I spent the last week contemplating Kristina Winther-Jacobsen’s new monograph: From Pots to People: A Ceramic Approach to the Archaeological Interpretation of Ploughsoil Assemblages in Late Roman Cyprus. (Peeters 2010). This slim volume is an important contribution to not only the archaeology of Late Roman Cyprus, but also intensive pedestrian survey archaeology in the Eastern Mediterranean. Her analysis rests on the analysis of pottery from the Troodos Archaeological and Environmental Survey Project (TAESP) which is a descendant of the Sydney Cyprus Survey Project (SCSP). My project PKAP, is an ugly cousin of SCSP as well, and we both PKAP and TAESP approached intensive survey in similar ways. Both projects recognized that intensive data collection was an essential first step in interpreting ploughsoil assemblages, both projects used version of the chonotype system to document the ceramic material from the field, and both projects placed significant emphasis on a transparent approach to our procedures, methods, and conclusions. The biggest difference, is that TAESP was conducted on a regional scale, whereas PKAP sought to document a large coastal site and its immediate hinterland.
The central argument in her book is that Late Roman sites on Cyprus produce ceramic material in certain predictable ways. The consistency in the relationship between light and heavy utility wares, table wares, cooking wares, and transport vessels allowed the author to draw conclusions regarding the function of the various sites and their relationship to wider productive landscape. From the TAESP survey area, Winther-Jacobsen identified farmsteads, mining settlements, an agro-church (a church that played a role in agricultural production), a seasonal settlement, and a market village. She reinforces her arguments for the utility of these settlement types through comparison with other projects on the island and in Greece.
As has become my practice, I am not going to offer a full review (although I think that I’ll probably divide my remarks on this important little book into two posts), but instead offer some observations on her methods and conclusions.
- Formation Processes. The author paid particular attention to the way that formation processes contributed to the production of surface and ploughsoil assemblages and summarizes a good bit of relevant scholarship on these matters. Here her work parallels some of the important contributions of David Pettegrew who argued that the full-range of discard behaviors, curation techniques, and natural and cultural activities contribute to the assemblage of material in the plough zone. Framing the discussion of ploughsoil assemblages in the context of formation processes is vital to understanding the meaning and distribution of artifacts in the landscape. Winther-Jacobsen makes some good observations regarding breakage rates, use, and discard practices of particular types of pottery suggesting that cooking wares, which are particularly common in her various assemblages, endured particularly difficult life-cycles with many opportunities for breakage and discard. Heavier vessels (with the possible exception of transport amphora) tended to be handled less frequently in the household, in contrast, would have had longer life-cycles and lower breakage rates making them appear less frequently in the ploughsoil assemblages.
- Method and Procedure. Winther-Jacobsen makes clear that the ceramicist and other field archaeologists participates in archaeological formation processes when the define and document an assemblage for analysis and interpretation. To this end, she includes a detailed meditation on her own sorting and analysis practices. While it is commendable that she recognize the archaeologist as another participant in the life-cycle of an object, I would have been keen to understand in a more specifc way how her practices – from sorting, to measuring, to documenting – had an impact on the kinds of analysis and interpretations found in her larger study. Like many projects that recognize the importance of reflective practices, Winther-Jacobsen seems to stop just short of demonstrating the fundamentally arbitrary nature of “archaeological material”. In other words, a cooking pot does not exist outside of the unique interaction between the ancient potter, the Late Roman cook, and 21st century archaeologist.
- Typologies. Observing the arbitrary nature of archaeological knowledge, does not in any way detract from its meaning (except among scholars committed to increasingly tenuous views that privilege the rhetoric of objective). The author understand that typologies are utilitarian things that facilitate the answering of particular questions. As a result, the team from TAESP modified the typology introduced by SCSP called the chronotype system. I have blogged on the strengths and weaknesses of this method for documenting pottery endlessly over the past several years (just run a search over at the archive). Whatever its weaknesses, its strength for our project has rested in two areas: 1. we are forced to identify each sherd that comes from the field and place it in some kind of chronological and roughly functional category; and 2. our dataset is in some way comparable to data collected by SCSP and the Eastern Korinthia Archaeological Survey where the chronotype system was developed. Winther-Jacobsen is clear that her efforts to refine the chronotype system eroded its comparability between projects. In particular, her creation of three categories Transport Amphora, Heavy Utility Ware and Light Utility Ware created two types of pottery that incompatible with the implementation of the chronotype system on other projects where we tended to use Medium Coarse Ware and Coarse Ware to identify utility ware sherds. While Winter-Jacobsen was obviously free sort her pottery however she wanted, it was odd that she didn’t make a greater effort to make her new chronotype categories “backward compatible” with similar categories from earlier chronotype projects. This was particularly problematic because her argument rested, in part, on comparing her assemblage to projects elsewhere. For this to be meaningful, some allowance must be made to compensate for the different typologies. The inherent flexibility of the chronotype system, which tends to parse assemblages into very fine categories, would seem to be ideal for this, but the author did not necessarily maximize this comparative potential in the book.
- On-Site and Off-Site. While TAESP was a “siteless” survey project, Winther-Jacobsen’s dataset derived almost entirely from dense concentrations of material identified as sites. On the one hand, this makes sense: she was interested in documenting and interpreting assemblages and sites produce sufficiently robust assemblages for interpretation. On the other hand, the interpretation of sites has never really been a massive problem for survey archaeology projects. Over the past 20 years, the more substantial issue has focused on how we understand off-site material. In fact, David Pettegrew’s efforts to link formation processes to ploughsoil assemblages had less to do with the interpretation of distinct sites in the landscape and more to do with how we understand the activities that produced off-site scatters. In short, the “continuous carpet” of low to moderate density artifact scatters in the countryside represent a far more challenging set of formation processes and require a more sophisticated set of interpretative practices than the robust assemblages produced by high-density concentrations of material.
- Scalability. This difference between on-site and off-site scatters and their interpretation shines light on issues of scalability in the methods that the author advocates. Winther-Jacobsen advocates for near total collection of material in order to produce assemblages susceptible to the kind of proportional analyses that she advocates in this book. Various forms of total collection are common practice for most survey projects when documenting a site. For off-site scatters or massively extensive, high-density scatters like those encountered by PKAP on Cyprus, such time-consuming, storage-straining, analysis-intensive practices are simply not viable. Taking nothing away from the author’s careful typologies of sites, large-scale, “large site” scatters and the ubiquitous and monotonous extensive low and medium density scatters require some form of sampling technique if they are to be documented at all. Any form of sampling will make problematic the proportional analysis of ceramic types that Winther-Jacobsen demonstrated because sampling will almost necessarily reduce the level of complexity present in the assemblage and create ambiguity in the relationship between the sample and the putative total assemblage present on the ground.
None of these issues should take away from the significance of Winther-Jacobsen’s book. It represents on of the most thorough and systematic treatments of the analysis of Late Roman material from the Eastern Mediterranean and establishes some valuable comparative standards that other projects will want to consider as they make the move from pots to people in their analysis.