You are here

Arslantepe Cretulae: An Early Centralised Administrative System Before Writing

Arslantepe Cretulae: An Early Centralised Administrative System Before Writing

By M. Fragipane, P. Ferioli, E. Fiandra, R. Laurito, and H. Pittman. With contributions by J. Blackman, E. Cristiani, M.B. D’Anna, C. Lemorini, P. Morbidelli, A.M. Palmieri, and C. Simonetti. Pp. 528, numerous figs., b&w pls. 11, color pls. 5, numerorous tables, cat. in DVD-ROM. Università di Rom a La Sapienza, Rome 2007. €175. ISBN 978-8-8901-7017-1 (cloth).

Reviewed by

Volume 5 of nine planned volumes in the Arslantepe series is a magnificent book. It describes in immense detail the thousands of meticulously excavated cretulae found in the late fourth-millennium palatial complex at Arslantepe (level VI A = Late Chalcolithic 5), southeastern Turkey. Given the wealth of evidence, there seems to be material for decades more of research.

First, the cretulae of the book’s title is Latin for “sealings” (Cic. Verr. 4.26.58). The authors hope that this term will replace the present confusion of nomenclature (e.g., “sealings,” “nodules,” “bullae,” in all languages) to express the full range of seal-impressed clay documents—whether securing, authenticating, guaranteeing, or recording transactions.

Arslantepe Cretulae (hereafter AC) may seem a monumental volume on a narrow subject, but—largely thanks to some authors of the current volume—it has become one of the most fruitful ways to explore the development of ancient urbanism and its underpinning economy and bureaucracy. AC should encourage a broad discussion of a subject that has too often been reserved for specialists. This comprehensive analysis sheds important new light on the “organisation of functions and administrative requirements in different [prehistoric and early historic] economic environments,” (15) on the development of centralized hierarchal societies, different models of urbanization, as well as the more traditional subject of glyptic iconography.

An introduction (Fiandra, Frangipane describes three main types of cretulae: (1) clay lumps pressed directly against containers, doors, or documents to guarantee or authenticate what was sealed or record a transaction concerning the sealed object; (2) ovoid cretulae hanging from objects, or solid ovoids unattached to any object (the latter known in the Aegean as noduli); (3) hollow clay balls having a seal-impressed surface, containing tokens (hitherto bullae). Pace AC, I would prefer to retain noduli and bullae (only for the hollow balls), as their uses and evolution seem quite distinct.

Chapter 1 (Frangipane) describes the findspots of 2,145 cretulae, with detailed plans, scatter maps, and (where relevant) reconstructions and stratigraphic sections.

The four main findspots are Temple A, west of the palace, 140 cretulae, mostly dumped along with waste materials—a “discarded archive”; Temple B, north of the palace, 38 cretulae kept in a sack near the vestibule entrance—a “temporary archive”; and Room A340, smallest of three storerooms west of the central courtyard, the largest group of in situ cretulae (n=175), left where they fell as containers were opened and resealed during redistribution—an “active archive.” The numbers of mass-produced bowls (similar to Mesopotamian “flowerpots”) in A340 suggest foodstuffs distribution. Given that 30 different seals were used in this small room (with no overlap elsewhere), AC argues that these were seals of people who withdrew goods rather than those of storeroom officials. This seems a possible, but not compelling, explanation. The last main findspot is Rubbish Pit A206, a long narrow cavity inside the palace containing more than 5,000 cretulae (1,728 with seal impressions)—along with vases, mass-produced bowls, animal bones, and a few “tablets” marked with round holes—all systematically discarded after accounting procedures or at the end of an administrative period. Three phases of dumping are identified in a succession of 20 thin alternating layers of burnt materials and soil. Each phase seemingly represents different types of transactions and different officials, rather than chronological differences. Some seals and some container or door cretulae occur in all three phases, which indicates that layers were dumped at very short intervals, probably (it is argued) filling the pit within a matter of months.

Chapter 2 (Ferioli, Fiandra, Frangipane), “Material and Functional Characteristics,” is of most interest to administrative specialists. Sections 1–4 explore, type by type and deposit by deposit, “the different impressions on the cretulae, both made by the seals that had been pressed onto them, and left behind by the containers and other objects that the cretulae had been used to secure” (61). While many observations are familiar from Ferioli and Fiandra’s earlier work, two new types are identified: sacks closed with cylindrical pegs (thus easily confused with door/chest closures); and doors closed with wooden “pin-tumbler” locks—the world’s oldest locks—probably securing rooms containing valuables. Section 5 (Frangipane, Simonetti) examines the “accounting tablets,” noting similarities with a “large token” from Tilbes Höyük, and small tablets from Uruk-Warka. Section 6 (D’Anna, Laurito) compares vases and in situ cretulae in A340, matching nearly all the pots to cretulae.

Chapter 3 (Pittman) presents a catalogue of the seal designs (211 seals). Most are stamp seals—as always in Anatolia—with both circular and rectangular faces; local-style cylinders make up 6–13% of various deposits. Section 3 (Frangipane) divides the local glyptic into four styles: naturalistic with soft outlines, coarser naturalistic, schematic, and abstracted schematic. All styles appear on stamps, while the more naturalistic style prevails on cylinders. Although lion iconography differs on the stamps and cylinders, craftsmen are otherwise distinguished by style, and not through depictions of animals or seal shapes. While many elements are common—indeed, particular—to Arslantepe glyptic, there are greater similarities in terms of style and subject within each deposit of cretulae than between deposits: “This might suggest that the officials working in a particular sector were linked [to] each other by some kind of bond that was expressed by the sharing of the same symbolic elements and iconographic conventions” (283). That may also be why so few seals were found even in two—and never more than two—different deposits (Seals 4, 30, 101 in A206 and A340; Seals 10, 12 in A206 and Temple B [pace three seals on p. 46], and Seal 214 in two minor dumps—one of only three Late Uruk imported cylinders [pace four on p. 285: adding S 157, an atypical animal file]).

Section 4 (Pittman) situates the corpus within the Upper Tigris tradition, subjected in this period to cultural forces emanating, albeit indirectly, from Late Uruk colony sites on the Euphrates. Local craftsmen borrowed the cylinder seal shape and a small number of iconographic motifs (e.g., the much-published S 1, a sledge scene) from the colonies. Except for four Late Uruk imports, virtually all the seals “were manufactured within a single tradition of seal carving, despite the considerable internal variety that can be seen within the corpus” (299).

Pittman divides the seals into two broad style groups: “Naturalistic” in a strictly relative sense, with generally correct proportions but lacking any internal articulations (304–5), and “Round Line,” the latter using the hemispherical cutting tool to form bodies that are not anatomically correct (308–9). A favored cylinder composition is the single file of animals and, on both cylinders and stamps, two or more images stacked one above the other (sometimes helter-skelter) or head-to-tail (tête-bêche). Arslantepe iconography is well defined; some parts developed from their own culture and traditions, other parts under the influence of new, closer contacts with the Late Uruk world to the south. Human imagery, although represented, is not the focus as in Late Uruk glyptic, but local craftsmen concentrate instead on animals, primarily horned quadrupeds, dogs, felines, snakes (either knotted or extended), and a four-legged, long-necked fantastic creature unique to Arslantepe. Disappointingly, imagery is not directly correlated with the seal user’s place in a functional hierarchy. A pyramid based on access to storage areas (securing doors)—from those who controlled three or more storage areas to a broad tier of those who sealed only containers and no doors—shows that the same iconographic themes were used at all levels of administration, from the highest to the lowest ranking. (There are two errors in the chart “Seals by Hierarchical Rank” [fig. 3.69]: the text puts S 15 in the second, not first, rank, and S 65 seems accidentally omitted from the second rank.) Similarly, an analysis of imagery and rank level by level in the great dump A206 merely concludes: “visual redundancy or elaboration can, with prominent exceptions, be considered fundamental for high rank” (338); more simply, bigger players usually have bigger, better seals.

Chapter 4 (Cristiani, Laurito, Lemorini) gives initial results of research into “Manufacture and Materials Used in Seal Production,” of the seals themselves that are, of course, entirely missing. Under the electronic microscope, the authors were able to interpret traces left on the best-preserved cretulae by different tools and techniques of manufacture, and identify raw materials of the original seals (stone, ivory, bone, wood).

Chapter 5 (Laurito), “Ropes and Textiles,” assesses traces of vanished materials on cretulae to reconstruct early handicraft activities. Identified materials range from cords and ribbons to small pieces of cloth (probably both flax and felt) and leather, and in one case, a highly processed skin similar to modern suede.

Chapter 6 (Ferioli, Frangipane), “Clay Used for Sealing Operations,” visually identifies four main groups of clay as well as two rarer groups; there is a direct correlation between Group B clay and the layers where these cretulae were found (A206), thus marking a distinct series of operations. Chemical analysis (Blackman) demonstrates the local origins of all cretulae clays, the same raw material also used for mudbrick, mortar, plaster, and some vases.

Chapter 7 (Frangipane, Fiandra), “A Complex Administrative System before Writing,” summarizes the results of almost 30 years of research. It is densely written but well worth studying in detail. The main points may be summarized as follows (466): (1) although only one complex of storerooms has been uncovered, the complex “situation in dump A206 reveals, with almost absolute certainty, that there must have been other storerooms in the public area”; (2) “every area of activity and every dump produced groups of cretulae which ... stand out in terms of their specific functional features and the presence of very characteristic, and mostly exclusive, seals”; (3) “every store or sector therefore had its own officials ... perhaps from within and from outside the palace, as suggested by the vast number of seals that were used in the operations”; (4) some few seals, however, were also found in other assemblages, “linking the various groups of materials in both chronological and functional terms.”

Even after the meticulous argumentation of this marvelous book, room still remains to question these last two conclusions. Regarding storeroom officials, one must keep in mind that this was still a world of mobile or semimobile pastoralists and farmers living in scattered rural villages. Arslantepe’s centralization took place, in fact, in the absence of real urbanization, “[f]or the site remained small during the period in which the public area was expanding enormously” (476). Vast numbers of seals used in a central place lacking an underlying urban structure suggest that the urban Mesopotamian model might not be the best fit; communal models might also be considered.

Regarding the length of time represented by the cretulae, if the great dump was truly filled in months (rather than years), the vast number of seals and cretulae must be strictly contemporary. But is this the only possible explanation? The user of seal S 10, for example (as far as known, the only ivory seal, and with unique border designs), was represented in cretulae layers from the very bottom to midway. Could S 10 have worked in the same department (represented by the door sealing S 2) over many years, his cretulae dumped from time to time? When he “retired,” the storeroom was either rarely used or his successor’s cretulae were dumped in other places. Name seals in stratified refuse deposits elsewhere demonstrate discarded cretulae spanning much longer periods (e.g., J. Wegner, “Social and Historical Implications of Sealings of the King’s Daughter Reniseneb,” in M. Bietak and E. Czerny, eds., Scarabs of the Second Millennium B.C. from Egypt, Nubia, Crete and the Levant: Chronological and Historical Implications [Vienna 2004] 232–36). I suggested such a model many years ago, meant to help interpret the sealings from Karahöyük (J. Weingarten, “Two Sealing Studies in the Middle Bronze Age,” in P. Ferioli et al., eds., Archives Before Writing [Turin 1991] 271–73, 291 n. 12). With all the data now available in the appendices, it should be possible to retest this model.

Whereas AC’s approach and argument will sometimes spark sharp debate among specialists, its originality and importance are not in doubt. This book has left me thinking about what it means to build an extraordinarily complex “village-based centralized power structure”—not a bad starting point for some fresh thinking.

Judith Weingarten
Via S. Croce 13
53030 Belforte (SI)

Book Review of Arslantepe Cretulae: An Early Centralised Administrative System Before Writing, by M. Fragipane, P. Ferioli, E. Fiandra, R. Laurito, and H. Pittman

Reviewed by Judith Weingarten

American Journal of Archaeology Vol. 113, No. 2 (April 2009)

Published online at

DOI: 10.3764/ajaonline1132.Weingarten

Add new comment

Plain text

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.