Bryn Mawr Classical Review

Reviewed by William E. Metcalf, New Haven, CT (wemetcalf@gmail.com)

Version at BMCR home site

Table of Contents

This book is the product of a conference held in Brussels in 2009. The papers, by many leading students of the ancient economy, are both wide-ranging and narrowly-focused. The range in fact extends from the eighth century B.C. (Ober) to the collapse of the Roman empire (Bagnall, Temin), but both the questions and the models for responding are a common thread. Is it possible to estimate ancient GDP? Were ancient Romans more or less prosperous than mediaeval and early modern peoples? What can we do in default of actual statistics and highly imperfect evidence? The answers are almost all unguardedly optimistic, though both Wilson and particularly Morley sound salutary cautionary notes.

Herewith a summary, limited by space considerations.

F. de Callataÿ, "Long-term quantification in ancient history: an historical perspective" (13-27). This is a wide-ranging essay on the development of historical studies in the 20th and early 21st centuries, with prognostications for the future. Specifically to the volume's title, quantification is likely, in the author's view, to be used more and more in ancient history, for four main reasons. 1) It is new, and supported by 2) massive amounts of new data from the archaeological world and 3) spectacular development of scientific tools; and 4) quantification is no longer taken simply as a technique used for investigation (or for lying), but has been recognized as a topic of investigation in itself. Quantification is particularly relevant to economic history where, even if our statistics are weak in modern terms, they are sufficiently robust to say something valid.

N. Morley, "Orders of magnitude, margins of error" (29-42). This is a largely cautionary essay. As Morley puts it, most studies of "quantification" focus on the construction of abstract models of high-level economic structures and "quantify" them in the sense of assigning quantities to different factors, regardless of the availability and quality of evidence. "They start not from a set of data amenable to quantitative analysis, but from an idea, usually derived from modern economic assumptions, of what they wish to know; they then seek suitable data, usually drawing on proxy evidence or comparative material because ancient sources are inadequate for the purpose" (34). This practice is compared, negatively, to that of Duncan-Jones, who started from data. This paper is too provocative to discuss at great length—it almost requires a review of its own—but its sentiments (largely skeptical) are salutary in exploring other sometimes extravagantly optimistic articles in this volume.

A. Bresson, "The ancient world: a climatic challenge" (43-62). Starting from Gibbon's supposition that the freezing of the Rhine facilitated barbarians crossing into the empire, Bresson examines whether this, or any other citation of remarkable weather from antiquity, will withstand scrutiny. He concludes that, for example, Herodotus' observation of the harsh weather in the Cimmerian Bosporus might be supported by large climatic phenomena, and cites numerous examples of the freezing of the Rhine. Thus no statement of an ancient author is to be dismissed outright.

G. G. Aperghis, "Creating a long-term model for an ancient economy" (63-78). This essay illustrates some of the pitfalls. No ancient source records the number of sheep and goats in ancient Mesopotamia, but prices are available for sheep, goats, and wool and their reproduction rates, their consumption by humans (these two no doubt derived from modern analogies) and tax rates. Now in modern Iraq the number of sheep is about 80% of the human population; as Aperghis puts it, "This figure would be subject to a great deal of uncertainty if applied to ancient Mesopotamia had it not been that about 80% also applies to the adjacent countries of Iran and Syria. This leads to the idea that this is likely to be an appropriate level in an agricultural society . . . under the prevailing conditions of terrain and rainfall, which have probably not changed significantly since antiquity (emphasis added) (65)." This last seems a gratuitous assumption. But it is of such stuff that the author proceeds to create the complicated models illustrated at pp. 71 and 73-75, leading to estimates of population, annual revenue and cost, and personnel numbers, both civilian and military, at the beginning and apex of the reign of Antiochus III (223, 197-190 B.C.). We might almost overlook the statement (77) that "these are best estimates produced by a model based on uncertain data and will naturally be different if the values of some parameters change."

R. J. van der Spek and B. van Leeuwen, "Quantifying integration of the Babylonian economy in the Mediterranean world using a new corpus of price data, 400-50 BCE" (79-101). The essay suggests, on the basis of prices, that Babylon had a limited long-distance trade. The argument benefits from the Astronomical Diaries, which include the prices of six commodities, most importantly for their purpose barley; in addition, eighteen new clay tablets containing lists of the same commodities have recently been recovered. The lack of trade made markets, which are volatile, susceptible to exogenous supply or demand shocks, since the apparatus for relieving either was not well-developed.

J. Ober, "Greek economic performance, 800-300 BCE. A comparison case" (103-122). Three arguments are made: the Greek economy grew steeply and steadily from ca. 800-300 B.C. both in size and aggregate consumption; despite dense population and remarkable urbanization by the fourth century, living standards remained high; and wealth was distributed relatively equitably, with a substantial middle class. In calculating growth rates, Ober draws heavily on the work of Morris ("Economic growth in ancient Greece," Journal of Institutional and Theoretical Economics 160.4, Dec. 2004, 709-742) and uses such proxies as house size, the progressive increase in number of names appearing in LGPN, and growth in size of coin hoards, both of which may be due to factors other than economic growth. His result is a per annum growth rate exceeding that of Rome; unfortunately no strictly comparable evidence is available for Rome, which in any case was a different (i.e. centralized) society.

G. Kron, "Comparative evidence and the reconstruction of the ancient economy; Greco-Roman housing and the level and distribution of wealth and income" (123-146). The author concludes that real wages could be quite high at Rome, making for a broad middle class, and that this is reflected in house sizes despite broad inequalities. He paints a picture similar to, and more broadly based, than that of Ober. Useful comparisons to modern economies are provided.

A. Wilson, "Quantifying Roman economic performance by means of proxies: pitfalls and potential" (147-167). The author is generally critical of uses heretofore made of proxies, either because of small sample size or the use of assumptions and guesswork; thus, reconstructions of the economy by such prominent scholars as Keith Hopkins or Richard Saller are dismissed out of hand. "All attempts to calculate Roman GDP per capita have relied on a battery of different figures as inputs to the calculation of which almost none are actually known. What you get out of such a calculation is entirely a function of the assumptions you put in" (149). Still, Wilson sees a future in ice core studies and in a couple of data sets developed by himself, fish-salting factories and water mills, the latter of which brings literary evidence to bear. His view is cautiously optimistic.

W. M. Jongman, "A new economic history of the Roman empire" (168-188). Jongman marshals too many arguments to discuss here in support of his view that the Roman world was one of increasing prosperity right up to the Antonine Plague, which began to destroy the late Republican and early imperial equilibrium of urbanization, market integration, agricultural specialization and prosperity for many. After this he sees the "second serfdom," in which landlords extracted by force what market forces could not give them.

P. Temin, "Price behaviour in the Roman empire" (189-207). Temin makes some useful observations about prices, distinguishing between market prices and administered prices: the former change frequently in a "random walk," the latter only seldom. An illustration of the latter is army pay, which he assumes to have been increased only when it became dysfunctional, i.e. was insufficient to meet the cost of goods. It therefore provides a kind of check on market prices. He takes the common view, of low inflation in the early empire, but also looks at the Antonine Plague as a watershed event: it reduced the population without reducing either arable land or the money supply; it left more money per capita, but there were fewer workers so that grain and other agricultural production were scarcer than before, relative to the amount of money in circulation. This "surplus" money was used to bid up the price of goods. It should be added by the way that Temin's is the only piece in this book that takes serious account of developments in the coinage.

W. Scheidel, "Roman real wages in context" (209-218). "What does all this tell us about "Roman" real incomes? In as much as we are able to say anything at all, there is no sign of historically elevated real wages during the Roman period that were driven by economic development rather than population loss" (214).

R. Bagnall, "Late Roman data collection" (219-228). This paper deals not with whether governments collected data (they did) but with whether or not they were, in the words of A. H. M. Jones, "statistically minded" (they were). They collected, and reported, data on a myriad of matters, ranging from population and army disposition to real property and numbers of cattle. One question is how far up the line this information travelled. Surely, the population of Egypt, say, was reported to Rome, so that someone knew the figure, and Josephus (who is thought to have overestimated it) could have known the right figure—or he may be right. The appetite for statistics grew with the later imperial bureaucracy, and various tabulations are produced in support of a statistically-minded imperial administration.

E. LoCascio and P. Malanima, "Ancient and pre-modern economies' GDP in the Roman empire and early modern Europe" (229-251). The authors attempt to determine whether the "ancient Greeks or Romans were richer or poorer than medieval and early modern European inhabitants." They acknowledge the commonly assumed rate of consumption of wheat (250 kg per capita per year) and its price (3 HS per modius), but question the way in which comparisons with pre-modern economies are constructed. They attempt better estimates for cereals, and conclude that the GDP in Roman times was higher than generally thought. This has an effect on the general model of economic development before the Industrial Revolution: of a long slow rise or stability with cycles, as the authors believe.

The volume's standard of production is high, though the review copy missed four pages in the final essay; more generally, it is easy to identify those contributions not written by native speakers of English—solecisms abound.