The Art of the Elamites


By: Edith Porada

 Columbia University

With the collaboration of R. H. Dyson and contributions by C.K. Wilkinson


The destruction of the great old city of Ur in Mesopotamia by the Elamites in about 2000 B.C. left a deep impression on the contemporary Mesopotamians. Two Sumerian lamentations on clay tablets reflect the memory of this event: the lament over the destruction of Ur and the lament over the fate of Ibbisin, the last king of Ur, who was led away into captivity. A few lines of the latter lament describing the fate decreed by the great gods Anu and Enlil follow in translation: ' . . . hostile Su people and Elamites will attain the inhabitants [of UR], the king [of Sumer] will have to leave the palace, Ibbisin will have [to go] to the country of Elam, [go] from the Sabu mountain, the "breast" of the mountain range, to the end of Anshan; like a bird which left its abode, like a stranger [he will not return] to his city.' [1]

Not only the king of Ur, whose dynasty had ruled over all of Mesopotamia and Elam, but also the patron goddess of Ur, Ningal, seem to have been led away into captivity. [2] This statue was not the only one which was taken to Elam. Several centuries later an Elamite king Shutruk-Nakhunte dragged two of the greatest works of Mesopotamian art from the town of Sippar [north of Babylon on, the Euphrates] to Susa: the stele of Naramsin of the Akkad dynasty and that of Hammurabi of Babylon. The French excavators of Susa discovered these monuments, as well as others which had been brought from Eshnunna in the Diyala valley, in north-eastern Mesopotamia. [3]

The divine and royal statues of the ancient Near East were meant to assure for the king the enduring protection of the deity, well being and a long life. Reliefs which showed a military victory of a ruler or his performance of a ritual action were surely intended to eternalize the effectiveness of such deeds. In the country of their origin works of art of this type must have been considered charged with beneficial power. Hostile intruders therefore would destroy them or lead them into captivity as representative of the conquered peoples--like Ningal of Ur. Often the conqueror had the original inscription erased and his own name and even a record of his conquest engraved on the captured statue or stele. To erase the name of a person literally meant to kill his memory.

With the destruction of Ur the Elamites under a king of Simash liberated themselves from Mesopotamian tutelage, but not for long. The successors of the Third Dynasty of Ur as rulers of Mesopotamia, the kings of Isin and of Larsa, continued a policy, developed earlier in relations with Elam, of 'military pressures and diplomatic marriages'. In the course of the second millennium B.C., however, some forceful Elamite ruler occasionally succeeded not only in establishing his independence from Mesopotamian interference but also in extending briefly his influence on regions lying on the western borders of Elam. In turn, some powerful kings of Mesopotamia like Hammurabi of Babylon claimed at least partial domination of Elam. Between these highlights in the political history of Elam and Mesopotamia there were long periods in which no major military engagement is recorded between the two countries. In some measure exchange of goods and ideas certainly must have taken place between Susa and several of the rich Mesopotamian towns, especially Lagash, Larsa and Eshnunna. But Babylonian texts from the first half of the second millennium indicate a decline in the assumed large-scale trade of earlier times, when ships are thought to have plied the Persian Gulf with timber, silver and tin from [p. 45] Susa and to have returned with agricultural products such as barely and oil. [4]

The clay tablets on which historical texts were written and on which merchants recorded their business transactions mention towns and countries of Elam, but they rarely give any indication of their geographical location. Anshan, which seems at times to have been the most important region of Iran aside from Elam, may have been situated in the Bakhtiari mountains. Susa probably dominated the entire plain irrigated by the Kerkha and Karun rivers. Strong rulers of Susa probably also reigned over the green pastures of the valleys of Luristan, which would have been of great importance for the supply of the capital with sheep--and later with horses for the army.

Little is known about the internal history of Elam. The texts which contain historical information are not numerous and are not yet fully understood by modern scholars. The language, Elamite, is neither Semitic nor Indo-European, and its relation to other languages is not yet clarified.

In periods of strong influence from Mesopotamia the texts, economic and legal records, were written in Sumerian and Akkadian, that is, in the languages of Mesopotamia. There are indications that the genuinely Elamite business practice was entirely oral, so that writing need not have been an integral element of Elamite culture. [5]

It is interesting to note some evidence of similarly oral business practice in the records found in the Hurrian region of northern Mesopotamia, in Nuzi near Kirkuk. Furthermore, other characteristics common to the legal documents of both Susa and Nuzi exist. This points to a relationship also manifested in the considerable proportion of Hurrian proper names found among the princes who played a role in the political history of the country. [6] Such relations between the populations of northern Mesopotamia and Elam are also reflected in the numerous Mitannian or Hurrian-style cylinder seals found at various sites in Iran. Only at Susa and in its immediate vicinity, where local Susian seal-cutters established a distinctive Elamite tradition was there a pronounced scarcity of seals of Mitannian or Hurrian style in the excavations.

These cylinder seals which were found at Susa and at the neighbouring site of Tchoga Zanbil serve to establish a framework of glyptic art in Elam from the Old Elamite period of the early second millennium B.C., through the Middle Elamite period of the second half of the second millennium, to the Neo-Elamite period of the early centuries of the first millennium B.C. [7] Occasionally this framework may serve for the classification of larger works of art which are uninscribed and undated. For this reason we begin the discussion of Elamite art with the cylinder seals. [p. 46]

Cylinder seals were produced in great numbers in the Old Babylonian period, about 1900 to 1600 B.C. The same is true of Susa, where we call the style of the seals Old Elamite. These Old Elamite cylinders conform to the Old Babylonian ones in the ubiquitous rendering of scenes of worship, a motif inherited from the time of the Third Dynasty of Ur. But certain details characterize seals as originating in Susa: for example, a tree at the end of a scene, or the placing before the deity of an offering-table bearing a bird, or some sacrificial animal, rarely a fish. [In Babylonia the tree design is not found in seals of that period, and the deity or its image is never shown partaking of food or even receiving a food offering.] Here distinctive ritual practices of Elam manifest themselves, practices which are reflected several centuries later in Assyria. These Old Elamite cylinders are often made of the black bitumen found near the oil-fields of the region. This material can be worked very easily, and the seal-cutter could indicate the surface of an object by a series of short incisions, as in the throne and the palmtree of Figure 20. The imprint of such a cylinder seal shows ragged outlines and looks crude. A second style is smoother. The example of the latter shown here has a curious tree growing from a knoll. The branches of the tree with their leaves or blossoms not only grow upward but also point downward. This might have been a means to fill the field, but one should not rule out the representation of a candelabra-like artificial construction. Perhaps the style of this cylinder is slightly later than that of Figure 20. [8]

A Middle Elamite style of the fifteenth and fourteenth centuries B.C. is here represented by a cylinder which can be dated approximately on the basis of similar imprints found on tablets of Nuzi. [9] It shows a carefully engraved scene with several figures. One of these is marked as a deity by a horned crown and sits on a throne, the back of which ends in an animal's head. This feature can be traced to other cylinders of the group, in which the god actually sits on an animal. We encounter here the characteristic Iranian practice of decorating and enlivening inanimate objects with animal heads. Both deity and worshipper [p. 47] have narrow waists; the worshipper who carries a sacrificial goat in his arms has his hair cut 'en brosse' or swept upward, a feature often observed in renderings of Elamites. In a subsidiary scene a worshipper appears before a standing deity, and in the upper register a lion pursues a horned animal, an ancient Mesopotamian and Iranian motif which appears in many different styles until the latest periods of Iranian art. This cylinder was found in a sanctuary in Luristan where it had probably been brought from Susa.

The next stage of Elamite cylinder seals became known through Ghirshman's discovery of a deposit of such seals in chapels of the sanctuary of Tchoga Zanbil. With few exceptions these cylinder seals probably belong to the latter part of the Middle Elamite period, in which King Untashgal [c. 1265-1245 B.C.] [10] built the sanctuary. A considerable number of these cylinders, of which we give one example here, resemble the early Kassite cylinders of Babylonia, dated in the fifteenth century B.C., in the use of attenuated figures carved with thin lines by means of a fine drill. However, the scene of adoration or worship shown here is characterized as typically Elamite by the servant who holds a fan behind the throne. The shelf with vessels in the upper field and the small goblin, more human than ape-like, are also long-lived Elamite motifs. This scene of worship of a deity may not have differed much from an audience with one of the great lords of Elam.

Most of these cylinders of Kassite style, which are among the finest found at Tchoga Zanbil, and others of good quality were made of deep blue glass. Such use of glass may ultimately go back to Egyptian influence. The seals of a much cruder style, here called common style, were made of a related composition, namely faience. An example of the common style shows a simplified version of the scene just described. Here the worshipper has taken over the function of the servant with the fan. The seated figure, probably a deity, raises a vessel to his mouth. The action here depicted and the shape of the vessel are most characteristic of this group of cylinder seals.

A later Neo-Elamite version of the scene is given in Figure 25. The smaller size of the cylinder and the proportions of the figures are comparable to those of [p. 48] Assyrian cylinders of the ninth century B.C. This dating is also suggested by the pointed headgear of the seated figure found in a Neo-Elamite relief from Susa [11] and in a rock relief at Naqsh-i Rustem, which will be discussed later.

Among the cylinders with religious themes we have chosen one which shows kneeling gods surrounded by streams of water that seem to issue from their shoulders and from their arms or their hands. [12]

From the fearful and destructive world of demons which can be successfully fought only with the help of the very same demons, if these are properly manipulated, comes the winged lion-headed figure of another cylinder seal. With his bird claws the demon stands on two kneeling ibexes and with each hand raises a gazelle by the hind legs. A related theme is shown in Figure 28, but there are several reasons for suggesting a later, possibly early Neo-Elamite origin for this seal. Such a date would be important for the classification of some of the Luristan bronzes which the demon resembles in abbreviation of form and in the curved slender neck. The figures are attenuated and lack the solid verticality of the preceding seal. Moreover, the wedge-shaped fillers are not found in any other seal from Tchoga Zanbil, but they are reminiscent of the single wedges of cuneiform writing occasionally scattered in the field of Assyrian cylinders of the early first millennium B.C. Lastly, the cylinder is made of bitumen and was not found in the chapels with the other cylinders, which are mostly of faience or glass. [p. 50]

Faience continued as a favoured material for seals in Elam , and a cylinder from Susa which again shows the pursuit of a horned animal by a lion belongs even more certainly to the early first millennium B.C. Here, too, the vertical composition has given way to oblique inclination of the animal bodies, and plants of a type common in Assyrian cylinders of the ninth and eighth centuries B.C. rise from the ground-line. Of special interest is the cross, which has branches between the arms. This motif occurs in somewhat related manner on Luristan bronzes and may still be found in Sasanian textile patterns.

The next cylinder shows two griffins hovering over a creature which looks like a snake with a bull's head but which may merely present one of those curious abbreviations of animal bodies that occur in Elamite and even in Proto-Elamite art from the earliest to the latest periods. [13] The fact that the griffins have again filled out and show more rounded forms suggests that they are to be placed in the first millennium, in the late ninth or eighth--perhaps even in the early seventh--century B.C.

Two cylinders which may serve to date other works, one of faience, the other of bitumen, show horned animals flanking a tree. The simplified example in which the branches of the tree end in globules was found in one of the chapels at Tchoga Zanbil; the more elaborate version was found at Susa. It shows a tree with a crown consisting of five pointed oval leaves which remind one of the outline of date -palm blossoms. This type of tree design is typical of the late and post-Kasite [p. 51] period of Babylonia, between the thirteenth and eleventh centuries B .C., and even survived into later times. Curiously enough, there are not many examples of such a tree design at Susa, and none has so far been found at Tchoga Zanbil. However, one does find the design on bronzes from Luristan. Perhaps the frequency of the motifs should be investigated for indications of stylistic links, but this is still a task for the future.

On the basis of the chronological division into Old, Middle and Neo-Elamite periods suggested for the cylinder seals, we may now discuss some other Elamite works of art.

No architectural remains from the first half of the second millennium B.C. were observed and described at Susa, and no traces of buildings have been preserved. It is therefore impossible to form an opinion of Elamite architecture at Susa during this period.

In the minor arts, however, a definite style manifests itself, a style characterized by the use of animal bodies and animal heads on vessels and other objects. Our plate shows the foot of an object carved in bitumen [14] with the foreparts of an ibex whose head and neck are worked in the round. Nose and beard of the animal are broken off; nevertheless, the animal sculpture is quite expressive, which is in part due to accentuation of the eyes with white shell inlays. Hair is indicated by rows of sharply engraved and short, often slightly curved lines. The style is reminiscent of the Old Elamite cylinder seals which are also made of bitumen and which show similar rows of incised lines to suggest surface patterns.

The same workshop which produced the object just discussed may have also made the ram-headed bowl found in a coffin between the hands of the deceased. [15] The sides of the bowl represent in side view the extended body of the ram whose neck and head are carved in the round at one end of the bowl. The position in which this vessel was found suggests that such vessels decorated with animal forms lent themselves well to ritual purposes.

The finest example of a bowl of this type was found in the northern Mesopotamian town of Ishchali, in the Diyala valley. [16] Three recumbent ibexes, their heads and necks turned at right angles to the rest of the body and partly worked in the round, were originally carved along the circumference of the bowl, but only one of the animals is preserved. The body is simplified to almost geometric forms, and the hair is stylized in rows of hatchings running in opposite directions like a herring-bone pattern. Although minor differences can be observed between the bowl from Ishchali and the bowl and fragmentary foot of an object [p. 52] from Susa here reproduced, the existence of a large number of such vessels in the finds excavated at Susa, as against the unique example from Ishchali, and the material of these vessels, which is typical of works of art made at Susa, offer sufficient indications on which to postulate the origin of the group in Elam. The appearance of the single piece in Ishchali can be explained by the fact that one of the trade-routes went from Susa to Mesopotamia over Kurdistan, the route nowadays crossing through Kermanshah and into Iraq over passes that lead into the Diyala valley. [17]

If our Elamite bowls are correctly dated in the time of the dynasties of Isin and Larsa in Babylonia, that is, in the twentieth and nineteenth centuries B.C., when a very naturalistic style prevailed in Babylonian art, the abstraction noted in the decoration of these bowls must correspond to a specifically Elamite taste at that period. A distinctively Elamite feature may also be the way in which the head of [p. 53] the animal decorating the bowl is turned at right angles to the body in relief and carved partly or entirely in the round. This device for enlivening the decoration of a bowl was used in Mesopotamia only in the time of the early urban development shortly before and after 3000 B.C. Perhaps its use at Susa in the Old Elamite period--and later--shows that there this early device was retained with the same tenaciousness which characterizes the retention of earlier features in the artistic production of Iran through the centuries.

Two metal objects dated by the excavator of Susa, De Mecquenem, in a period corresponding to our Old Elamite period, present other features which are not Babylonian and may therefore be Elamite or generally Iranian. A small golden falcon with spread wings has its claws pulled up close to the body as if the bird were seen in flight from below. The representation of the falcon with short thick neck and short beak is a characteristic of later Iranian art, as is the position of the legs. For example, a cauldron attachment from Hasanlu, made about one thousand years later, shows a bird of the same type, similarly positioned. The wings and tail of the bird from Susa seem to have been made in two pieces, of which the lower was a flat plate while the upper one was made in open-work. Together the two pieces formed cells for an inlay of a blue composition. [18] The technique in which cells or cloisons of gold or some other metal are made to hold inlays of some material like blue lapis lazuli, red carnelian or white shell was known quite early in Western Asia, as shown by finds from the Royal cemetery of Ur. [19] An even earlier origin was postulated for small pieces of jewelry, decorated in the same technique, found in a child's grave at Susa, but the date of that jewelry seems somewhat uncertain. [20] Nevertheless, there can be no doubt about the fact that the technique of decoration used for the falcon from Susa was known in Western Asia before the Old Elamite period. The blue composition used for the inlays of the falcon, however, is not attested for the early periods and may indicate that the object should not be dated before the second half of the second millennium B.C.

The second object supposedly of Old Elamite date is a socketed head of a bird or reptile made of silver, perhaps the top of a standard. Within the opened mouth of the creature a pattern of scales can be seen, [21] which makes it more likely that this is the head of a reptile, perhaps a tortoise. The 'tail' shown at the back of the head may imitate in metal the coloured cords, rolled up at the ends, used on standard-tops made of impermanent materials. [22]

No comparable work from the early second millennium B.C. is known from Western Asia. The plain geometric forms, given almost demonic life by the large eye peering out from under the thick brow, are comparable rather to later Iranian renderings of animals. It is possible that the head is incorrectly dated and that it was made only after the middle of the second millennium B.C., when Mitannians and Hurrians ruled in northern Mesopotamia and Kassites in the south and the prevalent taste favoured a geometric style there and also in Elam--as shown by the large number of Mittanian or Hurrian cylinders found in Iran and by the existence of at least one fine cylinder seal in Mitannian style with an Elamite inscription. [23] It is also possible, however, that the standard-top really belongs to a time before 1500 B.C. and that it prefigures the geometric style of a later age. This would mean, however, that the geometric style had its inception with standards and similar pieces and that it originated in Persia.

The Middle Elamite period is the only period of Elamite rule which has yielded [p. 54] coherent architectural remains--the sanctuary of Tchoga Zanbil a few miles from Susa. Here Ghirshman excavated a ziggurat and surrounding chapels and temples as well as a palace and various interesting installations. [24] The ziggurat, a temple tower, rises like a massive mountain from the flat and empty plain. Once fields and gardens probably surrounded the sanctuary and supplied the priests and other employees with grain and vegetables. The fact that the soil can be made exceedingly fertile by artificial irrigation is proved by an American and Dutch project for growing sugar-cane in fields located only a few miles from Tchoga Zanbil. In Tchoga Zanbil and Susa the temperature rises to 140 degrees in the summer. But in the desert the heat is dry and bearable in the shade, and nights are even cool. The great question whether or not the ancient inhabitants of Tchoga Zanbil and Susa sought out the cool and refreshing air of the mountain valleys of Luristan during the summer months cannot be answered today because we have too little evidence. The tendency of the present-day inhabitants of the towns of Khuzistan is to seek protection from the mid-day heat in subterranean rooms and to emerge only in the evening. Only the nomads wander with their herds of sheep and goats into the mountains in summer and return in winter to the plain around Susa and Tchoga Zanbil. [p. 55]

The sanctuary of Tchoga Zanbil was separated from the surrounding plain by an outer wall which measured 1200 x 800 metres. An inner wall enclosing the ziggurat and its courts measured 400 x 400 metres. It was pierced by seven gates of varying importance, all leading to the courts of the ziggurat. It is this inner wall which appears in the photograph in Plate 11; above it rise the three storeys of the ziggurat which remain of the original five-storeyed building. The single storeys look like square terraces built one above the other; in reality each storey rises directly from the ground. According to the excavator, R. Ghirshman, the two outer lowest storeys were first built around a central open court. Subsequently the higher storeys were concentrically encased in this court, with the highest storey rising in the middle.

In the second storey from the base several rooms were built behind the southeastern façade. Ghirshman interprets the complex as the lower temple [25] in which the god Inshushinak, to whom the entire Ziggurat was dedicated, was worshipped during the day. At night the god was thought to return to heaven, perhaps striding with gigantic steps up the ziggurat to the top, where a small temple is assumed to have stood. From this point the god would have ascended to heaven, and here he would have landed again in the morning. All this has been deduced from Mesopotamian parallels, however, and may or may not apply to Elamite beliefs. [p. 57]

The few human beings, priests and dignitaries who were admitted to the upper storeys would have had to climb narrow stairs with very high steps, partly covered by brick vaults and partly open to provide light for the stairs. The visual impression of the ziggurat was mainly determined by the horizontal lines of the terraces and by the regular alternation of salients and niches which formed the principal decorative elements here, as in the mud-brick decoration for which no earlier prototype is known is a triple-arched niche found to have decorated a round platform which had four such niches. [26] Two other platforms were found at Tchoga Zanbil, but they were too badly damaged to show any details. The curvature of these architectural forms contrasts strikingly with the prevalent rectangular forms employed in Babylonian architecture.

The principal gate which gave access to the courts surrounding the ziggurat was situated on the south-eastern side of the complex. It was called the 'Royal Gate' by the excavator both because of its large size and because of its decoration of glazed bricks and 'nails' with pommels holding flat tiles in place. The pommels bore the name of the builder of the ziggurat, Untashgal. They were covered with blue glaze like the tiles, which also had a restrained decoration of quarter-rosettes in the angles. Other tiles were decorated with disks of white and black glass in various sizes. Some of the bricks were blue, others green; some had circles with a white or blue centre, or white lozenge shapes on a lapis-lazuli blue background. The strong colour of these glazed bricks and tiles must have given a very festive air to the 'Royal Gate' as seen from the ziggurat. [27]

A new site for rock reliefs was chosen by the Sasanian king Ardashir II [379-383] in the last quarter of the fourth century. He had a hieratically stiff scene of investiture carved in a rock in Kurdistan, at a place today called Taq-i Bustan, near Kermanshah, not far distant from the mountain of Bisutun. There also Shapur III [383-388], Ardashir's successor, had a relief made at the back of an iwan which he caused to be carved in the rock. Shapur had himself represented beside his great father Shapur II [310-379] in a pictorial documentation of the legitimacy of the son, who had had to fight for the throne. The documentary intent of these reliefs is fully compatible with their rigid composition and with the stiff rendering of the bodies, which merely provide surfaces for the decorative patterns of the material and drapery of the robes.

The iwan with the relief of Shapur III was included almost a century later in plans for a monumental triple iwan, of which the left wing was never completed. In the tympanum of the large new and supposedly central iwan an investiture scene was carved which resembled in its scheme of three figures the composition of the relief of Ardashir II, but corresponded in style to the heavy block-like pair of figures made for Shapur III. Below this relief of an investiture is one of the most impressive works of Sasanian sculpture, a king in full armour on horseback. For this figure one may cite the description of Ammianus Marcellinus: 'Moreover, all the companies were clad in iron, and all parts of their bodies were covered with thick plates, so fitted that the stiff joints conformed with those of their limbs; and the forms of human faces were so skillfully fitted to their heads that, since their entire bodies were covered with metal, arrows that fell upon them could lodge only where they could see a little through tiny openings opposite the pupil of the eye, or where through the tips of their noses they were able to get a little breath.'

In front of the entrances to the ziggurat itself stood pairs of animals of which a long-legged, slender humped bull whose head and legs were reinforced by copper rods was reconstructed by Madame Ghirshman. [28]

Before the principal entrance to the ziggurat, on the south-eastern side, two rows of sacrificial tables were made of big bricks forming low truncated pyramids. Between the last two tables there was a pit which the excavator assumes was meant to receive the blood of the sacrificial animals. Nearby he found an installation which he interpreted very convincingly as intended for liquid offerings, and facing both this installation and the pyramidal offering-tables were three large square tables made of baked brick placed against the enclosure wall; beside them was a large vessel probably intended for ablutions. [29]

We cannot judge to what extent these ritual, architectural and decorative elements so far described are typically Elamite because no Mesopotamian ziggurat of the same period is equally well preserved and shows so many details. A unique element, however, seems to be formed by the vaulted chambers found on three sides of the second storey of the ziggurat. Each chamber was accessible by a stairway, and no chamber communicated with the next. Some chambers formed a second inner row parallel to that of the rooms lying against the outside wall of the ziggurat. These inner chambers had been carefully filled up with bricks at the time when the core of the ziggurat was built in the inner court. Some of the outer chambers contained tiles and enamelled bricks as well as other materials which were doubtless used during the building of the sacred [p. 58] complex of Dur Untash, as Tchoga Zanbil was called in ancient times. The excavator doubts, however, that this was the function intended for these chambers, which are carefully finished and painted white with lime. He thinks that they were store-rooms for offerings, and he also does not rule out the possibility that they were intended for royal tombs. That they were obviously used only during the building of the sanctuary speaks for the fact that the latter was not completed.

Opposite the south-western façade of the ziggurat several chapels were erected in which votive offerings were deposited, for example, more than one hundred cylinder seals. Furthermore, there were several groups of temples in the vicinity of the ziggurat, nine in all, dedicated to different gods of the Elamite pantheon. The most interesting structure aside from the ziggurat, however, is the palace, in the plan of which the position of the five tombs deep under ground level was obviously considered. The palace included a banqueting court and rooms which seem to have been pantries, to judge by the vessels which they contained. No living quarters, however, suggest themselves in the plan of this palace, which was obviously a hypogeum, perhaps comparable to the royal sepulchers at Ashur to which the Assyrian king Sennacherib [704-681 B.C.] probably referred when he spoke of the 'palace of repose, the eternal abode'. [30] The tombs at Tchoga Zanbil were vaulted with baked brick set in bitumen and a quick-setting cement-like plaster. The use of bitumen in place of lime mortar seems to have been specifically Elamite; in Mesopotamia this material was as a rule employed only for installations in which seepage of water was to be prevented, such as bathrooms and drains. This cannot have been the case here because the deceased were cremated and weapons were brought to the tombs. Yet Ghirshman has stated that in those cases where the tombs were intact the remains of two cremated persons could be distinguished. Only one skeleton was found which had not been cremated; it was lying on a bed-like platform beside the remains of two cremated corpses. Since the Elamites, like the Babylonians, were buried, not cremated, this evidence indicates that a different, perhaps western, custom prevailed in the royal house. One is reminded here of the cremation of Hittite kings and of one of the kings of Mitani. Nowhere else, however, is there evidence of the wife accompanying her husband into death--as the remains of two people in each of four instances at Tchoga Zanbil strongly suggest.

We must state here, moreover, that so far no inscription has been found which would definitely identify the tombs and the cremated corpses with the kings of Elam.

After the discussion of the ziggurat and the hypogeum one may view with special interest the unique model depicting a ceremony which might have taken place among surroundings resembling those of the ritual installations before the principal entrance of the ziggurat at Tchoga Zanbil. Most characteristic are the two rows of small knolls which closely resemble the pyramidal offering-tables excavated by Ghirshman. Two stepped structures in the model may be abbreviated renderings of ziggurats or large offering-tables in architectural form. The model was found at Susa, in one of two vaulted rooms which had been despoiled in antiquity but which could have been tombs like those described above. [31] The model was completely encased in gypsum so that it appeared in the masonry of one of these vaulted rooms as a large white tile with green spots of oxidization on the surface. Subsequently the gypsum was carefully removed and the entire [p. 60] scene appeared, though the foliage of the trees and other delicate details were forever lost. According to the inscription the model represents a scene at sunrise and was made for king Shilhak-Insushinak [1165-1151 B.C.]. [32] This is the only scene in Western Asiatic art preserved in three-dimensional form. It undoubtedly represents a ritual ceremony which would have been recorded in written form elsewhere in Western Asia.

The execution of the model shows that attention was paid to natural proportions but that the forms were greatly simplified. The bodies of the two men, for example, are rounded but show no subtle modelling of the surface. Tentatively we shall henceforth use this simplified rounded style of the model as a criterion for dating in the time of Shilhak-Inshushinak, that is, in the twelfth century B.C.

The two greatest works of Elamite sculpture, however, belong to a slightly earlier time. They are the bronze statue of Queen Napirasu, wife of Untashigal, in the Louvre and the copper head of an Elamite in the Metropolitan Museum. The upper part of the statue of Napirasu can almost be inscribed within a rectangular block, while the skirt which covers her lower body has the form of a tall, very slightly flaring bell. A pattern of small circles enclosing a dot covers the upper part of the garment; long fringes decorate the upper part of the skirt at the back and are brought forward to frame on both sides a rich pattern of columns filled with alternating strips of zigzag and hatching and bordered by zigzag lines, almost suggesting a derivation from patterns of architectural origin. All these lines run horizontally or vertically except for the softly undulating fringes at the bottom of the skirt. In the absence of the statue's head, its hands attract the viewer's attention even more than might have been the case in the original state. Even in the original state, however, these hands with their long and slender fingers, of which only one is adorned with a ring--and their quiet pose--must have imparted a certain calm and elegance to the statue.

It is not known how the statue was made. All that can be said here is that teh surface details just described indicate the delicacy of the execution; at the same time the exposed parts of the bronze interior--in which rods of an armature seem to be recognizable--suggest that the metal was not poured at one time. One may therefore doubt that the figure was hollow cast and then filled up, as stated in the original publication. [33] That this was not the practice of Iranian metal-workers is indicated also by the great head in the Metropolitan Museum, which is cast solid [Fig. 38]. The features have probably been coarsened by the disintegration of the copper. Thus the eyelids may seem heavier now than they originally were; the nose, which seems so thick as to suggest a feature characteristic of some individual, may again have been accidentally enlarged. The mouth, however, with the beautifully curved lips and the heavy moustache, can be fully appreciated even now. [p. 61]

The head is distinguished from Babylonian sculptures by the rendering of the eyebrows, which do not meet over the nose but terminate on either side of it in a sharp oblique line. This sets the eyes further apart and gives a distinctive cast to the man's face. Lines on the forehead were probably intended to indicate his mature age. The patterns formed by the curls of the beard as well as the ornaments of the asymmetrical headgear, which consists of a chequered piece of cloth over which ribbons are wound, stress the contrast between the inanimate materials and the face, which thereby seems more alive, more human than it actually is. The dating of this head was much discussed. It may belong to the second half of the second millennium, but it may also have been made somewhat earlier. [34]

The head has been called 'Head of an Elamite', which raises a problem since the object is said to have been found in Azerbaijan. It is possible that there existed in Azerbaijan a centre in which workshops produced objects related to those of Elamite style, but so far no archaeological expedition has discovered sculptures of such quality outside Susa.

The heads of two small figurines found at Susa, one of gold with a considerable amount of silver, occasionally called electrum, the other of silver, each carrying a sacrificial goat, somewhat resemble the 'Head of an Elamite', though they are more simplified and also show differences in detail. In the faces of the figures the eyes seem to be placed somewhat obliquely, an impression also created by the copper head. [35] The eyebrows of the figurines meet above the nose, which differentiates them from the copper head; instead of describing semicircles as in Babylonian sculpture, however, the eyebrows of the figurines are much straighter, so that they somehow create an effect related to that of the eyebrows in the copper head. Furthermore, the straight nose of the silver figurine may resemble that originally given the copper head, while that of the gold figurine is broader. Despite minor differences, eyes, eyebrows, nose and also the generous moustache worn by the persons portrayed seem to render a related ethnic type.

We can only guess that the large copper head belonged to a royal statue, but we are on more solid ground with the figurines, which surely represent a king in constant prayer and sacrifice before his deity. The dedication in a temple of a valuable statue of the ruling king was frequently used in Babylonia as an event after which the entire year was named. [36] No text, however, unequivocally mentions the dedication of a gold and a silver statue. If we are right in assuming that an Elamite king dedicated two such images of himself, this would be a specifically Elamite practice.

The two figurines and a whetstone topped by a feline head in gold were found buried near the large group of objects deposited under the temple of Insushinak, [37] built in the twelfth century B. C. by the Elamite king Shilhak-Inshushinak. One wonders, however, whether objects of such great value were really deposited in the foundations of a temple. In general such figures of gold or silver [p. 62] representing the king as an offerer not only served a pious purpose but also had a value in ostentation that should not be underrated. This fact also speaks against De Mecquenem's later assertion that the gold statuette and the objects associated with it were originally funerary deposits. [38]

While it is therefore not certain that the place where these figures were found indicates a date before or in the time of Shilhak-Inshushinak, this is likely on stylistic grounds. The hair style and costume of the figures, the latter strewn with dots [reminiscent of the small circles which cover the garments of Napirasu] and bordered with the short fringe at the bottom, and the precious material point to a date in the latter part of the second millennium B.C. rather than to the first millennium. It is possible that the large head was made about the time of the figurines; as mentioned above, however, a somewhat earlier date cannot be excluded.

The golden lion's head of the whetstone, found together with the figurines and reproduced with them in Plate 12, shows greater stylization than the human figures. The formation of the leonine features, especially the large eye with the upper lid raised almost to a point and the beaded patterns of muzzle and mane, as well as the rendering of the lips by a continuous cord-like line, manifests a definite style particularly suited in its simplicity to the decoration of tools. [39] A more rounded and naturalistic rendering of animals is found in a rein-ring with two ibexes standing on either side of a tree. This motif is found in related form in an Elamite cylinder seal, and it seems likely that the rein-ring shows a transposition of the more or less two-dimensional form of the cylinder into the three-dimensional one of the rein-ring.

The shape of the rein-ring continues one which was developed in the Early Dynastic period of Mesopotamia and is documented in the Royal Cemetery of Ur. Its survival in Elamite times would not be surprising in view of the tenacity of earlier tradition in the art of Iran. [40]

The reliefs of the Middle Elamite period offer several examples of survival of earlier motifs. The most important of these reliefs is the great stele dated in the thirteenth century B.C. by an inscription of Untashgal. While the king and his consort, Napirasu, and her mother appear to have been rendered in a flat and stiff manner, at least as far as the remaining parts of the figures permit one to judge, [41] greater interest is created by the varied outline and textures of the demon with moufflon's or ram's horns, who may be derived from the demon of prehistoric Iranian stamp seals and who may be linked to demons on bronzes from Luristan discussed below. One wonders whether the demon's beaked profile with receding forehead and chin, not found in other Elamite sculptures of human figures, could have been meant to characterize an inhabitant of the mountainous regions bordering Elam. [42] The profile of the demon seems more exaggerated than that of the water goddess of the register above. As on a cylinder seal from Tchoga Zanbil, the goddess seems to grasp streams which flow from bases but which may have been meant to originate in the fins that replace the feet of the water goddess. Like the figure of the demon, the water goddess may reflect earlier iconographic traditions, such as the rendering of the watercourses in rope-like form which can be seen on the steatite vase from Khafaje of the early third millennium B.C. In turn, much later concepts concerning a vital fluid [43] could have been influenced by renderings like those of the stele. While the stele of Untashgal was buried in the ruins of Susa after the destruction of the town by Ashurbanipal, some rock reliefs which I would tentatively assign [p. 65] to the Middle Elamite period survived and may have preserved motifs of Elamite iconography for a later time. The best preserved of these reliefs is the one on the rock of Kurangun in the Bakhtiari mountains, several hours northwest of Shiraz on a high cliff seen from afar. In the main scene, which is enclosed in a rectangular frame, a god sits on a throne formed by the coils of a serpent which he holds by the neck. He also holds a vessel from which two streams of water flow. One stream forms a canopy over the god and a goddess behind him and is probably caught in a vessel held by an attendant. The other stream flows toward the long-robed slender worshippers approaching the deities. A large number of squat pig-tailed figures in short kilts are carved on the rock as if descending toward the principal scene. There is a considerable difference in style between these figures and those of the main scene, which has been explained by assuming that this scene was re-cut at a later time than the procession of worshippers. It is not possible to be definite about this, however, or to fix the date of the main scene with any certainty. One can merely say that a god with a flowing vase first occurs in the Akkad period [c. 2370-2230 B.C.] but that the motif of the flowing vase survived in varied and extended form in the Middle Elamite period, as shown by the examples on the stele of Untashgal. It is not impossible therefore that the relief of Kurangun was made in the middle or even in the latter half of the second millennium B.C. [44]

The principal scene of the relief of Kurangun was copied several centuries later at Naqsh-i Rustrem, but this relief was almost completely eliminated by a relief of the Sasanian king Bahram II [A.D. 276-293]. Only the two figures at either end are well preserved; they probably represent the Elamite king and the queen who had the relief made. The king wears the pointed headgear typical of the Neo-Elamite period; [45] the queen wears the battlemented crown that is also worn by the Assyrian queens of the seventh century B.C. [46][p. 66]

To the same Neo-Elamite period as the relief of Naqsh-i Rustem also belongs the relief of a woman spinner from Susa, here reproduced to show the survival from Early Dynastic times of female figures in a seated posture with what must have been crossed legs. Another survival is shown in the servant with a fan of the same rectangular shape as that seen on Middle Elamite cylinder seals. The rounded relief of the piece brings out fully the artistÕs delight in observation of details of dress and jewelry, of hair style and furniture. None of the reliefs, however, approach the expressiveness found in the metal sculptures. One would like to conclude from this that the Elamites were principally metal-workers who favoured more than other techniques that of modelling in wax in preparation for casting. [p. 67]

Elamite work in glazed earthenware and faience is related to this technique of modelling in a soft material and was of equal if not of greater importance than metal in the decoration of Elamite palaces, temples, and probably also more modest homes. Unfortunately, most of the earthenware and faience [47]objects have lost their brightly coloured glaze and therefore look dull. Painted and glazed earthenware tiles, objects of faience, and glass all appear at the same time in the Middle Elamite period. [48] Glazed faience and glazed earthenware were widely produced throughout Western Asia around the middle of the second millennium B.C. In Nuzi, near modern Kirkuk in northern Mesopotamia, large animal sculptures were produced in the latter material, and the same was done in Tchoga Zanbil. [49] Outside Tchoga Zanbil glass is found only rarely and glazed tiles are not found at all at this time. They appear only later in Ashur, possibly under Elamite influence. [50] Conceivably there was in Elam a desire for and interest in production of cheap 'new' materials which had even greater brilliance and intensity of colour than the far more expensive natural stones like lapis lazuli. One may also think of the recurring indications--however slight--of direct contact by sea with Egypt, where glass was invented, probably at the beginning of the Eighteenth Dynasty, and where the Elamites might have obtained formulas for producing glass. At any rate, the deep blue colour which they used for their cylinders is not used at other Western Asiatic sites.

Glazed tiles began to be used in Elam in the time of Untashgal, as is shown by the excavations of Tchoga Zanbil [see p . 58 above]. The tiles are blue, green or white, or have two of these colours combined, but they are not multi-coloured, nor do they have figured representations. Tiles which show such features surely mark a later development. Fragments of such tiles were found at Susa with glazed earthenware nails inscribed with the name of King Shilhak-Inshushinak [c. 1165-1151 B.C.] and also with glazed tiles of green and yellow colour which bore the name of Sutruk-Nakhunte [c. 1207-1171 B.C.]. This evidence shows that there was in the twelfth century B.C. a further development in the decoration of glazed tiles. [52] We therefore date to this time a tile which shows a bird-footed demon standing on two griffins and probably holding two others in his hands. The composition resembles that of a cylinder from Tchoga Zanbil, Figure 27; the colours are strong, the forms carefully defined.

The second fragment represents a different style. The drawing is cursory; the forms are thin and pointed. The neck of the bull on the fragment rises sharply, [p. 68] then the line runs horizontally to the sharp bend of the horn. In contrast to this the curve in the neck of the griffin on the first fragment is smoothly rounded and terminates in the slight countercurve of the griffin's crest. Moreover, the colour of the second fragment is much duller than that of the first, although here accidental factors may have played a role in changing the original colours of the tile. [53]

The two styles represented here can scarcely belong to the same period. The cursory one is probably later and may be tentatively assigned to the tenth or ninth century B.C., which means that it would be Neo-Elamite. Another fragment from Susa which seems to belong to this later style shows a similar cursory treatment of the guilloche pattern. The rendering of the horned animal in this third fragment is interesting because it shows a curvature of the neck and an abstract division of the body which are reminiscent of the Luristan bronzes to be discussed below. The style here represented also seems to be reflected in Elamite cylinder seals, for the bull in the tile recalls the slender bulls with sharply bent horns in Figure 28. Moreover, the uncertain position of the bull's forelegs is comparable to the insecure postures of the demon and the animals in the cylinder seal. Such painstaking comparisons of small details are necessary in order to obtain some tentative outline of the development of Elamite art at the beginning of the first millennium B.C.

Somewhat later works of Neo-Elamite art can be recognized in several faience vessels and objects hammered in bronze, called repoussŽ work. The faience vessels are small jars similar to the small round boxes of ivory made in Syria in the ninth and eighth centuries B.C. One of them even has the same pattern of rosettes and double lotus-blossoms found in the diadem of an ivory head of the eighth century B.C. excavated at Nimrud. [54] These Syrian ivories probably indicate the date at which the Elamite jars were made. The jars rarely have a figured decoration, but one example has designs in flat relief and also heads worked in the round in place of the bosses seen on other jars for fastening the lid to the vessel. The reliefs show on one side a griffin and on the other a winged and bearded sphinx. The horned mitre of the sphinx has a curious shape; the horns are bent forward with the tip pointing down. At the back of the cap there is an inexplicable oblique strip, like the end of a band. Above the cap rises a globe. These three characteristics are found again on the horned mitres of bull-men seen on a bronze quiver of which the lower half is reproduced here. Again a small detail has led us from one work of art to another. The fact that we are not mistaken in assuming that the faience vessel and the bronze quiver belong to the same general style is indicated by the composition, which shows rosettes and other fillers placed in the larger intervals between the figures. In both the jar and the [p. 70] bronze, the figures are characterized by fleshy noses, large eyes and round mouths. Perhaps one may relate this striking facial type and the Syrian elements in this stylistic phase of the ninth to the seventh century B.C. with the presence of many Arameans in Elam at the beginning of the first millennium B.C.

A group of Neo-Elamite bronze objects which should be associated with the bronze quiver and the ivories are the so-called situlae, slender buckets of bronze with a conical base. The shape of these vessels was originally Egyptian, like the lotus pattern on the base. The style of the decoration of this group of situlae, however, is Neo-Elamite. [55] Our example shows a winged ibex with bearded human face beside a tree. Although the heavy forms and the fleshy nose of the demon remind us of the style of the quiver, the lines of the situla are nevertheless somewhat more fluid and the space is not as tightly filled. Conceivably this [p. 71] marks a slightly later date for the situlae, but it is also possible that these variations are due merely to different workshops.

The situlae and the quiver were usually dubbed 'Luristan' because no comparable object of bronze has so far been found at Susa, while many bronzes of different styles were found in the mountainous region of Luristan, north-west of Elam. The reason why no metal objects of the early second millennium were found at Susa is surely to be found in the thorough pillaging of the town by the Assyrians in about 640 B.C.

It is more difficult to explain how Elamite bronzes came to Luristan. They could have been pillaged by robbers from Luristan, for the people of these mountain regions must have been as difficult to control in ancient times as they have been until very recently. It is also possible that Elamites seeking refuge in Luristan from Assyrian attacks brought some of their valuables with them. Modern tourists who know the searing heat of Susa in summer like to assume that the Elamites moved to the cool mountains in these months. Reasons against such a hypothesis are the above-mentioned custom of the modern townspeople to seek protection from the heat in subterranean chambers and, secondly, the disinclination of the inhabitants of ancient towns to venture into such dangerous regions as open and mountainous country. It is possible, however, that the Elamite rulers did not belong to the same urban tradition as their townspeople, if we are right in ascribing to the kings burial rites which involved cremation in contrast to inhumation in various types of graves for the townspeople. The royal family and the court may therefore have followed a different pattern of life from the common people and may have enjoyed hunting in the mountains during the summer. For such royal hunting parties valuable objects could have been brought to Luristan and kept there. [p. 72]

Lastly, it may be suggested that there were sanctuaries in the mountains of Luristan in which Elamite works of art could have been deposited as offerings, either by Elamites themselves or by people who obtained Elamite objects for votive purposes. This possibility seems the most likely for two reasons: one, the objects which have been found, such as bronze coverings of quivers, were not objects of actual use, since it has been shown quite conclusively that bronze sheathing offers poor protection in comparison to leather [56] and two, a sanctuary with votive objects was actually excavated near Surkh Dum in Luristan. [p. 73]





1. A translation with discussion of the text was given by A. Falkenstein, 'Die Ibbísìn-Klage,' Die Welt des Orients [1950], pp. 377-384. I owe the Engish translation of the passage to D. O. Edzard.

2. For a discussion of the removal of the Ningal statue, see D. O. Edzard, Die 'sweite Zwischenzeit' Babylonians [Wiesbaden, 1957], p. 57.

3. For a discussion of these statues, see E. Strommenger, 'Das Menschenbild in der altmesopotamischen Rundplastik von Mesilim bis Hammurapi,' Baghdader Mitteilungen I [1960], pp. 72-74.

4. For a summary of the relations between Elam and Mesopotamia from the end of the third millennium B.C. to the beginning of the second, see Hinz, 'Persia . . .' [op. cit. in note IV/1], pp. 12-20; for the imports from Susa to Mesopotamia, see op . cit., p. 4. For the exports of barley and oil from Mesopotamia to Elam, see Leemans, op. cit. in note I/7, p. 116. The shrinking of trade can be deduced from the evidence presented by Leemans, especially p. 175.

5. The probable oral nature of the local business and legal practices in Susa was discussed by L. Oppenheim in a study of the legal records from Susa and their relations to records from northern Mesopotamia, from the Hurrian town of Nuzi, and from Assur; see 'Der Eid in den Rechtsurkunden aus Susa,' Wiener Zeitschrift für die kunde des Morgenlandes XLIII [1936], pp. 242-262.

6. R. Labat, 'Elam, c. 1600-1200 B.C., ' CAH II/XXIX [1963], pp. 4-6, stressed the ethnic changes which occurred in Elam toward the end of the First Dynasty of Babylon, in the seventeenth and sixteenth centuries B.C., and pointed to the likelihood of a considerable proportion of Hurrians among the newcomers.

7. The division into an Old, Middle, and Neo-Elamite period is based on a terminology suggested by H. H. Paper for Elamite texts; see his 'Elamite Texts from Tchogha-Zanbil,' JNES XIV [1955], p. 44. Paper merely made this suggestion for Middle Elamite texts, suggestng the use of Old Elamite for the 'proper names and isolated words and phrases embedded in Akkadian and Sumerian texts from Susa and elsewhere' before the beginning of Elamite literature. The term Neo-Elamite is introduced here, paralleling the general use of terms Neo-Assyrian and Neo-Babylonian.

8. A large group of Old Elamite cylinders was published by Delaporte, Louvre I, Pl. 34:2-10, which show the motif of a worshipper before an enthroned deity, with a vessel, an offering table, or a bird between the figures; Pl. 34:11-15, 17-21, and Pl. 35:1, 2, which are also Old Elamite but different in theme. Old Elamite cylinders with various themes were also published by M. Rutten in Revue d'Assyriologie XLIV [1950], Pl. IV: 35, Pl. V: 37, 40-51, Pl. VI: 52-54, 56.

9. For a discussion of imprints of Elamite style on tablets from Nuzi, see E. Porada, 'The Origin of Winnirke's Cylinder Seal,' JNES V [1946], pp. 257-259.

10. The discussion of the cylinder seals from Tchoga Zanbil and the drawings of a few examples are excerpted from a manuscript on this material which I am preparing for publication at the invitatin of R. Ghirshman. Drawings of cylinders from Susa were made by Paul Lampl after impressions which I was able to make at the Louvre with the kind permissin of A. Parrot. The dates of Untashgal and other Elamite things are given according to G. G. Cameron, History of early Iran [Chicago, 1936], pp. 230-231. R. Labat in 'Elam,' CAH, II/XXIX, pp. 3-13 and inside back cover, does not commit himself to absolute dates for the length of the rulers' reigns.

11. The Neo-Elamite relief from Susa showing a prince wearing headgear of a pointed type is best reproduced in Encyclopédie photographique de l'art I, p. 274, where it is erroneously dated in the end of the second millennium B.C. The correct date: 653-648 B.C., for Addahamiti-Inshushinak, the king represented, is given by Debevoise in JNES I [1942], p. 84.

12. That this was indeed the intention of the seal-cutter and not the representation of figures bound by the water-courses, as one might also think, is proved by the representation on a gold bowl in the Archaeological Museum, Teheran, reproduced in Archaeology 17 [Autumn 1964], p. 200.

13. For Proto-Elamite abbreviated animal forms, see Amiet, Glyptique, Pl. 32, Fig. 516; Pl. 35, Fig. 550.

14. What we call here bitumen is referred to as rock-asphalt in C. Singer et al., A History of Technology I [Oxford University Press, New York, London, 1954], p. 256, caption to Fig. 161.

15. The leg of an object with an ibex, Plate 8 above, was published by De Mecquenem in a drawing in MDPXXIX [1943], p. 111, Fig. 83:4. The animal-shaped legs of a bitumen basin from Susa, EncyclopŽdia photographique de l'art I, p. 248 C, show how the leg here reproduced was probably attached to a vessel. The bowl, Plate 8 below, was published in MDP XXV [1934], Pl. XII: 1. De Mecquenem said about it [op. cit., p. 211] that it was found 'A l'intérieur du sarcophage... pour l'utime purification des mains du mort . . . ' I have deduced from this statement that the object was found between the hands of the skeleton.

16. For a reconstruction of the vessel, see OIG 20 [1936], p. 100, Fig. 79.

17. This route seems to have been used even by Babylonian merchants when political disturbances interfered with the passage of valuable goods through central Mesopotamia; see Leemans, op. cit. [in note I/7], p. 171.

18. The falcon was published in MDP XXV [1934], p. 210, Fig. 53:3. The blue inlays, unfortunately, seem to have disappeared between the summer of 1960, when I saw them and photographed the object which was then still intact, and the summer of 1962 when they were no longer there. I was told that the objects in the hall had been moved about a great deal and we may presume that the inlays were lost at that time. The inlays were referred to as enamel by the the excavator, but they seemed to me to be of the powdery blue composition called 'Egyptian blue', described by F. R. Matson in Persepolis II, pp. 133-135. A small head which was found in the same group of graves as the falcon was made of the same composition; see ibid., Fig. 53:14.

19. See especially the rings [C. L. Wooley, The Royal Cemetary [Ur Excavations II, 1934], Pl. 138, U. 10878, U. 9778] but also a circular penant [ibid., Pl. 133, U. 8565] and even the silver hair ornaments inlaid with gold, sheel, lapis lazuli and red limestone [Pl. 136, text p. 240] which are decorated in this technique.

20. The jewellery was published in MDP XXIX [1943], p. 15, Fig. 12, and dated by Le Breton in the time of Susa Cb [Iraq XIX, 1957, p. 109, Fig. 27]. This date seems to have been based on the position of the tomb rather than on its contents, which should be studied more carefully before one can be certain of their date.

21. Unfortunately irregular lines were scratched on our plate; the regular pattern of the original is rendered in the drawing MDPXXV [1934], p. 210, Fig. 53:15.

22. Similar cords transposed into bronze are seen on bronze hammers, of which one has an inscription of King Shulgi of the Third Dynasty of Ur [c. 2097-2051 B.C.]; see J. Deshayes, 'Marteaux de bronze iraniens,' Syria XXXV [1958], p. 287, Fig. 3, bottom. The silver head from Susa was considered by De Mecquenem to be the head of a staff; see 'Têtes de cannes susiennes en métal,' Revue de'Assyriologie XLVII [1953], pp. 79-82. A drawing was reproduced on p. 81, Fig. 2:3.

23. A Cylinder of Mitannian style with an Elamite inscription is in the collectin of M. Foroughi and will be published in 1966 in Iranica Antique.

24. See especially the report on the excavations at Tchoga Zanbil in Arts asiatiques VIII [1961], pp. 251-254, where the ingenious installations for a reservoir are discussed.

25. For a summary in English of the architecture of Tchoga Zanbil, see Labat in 'Elam...,' CAH II/XXIX [1963], pp. 17-22. Labat writes the name of the founder of Tchoga Zanbil as Untash-[d] GAL; see ibidl, p. 9 and note 1. I have adopted this writing for the same reasons that he gives in the cited note, but have simplified the name in the text for smoother reading to Untashgal.

26. For the 'postament' with niches, see Ghirshman in Arts asiatiques II [1955], p. 172, Fig. 11, and pp. 176-177.

27. For the description of the 'royal gate,' see Ghirshman in Arts asiatiques IV [1957], pp. 116-119.

28. For the restoration of the bull by Mme. Ghirshmann, see Arts asiatiques VI [1959], pp. 279-280

29. For a description of these 'offering tables' and the other elements of the cultic installations here described, see Ghirshman in Arts asiatiques IV [1957], pp. 120-122.

30. For Ghirshman's description of the tombs at Tchoga Zanbil, see Arts asiatiques VI [1959], pp. 272-278. For the plan of the palace of Adad Nirari I with the royal sepulchres, see C. Preusser, Die Pálaste in Assur [WVDOG] 66, 1955], Pl. 4. For Sennacherib's inscription on a brick of the royal sepulchre at Assur, see D. D. Luckenbill, The Annals of Sennacherib [OIP II, 1924], p . 151, XIII.

31. De Mecquenem made this suggestion in Vivre et Penser, 3éme série, which corresponds to Revue Biblique 52 [1945], p. 141. P . Amiet kindly drew my attention to this article.

32. For the publication of the model by J. E. Gautier, see 'Le "Sit-Samsi" de Silhak In Susinak,' MDP XII [1911], pp. 143-151. [Note: Accents are not correctly placed in the Title of this article.]

33. The statue of Napirasu was described by G. Lampre, MDP VIII [1905], pp. 245-250. For an enlarged detail of the hands, see Parrot, Sumer, p. 323, Fig. 399.

34. The head of an Elamite was published as Pre-Achaemenid in Survey IV, Pls. 105, 106. I. M. Diakonov was the first to recognize that the head belonged to the early art of Iran; see 'On an Ancient Oriental Sculpture' [English summary of an article in Russian], Musée de l'Ermitage, trav. du départment oriental IV [Leningrad, 1947] [Gosudarstvennyi Ermitazh; Trudy Otdela Vostoka], pp. 117-118. Diakonov made a careful comparison of the stylization of the beard with those of other sculptures, mostly Mesopotamian, and pointed to the close relation which exists with the stylizations of the Akkad period. In view of the tendency of Iranian art to retain earlier traits over many centuries, I think that a date for the head after the Akkad period, in the middle of the second millennium B.C., is possible.

35. A rendering of the eyes with the lower lid rising perceptibly at the outer end, to create the impression of sightly oblique position, is also seen in the head of a bearded Elamite of clay from Susa, juxtaposed with a photograph of the bronze head by Parrot in Sumer, pp. 330, 331, Figs. 406 and 407.

36. The dedication of a golden statue of the king can be found amog the names of years from the reigns of the Babylonian kings Ammidiatna [c. 1684-1647 B.C.] and Samsuditana [c. 1626-1595 B.C.]; See Unger, 'Daten-listen,' in Ebeling-Meissner, Reallexikon der Assyriologie II, p. 189: 245 [Ammiditana], and JNES XIV [1955], p. 156: XI/21 and perhaps XI/23 [Samauditana]. The formula which mentions a statue of gold and of silver [Unger, op. cit., p. 192:288] probably was due to an error on the part of one of the modern translators of the text; see J. J. Finkelstein in Journal of Cuneiform Studies XIII [1959], p. 47:k.

37. For the description of the Inshushinak deposit and of the deposit in the vicinity; the gold statuette and associated finds, see MDP VII [1905], pp. 61-130 and 131-136.

38. This statement is found in the article in Vivre et Penser, cited in note V/31, p. 141.

39. The lion's head of the whetstone is not unlike that of the axe from Tchoga Zanbil [well reproduced in Godard, L'art de l'Iran, Fig. 14], despite the more rounded forms of the latter.

40. In an article entitled 'Archaische Zügelringe; zur Auflösung der Gattung "Luristanbronzen", 'to be published in the Festschrift for A. Moortgat, P. Calmeyerattempts to place all rein-rings of similar shape, including our Plate 13, in the third millennium B.C.' Without stratigraphic proof for a survival of earlier types into later periods, my argument for a later date cannot be pressed.

41. For the fragmentary figures of Napirasu, her mother and her consort in the upper register of the stele, see EncycopŽdie photographique de lÕart I, p . 270 C, also Vanden Berghe, Archéologie, Pl. 102c.

42. We find in the human-headed bulls of a bitumen vase from Susa, Encyclopédie photographicque de l'art I, p. 255, A, a profile related to that of the demon on the stele of Untashgal. Again inhabitants of the mountain regions may have been in the mind of the artist who carved these creatures of primeval strength recumbent among the characteristic Elamite pine-trees growing on mountains.

43. We can do no more here than point to the possibility of the survival of early iconograhical features in later literary concepts. Detailed comparisons, however, can be made only by scholars who have first-hand knowledge of the late literary texts .

44. The relief in th central panel was placed in the Guti period and the processionof figures was considered to have been made earlier by N. C. Debevoise in 'Rock Reliefs in Ancient Iran,' JNES I [1942], p. 78. Herzfeld placed the relief in the time of Gudea; see Iran, p. 188. Some indications, however, seem to point to a later date, for example, the hair style of the enthroned god, which seems to resemble that of warriors on a bronze relief of the late second millennium B.C.; see Encyclopdeie photographique de l'art I, p. 275. Though the latter relief is not inscribed, it is nevertheless probably that it was set up by Shilhak Inshushnak about 1130 B.C. in the temple of Inshushinak at Susa; see R. Labat, 'Elam and Western Persia, c. 1200-1000 B.C.,' CAH II/XXXII [1964], p. 17. Vanden Berghe, who followed the dating of earlier writers in Archéologie, p. 58, states in Iranica Antiquea III/1 [1963], p. 32 [note 3 continued from p. 31] that the central scene is much older than the procession of figures, which he tentatively dates in the eighth century B.C.

45. See the reference to the relief of 'Addahamiti-Inshushinak' in note V/11.

46. Naqia, the mother of Esharhaddon [680-669 B.C.], is represented with such a crown in Parrot, Assyria, p. 118, Fig. 133. The wife of Ashurbanipal, Ashusharrat, was shown with such a crown in the so-called Garden Scene reproduced by Frankfort, Art and Architecture, Pl. 114, and on her stele; for the latter, see Andrae, Die Stelenreihen in Assur [WVDOG 24, 1913], p. 7, Fig. 3 and Pl. X, stele I.

47. We differentiate here between faience, a composite material consisting of a body or core 'of finely powdered quartz grains cemented together by fusion with small amounts of an alkali or lime or both,' and earthenware in which the body contains clay. Both are glazed but earthenware objects have to have a sliceous covering to which the glaze can adhere. The quotation about the composition of faience was taken from J. F. S. Stone and L. C. Thomas, 'The Use and Distribution of Faience in the Ancient East and Prehistoric Europe,' Proceedings of the Prehistoric Society N.S. XXII [1956], pp. 37-84, especially p. 38.

48. Especially interesting are the doors, found at Tchoga Zanbil, which were decorated with glass rods; see Ghirshman, Arts asiatiques III [1956], p. 169, Fig. 8. De Mecquenem found one hundred kilos of such rods without recognizng how they had been used; see MDP XXIII [1953], Pl. B:1 and pp. 52-53, Fig. 20:1,2.

49. For the sculptures from Nuzi, see R. F. S. Starr, Nuzi [Cambridge, 1937], Pls. 110: A and 111; for an analysis of the glaze, see ibid., pp. 523-525. For the bull from Tchoga Zanbil, see note 28 above.

50. The earliest glazed bricks dated with any certainty are the orthostates of Tukulti Ninurta II [890-884 B.C.]; see W. Andrae, Coloured Ceramics from Ashur [Berlin, 1925], Pls. 7, 8.

51. It will be interesting to see whether this colour was produced by copper compounds as in Egypt or by cobalt, which was available in Persia; see A. L ucas, Ancient Egyptian Materials and Industries, 4th ed. revised by Y. R. Harris [London, 1962], p. 189.

52. To the same time, the reign of Kutir-Nahhunte after 1160 B.C., belong the brick reliefs, conveniently reproduced by Parrot, Sumer, p . 329, Fig. 405.

53. The fragment of the horned animal, Fig. 44, which I believe to come from a late tile, is shown in the reproduction [MDP I [1900], Pl. VI, upper left] with a strong yellow colour like Plate 14 below, which I consider an earlier tile.

54. Ivory boxes which resemble the Elamite faience jars in shape are represented among the Nimrud invories; see R. D. Barnett, a Catalogue of the Nimrud Ivories [British Museum, London, 1957], Pls. XVI-XXIV. The small ivory head with the diadem of rosettes and double lotus-blossoms was published in ILN [Aug. 16, 1952], p. 255, Figs. 8, 9. It is now in the Metropolitan Museum of Art, acc. no. 54.117.8. Perhaps the rosettes and the lotus-petals placed above and below a double band are derived from Hittite hieroglyphs with an auspicious meaning. This would help to explain the frequent occurence and longevity of the motif, especially its use on the crown of Darius at Bisutun, Fig. 85.

55. An undecorated situla was found at Hasanlu in a context of the ninth century B.C. Perhaps this means that the situlae decorated with repoussŽ were later. Mme. Maliki, who published the finest situla so far known, gave numerous reasons for dating the piece in the twelfth to tenth centuries B.C. and proposed to see in the situlae a group of objects from Luristan which extends over several centuries; see 'Situle à scène de banquet,' Iranica Antiqua I [1961], pp. 21-41. I am inclined to call the situla Elamite and to date it in the ninth or eighth century B.C. on the basis of the hair style of the principal male figure, which differs from that of King Marduknadinahhe cited by Mme. Maliki. Moreover, I do not know of another example of the empty honeycomb pattern of the figure's robe, which should be dated before the first millennium B.C.

56. The point that bronze shields were 'for ritual and show,' leather for use, was demonstrated by J. M. Coles in ILN [March 2, 1963], pp. 299-301.