The Circle of Ancient Iranian Studies
[ Home ]
[ About CAIS ]
[ Articles ]
[ Daily News ]
[ News Archive ]
[ Announcements ]
[ CAIS Seminars ]
[ Image Library ]
[ Copyright ]
[ Disclaimer ]
[ Submission ]
[ Search ]
[ Contact Us ]
[ Links ]
By: Dr. Georgina Herrmann & Dr Vesta S. Curtis
Fig. 1. the Investiture of Emperor Ardashir I at Naqsh-e Rostam
Little of the written record of the Sasanian period has survived: it consists in the main of a few inscriptions carved on rocks or buildings, some ostraca, usually economic, and some seals and bullae. As a result the information provided by the two primary sources, the coins and the rock commissioned by some kings, are of major importance. Of the two, the most significant are undoubtedly the coins for they present an unbroken sequence throughout the four centuries of Sasanian rule (Göbl 1971). They provide a wealth of information, not only about the sequence and chronology of the rulers but also reflect changes in the economy and religious preferences. Despite their small size, Sasanian coins measure on average between 28 -30 mm, they depict a wide range of religious symbols, which emphasize the association of the king and kingship with divinity.
From the point of view of the rock reliefs, the most important aspect of the coins is that they illustrate a remarkable idiosyncracy of the Sasanian kings: each king wore a different personal crown, which became more and more elaborate during the four centuries of the dynasty. Fortunately, the different crowns have been identified from the coins (Herzfeld 1928; Erdmann 1951) and form the foundation for establishing not only which kings commissioned the different reliefs but also help to identify other minor arts which show the monarch. The eighteenth century scholar, Antoine de Sylvestre de Sacy used earlier work on Parthian and Sasanian coins and their legends by fellow Frenchman Joseph Pellerin (1767:pls. I-II) to compare Sasanian coin legends and inscriptions on rock reliefs in 1793, and de Sacy's knowledge of Pahlavi was used by Ker Porter, who compared the royal images on the reliefs with early Sasanian coins (Ker Porter 1821, I, 552, 561, pl. 58).
Fig. 2. Coins of Ardeshir
In a few cases identification by crown can be reinforced by the presence of contemporary inscriptions. The best instance of this – and a very important example - is the Investiture of Ardashir I at Naqsh-e Rostam (Fig. 1). This magnificent relief shows two mounted horsemen, one handing the ribboned diadem or symbol of kingship to the other, and both with defeated enemies lying dead beside them. The left hand figure is wearing the skull-cap surmounted by korymbos, the final version of his personal crown developed by Ardashir (224-241), seen on both his gold and silver coins (Pl. 4c; Lukonin 1970: fig.1, Va-b; Göbl 1971: pl.1, 9-10).
The figure on the right wears the mural crown, his hair arranged in a top-knot as well as falling in luxuriant curls to his shoulders. Prior to the reading of the inscriptions carved on the horses' shoulders various interpretations of this scene had been proposed: the artist and traveler, Sir William Ouseley, for instance, considered the scene represented the peaceful handover of power from the father, Ardashir, to his son, Shapur, rather than a divine investiture (Ouseley 1819, I, 285-6; 1821, II, 294). Such an interpretation is eminently reasonable, for both figures are similar in size and presentation, as Ker Porter (1821 I: 556-7) also suggested. However, in this case, not only is the king identified by a trilingual inscription in Pahlavi, Parthian and Greek on the shoulder of his horse – 'This is the image of the Ohrmazd-worshipping Majesty Ardashir, whose origin is of the gods…', 'This is the image of the god Ohrmazd' (Bach 1978: 281-2). Both the identities of the two principal figures and the subject of the scene – the divine investiture of the king - are therefore certain (Luschey 1987).
(fig. 3.) 'Investiture of Narseh' at Naqsh-e Rostam
However, such positive contemporary identification is unfortunately rare, and there are frequent disputes as to interpretation. For instance, the 'Investiture of Narseh' at Naqsh-e Rostam (Herrmann & Howell 1977) is often thought to show a similar scene (fig. 3.)– though with figures on foot rather than mounted – to the Investiture of Ardashir I. The central figure of the king, identified as Narseh (293-302) by his crown, is offered the diadem by a woman, often considered to be the goddess Anâhitâ: she is wearing a mural crown with her hair arranged in a topknot. The figure of a child stands below the diadem, and there are two courtiers, one incompletely carved, following the king. The evidence for this interpretation stems from the identification of king and god on the Investiture of Ardashir I, the woman's wearing of the mural crown, Anâhitâ's special divine position during the reign of Narseh and his successful attempt to overthrow Bahrâm III with the help of the goddess, as referred to by name in his Paikuli inscription [Lukonin 1979:127].
Some scholars, however, consider that the scene represents the king and queen with their son and heir standing below the diadem, because Shahbazi argued that the hand of a goddess cannot be concealed (1983). The only other instance of a woman wearing a crown on a royal relief occurs in the investiture relief in the Great Grotto at Tâq-e Bostân (Fukai & Horiuchi 1972: pls. IV, XXIII-XXIV), where she is usually identified as Anâhitâ. Here, the crown is decorated with vegetation, similar to twigs attached to the crown of Narseh (cf. Göbl 1971: pl.5.75-76). The few other representations of women on Sasanian reliefs, such as on the Investiture of Ardashir I at Naqsh-e Rajab, or the scene showing Bahrâm II protecting a woman from an attacking lion at Sar Mashhad, or the double and triple busts on the coins of Bahrâm II and on his rock relief at Naqsh-e Rostam, regularly show women wearing a high hat tied with a diadem. The latter have been interpreted as royal images or as a mixture of royal and divine images, while a comparison with Roman coins of the third century has even led to a suggestion that the king used multiple portraits in the Roman fashion (Göbl 1952).
Similar problems affect a late fourth century investiture/victory scene carved at Tâq-e Bostân. The king, wearing the crown of Ardashir II (379-383), occupies the center and stands on a defeated enemy (Fukai & Horiuchi 1972: pl. LXXIV). In front of the king stands a figure wearing a mural crown, behind, Mithra in his distinctive rayed crown. Some scholars rely on the evidence of the crown and consider that this scene represents the investiture of Ardashir II by Ohrmazd and Mithra, and his commemoration of one of his victories. Others consider that it represents the victory of Shapur II over Julian the Apostate. The latter interpretation is possible if the rock relief is regarded as a posthumous commemoration of Shapur's victory over Julian. In this case, Ardashir II, who came to the throne after the death of Shapur, occupies the center, while the figure on the right is identified as the triumphant Shapur II, standing over the dead Roman emperor (Calmeyer 1977; Shahbazi 1987).
There is clearly considerable flexibility in the interpretation of scenes showing two persons, facing each other, one offering a diadem, the other accepting it. The only contemporary evidence that we have for interpreting such a scene is, of course, the Investiture of Ardashir I at Naqsh-e Rostam, where the principals are identified by inscriptions. 'Investiture' scenes also occur on some early Sasanian coins, for instance on the reverse of coins of Hormozd I (273). The king stands on the left of the fire altar, reaching out for a diadem offered by a deity on the right of the fire altar, possibly Mithra or Anâhitâ (Göbl 1971: pl.3.36, 38). Investiture scenes also appear on both obverse and reverse of coins of Bahrâm II (Choksy 1989), on coins of Jâmâsp in the late fifth century and finally on coins of K¨osrow I (531-579) (Göbl 1971, pls.3, 4,11, 12).
Unfortunately, problems over identification and interpretation are not confined to 'investiture scenes'. There are major disagreements over the identities of the Romans on the famous series of reliefs commissioned by Shapur I (242-273) to commemorate his remarkable victories over Rome: three examples are carved at Bishâpur and one at Naqsh-e Rostam. The central scenes of two multi-register reliefs, Bishâpur II and III (Herrmann & Howell, 1980, 1981), show the mounted figure of the king, wearing a mural crown and identified as Shapur I. He is shown triumphant over three Romans, one kneeling in front of him, one dead beside his horse and one at his side, whose hand he grasps. As early as 1811 Sir William Ouseley (1819, I, 284-5) suggested that the kneeling figure represented the emperor Valerian, whom Shapur I had captured and made prisoner. Subsequently the standing Roman was identified as Philip the Arab, who paid Shapur tribute (Caratelli 1947, 225), and the dead Roman as Gordian III, killed in battle. This identification is still accepted by many scholars, and extended to the very damaged Investiture Scene, probably of Shapur I, at Bishâpur (No. I – Herrmann & Howell 1983). The Roman kneeling between the opposed mounted figures and the similar figure on the great Victory Relief at Naqsh-e Rostam (Herrmann, MacKenzie & Howell, 1989) then also have to be identified as Valerian.
An alternative theory was put forward by B.C. McDermot (1954), when he noticed the parallelism between Shapur's description of the principal events of his reign inscribed on the walls of the Ka¿aba-ye Zardosht at Naqsh-e Rostam, and the Romans illustrated on his reliefs at Bishâpur. In ˆKZ, pars. 6-8, Shapur I described how the emperor Gordian assembled a Roman army, marched against the Aryans and was killed. He was succeeded by Philip the Arab who paid 500,000 dinars. Later on, in par. 22 Shapur refers to battles at Carrhae (Harran) and Edessa (Urha) with Valerian, whom he captured with his own hand (Huyse 1999, I, pp. 22-24). McDermot agreed that the corpse indeed probably portrayed the dead Gordion III. However, he proposed that the kneeling figure surely represented Philip the Arab, who sued for peace in 244, while the standing figure, his hand held by the king, must show Valerian, captured in 260.
(fig. 4.) Shapour I, Great Victory Relief at Naqsh-e Rostam
The dispute as to whether the kneeling or the standing Roman represents Philip the Arab or Valerian, continues to this day, and without fresh evidence is unlikely to be resolved to the satisfaction of all. It is unfortunate, for it has an obvious effect on the interpretation and dating of the reliefs and, therefore, of building a 'sequence' of the reliefs of Shapur I or understanding their significance. For instance, the very damaged Bishâpur investiture scene is similar in design but not style to that of Ardashir I at Naqsh-e Rostam. If, following McDermot, the kneeling figure between the horsemen is identified as Philip the Arab, then the scene would probably have been carved in the mid-240s, shortly after the king's investiture and the payment of tribute by Philip. If, however, the kneeling figure represents Valerian, then the relief must have been carved after his capture in 260 A.D., and would show the king still commemorating his investiture some 16 years later. It also requires all the Roman victory scenes to have been carved after 260. Only the two small reliefs at Naqsh-e Rajab, one showing the king with members of the court, the other his investiture, would then be attributed to the early years of Shapur I. However, some scholars consider that the court scene also belongs to Shapur's final years.
The same problem, but even more acute, occurs at Dârâbgerd (Hinz 1969, Herrmann 1969, Ghirshman 1971, Göbl 1974, Trümpelmann 1975, etc.). Indeed, in many ways this relief encapsulates the problems bedeviling the agreed interpretations of the reliefs, in the absence of which their value as major historical documents is diminished. The mounted figure of the king occupies the center of the scene. His crown is the skull-cap surmounted by korymbos, typical of Ardashir I. A number of important figures approach him, on the head of one of whom the king rests his hand, while a dead figure lies beside his horse. Above these figures and the chariot at the right are tiers of heads, thought to represent Roman soldiers. Behind him are rows of Persian nobles.
This scene was first interpreted in 1811 by Sir William Ouseley (1821, II, 146-48). He had identified the king and Valerian on the Bishâpur reliefs commemorating Shapur's Roman victories and was happy to see the same scene represented at Dârâbgerd. He proposed that the figure on whose head the king's hand rests was a real portrait of the unfortunate Valerian. Ouseley, however, was unaware of the significance of the crowns. Ouseley's identification has been followed by many scholars, including Ghirshman, Hinz, Vanden Berghe and Göbl. This is an extreme example of 'dating by event'. It infers that the relief must have been carved post-260 A.D., and explanations have to be sought as to why the aged king Shapur chose to wear his father's crown and, furthermore, carved the relief in a style typical of Ardashir. Equally, if the evidence of the crown is accepted and the king is considered to be Ardashir I, then the event has to be established. Relatively little is known of the victories of Ardashir in the late 230s. To try to resolve the conundrum, some scholars have suggested that Dârâbgerd is an early relief of Shapur I, or perhaps was carved during the period of joint rule of father and son.
The period of joint rule began in April 240 and coins were minted with the busts of Ardashir and his co-regent (Göbl 1971: pl.2.19-20; Schippmann 1990: 19). In all these cases the scenes are considered to be a composite record of the victories of Shapur I rather than of an actual event, a theory reinforced by scenes such as the Investiture of Ardashir I, which could certainly never have actually taken place. However, Jean Gage‚ (1965) proposed that the Roman victories at Bishâpur illustrated only Shapur's victory over Valerian, and showed the king with Valerian and other captives, i.e. that it was a pictorial representation of an actual event.
An additional approach to dating the reliefs by the crowns, inscriptions or scenes portrayed is to undertake detailed stylistic and technical studies. These can sometimes suggest the order in which the reliefs were carved, although not the event. For instance, the jeweled 'chest harness' seen on the Tâq-e Bostân sculptures is a fourth century introduction, identified from the coins of Shapur III onwards (Göbl 1971:pl. 8.131,139, 9.155; Fukai and Horiuchi 1972: pl. LXVI), and does not occur on the reliefs of the third century kings. Therefore, these reliefs cannot be earlier than the fourth century. As well as changes of fashion in clothing and harness, there are changes of technique and tool-working which can provide an additional level of evidence (Herrmann 1981).
Distribution. The popularity of rock relief in Iran has waxed and waned throughout the millennia, with the most recent examples being those of the early 19th century Qajar dynasty (Diba and Eftekhar 1999). Rock reliefs were employed by the Elamites and again by the Achaemenians. Some Seleucid and early Parthian examples are known. However, the form had reached a low ebb by the second century AD, although the number found in Elymais shows that this type of sculpture remained popular until the end of the Parthian period (Vanden Berghe & Schippmann 1985).
The renaissance of rock relief, and of numismastic art, under Ardashir I suggests the importance of both traditions as a propaganda medium for a new dynasty. Relatively few kings commissioned rock reliefs: there are in all only some 30 carved throughout the four centuries of Sasanian rule, and most of these are concentrated in the first 80 years. Furthermore, both their chronological and their geographical distribution are uneven. The majority can be dated to the third and early fourth centuries and were located in the Sasanian heartland – in Fars. Of these, eight were carved on the cliffs at Naqsh-e Rostam, near Istakhr, and three more in a nearby grotto, Naqsh-e Rajab; six were carved in a gorge adjacent to the new city Shapur founded at Bishâpur, while Ardashir carved two reliefs on the approaches to the Firuzâbâd plain, where he built his great circular city. Other less significant reliefs occurred, usually singly, in a variety of sites, on cliffs or rocks and near pools: most of these date to the reign of Bahrâm II.
The fashion for rock reliefs was once again on the wane by the early fourth century – in Fars only one of three jousting scenes at Naqsh-e Rostam can be certainly attributed to a king, and that is to Hormuzd II (302-309) (Herrmann & Howell 1977). The remaining three sculptures of the dynasty are at Tâq-e Bostân, formerly a 'paradise' near Kermanshah. One was a traditional 'investiture scene' carved on the surface of a cliff, the other two were very different and were set within ayvâns carved out of the rock (Fukai and Horiuchi 1972: pls. II, LXVI, LXXIV). Two belong to the late fourth century, while the last and most ambitious sculpture of the dynasty, the 'Great Grotto' dates to the sixth century.
This erratic distribution may reflect a change in the preferred media for official art. It is during the long reign of Shapur II (309-379) that evidence of standardization of the official image can be established both on silver (Harper 1981) and on the lavish stucco decoration employed within palatial buildings. While no rock reliefs can be certainly attributed to Shapur II, although some scholars consider that Bishâpur VI was commissioned by that king (Herrmann & Howell 1981), examples of stucco busts probably illustrating Shapur II have been recovered from Kish (Baltrusaitis 1938-9, 634, fig. 211; Moorey 1978, 136, pl. L, K1406, K1409, who considers the bust is more probably of Bahrâm V, 420-438) and Dârâbgerd (Azarnoush 1994: pls. VII-XII, XXI-XXV). There seems to have been a deliberate change of medium, with official art concentrating on silver and stucco. Late fourth century reliefs, all carved in a new location at Tâq-e Bostân, seem to reflect stucco originals in style.
The last two sculptural works of the dynasty copied a palatial setting by being set within ayvâns, in these cases being carved out of the living rock face at Tâq-e Bostân. The fourth century ayvân is relatively small and plain with two frontal figures placed on a shelf at the back (Fukai and Horiuchi 1972: pl. LXVI). For once, there is no dispute over whom they represent since they are identified by inscriptions as Shapur II and Shapur III (383-388). Like the nearby 'investiture scene' of Ardashir II, the modeled relief so typical of earlier Sasanian rock reliefs has been replaced by details incised on raised surfaces - a major technical change mirroring the change of location.
There was then a break of more than a century before the last and most ambitious work of the dynasty, the Great Grotto, was commissioned – dating once again is uncertain, for the elaborate late Sasanian crowns are hard to differentiate. The royal headgear consists of a mural crown adorned with a moon crescent and korymbos, placed within a pair of wings. Such a crown is first shown on coins of Pêrôz (458-484) and becomes the standard crown under K¨osrow II (591-628) and most of his successors (Göbl 1971: pls.10, 14 -15). The hairstyle of the king resembles that of K¨osrow II on his coins (Göbl 1971: pl. 14, 217-219). The work is usually ascribed to K¨osrow II and occasionally to Pêrôz.
Every surface of the ayvân was carved, the façade as well as the back and side walls: indeed, the ayvân appears to be a stone version of doubtless numerous stucco originals within the palaces of the time. Fragments of similar scenes and motifs have been found from numerous late Sasanian and post-Sasanian contexts (Kröger 1982). The curve of the arch of the ayvân is outlined with an untied diadem, thus framing the scenes on the back wall – above, another investiture scene, showing the king standing rather awkwardly between Ohrmazd and Anâhitâ, while below he is shown as a mighty warrior, mounted on his charger (Fukai and Horiuchi 1972: pl.II). There was no need to show a dead enemy – the Sasanian army at the time was invincible.
On the side walls are superb and lively representations of the hunt with a wealth of documentation, closely relating to descriptions at the time of K¨osrow in the Shâh-nâma or Book of Kings (VII: 2187-2188, ll. 3890-3924). The whole work is a celebration of well-known themes from Assyrian times: the divine investiture of the king, the mighty warrior protected by the gods, and the royal hunt with its multiplicity of roles, including the triumph of good over evil, the slaying of dangerous animals and training for war.
Themes and Techniques
The revival of rock relief by Ardashir I probably reflects his admiration for his Achaemenian 'forebears' with their superb tomb reliefs at Naqsh-e Rostam and his desire to link himself with them as one way to commemorate and authenticate his newly won imperial power. His striving for a distinctive 'Sasanian' crown and suitable sculptural style can be documented both on his coinage and his reliefs. His portraits on coins depict at least five types of headgear, the earliest being a tiara in the fashion of the coins of Persis, followed by a jeweled tiara similar to that of Mithradates II (124/3-87), otherwise only shown on an early sketch at Persepolis (Göbl 1971: pl.1.4-8; Calmeyer 1976: fig.3 ). Another version is a crenellated crown usually associated with Ohrmazd (Göbl 1971: pl. 1.14-15). Neither the tiara nor the mural crown, both dated to the beginning of Ardashir's rule, occur on his rock reliefs (Alram 2002): instead a skull-cap surmounted by a top-knotted hairstyle, either uncovered on the jousting relief at Firuzâbâd, or covered on his investiture reliefs, is depicted. The coinage also documents a major evolution of style from crude early coins showing the king full-face, similar to late Parthian examples to the superb, sculptural versions of his later types.
A similar evolution can be established for Ardashir's reliefs. His two earliest examples were probably carved at Firuzâbâd, the site of his superb castle commanding the road between Shiraz and the Persian Gulf, his circular city and his palace in the plain. Probably the earliest, rather crude effort was sited by a bridge over the river and showed his investiture by Ohrmazd. The god on the left offers the king the diadem, with the fire altar represented below. The king is followed by his fan-bearer, his son and other magnates. His victory over the last Parthian king was carved beside the Sasanian roadway high up the mountain and showed a magnificent series of jousts, three armed Sasanian knights, the king Ardashir himself, his son Shapur, and the king's fan-bearer, the latter two identified by their devices, defeating three Parthians, the king, slain, his Grand Vizier and another. This relief is beautifully carved in relatively low relief, with the figures polished against a roughened background. It commemorates the victory of Ardashir at Hormizdgan in April 224, but Parthian opposition to the new dynasty did not cease with this battle, as Parthian coins were still minted in Mesopotamia as late as AD 228 (Sellwood 1981: 290-96). Ardashir carved two more versions of his Investiture by Ohrmazd, one at Naqsh-e Rajab and the other, his finest work and arguably the finest Sasanian rock relief, at Naqsh-e Rostam, referred to above. This is carved in such high relief it is almost in the round, and once again the figures are highly polished and highlighted against a roughened background. It is this 'third way', the study of technical details, such as tool-working, finish and minor details, that may offer an additional level of evidence together with that of the crowns rather than the inconclusive identification of figures or events. For instance, only one of the reliefs of his son and successor, Shapur I, has a polished surface in the same way: it is the one showing the king and his followers at Naqsh-e Rajab. Equally on this relief, the horse is shown wearing the severe type of bit worn by horses on Ardashir's reliefs. The horses on Shapur's other reliefs all wear a simple 'snaffle' bit, the reins are held tighter, and the surface of both the sculpted figures and the background are left deliberately rough. Such changes suggest that 'Shapur and his followers' was the first of the king's reliefs. The themes selected by Ardashir I, divine investiture and victory, remained standard throughout the dynasty. Investiture reliefs were commissioned by his son Shapur I and and grandson Bahrâm I, as well as arguably by Narseh and Ardashir II. Victory scenes were commissioned by Shapur I and Hormuzd II. It was Ardashir's great-grandson, Bahrâm II, an unsuccessful king, who turned his attention from the regular themes to a variety of court and family scenes. We see him receiving a foreign delegation at Bishâpur, but otherwise reliefs show him with his nobles and his family at Sarâb-e Bahrâm, Naqsh-e Rostam, and fighting a lion at Sar Mashhad. This preoccupation with family, reflected on his coins and on silver bowls with their multiple busts, may have been an attempt to secure his son's succession – a plan to be defeated by his uncle Narseh. Perhaps another reflection of the king's insecurity is that a wide range of divine symbols appears on his coins. Religion played an important role during his reign, and the power of the arch magus Kirdêr was pre-eminent – he appears on nearly all Bahrâm's reliefs.
The reign of Bahrâm II saw a weakening of central control, and a number of what appear to be unofficial reliefs were carved at Tang-e Qandil, Barm-e Dilak and Guyum. Similarly at Naqsh-e Rostam some jousting reliefs, with the exception of one attributed to Hormuzd II, are hard to attribute to a specific monarch. There seems to have been a breakdown in monarchical control of this medium: this occurs at the same time as a cessation of relief carving in Fars and the move to the paradise at Tâq-e Bostân, where a very different school of carving is represented, one allied to a flourishing tradition of carved stucco decoration. This change of medium is all the more probable since at least some reliefs were finished with a coat of plaster, and then presumably painted (Bishâpur VI, Herrmann & Howell 1981, 21-22, Bishâpur II, Herrmann & Howell 1983, 12-13). Plastered reliefs would not have been practical on an open hillside – an architectural setting would have been more appropriate. However, despite their uneven distribution and the manifold problems of attribution and identification, the Sasanian rock reliefs remain an outstandingly important source for the Sasanian period, many of which still require detailed study and recording.
Copyright © 1998-2011 The Circle of Ancient Iranian Studies (CAIS)