Some New Thoughts on the Roman Economy
Over the weekend, I finally found a few hours to sit down with the relatively recent edited volume Quantifying the Roman Economy: Methods and Problems edited by A. Bowman and A. Wilson (Oxford 2009). The book brings together a number of different perspectives on the Roman economy in a broad response to later chapters of the Scheidel, Morris, and Saller edited Cambridge Economic History of the Greco-Roman World (Cambridge 2007). In my reading, the books stands in contrast to a recent work edited by M. Mundell Mango on Byzantine Trade (which I discuss here). Whereas Mundell Mango theorizes that it is possible to understand trade in the Byzantine world without necessarily appealing to wider considerations of the ancient economy, the authors in Quantifying the Roman Economy take the opposite approach and embed trade of all kinds within a theoretical and material critique of the Roman economy.
While I won’t review the entire book, I did want to point out some of its highlight to my loyal readers.
1. Bowman and Wilson’s introduction is among the best short summaries of the state of research in the Roman economy. Their considerations range from discussions of economic integration to survey of the potential of ancient economic growth and decline. They conclude their survey by focusing attention on four vital areas for analysis: demography and settlement, the agrarian economy, production and trade, and mining and metals. They argue that at present there exists sufficient evidence to support sustained analysis of these issues and that these issues can form the basis for an integrated view of the Roman economy.
2. Field Survey and Demography. Intensive pedestrian survey represents an important approach for establishing Roman settlement patterns, and these settlement patterns play a vital role in the organization of the Roman economy. In particular, the relationship between rural producers and urban dwellers structures the relationship between the primary production of food and centralized administrative, political, and population centers across the Roman Empire. As Jongman, Fentress, Mattingly, and Lo Cascio point out, the percentage of people living in both cities and in the countryside remains hotly contested. As a result, it is difficult to evaluate even the minimum and maximum productivity of the countryside required to sustain an urban population who is not engaged in primary agricultural production.
3. Peopling the Countryside. Elizabeth Fentress and David Mattingly provide valuable defenses of survey archaeology and its ability to shed light not only the structure of ancient settlement but ancient demography. Fentress argues on the basis of her intensive survey work on the island of Jerba and in the Albenga Valley that careful sampling of the landscape can provide a rough estimate of both the kinds and the distribution of sites in the countryside during the Roman period. The types of sites, ranging from urban areas to small villages and isolated farms, could then form the basis for basic demography. To summarize complex and nuanced study, Fentress argues that far fewer people lived in the countryside on Jerba than we might expect considering the potential density of urban settlement: 11% in single farms, 20% in villages, 20% in villas, and an impressive 49% in towns. She was then able to argue that the urban centers on Jerba (which is not a particularly fertile place) relied on imported grain.
In his response to the Fentress article, David Mattingly rightly offers a bit of caution by cleverly invoking Donald Rumsfeld’s category of “unknown unknowns” in intensive survey. For Mattingly, the unknown unknowns are those sites that do not manifest themselves in survey but may have a significant impact on how we understand ancient demography and settlement structure. Of course, Jerba with its light soils and relative geomorphological stability was less likely to produce the kinds of unknown unknowns than the more dynamic landscape of, say, the Rhone valley, but nevertheless, Mattingly is correct in reminding us that survey is better at demonstrating presence than absence.
4. Trade. Andrew Wilson’s summary of pressing issues with regard to Roman trade is another very useful contribution to any discussion of trade in the Mediterranean. He offers valuable critiques of evidence for trade ranging from shipwrecks to amphora and marble. In his study of shipwrecks, he uses aoristic analysis to create a more nuanced reading of Parker’s classic summary of shipwrecks by century. He shows that by plotting the possible date of “long-dated” Roman period (150 BC – AD 400) shipwreck by decade rather than by midpoint, it becomes possible to argue for a later peak in maritime commerce than Parker had estimated. In short, distributing the possible dates for long-dated shipwrecks helps to mitigate against a chronological pattern of trade biased by certain standard dating conventions.
Later in the same article, Wilson provides another useful model for understanding Roman period trade when he compares the production of certain classes of pottery (e.g. African Red Slip) to its frequency elsewhere in the Mediterranean. While such analysis is not particularly novel or innovative, he establishes quite clearly how the relationship between production and distribution is not fixed. Pottery supply represents only one aspect of the distribution of ceramics in the Mediterranean, and the quantitative gap between patterns of supply and distribution provide a useful basis for considerations of trading patterns as well as the vagaries of taste across the Mediterranean basin.
William Harris and Michael Fulford offer responses to Wilson’s contribution that expand the variables under consideration in his article to include the relationship between settlements in the Roman world and how the differences between overland and maritime trade and urban and ex-urban settlement types can significantly influence the distribution of material.
As my brief summary of this books probably makes clear, I liked this book and think it is the best single volume summary of the pressing issues and potential for using quantitative data to understand the Roman economy. As the availability of quantitative data from survey projects, excavations, and summary publications increases, scholars will need more robust models and approaches for producing synthetic analyses of trade, settlement structure, demography, and economic growth or decline. Despite the typical caveats surrounding the use of any quantitative data from antiquity, this volume has continued the optimistic trend begun with the Cambridge Economic History.