Bryn Mawr Classical Review 2005.01.17
Susan E. Alcock, John F. Cherry, Side-by-Side Survey: Comparative Regional Studies in the Mediterranean. Oxford: Oxbow Books, 2004. Pp. 251; figs. 114, tables 35. ISBN 1-84217-096-1. £45.00.
Contributors: Susan E. Alcock and John F. Cherry, Michael Given, Jack L. Davis, Nicola Terrenato, LuAnn Wandsnider, Stephen Thompson, Peter Attema and Martijn van Leusen, Tim Cunningham and Jan Driessen, James C. Wright, David L. Stone, Elizabeth Fentress, Sergio Fontana, Robert Bruce Hitchner and Philip Perkins, Robin Osborne, David Mattingly and Rob Witcher, T. J. Wilkinson, Jason Ur and Jesse Casana, Richard E. Blanton, and Jennifer Gates
Reviewed by Daniel R. Stewart, School of Archaeology and Ancient History, University of Leicester (firstname.lastname@example.org)
Word count: 3391 words
Arguably the greatest challenge facing survey archaeology is incorporating comparative data from various projects in order to build a meaningful regional perspective. The potential of such work has long been recognized, though rarely realized.1 The vagaries of individual projects' datasets and the lack of attention to the matter in the literature more often than not make this a Sisyphean task. Marked differences in collection methods and data recording, varied ecological, historical and political contexts, and diverse scales of operation and intensity all pose obstacles to direct survey comparison. So, while the questions have long been asked, this volume is the first substantive attempt at providing some answers. For those interested students and professionals with some background in the subject matter, it provides not only a very useful exegesis but also a starting point for much-needed discussion. It will no doubt prove to be as influential a volume as Keller and Rupp's Archaeological Survey in the Mediterranean Area.2
This volume comprises sixteen essays divided into six roughly thematic sections. These sections, though obviously linked in their reliance on survey data and their relevance to the broader issue of inter-survey comparison, are not so much guidelines for comparing surveys as groupings under which the problems of such an aim can be discussed. Consequently the sections are entitled: Introduction, Methodological Issues, Comparative Studies in the Mediterranean, Issues and Implications, Wider Perspectives, and Appendix.
The Introduction, written by the editors, provides a (very) brief overview of the nature of archaeological survey in the Mediterranean over the last twenty-five years. As introductions go, it does its job admirably, neatly stepping from the growth of survey to some of the methodological issues that have faced the discipline. The editors helpfully outline in ten bullet points the main procedural and theoretical advances, though admittedly some of these reflect their own research biases and might not necessarily appear on everyone's list of 'top ten survey advances.' It is obvious that some familiarity with the nature of the subject matter and the problems associated with survey is assumed, not unreasonably, as there is a minimum of explication concerning how these advances evolved or what problems they sought to address.
The Introduction then segues into an outline of some of the attempts at reconciling the multitude of problems related to data comparison. The success of this approach in other regions of the world, most notably Mesoamerica, serves to highlight the growing sense of disappointment that few successful studies have done the same for the Mediterranean. There are exceptions to this rule, but the broad consensus of Alcock and Cherry seems to be that, while comparative regional analysis can take many forms and utilise a variety of approaches, the dearth of substantive attempts in the Mediterranean needs to be addressed and is the logical 'next step' for the discipline.
After setting the stage in the Introduction, the volume moves to Methodological Issues. The first essay in this section is Given's, "Mapping and Manuring," which highlights obstacles to comparing sherd density figures between surveys. The title refers to the problem of reconciling varying occupation and land-use histories in disparate regions, specifically between regions that exhibit evidence for manuring, a process thought to facilitate the widespread scatter of ceramics as a component of household waste. The main thrust of the essay concerns the nature of pre- and post-depositional processes that affect the character of the surface scatter. Differences between regions in terms of visibility, chronology, survey intensity, and geomorphology will all affect the type and nature of the data recorded, and these problems can be roughly divided into those under the projects' control, such as interpretive and chronological framework and survey intensity, and those outside of the projects' control, such as visibility or ground-cover and geomorphology. Problems of chronology are perennial and not limited to survey, but, as the majority of the subsequent interpretation is based on the chronological framework employed, this paper would have benefited from a more in-depth discussion of the issue -- the notion of 'time' on an archaeological and cultural scale, and palimpsest models of sherd scatter, are thorny issues with no clear solutions but clear implications for the validity of interpretations. Given's solution is to rely on individual surveys to appropriately apply correction filters to their data, and he argues for an increase in the number of experimental tests, such as artefact seeding experiments, in order to improve the sophistication of these filters.
Jack L. Davis then discusses the issue of hidden landscapes in prehistoric Greece, the perceived imbalance between the prehistoric archaeological record as seen in surface finds and the supposed reality of the occupation. John Bintliff has long argued this point in print, and Davis seeks to address Bintliff's concerns directly by discussing the lithic data from three surveys and ceramic data from another survey, each with an eye on the data produced by the Boiotia project. Bintliff's main concerns seem to revolve around the reliability of data collected by other surveys, and Davis neatly addresses this difficulty by showing that, while direct comparison of data may not provide quantitative answers at this stage, qualitatively it allows for some resolution. This is not the most striking aspect of this paper, however. Instead, the implied critique of publishing practices, in terms of either completeness of data published or the timescales involved in moving from field-walking to press, seems to my mind to be the most telling argument. Davis' suggestion that the Internet is the key to meaningful comparative research is a theme that is picked up time and again throughout the volume.
The problem of survey intensity is discussed in Nicola Terrenato's contribution. Modern theoretical constructs, such as world-systems theory, highlight the need for regional analyses, yet practical advances in survey methodology necessarily constrain the sample size of recent projects. Increased survey intensity unavoidably leads to a slowing of the entire process, moving from hundreds into tens of square kilometres surveyed. This contradiction militates against the usefulness of survey in answering the regional-scale questions it sought to address in the first place. Terrenato helpfully discusses this issue in some depth, and suggests that instead of seeking regional solutions to regional interpretation problems, we should examine localized surveys to shed some qualitative light on global trends. He argues for keeping a macroscopic eye on factors that create bias while maintaining a large sample size, in essence bucking the trend established by the last twenty years of Mediterranean survey. The argument is certainly a compelling one, and it does seem obvious that larger surveys, in terms of sample size, are of more use in discussing larger regional issues than smaller surveys. It is unfortunate that Terrenato's insightful account glosses over the reasons why surveys have increasingly been sampling smaller and smaller regions; the survey of smaller regions evolved as a response to critiques of survey method and the reporting of bias in survey datasets. To return to that era without addressing the original concerns is less a solution and more another swing of the methodological pendulum.
Wandsnider discusses landscape paradigms and 'time perspectivism.' In essence, she highlights the shift in archaeological thinking in a general sense, and its particular effect on survey archaeology in the Mediterranean before moving on to an examination of current approaches to time. This latter aspect forms the most significant section of her essay as it makes explicit many of the problems and concerns surrounding the assumed temporal frameworks employed by many survey projects. For example, she outlines the problems and necessity of recognizing both long- and short-lived archaeological features that are introduced into the archaeological record at varying rates. She argues against the current paradigm of "regional/settlement pattern studies ... seated in flat-time functional metaphysics" (pg 59). In other words, it is not simply a matter of how an artefact or feature is deposited, but when and with what else. This is certainly a very interesting premise, though in the end this reviewer remained unconvinced by its wholesale rejection of the "flawed functional goal" of settlement pattern reconstruction. As a point of departure from the standard temporal analyses, however, this contribution certainly makes one think, and, while her solutions are not entirely satisfying, the problems raised are certainly systemic.
So ends the section on Methodological Issues, and the next section, Comparative Studies in the Mediterranean, presents five case studies that attempt to tackle some of the methodological issues outlined in the previous section. It certainly begins on a high note.
Stephen Thompson begins the section on Comparative Studies in the Mediterranean, with the first of five studies that attempt to tackle some of the methodological issues outlined in the previous section. His research was undertaken specifically to assess the validity of comparing data between and within surveys, using examples from Metaponto, southern Italy, to study the influence of variable visibility on the archaeological record. He traces the development of survey in the region over the last twenty years, highlighting the discrepancies and changes in methodology along the way (such as 'site' and 'non-site' recording, and the increased reliance on computerization of analyses). Before moving onto the actual case studies, he makes a strong case for re-survey -- especially in areas covered by surveys in the early 1980s -- as a tool for levelling the playing field between different datasets. His case studies stress the multitude of variables within the concept of 'archaeological visibility' and present a very sound approach that redresses some of the imbalance inherent in survey comparison.
Peter Attema and Martijn van Leusen present initial findings from the Regional Pathways to Complexity Project, a project specifically designed to compare survey data from three different regions of Italy: the Pontine, the Sibaritide, and the Salento Isthmus. Interspersed with the general discussion of the nature and background of their project are five 'boxes', each of which presents some preliminary findings on a specific theme. For example, one box deals with land systems and social strategies, while another discusses 'core' processes as evidenced in the three regions (these processes being primarily centralization, urbanization and colonization). It is an interesting layout that at times conflicts with the flow of the main argument, but this is a mere quibble. Especially useful is the explicit description of problem areas, as identified by the survey directors. These include problems concerning the definition of the object of comparison, the significance of bias, the limits of chronological frameworks, and the leap between archaeologically observable patterns and high-level socio-cultural processes. It is exactly this sort of explicit outlining of thought processes, problems, and attempted solutions that makes survey comparison easier for all involved. The only gripe would be that, as this is an interim report, the utility of some of what they suggest is as yet unproven.
One of the most frequent critiques aimed at archaeological survey is just how the surface and sub-surface remains relate. Tim Cunningham and Jan Driessen attempt to address this issue for Bronze Age Crete, by studying survey and excavation data relating to a specific theme. In short, they argue for the explicit framing of research hypotheses as a means of testing the relevance and utility of survey data for a specific research question. In other words, instead of undertaking a survey project in order to study the multi-temporal nature of settlement and land-use in a particular region, they argue for the application of survey data, in partner with excavation, to a specific question limited to a more constrained temporal framework: here, the development of 'palatial' culture on Crete. This is certainly something of an inversion of the standard survey research model. Unfortunately, it presumes the existence of multi-period survey projects in the first place, and therefore could only ever be used in tandem with standard approaches to survey. One interesting theme that runs through their paper, and is made explicit in the conclusion, is their rejection of the canonical 'final publication.' Instead, they argue for the general release of raw data, perhaps akin to that undertaken by the Pylos Regional Archaeological Project (PRAP), who have made their data available on-line.
One of the major developments in survey methodology occurred over twenty years ago, when surveys largely switched from extensive to intensive, essentially sacrificing large-scale inquiry for statistical reliability in terms of distribution of sites or scatters. James C. Wright seeks to integrate data from the two different forms of surface survey in order to assess the settlement pattern during the Bronze Age in the north-eastern Peloponnese in Greece. Wright argues that several considerations must be taken into account when attempting such a project: scales of examination in terms of space and time, vagaries in 'site' size (both absolutely and comparatively) and function, and a broad category relating to the disparities in the quality of the evidence. Like the authors of the previous paper, he too seeks to reconcile the survey and excavated material to help paint a coherent picture of activity within his study period. This paper certainly highlights the benefits inherent in splitting longer periods of time into smaller, workable sections, but it is unclear exactly how workable an approach this is for periods other than 'prehistory.' It is a method that is certainly highly dependent on a tight ceramic chronology. For example, it would by well-nigh impossible to divide the Roman period for many survey projects into Early, Middle, and Late, given the long use periods of some specific ceramic forms and the unclear relationship between the introduction of new ceramic types and regional histories.
David L. Stone's discussion of the state of survey archaeology in Tunisia, North Africa, redresses the northern Mediterranean bias of the case studies. He argues that Tunisia has one of the most comprehensive survey databases in the Mediterranean, with no fewer than 19 recent projects. Of these, however, only one has issued a final publication. Thus, the picture he paints of landscape, land use, and settlement history is necessarily built on interim data and interpretations. This is not a criticism, far from it; this case study underlines the usefulness and potential of this data and makes a strong case for comparative studies of preliminary data. Indeed, considering the usual time-scale between when the data is collected and when it is released for public consumption, Stone successfully underlines the need for exactly this sort of research.
The next thematic section concerns Issues and Implications, and deals with such varied topics as African Red Slip (ARS) ware, demography, and the cartographic representation of survey data and its interpretations. Fentress, Fontana, Hitchner and Perkins open the section with a discussion of the distribution of ARS and its utility as a marker, both in terms of chronology and settlement patterning. Central to the essay is a re-assessment of a production curve produced by two of the authors fifteen years ago. This results in the correction of some of the dates associated with the ARS supply, though, overall, general trends remain. The authors then move onto testing the new data against what is known of settlement patterns and site numbers in both Sicily and Tripolitania. The most interesting trend to come out of this analysis is the almost total lack of correlation between site numbers and the number of sherds recovered per site. The '3rd century crisis', as evidenced in the ARS production curve, is never reflected in the occupation of sites, although differences in overall sherd density and scatter size are not discussed and may be useful here. The authors conclude with a brief discussion on the nature of the ancient economy, showing how the distribution of ARS, and likely other, as yet unstudied, artisan productions, is a key demonstration of the integrated nature of the Roman economy.
One of the perennial problems with survey data is its application to demographic analyses. While levels of population are undoubtedly important to issues such as production and consumption, urbanization, and centralization, the place of survey in assessing the quantitative levels of a population is far from secure. Robin Osborne attempts to gauge the applicability of the survey data for just this purpose in the Greek world. The reluctance with which most recent survey projects in this region attempt to quantify population is a measure of how successful he is. He certainly provides a rather cogent analysis of the various approaches used in quantifying demography and the multitude of problems. His conclusion is a rather wary caution about the utility of survey in this respect: it is no more useful than the traditional historian's demographic tools, and more importantly the present use of survey data for this purpose cannot be divorced from these traditional tools. The study certainly presents one of the best examples among recent publications of the integration of historical and archaeological information, underlining the point that comparative studies are not necessarily just inter-survey, but inter-disciplinary.
This section concludes with David Mattingly's and Rob Witcher's paper, a critical assessment of the current state of cartography relating to the ancient world. It begins with a discussion of the Barrington Atlas and its relationship to landscape archaeology, before moving onto issues of representing settlement density visually. The authors attempt to reconcile the sometimes contradictory requirements of mapping at a micro- and macro-scale, while incorporating new theoretical points of view. For example, they argue for a shift away from the mapping of 'Roman' finds in 'native' areas to the exclusion of local traditions. They make the logical suggestion that the correction of these implicit biases begins with the survey.
The penultimate thematic section, Wider Perspectives, presents case studies from other areas of the world, and is a welcome opportunity for readers to correct their 'Mediterranean myopia', to steal a phrase. The first contribution, by Wilkinson, Ur and Casana presents a study of settlement patterns in the northern Fertile Crescent of the Near East in the Bronze Age. As the majority of the discussion is drawn from an analysis of extensive surveys, this paper is an interesting counterpoint to the intensive survey imbalance seen in earlier selections. The authors note a significant disparity in settlement densities in the northern fertile crescent when compared with both southern Mesopotamia and the bulk of the Mediterranean. Their incorporation of a wide variety of environments, with resultant differences in site formation and preservation, should be something of a wake-up call for projects in the Mediterranean that purposefully exclude areas deemed 'marginal'. As a case study for comparative research, in fact, it is hard to find fault with either their methods or their conclusions.
The final paper (barring the Appendix) is Richard E. Blanton's cross-cultural examination of settlement patterns and population change in Mesoamerica and the Mediterranean. He states that the majority of studies of settlement patterns focus on regional and temporal differences within civilizations, and he seeks to assess the validity of doing this between civilizations. He takes data from thirteen Mediterranean and eight Mesoamerican surveys in order to compare changes in settlement density among similar lines (center and nearest-neighbour analyses). His analyses focus on agriculture, population, and the evolution of what he terms 'commercialized landscapes'. While cross-cultural comparisons sometimes boil down to a catalogue of differences without interpretation, this is an impressive study into the application of similar research methods and theories to disparate areas. The inclusion in two mini-appendices of the raw data heightens its appeal.
The volume concludes with the Appendix, complied by Gates, Alcock, and Cherry. This lists Internet resources for survey in the Mediterranean, ranging from individual projects' websites to practical handbooks on the presentation of data. It is arranged alphabetically by region, and, while this list will almost certainly become out of date fairly rapidly (it was compiled in 2002-2003), it nonetheless offers an excellent starting point for people seeking more information on survey online. Ideally, the list would exist online itself, allowing for continued updating and making it a far more useful resource.
In conclusion, this volume is a welcome addition to the all too scanty corpus of works dealing directly with survey comparison. The book is handsomely produced, and the tables, plans, and maps are all clear and relevant -- a point of praise that is all too rare in survey literature. While some prior knowledge of the broader issues of survey is largely assumed, the vast majority of the papers are fairly accessible and well documented. And though the contents of this volume may not find agreement among all, it will be impossible to ignore.
1. See Cherry's (1983) 'Frogs Round the Pond', in D. Keller and D. Rupp (edd.), Archaeological Survey in the Mediterranean Area, Oxford: BAR International Series 155, pp. 375-415.
2. See note 1.