These two volumes are part of the Oxford Studies on the Roman Economy, edited by Alan Bowman (Camden Professor Emeritus of Ancient History) and Andrew Wilson (Professor of the Archaeology of the Roman Empire), both at the University of Oxford. Miko Flohr currently is lecturer in ancient History at Leiden University. The volumes, which together contain 34 studies on crafts, commerce, and trade in the Roman world, have their origin in conferences held in Oxford (in 2009 and 2011). The focus of Urban Craftsmen and Traders in the Roman World is on the social and physical environment of craftsmen and traders and the ways these shaped the decisions they made to realize their social and economic ambitions. Trade, Commerce, and the State in the Roman World has a related perspective, as the intended focus is on the institutions and requirements of the state. Inevitably, the individual studies approach these subjects in a widely varying way, and in particular in the latter volume, the role of the state is sometimes only vaguely present. The editors rightly emphasize that archaeological research over the past decades has caused a fundamental shift in our perception of the Roman economy, and hence most studies rely heavily on archaeological evidence. Rather than discussing the chapters of the two volumes one by one, I will examine a few important themes that are in my view central to the debate.

One such theme is the role of the state in shaping trade networks. Not surprisingly, several studies refer to the annona as an important factor in shaping trade networks. Tran (Urban) emphasizes that the social status of the navicularii in Arles was much higher than of the members of other collegia (e.g., stone cutters), so we should not generalize from this group of businessmen to all traders. The navicularii largely owed their wealth and public position to their involvement in the activities of the annona. Lo Cascio (State) argues that central and urban authorities were determined to guarantee the workings of a competitive market based on supply-and-demand mechanisms. While it is true that urban authorities often had no other choice than to rely on market mechanisms, because they had insufficient control over operators on the market beyond their territory, Lo Cascio’s argumentation offers a somewhat one-sided selection of sources, as the heavy-handed, and sometimes violent and random, intervention by urban magistrates in the urban food market shows.

The state distribution of staple foods like grain and olive oil is closely linked to the private trade in other commodities, if only because private trade often made use of the same shipments and facilities. Bowman (State, p. 42) suggests that transporters compensated for the low earnings of state contracts by maximizing the opportunities for “high-profit private” trade along the same routes. Whether state contracts were always low earning and private trade always high-profit is debatable, though. Adams, who offers a critical analysis of the evidence on state and private transportation in Roman Egypt, points out that transporters bringing grain to Alexandria were guaranteed profitable return cargoes (State, p. 203). Return cargoes are indeed an important element of trade, but one wonders what products were shipped from Alexandria to the grain-tax paying communities upriver.

Bonifay (State, in part. pp. 336–40) points out that ARS (African Red Slip) was not distributed along African amphorae, but along African grain. Hence, the founding of Constantinople, which meant that Egyptian grain was now largely diverted to the new capital city, significantly altered distribution patterns of ARS. However, he continues, how to explain in that case the wide distribution of ARS in the East after the end of the Egyptian grain supply to Rome? Bonifay suggests that ARS travelled with different cargoes over time and that we need more complex models to understand the link between state transport of foodstuffs and the commercial distribution of commodities. Privately traded commodities may have been redistributed in transit-harbors like Ostia and, thus, not directly reflect state shipping routes. Against the latter supposition, Reynolds (State, p. 378) points out that redistribution would have been burdensome and costly.

Analyzing the geographical distribution of stamped Samian ware, Fulford (State) observes that the emerging pattern is inconsistent with what one would expect purely on the basis of transportation costs. Rheinzabern, Lezoux, La Madeleine, and Trier all show the same pattern: Distribution was almost exclusively downriver on the Rhine and upper Danube, while increasing distance had no impact on distribution. Distribution is more widespread in Britain than in the hinterland of the production centers. There is clearly an involvement of military supply, but the pottery reached civilian as much as military consumers. So, why did these products not go downriver as return cargo for Mediterranean goods, one may ask? Fulford suggests that the stamps are related to state contracts to supply the army, but he also warns that the observed pattern would therefore only be that of stamped ware. This is an interesting suggestion, but leads to the question why the state would offer contracts for the supply of finewares, which was apparently not only meant for military consumers. The role of the army supply is clear, though, which is confirmed by the observation of Papi (State, pp. 433–34) that finewares ceased to arrive at Thamusida (Mauretania Tingitana) after the departure of the army units there.

Taking up Hopkins’ taxes-and-trade model, Bowman (State) argues against the widespread assumption that the overall tax burden of the Roman state did not exceed 10 percent, suggesting it may have been about 20 percent in the High Empire. He also argues against a general shift towards taxation-in-kind. Sirks (State) offers the intriguing theory that changes in tax-collection and the organization of state transport reduced their stimulating effects on trade. Private tax-companies accrued vast capital funds that were available for investment in commerce, and so the shift to direct state taxation took away these private capital reserves. Likewise, the development of shipowners’ involvement in transportation for the state from voluntary membership of collegia naviculariorum into a munus implied the tying up private capital in semi-public investment (see also Adams regarding Egypt).

Rice and Broekaert deal with capital investment in trade and the relationship of the political and social elites with the freedmen involved in business. In her study of traders’ strategies, Rice (Urban) warns not to overestimate the dependence of these freedmen on elite patrons. However, on the basis of epigraphic evidence Broekaert (Urban) convincingly argues that elite patrons invested in business and that, while independent freedmen undoubtedly existed, most freedmen worked for these patrons in a wide variety of forms and relationships and were dependent on their capital.

Another interesting debate concerns the clustering of crafts in urban contexts. On the basis of comparisons with modern examples, Goodman (Urban) points to the economic advantages of clustering and argues in favor of clusters of crafts and trades in certain Roman cities (Pompeii, Timgad, Rome). In the same volume, however, Dross-Krüpe notes an absence of clustering in the spatial patterns of cities like Ostia, Pompeii, and Oxyrhynchus. In a very informative study of bakeries, Monteix (Urban) analyzes the features of bakeries in Pompeii and other cities. Not one bakery showed all the features that can be found in bakeries in general. Hence, Monteix concludes that practical and economic considerations concerning capital investment and the use of space determined the outlay of individual bakeries.

While the attention of economic historians of later periods has long shifted to consumption as a vital factor in economic performance, regarding Antiquity production and distribution tend to receive more attention than consumption, owing to the nature of the archaeological evidence. It is therefore good to see that Van Driel-Murray (Urban) analyzes forms of footwear from the perspective of consumerism and fashion. The geographically widespread interconnectedness of the shoe-trade is visible in fashions appearing in the entire Roman world. Lacking data on how forms were distributed, interpretations regarding commercial aspects are inevitably somewhat speculative. For example, what is the effect of the mobility of people wearing certain forms on the spread of forms and archaeological patterns? A fascinating insight into the textile sector, whose products generally did not survive, is offered by lead tags that have been found in large numbers in Siscia and which are discussed by Radman-Livaja (Urban). These tags generally mention garment and price. Compared to prices known from literary sources and Diocletian’s Price Edict, the prices on the tags are too low to be purchasing price, so the author interprets them as services such as fulling.

Finally, several studies deal with the long-distance trade connections between the Mediterranean region and southern/eastern Asia and (sub-)Saharan Africa. The volumes and values involved should certainly not be underestimated. Graf (State) offers a detailed overview of the development of the so-called Silk Road, which in Roman times (probably already in the first century bc) consisted of a chain of exchanges involving middlemen such as the Sogdians, Indians, and Parthians. Traders from Palmyra were involved in trade relations with the East overland and over sea. Tomber (State) makes the interesting observation that the Red Sea ports of Myos Hormos and Berenike lacked the monumental facilities characteristic of Mediterranean harbors. The imperial authorities may have been interested in the custom dues levied on the trade through these ports (and in the marbles from the nearby desert) but apparently not in the ports themselves. Nappo (State) analyzes the finds of Roman coins in India and shows much more continuity of trade with India into the third century than previously assumed on the basis of the uneven chronological spread of the number of coins. As he shows, the picture changes if the value of gold and silver coins is taken into account.

Wilson (State) emphasizes the importance of the trade between the Sahara and Roman territory in North Africa. While he shows Roman interest and involvement in the region and the presence of Roman goods and the activities of traders, the element of scale is lacking in his argument and much of the proposed trade is in archaeologically invisible goods, in particular slaves. For example, we read that “appreciable quantities of pottery” are found. Realizing the difficulties of absolute estimations, this still begs the question in what sense exactly these volumes are “appreciable.” Papi (State, p. 437) rightly emphasizes that we cannot reconstruct the volume of imports on the basis of the pottery and amphorae evidence currently available.

The lack of quantitative modelling and impressionistic use of scale is a feature of several contributions. Rice (State, p. 97), for example, writes that the archaeological material leaves little doubt as to the significance of maritime transport in the Roman Mediterranean, pointing to thousands of shipwrecks, tens of thousands of ceramic sherds, glass products, etc. Not denying the significance of trade, one does wonder how many sherds are needed to show “significant” trade. Likewise, Adams (State, p. 204) states that the trade and transport on the Nile was “on a large scale.” Again, what does that mean? The rapid development of the size and quality of glass items does indicate the steeply growing importance of this sector, but one would like to know the significance of the fact that furnaces which produced “raw” glass had a capacity of 7 or even 15 tons. The latter amount, Foy (Urban, p. 267) adds, is sufficient for 10,000 large glass items. Again, what does this figure tell us about the level of consumption? Moreover, one should not underestimate the role of those sectors that did not leave many visible traces, such as the textile sector, which in later economies was the single greatest market of non-comestible goods.

Reading the diverse and wide-ranging studies in both volumes leads to the observation that the investigation of the Roman economy is still often caught in the primitivist-modernist debate. Many studies intend to show the incorrectness of the primitivist position, and their aim is therefore to point to the successes and achievements of the Roman economy, due to such factors as political and monetary unity, legal and social institutions, and the Pax Romana. Again, it is in particular due to archaeologists that we now realize the growth of the Roman economy, but the anti-primitivist stance occasionally leads to an unbalanced black/white approach. The objective should not be to determine if, but the extent to which growth was achieved, markets were integrated, and constraints were overcome. In 2005, I warned against overestimating the extent of the integration of the grain market in the Roman (Mediterranean) world. Rice (State, p. 110 n. 8) writes: “the integration of markets and subsequent specialization of production and trade are, however, visible in much greater quantities across the Roman world than Erdkamp allows.” This depends very much on how you define market integration (rather than simply as “commercial connectedness”). As just observed, which quantities of sherds are sufficient to indicate market integration? In the absence of good quantitative data on prices and volumes, an important criterium consists of the impact of trade connections on the structure of the economy. The integration of the grain market in Europe in the early modern period meant that it determined the structure of the economy of both Holland and the Baltic region. So, when archaeologists and historians talk about “integration” of the Roman economy, they should be clear what they mean by it. If the Kingdom of Saba’ introduces a coin type with the portrait of Augustus (Davidde, State, p. 582), this certainly proves Roman influence, but what does it mean economically? Other studies rightly point out the limitations of market integration. Harris (State, p. 232) concludes that the wood sector was regionally integrated rather than on an empire-wide scale. Most interestingly, Reynolds (State) points out that the well-known separation in western and eastern economic blocs in the Later Roman Empire originated in the mid-, possibly early third century. Focusing on the Levantine provinces, he shows that this trend was accompanied by a shift towards greater regional production and trade, but that there is no reason to see this in the light of an economic “crisis.”

All in all, the recent archaeological data, the integration of archaeological and historical evidence, the debates on its interpretation, and the many different approaches make these two volumes important contributions to the thriving field of research into the Roman economy.