the elite guardians of the new regime.87 But as yet there is no evidence that such an institution did in fact survive.88 The improbabilities of this inscription being a product of this traumatic episode at least balance the circumstantial evidence that makes the possibility worth consideration. If a stronger case without counterbalancing improbabilities can be made, then that one should prevail. The case for this inscription being an end-of-service memorial and dedication by ephebes of the first cohort to fulfill their duties under the newly instituted ephebic reforms, those enrolled under Ktesikles in 334/3, is more consistently plausible than either of the two alternative possibilities. With the hermaic image that it once supported, this base will have stood near the gate of Panakton in a place of prominence, appropriate to one of the first such monuments of an institution that would persist for another decade. The small size of this ephebic cohort comports with an early date, and is not out of line with the sizes attested for other ephebic classes. The prosopographical evidence does not provide sufficient grounds to associate any of the individuals named on this stone with the ephebeia in either 87. Phokion, together with Xenokrates, head of the Academy, was largely responsible for negotiating the terms of the settlement that Antipater imposed on Athens in 322 (Plut. Phoc. 27, cf. 29.4). Phokion had overseen the ephebes as στρατηγὸς ἐπὶ τὴν χώραν on many occasions, and most immediately in the year 323/2, which ended in this settlement (see n. 75, above). Phokion was influenced by the Academy and certainly familiar with Plato’s concept of a class of guardians as protectors of the state against threats both internal and external (see Tritle 1988, pp. 50–55, 141–145; on the potential exaggeration of this connection by posterity, see Gehrke 1976, pp. 180–196; Bearzot 1985). Phokion might, therefore, have been an advocate of an ephebeia that could conform to the values of the government imposed by Antipater. But a different scenario is equally plausible. Friend (pp. 174–176) argues that Phokion would have viewed the ephebes, inducted under the democracy, to be natural opponents of the oligarchy he had helped to create, and therefore would likely have worked to dissolve this institution. 88. Mitchel (1964) is skeptical that the ephebeia would have survived. He interprets IG II2 1187 (IG II3.4 339; I.Eleusis 99) as indicating that vestiges of the ephebeia had to be maintained through private initiative in the era following the battle of Krannon. 313 insc r ip t ions f r om panakton 334–332 or 323–321,89 although the possible restoration of the name in the first wreath as [στρατηγὸν / Σώφιλ]ο̣[ν / Φυλάσι]ο̣[ν] favors the earlier dating.90 The oddity of the sophronistes remains a peculiar feature, but can be understood as an irregularity arising from the formative process of establishing this institution. Decisive, in my view, is the conclusion noted above that this stone was inscribed by the same hand that inscribed the dedication at Eleusis by the same tribal cohort that was enrolled in the archonship of Ktesikles.91 Panakton was closely dependent on Eleusis, and was within the same regional command when the generalship of the countryside was later divided between the στρατηγὸς ἐπὶ τὴν χώραν τὴν ἐπ’ Ἐλευσῖνος (general in charge of the countryside toward Eleusis) and the στρατηγὸς ἐπὶ τὴν χώραν τὴν παραλίαν (general in charge of the countryside toward the coast).92 Ephebes who were stationed at Eleusis were certainly also stationed at Panakton during the year in which they “patrol[led] the countryside and perform[ed] duties in the garrison posts.”93 The ephebes of Hippothontis enrolled under Ktesikles in 334/3 appear to have made end-of-service dedications at the beginning of the archonship of Niketes, in 332/1, at each of these garrison posts.94 3 Dedication with honorific decrees and roster of the ephebes of Leontis Fig. 11 Panakton 1988-1. Fragment of a white (“Pentelic”) marble stele found on July 16, 1988, in the rubble filling the nave (east room) of the church in 20 m square J-11 (see Fig. 2). Left edge preserved, broken on all other sides. Face is very worn, chipped, and pitted. H. 0.240; W. 0.120; Th. 0.055–0.060 m. L.H. lines 1–9 0.0055 m, lines 10–21 0.0040 m. Notices in Munn 1988, p. 366; Munn and Munn 1991, p. 14; SEG XXXVIII 67; PAA 1995–2012, sub nominibus; Friend T23. 89. Ἰσοκράτης Παλληνεύς might just as likely have been the same as the thesmothetes from the cleruchy on Samos whether he was appointed in 334 or in 322. 90. See p. 307, above, with n. 61. 91. IG II2 1189 (I.Eleusis 84; Friend T3), by the Cutter of IG II2 337, see p. 295, above, with nn. 29, 30. 92. The division of commands was instituted sometime between the 280s and the 260s; see Oliver 2007, pp. 164–165, n. 42; Oetjen 2014, p. 11. For Panakton in the Eleusinian district, see, e.g., IG II3.4 281 (IG II2 2971; I.Eleusis 195) and IG II2 1299 (I.Eleusis 196). 93. Ath. Pol. 42.4: περιπολοῦσι τὴν χώραν καὶ διατρίβουσιν ἐν τοῖς φυλακτηρίοις. 94. A variety of factors could result in multiple dedications at the end of service. One consideration is the recording of honors granted by different authorities at different locations. So, e.g., the ephebes of Kekropis enrolled under Ktesikles received honors awarded by their tribe, by the council, by the demesmen of Eleusis, and by one of the tribal demes (all recorded together on the dedication to the tribal hero, IG II2 1156; Reinmuth 2; Friend T2; Rhodes and Osborne 2003, no. 89; the possibility that one or more of these awards was also recorded elsewhere cannot be excluded). Another factor is the deity or hero whom the ephebes wish to gratify with a dedication. Several ephebic dedications are made “to the Hero,” which is either to the eponym of the age class or to the tribal eponym (see n. 35, above), and could appropriately have been the practice of every ephebic class. But other dedications are made to the deity or hero most venerated at one of their stations of service, such as the dedications at Rhamnous, which are presumed to be to Nemesis (IG II3.4 336; Reinmuth 13; I.Rhamnous 98, with Palagia and Lewis 1989; also Petrakos 2004; also IG II3.4 339; I.Rhamnous 99; Friend T16); at Oropos to Amphiaraos (I.Oropos 353; Reinmuth 15; Humphreys 2004–2009; Friend T15), and at Eleusis to Demeter and Kore (IG II2 1189; Reinmuth 3; I.Eleusis 84; Friend T3; also IG II3.4 337; Reinmuth 5; Clinton 1988; I.Eleusis 86; Friend T6). There is no evident reason why an endof-service dedication at one sanctuary should preclude the possibility of others elsewhere. The variety in ephebic dedications, most of them end-of-service, is reviewed by Humphreys 2004–2009, pp. 87–90. 314 mar k munn 332–323 b.c. Non-stoich. Unknown number of lines missing. [κ]αλῶ̣[ς καὶ φιλοτίμως? - - - - - - - - ἐπαινέσαι - - - - - - - - - - - - - - - - - - ἀρετῆς?] ἕνεκα̣ [καὶ - - - - - - - - - - - - - - - καὶ στεφανῶσαι χρυσῶι στεφάνωι ἀπὸ - - - - - - -] [δ]ραχμ[ῶν - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -] vacat 0.008 m [. . .]σιάδ[ης εἶπεν· ἐπειδὴ ?- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - τοῖ]5 [ς ἐφ]ήβοις το[ῖς - - - - - - - - - - - - - - - - - - - - - - - - - - - - ἐπαινέσαι - - - - - - - - - - -] [. . σ]τρατου [- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ἀρετῆς] [ἕν]εκα καὶ φιλ̣[οτιμίας - - - - - - - - - - - - - καὶ στεφανῶσαι χρυσῶι στεφάνωι ἀπὸ] [χι]λίων δραχμ[ῶν - - - - - - - - - - - - - - - - - - ἀναγράψαι δὲ τόδε τὸ ψήφισμα εἰς] [τὸ] ἀνάθημα τὸ τ[ῶν ἐφήβων τῆς Λεωντίδος φυλῆς - - - - - - - - - - - - - - - - - - - - -] 10 [Δ]ειραδιῶται [Φ]ανόστρατος Δημο[- - - - - - - -] [Φ]ιλωνίδης Φιλο[- - - - - - - - -] [Ἀ]λκ̣ιβιάδης Αἰσ[- - - - - - - - -] [Κ]ρ̣ω̣π̣ίδαι 15 [Κά]λ̣λιστο[ς] Προστίμου [Ἱπ]ποκρά[τη]ς Ἱπποκρά[τους] [Κ]ηφισό̣δ̣ω̣[ρο]ς Κηφισοφῶ[ντος] Σόφων Ξενάρχου [.]έανδρος [Σ]ωδάμου 20 [ἐ]ξ̣ Οἴου̣ [.]λ̣οδη̣[. . .] Χ̣αρ̣ίου Θεοτιμίδ̣η̣ς Ἀστυνόμου Π]οτάμιοι Εὐθύμαχος Πύρρου 25 [. . . .]τρ̣[. . .]ς Πύρρου [.]α̣[­ ­ ­ ­ ­ ­ ­] Καλλίου [.]α̣ρ̣[.]α̣[­ ­ ­ ­ ­ ­]ονος [.]ο̣ς̣[­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­] [.]ο̣[­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­ ­] Ep ig rap hic al Not e s Letters are slightly larger and more widely spaced in the body of the decrees, lines 1–9, than they are in the roster below. Where letter shapes are clearly discernible they resemble the close spacings and shapes described by Tracy as the work of the Cutter of IG II2 1176.95 Rho (lines 15, 24, 25), upsilon (lines 19, 22, 24, 25, 26), and phi (lines 12, 17, 18) are consistently slightly taller than other letters. Omicron is more deeply cut and sometimes has an ovoid shape that Tracy likens to the shape of a rugby ball (end of line 19). Letters in the lines with deme names (10, 14, 20, 23) are more widely spaced than elsewhere in the roster. Line 1: At right, the left splayed foot of omega is discernible. Line 2: At right, the lower part of the left diagonal of alpha is discernible. Line 4: The sigma is secure, so [. . .]σιάδ[ης] must be a personal name since no demotic ends in -σιάδης. If the spacing is not uniform, [Λυ]σιάδ[ης] is a possibility. 95. Tracy (1995, pp. 129–131) places his work ca. 330–324/3. Based on his examination of a squeeze, Tracy is not able to confirm this attribution, but notes (pers. comm.) “[t]his lettering is absolutely characteristic of the very small lettering of the accounts and inventories of the 340’s and 330’s,” and refers to Tracy 1995, p. 79. insc r ip t ions f r om panakton 315 Figure 11. Inscription 3 (1988-1). Scale 1:2 Line 7: At right, the lower end of the left diagonal of lambda is discernible. Line 13: At left, the upright of the kappa is discernible. Line 14: At left, the top of the rho, the top of a round letter, and the upright at the top of the pi are discernible. Line 15: At left, the lower end of the right diagonal of lambda is legible. All letters in the patronymic are legible. Line 17: In the sixth to eighth space a circular, triangular, and circular letter are discernible. Line 20: This line occupies a smaller space than normal between the lines above and below. At left, a rectangular letter with cross-strokes at top and bottom, probably xi, and at right a vertical of the upsilon are discernible. Line 21: At left, the first and third discernible letters are triangular; after the third two verticals can be discerned; [Φι]λ̣όδη̣[μος] is a possible restoration. After the break there are traces of the right diagonals of chi and a vertical and top cross stroke of rho. Line 22: The eighth letter is triangular, followed by the lower ends of two vertical strokes. Line 25: Possibly either [. . . σ]τρ̣[ατο]ς or [. . . .]τρ̣[ιδη]ς, the former much more likely than the latter. Among many possibilities the best fits are [Ἡγέσ]τρ̣[ατο]ς and [Λεώσ]τρ̣[ατο]ς. 316 mar k munn Line 26: Possibly [Κ]ά̣[λλιππος] Καλλίου; see below on the prosopography of Καλλίας Ποτάμιος. Line 27: All letters are faint and uncertain except for the last four letters of the patronymic, which is possibly restorable as [Φιλόφρ]ονος; see below. Lines 28–29: All letters are very faint and uncertain. Pr os op o g rap hy Traill was able to examine this inscription soon after its discovery was announced, and I am indebted to him for improvements in the readings of names in the roster. I list the readable names, including patronymics, in alphabetical order with references to PAA, LGPN where applicable, and the current numbering in AO. Ἀ]λκ̣ιβιάδης Αἰσ[- - - - - -], Δειραδιώτης (line 13): ephebe, PAA 121550; AO 7. Εὐθύμαχος Πύρρου, Ποτάμιος (line 24): ephebe, PAA 433685; AO 26. He and his brother listed after him, [. . . .]τρ̣[. . .]ς Πύρρου (line 25), were members of the section of the deme Ποταμός known as Ποταμὸς Δειραδιώτης, as demonstrated by the demotic of their father, Πύρρος [Εὐθ]υμάχου, Ποτάμιος Δειραδιώτης, PAA 796855; LGPN II, 38; AO 44, known from IG II2 1752, line 27; Agora XV 52, line 29, a prytany list of Leontis of ca. 325.96 The father of our ephebes, Πύρρος [Εὐθ]υμάχου, is kinsman of Παγκλῆς Ποτάμιος Δειραδιώτης, PAA 760830; LGPN II, 2; AO 2, who is father of two brothers, Πύρρος Παγκλέους Ποτάμιος Δειραδιώτης, PAA 796852; LGPN II, 39; AO 45, and Φιλόφρων Παγκλέους Ποτάμιος Δειραδιώτης, PAA 952070; LGPN II, 9; AO 11, who were ephebes sometime between 332 and 324; see Reinmuth 15, column II, lines 18–19; I.Oropos 353, lines 53–54; Friend T15, and see n. 98, below, on the date of this inscription from the Amphiaraion of Oropos. Θεοτιμίδ̣η̣ς Ἀστυνόμου, ἐξ Οἴου (line 22): ephebe, PAA 510143; AO 4. Father cited as PAA 223205; AO 6, possibly the same as Ἀστύνομος ἐξ Οἴου, PAA 223200; LGPN II, 2; AO 2, recorded in IG II2 1553, lines 18–20, as a prosecutor of a metic woman for ἀπροστασία ca. 330, as interpreted by Meyer (2010, p. 144); and probably the same Ἀστύνομος Λυσιάδου ἐξ Οἴου, PAA 223210; LGPN II, 2; AO 2, dedicant in the theater of Dionsysos, mid-4th century, IG II2 3831. See also Λυσιάδης ἐξ Οἴου, PAA 613620; LGPN II, 30; AO 33, recorded in IG II2 1593, line 18, as a guarantor of a tax-buyer, mid-4th century, as this text is interpreted by Papazarkadas (2011, pp. 285–290). [Ἱπ]ποκρά[τη]ς Ἱπποκρά[τους], Κρωπίδης (line 16): ephebe, PAA 538540; AO 26. Father cited as PAA 538539; AO 25. Καλλίας Ποτάμιος (Δειραδιώτης) (line 26): for the demotic, see under Εὐθύμαχος Πύρρου, Ποτάμιος above. Father of an ephebe, PAA 555095; AO 200, probably the same as Καλλίας Καλλίππου Ποτάμιος, PAA 555100; LGPN II, 185; AO 199, known from a grave stele of the 4th century, IG II2 7264, as the father of Κάλλιππος, PAA 559460; LGPN II, 63; AO 70, who may be the ephebe named here if this line is correctly restored as [Κ]ά̣[λλιππος] Καλλίου. 96. There were three sections of Ποτάμιοι in the tribe Leontis (Π. καθύπερθεν, Π. ὑπένερθεν, and Π. Δειραδιῶται; see Traill 1975, pp. 44–45 [n. 18], 127–128). They are usually distinguished in Leontid lists (as in IG II2 1742 = Agora XV 13; IG II2 1752 = Agora XV 52; I.Oropos 353 = Reinmuth 15, Friend T15), but otherwise members of these deme sections are simply called Ποτάμιοι, as they evidently are on this roster. insc r ip t ions f r om panakton 317 [Κά]λ̣λιστο[ς] Προστίμο[υ], Κρωπίδης (line 15): ephebe, PAA 560595. Father cited as PAA 790550. Πρόστιμος is a name not otherwise attested in Attica or elsewhere. Προσ- names are rare until the Imperial period, when Προσδόκιμος, semantically similar to Πρόστιμος, becomes fairly common. The reading of ΠΡΟΣΤΙΜΟ on this stone is secure. [Κ]ηφισό̣δ̣ω̣[ρο]ς Κηφισοφῶ[ντος], Κρωπίδης (line 17): ephebe, PAA 568530. Father cited as PAA 569262; AO 35. [.]έανδρος [Σ]ωδάμου, Κρωπίδης (line 19): ephebe, cited in PAA 602730 as [ Λ?]έανδρος, and as [ Ν?]έανδρος in PAA 703005; AO 6. Father cited as PAA 854605; AO 3. Σόφων Ξενάρχου, Κρωπίδης (line 18): ephebe, PAA 829535. Father cited as PAA 731150; AO 1. [Φ]ανόστρατος Δημο[- - - - -], Δειραδιώτης (line 11): ephebe, PAA 917070; AO 14. [Φ]ιλωνίδης Φιλο[- - - - - -], Δειραδιώτης (line 12): ephebe, PAA 957265; AO 32. Χαρίας ἐξ Οἴου (line 21): father of an ephebe, PAA 981470. The son’s name might be restored as [Φι]λ̣όδη̣[μος] (the spacing of ΦΙ being a plausible fit), a name attested for Οἶον (Κεραμεικόν) in the 1st century b.c., IG II2 2462, line 6. [- - - -]ονος, Ποταμίου (line 27): father of an ephebe. The father is likely to be restorable as [Φιλόφρ]ονος, as a Φιλόφρων is attested as a kinsman of the two brothers, sons of Πύρρος Ποτάμιος, listed above in lines 24 and 25 (see Εὐθύμαχος Πύρρου, Ποτάμιος, above), and a grave stele of the 4th century, IG II2 7269, names a Φιλόφρων, PAA 951980; LGPN II, 4; AO 4, as a relative by marriage of [Παγ]κλῆς Πύρρο⟨υ⟩ Ποτάμιος, PAA 760835; LGPN II, 2; AO 2. [. . .]σιάδ[ης], proposer of the second decree (line 4), could be restored as one of several attested names: [Κτη]σιάδ[ης], [Μνη]σιάδ[ης], [Παυ]σι­ άδ[ης], [Τει]σιάδ[ης], and among those attested as members of Leontis, [Δαι]σιάδ[ης], [Σω]σιάδ[ης], and [Λυ]σιάδ[ης], among whom Λυσιάδης (allowing slightly wider letter spacing at the head of this decree) is the most probable, as the name is attested in the family of the ephebe Θεοτιμίδ̣η̣ς Ἀστυνόμου, ἐξ Οἴου (line 22); see above. 97. Of Kekropis, under Ktesikles: IG II2 1156; Reinmuth 2; Friend T2; Rhodes and Osborne 2003, no. 89; under Nikokrates: IG II3.4 337; I.Eleusis 86; Reinmuth 5; Friend T6. Of Leontis, under Nikokrates: Meritt 1940, no. 8; Reinmuth 9; Friend T9. 98. I.Oropos 353 (Reinmuth 15; Friend T15) is dated by Reinmuth to the enrollment class of 324/3, but that dating is problematic. It is more likely that it dates to an enrollment class between 332/1 and 326/5, and was dedicated between 330/29 and 324/3, as argued by Humphreys (2004–2009) and Friend (T15). Commen tary This fragment is part of an ephebic monument similar to those of the ephebes of Kekropis enrolled under Ktesikles and under Nikokrates, and the ephebes of Leontis enrolled under Nikokrates, which are the better preserved among other examples of such monuments to ephebic service.97 Like those, this is certainly a monument of the Lykourgan era, and might belong to enrollment classes from the year of Ktesikles (334/3) to that of Antikles (325/4), excluding only the year of Nikokrates (333/2) and the year of I.Oropos 353 (Reinmuth 15; Friend T15), which is uncertain.98 As an end-of-service dedication, this inscription is datable either to 332 or between 330 and 323. Preserved in the first nine lines are fragments of two honorific decrees, the last line of which refers to the dedication where these decrees are to 318 mar k munn be inscribed, indicating either the present stele or another monument elsewhere, of which the present stele is a copy. Other decrees may have preceded these two on the stele, as its full dimensions are unknown. An estimate of the width of the stele, as well as the number of ephebes, can be made based on the space occupied by the surviving portion of the ephebic roster. At least 15, probably 16, names of ephebes are preserved in whole or in part in the left column, and we may reasonably conjecture that the column originally contained four or five more names. If there were three similar columns on the original stele, then the number of ephebes would have been about 38 at a minimum, if the third column were only half full, and could have been up to around 62. Based on the nearly complete column width preserved here we may estimate that the width of a column, with small margins, was 0.13 to 0.14 m.99 Assuming that there were three such columns on this stele, the original width would have been ca. 0.39–0.42 m. This would indicate that the line length in the decrees was approximately 50–54 letters.100 On the basis of the wording and names likely to have been included in the formulaic language of the decrees, such a line length seems a reasonable minimum estimate. Support for such an estimation of the original dimensions of this stele comes from another method of calculating the likely size of the ephebic roster. A rough calculation can be derived from comparing the numbers of ephebes attested by deme here against another fully preserved roster of 62 ephebes of Leontis, the dedication from the Amphiaraion of Oropos.101 A total of 14 ephebes on the present roster compares with 11 from the same demes on the roster from Oropos, which, if the sample were statistically significant, would suggest that this roster was larger than the 62 ephebes on the roster from Oropos.102 Given the small comparative sample from small demes, however, an extrapolation from these proportions should only be considered approximate within a very wide margin of error, especially as a number larger than 62 is unlikely in view of the probable space available on the stone. The calculation should at least indicate that the original roster on this inscription was not very far removed from the number attested in the Oropos inscription, among the largest ephebic cohorts so far attested, and therefore larger than the roster of 38 ephebes of Leontis attested for the enrollment year of Nikokrates, 333/2. This would comport with a date in the 320s, which is also supported by the likely identification of the cutter of this inscription as Tracy’s Cutter of IG II2 1176 (see above), whose work is placed ca. 330–324/3. 99. This estimate is based on assuming that line 17, restored as containing 23 letters, represents the minimum for the longest line, with an average letter space occupying 0.054 m, and adding an equivalent space as a margin, and estimating that a maximum might be as much as 25 letters. 100. This estimate is based on an average letter spacing in lines 1–9 of 0.078 m. 101. I.Oropos 353 (Reinmuth 15; Friend T15); see n. 98, above, on its date. 102. The present inscription lists three Deiradiotai, five Kropidai, two from Oion, and at least four Potamioi (counting all Potamioi demes together; see n. 96, above), for a total of at least 14, while I.Oropos 353 lists one Deiradiotes, two from Oion, eight Potamioi, and none from Kropidai, for a total of 11. These are all relatively small demes, among which the number of Kropidai attested here is remarkable. If Kropidai were removed from the comparison, the proportions would be nine here to 11 from Oropos. insc r ip t ions f r om panakton 319 Preserved wording suggests formulaic language for the honorific decrees, but not enough survives to indicate specifically who is being honored for what actions. Ephebes collectively, or their lochagoi, or an individual officer may have been the recipient(s) of a gold crown in the decree ending with lines 1–3. The gold wreath of a thousand drachmas awarded in the last decree is appropriate for an officer, typically the sophronistes but also possibly the strategos or the kosmetes.103 Part of the name of the proposer of the last decree survives, [. . .]σιάδ[ης. If spacing were not strictly observed, this might be restored as [Λυ]σιάδ[ης, who could be the grandfather of the ephebe Θεοτιμίδης Ἀστυνόμου ἐξ Οἴου (see above). The reference to the ephebes in lines 4–5 could be part of the phrase [χρήσιμον αὑτὸν παρέσχηκεν τοῖς ἐφ]ήβοις το[ῖς] (who has rendered good service to the ephebes who . . .) referring probably to the sophronistes, after which either the archon of the ephebes’ enrollment year ([τοῖς ἐφ]ήβοις το[ῖς ἐπὶ τοῦ δεῖνος]) or the tribe of the ephebes ([τοῖς ἐφ]ήβοις το[ῖς τῆς Λεοντίδος φύλης]) would be specified. The phrase [δεδόχθαι τοῖς ἐφ]ήβοις is possible (several ephebic decrees specify that the ephebes have awarded crowns to their officers), but this expression is not found in any existing ephebic decrees. In line 6, [. . σ]τρατου is best understood as the patronymic of the honoree, probably the sophronistes. Mention of the ἀνάθημα in line 9 confirms that these records were part of a dedicatory monument, but to which deity is unknown. Among the ephebes it is notable that there is a pair of brothers, Εὐθύμαχος Πύρρου and [. . . .]τρ[. . .]ς Πύρρου in lines 24–25, who must be relatives of another pair of brothers from Leontis, Πύρρος Παγκλέους and Φιλόφρων Παγκλέους from the same deme, Πόταμος Δειραδιώτης, in the monument of the ephebes of Leontis dedicated at the Amphiaraion of Oropos.104 That monument contains at least two additional sets of brothers enrolled as ephebes in the same year, occasioning discussion of whether they could be twins. The more likely explanation is that, when two brothers were close in age, the elder waited for the younger to become eligible so that they could enroll together and serve together.105 4 Roster of the ephebes of Leontis Fig. 12 Panakton 1992-400. Fragment of a white (“Pentelic”) marble stele discovered on July 28, 1992, built into the threshold of a door into the north room of House I in 20 m square J-10 (see Fig. 2). Broken at top and on left; right side and bottom rough-picked, back rough-hammered. Inscribed face worn, especially toward top, and friable. H. 0.315; W. 0.205; Th. 0.095 m. L.H. 0.005–0.006 m. Notices in PAA 1995–2012, sub nominibus; Friend T24. 103. A gold wreath of 500 drachmas is awarded (collectively) to the ephebes of Kekropis enrolled under Ktesikles (334/3) and another wreath of 500 drachmas is awarded to their sophronistes (IG II2 1156, lines 30, 32–33, 57–58, 60; Reinmuth 2; Friend T2; Rhodes and Osborne 2003, no. 89); the sophronistes of Leontis received a gold wreath of 1,000 drachmas for his services to the ephebes enrolled under Nikokrates (333/2: Meritt 1940, no. 8, column I, line 17; Reinmuth 9; Friend T9). Wreaths are regularly awarded to ephebes, and by ephebes to their officers, without the values being specified; see IG II3.4 337 (I.Eleusis 86; Reinmuth 5; Friend T6); I.Oropos 353 (Reinmuth 15; Friend T15); IG II3.4 338 (I.Rhmanous 100; Reinmuth 11; Friend T21). 104. Reinmuth 15, column II, lines 18–19 (I.Oropos 353; Friend T15 lines 53–54); on the date, see n. 98, above. 105. On the occurrence of brothers in ephebic texts, see Reinmuth, pp. 64, 72–73; 1948, p. 215; Sekunda 1992, pp. 321, 329–330, 335, 339; Hansen 1994, pp. 303–304. 320 mar k munn 332–323 b.c. Non-stoich. Unknown number of lines missing. [- - - - - - - - - traces of letters - - - -] ca. 10 [- - - - - - - - - -]ος Πολυστρά[του] ca. 4 ca. 3 Ἡ̣[­ ­ ­ ­]α̣ν̣[­ ­ ­]ι̣ος Φανοστ[ράτου] ca. 4 [­ ­ ­ ­]ε̣υ̣[ς] Διοπείθους 5 Ὑ̣β̣άδ̣αι Φι[λ]ονικ[ο]ς Φιλονίκου Παιονίδαι Ἡγέστρατος Ἡγεμάχου Βλέπυρ[ο]ς Πολυμνήστου 10 Εὐθύκριτος Εὐτέλους Αἰ[θα]λ̣ίδαι Εὐ[θ]ύκριτος Σμ̣[ί]κρου Ἀντιφάνης Φα[ιδ]ρίου Π̣ή̣λ̣η̣κ̣ες 15 Λυσικράτης Ἀριστίωνος Εὐπυρί[δ]αι Θεόδωρος Θεοδότου [Π]αυσανίας Λυσίου [Κ]ολωνῆς 20 Θηρικλῆς Προκλέους [ἐ]ξ Οἴου Θεοφάνης Χαιρεφάνους Πασίων Πασίωνος vacat Ep ig rap hic al Not e s Certain letters are consistently slightly taller (Α, Γ, Η, Μ, Ν, Π, Ρ, Σ, Τ, Φ, and Υ), a few are consistently slightly shorter (Δ, Λ, and Ω), and some vary between these (Ε, Ι, and Κ), while omicron and theta are consistently very small. Alpha has a wide stance by contrast with delta and lambda. Delta and lambda both have a nearly vertical left stroke and a more splayed right stroke. Epsilon often has the vertical descending below the bottom cross-stroke, and sometimes also above the top stroke, while the horizontals sometimes all angle down to the right, sometimes splay upward and downward, usually with horizontal strokes of equal length but sometimes with a shorter middle stroke. The diagonal strokes of kappa never splay as far as the full height of the vertical. Mu splays widely. Most distinctive is upsilon, which is uniformly cut in two strokes, a slightly leaning left vertical and a short splayed right stroke. Line 1: Wear on the stone is too great to ascertain any letters, or to be sure if the space of 0.03 m above was inscribed. Line 3: At left, very faint traces of two verticals; after a gap, a widely splaying likely alpha is closely followed by a vertical; after a gap, part of a vertical, after which letters become distinct. Line 4: Faint traces of epsilon and the upper half of an upsilon are discernible. insc r ip t ions f r om panakton 321 Figure 12. Inscription 4 (1992-400). Scale 1:2 Line 5: Faint traces of upsilon, beta, and delta are discernible. Here and in following deme names, the beginning of the name is offset one space to the left of the roster of names of ephebes. Line 11: After a gap, the right stroke of lambda is visible. Line 12: In the patronymic, the left stroke of mu is visible, and after the gap the splaying diagonals of kappa are visible. Line 13: At left, the first two letters are faint but visible. Line 14: At left the first three letters are faintly discernible; a vertical of the second eta shows through damage to the stone, after which the splaying diagonals of kappa are faintly discernible. Line 15: At left, the first three letters are faint but visible. Line 16: At left, the first two letters are faint but visible. 322 mar k munn Pr os op o g rap hy A preliminary reading of this inscription was shared with Traill, and names from that reading have been entered into his Persons of Ancient Athens (PAA). Here I list individuals in alphabetical order with references to PAA, LGPN where applicable, and the current numbering in AO. Ἀντιφάνης Φα[ιδ]ρίου, Ἀἰ[θα]λ̣ίδης (line 13): ephebe, PAA 137350; AO 30. Father cited as PAA 911900; LGPN II, 9; AO 11, identical with Φαιδρίας Αἰθαλίδης, father of another ephebe, Ἐξώπιος, I.Oropos 353, line 51 (Reinmuth 15, column II, line 16; Friend T15), dating between 332 and 324. Bearing a comparatively rare name, another Ἐξώπιος (no demotic), PAA 388505; LGPN II, 1; AO 1, lessee of a mine in 367/6, Agora XIX, no. P5, lines 43–44, 62, might be this ephebe’s grandfather. Βλέπυρ[ο]ς Πολυμνήστου, Παιονίδης (line 9): ephebe, PAA 267035; AO 5. Father cited as PAA 780267; AO 19. Π[ο]λύμνηστο[ς] Παι(ονιδης), LGPN II, 16; AO 20, named on a curse tablet from the Kerameikos, dated to ca. 350, Kerameikos III 6, line 1, is likely to be the ephebe’s father, and Πείθανδρος Παιονίδης, LGPN II, 2; AO 3, named on a related curse tablet, Kerameikos III 3, lines 75–77, is likely to be the ephebe’s uncle. Inasmuch as the ephebe, Βλέπυρος, bears a comparatively rare name, Βλέπυρος Πειθάνδρου Παιονίδης, PAA 267030; LGPN II, 5; AO 4, proposer of a decree, IG II2 189, line 7, and secretary of the boule and demos, Agora XV 36 (IG II3.4 75), line 35, and no. 37, line 2, mid-4th century, is likely to be this ephebe’s grandfather, father of both Πολύμνηστος and Πείθανδρος. For other kinsmen from the deme Paionidai named on these curse tablets, see under Ἡγέστρατος Ἡγεμάχου, Παιονίδης, below. ca. 4 Διοπείθης ([­ ­ ­ ­]ε̣υ̣[ς] Διοπείθους, line 4): father of ephebe (demotic not preserved), PAA 363345; AO 36. Εὐθύκριτος Εὐτέλους, Παιονίδης (line 10), ephebe, not identified in PAA. His father, Εὐτέλης Εὐθυκρίτου, is PAA 445765; LGPN II, 3; AO 4, named in IG II3.4 77, line 13 (IG II2 3133), a list of agonistic dedicants by tribe dated to 339/8. The grandfather, Εὐθύκριτος, is PAA 433455; LGPN II, 20; AO 22. Εὐ[θ]ύκριτος Σμ̣[ί]κρου, Ἀἰ[θα]λ̣ίδης, (line 12): ephebe, PAA 433375; AO 11. Father is PAA 825865; AO 17. Ἡγέστρατος Ἡγεμάχου, Παιονίδης, (line 8): ephebe, PAA 480647; AO 12. Father is PAA 480502; LGPN II, 10; AO 15, probably to be identified as Ἡγέμαχος Πα[ι](ονίδης?), PAA 480500, named with other fellow demesmen in curse tablets from the Kerameikos, dated to ca. 350, Kerameikos III 3, line 6, and no. 6, line 3; see above, under Βλέπυρ[ο]ς Πολυμνήστου, Παιονίδης. Θεόδωρος Θεοδότου, Εὐπυρί[δ]ης, (line 17): ephebe, PAA 506842; AO 123. Father is PAA 9 505416; AO 45, possibly the same as Θεόδοτος, Εὐπυρίδης, PAA 505415; LGPN II, 41; AO 44, bouleutes in 336/5, Agora XV 42, line 253. Θεοφάνης Χαιρεφάνους, [ἐ]ξ Οἴου, (line 22): ephebe, PAA 510673; AO 21. Father cited as PAA 975525; AO 17 (both mistakenly assigned to Hippothontis instead of Leontis in PAA). insc r ip t ions f r om panakton 323 Θηρικλῆς Προκλέους, [Κ]ολωνεύς, (line 20): ephebe, PAA 513992; AO 4. Father is PAA 789085, possibly same as Προκλῆς Κολωνῆθ[εν], PAA 789083; LGPN II, 36; AO 39, epimeletes of docks in the archonship of Lysistratos, 369/8, IG II2 1617, line 75. Λυσικράτης Ἀριστίωνος, Πήληξ, (line 15): ephebe, PAA 615442; AO 39. Father cited as PAA 166730; AO 52. Πασίων Πασίωνος, [ἐ]ξ Οἴου, (line 23): ephebe, PAA 768185; AO 12. Father cited as PAA 768184; AO 11. [Π]αυσανίας Λυσίου, Εὐπυρί[δ]ης, (line 18): ephebe, PAA 769658; AO 22. Father cited as PAA 613972; AO 51. ca. 10 Πολύστρατος ([- - - - - - - - - -]oς Πολυστρά[του], line 2): father of ephebe (demotic not preserved), PAA 781050; AO 47. ca. 4 ca. 3 Φανόστρατος (Ἡ̣ [- ­ ­ ­]α̣ν̣[­ ­ ­]ι̣oς Φανοστ[ράτου], line 3): father of ephebe (demotic not preserved), PAA 917020; AO 43. Φι[λ]όνικ[ο]ς Φιλονίκου, Ὑβάδης (line 6): ephebe, PAA 940480; AO 8. Father is PAA 940479; AO 7. Com men tary 106. See Ἀντιφάνης Φα[ιδ]ρίου, Ἀἰ[θα]λ̣ίδης (4, line 13), probably the brother of Ἐξώπιος Φαιδρίου, Ἀἰθαλίδης (I.Oropos 353, lines 50–51); the two sons of Πύρρος (Εὐθυμάχου), Ποτάμιος (3, lines 24–25), are probably cousins of Πύρρος [Π]αγκλέους, Ποτάμιος and his brother, [Φ]ιλόφρων Παγκλέους, Ποτάμιος (I.Oropos 353, lines 52–54) and likewise probably cousins of another Ποτάμιος who is likely the son of [Φιλόφρ]ων (3, line 27); see prosopographical notes. As with 3 above, this ephebic roster of Leontis might belong to enrollment classes from the year of Ktesikles (334/3) to that of Antikles (325/4), excluding only the year of Nikokrates (333/2) and the year of I.Oropos 353 (Reinmuth 15; Friend T15), which is uncertain. As an end-of-service dedication, this inscription should date either to 332 or between 330 and 323. Prosopography does not allow any more precision, but the nexus of brothers and cousins detected among the three Leontid rosters—the two from Panakton and the one from the Amphiaraion—confirms the general dating.106 Inscribed by a different hand on a thicker stone, and with different rosters for ephebes ἐξ Οἴου, this is certainly the record of a different cohort from Leontis than that listed in 3 above. About the size and scope of the original stele we have a few indicators. The thickness of the stone and the size and spacing of the letters are both more ample than that of 3, above. The present roster is the last of at least two columns that contained at least 16 ephebic names, and a third column is not unlikely. The inscribed column occupies a width of approximately 0.17 m, so estimating two columns with a margin would give a stele about 0.36 m wide; if there were three columns the stele would be approximately 0.53 m wide. As for the number of ephebes whose names might be recorded, the minimal estimate would consist of two columns of approximately 16 names each, or 32 in all. The two columns could be longer than 16 names, however, so the upper limit is a matter of conjecture. If there were three columns, then a lower limit would be 48 names, with the upper limit again a matter of conjecture. To compare the number of ephebes in each of the demes preserved here to those in the full roster of 62 ephebes of Leontis from the Amphiaraion at Oropos (I.Oropos 353; Reinmuth 15; Friend T15; see p. 318, with n. 101, above): in this inscription from Panakton we have one ephebe from Hybadai (compared to five), three from Paionidai (compared to none), two from Aithalidai (compared to one), one from 324 mar k munn Pelekes (compared to two), two from Eupyridai (compared to one), one from Koloneis (compared to two), and two from Oion (compared to two), for an overall comparison of 12 ephebes at Panakton and 13 from the Amphiaraion. Proportional representation from among these small demes cannot be considered an exact indicator of the overall tribal cohorts, but as a rough indicator these numbers suggest a total tribal cohort in the vicinity of 57 ephebes of Leontis at Panakton, a plausible figure which should be taken as the probable mean of a range that could conceivably be as few as 32 or possibly as many as 66. 5 Dedication of hypaithroi at Panakton Figs. 13, 14 Panakton 1992-250. Fragment of a rectangular base of white (“Pentelic”) marble found on July 25, 1992, in the debris of collapsed wall and roofing material filling the narthex (western room) of the church in 20 m square J-10 (see Fig. 2). The base is broken on the left side, hammer-dressed on the right, broken and partially hammer-dressed on the back, roughly picked on bottom. The top surface is smooth, with a nearly circular cutting partially preserved on the left edge, above the partially preserved column of names on the face. Cutting is 0.025 m deep and 0.130 m in diameter, with a circular tenon hole 0.009 m in diameter and 0.016 m deep centered in it. The inscribed face is weathered and friable. H. 0.223; W. 0.320; Th. 0.138 m. L.H. 0.005–0.008 m. Previously published in Munn 1996, pp. 53–55; SEG XLVI 249; IG II3.4 324; AIO 1783. Ca. 255–200 b.c. 1 5 10 15 20 Non-stoich. [- - - - - - - - - - - - - - heading broken away - - - - - - - - - - - -] [- - - - - - - - - - - - - - - ὕπα]ι̣θ̣ροι οἵ{ι}δε ἀ[ν]έθ̣εσαν [- - - - - - - - - - - - -]ριος [- - - - - - - - - - Ἀγ]γελῆθεν [- - - - - - - - - - - Ῥ]α̣μ̣νούσι⟨ο⟩ς [- - - - - - - - - - - Ἕ]ρμειος [- - - - - - - - - - -Ἀλ]ωπεκῆθεν [- - - - - - - - - - - - ἐ]ξ Οἴου [- - - - - - - - - - - -] Κυδαντίδη[ς] [- - - - - - - - - - - Ἀ]φιδναῖος [- - - - - - - - - - - -] Χολαργεύς̣ [- - - - - - - - - - - -]νίδης [- - - - - - - - - Φυλ]⟨ά⟩σιος [- - - - - - - - - - - -]ς ⟨Θ⟩ηβαῖος [- - - - - - - - - - - -] Ἀφιδναῖος [- - - - - - - - - Ἁμα]ξαντε̣ύς [- - - - - - - - - - - -]ναῖ[ο]ς [- - - - - - - - - - - -]νεύς [- - - - - - - - - - -Κυ]δαντίδης [- - - - - - - - - - - -]νεύς [- - - - - - - - - -Πα]ιανιεύς [- - - - - - - - - - -Ἀ]μφιτροπῆθεν [- - - - - - - - - - - Σο]υν[ι]εύς [- - - - - - - - - - - -]ιος insc r ip t ions f r om panakton 325 Figure 13 Inscription 5 (1992-250). Scale 1:3 Figure 14. Inscription 5, oblique view showing cutting on upper surface. Scale 1:4 E p ig rap h ic al N ot e s Alpha, delta, and lambda are all similarly equilateral. Gamma, epsilon, eta, pi, and tau all occupy a uniformly rectangular space; the center cross-strokes of epsilon are usually equal to top and bottom strokes, occasionally shorter. Nu is usually also rectangular, but sometimes the final vertical leans slightly to the right. Omicrons and thetas are cut more deeply than other letters; they vary in size, are sometimes angular (diamond shaped), sometimes oval, usually round. The diagonal arms of upsilon sometimes splay widely. Letter sizes are the same in the heading, line 2, and in the roster of names below, but the spacing of letters in the heading is wider, an average of 0.015 m per letter in line 2 compared to an average of 0.01 m per letter in the roster. 326 mar k munn Line 1: The top of this stone very likely contained a line of inscription, of which no definite trace survives. I originally considered this line to be a deliberate rasura, but Tracy has pointed out to me (by personal communication) that this does not look like a regular excision of inscribed letters. I agree that this does not have the appearance of a regular rasura, but the breakage is too regular to have been purely accidental damage to the edge of the stone. The breakage occupies a nearly uniform height of 0.025 m for the full 0.27 m length of the top edge of the face of the stone, broken to an almost uniform depth, and with what appear to be slightly deeper vertical strokes at intervals of between 0.012 and 0.007 m. I consider this most likely to be deliberate breakage by hammer or chisel strokes designed to efface the top line, which likely contained the name of an honorand in whose name the original monument on this base was dedicated.107 Line 2: At left, the first distinct letter is a circular letter damaged in its upper right half, but certainly either omicron or theta; context dictates the latter reading.108 To left of this, at the edge of the break, there is a distinct horizontal stroke at what could be the bottom of an epsilon; I regard this as an accidental mark, likely the product of the later breaking of the stone. Immediately to the right of this stroke the faint trace of a vertical iota; iota elsewhere is lightly incised, cf. the iotas in the fifth and seventh visible letter spaces in line 2, and immediately below in line 3. Between the iota and the delta of οἵδε there is a stray vertical stroke; it is possible that the stone cutter originally cut a nu instead of iota. The left diagonal and apex of alpha is visible before a break, after which two epsilons, alpha, and nu are clear, while faint traces of theta and sigma can be discerned. Line 3: At right, the sigma is almost effaced, but the right ends of the diagonal strokes are discernible. Line 4: At right, it is possible that the tip of the first gamma shows at the break; the second gamma and all following letters are clear. Line 5: At right, trace of the lower half of the right diagonal of alpha and the first stroke of mu are discernible. Portions of the sigma in this line, and in the line below, are effaced by a break. The cutter has incised a theta instead of omicron in the penultimate letter space. Line 9: The letters of this line appear to be slightly larger than in other lines, possibly indicating a correction.109 Line 11: At right, most but not all traces of the sigma are effaced by a blow to the stone, similar to the blow that has effaced the sigma at the end of line 9. Line 12: At right, the top stroke of sigma is visible. Line 13: At left, the cutter has not incised the cross-bar of the alpha. Line 14: The cutter has not incised the central dot of the theta. Line 16: At right, the vertical of the epsilon is visible next to a break. Line 24: The top half of these letters are visible. Comme n tary As revealed by the cutting on its top surface, this stone is almost exactly half of a base that originally supported a marble statue or columnar monument with circular base or plinth 0.13 m in diameter, and with a central tenon. 107. An alternative possibility, that the top edge of this stone was reworked to fit it into a secondary use, seems unlikely. There is little likelihood that a base of this sort had a projecting molding that required removal; it was most likely a simple squared block. Furthermore, reuse would most likely occur not before the destruction of the Hellenistic fortress in 200 b.c., but in the late medieval reoccupation, when this and many other earlier blocks were reused. The late medieval building technique did not require nicely squared and fitted blocks, but employed stones of many random sizes in mortared construction, so careful trimming here is improbable. 108. In my 1996 publication of this inscription I entertained the possibility that this rounded letter could be beta, but where a vertical stroke should exist the character is clearly round, so that possibility should be excluded. 109. I owe this observation to the careful inspection of a squeeze by Stephen Tracy. insc r ip t ions f r om panakton 327 Doubling the preserved width of 0.32 m allows us to estimate the number of letters missing in the effaced first line, which presumably occupied the full original width of the stone, and the number of letters missing from the beginning of line 2. In the preserved portion of line 2 there are 17 letters occupying approximately 60% of the width of 0.32 m. Assuming that the missing left half of the line began close to the left edge of the original face, there would be room for approximately 28 more letters to be restored at the beginning of line 2. Spacing of letters in line 2 is uneven. There is a difference of approximately 12% between the wider spacing of the four letters POIO toward the left and the closer spacing of EΣΑΝ at the right end of line 2. Such a difference over the space for 28 letters yields a margin of error of ±3 letters. Adding a text of 25 letters to this line, a plausible restoration could therefore be [οἱ ἐμ Πανάκτωι τεταγμένοι ὕπα]ι̣θ̣ροι οἵδε ἀ[ν]έθ̣εσαν (These [hypa]ithroi [stationed in Panakton] dedicated [this monument]).110 Following the examples of other garrison texts, including 2 here, the men making the dedication were likely doing so to honor a commanding officer, who therefore would have been named in the first line, where his office (most likely τὸν στρατηγόν or στρατηγήσαντα), his name, patronymic, and demotic, and possibly the participle στεφανώσαντες would occupy between 50 and 62 characters. The deity or hero to whom this dedication was made may have been named as well, but need not have been. Hypaithroi, soldiers assigned the duty of patrolling the countryside “under the open sky,” appear as regular units in the garrisons of Attica late in the 3rd century.111 They served alongside οἱ στρατευόμενοι τῶν πολιτῶν (citizens performing military service) and with them they are recorded as voting honors for their commanding officers.112 This inscription demonstrates that the hypaithroi could comprise a large proportion of Athenian citizens, but the presence of the Theban (line 14) indicates that citizenship was not 110. Approximate parallels for this wording are found in inscriptions of the second half of the 3rd century. In IG II3.4 281 (IG II2 2971, I.Eleusis 195), a garrison described as Ἀθηναίων οἱ τεταγμένοι ἐμ Πανάκτωι is among those groups who dedicate a statue of their commanding general at Eleusis. In IG II2 1299 (I.Eleusis 196), a similar group of Athenians as well as οἱ στρατιῶται οἱ παρὰ τεῖ πόλει στρατευόμενοι (those performing military service for the city), who are non-Athenian soldiers, likewise dedicate a statue in honor of their commanding general at Eleusis. 111. For the formation of the corps of hypaithroi as a function of conditions in Attica during the 3rd century, see Kent 1941; Garlan 1978; Oliver 2007, pp. 180–181; Couvenhes 2011, pp. 301–303; Oetjen 2014, pp. 19, 25–26. The date of the creation of this military body is a matter of discussion; see below. 112. So SEG XXXI 20 from Rhamnous (I.Rhamnous 49), of 207/6, records the resolution of τοῖς στρατευομένοις τῶν πολιτῶ[ν καὶ τ]οῖς ὑπαίθροις (lines 1–2), granting honors to Theotimos son of Theodoros of Rhamnous for exemplary service during six years as an elected general; IG II2 1306 from Eleusis (I.Eleusis 211), of 209, records resolutions δεδόχθαι τοῖς τετ[α]γμένοις τῶν πολιτῶν ἐν Ἐλευσῖνι καὶ Πανάκ­ τωι κ[αὶ Φυ]λεῖ καὶ τοῖς ὑπαίθροις (resolved by the citizens stationed at Eleusis and Panakton and Phyle, and by the hypaithroi: lines 41–42; I.Eleusis 211, lines 42–43) giving honors to their general, Demainetos son of Hermokles of Athmonon, that included the erection of a bronze statue of him at Eleusis. The fact that noncitizen mercenaries could join in corporate resolutions together with citizen soldiers is explicit in IG II2 1299 from Eleusis (I.Eleusis 196), of ca. 234, where honors for the Athenian general Aristophanes son of Aristomenes of Leukonoion are voted by τοῖς τεταγμένοις τῶν πολιτῶν Ἐλευσῖνι καὶ ἐμ Π[ανάκτω]ι καὶ ἐπὶ Φυλεῖ καὶ τοῖς στρατιώταις τοῖς παρὰ τεῖ πόλει στρατευομ[ένοις καὶ] τεταγμένοις Ἐλευσῖνι (the citizens stationed at Eleusis and in Panakton and at Phyle and by the soldiers performing military service for the city and stationed at Eleusis: lines 20–22) and subscribed to also by το[ῖς ξένοις τοῖ]ς μετὰ Γνωσίου τεταγμένοις (the foreign troops stationed with Gnosias, lines 41–42, cf. 93–117). See also IG II2 1286 (I.Rhamnous 11) from Rhamnous, probably of the 240s. See Launey 1950, pp. 1037–1054; Oliver 2007, pp. 187–188; and Oetjen 2014, pp. 127–171, for discussions of the manner in which these mixed military units adopted certain of the corporate practices characteristic of democratic civic life. 328 mar k munn a prerequisite for service in this unit.113 This and the fact that the hypaithroi are distinguished from the “citizens performing military service” indicates that the hypaithroi must be considered a mercenary force, recruited perhaps because their duties exposed them to more dangers than those faced by citizen soldiers performing garrison duty. Hypaithroi performed the same duties of patrolling the open countryside that had been the assignment of peripoloi and ephebes a century earlier.114 Their earliest firmly dated attestations, in the inscriptions from Rhamnous and Eleusis, belong to the 220s and later.115 The chronology of datable texts has allowed the suggestion, first put forth by Garlan, that the hypaithroi were a unit created by the Athenians only after the final evacuation of Macedonian forces in 228.116 It remains possible, however, that some attestations of uncertain date might belong before 228, and that the hypaithroi were already in existence in the 230s or even earlier, during the period when Macedonian forces still occupied garrison posts in Piraeus, Mounichia, Salamis, and Sounion.117 The Macedonian occupation of Attica as a whole, established in 261 after the surrender of Athens at the end of the Chremonidean War, was phased out in stages, and after Antigonos Gonatas withdrew soldiers under Macedonian command from the Mouseion fort in Athens in 255 it is highly likely that Macedonian officers no longer commanded the garrisons at Eleusis, Panakton, Phyle, and Rhamnous.118 With the defenses of the land frontiers of Attica in the hands of officers and troops under Athenian command, there is no reason why the hypaithroi, composed primarily of 113. An inscription of ca. 228 from Rhamnous (SEG XXVIII 107; I.Rhamnous 26) is a resolution passed by the Athenians of the hypaithroi (lines 19–20: δεδ[όχθαι] Ἀθηναίων [τ]οῖς στρατευομένοις ἐν Ῥαμνοῦντο [τῶν ὑπαί]θρων), indicating that the citizens of this unit could take corporate action apart from noncitizens who presumably were part of their unit. Garlan (1978, pp. 106–107) took the formulation in this inscription to mean that the hypaithroi were exclusively Athenian citizens, which led him to suggest that they were a creation of the newly liberated Athenian state after 228. The present inscription attests to the presence of foreigners among the hypaithroi and allows the possibility (which was never precluded) that this unit was created before the evacuation of all Macedonian troops from Attica. 114. On the peripoloi¸ see the historical commentary to 1 with n. 25, above. On the duties of ephebes to patrol the countryside in their second year, see Arist. Ath. Pol. 42.4: περιπολοῦσι τὴν χώραν καὶ διατίβουσιν ἐν τοῖς φυλακτηρίοις. Cf. Aeschin. 2.167: ἐκ παίδων μὲν γὰρ ἀπαλλαγεὶς περίπολος τῆς χώρας ταύτης ἐγενόμην δὺ’ ἔτη (when I came of age I became a peripolos of this country for two years). 115. The earliest firmly dated decree and dedication by the hypaithroi is from Eleusis, IG II2 2978 (I.Eleusis 200), dated to 224/3. Oetjen (2014, pp. 203–205) lists four decrees mentioning the hypaithroi from Rhamnous (I.Rhamnous 26, 46, 49, 55), all likely or certainly dating between the 220s and the end of the 3rd century. 116. Garlan 1978; followed by I.Eleusis, vol. 2, p. 267; Couvenhes 2011, pp. 301–303; Oetjen 2014, pp. 19, 25–26. Couvenhes sees an evolution of military units deployed in the countryside, from the peripoloi of the 5th and 4th centuries (see nn. 25, 27, above) to the kryptoi who appear in inscriptions from Rhamnous in the 260s, eventually replaced by the hypaithroi as they are attested after 228. It is entirely possible, however, that units such as the kryptoi and the hypaithroi could have fulfilled different sorts of duties (the one being primarily stationary lookouts, the other mobile patrols, for example), and could therefore have operated in the same years. 117. Kent (1941, p. 349, n. 12) has suggested that the hypaithroi in Attica were formed under the Macedonians; Oliver (2007, p. 18) allows that the hypaithroi may be attested in the 230s; cf. the uncertain dating of SEG XXII 129 (I.Rhamnous 55), placed by Petrakos in the second half of the 3rd century. 118. Habicht (2003, p. 54) acknowledging the work of Oetjen, has recognized that after 255 the fortresses of Eleusis, Panakton, Phyle, Rhamnous, and Aphidna were “under Athenian jurisdiction. From that date on, Athenian generals, elected by the assembly without any interference from the king (as had happened before 255), were in charge in all these places.” See the review of the evidence by Oetjen (2014, pp. 48–70). insc r ip t ions f r om panakton 329 Athenian citizens but including foreigners as well, could not have been created soon after the partial Macedonian withdrawal of 255. In fact, there were good reasons for organizing such a mobile defense force for the protection of Attica at that time. The late 250s, 240s, and 230s saw raids into Attica by forces opposing Antigonos and his Athenian subject-allies.119 Epigraphic records of money raised for the defense of the countryside, and honors accorded to officers who saw to the safety of farmers in the countryside, indicate conditions that may have prompted the formation and deployment of the hypaithroi.120 Unlike the inscriptions that refer to the hypaithroi serving along with citizen soldiers, or referring specifically to the citizens serving among the hypaithroi (most of them datable after 228; see nn. 112, 113, above), the present inscription is a dedication of the hypaithroi alone, as a mixed unit, with no explicit reference to Athenian citizenship (outside of the demotics of the men making the dedication). This may suggest that this dedication was a monument of the period of hybrid Athenian and Macedonian authority over the defenses of the countryside. Between 255 and 228 Athens and Athenian forces in Attica remained aligned with the Antigonid kings of Macedon. In this period the Athenians publicized honors that they awarded to Athenian commanders, some of whom were serving on the authority of the Athenian state and some of whom were Macedonian appointments.121 Macedonian supremacy, in either case, was a condition of their service. It is quite possible that some of these men were held in genuine public esteem among Athenians regardless of their subservience to Macedonian authority, while others may have been perceived as too obsequious, or more concerned to please their Macedonian overlords than to serve true Athenian interests—at least in retrospect, after the final and full withdrawal of Macedonian forces from Attica in 228. If such an individual were named as an honorand on this dedicatory monument from the period 119. Habicht (1997, pp. 162–166) summarizes the effects on Attica of the wars first between Antigonos Gonatas and Alexander of Corinth, later against Aratos of Sikyon and the Achaian League, and later still, after the death of Antigonos and the succession of Demetrios II in 239, the continuing war with the Achaian and Aitolian leagues. Kralli (2003) suggests that divine honors voted by the Rhamnousians to Antigonos as σωτὴρ τοῦ δήμου (I.Rhamnous 7) were given in thanks for financial support that the king provided for the defense of the countryside in the late 250s. Honors given to the Athenian general Aristophanes son of Aristomenes in IG II2 1299 (I.Eleusis 196), ca. 234, attest to the conditions of war that threatened Attica during his tenure of military office in the preceding years; see the discussion of Oliver (2007, pp. 150–152). 120. See Oliver 2007, pp. 131–133, referring to an unpublished inscription from Rhamnous, and see Agora XVI 213 (IG II3.1 1011) with Oliver 2007, pp. 200–204, on the special contributions raised for the stratiotika in 248/7, εἰς τὴν σωτηρίαν τῆς πόλεως καὶ τὴν φυλακὴν τῆς χώρας (for the safety of the city and the defense of the countryside: lines 14–15, 30–31). 121. Note the honors voted by the Salaminians for Hierakleitos son of Asklepiades of Athmonon, the Athenian general appointed by Antigonos Gonatas to command in Piraeus, who is congratulated because he saw to it that μηθὲν ἀδίκημα γίνηται κατὰ τὴν χώραν (no wrongdoing should take place in the countryside: IG II2 1225, lines 10–11, 14–15, ca. 245), indicating that the protection of the countryside was within the purview of Macedonian authority. Note also how Aristophanes son of Aristomenes, Athenian commander at Eleusis, is honored for the good he has done for the Athenian people and for King Demetrios and his family, in IG II2 1299 (I.Eleusis 196), ca. 234. Tracy (2015) describes an unpublished decree, Agora I 7572, honoring Eurykleides of Kephisia for his services as hoplite general of the Athenians both before and after the Macedonian withdrawal in 228, where three lines of the decree were later carefully erased (presumably in 200; see Byrne 2010) to remove any reference to dutiful service under Macedonian overlords. 330 mar k munn of Macedonian presence, this could explain why the first line appears to have been deliberately effaced.122 In any event, the inscription cannot be any later than the end of the 3rd century, as the war with Philip V that began in 200 saw both the destruction of Panakton, as indicated by the archaeological evidence, and the disappearance of the hypaithroi as a regular military unit, as indicated by the epigraphic evidence. 6 Pediment of a stele Figs. 15, 16 Panakton 1992-4. Fragment of the pediment of a white (“Pentelic”) marble stele found on July 8, 1992, in debris associated with House III immediately east of the church in 20 m square J-11 (see Fig. 2). Horizontal cornice with ovolo below, single letter on the face of the cornice; raking cornice with small ovolo below taenia, larger ovolo above. Broken at back, left, and right, and below horizontal cornice. H. 0.083; W. 0.063 m. L.H. 0.007 m. Figure 15. Inscription 6 (1992-4). Scale 1:2 4th–3rd century b.c. [ΘΕ]Ο[Ι] Comme ntary This fragment preserves an invocation found at the top of stelai of all sorts from the mid-5th century on.123 This form of pedimental stele became common by the mid-4th century.124 With two moldings and a taenia, this pedimental stele has a slightly more elaborate raking cornice than most examples. The break on the right is close to the junction of the horizontal and raking cornices, and probably corresponds roughly to the point at which a palmette akroterion was originally attached. The location of the omicron in relation to the projected junction of the two cornices determines the approximate spacing of the letters. It seems likely that the break on the left 122. Habicht (1997, p. 179) notes evidence of an effort to efface the record of dependence on Macedon soon after the liberation of Athens in 228. While not as systematic as the careful excision of the language of obeisance to the kings of Macedon that took place upon the outbreak of war with Philip V in 200 (see Byrne 2010), such an effort may have occasioned the deletion of honors to certain individuals whose service under Macedonian domination was no longer deemed appropriate for public commemoration. In Munn 1996 I suggested that the name deleted from the first line of this monument could have been that of Demetrios of Phaleron, put in power in 317 by Kassander, whose memory was later condemned by the Athenians according to later sources. I now recognize that the present inscription is much later than the lifetime of that Demetrios, and that the principal evidence I cited in making this suggestion, IG II2 2971 (I.Eleusis 195; now IG II3.4 281), has been shown by Tracy (1994; 2000; 2003, p. 132) to be a monument to his grandson of the same name, who rose to prominence in the mid-3rd century. The younger Demetrios did in fact have a military career spanning the later 250s and early 240s that included command over, and honors awarded by, Athenians garrisoned at Eleusis, Panakton, and Phyle (IG II3.4 281; I.Eleusis 195; cf. I.Eleusis 194). Although the role of deputy and overseer of Athens on behalf of Antigonos Gonatas attributed to the younger Demetrios by Habicht (1983, pp. 18–20; 1997, pp. 153–154) has been shown to be implausible by Oetjen (2000), it is not impossible that he, or someone with a similar career, might have been more eager to win favor with the Macedonian court than to serve for the good of the Athenian people, as seen by the Athenians serving in the defenses of Attica after 228. 123. Pounder (1975) examines the origin and significance of this and related invocations in public inscriptions. 124. For examples, see I.Eleusis 70, 100, 142 (pls. 30, 45, 58), and Reinmuth 19 (pl. XXX), from the mid- to late 4th century; and I.Rhamnous 3, I.Eleusis 197 (pl. 98) from the midto late 3rd century. insc r ip t ions f r om panakton Figure 16. Inscription 6, as reconstructed within the stele pediment. Scale 1:3 331 was close to where a peak akroterion would have sprung from the raking cornice, such that the break was a little to the right of the midpoint of the pediment. If this is correct, the stele was not very wide, at approximately 0.30 m. A 4th- or 3rd-century date is inferred from growing popularity of this style in the 4th century, and the fact that the site was destroyed and abandoned at the beginning of the war with Philip V of Macedon in 200. CO N C LU S I O N S 125. Inscriptions mentioning Panakton from Eleusis include IG II2 1285, 1299, 1303–1307, 1672, 2971 = IG II3.4 281 (I.Eleusis 194, 196, 207, 211, 197, 198, 205, 177, and 195 respectively); from Rhamnous, SEG XXV 155 (I.Rhamnous 17). The more numerous garrison inscriptions from Rhamnous, which unlike Panakton was a deme as well as a fortress, have been the particular subject of study by Oetjen (2014). The epigraphical silence from the northwest frontier of Attica observed by Vanderpool has been broken. The variety of inscribed records that have been recovered from Panakton thus far provide new data for our understandings of the organization of military matters and related social and political institutions at Athens as it transitioned from an influential and independent state of the later Classical period into its role as a subordinate in the period of Macedonian ascendency. Out of the approximately two and a half centuries of active occupation at Classical and Hellenistic Panakton, the inscriptions found so far, together with those found at Eleusis and Rhamnous that refer to Panakton, present a picture of garrison life that spans the last century and a half of the life of the fortress.125 The chronological distribution of these inscriptions is roughly similar to that of the garrison texts from Eleusis, Rhamnous, and Sounion. If anything, the sampling from Panakton is somewhat richer at the earlier end of the collective corpus of garrison inscriptions, as most of them belong to the 4th century, chiefly to the Lykourgan era. We can reasonably expect that further excavation on this site will reveal still more inscriptions, most likely of the Lykourgan and Hellenistic eras, and perhaps from earlier than the mid-4th century. Unlike Eleusis and Rhamnous, Panakton was never a deme with an officially recognized corporate civilian population. Nevertheless, the garrison community, which likely fluctuated in size over the life span of the fortress, was more or less continuously involved in the sorts of formal activities that generated inscribed records concerning military or ritual matters. In the Hellenistic era, and in periods of significant military activity from the Peloponnesian War down to the confrontations with Macedon, the 332 mar k munn Figure 17. Stele bases recovered from the rubble of the church at Panakton garrison at Panakton is likely to have been as large as the garrison posted at Eleusis. This is, at least, the conclusion that can be drawn from IG II2 1299, an inscription of the 230s that authorizes the appointment of representatives from the garrisons at Eleusis, Panakton, and Phyle to collect money for the commissioning of the honorific statue of their general. The inscription specifies that five men are to be elected from Eleusis, five from Panakton, and one from Phyle.126 The six inscriptions presented here (five of them substantive texts) represent the findings of surface survey (3), one season of small soundings (2), and one season of more extensive clearing of surface remains on the site of Panakton (1, 4–6). In the course of this work, seven limestone stele bases and three limestone beddings for monument bases of various sizes have been found, most of them either built into the late medieval church or recovered from its rubble (a selection is shown in Fig. 17). The majority of texts found so far have the character of dedicatory and honorific monuments (3–5 and probably 6), of the sort that would have been erected at one or more shrines on the site—one of which may well have been in the central location now occupied by the remains of the church.127 The hermaic 126. IG II2 1299 from Eleusis (I.Eleusis 196), ca. 234, lines 37–39. 127. Mention of the celebration of the Apatouria at Panakton as an appropriate occasion for the publication there of honors indicates that the central cult place at Panakton may have been a shrine of Zeus Phratrios and Athena Phratria. On the Apatouria at Panakton, see IG II2 1285 (I.Eleusis 194), line 22 (restored), and IG II2 1299 (I.Eleusis 196), lines 29–30. The association of Panakton with the foundation legend of the Apatouria can be traced as far back as the Ἀττικὴ συγγραφή of Hellanikos, at the end of the 5th century (FGrH 323a F23), discussed in Munn 1989, pp. 236–238; Lambert 1998, pp. 148–152. The Apatouria has strong associations with rituals of coming of age, and so with ephebes; on this, and the several other deities associated with this festival (Dionysos, Hephaistos, Apollo, Artemis), see Deubner 1966, pp. 232–234; Parke 1977, pp. 88–92; Lambert 1998, pp. 152–161. insc r ip t ions f r om panakton 333 Figure 18. Niches cut in a limestone face immediately below the circuit of Panakton base (2) was also a dedicatory monument, which originally must have stood not far from where it was found, by the gate. The examples of the stele and monument bases found at the fortress of Rhamnous suggest that other monuments (such as the arsenal inventory, 1) may have been located along other avenues within the fortress. The evidence of two rock-cut niches in a vertical limestone face in 20 m square M-11 (see Fig. 2) below the fortress at Panakton indicates that other votive monuments may lie outside the walls of the fortress (Fig. 18). Given Panakton’s non-deme role as a garrison fort, the texts published here are a valuable complement to the inscriptions that come from the garrison demes and from Athens itself. They reveal measures taken by the Athenians to assert their claims and to plant their institutions in a contested borderland. As such, they bring us a clearer picture of the history of the frontiers of Classical and Early Hellenistic Attica, a picture that will likely be enlarged and enriched by further exploration of this important site. 334 mar k munn R EF ER EN C E S Agora = The Athenian Agora: Results of Excavations Conducted by the American School of Classical Studies at Athens, Princeton X = M. Lang and M. Crosby, Weights, Measures and Tokens, 1964. XI = E. B. Harrison, Archaic and Archaistic Sculpture, 1965. XV = B. D. Meritt and J. S. Traill, Inscriptions: The Athenian Councillors, 1974. XVI = A. G. Woodhead, Inscriptions: The Decrees, 1997. XIX = G. V. Lalonde, M. K. Langdon, and M. B. Walbank, Inscriptions: Horoi, Poletai Records, Leases of Public Lands, 1991. AIO = S. Lambert, ed., Attic Inscriptions Online, www.atticinscriptions.com. Anokhin, V. O., S. N. Bibikov, A. I. Terenožkin, and V. D. Baran. 1971. Skifo-Sarmats’ka ta antična archeologija (Archeologija Ukraïns’koï RSR 2), Kiev. AO = S. G. Byrne, ed., Athenian Onomasticon, www.seangb.org. APF = J. K. Davies, Athenian Propertied Families 600–300 B.C., Oxford 1971. Baitinger, H. 2001. Die Angriffswaffen aus Olympia (OlForsch 29), Berlin. Bearzot, C. 1985. Focione tra storia e trasfigurazione ideale (Scienze storiche 37), Milan. Burckhardt, L. A. 1996. Bürger und Soldaten: Aspekte der politischen und militärischen Rolle athenischer Bürger im Kriegswesen des 4. Jahrhunderts v. Chr. (Historia Suppl. 101), Stuttgart. Byrne, S. G. 2010. “The Athenian Damnatio memoriae of the Antigonids in 200 b.c.,” in ΦΙΛΑΘΗΝΑΙΟΣ/Philathenaios: Studies in Honour of Michael J. Osborne, ed. A. Tamis, C. J. Mackie, and S. G. Byrne, Athens, pp. 157–177. Chankowski, A. S. 2010. L’éphébie hellénistique: Étude d’une institution civique dans les cités grecques des îles de la Mer Égée et de l’Asie Mineure, Paris. ———. 2014. “L’éphébie athénienne antérieure à la réforme d’Epikratès: À propos de Reinmuth, Eph.Inscr. 1 et de la chronologie des premières inscriptions éphébiques,” BCH 138, pp. 15–78. Clinton, K. 1988. “The Ephebes of Kekropis of 333/2 at Eleusis,” ArchEph 127 [1991], pp. 19–30. Cole, P. J. 1981. “The Catapult Bolts of IG 22 1422,” Phoenix 35, pp. 216– 219. Cooper, F. A. 2000. “The Fortifications of Epaminondas and the Rise of the Monumental Greek City,” in City Walls: The Urban Enceinte in Global Perspective (Studies in Comparative Early Modern History), ed. J. D. Tracy, Cambridge, pp. 151–191. Couvenhes, J.-C. 1999. “La réponse athénienne à la violence territorial aux IVe et IIIe siècles av. J.-C.,” Cahiers du Centre Gustave Glotz 10, pp. 189–207. ———. 2007. “La fourniture d’armes aux citoyens athéniens du IV e au III e siècle avant J.-C.,” in Les armes dans l’Antiquité: De la technique à l’imaginaire. Actes du colloque international du SEMA, Montpellier, 20 et 22 mars 2003, ed. P. Sauzeau and T. Van Compernolle, Montpellier, pp. 521–539. ———. 2011. “Les patrouilleurs de l’Attique: Péripoloi, kryptoi, et hypaithroi de la cité athénienne,” in Pratiques et identités culturelles des armées hellénistiques du monde méditerranéen: Hellenistic Warfare 3 (Scripta antiqua 38), ed. J.-C. Couvenhes, S. Crouzet, and S. Péré-Noguès, Bordeaux, pp. 295–306. Delorme, J. 1960. Gymnasion: Étude sur les monuments consacrés a l’éducation en Grèce (des origines à l’Empire romain) (BÉFAR 196), Paris. Deubner, L. 1966. Attische Feste, 2nd ed., Darmstadt. Dontas, G. S. 1983. “The True Aglaurion,” Hesperia 52, pp. 48–63. Dow, S. 1969. Conventions in Editing: A Suggested Reformulation of the Leiden System (GRBSA 2), Durham. Eretria XI = D. Knoepfler, Décrets érétriens de proxénie et de citoyenneté (Eretria XI), Lausanne 2001. Fachard, S. 2013. “Eleutherai as the Gates to Boeotia,” in Pratiques militaires et art de la guerre dans le monde grec antique: Études offertes à Pierre Ducrey à l’occasion de insc r ip t ions f r om panakton son 75 e anniversaire (Revue des études militaires anciennes 6), ed. C. Brélaz and S. Fachard, Paris, pp. 81–106. Fachard, S., S. C. Murray, A. R. Knodell, and K. Papangeli. 2020. “The Fortress of Eleutherai: New Insights from Survey, Architecture, and Epigraphy,” Hesperia 89, pp. 475–549. Ferguson, W. S. 1911. Hellenistic Athens: An Historical Essay, London. Friend = J. Friend, The Athenian Ephebeia in the Fourth Century B.C.E. (Brill Studies in Greek and Roman Epigraphy 13), Leiden 2019. Gabrielsen, V. 1994. Financing the Athenian Fleet: Public Taxation and Social Relations, Baltimore. Gaifman, M. 2012. Aniconism in Greek Antiquity (Oxford Studies in Ancient Culture and Representation), Oxford. Garlan, Y. 1978. “Études d’histoire militaire et diplomatique XVI: Décret de Rhamnonte voté par des hypaithroi,” BCH 102, pp. 103– 108. Gauthier, P. 1976. Un commentaire historique des Poroi de Xénophon (Centre de recherches d’histoire et de philologie 3; Hautes études du monde gréco-romain 8), Paris. Gehrke, H.-J. 1976. Phokion: Studien zur Erfassung seiner historischen Gestalt (Zetemata 64), Munich. Gerstel, S. E. J., M. Munn, H. E. Grossman, E. Barnes, A. H. Rohn, and M. Kiel. 2003. “A Late Medieval Settlement at Panakton,” Hesperia 72, pp. 147–234. Habicht, C. 1961. “Neue Inschriften aus dem Kerameikos,” AM 76, pp. 143–146. ———. 1983. Studien zur Geschichte Athens in hellenistischer Zeit (Hypomnemata 73), Göttingen. ———. 1997. Athens from Alexander to Antony, trans. D. L. Schneider, Cambridge, Mass. ———. 2003. “Athens After the Chremonidean War: Some Second Thoughts,” in The Macedonians in Athens 322–229 B.C. Proceedings of an International Conference Held at the University of Athens, May 24–26, 2001, ed. O. Palagia and S. V. Tracy, Oxford, pp. 52–55. Hansen, M. H. 1985. Demography and Democracy: The Number of Athenian Citizens in the Fourth Century B.C., Herning. ———. 1994. “The Number of Athenian Citizens secundum Sekunda,” EchCl 38, pp. 299–310. ———. 2006. Studies in the Population of Aigina, Athens and Eretria (Historisk-filosofiske meddelelser 94), Copenhagen. Harris, D. 1995. The Treasures of the Parthenon and the Erechtheion (Oxford Monographs on Classical Archaeology), Oxford. Henderson, J. 1975. The Maculate Muse: Obscene Language in Attic Comedy, New Haven. Humphreys, S. C. 2004–2009. “Ephêboi at Oropos,” Horos 17–21, pp. 83–90. I.Délos = F. Dürrbach, P. Roussel, M. Launey, A. Plassart, and J. Coupry, eds., Inscriptions de Délos, Paris, 1926–1972. I.Eleusis = K. Clinton, Eleusis: The Inscriptions on Stone: Documents of the Sanctuary of the Two Goddesses and Public Documents of the Deme, Athens. 1 = Text (1A); Plates (1B) (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας 236), 2005. 2 = Commentary (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας 259), 2008. I.Oropos = V. Petrakos, Οἱ ἐπιγραφὲς τοῦ Ὠρωποῦ, Athens 1997. I.Rhamnous = V. Petrakos, Ὁ δῆμος τοῦ Ραμνοῦντος 2: Οἱ ἐπιγραφές (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας 182), Athens 1999. Jordan, B. 1972. The Athenian Navy in the Classical Period: A Study of Athenian Naval Administration and Military Organization in the Fifth and Fourth Centuries B.C. (University of California Publications: Classical Studies 13), Berkeley. Kaltsas, N. 2002. Sculpture in the National Archaeological Museum, Athens, trans. D. Hardy, Athens. Kearns, E. 1989. The Heroes of Attica (BICS Suppl. 57), London. Kennell, N. M. 1995. The Gymnasium of Virtue: Education and Culture in Ancient Sparta (Studies in the 335 History of Greece and Rome), Chapel Hill. Kent, J. H. 1941. “A Garrison Inscription from Rhamnous,” Hesperia 10, pp. 342–350. Kerameikos III = W. Peek, Inschriften, Ostraka, Fluchtafeln (Kerameikos III), Berlin 1941. Kralli, I. 2003. “The Date and Context of Divine Honours for Antigonos Gonatas: A Suggestion,” in The Macedonians in Athens 322–229 B.C. Proceedings of an International Conference Held at the University of Athens, May 24–26, 2001, ed. O. Palagia and S. V. Tracy, Oxford, pp. 61–66. La Follette, L. 1986. “The Chalkotheke on the Athenian Akropolis,” Hesperia 55, pp. 75–87. Lambert, S. D. 1998. The Phratries of Attica (Michigan Monographs in Classical Antiquity), rev. ed., Ann Arbor. ———. 1999. “IG II2 2345, Thiasoi of Herakles, and the Salaminioi Again,” ZPE 125, pp. 93–130. Launey, M. 1950. Recherches sur les armées hellénistiques 2: Recherches sociologiques (BÉFAR 169), Paris. Lewis, D. M. 1973. “Attic Ephebic Inscriptions,” CR 23, pp. 254–256. LGPN = A Lexicon of Greek Personal Names, Oxford 1987– I = P. M. Fraser and E. Matthews, eds., The Aegean Islands, Cyprus, Cyrenaica, 1987. II = M. J. Osborne and S. G. Byrne, eds., Attica, 1994. [repr. 1996] IIIA = P. M. Fraser and E. Matthews, eds., The Peloponnese, Western Greece, Sicily, and Magna Graecia, 1997. Lhôte, É. 2006. Les lamelles oraculaires de Dodone (Centre de recherches d’histoire et de philologie 3; Hautes études du monde gréco-romain 36), Geneva. MacDowell, D. M., ed. 2000. Demosthenes: On the False Embassy (Oration 19), Oxford. Maier, F. G. 1959. Griechische Mauerbauinschriften 1: Texte und Kommentare (Vestigia 1), Heidelberg. Marsden, E. W. 1969. Greek and Roman Artillery: Historical Development, Oxford. 336 Meritt, B. D. 1940. “Greek Inscriptions,” Hesperia 9, pp. 53–96. Meyer, E. A. 2010. Metics and the Athenian Phialai-Inscriptions: A Study in Athenian Epigraphy and Law (Historia Suppl. 208), Stuttgart. Miller, S. G. 1977. “Excavations at Nemea, 1976,” Hesperia 46, pp. 1–26. Mitchel, F. W. 1961. “The Cadet Colonels of the Ephebic Corps,” TAPA 92, pp. 347–357. ———. 1964. “Derkylos of Hagnous and the Date of I.G., II2, 1187,” Hesperia 33, pp. 337–351. ———. 1975. “The So-called Earliest Ephebic Inscription,” ZPE 19, pp. 233–243. ———. 1984. “An Ephebic Dedication of 334/3 Reconsidered,” AncW 9, pp. 114–118. Mitsos, M. T. 1965. “Ἐκ τοῦ Ἐπιγραφικοῦ Μουσείου (VII),” ArchEph 104 [1967], pp. 131–138. Munn, M. 1988. “ Ἐν μεθορίοις τῆς Ἀττικῆς καὶ τῆς Βοιωτίας,” in Επετηρίς της Εταιρείας Βοιωτικών Μελετών: Α′ Διεθνές Συνέδριο Βοιωτικών Μελετών, Θήβα, 10–14 Σεπτεμβρίου 1986 1: Προϊστορία και αρχαίοι χρόνοι, ed. A. P. Bekiaris, Athens, pp. 363–371. ———. 1989. “New Light on Panakton and the Attic-Boiotian Frontier,” in Boiotika: Vorträge vom 5. Internationalen Böotien-Kolloquium zu Ehren von Professor Dr. Siegfried Lauffer, Institut für Alte Geschichte, Ludwig-Maximilians-Universität München, 13.–17. Juni 1986 (Münchener Arbeiten zur Alten Geschichte 2), ed. H. Beister and J. Buckler, Munich, pp. 231–244. ———. 1993. The Defense of Attica: The Dema Wall and the Boiotian War of 378–375 B.C., Berkeley. ———. 1996. “First Excavations at Panakton on the Attic-Boiotian Frontier,” Boeotia Antiqua 6, pp. 47–58. ———. 2010. “Panakton and Drymos: A Disputed Frontier,” in Attika: Archäologie einer “zentralen” Kulturlandschaft. Akten der internationalen Tagung vom 18.–20. Mai 2007 in Marburg (Philippika 37), ed. H. Lohmann and T. Mattern, Wiesbaden, pp. 189–200. mar k munn Munn, M., and M. L. Z. Munn. 1990. “On the Frontiers of Attica and Boiotia: The Results of the Stanford Skourta Plain Project,” in Essays in the Topography, History, and Culture of Boiotia (Teiresias Suppl. 3), ed. A. Schachter, Montreal, pp. 32–40. Munn, M. L. Z., and M. Munn. 1991. “Stanford Team Explores the AtticBoiotian Frontier,” Newsletter of the American School of Classical Studies 27, pp. 3, 14–15. Ober, J. 1985. Fortress Attica: Defense of the Athenian Land Frontier, 404– 322 B.C. (Mnemosyne Suppl. 84), Leiden. ———. 1992. “Towards a Typology of Greek Artillery Towers: The First and Second Generation (c. 375–275 b.c.),” in Fortificationes antiquae (McGill Monographs in Classical Archaeology and History 12), ed. S. Van de Maele and J. M. Fossey, Amsterdam, pp. 147–169. Oetjen, R. 2000. “War Demetrios von Phaleron, der Jüngere, Kommissar des Königs Antigonos II. Gonatas in Athen?” ZPE 131, pp. 111–117. ———. 2014. Athen im dritten Jahrhundert v. Chr.: Politik und Gesellschaft in den Garnisonsdemen auf der Grundlage der inschriftlichen Überlieferung (Reihe Geschichte 5), Duisburg. Oliver, G. J. 2007. War, Food, and Politics in Early Hellenistic Athens, Oxford. Olynthus X = D. M. Robinson, Metal and Minor Miscellaneous Finds: An Original Contribution to Greek Life (Olynthus X), Baltimore 1941. Orlandos, A. K. 1954. “ Ἀνασκαφαί Ἐλευσίς,” Ergon 1954 [1955], pp. 3–10. PAA = J. S. Traill, Persons of Ancient Athens, 22 vols., Toronto 1994–2016. Palagia, O., and D. M. Lewis. 1989. “The Ephebes of Erechtheis, 333/2 b.c., and Their Dedication,” BSA 84, pp. 333–344. Papazarkadas, N. 2011. Sacred and Public Land in Ancient Athens (Oxford Classical Monographs), Oxford. Parke, H. W. 1977. Festivals of the Athenians (Aspects of Greek and Roman Life), London. Parzinger, H. 2004. Die Skythen (Beck’sche Reihe 2342), Munich. Pélékidis, C. 1962. Histoire de l’éphébie attique des origines à 31 avant 337 insc r ip t ions f r om panakton Jésus-Christ (École française d’Athènes: Travaux et mémoires 13), Paris. Petrakos, V. C. 1976. “Ἀνασκαφὴ Ῥαμνοῦντος,” Prakt 132, pp. 5–60. ———. 1999. Ὁ δῆμος τοῦ Ραμνοῦντος: Σύνοψη τῶν ἀνασκαφῶν καὶ τῶν ἐρευνῶν (1813–1998) 1: Τοπογραφία (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας 181), Athens. ———. 2004. “Οἱ ἔφηβοι τῆς Λεοντίδος τοῦ 333/2 π.Χ.,” PraktAkAth 79, pp. 167–176. Pounder, R. L. 1975. “The Origin and Meaning of ΘΕΟΙ in Greek Inscription Headings” (diss. Brown Univ.). Pritchett, W. K. 1949. “The Ephebes of Oineis,” in Commemorative Studies in Honor of Theodore Leslie Shear (Hesperia Suppl. 8), Princeton, pp. 273–278. ———. 1955. “Dotted Letters in Greek Epigraphy,” AJA 59, pp. 55–61. Reinmuth = O. W. Reinmuth, The Ephebic Inscriptions of the Fourth Century B.C. (Mnemosyne Suppl. 14), Leiden 1971. Reinmuth, O. 1948. “The Ephebate and Citizenship in Attica,” TAPA 79, pp. 211–231. Rhodes, P. J. 1972. The Athenian Boule, Oxford. ———. 1981. A Commentary on the Aristotelian Athenaion Politeia, Oxford. Rhodes, P. J., and R. Osborne, eds. 2003. Greek Historical Inscriptions 404–323 B.C., Oxford. Rihll, T. 2007. The Catapult: A History (Weapons in History), Yardley, Pa. Rouse, W. H. D. 1902. Greek Votive Offerings, Cambridge. Roussel, P. 1941. “Sur quelques inscriptions attiques,” RA 18, pp. 209–232. Schmitt, H.-O. 2007. “Die Angriffswaffen,” in Kalapodi II: Ergebnisse der Ausgrabungen im Heiligtum der Artemis und des Apollon von Hyampolis in der antiken Phokis, ed. R. C. S. Felsch, Mainz, pp. 423–551. Schwenk, C. J. 1985. Athens in the Age of Alexander: The Dated Laws and Decrees of “The Lykourgan Era,” 338–322 B.C., Chicago. Sekunda, N. 1992. “Athenian Demography and Military Strength 338– 322 b.c.,” BSA 87, pp. 311–355. Snodgrass, A. M. 1967. Arms and Armour of the Greeks (Aspects of Greek and Roman Life), Ithaca. Stais, B. 1891. “Ἀγάλματα ἐκ Ῥαμνοῦντος,” ArchEph 30, pp. 44–62. Steinbock, B. 2011. “A Lesson in Patriotism: Lycurgus’ Against Leocrates, the Ideology of the Ephebeia, and Athenian Social Memory,” ClAnt 30, pp. 279–317. Tracy, S. V. 1994. “Hands in Greek Epigraphy: Demetrios of Phaleron,” Boeotia Antiqua 4, pp. 151–161. ———. 1995. Athenian Democracy in Transition: Attic Letter-Cutters of 340 to 290 B.C. (Hellenistic Culture and Society 20), Berkeley. ———. 2000. “Demetrius of Phalerum: Who Was He, and Who Was He Not?” in Demetrius of Phalerum: Text, Translation, and Discussion (Rutgers University Studies in Classical Humanities 9), ed. W. W. Fortenbaugh and E. Schütrumpf, New Brunswick, pp. 331–345. ———. 2003. Athens and Macedon: Attic Letter-Cutters of 300 to 229 B.C. (Hellenistic Culture and Society 38), Berkeley. ———. 2015. “Eurykleides I of Kephisia and the Office of Hoplite General at Athens,” in ΑΞΩΝ: Studies in Honor of R. S. Stroud (Ελληνική Επιγραφική Εταιρεία 17), ed. A. P. Matthaiou and N. Papazarkadas, Athens, pp. 291–298. ———. 2016. “Sophilos, Son of Aristotle, of Phyle,” Hyperboreus 22, pp. 263–269. Traill, J. S. 1975. The Political Organization of Attica: A Study of the Demes, Trittyes, and Phylai, and Their Representation in the Athenian Council (Hesperia Suppl. 14), Princeton. Tritle, L. A. 1988. Phocion the Good, New York. Vanderpool, E. 1978. “Roads and Forts in Northwestern Attica,” CSCA 11, pp. 227–245. Mark Munn The Pennsy lvania S tat e Univ ersit y department of c l assics and anc ient medit er ranean studie s w eav er b uil ding univ ersit y par k, pennsy lvania 168 02-5500 [email protected]