Ricerche sui sigilli a cilindro Vicino-Orientali del periodo di Uruk/Jemdet Nasr.

Collon, Dominique

This study of the earliest cylinder seals of the second half of the fourth millennium B.C., begins with an introductory section summarizing the development of the field, the various studies relating to these seals and the necessity for a new study incorporating recently excavated material. This new material has led to a chronological reassessment of two styles, the "naturalistic" Uruk style and the "schematic" Jemdet Nasr style, which had originally been thought to be consecutive but which are now shown to have been contemporary. The first is known mainly from impressions found in excavations, while the second is attested predominantly from actual seals, many of which are unprovenanced: this must reflect a very different pattern of use. According to Rova (p. 12), "The fundamental problem . . . is distinguishing, within the huge field of variations presented by the seals. . . . between the aspects which can be attributed to chronological differences, to local or regional factors or, finally, to the presence of different lapidary styles and workshops, and those [aspects] which can perhaps be linked to their [the seals'] specific function and the social position of their owner."

In order to attempt to resolve this problem, the author has submitted the material to statistical analysis. She defines her largely iconographical criteria for selection of the 963 seals and impressions (including fragmentary impressions) which make up her corpus: these had to be published, excavated or fairly reliably provenanced, in the Uruk and Jemdet-Nasr figurative style, from Mesopotamia, southwestern Iran, Syria, and Anatolia (but excluding Egypt), chronologically linked to the Uruk period, but excluding "the groups which are clearly attested only within the Jemdet Nasr period or later which have no precedents within the Uruk phase" (p. 18; such a statement reveals the unresolved ambiguity of using the term "Jemdet Nasr" to describe the so-called "schematic" seals of the Late Uruk period and the necessity of having a term to describe the phase between Late Uruk and Early Dynastic I - see also n. 91). The inclusion or exclusion of further groups and the criteria for counting an image as single or multiple, and for dealing with repeats on seals and sealings, are discussed on pp. 19-24.

The programme used for codifying the material (SPAD.N) is described in part II, and the variables or fields are listed on p. 26. The remainder of part II consists of a detailed discussion of these fields: region or site (pp. 31ff.); archaeological context (pp. 52 ff.); type of objet (seal or impression) (pp. 57ff.); elements of composition - humans (pp. 70ff.), animals (pp. 78ff.), hybrids (i.e., monsters, etc.) (pp. 88ff.), inanimate (buildings, plants, objects) (p. 91ff.) - and their relationship to each other (pp. 121ff.). Numbers refer to the seals in the corpus, the latter illustrated by clear line drawings arranged on fifty-nine plates at the end of the volume and catalogued on pp. 309-31. The seals were also described in running text (discussed pp. 13845, with examples) so as to allow for a "lexical" as well as a numerical analysis of the material.

Part III records the results of the actual analysis, illustrated with graphs and charts, and the results are classified in part IV. The methodology and underlying mathematics are discussed by Sergion Camiz (pp. 279-93). Much of this is extremely detailed and technical but fortunately for the reader, the whole is summarized and viewed in context on pp. 253-70.

The iconographical analysis (internal variables) showed a division into two main categories: scenes with animals and scenes with humans, the latter with three main thematic groups: religious scenes, hunting scenes, and scenes from daily life. Scenes with animals are less varied and can be divided into three groups: the largest depicting common animals, a smaller group depicting rare animals, and a third group with snakes, birds of prey, lions, and monsters, often in heraldic compositions and "di probabile significato simbolico." Schematic seals combining spider and pig-tailed women and scenes consisting entirely of inanimate objects are considered to be two further categories; surely in the former, which Rova correctly relates to seals representing scenes from daily life, the spider stands for Uttu, the Sumerian spider-goddess of weaving, or her earlier counterpart. Certain types of vessel appear more often in religious scenes. Certain postures are restricted to certain categories - prisoners have their arms tied behind their backs! The structure of "schematic" seals is simpler than that of "naturalistic" seals. This is disappointing and one feels that similar conclusions could have been reached more simply.

The combination of the iconographic and geographic analyses (internal and external variables, pp. 258ff.) is also disappointing since it demonstrates no clear geographical variations; only the seals from Anatolia (Arslantepe) are distinctive. The Uruk style seems to have been international, with a greater emphasis in southern Mesopotamia on scenes of religion and warfare. The archaeological context demonstrates that certain scenes are more common in temples, while scenes from daily life and "schematic" scenes are found in domestic contexts. Seal impressions attached to movable containers show a preponderance of religious and complex or heraldic animal scenes, spherical bullae display a varied repertoire and elongated bullae are sealed with simpler animal scenes. Alas, none of this is new.

Equally disappointing is the lack of information on chronological and stylistic variations; the analysis merely confirms that the spherical bullae are early and the Jemdet Nasr sealings are late within the period. Most iconographical motifs seem to have had a broad chronological span. There is perhaps a tendency for "schematic" seals to replace more complex scenes from daily life or war, but religious scenes continue. "From the point of view of style, [the 'schematic' seals] are absolutely homogeneous and they do not therefore present any useful characteristic for establishing their chronological evolution" (p. 261). Rova (p. 261 n. 124) is obliged to resort to the conclusions in P. Amiet's Glyptique mesopotamienne archaique (Paris: CNRS, 1961). The author (p. 262) considers that it is "unlikely that the images on the seals represent a true and formalized language of their own, aiming at providing precise information concerning the social position of the seal owner or the type of goods sealed." However, the types of scenes do reflect a heavily stratified and specialised society. She states (p. 264) that the "schematic" scenes represent most of the themes of the "naturalistic" seals, of which they are "the synthethizing version, pruned of every unnecessary iconographical element, with simplified syntax and a less accurate representation of details" and they must therefore, at least "conceptually," be later. She then goes on to list the numerous and telling differences in the repertoire of the two groups: scenes depicting warfare, the priest king, and heraldic animals, found on "naturalistic" seals, have no equivalent on "schematic" seals which, however, frequently depict women.

These "schematic" seals are, I believe, the seals of that section of society which was concerned with manufactured goods (textiles, pottery-making), which were largely in the hands of women and probably carried out in the home. Few impressions have survived, because relatively few domestic contexts have been excavated, and it would be mostly in domestic contexts that bales of cloth would be unpacked. In contrast, jars of oil, sacks of cereal, war booty, and administrative documents would have been stored in the temples, sealed by the seals of the priestly elite, and these buildings have been excavated.

The book is well presented and legible; the collection of illustrations and associated data will be useful as a basis for future research, but we must commiserate with the author who devoted several years to such an exhaustive analysis that has produced such negligible results. We can only conclude, with her, that the sample is still too small and too unrepresentative and that new excavations are needed.

DOMINIQUE COLLON THE BRITISH MUSEUM