monuments generally included in discussions of Parthian art come from the periphery of the Parthian world—Syria, Mesopotamia, the edges of the Iranian plateau.


ART IN IRAN, History of

iv. Parthian Art

The monuments generally included in discussions of Parthian art come from the periphery of the Parthian world—Syria, Mesopotamia, the edges of the Iranian plateau—for the art of the Parthian capitals at Hecatompylos (Šahr e Qūmes), Hamadān, and Ctesiphon is almost totally lost. What survives are those works produced for vassals, independent noblemen, dwellers in cities under loose Parthian control, or in some cases, as with Palmyra and Commagene, in areas that lay totally outside Parthian political control. Nonetheless, for the period from the 3rd century B.C. to about the middle of the 3rd century A.D., the region extending from the Syrian desert through Iran and into Central Asia forms an artistic unit with certain definable characteristics: frontality, rigidity, great interest in representing details, especially the elaborately decorated “Parthian dress” and jewelry worn throughout the region, and conceptually, an intellectual rather than a visual approach to the depiction of figures and costumes (H. Seyrig, Syria 20, 1939, p. 180; see also M. I. Rostovtzeff’s classic article, “Dura and the Problem of Parthian Art,” Yale Classical Studies 5, 1935, pp. 155 304). As is to be expected across such a vast region, both the forms of art and their functions differ from site to site; in many cases, these differences can be explained by the influence of earlier local traditions. In general, Parthian art is characterized by eclecticism, a willingness to borrow style and motifs from Greek and earlier Near Eastern cultures and to recombine them to create new forms. In order to appreciate both the underlying unity and the regional variations, it is necessary to survey the art of areas under Parthian political control site by site. This essay will not discuss Palmyra, which was never under Parthian political control, Dura Europos, which, although ruled by the Parthians for over 200 years, was a remote outpost, or the kingdom of Commagene, which remained independent until it became a Roman province; material from these western sites will be used only for comparative purposes.

The Seleucid period is poorly represented in Iran and Central Asia. The most important site, the Greek colony of Āy Ḵānom (Aï Khanum) in Bactriana (modern Afghanistan) was destroyed about the middle of the second century B.C. The art of the earliest Parthian capital, Nisa, which is located in Soviet Central Asia, is strongly Hellenizing. Two small marble statues may indeed be Greek imports. One represents a draped goddess; the other, an Aphrodite Anadyomene type, is identified by the excavators, probably incorrectly, as Rhodogyne, a daughter of Mithradates (M. E. Masson and G. A. Pugachenkova, “Mramornye statui parfyanskogo vremeni iz Staroĭ Nisy (Predvarit. publikatsiya),” [Marble statues of the Parthian period from Old Nisa], Ezhegodnik Instituta Istorii Iskusstv Akademii Nauk 7, 1956, pp. 465 83; G. A. Koshelenko, Kul'tura Parfii, Moscow, 1966, pp. 36 38; R. Ghirshman, Persian Art, 249 B.C. A.D. 651, New York, 1962, p. 29, fig. 38). The “Square Hall” was apparently the center of a dynastic cult. The clay statues which stood in the gallery on the second level probably represent ancestors of the Arsacids. Some figures wore Greek, others Parthian dress (M. E. Masson and G. A. Pugachenkova, Iuzhno Turkmenistanskaya Arkheologicheskaya Kompleksnaya Ekspeditsiya, Trudy, IV, Moscow, 1958, pp. 355 62; D. Schlumberger, L'Orient hellénisé, Paris, 1970, pp. 35f., fig. 11; Colledge, Parthian Art, p. 75, fig. 13). The technique of sculpture in clay or stucco over a wooden core appears to have been particular to Bactriana, as shown by its use for statues of the Greek period from the “temple à redans” at Āy Ḵānom in Afghanistan (P. Bernard, Comptes rendus de l'Académie des inscriptions et belles lettres, 1969, p. 244, figs. 19 20) similar sculptures in high relief, also probably representing heroized ancestors, are known from the Kushan complex at Khalchayan, also in Central Asia (G. A. Pugachenkova, “La sculpture de Khaltchayan,” Iranica Antiqua 5, 1965, pp. 116 27; idem, Skul'ptura Khalchayana, Moscow, 1971), and the technique survives in the Buddhist sculptures of Hadda. The ivory rhytons found in the treasury show more clearly a mixture of elements derived from different cultures. The small figured friezes that decorate the open ends represent Greek scenes, in a roughly Greek style, largely Dionysiac, and must have been based on Hellenistic models. Some of the fantastic animals carved on the small ends are related to Achaemenid forms (e.g., a horned lion), others to Greek ones (e.g., a centaur) (Masson and Pugachenkova, Ekspeditsiya; Ghirshman, Persian Art, pp. 29f., fig. 4; Schlumberger, L’Orient hellénisé, pp. 36, 170, 173; Colledge, Parthian Art, Ithaca, 1977, pp. 115f., fig. 43). The Nisa objects probably date to the 3rd and 2nd centuries B.C.

It is particularly difficult to discuss the art of Iran during the period of Greek and Parthian rule, since so little can be accurately dated, and the cultural context is often unclear. Greek iconography was introduced into Iran in the Seleucid period; the stone lion of Hamadān has been convincingly compared to the lions of Charonea and Amphipolis and tentatively interpreted as a cenotaph for Hephaistion (H. Luschey, “Der Löwe von Ekbatana,” AMI, N.S. 1, 1968, pp. 115 22, pls. 45 50).

Rock reliefs were a traditional form of royal expression in Iran and continued during the Seleucid and Parthian periods. On a relief at Bīsotūn (Behistun) dedicated to Herakles Kallinikos by a Seleucid satrap in 148 B.C., the reclining hero is placed above a lion from a much earlier relief, which is made to function like a lion skin. Herakles holds a cup, and his club and bowcase are carved on the background (W. Kleiss, “Zur Topographie des "Partherhanges" in Bisotun,” AMI, N.S. 3, 1970, pp. 145 47, fig. 11, pl. 66). Also at Bīsotūn are two very badly damaged Parthian reliefs, one probably commemorating Mithradates II (ca. 110 B.C.), the other Gotarzes (ca. A.C. 50) (E. Herzfeld, Am Tor von Asien, Berlin, 1920, pp. 36ff.; Colledge, Parthian Art, p. 91, pl. 16; Ghirshman, Persian Art, p. 53, fig. 65). While these reliefs are carved against the face of the cliff, another group of reliefs, dated to the 2nd century A.D. on the basis of comparisons with Hatra, are cut on an isolated boulder. On the south face is a dignitary offering an incense sacrifice, while on each flanking face is a worshiper or an officer; all wear an unadorned form of Parthian dress (Kleiss, “Zur Topographie,” pp. 147 49, fig. 13; Ghirshman, Persian Art, 1962, p. 53, fig. 66; Colledge, Parthian Art, p. 91, pl. 16).

A relief carved on a boulder at Ḵong e Nowrūzī in Ḵūzestān seems to provide new material for the discussion of the vexed question of the origins of “Parthian frontality.” The relief shows a rider, whose head resembles that of Mithradates I (ca. 171 138 B.C.), followed by a page and advancing toward a group of four nobles. The rider and the page are in profile and apparently wear Greek dress. The nobles, clad in a simple form of Parthian dress—trousers and an undecorated tunic, the hem of which falls in three points—stand frontally. The rider and the page wear their hair relatively short, in the Greek fashion, but the nobles have a Parthian tripartite hairdo. The man closest to the rider is distinguished by his greater size, his sword, and the fact that an eagle flies toward him. The rendering also follows two different conventions: the frontal figures are carved in a more linear fashion than are the rider and his page (L. Vanden Berghe, “Le relief parthe de Ḫung i Naurūzī,” Iranica Antiqua 3, 1963, pp. 158 68, pls. LII LVI; Schlumberger, L’Orient hellénisé, pp. 40f., 176, S.I. 6). This frontality and linearity also characterize the later reliefs of Artabanus V from Susa, dated to A.D. 215 (Ghirshman, Persian Art, pp. 56f., fig. 70), as well as the rock reliefs of Tang e Sarvak (see below), Tang e Botān (A. D. H. Bivar, “The Inscriptions at Shimbar,” BSOAS 27, 1964, pp. 265 90), and Bīd Zard (Vanden Berghe, “Le relief parthe,” p. 167, pl. LVI.2), all in Elymais, and the figures on the isolated boulder at Bīsotūn. The contradictions in style and composition make the dating of the relief from Ḵong e Nowrūzī difficult; it is possible that the figures of the rider and the page were deliberately archaized for some now unknown historical reason. The scenes carved on one of a group of isolated boulders in Tang e Sarvak (Ḵūzestān) provide a good illustration of the range of subject matter considered suitable for rock reliefs in the Parthian period. On the north face the investiture of a king identified in the inscription as Orodes is shown. He reclines on an eagle footed couch, holding a wreath in one hand. He wears a tunic, trousers, and a helmet. Behind him stands a god, perhaps Bel, who also wears a helmet; the two standing figures at the left side of the relief are perhaps Athena Anahita and Mithras. The male figures all wear their hair in two bunches over the ears, a characteristic Parthian hairdo. Below the investiture scene is a panel containing three standing figures. To the right of the investiture, and on a corner of the boulder, is carved an altar on which stands a huge baetyl tied with a fillet. This baetyl is being worshiped by a huge figure on the adjacent face of the rock. This face is divided into two major registers. At the ends of the upper register are two seated figures, perhaps a king and queen, between whom are carved standing figures. The lower register contains four figures standing with upraised right arms, probably a gesture of homage to the seated figures above them; to the right of this group is a hunter on horseback. Below this register an isolated panel contains a scene of a man strangling a lion. These scenes, then, include investiture, worship, homage to royalty or nobility, and animal combats illustrating the prowess of a ruler or nobleman. The similarity of style to the stele of Artabanus V from Susa suggests a date in the later 2nd or early 3rd centuries A.D. (N. C. Debevoise, “Rock Reliefs of Ancient Iran,” JNES 1, 1942, pp. 97 101, pl. IIb; W. B. Henning, “The Monuments and Inscriptions of Tang i Sarvak,” Asia Major 2, 1952, pp. 151 71, pls. i xx; H. Seyrig, “Sur un basrelief de Tang i Sarvak,” Syria 47, 1970, pp. 113 16, pl. IX.3).

A number of votive images of worshipers, both in relief and in the round, apparently decorated the terraces of the sanctuary at Bard e Nešānda and, to a lesser extent, that of Masjed e Solaymān, located in the southern part of Elymais. The images at Bard e Nešānda represent for the most part worshipers clad in a rather unadorned form of Parthian dress and standing either with the right hand raised in worship or offering sacrifice at an altar. The most complex relief, which is incomplete, contains five figures. The principal figure, probably identified as a king or noble by his elaborate Parthian dress and high tiara, offers sacrifice. Behind him stand two figures in decorated Parthian dress, while another figure in simpler garments salutes him across the altar. The fifth figure is probably an attendant. All of the figures from Bard e Nešānda are rigidly frontal.

On the basis of comparisons with Hatra and Dura Europos, Ghirshman suggests that the sculptures date to the 2nd century A.D. at the latest and that some might belong in the 1st century (“Bard e Néchandeh,” Syria 41, 1964, pp. 315 21, figs. 14 16, pl. XX; Syria 42, 1965, pp. 300 10, figs. 9 15, pls. XX, XXI; Terrasses sacrées de Bard è Néchandeh et Masjid i Solaiman [MDAI 45], Paris, 1976, I, pp. 2l 23, fig. 11 and pl. XIII; for other sculpture from the site see pp. 30 38). The workmanship is simpler than at Hatra, presumably because the sites were located far from the centers of the Parthian court. Vestiges of Greco Roman influence, in motif if not in style, are present in the high relief sculpture of Herakles and the Nemean lion from Masjed e Solaymān (Ghirshman, “Masjid Solaiman ou Mosquée de Salomon,” CRAI, 1968, pp. 11 15, fig. 3; Terrasses Sacrées, pp. 119 22, pl. LXXXI, 5; 23, 24, GMIS 30; for other sculpture from the site, see pp. 122 30). The popularity of Herakles throughout the Parthian world is notable. The relief of Herakles from Bīsotūn shows that he was introduced into Iran during the period of Seleucid control. In addition to Masjed e Solaymān, Herakles is attested during the Parthian period at Tang e Botān in the Šīmbār valley (Elymais) (Bivar, “Inscriptions,” pp. 265 90; Colledge, Parthian Art, p. 92, pl. 18), and the figure strangling a lion from Tang e Sarvak is probably based on Herakles as well. He is one of the few deities of Greek origin to be popular at Hatra, Dura Europos, and (to a lesser extent) Palmyra (S. B. Downey, The Excavations at Dura Europos, Final Report III, part 1, fasc. 1: The Hercules Sculpture, New Haven, 1969). In Iran, he may have been associated with Vərəθraγna, in Mesopotamia, perhaps with the old Mesopotamian “nude hero.”

Sculpture of the Seleucid and Parthian periods in Iran is also represented by chance finds from a number of sites, including Susa, the Mālamīr mountains, Nehāvand (Ghirshman, Persian Art, pp. 18f., figs. 23, 24), and Dīnāvar (ibid., p. 18, figs. 21, 22). A female head from Susa, signed by Antiochus, son of Dryas (Ghirshman, Persian Art, p. 96; Colledge, Parthian Art, pp. 83f., pl. 9c), demonstrates the presence of Greek art, perhaps even of Greek artists. The head, frequently identified as a portrait of Musa, the consort of Phraataces (ca. 2 B.C. 4 A.D.), is more likely a Tyche, since she wears a mural crown. Its stepped battlements resemble those of Achaemenid and Parthian architecture (H. Frankfort, The Art and Architecture of the Ancient Orient, Baltimore, 1969, pls. 181, 183; Ghirshman, Persian Art, fig. 42) and suggests that in spite of its superficially Hellenistic appearance, the head was carved in Iran. Another sculpture of Hellenistic style, representing a goddess or a woman clad in a distinctively Greek garment, was found in the Baḵtīārī mountains (Ghirshman, Persian Art, p. 22, fig. 28).

The most important group of sculpture in the round from Iran, including works in both Greek and Parthian styles, was found in an open air complex, perhaps a dynastic sanctuary, at Šāmī in Ḵūzestān. The date of both the complex and the sculpture is uncertain (K. Schippmann, Die iranischen Feuerheiligtümer, Berlin and New York, 1971, pp. 277ff.). Ghirshman has argued, on the basis of the possible identification of a bronze head as a portrait of Antiochus IV Epiphanes (175 164 B.C.), that the complex is one of his sanctuaries destroyed by Mithradates I (Persian Art, pp. 19 21, fig. 26). The bronze statue of a noble, probably a local dynast, is hard to date far lack of securely dated comparative material from Iran. Ghirshman places it in the 2nd century B.C. (Persian Art, pp. 87 89, fig. 99), Godard in the 1st century A.D. (“Les statues parthes de Shami,” Athār é Irān 2, 1937, pp. 285 305). The figure wears a rather peculiar form of Parthian dress; a tunic crossing on the chest, and chaps worn over a sort of undergarment. This form of tunic is known also at Hatra (e.g., Ghirshman, Persian Art, p. 86, fig. 98), and chaps worn over trousers appear also on an unusual figure from Hatra (F. Safar and M. A. Mustapha, Hatra: The City of Sun God, Baghdad, 1974, p. 78, fig. 24) and on two figures from Qaṣr al Abyaż at Palmyra (H. Seyrig, “La grande statue parthe de Shami et la sculpture palmyrénienne,” Syria 20, 1939, pp. 180f., pl. XXX). The head is too small for the body and was cast separately, perhaps at Susa, the nearest city. As Seyrig notes, the general stylistic features of the statue—its stiff frontality, the conventional rendition of the folds of the tunic and trousers, the generalized facial features—link it to the art of Palmyra, far to the west (Seyrig, “La grande statue parthe,” pp. 177 81). A stone figure of lesser quality shows a nobleman in a tunic with his right hand raised, presumably in a gesture of worship or respect; his features are also treated in a generalized, non portrait like manner (Ghirshman, Persian Art, p. 27, fig. 36). A female head in Greek style, perhaps representing Aphrodite, was also found in the complex (ibid., p. 19, fig. 25).

The so called Frātadāra temple at Persepolis probably dates to the 3rd century B.C. (Schippman, Feuerheiligtümer, pp. 177 85, but see Boyce, Zoroastrianism II, Leiden, 1982, pp. 226f.). Low relief carvings of a prince (or a priest) and a princess, each holding barsoms (a bundle of sacred rods), are placed on the insides of a window frame (Ghirshman, Persian Art, p. 26, fig. 34). The placement of sculptures in this position clearly recalls Achaemenid practice, as does the figure style, although the relief is lower and the carving simpler.

The only surviving paintings in Iran attributed to the Parthian period were found in a palatial religious complex at Kūh e Ḵᵛāja in Sīstān, at the eastern limits of Iran. The paintings, which decorated a vaulted corridor, are attributed to a relatively early phase, since they were later covered by a layer of bricks. Stein considered them Buddhist, while Herzfeld associated them with the first building phase, which he dated to the 1st century B.C., the period of Arsacid rule in Iran and Saka rule in the East (A. Stein, Innermost Asia, Oxford, 1928, II, pp. 913 21; E. Herzfeld, Iran in the Ancient East , London and New York, 1941, pp. 291 97). Gullini, who had divided the site into five phases (one Achaemenid, two Parthian, one Sasanian, and one Islamic), confirms Herzfeld’s attribution of the paintings to the period of Parthian political control in Iran, loosely defined between the 3rd century B.C. and the 3rd century A.D. (G. Gullini, Architettura iranica dagli Achemenidi agli Sasanidi, Turin, 1964, pp. 443 53). Herzfeld recognized the mixed character of their style and iconography: a ceiling with painted coffers containing rosettes and riding Erotes, derived from Greek art; a group of three male figures in three quarter view, perhaps gods, wearing Greek dress and in one case a winged helmet. Another group, traditionally identified as a king and queen, is seated in three quarter view on an elaborate throne and wears dress and ornaments akin to those of Parthian Hatra and Dura Europos. Herzfeld assigned these last two groups, as well as a remarkable head of a man in three quarter view, to a late phase of Greco Bactrian art. Recently, the eclectic style and the lack of frontality, as well as a reevaluation of Gullini’s stratigraphy, have led some scholars to consider lowering the date to the end of the Parthian or the beginning of the Sasanian period (Schlumberger, L’Orient hellénisé, pp. 53 59). On the other hand, the stucco decoration of Kūh e Ḵᵛāja, with its all over patterns of interlocking circles and meanders alternating with stepped battlements, finds parallels in the decoration of the Parthian structures of Qalʿa ye Yazdegerd, Warka, and Assur (Ghirshman, Persian Art, pp. 37, 41, fig. 54; Schlumberger, op. cit., pp. 55f., fig. 25).

Qalʿa ye Yazdegerd lies on the westernmost slope of the Iranian plateau and belongs culturally to the borderland between Iran and Mesopotamia. Keall has dated the site to the Parthian period on the basis of its stucco decoration and has suggested that it might have been the stronghold of an independent nobleman. At this stage of the excavations, it is not possible to analyze the plan of the building (a palace?), but Keall’s hypothesis that some of the rooms were ayvāns is plausible. The walls of important rooms were covered with decorative stuccoes, some of which are used to articulate the wall surface. The patterns used in the stuccoes of Qalʿa ye Yazdegerd—meanders, stylized vegetal patterns framed in squares or semicircles, battlements—are similar to those seen at Parthian Uruk Warka and Assur. Likewise, the pseudo Corinthian capitals found at Qalʿa ye Yazdegerd are similar to those from a Parthian house at Warka. An unusual capital from Qalʿa ye Yazdegerd shows a female nude, probably Aphrodite, between dolphins which take the place of volutes. Unlike the stuccoes from Warka and Assur, which utilize purely abstract patterns, those of Qalʿa ye Yazdegerd include many figural motives derived from Greco Roman art, such as Erotes, nude females, Attis like figures, dancing figures, both nude and draped, and animals being hunted. Heraldically confronted griffins are also derived from the Greco Roman repertoire, but another type of griffin has a curved wing that suggests the sēnmurw of Sasanian art; unfortunately, all specimens discovered so far are broken just behind the wing, so that it is impossible to tell whether the rest of the body was present. Other types, such as a ring enclosing a male bust with a billowing Parthian hairdo and wearing a torque, and a panel with intertwined dragons, recall motives used at Hatra. Enough traces of the bright paint that once covered the stuccoes remain to make possible an eventual reconstruction of the color scheme (E. J. Keall, “Qalʿeh i Yazdigird: The Question of its Date,” Iran 15, 1977, pp. 1 9; W. Andrae and H. Lenzen, Die Partherstadt Assur, Leipzig, 1933, repr. Osnabruck, 1967, pp. 53f., pls. 14 21; J. Schmidt, XXVI. und XXVII. vorläufiger Bericht über die Ausgrabungen in Uruk Warka, Berlin, 1972, pp. 46 55, pls. 32 39).

The three most important sites of the Parthian period in modern Iraq are Hatra and Assur in the north and Uruk Warka in the south. Hatra is the most impressive; though there is some evidence, so far unpublished, of remains from the Assyrian period. The city as we know it was probably founded in the late 1st century B.C. and flourished until its fall to the Sasanians in A.D. 240/41. Excavation has been concentrated on the religious structures that dominate the site—the great temple of Šamaš and the smaller temples grouped around it. There are vast quantities of stone sculpture, including architectural decoration; religious sculptures, both in relief and in the round, representing the numerous deities of the Hatran pantheon; and statues of the kings, princes, and nobility of Hatra.

The temple of Šamaš was heavily decorated: The voussoirs of the arches of the five ayvāns were filled with the high reliefs, presented as full figures, isolated heads, or animal protomes. Subjects include deities, nobles, and satyrs. The idea of placing sculpture in the voussoirs of arches is probably derived from the Roman world, but the sheer quantity surpasses that of Roman examples (H. von Gall, “Zur figuralen Architekturplastik des grossen Tempels von Hatra,” Baghdader Mitteilungen 5, 1970, pp. 7 33). The arches of the so called temples of the triad are similarly decorated. Other sculpture from the temple of Šamaš is probably apotropaic, including a bizarre male Medusa, the consoles near the entrances to the ayvāns, the victories on the upper walls, and the masks placed high on the walls inside the main ayvān. The eagles on the cornice of this same ayvān surely symbolize either Šamaš or the eagle god Samya, while the bull protomes in the northernmost ayvān perhaps stand for Hadad. (Safar and Mustapha, Hatra, figs. 92, 394f.; Ghirshman, Persian Art, fig. 49).

The divinities represented at Hatra fall into two main categories: sculpture in the round and cult reliefs. Some freestanding sculpture was probably imported from the Greco Roman world, most notably a group of marbles representing Poseidon, Apollo, and Eros, found near the temple dedicated to Bar Maryn, the son of the triad (Safar and Mustapha, Hatra, pp. 120 23). Most of the sculpture was probably made locally and is eclectic both in motif and in style. In several sculptures of Herakles the nude hero is shown in his Greek form, but the patterning of the lion skin recalls Assyrian work, and a statue dedicated by the Roman military tribune Petronius Quintianus in the mid 3rd century A.D. shows Herakles wearing local necklaces (Downey, Dura Europos, pp. 83 96, pls. XVII XXIV). Another statue represents a god identified as Assurbel on the basis of inscriptions, but the image bears a striking resemblance to the bearded Apollo of Hierapolis as described by Macrobius. The god wears a Greek cuirass but also a Parthian torque. On his back is a cloak with scales and a Medusa mask, a variant on the Greek aegis. The cylindrical character of the statue and the abstract patterning of the beard recall Assyrian art. At his feet kneels a female figure in a mural crown, obviously derived from a Greek Tyche but transformed in style; she is flanked by eagles, a symbol of a local god (Ghirshman, Persian Art, fig. 1; R. Du Mesnil du Buisson, “De Shadrafa, dieu de Palmyre, aà Baʿalshamin, dieu de Hatra,” Mélanges de l'Université St. Joseph 38, 1962, pp. 151 55; Schlumberger, L’Orient hellénisé, pp. 143f.). A high relief sculpture of Allāt, dressed as Athena and flanked by two goddesses in the local dress, all standing on the back of a lion, again illustrates the eclectic character of Hatran art (Ghirshman, Persian Art, p. 92, fig. 103). The so called “Cerberus relief” shows a god, his consort, and a three headed dog. The god stands frontally, clad in Parthian dress somewhat like that worn by the bronze figure from Šāmī; an eagle perches on his horned diadem (which surely recalls the earlier Mesopotamian horned crown), and snakes spring from his waist and shoulders. In his upraised right hand he holds an axe; in his left, a sword and the leash of the dog. While the triple headed dog is clearly based on the Greek Cerberus, here he identifies the god as the underworld god Nergal. His consort, seated frontally on a throne flanked by griffins, is crowned by an eagle. In her right hand she holds a feather; in her left, the standard, an important Hatran religious symbol. Her footstool is decorated with fish. The standard, the eagle, and the fish probably identify her as Atargatis, the goddess of the Syrian city of Hierapolis. Additional animal attributes are associated with these and other Semitic deities (H. Ingholt, “Parthian Sculptures from Hatra,” Memoirs of the Connecticut Academy of Arts and Sciences 12, New Haven, 1954, pp. 17 33; Ghirshman, Persian Art, p. 87, fig. 98).

Other types of religious sculpture at Hatra include model shrines, some with standards carved on the columns, and cult banks with images of divinities, such as Herakles and a centaur, or Herakles, Allāt, and another goddess (S. B. Downey, “Cult Banks from Hatra,” Berytus 16, 1966, pp. 97 109; idem, Dura Europos pp. 88 90). The Hatran kings and nobility dedicated statues of themselves in the numerous temples of city, a practice that presumably continued the old Mesopotamian custom of placing surrogate statues in temples to offer prayers for the dedicant. Most figures, both male and female, raise the right hand in a gesture of worship; some hold a leaf, probably an attribute of holiness, in their left hand. The male figures, normally bearded, are clad in richly decorated Parthian dress; they normally carry a sword and a dagger and may wear headdresses of various types, such as a high, decorated tiara (Uthal, an unknown figure) or an eagle crowned diadem (Sanatrūk). The headdresses are presumably an indication of status, but too many figures remain unidentified to allow secure interpretations. The garments are often elaborately decorated in abstract patterns meant to represent embroidery and applique (Ghirshman, Persian Art, pp. 90 94, figs. 100, 102, 105; J. Teixidor, “The Kingdom of Adiabene and Hatra,” Berytus 17, 1967 68, pp. 1 11). Similarly elaborate Parthian dress appears, in slightly varying forms, as far east as Sorḵ Kotal and even Mathurā in the Kūšān kingdom (Colledge, Parthian Art, p. 87, figs. 14 a, b; D. Schlumberger, “The Excavations at Surkh Kotal and the Problems of Hellenism in Bactria and India,” Proceedings of the British Academy 47, 1961, pp. 77 95) and as far west as the Anatolian kingdom of Commagene and the city of Palmyra in the Syrian desert (F. K. Dörner and T. Goell, “Arsameia am Nymphaios,” Istanbuler Forschungen 23, Berlin, 1963, pp. 197 227, pl. 48; M. A. R. Colledge, The Art of Palmyra, Boulder, Col., 1976, pp. 64 80, and passim; Ghirshman, Persian Art, pp. 57 68, figs. 79 80). Both the poses and the features of the Hatra statues tend to be stereotyped: youths are beardless, with aureole hairdos and bland features, while men wear mustaches, spade shaped beards, and either aureole or tripartite hairdos. While these statues are not portraits in the Roman sense, there are signs of individuality. For example, Sanatrūk’s face is extremely flat, and the head of an unidentified noble has a wrinkled forehead, which may be a sign of Roman influence. Most of the men have distinct pot bellies, perhaps a sign of affluence in a region in which food cannot have been plentiful.

The statues of women also place great emphasis on the detailed depiction of distinctive local dress and an abundance of rich jewelry. The degree of elaboration varies considerably and probably reflects differences in status. Šāprī (Shapry), the daughter of Sanatrūk, wears an elaborate series of necklaces, and her high tiara is decked with strings of jewels. Ubal, daughter of Jabal, wears a simplified version of the same costume, and the stylizations of the drapery, especially the whorl patterns over the breasts, are more simplified also (Ghirshman, Persian Art, pp. 93 95, figs. 104, 106; D. Homès Fredericq, Hatra et ses sculptures parthes, étude stylistique et iconographique, Istanbul, 1963, pp. 17 36). This female costume is known, with slight variations, from Palmyra (Colledge, The Art of Palmyra, pp. 69 72) and Dura Europos (A. Perkins, The Art of Dura Europos, Oxford, 1973, pp. 38 41, pl. 10) to Kūh e Ḵᵛāja in the east (Ghirshman, Persian Art, p. 42, fig. 56). As in the male statues, it appears that the sculptors made some attempts to suggest individual traits within the confines of their highly schematic system. What evidence there is suggests that most of the sculpture was the work of local artists. It varies considerably in quality, from poor adaptations of western types (e.g., some of the figures of Herakles), to works of considerable originality judged on their own terms (e.g., the statues of Sanatrūk and Šāprī, and especially the figure of Assurbel).

After a period of decline following its capture by the Babylonians, Assur revived as a Parthian city, apparently in the 1st century B.C. The major remains of the Parthian period are architectural. The palace, probably built in the 1st century A.D., has as its main feature four ayvāns facing onto a central court. The façades of these ayvāns are covered with an elaborate three tiered decoration of engaged columns and pilasters flanking doors on the ground level and blind niches on the upper levels. The columns are topped with pseudo Ionic capitals, the pilasters with moldings decorated with motives of classical origin used in unclassical combinations. The columns “support” friezes covered with geometric and floral all over designs similar in concept to the stucco decoration of Kūh e Ḵᵛāja, Qalʿa ye Yazdegerd, and Warka (Andrae, Assur, pls. 14 21). These architectural elements were brightly painted. Little sculpture was found at Assur but three stelae from the Gate House provide important evidence about the date of the introduction of frontality into Parthian art. Each stele shows a standing man in undecorated Parthian dress. On two of the stelae the figures are in profile; one of these bears the date of 89/8 B.C. (if Seleucid era) or A.D. 12/13 (if Arsacid era). This figure holds a feather in his left hand, while his right hand is raised toward images of the sun and the moon carved in low relief on the background. The style of the third stele, in which the figure is frontal, is so similar to that of the other two as to suggest that it is roughly contemporary (ibid., pp. 105f., pl. 59a d; Schlumberger, L’Orient hellénisé, pp. 120f., fig. 42). A graffito on a pithos is perhaps based on a painting in a temple. It represents two men clad in Parthian dress offering incense sacrifices to two deities, one seated and the other (perhaps a nude goddess) reclining. All figures are rigidly frontal. Bits of figured wall paintings were found in the palace but are too badly preserved to allow reconstruction (Andrae, Assur, pp. 109 12, fig. 46, pls. 61f.).

The remains of the Parthian period at Warka have been little explored. The temple of the otherwise unknown god Gareus, dated to about A.D. 110 on the basis of an inscription, is decorated on the outside with engaged pilasters topped with brick reliefs of monsters, an elongated, long necked griffin like those on the frieze of the temple of Bar Maryn at Hatra, and a dog in a similar style. Part of a life size bronze foot clad in a soft shoe is probably all that remains of the cult statue (E. Heinrich, “Sechster vorläufiger Bericht über die von der Deutschen Forschungsgemeinschaft in Uruk Warka unternommenen Ausgrabungen,” Abh. der Preussichen Akad. der Wissenschaften, phil. hist. Kl. 2, Berlin, 1935, pp. 33 35, pls. 25a; 26a, b; 30; Safar and Mustapha, Hatra, p. 347). The only other artistic remains at Warka are the decorative stuccoes from a Parthian house, already discussed in connection with Qalʿa ye Yazdegerd.

Our uneven knowledge of Parthian art, especially the great lacunae in the evidence from their Iranian homeland, make the assessment of the Parthian achievement difficult. The functions of art varied widely from site to site, apparently responding in large part to differing economic and social conditions. Thus, at Hatra, Šāmī, and probably also Nisa and Kūh e Ḵᵛāja, portraits of rulers are prominent. At major religious centers, such as Hatra and Bard e Nešānda, much of the sculpture has religious subjects, and the nobility is strongly represented. Art forms also appear to have been influenced by the traditions of previous cultures in the wide area that fell under Parthian control: rock reliefs with religious and dynastic subjects are particularly prominent in Iran, while sculptured images of worshipers most often appear in temples at Hatra. Perhaps because so much of the art, whether religious or aulic, was ceremonial, the sculpture and painting has a solemn, almost hieratic quality. The emphasis on details of decoration of clothing and on the depiction of ornaments is perhaps the result of a desire to emphasize rank.

The almost constant frontality of figures even in narrative scenes, whether it is of Greek origin (E. Will, Le relief cultuel gréco romain, Paris, 1955, pp. 219 55; Schlumberger, “Descendants non méditerranéens de l'art grec,” Syria 37, 1960, pp. 253 81) or Iranian (Rostovtzeff, “Dura and the Problem of Parthian Art,” pp. 238 41), or (as seems more likely) an invention of the peoples of Syria and Mesopotamia during the period of Parthian rule (Colledge, The Art of Palmyra, pp. 126 28; S. B. Downey, “The Stone and Plaster Sculpture: Excavations at Dura Europos,” Monumenta Archaeologica 5, Los Angeles, 1977, pp. 283 87), seems to result from a desire for direct contact between the viewers and the figures—whether divine or human—in a work of art. This frontality gives a ceremonial quality to even relatively humble works of art. Perhaps the Parthian achievement lies in the ability to select from art produced by other peoples the elements best suited to its purposes and to create from these diverse elements an art that responds to the varying needs of the rulers of the Parthian empire.



See also W. Andrae, Hatra I-II, Leipzig, 1908-12.

M. A. R. Colledge, The Parthians, London, 1967.

Survey of Persian Art. E. Porada, Ancient Iran, London, 1965.

G. A. Pugachenkova, Khalchayan, Tashkent, 1966.

L. Vanden Berghe, L'archeologie de I'Iran ancien, Leiden. 1958.

(S. B. Downey)

Originally Published: December 15, 1986

Last Updated: August 15, 2011

This article is available in print.
Vol. II, Fasc. 6, pp. 580-585