Wednesday, July 18, 2007


[This article appeared in The Oriental Institute News & Notes, No. 194 (Summer 2007), and is reproduced here with the kind permission of the Oriental Institute. News & Notes is a quarterly publication, printed for members as one of the privileges of membership in the Oriental Institute]


by Mark B. Garrison

Presented at the Oriental Institute, in honor of the retirement of Raymond D. Tindel (March 26, 2007)

Archives of clay tablets have been found at a wide variety of sites across the great sweep of what we today call the ancient Near East, or ancient Western Asia. Archives are highly prized owing to the rich textual evidence that they provide on various matters social,political, economic, religious, etc. What is perhaps less well known about archives of clay tablets is that they are often conveyers of visual imagery. In many instances part of the protocols involved in administrative activity was the application of one or more seals across various surfaces of the tablets. These seal impressions functioned administratively on a variety of levels, acting as markers of witnesses to the transactions,identifiers of personnel/offices involved,tokens of authority and security, etc. Some “archives” consist in fact solely of uninscribed clay artifacts that carry seal impressions.

Persepolis Fortification tablets in situ

Seals in ancient Western Asia consisted of two types, cylinders and stamps. Since their invention in the late fourth millennium B.C., cylinder seals were by far the preferred medium in ancient Iran and Iraq. In the first millennium B.C., stamp seals re-appeared and were used contemporaneously with cylinder seals. When cylinder and stamps were applied to the still-moist tablets, the imagery, carved on the reverse in the actual seal matrix, would appear in the positive on the surfaces of those tablets. One of the interesting aspects about seals is that the seal user created imagery through the application of his/her carved glyptic artifact. While some seal users may have understood that imagery in a very practical manner, i.e., a signature, it is clear that many seal users prized the often virtuosic carving and must have delighted in the potential imagery that lay literally at their fingertips.

The importance of glyptic imagery has long been recognized. Indeed, seals are the most commonly occurring artifact that carries visual imagery in the archaeological record of the ancient Near East, surviving by the tens of thousands. As such, seal images have been studied since the beginnings of modern archaeological expeditions in the nineteenth century.

It is a curious aspect of glyptic imagery that it survives in two distinct forms: as an actual seal artifact (usually of stone,but sometimes of clay, bone, metal, shell, etc.), and as an impression in clay. It is exceptionally rare to have both an ancient seal and an ancient impression of it. This may be simply the result of archaeological serendipity, or it may reflect real patterns of human behavior. That is, seals that were used as administrative markers may have been different from those that were used as items of personal adornment and/or amulets; thus, the two types of seals may survive in the archaeological record differently owing to different patterns of use and deposition.

Cylinder seal with modern impression. OIM A25308

Cylinder seals can be manipulated in various ways to provide oftentimes strikingly beautiful modern impressions. Indeed, these seals have to be impressed in some type of modern sealing medium in order for us to study their visual imagery since the carved surface of the cylindrical piece of stone is often difficult, if not impossible, to read. By contrast, seal impressions almost never preserve the full scene and are often difficult to see. In many archives a seal will be applied to more than one clay artifact, meaning that all impressions of that seal must be surveyed and then a collated composite image generated. For these reasons, even today, surveys of the glyptic arts focus almost exclusively on modern impressions of ancient seals, rather than impressions of seals from archives.

It is a sad fact that the great bulk of these seals do not have an excavated provenance. Because of their small size and often beautiful carving, seals have long been highly prized by collectors. The resulting loss of contextual information is a devastating blow to our research endeavors. While the images on these unprovenanced seals may be beautiful, they float in time and space; serious questions of authenticity are often part of the calculus, especially with a seal image that does not fit the canon as traditionally defined. A traditional venue of the publication of glyptic imagery in fact is the catalogue of seals now found in museum and private collections. The scale of the problem of provenance may be seen in the recently published volume six of the British Museum series (Catalogue of the Western Asiatic Seals in the British Museum: Cylinder Seals 6: Pre-Achaemenid and Achaemenid Periods, by Parvine H. Merrillees). Of the ninety-two seals in volume six, only three seals have an excavated provenance.

Aerial photograph of the Persepolis terrace; the Fortification archive was discovered in chambers of the fortification wall visible at upper left in the photograph, where the wall abuts the mountain

By contrast, seals preserved as impressions on clay artifacts within the context of archives have less often attracted the attention of collectors (there are, nevertheless, thousands of clay tablets of unknown archaeological provenance in museum and private collections around the world). We are especially fortunate to have many large archives of sealed tablets discovered through controlled archaeological excavations. When we have a seal preserved as an impression on a tablet, we are seeing the imagery in its functional context. That functional context (i.e., the surface of a tablet) is enhanced by the fact that the tablet is related to other tablets via the archival context. The seal impression is thus related via function to other seal impressions in that archive. That archive, moreover, can almost always be specifically located in time and space. Seal images preserved as impressions in archives thus constitute one of the more remarkable contexts for the study of the visual imagery of the ancient world.

The richness of this evidence and its potential for providing unique insights into the lived human experience/interac-tion with images may be glimpsed in a large archive of administrative documents currently at the Oriental Institute on loan from the Iranian government. This archive is known as the Persepolis Fortification archive. The Persepolis Fortification archive was found in chambers of the northern fortification at Persepolis (whence the name of the archive) in 1933 by a team from the Oriental Institute. The particulars of the archive and its importance were well articulated by M. W. Stolper in the Winter 2007 edition of News & Notes. In brief, the archive represents the administration of a food rationing system that covered an amorphous area consisting of the environs of Persepolis (Parsha), Pasargadae (Batrakatsh), and Shiraz (Tirazzish) and a broad(?) expanse to the northwest along the royal road to Susa. Date formulae preserved in many texts date the archive to years 13–28 (509–493 B.C.), in the reign of Darius I. There are three major components of the archive: tablets that carry Achaemenid Elamite inscriptions in cuneiform and, very often but not always,impressions of seals; tablets that carry Aramaic inscriptions in ink (and/or incised) and, very often but not always, seal impressions; and tablets that carry only seal impressions (what are designated as the uninscribed tablets). The exact number of tablets and fragments is not known, but recent work by Stolper places the tablet count at approximately 15,000–18,000 distinct documents in toto. That is a huge number of artifacts, constituting one of the largest archives to have survived from ancient Western Asia.

The seal images are the only aspect of this administrative system that can be documented across all three components of the archive: Elamite tablets, Aramaic tablets, and uninscribed tablets. The seals applied to the tablets represent the officials and offices delivering and receiving commodities,and the officials and offices responsible for overarching administrative accounting and oversight. Just as there are many tablets, so, too, there are many seals preserved in the archive. On the 2,087 Elamite tablets published by R. T. Hallock in 1969, approximately1,148 different seals can be recognized (we distinguish seals preserved on the Elamite tablets with the siglum PFS). The first volume of the publication of the seal s preserved on the PF tablets has now been published: M. B. Garrison and M. C. Root, Seals on the Persepolis Fortification Tablets, Volume 1: Images of Heroic Encounter (Oriental Institute Publications 117; Chicago: The Oriental Institute, 2001). From preliminary research by E. Dusinberre on the seals applied to the Aramaic documents and by the author on the unpublished Elamite tablets and the uninscribed tablets, there are probably at a minimum another 1,000–1,500 seals.

We have thus preserved in the Fortification archive one of the most densely concentrated collections of visual imagery from the whole of ancient Western Asia and the Mediterranean worlds. Thinking for a moment about the multi-layered contexts, this imagery is closely circumscribed in place (the environs of Persepolis and an area extending to the north and west), time (509–493 B.C.) and function (all the images are being used by officials/offices to seal tablets in the archive). Each of these some 2,000 images is thus linked to all other images in a most intimate and direct way. These images are also objects actively in use by administrators, thus linked to a specific type of activity by specific people. Those individuals, moreover, run the social gamut from lowly workers to the imperial family. The social position and administrative rank of many of these individuals are, furthermore, fairly well known. These images are also linked in time and place with the massive construction activities associated with the building and ornamenting of Persepolis and rock-cut tomb of Darius at Naqsh-e Rustam. Finally, this particular time and place happen to be exceptionally critical, marking the initiation of administrative, political, social, and ideological programs associated with Darius’ consolidation of an empire the scale of which the world had never seen.

Photo of PFS 9*. Drawing and photo by Mark Garrison

By examining, for example, how individuals select and use imagery, we may pursue in these seal images a social history of art that in most other times and places would be an impossibility. One way that we may explore the richness of the multilayered contexts of the Persepolitan seal images is via a phenomenon that I have called “replacement seals.” In numerous instances we are able to track an individual’s replacement of one seal by another. Tracking this type of phenomenon in and of itself would of course be next to impossible with any glyptic artifact found outside of an archival context. Most of the time at Persepolis the adoption of a new seal goes unmentioned in the textual record. In one now-famous case involving the chief administrator of the archive, Parnaka, probably the uncle of Darius, two Elamite texts (PF 2067 and PF 2068) specifically note the replacement of the earlier seal, PFS 9* (Cat.No. 288) with the new seal, PFS 16* (Cat.No. 22): “also, the seal that formerly (was) mine has been replaced — now this seal (is) mine that has been applied to this tablet” (PF 2067).

Parnaka’s right-hand man, the second in command of the archive, Zishshawish, also replaced his earlier seal, PFS 83*, with a new seal, PFS 11*. Zishshawish is the Elamite form of the Old Persian name *Ciça-vahu- “of good lineage.” His two seals provide an especially interesting case of the replacement of seals concerning an individual at the very highest levels of the Persepolitan administration. As indicative of his high administrative rank, Zishshawish receives very high food rations, issues letter orders, employs scribes, and never needs a counterseal on his transactions.

Based on the preserved texts, Zishshawish can be documented using his first seal, PFS 83*, between May/June 507 B.C. and November/December 504 B.C. The seal is a unique and intriguing design. Two figural compositions constitute the major elements of the design field; a winged bovine with suckling calf, and a four-winged human headed-bull supporting a winged disc. An Aramaic inscription along with the lower three prongs of a star remain in the upper field.

If preserved as an unprovenanced artifact in a museum, this seal would probably be classified as an Assyrian product. Certainly, in comparison to the conventional understanding of how Achaemenid glyptic ought to appear, an understanding built almost exclusively upon unprovenanced artifacts, the imagery, iconography and style of carving of PFS 83* certainly do not seem “Achaemenid.” Several features of its design are traditionally associated with Assyrian art. For example, the cow and calf motif was especially popular in Assyrian and Syrian glyptic, ivory carving and metalwork of the early first millennium B.C. So, too, the bull-man in the form of an atlantid, often associated with a winged ring/disk or the half-figure in the winged ring/disk, was very popular in Assyrian glyptic. PFS 83* is not, however,an unprovenanced object. Embedded in a pool of imagery owing to its archival context, we can see that its carving style is completely at home within the seals from the Fortification archive.

PFS 16*. Drawing and photo by Mark Garrison

Lastly, the Aramaic inscription on PFS 83* is quite at home within a Persepolitan context. Only the first word in what appears to have been a one line inscription, enclosed in a panel, is preserved: HTM…, “Seal (of ) …”. The preservation of the vertical edge for a panel at left would seem to indicate that there may have been as many as four or five more letters in the line. One assumes that the missing section of the inscription contained a personal name. Inscribed seals are fairly rare in the archive, less than 10% of the seals carry inscriptions, and the majority of those inscriptions are in Elamite. There is, however, a substantial corpus of seals inscribed in Aramaic, and the formula “Seal of PN” is a common one among those seals. Indeed, it is interesting to note that Parnaka also uses Aramaic for the inscriptions on both of his seals.

PFS 83* thus may be related in style,imagery and inscription to other seals in the archive. In general, the seal takes its place as one example among hundreds of Persepolitan seals that exhibit archaizing imagery and style that are deeply indebted to Assyrian models. Nevertheless, it is clear that this seal is a very special one. The scene of the cow and calf is unique among the seals studied to date in the archive. The Aramaic inscription also marks the seal as special; the combination of the placement of the inscription in the upper field and its enclosure within a panel cannot be paralleled in any other seal studied to date in the archive. Several characteristics thus point to this seal being a commissioned piece. This ought not to surprise us, given the exceptionally high administrative rank of Zishshawish.

Photo of PFS 83*. Drawing and photo by Mark Garrison

On a tablet dated to December 503 B.c./January 502 B.C., Zishshawish for the first time uses his new seal, PFS 11*. He continued to use this seal until Febru-ary/March 496 B.C. Whether the approximately one year hiatus between his last use of PFS 83* and his first use of PFS 11* is real, or simply due to accident of record survival, cannot be determined.

PFS 11* is a magnificent seal, one of the great masterpieces of glyptic carving from the Fortification archive. The scene consists of a central “altar” above which floats a half-figure in a winged disk; to either side of the altar is disposed a crowned figure in Persian court dress. This central scene is then flanked by date palms followed by a paneled, trilingual (Old Persian, Elamite, and Babylonian) royal-name inscription in the terminal field.

Unlike PFS 83*, PFS 11* conforms to our conventional understanding of how Achaemenid glyptic ought to appear, in style, iconography, and composition. The figures wear the Persian court robe and the dentate crown that are so often associated with aspects of Achaemenid imperial imagery. The scene is relatively rare, but,nonetheless, one well known from unprovenanced glyptic examples.

Like PFS 83*, PFS 11* is also a very special glyptic product. The seal is one of four seals from the archive that carry the standard royal-name inscription of Darius. None of these seals belonged to Darius; all were used by high-ranking officials/offices. Of course, without its archival context, we would have no idea that the seal in fact belongs to Zishshawish, and, I suspect, there would be speculation that this was in fact the personal seal of the king. The scene of two figures flanking an “altar” is rare in the archive; I know of only four other examples in the seals studied to date. So, too, the carving style, what I have called the “Court Style,” is rare in the Fortification archive. The scene, and aspects of its iconography, recall the famous tomb relief of Darius at Naqsh-e Rustam, but the relationship is not as close as one may think at first glance. In fact, in its mirror doubling of the royal figure and its static, idealized composition, PFS 11* provides in many ways a very different statement about the nature of Achaemenid kingship.

Owing to the rich archival context, we may offer some speculation on the nature of the relationships of the two seals of Zi‰‰awi‰ to each other, and what those relationships may say about Zishshawish himself. The two seals provide an in-depth view of patron taste/needs at the very highest levels of the imperial administration at the heart of the empire. The precise social/ad-ministrative/political dynamics that lead to Zishshawish’ initial selection of the imagery on PFS 83*, and then the replacement of that imagery with PFS 11*, are, of course, lost to us. Nevertheless, we may be able to infer some aspects to these processes owing to the rich archival contexts of both Zishshawish the administrator and the seals that he uses.

Both seals, as we have seen, are special artifacts, possessing iconographic, stylistic, and compositional traits that are either rare or unique. Both must be commissioned objects, as one may have expected for an individual of Zishshawish’s administrative rank, and, thus, we may infer that Zishshawish played some role in the selection of their style and imagery. PFS 83*, despite its unique features, takes its place first and foremost as one of many hundreds of examples of strongly Assyrianizing imagery in the seals from the Fortification archive. Rather than trying to decode the individual elements (and/or their combination) of the figural imagery, I suggest that the primary signification of the imagery lay in this Assyrianizing “flavor.” It is striking that many of these now very well-known administrators who have direct ties to the royal family seem to prefer seals executed in this archaizing manner. Zishshawish’s immediate superior, Parnaka exhibited the same predilection for Assyrianizing style and imagery in both of his seals, PFS 9* and PFS 16*. The royal woman Irtashduna, the daughter of Cyrus and favored wife of Darius (Herodotus 7.69.2), uses a seal, PFS 38, with such exceptionally strong Assyrianizing elements that several commentators have actually dated its carving to the Assyrian period. As far as we know,Zishshawish had no direct ties (by marriage or blood) to the royal family. Perhaps the imagery of his seal PFS 83* is an attempt to emulate the taste of his immediate superior and the royal family as a whole.

Photo of PFS 11*. Drawing and photo by Mark Garrison

The sudden appearance of PFS 11*,a seal bearing a royal-name inscription no less, must mark a critical point in the biography of Zishshawish the administrator and courtier. The seal appears, along with a handful of other beautifully executed Court Style seals, three of them also bearing royal-name inscriptions of Darius, in the last decade of the sixth century B.C. None of these seals belongs to members of the royal family. The very specific and consistent style and iconography of these seals articulate very clearly the new imperial message. The seals also would seem to act as foci of a dialogue between the king and his administrative elite. They communicate both the king’s recognition of these individuals as closely linked by loyalty (in lieu of blood and/or marriage) to the king/ royal family, and those individuals’ affirmation of membership/loyalty to the newly(re)constituted royal order. As such, the imagery and style of these seals convey a dialogue between king and administrative elite having more to do with personal relationships than abstracted concepts of imperial ideology. Images and image making,while dominant features of the physical and intellectual landscape of the Persepolis region in the late sixth century B.C., thus may have also played a critical role in the social and political lives of individuals.

[Following are the pages of this articles in the format in which it originally published in The Oriental Institute News and Notes]