This book emerges from one of the many fruitful colloquiums organised as part of the Oxford Roman Economy Project (OXREP), an effort led by the editors of the volume, Alan Bowman and Andrew Wilson, now in its seventh year. The general aim of OXREP is to appraise and interpret quantitative data for the Roman economy. This volume specifically addresses this mission within the context of urbanisation in the hopes that it will produce proxies – limited, specific sets of data that can be studied in detail in the place of bodies of evidence which are more complex and comprehensive – thus, ‘we can begin to analyse the major trends and patterns in economic behaviour’ (p. 6). The editors, and many of the volume’s contributors, believe that urbanisation is intrinsically linked with economic growth and prosperity. It follows that by determining the size of cities, their population, types of urbanisation, settlement patterns and, of course, changes in these and other measures over time, an indication of wider economic structure and scale can be obtained. However, the editors make it clear that the volume and type of data required for answering the critical macro-demographic questions of the subject neither are, nor ever will be, obtainable. Hence, the next best alternative is to engage in a cautious analysis based upon the available, albeit limited, quantitative data.
The first section features four chapters which are unified around the theme of ‘survey method and data.’ Simon Price’s contribution explores the value of archaeological field surveys and suggests that they can act as a reliable basis for estimating the population of Ancient Greek states, and, of course, by implication, urban settlements in the Roman empire. Price takes up the all-too-often-heard (and largely unanswered) call for more data before outlining several of the problems with previous attempts at creating proxies, such as Anthony Snodgrass’s surveys in Arcadia. His own proxy-based study is comprised of Mogens Herman Hansen’s area and remains-based demographic estimates combined with archaeological field surveys of surface scatters of pottery.1 In order to make his analysis work, the author assumes five persons per house. Other figures are extrapolated from census data from Hellenistic and Roman Egypt. While the author admits that this approach is an inexact science (indeed, harsh critics might even call these numbers arbitrary), Price considers this a better method than naive inductivism and putting ‘lots of dots on maps’ (p. 21). Indeed, Price’s method does give readers new material to work with, and it bypasses some of the problems with previous studies. His use of survey data will be of great interest to scholars who study population density in the Roman world. He ultimately concludes that estimates of population density, on the whole, need to be lowered.
Robert Witcher follows Price’s chapter with a contribution sensitive to the theoretical debates that often overshadow quantitative studies. He helpfully quotes Jack Davis, for instance, who expresses concerns about ‘massaging away’ real differences in empirical results in order to force data into prior expectations.2 At the same time, Witcher ultimately believes that the potential fruit of quantitative study is too beneficial to leave alone. Drawing upon case studies from a wide range of Mediterranean sites, Witcher’s study should cause scholars to re-examine the way that recovery rates are treated within demographic models. This chapter is particularly accessible to those with less than specialist knowledge, as the author takes the time to cover the state of existing debates before introducing his own model, which applies recovery rate estimates to survey results in order to come up with population figures. While Witcher stresses in numerous places that his model is not intended to be definitive, but rather is meant to allow opportunity to explore alternative variables, he ultimately produces population numbers which are lower than those in previous studies.
Peter Attema and Tymon de Hass collaborate on the final chapter in section one, which employs a case study from central Italy in order to, in the authors’ words, ‘contribute to the methodological debate on the use of survey data for past (Roman) population reconstructions’ (p. 97). The University of Groningen, the home institution of both scholars, recently surveyed 216 km² of the Tyrrhenian seaboard approximately 50 miles south of Rome. The authors take full advantage of this data by providing maps and charts which clearly and authoritatively communicate complex and varied quantitative data. In considering settlements in the area of Nettuno, for instance, the types of pottery, building debris and standing remains found place sites within categories of ‘possible’ or ‘certain’ in terms of whether they were occupied. Within these categories, further distinctions are made about the type(s) of settlements: villages, villas and farms, for example. Finally, population numbers are extrapolated from a range of periods from 350 B.C. to A.D. 400. These seem to indicate accelerated growth until the mid-first century B.C., when population plateaus before a decline of over fifty percent after A.D. 250. In the midst of such a stunning display, however, some concerns with the authors’ data and their interpretations remain. Attema and de Hass admit, for example, that their classifications do not necessarily reflect the original variation which was present from 350 B.C. – A.D. 400. Retrieval rates are given in rough percentages without an indication of where they originated. This leaves the reader wondering to what extent the charts and maps, though impressive, reveal reality or, in fact, dictate it.
The second section of the book, entitled ‘urbanisation’, commences with chapters by Neville Morley and Wilson respectively. These contributions represent the rhetorical heart of the volume, as within these two chapters is the debate most lively, engaging and concerned with the broader thematic and methodological issues outlined in the editors’ introduction. It must be remembered that the editors’ desire for the volume is to demonstrate ‘the current state of the debate’ (p. 4). Nowhere is this clearly shown than here; and indeed, the inclusion of more discussion and intersection along these lines may have only improved the collection as a whole.
Morley chooses to contextualise his contribution within larger debates, most notably the primtivist/modernist dispute – an unsurprising strategy, considering his theory-driven approach to the issue. After spending much of the chapter critiquing what the author sees as unsupported, if not assumed, connections between urbanisation and ‘economic development’ (even the term, which is itself argued to be laced with value judgements, does not escape criticism), he announces that what is needed is ‘a theory of (ancient) urbanisation and of the nature of the processes that supported the development of urban centres before we can attempt to delineate and quantify the parameters of the development, let alone begin to discuss the implications of this for the economy’ (p. 153). This chapter is a clear outlier among the volume’s other contributions, many of which, to varying degrees, press on with quantitative analysis rather than engage in the kind of theorising which Morley believes is vitally important. Only in the last page and a half does Morley finally ‘sketch’ his conclusions, perhaps the most apt term considering the rather deductive nature of his argument. Among other things, changes in urban systems may have had less to do with an increasing surplus and greater aggregate productivity, and may actually indicate differences in the way that existing surplus wealth was distributed.
In direct contrast to Morley, Wilson advocates a more positivist approach. The burden of proof, he argues, should fall to scholars who seek to show that urbanisation and economic performance are not linked (universality here is implied). With this preference in mind, Wilson thus attempts to estimate population within models which are ‘the least bad’. His chapter overwhelms the reader with proxies, charts and other data, including a count of building inscriptions over a seven hundred year period, several individual case studies and, finally, tables of urban area and population for cities throughout the Roman world. Wilson emphasises that population should directly affect our view of the size of the empire and the scale and performance of its economy, while simultaneously, and helpfully, offering scholars the very latest quantitative material for the Principate.
In a contribution chiefly concerned with how human behaviour affects material culture, Simon Keay and Graeme Earl base their analysis upon their AHRC-funded project which studied urban connectivity in Iron Age and Roman Southern Spain. Their chapter features an abundance of richly designed maps, charts and graphs, all of which support their method of examining the integration of archaeological and epigraphical evidence within a specified geographic area. This evidence casts doubt upon several relationships often assumed by scholars, such as the idea that greater ‘urban’ attributes in towns and cities should be automatically correlated with privileged legal status. Keay and Earl’s chapter is also especially sensitive to the methodological problems which have plagued past studies of similar topics: the unsuitability of the data for statistical analysis, over-simplistic interpretations, etc. From this small window into a much larger project, readers may find themselves interested in seeing the project’s final write- up.
The volume concludes with Bowman’s take on population and settlement in Roman Egypt. In a contribution which is both accessible and also helpfully contextualised within the past few decades of work on the subject, Bowman argues that population numbers in the region surged from the early Ptolemaic period until the mid-second century AD, when the Antonine Plague caused a decline. By the mid-third century a recovery is evident. Otherwise, Bowman is rather cautious, even in light of the rank-size correlations presented by Annalisa Marzano and John W. Hanson within this volume. The final paragraph of Bowman’s chapter also acts as a conclusion for the volume as a whole, where the author reminds readers of the central expectations of the project: ‘the cumulative evidence of a significant number of case-studies will give us the evidence, direct or proxy, for making more reliable quantitative estimates of key features of the Roman imperial economy’ (p. 351). This volume underscores the successful fulfillment of the first criterion, as an abundance of quantitative evidence has been accumulated and distributed – an outcome which will bear fruit for years to come. Much of this volume could also be seen to improve the potential for ‘reliable’ quantitative estimates, although more methodological debate and diversity would have greatly strengthened the collection.
As with previous OXREP outputs, this volume purposefully bypasses some of the grand theoretical and methodological debates about the very nature of the ancient economy itself. It will be up to each reader to determine whether this approach is ultimately appropriate; after all, theory arguably has a direct bearing upon the aims, methods and findings named in the book’s introduction as those of primary importance: the quantification of regional variation and integration, supply, demand, distribution and the structure and scale of ancient markets. However, within the quantitative framework provided by the editors and shared by many of its contributors, this volume ultimately achieves what it intends; that is, to assess and analyse quantifiable data on the Roman economy as well as to provide interpretations for how these data fit within wider categories of economic behaviour, institutions and processes.
1. M. H. Hansen (2006), The Shotgun Method: the Demography of the Ancient Greek City-state Culture, London; M. H. Hansen (2008), ‘An update on the shotgun method’, Greek, Roman and Byzantine Studies, 48: 259-86.
2. J. L. Davis (2004), ‘Are the landscapes of Greek prehistory hidden? A comparative approach’, in S. E. Alcock and J. F. Cherry (eds.) Side-by-Side Survey. Comparative Regional Studies in the Mediterranean World, Oxford: 22-35, at 33.