Analysing Pottery: Processing, Classification, Publication

The collection of papers under review originated in a 2003 workshop in Bratislava, followed three years later by a session at the annual conference of the European Association of Archaeologists. Both meetings were concerned with the intellectual and practical challenges of processing and analyzing large amounts of archaeological pottery. As an aid to those embarking on such projects, the resultant volume offers a series of exempla, 15 papers (9 in English, 6 in German) that highlight problems, outline systems for description and documentation, and describe and occasionally critique the application of various analytical techniques and the results achieved through their use. The geographical range of the contributions encompasses central and southern Europe, the Aegean, and Anatolia, and most treat material from settlements of the Bronze Age, with only three focusing on later periods (La Tène, Roman, and Early Medieval). The methods that they demonstrate, however, are widely applicable. Most of the contributors share a concern with extracting information from pottery that is fragmentary and often from disturbed contexts, and they outline a variety of ingenious strategies for doing so.

Like archaeological pottery itself, these studies are highly diverse and difficult to pigeonhole. The editors have divided them into four categories—“Ware Definitions” (two papers), “Multivariate Analyses” (four papers), “Other Statistics” (eight papers), and “Database Systems” (one paper)—but there are many areas of overlap. Some contributions focus sharply on one part of the process, but many touch on almost every stage of the investigation. A thoroughgoing critique or even a summary of their contents is beyond the scope of this review, and I therefore confine my comments to recurring themes and a few examples. (The table of contents is currently available at http://archeologia.fphil.uniba.sk/attachments/Horejs_Jung_Pavuk_press.pdf.)

Most authors outline the gathering and recording of the same basic data (e.g., details of ware/fabric, surface treatment, decoration, form, dimensions, preservation, context). A clear picture of the essential ceramic attributes emerges, but there are some instructive disagreements about their description. For instance, in many studies, temper is identified with geological precision, but some authors (e.g., Gauss on Bronze Age ceramics at Kolonna, Greece; Mielke on Hittite ceramics) advocate instead general descriptions, and Gauss warns that correct geological identifications are beyond the capabilities of most archaeologists and create a false sense of scientific precision (308). The Munsell color chart is adopted by some, while others argue for subjective descriptions of fabric color. Similarly, the calculation of estimated vessel equivalents (variously abbreviated eres, ERE, and GE) is found to be essential for some projects but inappropriate to others. These differences are generally not matters of principle but stem from the nature of the material under study, and they are useful in prompting readers to consider which procedure is appropriate to their own projects.

While the collection is explicitly descriptive rather than prescriptive, it includes a number of tools that could be adopted as is or easily adapted to a planned project. Berger, for instance, offers a list of precise (German) terms for the surface treatments encountered in the Early Helladic pottery at Kolonna, although her black-and-white images are not quite adequate to the task of illustrating the subtleties captured by her terminology. Widely adopted, her system would facilitate intersite comparisons. Several authors illustrate the forms used for data entry: for Bronze Age pottery at Tell el Dab’a (Egypt), Troy (Anatolia), Kolonna (Greece), and Bruszczewo (Poland); and La Tène pottery from Straubing-Bajuwarenstrasse (Bavaria). Gauss’ account of the recording system developed at Kolonna constitutes a particularly clear paradigm for the whole process, including the practical details of numbering, equipment, and software used for photography and drawing, and concluding with a list of factors to be taken into consideration in planning (319).

Analytical techniques range from traditional comparative studies through simple statistics to sophisticated multivariate analyses, and more than one method is usually employed. For example, Tappert’s study of pottery from the La Tène settlement at Straubing-Bajuwarenstrasse uses quantitative analysis and plots of feature combinations side by side with traditional comparisons with material elsewhere to develop a chronology for the site. Similarly, Kniesel and her colleagues, working with pottery from a fortified Bronze Age village in Bruszczewo (Poland), employ seriation and correspondence analysis for the development of a form sequence, correspondence analysis for the classification of wares, and GIS for exploration of site formation processes and changes in the use of a specific area over time.

Some studies offer encouragement in presenting ceramic material that at first sight seems very unpromising. Mielke, for example, reports the analysis of fragmentary Hittite ceramics, markedly deficient in overt chronological markers, from a steeply sloping site with a severely disturbed stratigraphy. Nonetheless, through a combination of plotting the frequency of wares and forms with comparison to stratified finds from Boğazköy-Hattuša, he was able to sort the material into chronological groups that could then be correlated with the site’s architectural phases.

Most of the papers use simple statistics that will be familiar to most archaeologists (e.g., estimates of vessel numbers, percentages). The four papers on multivariate analyses, however, are another matter, and are heavy going for those (like myself) with limited experience of this highly mathematical area of archaeological analysis. Macháček offers a detailed discussion of key concepts and terms and an explanation of statistical techniques (principle components analysis, cluster analysis); it constitutes a useful overview, though the German technical vocabulary will pose problems for Anglophone readers. Most enlightening, however, is Pavúk’s account of the history of multivariate analysis at Troy over a decade, providing insight into the evolution of the application of these techniques at the site along with a hardheaded assessment of their successes and failures. Pavúk’s paper is the clearest in showing how the analytical results, like the responses of the Delphic oracle, themselves require interpretation and pose new questions that can lead to significant breakthroughs in the understanding of the archaeological record. His summary (95) of what seriation and correspondence analysis can and cannot do should be studied with care by anyone planning to embrace these techniques. Even to the greatest skeptic, however, correspondence analysis emerges as a powerful heuristic device, identifying patterns and anomalies and drawing attention to areas that need further study.

These papers succeed admirably in providing a menu of approaches to ceramic analysis currently in vogue. Their ideal audience is a thoughtful ceramic researcher with significant hands-on experience with pottery and a firm command of both English and German, who is about to embark on a project involving large amounts of archaeological ceramics, especially prehistoric ceramics. Such a reader will find many explanations, models, case studies, and cautionary tales that will serve as a guide to choosing methods and approaches suitable for his or her project and in deciding “what should be done with a pile of sherds” (9).

Susan I. Rotroff
Department of Classics
Washington University in Saint Louis
Saint Louis, Missouri 63130
srotroff@wustl.edu