Greek Statue Terms Revisited:

What does ἀνδριάς mean?

Catherine M. Keesling

I T HAS LONG BEEN a truism in scholarship on Greek

sculpture that the ancient Greeks had no word that meant
“statue.” Rather, the Greek terms used to describe or refer
to statues depended upon their functions and the contexts in
which they stood.1 Though recent scholarship has argued
against any clear-cut ontological distinction between cult
statues and other divine images, a series of specialized terms—
ξόανον, βρέτας, and ἑδός—seem to have been used beginning
in the sixth century BCE to denote a divine statue displayed on
axis in a temple building.2 In the early fourth century, a distinc-
tion began to be made between a divine statue called an
1 J. Ducat, “Fonctions de la statue dans la Grèce archaïque. kouros et

kolossos,” BCH 100 (1976) 246: “Le grec n’a pas, on le sait, de terme
générique pour désigner ‘la statue’; il a plusieurs substantifs, correspondant
soit à tel ou tel aspect (‘objet poli’), soit à telle ou telle fonction (‘siège’ de la
divinité, offrande, représentation d’homme, image) de la statue.” A. Du-
plouy, Le prestige des élites. Recherches sur les modes de reconnaissance sociale en Grèce
entre les Xe et Ve siècles avant J.-C. (Paris 2006) 186: “Or en grec, il n’existe pas
de mot générique pour désigner la ‘statue’.” A typical generalizing state-
ment from a handbook on Greek sculpture is C. Rolley, La sculpture grecque I
(Paris 1994) 22: “Les textes classiques distinguent souvent la statue d’une
divinité, agalma, de celle d’un être humain, ἀνδριάς (andrias).”
2 For cult statues and the Greek terms used to describe them see

especially I. B. Romano, Early Greek Cult Images (diss. Univ. of Pennsylvania

1980) 42–57; A. A. Donohue, Xoana and the Origins of Greek Sculpture (Atlanta
1988), and “The Greek Images of the Gods: Considerations on Ter-
minology and Methodology,” Hephaistos 15 (1997) 31–45; T. S. Scheer, Die
Gottheit und ihr Bild, Untersuchungen zur Funktion griechischer Kultbilder in Religion
und Politik (Munich 2000) 8–34.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
2017 Catherine M. Keesling
838 GREEK STATUE TERMS REVISITED

ἄγαλµα and a portrait statue called an εἰκῶν.3 From the third

century onward, a statue representing a Hellenistic king or a
prominent civic benefactor might be referred to as an agalma if
its function was primarily religious, but an eikon or an andrias if
it was likened to portraits of other persons.4 As Simon Price
pointed out, agalma, andrias, and eikon could all be used to de-
scribe statues representing the Roman emperor, and so “the
observer could use different terms depending on what aspect of
the object [statue] he wished to stress.”5
One of the most common Greek terms for a statue, andrias, is
also the least discussed and the least understood.6 Recent
studies of Greek portrait sculpture raise key questions about
terminology that thus far remain unanswered.7 When did

3 Another important Greek statue term whose meaning has been debated

in recent scholarship, kolossos, lies beyond the scope of this article. See most
recently M. Dickie, “What is a Kolossos and How Were Kolossoi Made in
the Hellenistic Period?” GRBS 37 (1996) 237–257, and E. Kosmetatou and
N. Papalexandrou, “Size Matters: Poseidippos and the Colossi,” ZPE 143
(2003) 53–58.
4 See D. Fishwick, “Statues Taxes in Roman Egypt,” Historia 38 (1989)

335–347, esp. 340–344 (portraits of the Ptolemies).

5 S. R. F. Price, Rituals and Power (Cambridge 1984) 176–179 (quotation at

176); see also P. Veyne, “Les honneurs posthumes de Flavia Domitilla et les
dédicaces grecques et latines,” Latomus 21 (1962) 49–98; A. Oliver, “Honors
to Romans: Bronze Portraits,” in C. C. Mattusch (ed.), The Fire of Hephaistos:
Large Classical Bronzes from North American Collections (Cambridge [Mass.] 1996)
144–145. K. Tuchelt’s assertion, Frühe Denkmäler Roms in Kleinasien I (IstMitt
Beih. 23 [1979]) 68–71, that the material distinction between marble statues
and bronze ones was paramount in determining which were called agalmata
(marble) and which were called eikones (bronze) has been criticized by K.
Höghammar, Sculpture and Society. A Study of the Connection between the Free-
Standing Sculpture and Society on Kos (Uppsala 1993) 68–70, and D. Damaskos,
Untersuchungen zu hellenistischen Kultbildern (Stuttgart 1999) 304–309.
6 For andrias see, in addition to the citations above, H. Philipp, Tektonon

Daidala: Der bildende Künstler und sein Werk im vorplatonischen Schrifttum (Berlin
1968) 106–107, and S. Bettinetti, La statua di culto nella pratica rituale greca (Bari
2001) 37–42, both with an emphasis upon Classical literary references.
7 R. Krumeich, Bildnisse griechischer Herrscher und Staatsmänner im 5. Jahr-

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 839

andrias, a term whose etymology naturally connects it with male

figures, begin to be used to refer to portraits of women?8 Why
did andrias eventually become a more common term for
portrait statues in Greek inscriptions than eikon? And when did
andriantopoios come to denote a maker of bronze statues
specifically? As I hope to show, though it is clear that andrias,
beginning in the fifth century BCE, was sometimes used in
both literature and inscriptions to refer to portrait statues, it
can only be said to mean “portrait” from the second century
BCE onward. Appealing to later Greek usage risks obscuring
several important points about what the Greeks called their
statues before the second century BCE. In the Archaic and
Classical periods, both inscriptions and literary sources used
andrias for any male statue. Eikon became a standard term for a
portrait statue once honorific decrees awarding portraits began
to be inscribed on stone; thereafter, andrias was used sporadi-
cally in such decrees to convey the technical specifications of
male portrait statues called eikones. The earliest inscribed de-
crees in honor of women in the Greek world, dating to the
fourth and third centuries BCE, avoided referring to female
portraits as andriantes. The inscribed Hellenistic temple inven-
tories from Delos consistently used a constellation of other
terms (ζῷον, ζῴδιον, and ζῳδάριον especially) to avoid calling
___
hundert v. Chr. (Munich 1997); F. Queyrel, Les portraits des Attalides: Fonction et
representation (Paris 2003); S. Dillon, Ancient Greek Portrait Sculpture: Contexts,
Subjects, and Styles (Cambridge 2006); P. Schultz and R. von den Hoff (eds.),
Early Hellenistic Portraiture: Image, Style, Context (Cambridge 2007); S. Dillon,
The Female Portrait Statue in the Greek World (Cambridge 2010); J. Ma, Statues
and Cities: Honorific Portraits and Civic Identity in the Hellenistic World (Oxford
2013).
8 For andrias as a term applicable to female portraits see Dillon, The Female

Portrait Statue 36–37 and 189 n.133, citing R. R. R. Smith, Hellenistic Royal
Portraits (Oxford 1988) 16 and 35, and LSCG Suppl. 107, a decree of the
third century BCE from the Asklepieion on Rhodes regulating the place-
ment of offerings. The decree makes a distinction between andriantes (statues)
and anathemata (non-statue offerings); in this particular case, andrias means
any statue, not only portraits.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
840 GREEK STATUE TERMS REVISITED

female figures andriantes, down to ca. 200 BCE. Thereafter,

both inventories and honorific decrees frequently made a
distinction between painted portraits, called eikones, and portrait
statues, called andriantes. This seems to be the origin of the use
of andrias in the Roman imperial period as a standard Greek
term for a portrait statue, either male or female. Finally, the
terms andriantopoios and agalmatopoios seem to have been used
interchangeably to mean “sculptor” until the time of Aristotle.
Of the terms for statues enumerated in my introductory
paragraphs, agalma by far dominates the literary and epigraphi-
cal lexicon of the Archaic period and the fifth century. As
Joseph Day has most recently shown, in inscribed dedicatory
epigrams agalma connoted “an ornament that generates friendly
responses with its beauty.”9 Down to the end of the fifth cen-
tury, at the same time that agalma in literature often denoted
the statue of a god or a hero, it continued to be used in epi-
grams inscribed on votive offerings that were not statues, such
as ceramic tiles and vases. In contrast, andrias occurs in only
three inscriptions on statues or statue bases before the end of
the fifth century, each time in reference to a schematic, nude,
standing, male figure of the kouros type. The first appears on the
base of the so-called Naxian colossus on Delos, a massive
marble kouros of ca. 590–580 representing Apollo (CEG 401):10
[τ]ο͂ ἀϝυτο͂ λίθο ἐµὶ ἀνδριὰς καὶ τὸ σφέλας.
I am (made) of a single stone, both statue and base.
The second inscription with andrias is a late Archaic (ca. 500)
prose dedication to Apollo on the base for a kouros from

9 J. W. Day, Archaic Greek Epigram and Dedication, Representation and Reperfor-

mance (Cambridge 2010) 124–129 (quotation at 125).
10 For the date see P. Bruneau and J. Ducat, Guide de Délos3 (Paris/Athens

1983) 125–128, no. 9, who interpret the verse as a riddle or adynaton; see
also P. Bruneau, “Deliaca (VII),” BCH 112 (1988) 577–579. According to
Herodotus (2.176), the pharaoh Amasis dedicated at Memphis a colossal
statue and two smaller ones on the same base, all three statues cut from the
same block (τοῦ αὐτοῦ ἐόντος λίθου).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 841

Neandria in the Troad:11

τόνδε τ̣ὸν ἀνδ[ριάντα Ἀπό]λλονα̣ ὀνέθε̄-
κε Ἑρµέ̣ας ἀρα[σαµένο] το͂ παῖδος, vacat
Ὀγεµάχ[ο] (or ᾿Ογεµάχ[ειος]). vacat
Hermeas son of Οgemachos(?) dedicated this Apollo statue, his
son having vowed it.
The third inscription, also in prose, was carved on the thigh of
a colossal kouros of the mid-sixth century found along the
Sacred Way leading from Miletus to the sanctuary of Apollo at
Didyma. It refers to a group of andriantes displayed together:12
τόσ̣<δ>ε τὸς ἀνδριάντ[ας]
[Λά]τµιοι ἀν[έθε]σαν (κτλ.)
The Latmians dedicated these statues …
The andriantes might have consisted either of male statues ex-
clusively or of a combination of male and female figures; pairs
of votive kouroi and korai are attested elsewhere.13 In light of a
similar reference to multiple statues in the inscription on the
Archaic seated male figure dedicated by Chares at Didyma,
however, we should perhaps reconstruct the Latmian dedi-
cation as a group representing either a family or the members
of a genos, predominantly male. Just such a statue group is now

11 E. Schwertheim, Neue Forschungen zu Neandria und Alexandria Troas (Asia

Minor Studien 11 [1994]) 40–41, no. 2 [SEG XLIV 986]. Earlier editions
are R. Koldewey, Neandria (Berlin 1891) 27–28 (who reported the discovery
of fragments of a kouros near the base); Ad. Wilhelm, Beiträge zur griechischen
Inschriftenkunde (Vienna 1909) 7–8; M. L. Lazzarini, Formule delle dediche votive
(Rome 1976) no. 767 (who emended the accusative Ἀπό]λλονα̣ to dative
Ἀπό]λλον⟨ι⟩); and LSAG 2 360 and 362, no. 9 (ca. 500–475 BCE?).
12 I.Didyma 12–13 [SEG XVI 711], with M. Wörrle, CRAI 2003, 1371

n.44; cf. LSAG 2 332–333 and 342, no. 25.

13 Two nearly identical korai of ca. 570–560 were dedicated by Chera-

myes in the Samian Heraion, the so-called Hera from Samos now in the
Louvre and a second statue found in 1984: J. Franssen, Votiv und Repräsen-
tation: Statuarische Weihungen archaischer Zeit aus Samos und Attika (Heidelberg
2011) 65–69, nos. A5 and A6.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
842 GREEK STATUE TERMS REVISITED

attested in the so-called Temenos along the Sacred Way be-

tween Didyma and Miletus, where several seated figures were
placed on a semi-circular base in an enclosure along with a cult
dining room.14
As is well known, the term kouros for an Archaic, schematic,
nude male figure in marble is a modern coinage.15 On the rare
occasions when Archaic Greeks needed a term to refer to a
kouros statue, they chose andrias.16 But that andrias simply meant
“statue” is demonstrated by Herodotus’ usage in the last
quarter of the fifth century.17 He used agalma to mean “divine
image,” and this term occurs in his text about four times more
frequently than either andrias or eikon; his particular interest in
agalmata is consistent with his preoccupation with the divine
and divine agency in history. Yet he also called the colossal
bronze Apollo dedicated at Delphi by the Greeks from the
Persian spoils from the battle of Salamis an andrias (8.121.2).18
14 Seated statue of Chares and its inscription: LSAG2 332, no. 29; K.

Tuchelt, Die archaischen Skulpturen von Didyma (IstForsch 27 [1970]) 78–80, no.
K47 and pls. 43–46; C. M. Keesling, The Votive Statues of the Athenian Acropolis
(Cambridge 2003) 102–106. Temenos on the Sacred Way and its statues:
K. Tuchelt, H. R. Baldus, T. G. Schattner, and H. P. Schneider, Ein Kult-
bezirk an der Heiligen Strasse von Milet nach Didyma (Didyma III.1 [1996]); Du-
plouy, Le prestige 203–214 and 228–233; A. Herda, Der Apollon-Delphinios-Kult
in Milet und die Neujahrsprozession nach Didyma (Milesische Forschungen 4
[2006]) 343–350.
15 For the modern use of the term see M. Meyer and N. Brüggermann,

Kore und Kouros: Weihgaben für die Götter (Vienna 2007) 93, with references to
earlier scholarship. An ancient precedent is the reference to golden kouroi at
Homer Od. 7.100–103.
16 Pace Meyer in Meyer and Brüggermann, Kore und Kouros 29, given the

very small number of inscriptions associated with kouroi it is probably not

significant that only one kouros inscription (no. 290) uses the term agalma.
17 On Herodotus’ statue terminology see A. Hermary, “Les noms de la

statue chez Hérodote,” in M.-Cl. Amouretti and P. Villard (eds.), ΕΥΚΡΑΤΑ.

Mélanges offerts à Claude Vatin (Aix-en-Provence 1994) 21–29.
18 Herodotus’ contemporary Aristophanes refers to a statue of the Athen-

ian hero Pandion as an andrias (Pax 1183). In the inscribed accounts of the
naopoioi at Delphi for 340/39 BCE, a statue of Apollo is referred to simply as

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 843

He used both agalma and andrias to describe another Apollo,

this one gilded (6.118.3), but elsewhere distinguished an agalma
representing Apollo from the andrias representing the legendary
mystic Aristeas of Proconnesos that stood beside it in the agora
of Metapontum (4.15.4). In the temple of Bel (Zeus) at Bab-
ylon, Herodotus compared the seated gilded agalma of Zeus
with a reported solid gold andrias twelve cubits tall that had
previously stood there before Darius removed it (1.183); the
comparison makes it clear that the andrias too was an image of
Zeus.19 Mixed male/female statue pairs and groups in Herodo-
tus can also be andriantes: the portraits of the pharaoh Sesostris
and his wife in front of the temple of Ptah (Hephaistos) at
Memphis (2.110.1–2); Mikythos’ dedication at Olympia of
gods, goddesses, divine personifications, and portraits (7.170.4);
and the Phocian dedication at Delphi consisting of several
statues of unknown type surrounding a tripod (8.27.5).
Just as andrias in Herodotus means “statue,” not “portrait,”
he uses eikon for statues representing a variety of subjects:
images of heroes (2.106, Memnon), animals (1.50.3, Croesus’
___
the µέγας ἀνδριάς, the big statue (FD III.5 22.30 = CID II 43, cited by A.
Hermary, “Témoignage des documents figurés sur la société chypriote
d’époque classique,” in E. Peltenburg [ed.], Early Society in Cyprus [Edin-
burgh 1989] 195 n.9).
19 The bilingual Phoenician-Greek dedicatory inscriptions on three

fourth-century statuette bases from sanctuaries of Apollo (Reshef Mikal) on

Cyprus use the term andrias: O. Masson, Les inscriptions chypriotes syllabiques
nos. 215, 216, 220. Masson thought these andriantes were portraits of their
dedicators: “le mot ἀνδριάς désigne évidemment la statuette à l’effigie du
dédicant jadis fixée sur la base” (p.225); cf. Hermary, in Early Society in
Cyprus 187–189 and 195 n.9. Similarly, P. Briant, “Droaphernès et la statue
de Sardes,” in M. Brosius and A. Kuhrt (eds.), Studies in Persian History: Essays
in Memory of David M. Lewis (Leiden 1998) 205–226, assumed that an andrias
in a late fifth- or fourth-century document recording the dedication of a
statue by a Persian named Droaphernes at Sardis (I.Estremo oriente 235) was a
portrait of Droaphernes himself; cf. K. Rigsby, “A Religious Association at
Sardes,” AncSoc 44 (2014) 3–5, who cites five post-fifth-century dedicatory
inscriptions in which andrias refers to a statue of the recipient deity.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
844 GREEK STATUE TERMS REVISITED

gilded lion at Delphi), as well as men and women, both

sculpted (1.31.5, the statues of Kleobis and Biton at Delphi;
2.182.2, the wooden portraits of the pharaoh Amasis and his
wife) and painted (2.182.1, an εἰκόνα ἑωυτοῦ γραφῇ
εἰκασµένην of Amasis). It may be significant that the only time
Herodotus chose to mention a non-agalma female statue on its
own, he called it not an andrias but an eikon: he notes that a
small golden εἴδωλον at Delphi—a term he uses only three
times—was identified by the Delphians in his own time as an
eikon (representation) of Croesus’ female bread baker (1.51.5).20
Though Herodotus did not use it, the term κόρη for female
figures appears in some fifth- and fourth-century Athenian in-
scriptions as an alternative to andrias.21 The earliest occurrence
of kore in this sense, on an inscribed statue base of ca. 480 on
the Acropolis, inspired Theodoros Sophoulis and Henri Lechat
to coin the term kore for the female equivalent of the kouros
when large numbers of Archaic marble female statues were
coming out of the Acropolis excavations.22 The inscription
reads (IG I3 828 = DAA 229 = CEG 266):
[τέ]νδε κόρεν ἀ[ν]έθεκεν ἀπαρχὲν
[Ναύ?]λοχος ἄγρας : ἓν οἱ ποντοµέδ-
[ον χρ]υ̣σοτρία[ι]ν’ ἔπορεν.

20 For eidolon in Herodotus see also 6.58 (a body-replica of a Spartan king

killed in battle is carried to the tomb on a bier) and 5.92 (the ghost of the
Corinthian tyrant Periander’s murdered wife Melissa). For the possible
significance of the appellation bread-baker for the female figure dedicated
by Croesus see L. Kurke and A. Garrett, “Pudenda Asiae Minoris,” HSCP
96 (1994) 75–83, and A. Jacquemin, Offrandes monumentales à Delphes (Paris
1999) 198, no. 344. In reality, Croesus’ bread-baker may have been a figure
of Artemis, or even a female support figure detached from one of the num-
erous precious metal vessels Croesus dedicated in the sanctuary.
21 Keesling, The Votive Statues 110–114 and 241–242; cf. Meyer in Meyer

and Brüggermann, Kore und Kouros 31–32.

22 T. Sophoulis, Τὰ ἐν Ἀκροπόλη ἀγάλµατα κορῶν ἀρχαικῆς τέχνης

(Athens 1892) 6 and 12–15; H. Lechat, Au Musée de l’Acropole. Etudes sur la

sculpture en Attique (Paris/Lyon 1903) 276–277.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 845

[Nau]lochos dedicated this female statue, first fruits of the catch

which the sea-ruler with the golden trident provided to him.
The statue dedicated by [Nau]lochos, now lost, might in fact
have been made of bronze; in any case, the term kore reveals a
female as opposed to a male figure.23 Like andrias in the Archaic
inscriptions quoted above, kore did not in itself imply that the
statue so described was anonymous. Athenian temple inven-
tories and building accounts of the Classical period also used
kore for female figures whose identity (for whatever reason) was
not specified. Two examples of a gilded (κατάχρυσος or χρυσῆ)
kore on a stele were listed among the golden objects stored in
the Parthenon and the Hekatompedon beginning in 434/3;
here stele refers to a small pillar base.24 Similarly, the Par-
thenon inventories of 369 (IG II2 1424a.279) and 368
(1425.380–381) mention eleven “korai from the baskets.”25 The
building account for the Erechtheion dated to 409/8 (IG I3
474.86) refers to the six marble caryatids holding up the south

23 Similarly, the funerary epigram by Kleoboulos of Lindos for King

Midas of Phrygia (quoted by Diogenes Laertius 1.89; PMG 581) refers to a
female figure standing over the tomb as a parthenos (maiden): καὶ τὸ ἐπί-
γραµµά τινες τὸ ἐπὶ Μίδᾳ τοῦτόν φασι ποιῆσαι· χαλκῆ παρθένος εἰµί, Μίδα
δ’ ἐπὶ σήµατι κεῖµαι (“And some say he [Kleoboulos] composed the epi-
gram for Midas: ‘I am the bronze maiden set up over the tomb of Midas’ ”).
24 D. Harris, The Treasures of the Parthenon and Erechtheion (Oxford 1995),

nos. IV.20 (Parthenon inventories, 434/3 through 412/1) and V.90 (Heka-
tompedon, 434/3 through 408/7). For a “stele” as a statue base see R. H.
W. Stichel, “Columella-Mensa-Tabellum. Zur Form der attischen Grab-
mäler im Luxusgesetz des Demetrios von Phaleron,” AA 107 (1992) 436–
438.
25 R. Hamilton, Treasure Map, A Guide to the Delian Inventories (Ann Arbor

2000), Athena Treasure B.246. Explanations for what these korai were range
from bronze statuettes dedicated by the girls who served as kanephoroi, stored
in baskets (J. Schelp, Das Kanoun: Der griechische Opferkorb [Würzburg 1975]
20) to figural attachments for the baskets carried by these girls in festival
processions (B. S. Ridgway, Hellenistic Sculpture I [Madison 1990] 587–588
n.13).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
846 GREEK STATUE TERMS REVISITED

porch as korai.26
Herodotus used both andrias and eikon for portrait statues;
eikon became the standard Greek term for a portrait statue only
when honorific portraits emerged at the beginning of the fourth
century BCE. The only pre-Herodotean epigraphic example of
eikon occurs at Olympia in the epigram for Euthymos of Locri,
a three-time victor in boxing who won his last known victory in
472 (I.Olympia 144 [CEG I 399]):
Εὔθυµος Λοκρὸς Ἀστυκλέος τρὶς Ὀλύµπι’ ἐνίκων.
εἰκόνα δ’ ἔστησεν ⟦τήνδε βροτοῖς ἐσορᾶν.⟧
Εὔθυµος Λοκρὸς ἀπὸ Ζεφυρίο ⟦ἀνέθηκε⟧.
Πυθαγόρας Σάµιος ἐποίησεν.
“Euthymos the Locrian, son of Astykles, I won three times at
Olympia.”
He set up this portrait for mortals to wonder at.
Euthymos the Zephyrian Locrian dedicated it.
Pythagoras the Samian made it.
The final words of the second and third lines were changed,

26 Nymphs, and statues of them, could also be called korai. In Pl. Phdr.

230B–C Socrates describes a rural sanctuary belonging to the nymphs and

the river god Acheloös full of “korai and agalmata.” Plutarch (Them. 31)
reports that when Themistokles was in exile in the Persian Empire he visited
the temple of Cybele in Sardis and saw a two-cubit tall bronze kore called
the “water-carrier” (ὑδροφόρον κόρην χαλκήν) that he himself had dedi-
cated on the Acropolis from fines he collected as commissioner of the
Athenian water supply before the Persian Wars. For nymphs as korai see also
CEG 331 = SEG XXII 404 (dedication of ca. 500–480 from Boiotia) and T.
Hadzisteliou-Price, “Double and Multiple Representations in Greek Art
and Religious Thought,” JHS 91 (1971) 56–57. Kore in a statue base inscrip-
tion of the second century CE from Thrace may refer to the dedication of a
statue of a nymph (I.Thrac.Aeg. 431 = SEG LV 780): [Τι(βέριος)?] Κλαύδιος
Φιλόµο[υ]σ̣ος κατ’ ὄναρ ἀνέ̣θηκεν τὴν κόρην χαριστήριον θεοῖς συ[ν]νάοις
µετὰ τέκνω[ν κ]αὶ συ[µβ]ίων, ἰερητ̣[εύοντος – –]ράτου [τοῦ δεῖνος]
(“Tiberius Claudius Philomousos dedicated the kore, in accordance with a
dream, as a thank-offering to the temple-sharing gods, with his children and
retainers, the priest being – –”). All that remains of the statue are the feet of
a small female figure carved in one piece with the base.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 847

probably soon after the portrait was set up in ca. 470, and the
reason why is not clear.27 All the same, eikon was there from the
beginning, and the unprecedented use of this term for Eu-
thymos’ portrait statue seems intended to stress its resemblance
to its subject.28 The earliest surviving inscribed decrees award-
ing the honor of a portrait statue, those for Konon in 394/3,
use eikon as the technical term for this new category of civic
honor. The relevant part of the decree of Erythrai reads (I.
Erythrai 6.13–16 = GHI 8):
ποήσασθαι δὲ
[αὐτοῦ ε]ἰκόνα χαλκῆν
[ἐπίχρυσον] καὶ στῆσαι
[ὅπου ἂν δόξηι] Κόνωνι·
And make a gilded bronze portrait of him and set it up wherever
Konon decides.29
The adoption of eikon as the preferred term for an honorific
portrait led to an explosion in its use in private dedicatory
epigrams of the first half of the fourth century: CEG II includes
28 examples, while Euthymos’ inscription is the only example
in the first volume.
Francis Piejko remarked that in the extant honorific decrees
awarding a statue from the fourth century through ca. 50 BCE,
27 On the inscription see E. Loewy, Inschriften griechischer Bildhauer (Leipzig

1885) no. 23; J. Ebert, Griechische Epigramme auf Sieger an gymnischen und hip-
pischen Agonen (AbhLeip 63.2 [1972]) no. 16; Lazzarini, Formule delle dediche no.
853; LSAG 2 342, no. 19.
28 Eikon as evidence that Euthymos’ portrait was a ‘true’ likeness: M. L.

Lazzarini, “Epigrafia e statua ritratto: alcuni problemi,” AAPat 97 (1984/5)

89–91; J. P. Barron, “Pythagoras’ Euthymos: Some Thoughts on Early Clas-
sical Portraits,” in R. Mellor and L. Tritle (eds.), Text and Tradition, Studies in
Greek History and Historiography in Honor of Mortimer Chambers (Claremont 1999)
37–59.
29 Cf. the fragmentary Athenian decree of 393 for Konon’s patron

Euagoras of Salamis on Cyprus (GHI 11), where no term for the statue has
been preserved on the stone; cf. D. M. Lewis and R. S. Stroud, “Athens
Honors King Euagoras of Salamis,” Hesperia 48 (1979) 180–193.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
848 GREEK STATUE TERMS REVISITED

the term andrias for an honorific portrait is subordinated to

eikon: “the honor conferred is εἰκών, whereas contracting, fabri-
cation, engineering, setting up, supervision, and costs of such
an εἰκών will be often specified by the alternative word
ἀνδριάς.”30 An early example of this pattern is the decree of
Priene in honor of the Ephesian priest Megabyxos, son of
Megabyxos, dating to 334–323 (I.K. Priene I 16 = I.Priene 3 =
Syll.3 282.II):
(8–9) ἐστεφανῶσθαι ὑπὸ τοῦ δήµου χρυ[σ]ῶι στε[φάνωι]
καὶ εἰκόνι χαλκῆι ὡς καλλίστηι …
(15–24) εἶν[αι δὲ]
[α]ὐτῶι καὶ ἐµ πρυτανείωι σίτησ[ιν,] τὸν δὲ ἀνδ[ριάν]-
[τ]α̣ ἐγδοῦναι µετὰ Μεγαβύξου̣ τοὺς νοµοφύ[λακας]
[το]ὺς νοµοφυλακοῦντας µῆ[ν]α Βοηδροµιῶ[να καὶ]
[Πυ]ανοψιῶνα ἐπὶ στεφανηφόρο[υ] Διοφάνευς· σ[τῆσαι]
δὲ τὸν ἀνδριάντα ἐν τῶι ἱερῶι [τῆ]ς Ἀθηνᾶς πρὸ [τοῦ µε]-
τωπίου τοῦ ναοῦ καὶ στήλην π̣[αρα]στῆσαι τοὺ[ς νοµο]-
φύλακας ἀναγράψαντας τόδ[ε τὸ ψ]ήφισµα, τὸ δ[ὲ ανά]-
λωµα εἰς τὸν ἀνδριάντα καὶ τ[ὴν] σ̣τήλην ὑπηρ[ετῆσαι]
τοὺς νεωποίας Ἄδµητον καὶ ․․ᾶδα…
Let him [Megabyxos] be crowned by the demos with a golden
crown and a bronze portrait [eikon] as fine as possible … And let
there be for him also free meals in the prytaneion, and let the
nomophylakes in office in the months Boedromion and Py-
anopsion when Diophanes is stephanephoros contract for the
statue [andrias] in conjunction with Megabyxos; and let the
statue [andrias] stand in the sanctuary of Athena in front of the
façade of the temple and the stele upon which the nomophylakes
have inscribed this decree stand alongside, and the neopoiai Ad-
metos and – – will undertake to cover the expense for the statue
[andrias] and the stele.
All the examples of andrias in the honorific decrees cited by
Piejko refer to portraits of men, which raises the question:

30F. Piejko, “Antiochus Epiphanes Savior of Asia,” RivFil 114 (1986)

425–436 (quotation at 431).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 849

when did female portraits begin to be called andriantes? One

way to answer this question is to look at the small corpus of
Classical and Hellenistic honorific decrees for women.31 The
earliest decree that mentions an honorific portrait of a woman
seems to be one from Erythrai in honor of Mausolus and his
sister/wife Artemisia from the 360s or mid-350s (I.Erythrai
8.11–14 = GHI 56):32
στῆσαι δὲ α[ὀ̣]-
[τοῦ κ]αὶ εἰκόνα χαλκῆν ἐν τῆι ἀγ[ο]-
[ρῆ]ι καὶ Ἀρτεµισίης εἰκόνα
[λιθί]νην ἐν τῶι Ἀθηναίωι·
And set up both a bronze portrait of him [Mausolus] in the
agora and a stone portrait of Artemisia in the Athena temple.
Though statue honors for women are otherwise attested by
statue bases, as we can see from the list in the Appendix below
extant honorific decrees mentioning portrait statues of women
are very few before ca. 200 BCE; none of these uses andrias.33
Even after 200, when honorific portraits of women became

31 For discussion of selected examples see R. van Bremen, The Limits of

Participation: Women and Civic Life in the Greek East (Amsterdam 1996) (second
century BCE and later); A. Bielman, Femmes en public dans le monde hellénistique
(Lausanne 2002).
32 For portraits of the Hekatomnids of Caria see J. Ma, “The History of

Hellenistic Honorific Statues,” in P. Martzavou and N. Papazarkadas (eds.),

Epigraphical Approaches to the Post-Classical Polis, Fourth Century B.C. to Second
Century A.D. (Oxford 2013) 165–181: “the statues for Artemisia and Ada,
female members of the dynasty, are the first securely known public honorific
statues for women” (168).
33 For early honorific portraits of women generally see C. M. Keesling,

“Syeris, Diakonos of the Priestess Lysimache on the Athenian Acropolis (IG

II2 3464),” Hesperia 81 (2012) 490–498. G. J. Oliver’s catalogue (“Space and
the Visualization of Power in the Greek Polis: The Award of Portrait
Statues in Decrees from Athens,” in Early Hellenistic Portraiture 184–188) of
86 Athenian honorific portraits before the Roman imperial period includes
only four of women: one of these (no. S56 = Agora XVI 277 and XXXI 35,
ca. 180 BCE) is a painted portrait, and only one (no. S49, the portrait of
Glaukon included in the Appendix below) dates before ca. 200.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
850 GREEK STATUE TERMS REVISITED

more common, it is difficult to find any epigraphic reference to

a female portrait as an andrias. In the Testament of Epikteta
from Thera (IG XII.3 330), inscribed on the base for portraits
of Epikteta and her two sons ca. 210–195, it is doubtful that
references in the inscription to andriantes include Epikteta’s own
portrait; instead, the statues of Epikteta and her sons are called
agalmata because after Epikteta’s death all three were recipients
of hero cult.34 In the dossier of decrees in honor of the local
benefactor Archippe from Kyme in Aeolis soon after 130
(I.Kyme 13), a group of statues representing Archippe’s deceased
father Dikaiogenes together with Archippe and a personifica-
tion of the Demos crowning her are referred to as andriantes, but
an individual portrait of Archippe is referred to only as an
eikon.35 The earliest honorific text to call an individual female
portrait statue an andrias may in fact be the one for Nikassa, a
priestess of Athena Lindia on Rhodes, dated to 10 CE (I.Lindos
II 392.7).36
Inscribed temple inventories support the notion that, though
in the Archaic and Classical periods andrias meant “statue,” it
was avoided in reference to female portraits, and female figures

34 See A. Wittenburg, Il testamento di Epikteta (Trieste 1990). The andriantes

in lines 12 and 15 are those of Epikteta’s deceased husband, Phoinix, and
son Kratesilochos (Wittenburg 144–147). When her other son, Andragoras,
died two years later, Epikteta set up an andrias of him as well (line 21). Agal-
mata of the whole family in the Mouseion: lines 44–45 and 274–276.
35 The first decree (col. i.1–20), after voting a bronze eikon of Archippe

crowned by Demos and a bronze eikon of Dikaiogenes, calls the three statues
together andriantes (15). The sixth decree inscribed on the same stone (vi.17,
29–30, 39) mentions the subsequent award on a gilded eikon to Archippe.
For discussion of the honors for Archippe see van Bremen, The Limits of Par-
ticipation 13–18. In a decree of Kyzikos in Mysia from the late first century
BCE (CIG 3657), Kleidike is awarded a bronze eikon that will stand beside
the andrias of her brother Dionysios in the men’s agora of the city (van
Bremen, The Limits of Participation 171–172 and 187).
36 Cf. IG V.2 436, a second-century BCE decree of Megalopolis in honor

of a woman named Xenokrate, where the restoration (line 11) of andrias

rather than eikon in a very fragmentary text seems doubtful.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 851

in general, until late in the Hellenistic period. As we have seen,

inventories of the temples on the Athenian Acropolis include
female statues and statuettes called korai; they also include male
figures called andriantes. In the Acropolis inventories of 369/8
and 368/7, for example, the helmets (κύναι) and helmet crests
(λόφοι) detached from andriantes should come from bronze male
portraits; other andriantes on the Acropolis, described as support
figures attached to a lampstand or holding a water vessel, are
similar to male figures mentioned by Pausanias and attested by
extant Hellenistic and Roman bronzes.37 In an inventory of
bronze statues on the Acropolis from the Lycurgan period (IG
II2 1498–1501), a distinction is made between bearded (γενει-
ῶν) and beardless (ἀγένειος) andriantes.38 On Delos, temple in-
ventories began to be inscribed on stone in the fourth century
when the island was controlled by an Athenian-dominated
amphictyony and continued through the period of Delian in-
dependence (from 314 to 167), and on into the period of direct
Athenian control (after 167); inventories of the contents of
upwards of twenty different buildings survive.39 As in the

37 Helmets from andriantes: IG II2 1424a.281; helmet crests: 1424a.284.

Bronze lampstand and the andriantiskos from it: 1424a.271; gold unweighed
aporrhanterion which the andrias holds: 1424a.362; gold perirrhanterion
which the andrias holds: Harris, The Treasures VI.29. The mid-fifth century
sculptor Lykios, son of Myron, made statues of a youth holding a perirrhan-
terion on the Athenian Acropolis (Paus. 1.23.8) and of a boy holding an
incense burner (Plin. HN 34.79). For adolescent male support figures in Hel-
lenistic and Roman bronze sculpture see e.g. B. S. Ridgway, Roman Copies of
Greek Sculpture (Ann Arbor 1984) 83–84.
38 See D. Harris, “Bronze Statues on the Athenian Acropolis: The Evi-

dence of a Lycurgan Inventory,” AJA 96 (1992) 637–652, and D. Harris-

Cline, “Broken Statues, Shattered Illusions: Mimesis and Bronze Body Parts
on the Akropolis,” in C. C. Mattusch et al. (eds.), From the Parts to the Whole I
(JRA Suppl. 39 [2000]) 135–141. Cf. E. Kosmetatou, “Reassessing IG II2
1498–1501A: Kathairesis or Eksetasmos?” Tyche 18 (2003) 40–41.
39 For the Delian temple inventories see especially Hamilton, Treasure

Map, and E. Kosmetatou, “Ζῴδια in the Delian Inventory Lists,” Mnemosyne

57 (2004) 481–484.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
852 GREEK STATUE TERMS REVISITED

Athenian inventories of the fifth and fourth centuries, in the

Delian inventories before 167 there is no reason to doubt that
any of the individual statues or statuettes called ἀνδριάς, ἀν-
δριαντίσκος, or ἀνδριαντίδιον was a male figure.40 On Delos,
though andrias and its diminutives appear frequently through-
out the inventories, divine images were typically called either
agalma or by the god’s name, as in the post-167 inventories of
the Serapieion that include a Zeus, an Eros, an Aphroditiskos,
an Apolloniskos, a Palladion, and a Paniskos, among other
statues.41 The inventories frequently employ a cluster of related
terms—ζῷον, ζῴδιον, and ζῳδάριον—to refer to human
figures generically, without implying anything about their
identity, gender, or material.42
Until ca. 200 BCE, the Delian inventories in their usage of
andrias are consistent with Greek authors and with other in-
scriptions: an andrias was a male figure, either a portrait or the

40 Cf. the frequent collective references to fragments of gold and silver

fallen from andriantes in inventories of the Apollo temple beginning with

I.Délos 379. Hamilton’s translations (Treasure Map 349–360) obscure the use
of andrias-terms in the inventories. He translates eikon and zoidarion as
“figure,” agalma as “statue,” and andriantiskos as “statuette.” Andrias and its
diminutives, not included in Hamilton’s glossary, are generally translated as
either “statue” or “figure.” For the use of diminutives as an example of
linguistic creativity in the Delian inventories see C. Prêtre, “Un collier
délien,” REA 99 (1997) 371–376, and “Imitation et miniature. Etude de
quelques suffixes dans le vocabulaire délien de la parure,” BCH 121 (1997)
673–680.
41 See Hamilton, Treasure Map 223–240 (Serapieion Treasure D): I.Délos

1416.A.i.1 ff., 1417.A.ii.141 ff. (155 BCE), and 1442.A.1 ff. (146 BCE).
42 See for example the ζῳδάρια παιδικά (figures of children) in the

Serapieion D inventories. ζῷον and ζῴδιον in Herodotus refer to figures

carved in relief, painted, or used as textile decoration (e.g. 1.70.1, 2.148.7,
3.47.2, 3.88.3). In later periods, however, zoia were clearly freestanding
statues. Aristotle’s will (Diog. Laert. 5.15–16) included provisions for the
dedication both of portraits (eikones) of family members and of stone zoia four
cubits tall of Zeus Soter and Athena Soteira. In the Testament of Epikteta
the zoia of the Muses interpreted by Wittenburg as reliefs should also be
statues (Il testamento di Epikteta 144–147).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 853

representation of a god or a hero. After ca. 200 on Delos, we

see andrias used for both male and female figures, and this
change can be related to important developments in Greek
portraiture. In one notable instance, the same pair of silver
statuettes of Apollo and Artemis dedicated by a woman named
Kleino, stored in the temple of Apollo, were called ζῴδια from
279 through 224 BCE, but ἀνδριαντίδια from 195 onward.43
In the inventories from the period of Athenian domination
after 167, we find for the first time the collocation ἀνδριὰς
γυναικεῖος for a female figure, which confirms both the nor-
mative assumption that an andrias was male and a new desire at
this time to apply the term to female statues.44
This broadening of the usage of andrias was not merely a
response to the increasing popularity of female portraiture after
ca. 200; rather, it seems to result from a desire to make a clear
distinction between portrait statues and portrait paintings.
Though painted portraits as votives are attested as early as the
fifth century, when the sons of Themistokles dedicated a
painting (γραφή) depicting their father inside the cella of the

43 Kleino’s Apollo and Artemis appear in 12 inventories, from 279 BCE

(IG XI.2 161.B.1) through 145 (I.Délos 1449.e). Kosmetatou (Mnemosyne 57

[2004] 481) noticed the change in terminology, but concluded that the
terms zoidia and andriantidia were interchangeable. This seems to be the case
only in the second-century inventories, where both ζω-terms and ἀνδρια-
terms were used for female figures. For example, in an inventory of the
temple of Agathe Tyche from 146 (I.Délos 1442.B.35) we find in line 44 a
ζωιδάριον Ἀφροδίτην λίθινον (stone Aphrodite figure) paired with an
ἀνδριαντίδιον χαλκοῦν Ἀγαθῆς Τύχης (small bronze andrias of Agathe
Tyche).
44 In the Thesmophorion in 155 (I.Délos 1417.A.i.49) appear an ἀνδριαν-

τίδιον γυναικεῖον (59) and an ἀνδριὰς γυναικεῖος (114–115). In the same

year the inventory of the Letoion (1417.A.i.100) includes an ἀνδριὰς
γυναικεῖος on a base (114–115), and the Aphrodision (1417 A.ii.1) has a
γυναικεῖος ἀνδριάς dedicated by Stesileos (119). The 155 BCE inventory
associated with the gymnasium (1417.A.i.118) includes an ἀνδριὰς τέλειος
γυµνός (full-sized nude statue, 123), two ἀνδριαντίδια (128, 132), and an
ἀνδριὰς γυναικεῖος with a cup in her hand (141).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
854 GREEK STATUE TERMS REVISITED

Parthenon (Paus. 1.1.2), officially awarded painted portraits are

first attested in the third century, becoming far more common
in the second.45 As painted portraits were being awarded as a
lesser honor than a portrait statue, honorific decrees began to
refer to a portrait statue as an andrias and a painted portrait as
an εἰκῶν γραπτή. The earliest examples of honorific decrees
awarding both a statue and a painting, which date to the sec-
ond century BCE, contradict Piejko’s generalization about the
use of andrias in pre-50 BCE decrees. For example, in a decree
of the Poseidoniasts of Berytus on Delos soon after 153/2 in
honor of a Roman (I.Délos 1520), the γραπτὴ εἰκῶν (26–27 and
31) clearly refers to a painted portrait awarded in addition to
an andrias (24 and 27), a portrait statue.46 In some inscriptions
of the first century BCE and the first century CE, it also seems
likely that an eikon without further specification, awarded at the
same time as an andrias, should be understood as a portrait
painting.47

45 For painted portraits see M. Nowicka, Le portrait dans la peinture antique

(Warsaw 1993) 121–126. The Delian inventories make a distinction be-

tween πίνακες ἀναθηµατικοί (votive painted plaques) and πίνακες εἰκονικοί
(portrait paintings); the latter are also attested in the Athenian Asklepieion
(S. B. Aleshire, The Athenian Asklepieion. The People, their Dedications, and the In-
ventories [Amsterdam 1989] 148). The earliest Athenian honorific painted
portrait is the “eikon on a pinax, according to custom” awarded by a thiasos
of Carian Zeus to its treasurer, Menis Mnesitheou of Herakleia, in 298/7
(IG II2 1271). See H. Blanck, “Porträt-Gemälde als Ehrendenkmäler,” BJb
168 (1968) 1–12, who however cites the decree of 178/7 for Hermaios
Hermogenou Paionides (IG II2 1327) as the earliest example. Satyra, a
priestess of the Thesmophoroi, was honored by her deme ca. 180 with an
eikon on a pinax (Agora XVI 277 = XXXI 35).
46 Other early examples are CIG 3068.B.28 (Teos) and I.Kourion 34.23

(both mid-second century BCE).

47 Likely examples are IG II2 4193 (first century BCE), IOSPE I2 34 from

Olbia (early first century BCE), TAM V.2 920 from Thyateira (49 BCE?). In
the Lindian statue base of 10 CE for the priestess Nikassa (I. Lindos II 392)
the honors mentioned include an eikon, a gilded eikon, and a bronze andrias:
the former two could be painted portraits, one with a gold background. An
early imperial decree in honor of a female sacred official called a hydrophoros

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 855

From the late Hellenistic period onward, epigraphic usage

diverges somewhat from that of literary texts. In Polybius,
andrias refers to any statue, divine or portrait, male or female,
but Pausanias in the second century CE still preferred to call
portrait statues eikones, not andriantes.48 In hundreds of Roman
imperial inscriptions of various genres from mainland Greece,
the Aegean islands, and Asia Minor, however, andrias should be
translated “portrait statue.” These include, I would argue,
some key epigraphic texts of the first century CE. In a decree of
the gerousia of Cos (IG XII.4 471 = Iscr.Cos ED 230)
authorizing the melting down of andriantes, the 17 male Coan
citizens whose names are listed in the genitive case are best
understood as the subjects of portrait statues to be removed
___
from Didyma (I.Didyma 381) mentions the award of both eikones and andri-
antes; in a second, roughly contemporary hydrophoros decree of 17/6 BCE
(I.Didyma 378) an εἰκῶν χρυσή awarded by the demos is mentioned: is this a
gilded bronze statue or a painted portrait with a gold background? Cf. T.
Pekáry, “Statuen in kleinasiatischen Inschriften,” in S. Şahin et al. (eds.),
Studien zur Religion und Kultur Kleinasiens, Festschrift F. K. Dörner II (Leiden 1978)
730, who cautioned that “eine immer und überall gültige Definition von
εἰκών und der angefügten Adjektive ist wohl nicht möglich. Das Wort be-
deutet grundsächlich Porträt, bildnishafte Darstellung, und ohne Adjectiv
wohl in den meisten Fällen, jedoch nicht ausschliesslich, eine Statue.”
48 In at least two cases, Polybius seems to use andrias for portrait statues as

opposed to divine images: 21.30.9 (agalmata, andriantes, and graphai taken as

plunder from Ambracia) and 32.15.3 (both andriantes and stone agalmata). Cf.
4.78.3, where he refers to a bronze andrias of Athena. For statue terms in
Pausanias see V. Pirenne-Delforge, “Image des dieux et rituel dans le dis-
cours de Pausanias. De l’ ‘axiologie’ à la théologie,” MEFRA 116 (2004)
811–825: “Les termes eikôn et andrias sont inadaptés pour décrire la statue
d’un dieu parce que leur signification intègre l’idée de portrait” (816). The
eikonion of Themistocles inside the temple of Artemis Aristoboule in Athens,
described by Plutarch (Them. 22.1–2), could be either a statuette or a small
painted portrait (P. Amandry, “Thémistocle à Mélitè,” in Χαριστήριον εις
Αναστάσιον Κ. Ορλάνδον IV [Athens 1967–1968] 276–277; cf. Krumeich,
Bildnisse 78–79). This diminutive is used for a cypress wood statuette in
I.Délos 1442.A.i.56 (146/5–145/4 BCE), but all literary attestations date to
the Roman imperial period.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
856 GREEK STATUE TERMS REVISITED

from display and their bronze melted down to mitigate a

financial crisis.49 Lindos II 419 (= LSCG Suppl. 90), a long and
difficult decree of 22 CE, proposes auctioning off the right to
put new inscriptions on the bases of andriantes on the acropolis
of Lindos to generate revenue to pay for sacrifices and festi-
vals.50 Though andriantes here has most often been taken as a
general reference to statues of all kinds, it is doubtful how many
statues other than portraits would have been standing on the
Acropolis and its approaches at this late date (30–44):
ἐπειδὴ δὲ καὶ ἀνδριάντες
[τ]ινές ἐντι ἐν τᾷ ἀναβ[ά]σει καὶ αὐτᾷ τᾷ ἄκρᾳ ἀνεπίγραφοι καὶ
ἄσαµοι, συνφέρον δέ [ἐ]στι καὶ τούτους ἤµειν ἐπισάµους ἐπιγρ[α]-
[φ]ὰν ἔχοντας ὅτι θεο<ῖ>ς ἀνάκεινται, δεδόχθαι Λινδίοις,
κυ(ρωθέντος) τοῦδε
[τ]οῦ ψα(φίσµατος)· τοὶ αὐτοὶ ἐπιστάται µ[ισθω]σάντω ἑκάστου
ἀνδριάντος τὰν
[ἐ]πιγραφάν, διαχειρο[τονησ]άντων Λινδίων εἰ δεῖ τοῦ εὑρίσ-
κοντος κατακυροῦ[ν ἢ µ]ή, καὶ [εἴ κ]α [δ]όξῃ τοῦ εὑρίσκοντος κα-
[τ]ακυροῦν τὸ πεσὸν ἀργύριον· [ἀ]πὸ τού[τ]ων καταβαλόµε-
[ν]οι λ[όγ]ον π[ό]σου ἑ[κ]ά[σ]το[υ ἁ] ἐπιγραφ[ὰ ἀπε]δόθ[η]
παραδόντω ἰερὸν
[ἤ]µ[ειν εἰς] πα[ρ]ακα[τ]α[θ]ήκαν τᾶς Ἀ[θ]άνας τ[ᾶ]ς Λινδία̣ς καὶ
τ[οῦ]
[Διὸς τοῦ Πολιέ]ω̣ς·̣ [τοὶ δὲ] ὠνησά[µ]ε[ν]οι τὰς ἐπιγραφὰς µὴ
[ἐχόντων ἐξουσίαν ἀπ]ε[νε]νκεῖ[ν] ἐκ τᾶς ἄκρας ἀνδριάν[τας]
[τρόπῳ µηδ]ενὶ µηδὲ παρευρέσει µηδεµιᾷ ἢ ἔνοχοι ἐόντ[ω]

49 See the remarks of Chr. Habicht, “Neue Inschriften aus Kos,” ZPE

112 (1996) 86. Cf. IG XII.4 353 (= Iscr.Cos ED 257: first or second century
CE), which prohibits the dedication of an eikon, agalma, or andrias on any
exedra in the gymnasium of Cos, where eikon may mean a portrait painted
on a pinax.
50 The most thorough discussion of the relevant section is M. Kajava,

“Inscriptions at Auction,” Arctos 37 (2003) 69–80; cf. H. Blanck, Wieder-

verwendung alter Statuen als Ehrendenkmäler bei Griechen und Römern (Rome 1969)
101–103, and L. Migeotte, Les souscriptions publiques dans les cités grecques
(Geneva/Quebec 1992) 121–126, no. 41.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 857

[ἀσεβεί]ᾳ· πο̣ιησάµενοι δὲ τὰν αἴτησιν ἐχόντων ἐξουσ[ίαν]

[ἀπενενκ]εῖν ἅ κα συνχωρήσωσι διὰ τᾶς αἰτήσιος Λίν[δ]ιοι·
And since there are some portrait statues [andriantes] along the
ascent and on the top itself [of the acropolis], which are without
inscription [ἀνεπίγραφοι] and undistinguished [ἄσαµοι], and it
is expedient that these too shall be distinguished [ἐπισάµους],
bearing inscriptions (saying) that they are dedicated to gods, it
was voted by the Lindians: when this decree has been sanc-
tioned, the same epistatai shall lease out the inscription of each
portrait statue [andrias], the Lindians deciding by vote whether
the winning bid should be confirmed or not, and if it will be
decided that the winning bid should be confirmed, they [the
epistatai], after having made an account of the rate for which the
inscription of each portrait statue [andrias] has been ceded, shall
hand over the money accrued from these to be sacred to the
fund of Athana Lindia and Zeus Polieus. Those who have pur-
chased the inscriptions shall not have permission in any wise nor
under any pretext to remove portrait statues [andriantes] from the
top; otherwise they shall be liable to be accused of impiety. But if
they make a request, they shall have permission to change (por-
trait statues) [ἀπενενκεῖν, largely restored]51 according to what
the Lindians agree on account of the request (transl. Kajava,
with “statue” changed to “portrait statue”).
Here the uninscribed and undistinguished andriantes that are to
receive new inscriptions dedicating them to the gods should be
interpreted as portraits whose subjects are no longer iden-
tifiable.52 Though the Lindian inscription specifies only that the

51 Alternatively, L. Robert, Hellenica 2 (1946) 110–111, restored

[µετενενκ]εῖν, which would imply the exchange of the statue standing on

the auctioned base for another statue.
52 Cf. Kajava, Arctos 37 (2003) 74 (“In fact, considering that it was normal

for cultic and votive statues of deities to be without inscription, one may
assume that the Lindian andriantes also included some belonging to this
category”); J. Mylonopoulos, “Odysseus with a Trident? The Use of Attributes
in Ancient Greek Imagery,” in Divine Images and Human Imaginations in Ancient
Greece and Rome (Leiden 2010) 171–174, who takes the asamoi andriantes to be
divine images unrecognizable owing to the absence or loss of attributes.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
858 GREEK STATUE TERMS REVISITED

andriantes are to be inscribed with dedications to the gods, the

terms used to describe the statues eligible for selection, ἀνεπί-
γραφοι and ἄσαµοι, anticipate the language used by Dio Chry-
sostom (31.72–74) later in the first century to describe honorific
portrait statues in Rhodes town chosen for reinscription with
the names of new portrait subjects. The same practice was
likely envisioned at Lindos.
Finally, the related words ἀνδριαντοποιός and ἀνδριαντο-
ποιϊκά take on new significance in light of the recent pub-
lication of Poseidippos’ poetry book of ca. 280 BCE, preserved
on papyrus: the epigrams of the Andriantopoiïka section concern
bronze statues, mostly male, none of them female portraits.53
When Pindar in the first half of the fifth century compared
himself to an andriantopoios who makes agalmata (Nem. 5.1–6), the
natural inference is that he meant the sculptors of athletic
victor portraits, but the reference is not yet specifically to
bronze statuary, for he elsewhere (Pyth. 5.40) called a wooden
statue (of Apollo?) at Delphi an andrias.54 Later in the fifth
century, Gorgias’ Encomium of Helen (18) included the antithesis
ἡ δὲ τῶν ἀνδριάντων ποίησις καὶ ἡ τῶν ἀγαλµάτων ἐργασία
νόσον ἡδεῖαν παρέσχετο τοῖς ὄµασσιν (“the fashioning of
andriantes and the working of agalmata provides a pleasant

53 On the Andriantopoiïka see especially A. Stewart, “Posidippus and the

Truth in Sculpture,” in K. Gutzwiller (ed.), The New Posidippus. A Hellenistic

Poetry Book (Oxford 2005) 183–205, and E. Prioux, Regards alexandrins, histoire
et théorie des arts dans l’épigramme hellénistique (Louvain/Paris 2007) 109–113.
Cf. K. Gutzwiller, “Posidippus on Statuary,” in G. Bastianini and A. Casa-
nova (eds.), Il papiro di Posidippo, un anno dopo (Florence 2002) 44: “Even the
title ἀνδριαντοποιικά, though formed from one technical term for sculpting,
conveys through its living etymology an emphasis on the human elements in
the statues presented.” The andriantes themselves seem all to have been
made of bronze, a material correlation lacking in the Hellenistic temple in-
ventories from Delos.
54 For the relationship between Pindar’s epinician odes and contem-

porary Greek sculpture see D. Fearn, “Kleos Versus Stone? Lyric Poetry and
Contexts for Memorialization,” in P. Liddel and P. Low (eds.), Inscriptions
and their Uses in Greek and Latin Literature (Oxford 2013) 231–253.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 859

disease for the eyes”). This seems to constitute a case in which

“matter which does not contain any inherent contrast is split in
half or duplicated for the sole purpose of producing a pair.”55
Though Xenophon employed the term andriantopoios in refer-
ence to bronze statues and their sculptors, Plato clearly used
andriantopoios and agalmatopoios interchangeably: the fifth-century
sculptor Pheidias is called either an andriantopoios (Meno 91D) or
an agalmatopoios (Prt. 311E); Pheidias and his contemporary
Polykleitos of Argos are referred to in tandem as agalmatopoioi
(Prt. 311C).56 The earliest instance in Greek literature in which
an andriantopoios as a sculptor of human figures in bronze is
clearly distinguished from a maker of divine images in stone
occurs in Aristotle’s Nicomachean Ethics (1141a11): τὴν δὲ σοφίαν
ἔν τε ταῖς τέχναις τοῖς ἀκριβεστάτοις τὰς τέχνας ἀποδίδοµεν,
οἷον Φειδίαν λιθουργὸν σοφὸν καὶ Πολύκλειτον ἀνδριαντο-
ποιόν (“wisdom in the arts we ascribe to those who practice the
arts most perfectly, for example Pheidias the stone sculptor and
Polykleitos the andriantopoios”). Lists of Greek sculptors that
divide them into agalmatopoioi, sculptors specializing in divine
images, and andriantopoioi, bronze portrait sculptors, seem to
originate no earlier than the second century BCE, the same
55 D. M. MacDowell, Gorgias, Encomium of Helen (London 1982) 19.
56 See especially Xen. Mem. 1.4.2–4 (Polykleitos excels in sophia among
andriantopoioi) and 3.10.6–8 (Kleiton the andriantopoios makes statues of run-
ners, wrestlers, boxers, and pancratiasts). For the literary and epigraphical
sources on Pheidias and Polykleitos see S. Kansteiner et al. (eds.), Der neue
Overbeck: Die antiken Schriftquellen zu den bildenden Künsten der Griechen (Berlin
2014) nos. 841–1075 (no. 1047 = Arist. Eth.Nic. 1141a9–12, see below) and
1205–1294. Agalmatopoioi appear in the Parthenon and Erechtheion building
accounts (IG I3 445–449 and 476); in the list of individuals and their occu-
pations in the heroes of Phyle decree of 401/0 (IG II2 10); and in inscribed
accounts of the mid-fourth century (IG II2 216, 217, 1508). In ca. 336–330 a
Boiotian andriantopoios was hired by the Athenian state to repair a fifth-cen-
tury (bronze?) agalma of Athena Nike (IG II3 444, discussed by S. Lambert,
“Connecting with the Past in Lykourgan Athens: An Epigraphical Per-
spective,” in L. Foxhall et al. [eds.], Intentional History: Spinning Time in Ancient
Greece [Stuttgart 2010] 226–228).

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
860 GREEK STATUE TERMS REVISITED

time that andrias began to be used in Greek inscriptions as a

standard term for a portrait statue, male or female.57
The Greeks did have a word for statue: ἀνδριάς. In origin an
andrias’ male gender mattered more than its material or whom
it represented. In the Archaic and Classical periods at least,
andrias was simply less common in literature and inscriptions
than other terms, agalma and eikon, with more specific deno-
tations. In these periods female portraits, uncommon before ca.
200 BCE, were not referred to as andriantes. Though a full study
of the Delian inventories as a source for Greek sculpture re-
mains to be written, their language reflects a shift in the mean-
ing of andrias over time that comes through most clearly in late
Hellenistic and Roman imperial honorific decrees, where we
see a new concern to make a distinction between portrait paint-
ings and portrait statues. The painted—and sometimes gilded
—eikones that proliferate in honorific texts of these periods set
the stage for the icon paintings of Byzantine Christianity.58
APPENDIX: Pre-200 BCE honorific decrees
mentioning portrait statues of women
1. RO 56 = I.Erythrai 8 (360s or mid-350s BCE): Mausolus honored
with a bronze eikon in the agora of Erythrai and Artemisia with a
stone eikon in the sanctuary/temple of Athena (Athenaion).
2. IG XI.4 514; H. Kotsidu, Τιµὴ καὶ δόξα: Ehrungen für hellenistische
Herrscher (Berlin 2000), no. 123 (soon after 300): agalmata of Askle-
pios and of Stratonike, daughter of Demetrios I, on Delos.
3. I.Didyma 480 = SEG XXXIV 1075; Kotsidu no. 269; Bielman,
Femmes en public 64–68, no. 10 (299/8): Apame (first wife of Seleu-

57 Such lists appear in the Laterculi Alexandrini, P. Berol. inv. 13044r col.

vii.6–9 (H. Diels, AbhBerl 1904, 7), a papyrus of the second or first century
BCE, and in P.Oxy. X 1241 (second century CE). For arguments in favor of
a post-Ptolemaic date for the latter see J. Murray, “Burned after Reading:
The So-Called List of Alexandrian Librarians in P. Oxy. X 1241,” Aitia 2
(2012) (http:// aitia.revues.org/544).
58 S. Sande, “The Icon and its Origin in Graeco-Roman Portraiture,” in

L. Rydén and J. O. Rosenqvist (eds.), Aspects of Late Antiquity and Early Byzan-
tium 75–84, esp. 77–80.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861
 CATHERINE M. KEESLING 861

kos I) honored by Miletus with an eikon.

4. SEG XXXVI 1218; Kotsidu no. 293 (243/2): eikones of Ptolemy
III and Berenike II in the Letoön at Xanthos.
5. I.Oropos 175 = IG VII 297 = OGIS 81; Kotsidu no. 82 (215–204):
decree ending with reference to eikones of Ptolemy IV and Ar-
sinoe, inscribed on the base for their portrait statues (I.Oropos
427).
6. IG II2 1314 (213/2): eikon of Glaukon, chosen by lot to be annual
priestess, in the temple (naos), awarded by orgeones.
7. SEG XLI 1003.8, 32, 45; Kotsidu no. 239 (204/3): marble
agalmata for Antiochos III and Laodike III in Teos; also an agalma
in bronze and a gilded eikon for Antiochos.59

July, 2017 Department of Classics

Georgetown University
keeslinc@georgetown.edu

59I would like to offer my warmest thanks to J. W. Day, who made com-
ments on a draft of this paper, to an anonymous reader, and to the editors
of GRBS. Any mistakes or misunderstandings that remain are my own.

—————
Greek, Roman, and Byzantine Studies 57 (2017) 837–861