• r2-7-1
    "Cybele shrine," front. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-20
    "Cybele shrine,” top right. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-21
    "Cybele shrine," in situ with David Mitten. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-23
    "Cybele shrine," top view. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-27
    "Cybele shrine,” three quarter view of monument, front. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-28
    "Cybele shrine," three quarter view of monument, back and right. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-29
    "Cybele shrine," detail of torso showing broken attribute (lion?) outlined. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-30
    "Cybele shrine," right side. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-40
    "Cybele shrine," plan: actual state and restoration, drawing. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-45
    "Cybele shrine," restored plan as actual building. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-50
    "Cybele shrine," left side, drawing. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-60
    "Cybele shrine," right side, drawing. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-70
    "Cybele shrine," back, drawing. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-80
    "Cybele shrine," back. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-100
    "Cybele shrine," panel A, detail of standing human figures on side of shrine. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-110
    "Cybele shrine," panel B, detail of kore. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-120
    "Cybele shrine," panel C, detail of kneeling boy. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-130
    "Cybele shrine,” panel D, detail of kneeling boy. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-140
    "Cybele shrine,” panel E, side. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-170
    "Cybele shrine," panel H, detail of striding female figure. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-180
    "Cybele shrine," panel I, detail of striding female figure. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-190
    "Cybele shrine," panel J, detail of striding female figure. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-200
    "Cybele shrine," panel K, detail of lion. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-210
    "Cybele shrine," panel L, detail of lion. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-220
    "Cybele shrine," panel M, side. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-230
    "Cybele shrine," panel N, detail of boar and lion. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-240
    "Cybele shrine," panel O, detail of warrior fighting lion. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-250
    "Cybele shrine," panel P, detail of charioteer. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-260
    "Cybele shrine," panel Q, detail of horse. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-270
    “Cybele shrine," panel R, detail of man. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-320
    "Cybele shrine," preserved capital fragment. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)
  • r2-7-500
    "Cybele shrine," in situ. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

Marble Naiskos of Cybele

540-530 BC, Late Lydian (Persian)
Manisa, Archaeological and Ethnographic Museum, 4029
Museum Inventory No.
Sardis or Museum Inv. No.
Marble, Stone
Object Type
Sculpture Type
Human Figure, Naiskos, Votive Relief, Draped Woman
Syn 63
Syn MH Spolia
B-Grid Coordinates
E73.5 / N17.2 - N17.7 *97.25 - 96.75
Found Aug. 9, 1963, by D. G. Mitten ''among the tumbled masonry of a shattered pier" (N4) of Syn MH where it had been reused, E73.5/N17.2-17.7 *97.26-96.75 (Figs. 21-22, in situ). The findspot is indicated on the plan, Fig. 5; it should be noted, however, that the system of numbering the Syn piers has changed to count from the W instead of E, so that former pier N5 is now N4.

Monument in the form of a shrine decorated with reliefs, with goddess standing in front, henceforth referred to as “Cybele shrine.”

A frontal female figure wearing a girt chiton stands between two snakes (or plants?) in the entrance of a shrine. The shrine had three-quarter columns at the four corners and engaged half columns at the center of the back and sides, a total of seven columns. As the preserved capital fragment indicates (Figs. 20, 26, 30), the shrine was of Ionic or Proto-Ionic order. Wide bands of incised and painted meanders separate each side and the back into three horizontal zones, subdivided into six panels by the central column, a total of eighteen reliefs. The representations show maidens in procession, komasts, and dancers on the left side; maidens in procession, and lions on the right side; and six mythological scenes on the back.

Anathyrosis on the bottom shows that the monument was intended to be set on a base or pedestal of stone. The smooth outer strip is done with large oblique strokes of a flat chisel or drove, similar to those on the masonry of the Alyattes chamber (BASOR 170, 55). The strip is 0.10 in W. at the back, 0.05 on the left, and only 0.025-0.03 at the front.


The low platform is envisaged as a stylobate. The shrine had two piers (antae) in front, with Ionic three-quarter columns attached. The pier faces and columns are lost except for the lower part of the right column base. Taken literally, the plan would show an oblong, more wide than deep (Fig. 25). Our tentative reconstruction (Fig. 24) presents a nearly square small temple with all columns attached or engaged and the image base occupying most of the interior.

The corner columns stand on high plinths, slightly rounded in front and tapering upward (H. 0.10; W. at bottom 0.115, at top 0.10; Figs. 20, 24, 25, 28-32, 37, 43, 49). A plinth is also indicated for the central column in back. The bases are plain tori, with the largest diameter in the lower half (H. 0.035; lower diam. 0.105, 0.12 for central columns; upper diam. 0.10; max. diam. for corner columns 0.12). The columns have fluted shafts with 9 fillets (ridges) for the half and 13-14 fillets for the three-quarter columns, hence a total of 18 for a column in the round. The flat fillets are 5-6 mm., the grooved flutes 4 mm. in W. Fillets and flutes run straight against the base and do not end in ovals. There is no intervening molding (apophyge) at the foot of the shaft. The column outlines and ridges were drawn with a ruler. (Max. H. of unrestored parts: left side corner 0.445, center 0.46; back, right [from front] corner 0.485, back center 0.48. Total shaft H. of restored column on right side 0.57; lower shaft diam., corner and central 0.10, 0.062 at break. Marked taper on left side 0.038 over a H. of 0.44.)

Only one incomplete fragment of a capital (Figs. 20, 26, 30) survives (P.H. 0.15; radius ca. 0.037). The echinus at the neck forms a counterpart of the base and looks quite like a Doric capital. It is surmounted by an unmistakable Ionic spiral band with a rather large and steep center framed by raised edges and linking what must have been two volutes. It looks as if there were a small palmette in the reentrant of the junction of the right volute with the echinus. The volutes were topped by a large, undecorated, swelling abacus.

The division of the external walls is by bands 0.025-0.027 in H. and 0.07-0.08 in L. on the sides, 0.10-0.15 on the back. The H. of the lower and the middle panels is 0.15; none of the top panels is complete.

The shrine presents unusual and experimental traits. The most striking of these is a plan with engaged Ionic columns, of which apparently no actual archaic example has come down to us. An early sarcophagus in Samos seems to imitate a freestanding Ionic colonnade rather than a building with engaged columns (cf. I. Kleeman in Matz Festschrift, 48, 52f., pls. 12, 15). The placing of a central column on axis in the back occurred at the colossal temples at Samos (Rhoikos) and Ephesus; the Sardis shrine may reflect a third example of the same era (Samos, 570 B.C.?; Ephesus, 560 B.C.?, cf. Dinsmoor, Architecture of Ancient Greece 125, n.1).

Though simplified, the columns are evidence of an early formative phase of the Ionic style. The tall plinths, the heavy plain cushion bases, with the greatest diameter in the lower half, and the dense flutes which run straight down against the base are all characteristic of the pre-Croesan and Croesan eras (cf. for flutes with straight ends, the Oikos of the Naxians, Delos, ca. 600 B.C.; the Naxian sphinx at Delphi, ca. 580-570 B.C.; and the reconstructed columns of Smyrna, ca. 610-600 B.C.). The very large echinus is unparallelled but is nearest to some column capitals of the Croesan Artemisium at Ephesus. If it was painted with egg and dart, it would look much closer to standard Ionic.

Literature: for plinths and bases, cf. Wesenberg, Kapitelle und Basen, 111f., 114ff., figs. 169, 176, 233-235; Nylander, Ionians in Pasargadae, 103ff., figs. 35f., Building P; and 6 (Figs. 16-19). Gruben, Das archaische Didymaion, fig. 38, comparative chart illustrates major archaic Ionic columns, and text 115ff., with literature, and fig. 19, Didymaion capital; Smyrna, Akurgal, Civilizations, 119, fig. 41.

If the capital fragment is correctly joined, the proportion of total H. to lower diam. is ca. 5.5:1, stocky for early Ionic. The most conservative proportioning assumes 8:1 for Ephesus; Gruben calculates 12:1 (Didymaion, 153ff.). The taper, 0.10-0.062 =1:0.6 (38%), is probably exaggerated. It exceeds even the ratio of 1:0.7 for the Oikos of the Naxians and 1:0.76 for Croesan Ephesus (ibid., 148, chart). The artist was obviously hard put to reproduce on a small scale what seem to have been slender, tapering columns with densely spaced fillets.

This is the first known example of a Greek building with sculptured decoration distributed over the full height of the wall. The principle is known in near-contemporary Babylon as in the Ishtar Gate (602-562 B.C.), but the Babylonian example has a tapestry-like arrangement without architectural registers or panels. The Sardis design looks like a bold attempt to combine the Mesopotamian figurative tradition with colonnaded Greek temple architecture. It bespeaks an experimental pioneering spirit; the adaptation of “Late Hittite” or Assyrian figurative orthostates to the column bases of the Croesan columnae caelatae in Ephesus was a similar bold invention. Presumably, the sculptor was copying an Alyattan or Croesan building of ca. 570-560 B.C. (For the beginning of frieze-like architectural decoration, cf. R. Demangel, La frise ionique, 15-113, 119-130; Dinsmoor, Architecture of Ancient Greece, 64.)

A similar shrine with an image but with Aeolic columns and a flat roof is shown on a black-figured amphora, British Museum B 49, of ca. 530 B.C. (Beazley, Attic Black-Figure Vase-Painters, 326, 715, without number, “Cybele?”; CVA British Museum III, pl. 35:2a; K. Schefold, Die Griechen 52, 338, fig. 5. Wesenberg, Kapitelle und Basen 84, fig. 175. Ph. Oliver-Smith in Lehmann Essays, 235, fig. 4. Nilsson, Greek Popular Religion, 91f., fig. 35, detail, interpreted as a Magna Mater).

The Image

Broken are the head, both arms from below the shoulders, the feet, and the front part of the fold on the proper right (Figs. 20-22, 27, 28). “In front, in high relief is shown the figure of woman standing with feet close together, the right arm along the side to grasp a fold of the paryphe of the chiton [which swings to her right], the left arm bent, with the forearm brought to the front of the body and the hand holding some offering [hand and arm broken off]. She wears a chiton pulled over a belt to form a pouch (kolpos) on each side. An epiblema [veil worn over the head] descends shawl-like from both shoulders” (Richter, Korai, no. 164, 92f.). The upper chiton has fine, linear, wavy folds; the lower part has rope-like folds meeting in the center. The folds of the paryphe (at her left foot) end in a “swallow-tail” pattern. She wears a thick necklace with some cross-rings indicated.

Two vertical, wavy forms rise on either side of the image, thickening upward. These are probably snakes rather than trees; compare the terracotta plaque from the Agora, Athens (D. Burr, Hesperia, 604, 608f., figs. 72, 73; Matz, Geschichte der griechischen Kunst I, 482ff., pl. 283a), restored as a goddess with raised arms, flanked by two snakes and interpreted by Burr as one of the Semnai (Erinys).

There has been argument among Sardis excavators about the broken-off attribute on her chest. The general posture is similar to that of the Cheramyes Aphrodite who held a hare (Richter, Korai no. 56, fig. 189). The statement that the lines on her breast represent the tips of a lion's mane (Hanfmann and Waldbaum, “Kybele and Artemis”, 268) is, upon recheck, mistaken. These are chiton folds radiating from the breast. However, the same re-examination in 1975 showed that a rounded piece at the left edge of the break once belonged to a paw, and that the sharp, horizontal lower break cannot be an indication of drapery. The line of the break suggests that the front paws and belly of a recumbent lion similar in type to Cat. 29 and Cat. 31 (Figs.115, 119) may have been broken off. In some photographs the outline of an animal with its head against the goddess’ right breast is clearly seen (Fig. 28). The type, then, would be an archaic forerunner of the standing Cybele type seen in the classical stele Cat. 20 (Figs. 78, 82). Closest in dress and general appearance is Richter's no. 153, found at the W end of the Heraion of Samos. She also wears the veil, apparently the same type which is well illustrated by the priestesses of the Didymaion (Richter, Korai 90, no. 153, figs. 491ff.; Didyma, ibid., 92, no. 162, figs. 516ff.) and has the same swinging central fold. She was dated by Buschor to 540-530 B.C. (Altsamische Standbilder, 5, 93, fig. 373).For the fine crinkly chiton folds, cf. Richter, Korai, no. 109, fig. 331, Acropolis no. 669, "530"; Tuchelt, Die archaischen Skulpturen 105-107, K 82, pl. 76, Didyma gorgons dated to 530-525 B.C. For the rope-like folds, Croesus column fragment, Pryce, Catalogue of Sculpture in the British Museum, 59, B 121, fig. 63; Richter, Korai, no. 120, figs. 381-384, Acropolis no. 683; Tuchelt, Die archaischen Skulpturen , no. K 29, pl. 31. The kore fragment from the Croesus column, Pryce, Catalogue of Sculpture in the British Museum, no. 57, figs. 58 (photo) and 59 (decoration), shows that the broad paryphe strip was an elaborately decorated hem. It may have been painted on the Sardis goddess. Tuchelt, Die archaischen Skulpturen , 129, no. L 111; 155f., n.102, 185f., assigned to the last quarter of 6th C.

Reliefs, L. Side: Panels

A (fig. 33). Incomplete kore walking to right. W. of panel 0.069; P.H. 0.125; H. to waist 0.092. The panel is broken on the top and in front. Missing are the head, part of the chest, part of the right foot, and the ornamental band. The figure is sharply cut, and the surface of the back part of the body is preserved in nearly the original polished state. She wears a chiton and a short cloak over the shoulders which ends at knee level in a pointed tip. The lowered right arm is bent forward at the elbow then down at the wrist. The fingers are indicated. The right hand grasps a lightly incised vertical fold which extends to the bottom of the garment. There is a suggestion that, as in the panel B kore, the upper left arm was brought forward with part of the cloak and a falling vertical fold. The ankles of both feet are indicated.

B (fig. 34). Almost complete kore walking to right. W. of panel 0.078; H. of figure 0.158, of head 0.03, neck to waist 0.035, waist to foot 0.093. The top of the head is broken off. Though the panel is very worn, the original work with light chisel strokes and abrasive can be seen in a few places. More can be seen with the naked eye than either the photograph or drawing brings out. The head has a large aquiline profile with a big eye and "stepped wig" hair falling on the shoulder. A diadem (mitre?) decorated with a dot rises high over the forehead. A high soft breast and long lower body, as in the Samian Cheramyes group, are indicated. She wears a long (belted?) chiton and a short cloak (epiblema), "Ionic mantle" (Richter), which falls from the shoulder and ends at knee level in a pointed tip. The right arm comes out from under the cloak to grasp two fine vertical folds of chiton. The lower hem of the chiton dips between the feet in a curve. The left arm is bent upward, the palm stretched out horizontally, holding some small object (fruit? flower?). Part of the cloak fell in folds on both sides of the left upper arm.

Published: BASOR 174, fig. 26; Richter, Korai 93, no. 164, fig. 524. For "stepped wig" effect, cf. ibid., nos. 29, 31, 42, 54, figs. 108, 110, 144, 176 (Corinth, dated 625-600; Palma Montechiaro, Sicily, late 7th C.; Berlin goddess, 580-570; Palma Montechiaro II, 600-575 B.C.; Tenea kouros, pl. XId-e, 575-550 B.C.). The dipping curved hem of the chiton occurs on the Amphiaros crater (ibid., pl. Xllc, Corinthian, 570 B.C.); proportions and gestures, too, are similar.

C (fig. 35). Komast (rather than silen) running to right. Panel: H. 0.15, W. 0.07. Broken off are the top of the frame, the front of the panel, the face, and part of the hand and left foot; the relief is very battered. Very long hair with a diadem ribbon extends down the back. A roll of hair and a beard are recognizable (not quite correct in the drawing Fig. 32). The komast holds a cup or kantharos in his raised right hand; the wineskin (askos) on his back would be held in his left. Part of the penis is visible; the rest may have been painted. His right foot and part of the wineskin disappear into the ground and behind the column respectively.

D (fig. 36). Similar but smaller komast in typical "Kneelauf' to right. Panel: H. 0.15, W. 0.07; H. of komast 0.115. The surface is very battered. Over his right shoulder he wears a short cloak which then passes over his right arm and midriff (incorrect in drawing). With his right hand he raises to his lips a goblet, a large open cup, the foot of which appears below his hand. On his shoulder is perched a jumping animal which leans its forepaws on top of his head. In the original, one distinguishes a horn; presumably the animal is a goat. The head of the goat and the left foot of the komast disappear behind the frame and column, respectively.

As the running drinkers have neither animal ears nor animal tails, they cannot be sileni, cf. Oxford Classical Dictionary2, 956 s.v. "Satyroi." For similar but dressed komasts, cf. Greifenhagen, Antike Kunstwerke, 11, 45, fig. 24, Ergotimos cup; A. Seeberg, Corinthian Komos Vases, 66, no. 236, pl. 15, "Late Corinthian," similar in style. Several vases by the Amasis Painter show men or youths with wineskins and small game: Munich 8763, amphora type B, side A: Dionysus with four youths, one carrying wineskin over his shoulder, two youths carry sticks, each with a fox and a hare (Beazley, Paralipomena, 65). Geneva 1.4, amphora type B, side A: Dionysus with four youths; one brings a hare, the other a wineskin (idem, Attic Black-Figure Vase-Painters, 150, 8; Karouzou, The Amasis Painter, pls. 4-5). Amphora, Berlin 1690, side B: sacrificial procession where one man holds an oinochoe and wineskin, another carries a piglet (Beazley, Attic Black-Figure Vase-Painters, 151, 11; Karouzou, The Amasis Painter, pl. 9). Lekythos, Kerameikos, shows a man with a wineskin over his shoulder (Beazley, Attic Black-Figure Vase-Painters, 155, 61; W. Kraiker, Eine Lekythos des Amasis im Kerameikos, Beil. II). I owe the above references to Joan Mertens. Big thighs, big heads, and a running jump are akin in vases ca. 550-530 B.C., both by the Amasis Painter and others (e.g. Beazley The Development of Attic Black-Figure, 60, pls. 24-25).

E (fig. 37).Female dancer running to right. Panel: W. at top 0.064, at bottom 0.055; H. 0.12. Her arms are raised and her head turned to the right (toward the front of the monument). The top of the strip is broken off; the entire front including her right leg is missing. The panel is badly worn. Her hands are spread as if to support the top strip. Her hair falls slightly below her neck. She wears a belted chiton (belt omitted in drawing Fig. 32) and a cloak which falls from the shoulders with the lower edge forming an arc. The left foot disappeared behind the column.

F. Similar female dancer. Panel: W. at top 0.0675, H. 0.125. She is running with her left knee bent, right foot well back. The hem of her garment slips above the ankle of the back foot.

For cloaks, cf. Artemis of Olympia, Lamb,Ancient Greek and Roman Bronzes, 89, pl. 28 d, considered Corinthian, 6th C.; pl. 7, no. 55. One might see in the two dancers a simplification of the maenadic dance. L. B. Lawler, “The Maenad: A Contribution to the Study of the Dance in Ancient Greece,” MAAR 6 (1927), 17, for cloaks, 28 poses, perhaps most similar to castanet player on (Oltos?) kylix, Orvieto, pl. 17. Cf. also dancer with raised arms, Etruscan from Vulci, Neugebauer,Antiquarium, 69, pl. 29.

Reliefs, R. Side: Panels G-L (Figs. 29, 30, 38-43)

G. Lower part of kore walking to left. P.H. 0.10; W. (lower) 0.075. Two thirds of the panel is broken off. The finely polished, flatly cut garment has a vertical fold in front; the lower hem descends obliquely from front to back.

H (Fig. 39). Incomplete kore walking to left H. at left side 0.125, right side 0.118; W. upper and lower 0.075. Missing is the upper part with the head and some of the hand. The right arm of the figure is bent with the open palm horizontal and facing upward (as in the kore, panel B). The lowered left hand grasps the vertical folds of the chiton, two of which are indicated. A small "Ionic" mantle falls from the shoulders down the back, ending apparently just below knee level. On the other side, the cloak is apparently envisaged as falling from the right arm (not shown in drawing Fig. 38). The body contours of the back show very clearly through the drapery.

I (Fig. 40).Complete kore walking to left. W. lower 0.073, upper 0.075; P.H. 0.165. The front of the panel is broken off; it is very battered. The outside arm, presumably the left (she was holding the skirt of the chiton with the invisible right arm?), is raised, holding a pinecone. She wears a wide ribbon (diadem?) over her forehead; the hair falls to her waist in a stepped pattern. The mantle descends from the back and passes under the upper arm. Vertical folds of the mantle are flung over her arm, then fall to above knee level, ending in a pointed tip. Its lower edge is indicated as going from the knee to above the back ankle (seen more clearly in panel J).

Published: Richter, Korai, 93, no. 164, fig. 527.

J (Fig. 41). Complete kore walking to left. H. 0.15; W. lower 0.07, upper 0.0775. This panel is better preserved than panel I. The radiating arrangement of the hair waves over the skull and the way in which the mantle is draped around the body, passing first under the left arm and then back over it, is clearer. Hair arrangement: Richter, Korai, no. 44, fig. 153, Olive Pediment, 580-570 B.C.; no. 89, fig. 275, Lyons, ca. 550 B.C. The posture and costume in- tended is clearly that of later korai. Clearest explanation, ibid., pl. XVd, drawing of woman by Andokides Painter, sculpture ca. 520 B.C., figs. 358ff. The foldless mantle and stepped hair, however, are early traits.

K (Fig. 42). Lion sejant. H. 0.178; W. at top 0.073. The front part of the panel is broken off, and the lion's haunch is battered. The forelegs rest on a volute growing obliquely from the ground to the column. The head is turned backward, toward the front of the monument. The hindquarters are raised as if the lion is preparing to jump. The large simple mane is rendered in relief with a slight rise over the forehead and reaching just below the lower jaw. The upper jaw is pointed, giving an almost eagle-like appearance. There is a double outline around the right foreleg from the shoulder to the foot (not shown in drawing Fig. 38).

L (Fig. 43). Lion sejant. H. 0.17; W. at top 0.07. The forelegs are placed on an indistinct object, perhaps a rock. The hindquarters are concealed behind a column. The pointed upper muzzle, open mouth with tongue or teeth, ear, and eye are incised. There is no double outline on the leg.

The lion standing with forelegs on a volute is paralleled on terracottas from Sardis X, 19-21, inventory number Ter. 36, color. Lions are standard attributes for flanking images of Cybele: cf. Keil, Ephesos, fig. 26; 25 (Figs. 92-101), 256 (Fig. 442); Ch. II, "On Lions." For the type of the lion with head turned back and standing on a volute, cf. Corinthian bronzes and vases datable before the mid-6th C. B.C., G.M.A. Richter, Greek, Etruscan and Roman Bronzes, 7f., no. 13; Neugebauer, Antiquarium, 69, pl. 6. For comparable Middle Corinthian lions see D. G. Mitten in Master Bronzes, no. 59, 550-500 B.C.; Payne, Necrocorinthia, 69f., fig. 15, and Late Corinthian, pl. 37:6. The related Perachora lion in Boston is dated by Gabelmann as Middle Corinthian, 600-575 B.C. (Gabelmann, Studien, 50, no. 29, pl. 5). Similar but slightly later, Assos, Charbonneaux, SGR, 103, no. 2834, ill.

Reliefs, Back: Panels M-R (Figs. 31, 44-50)

The upper part is broken, including parts of two reliefs. The lower right corner is broken and reattached; there is also damage to the "platform." The surface of all the reliefs is very severely worn. There are traces of an incised meander preserved on the top and middle bands.

M (Fig. 45). Two birds (eagles?) attacking one or two animals climbing a tree. P.H. 0.07; W. 0.17. The left eagle is not complete in the drawing (Fig. 44); in the original the head with an eye and beak can be made out, striking at the indistinct head of an animal. The bird has a rounded body, a tail with three feathers, and a wide curving wing (tip behind column); the top wing is mostly lost. The head of the right eagle is striking at the round curving object almost entirely eroded. The downward pointing wing, part of the body, and the tail (tip behind column) of the eagle are still preserved. On the left of the tree is a figure of an animal, indistinct except for one paw (right) brought up against the tree J. K. Anderson, verbal communication Oct. 1976, feels it may be a fox). The round curving object on the upper right of the tree is possibly an animal with right paw brought upward. The subject is very enigmatic. Are the eagles protecting the sacred tree (of Cybele?) against animals which attack it?

N (Fig. 46). Lion and boar approaching a tree in which Peleus (now lost) has taken refuge. On the left is a massive boar with long spinal bristles rising without a break to the incised “mane.” The tail curls in a circle over the rump. The head is rubbed off. On the right a heavy-bodied lion jumps against the tree. His hindquarters disappear behind the column, the tip of the tail reappearing above? There are tips of tree branches (palm?) above the boar. This is the best preserved panel, very vigorously drawn and coming close to the original effect of the slightly raised relief. The subject is Peleus, whom the wild animals chased up the tree; subsequently the sage centaur Chiron appeared and threw a knife to Peleus, who was then able to escape. A centaur is seen in panel Q, but he probably does not belong to the Peleus scene.

Oinochoe, BMMA (1946-47) 255ff. = Richter, Handbook, 62f., fig. 44e; same scene, amphora Villa Giulia, 22247, Willemsen, Lowenkopf-Wasserspeier, 36. For other Sardian representations of boars, cf. BASOR 162, 39, fig. 24; D.P. Hansen, “Bronze Boar”; Greenewalt, “Wild Goat”, 72f., 88, no. 18, pl. 17; also Hanfmann, “Sardis und Lydien”, 521, pl. 7, fig. 10. Payne, Necrocorinthia, 70, n. 3, remarks that Corinthian boars never show the break in the center of spinal bristles, which is characteristic of the Ionian type. However, such a break does not really happen in the shrine relief on which the bristles seem to be of uniform H. without “mane.” For massive type and posture of lion see W. L. Brown, Lion, 74f., pl. 26c, though he thinks the animal is a she-bear (Caeretan hydria). Cf. also relief from Syme, Istanbul, Berger, Basler Artzrelief fig. 58, 510-500 B.C.

O (Fig. 47). Herakles fighting Nemean lion. H. at right 0.175, at left 0.17; W. at top 0.16. The panel is partly worn. The running Herakles swings his club (?) in his raised right hand and grasps the lion's throat with his left. He has a large nose, a beard, and long hair falling in a broad wave on the shoulders. The hair is tied with a ribbon which shows traces of red pigment. A V-shape below the neck seems to indicate a garment (short chiton?), but its lower edge is worn away. The roaring lion tears Herakles' left arm with his right forepaw, extends the left foreleg just below the right, and stems the left hind leg against Herakles' upper right leg. The tail swings in an arc upward behind the right hind leg. The rear of the lion disappears behind the column, as does part of Herakles' club. The scale of the figures is larger than in the two top panels.

Herakles and the Nemean lion is a favorite subject of archaic vase painters and bronze workers. Cf. Brommer, Herakles, 81f.; Payne, Necrocorinthia, 126, pl. 45:8; Kunze, AS, 95ff., esp. 101, pls. 19.IV g; 53.XXVIII y; 66.XLII j. The so-called Attic scheme which appears on "Argivo-Corinthian" bronze reliefs is the one adopted at Sardis. Payne links this scheme with black-figure painter Tleson and dates it 550-500. The Ionian example, CVA RodiItalia, pl. 430, is different. For an Anatolian lion fighter shown with Cybele in Cappadocian reliefs see Gall, Felsgraber, 590; in Phoenicia and Syria, Barnett, Nimrud Ivories, 66-68.

P (Fig. 48). Charioteer (Pelops) in chariot to left. W. of panel at top 0.155, at bottom 0.145; H. 0.16. The very worn panel shows a small horse, with nobly curving neck and mane, trotting (?) with the left foreleg raised. The contours are doubled to show a second horse (of a biga?). The chariot has an "Oriental" eight-spoked wheel with wide tire. The chariot box is of "Monteleone" shape with the upper edge slanting in front, then becoming horizontal before the round hand grip. The reins are indicated first by a broad strip from the neck of the horse to the hands of the charioteer, then by three lines below his hands. The charioteer stands erect, both forearms extended forward, and his back not quite touching the column. His long hair falls in "steps" down the back, and he has a beard. He wears a chiton incised with thin lines; a cloak is flung around the body in curving folds and then thrown over the left arm in two long, vertical folds. Other thin folds go from the stitched sleeve down the back. He is riding in parade, not racing.For the style of the folds, cf. the Croesan columns, Pryce, Catalogue, 59, no. B 121, fig. 63. For horses and charioteer with cloak thrown over arm, see Andokides Painter, Hanfmann, Ancient Art, no. 254, pl. 76; François vase, Zeus, Richter, AGA, fig. 107. For wheels, tires: Borchhardt, Epichorische Reliefs, pls. 47:2, 50, but with carts. It is tempting to see in the charioteer the Lydian Pelops, inventor of the Olympic chariot race. The Sardians of Hellenistic times and possibly earlier celebrated Pelops by calling one of the Sardian phylai Pelopis. In A.D. 29, they claimed him as their own when petitioning Tiberius for the Imperial cult (Tacitus Ann. 4.55 = Sardis M2, 64, no. 221), and he is shown on Sardian coins of Marciana and Platina, galloping in a chariot (BMC Lydia, nos. 132-133); the coins will be discussed in a forthcoming volume in this series by A. E. M. Johnston. Cf. for the traditional myths, OCD, 797, with Pindar Olympia 1.46, calling him Lydian. The Sardis relief would be the earliest representation: race with Oinomaos, close to Sappho Painter, and others listed by Brommer, Vasenlisten, 369ff. are of the 5th C. B.C.

Q (Fig. 49). Centaur walking or galloping to right. H. 0.155; W. at top 0.145, at bottom 0.13. Most of the surface is gone. There is a clearly visible outline of the upper hind legs, the left set back in walking posture; and the tail is very decoratively arranged in an upward spiral. By strengthening the lines that are only visible in special light, one can discern the upper outline of the body with the right arm raised in back, part of the top of the hair with a roll over the forehead, the beard, and a peculiar curving object, which might be the left hand, or an object held in it.

Nearest in time and location are the Larisa terracottas and the frieze from Assos, Åkerström, Terrakotten Kleinasiens, 55f., pls. 26-27; Assos, Charbonneaux, SGR, 103, ill.

R (Fig. 50). Three-figure group, subject uncertain. H. at left 0.165, at right 0.155; W. at top 0.142. The panel is broken across the lower right and very worn. Only the upper parts of the figures are discernible. On the right is a seated (?) figure, perhaps nude and facing left. Visible are the forehead, nose, moustache, pursed mouth, beard, and long hair falling down the back to below the shoulders. His left arm is bent and brought forward. Next to him a female figure facing right extends her small arm to grasp the chin (or reach just under the chin) of the seated figure. This may be the well-known gesture of supplication, or she might be stabbing him. She wears her hair in a bun; she is probably clothed, as an incised fold appears below the hand of the seated figure. From the left a young man bends his right arm and swings a weapon. He wears his hair in a roll (krobylos). Biceps seem to be indicated. One has the impression of a nude, athletic youth striding to the right. His weapon looks like a club, but there is a horizontal bar behind the head of the female which may be the tip of a spear. Possible subjects are the murder of Agamemnon by Clytemnestra and Aegisthus, or of Aegisthus by Electra and Orestes if the female is stabbing. Less likely is the death of Priam. If the female is supplicating, the subject is harder to guess.


The korai (priestesses), komasts, and dancers on the two side walls all come to the front of the building where the goddess receives them. The two lions are thought of as permanent attributes. A procession was also represented in Ephesus (Hanfmann, Croesus, 12f., figs. 20, 27-29). Processions coming along two sides to the front and center occur in late archaic Xanthos and early classical Thasos, most strikingly in the relief from the acropolis of Thasos (Getty Collection) where the seated goddess in the center is shown in a shrine (Berger, Basler Artzrelief, figs. 51-53; cf. also Picard, Manuel I, 88, fig. 41; Charbonneaux, SGR, 14 ill.; Richter, Korai, no. 192, figs. 612f.) and of course on the Parthenon. Like the reliefs of the Athena temple of Assos, the mythological reliefs on the back of the Sardis piece show no clear relation to the owner of the shrine, except possibly a panel at the top (panel M, eagles protecting sacred tree?). However, both Pelops and Herakles have Sardian associations, Herakles because of Omphale, and as the ancestor of a mythical royal dynasty.

A matter of great interest is the presence of Dionysiac komasts and dancers. It confirms the close connection of the orgiastic rituals of the Great Mother Cybele and the Lydian Bacchus (Bakz) celebrated by the chorus of Lydian women in the ecstatic hymn of Euripides' Bacchae 72-82. E. R. Dodds (Euripides' Bacchae, 76f.) makes the interesting suggestion that "from very early times a Divine Mother and a divine son were worshipped with dances and mountain rituals" (oreibasia) and that in Asia Minor the “Mother was Cybele and the son Bakkos.” (On the hymn cf. also R. P. Winnington-Ingram, Euripides and Dionysus, 32ff., 153ff.; J. Harrison, Mythology and Monuments of Ancient Athens, 49f.)

One may ask — but not answer with any certainty — whether the artist envisaged procession and rituals at one definite festival, or more likely, expressed the various essential aspects of the cult: procession to the temple and dancing and drinking on the mountain Tmolus, where Bacchus was born and where the lions of the goddess roared.

Function and Dating

The shrine with the goddess is certainly the most important piece of sculpture found at Sardis. Most probably this was a votive comparable to the models of buildings dedicated from Geometric times on. One might easily imagine that the patron or architect of a new temple might dedicate such a model; some scholars believe this is what Herodotus (5.60) said the Alcmaeonidae had done at Delphi. A temple so elaborately decorated might have been a royal present. As to architects, Theodoros, famous for developing with Rhoikos the Ionic order, worked for Croesus. It is less likely that the shrine model served as a supporting base for another offering.

Cf. O. Benndorf, “Antike Baumodelle”; S.D. Markmann, “Building Models”; A. Bundgaard, Mnesicles, 217-218; R. A. Staccioli, Modelli, 81ff., pls. 7-10, 12-16, 45.

Because of the two lions as attributes on the right side, the building may be the temple of Cybele, which was burned by the Ionians in 499 B.C., and the standing figure her archaic image. It is, of course, also possible that there were several shrines of Cybele, and this was intended to represent a smaller one, not the main temple.

The technique is highly unusual and experimental. It combines nearly round sculpture (goddess) and very light and shallow reliefs outlined with engraved lines and polished. As panel O proves, the reliefs were also painted, probably in strong, abundant polychromy like the contemporary poros buildings of the Acropolis in Athens (R. Heberdey, Altattische Porosskulptur, 46, n. 5) or the marble reliefs of the Artemis temple in Ephesus (Reutersward, Studien zur Polychromie, 35-37, 74-75).

The architecture and architectural sculpture are equally unusual and experimental. The plan, details of columns, and distribution of sculpture the entire height of the wall seem like an experiment on the way to the design of the Artemisium of Ephesus (ca. 560 B.C.?) where only the bottoms of columns and the balustrade above the roof edges were decorated with the sculptured friezes. Architecture and the parallels for the shrine reliefs, many of which are comparable to Corinthian art and to the Croesan columns at Ephesus, argue for a date in the Croesan era, ca. 550-540 B.C. The kore-like image, however, compares most closely with a Samian maiden dated 540-530 B.C. by Buschor (Altsamische Standbilder 5, 93; Richter, Korai, no. 153). Richter dates our piece "early in the last third" of the 6th C. B.C.; Tuchelt, Didyma (155f., 185f.) to 525-500 B.C. The style of the image and the shrine reliefs, however, makes a date later than 530 B.C. unlikely. Since not only the arrangement of costume but also the "straight-armed" way of holding an animal with the left arm are best paralleled among Samian sculpture (Buschor, Altsamische Standbilder 5, 90, 92, fig. 357, 550-540 B.C.), it is Samian and possibly Ephesian, not Milesian, inspiration that one perceives in the style of the image. One might discern a local Lydian exaggeration in the squatter proportions and in the over stressing of rope stylization of folds; but there is refinement in the rendering of the chiton and a slightly hesitant but evocative power in the outlines of the figures of the shrine reliefs with their unique dainty flatness.


Incompletely crystalized "half marble," coarse-grained, friable.

Heavily weathered. Missing upper part, lower I. corner, bottom of central corner of I. side. The frontal three-quarter columns are lost except for one base. Column and capital of middle column on r. side reattached. Lower r. front corner broken and reattached. Surface worn. Much cement from reuse in wall at lower front and back was subsequently cleaned off. Top part, including head of the goddess was split off forcibly with large pointed chisel (Fig. 23). Apart from a bit of red pigment in a meander and in the hair band of Herakles (panel O), original painting gone.

H. 0.62; W. at back 0.57; P.W. at front 0.41; P.D. at I. 0.33, at r. 0.442.
For new material on Cybele in Phrygia, see H. Güterbock, Ankara-Reliefs.
See Also
See also: LATW Cat. 34.
Published: BASOR 174, 39-43, figs. 25, 26; Hanfmann-Detweiler, Ages, 93, fig. 4; Hanfmann, Rayonnement, 494-496, figs. 1-3, pls. 124:3, 125:3; Pedley, Age of Croesus, 102ff.; Hanfmann, Classical Sculpture, 311, fig. 84, dated 550; Hanfmann-Waldbaum, “Kybele and Artemis”, 268, 2 ill.; Schefold, Die Griechen, 284, fig. 333; Richter, Korai, 66, 92, no. 164, figs. 524-527, "Siphnian Treasury group" sub-group V: 5, "Korai from the East, last third of the sixth century." Wesenberg, Kapitelle und Basen, 111f., no. 3, fig. 232. H. von Gall, Paphlagonischen Felsgraber and in a letter to G. M.A. Hanfmann, May 6, 1967, notes an "interesting continuity of this monument with Hellenistic Cybele terracotta Sabouroff." Salviat, Steles, 251, on Sardis monument, confirming that the other Cybele stelai show simplified naiskoi. Hanfmann, Croesus to Constantine, 12, figs. 23-26. M.J. Vermaseren, Corpus Cultus Cybelae Attidisque.