Roman Seas: A Maritime Archaeology of Eastern Mediterranean Economies

As a point of departure, this comprehensive volume asks how maritime contacts influenced the economic and social development of ancient Mediterranean communities. Leidwanger draws inspiration in the overlapping and contrasting traditions of F. Braudel, P. Horden, N. Purcell, and C. Broodbank: can the ancient Mediterranean be conceived as a singular unifying sea, one of “intense fragmentation” and “microregions,” or perhaps connected through interdependent seas (1–6)? Leidwanger asserts that the mare nostrum facilitated various roles in shaping life along its shores, emphasizing its diverse marine environments—the “Roman Seas” of the title. He contributes to a new generation of scholarship on connectivity, however, in stressing that understanding different scales of movement within and between these seas is of fundamental importance to answering the study’s central question.

Leidwanger’s primary objective, laid out in chapter 1, is to illuminate how, in the northeastern Mediterranean, economic regions emerged around maritime space and how they changed over time from the second century BCE to seventh century CE. He is one of a growing group of scholars who see the applicability of social network analysis (SNA) for understanding such maritime connectivity. In particular, SNA can be used to model the scales and diachronic economic patterns generated in specific maritime topographies. Network-based approaches assume that relationships between entities matter: Leidwanger argues that ports represent but one useful dataset in this inquiry, while shipwrecks (and the geographical origins of their cargoes) represent another.

In chapter 2, Leidwanger presents his model’s physical parameters: the region’s marine environment of coastal topography, currents, and winds and how these could impact ancient seafaring; and the types, size, cargo capacity, construction, and crew size of fundamentally similar Roman and Late Antique ships—with the latter group’s source material derived from shipwreck assemblages supplemented by iconography and ancient texts. The chapter’s second half applies these parameters to illustrate the effective speed, probable distances, temporality, and seasonality of navigation in the northeastern Mediterranean (assisted by data from the sailing trials of Kyrenia II, a replica of the fourth-century BCE ship wrecked off the northern coast of Cyprus). Leidwanger does not subscribe to the long-held concept that sailing ceased during the winter months (mare clausum [64]) but rather follows more recent scholarship. Although prevailing offshore winds dominate, coastal topography, local winds, and currents could be exploited year-round based on the experience and local knowledge of mariners and the availability of diverse vessel types. 

In chapter 3, the theoretical background that Leidwanger applies to construct his network model of maritime interaction is problematized through concepts of “regions” and “regionalism.” Building on the previous chapter’s parameters, the perspective is further developed from the nomenclature given by ancient authors to seas within the Mediterranean, such as the Mare Carpathicum and Mare Cyprium (70), which were assumed in the administrative conventions of the Roman empire. He argues that this reflects a “common sense geography” (76) where the topography and patterns experienced by ancient mariners were reflected in shared maritime spheres with informal boundaries that affected navigational practices. These, in turn, could impart shared identities to communities and distinct related economic mechanisms. Leidwanger harkens back to C. Westerdahl’s pioneering maritime cultural landscape approach in stressing that models must include the physical environment to underscore the variable scales, temporalities, and geographies of these regions. Therein, shipwrecks and ports illustrate the range of human mobilities through trade, travel, and fishing. 

Complementary approaches are applied in chapter 4 to analyze the study’s main dataset of 67 shipwrecks (34 Roman, 33 Late Antique). The first approach quantifies the shipwreck cargoes, from which two discrete groups appear, separated by a drop in numbers between the third and early fourth centuries. Leidwanger points out that this lacuna is not present in Mediterranean-wide trends, highlighting the disparate regional reactions to the events of the third century. The second, using the network analysis software Gephi, creates social network models of the geographical links derived from ships’ cargoes from numerous origins. During the Roman period, the Aegean appears to be central, with a few connections extending across Cyprus, Cilicia, and the northern Levant to southern Gaul, the Black Sea, and the Adriatic. The Late Antique network has a more diverse number of connections, with a focus on the Aegean, but with Cyprus, Cilicia, and the northern Levant more central. The third approach applies ArcGIS-based analysis of these networks to illustrate the types and lengths of journeys. This is where the environmental parameters and navigational capabilities come together, representing the network links using likely sailing times. For example, a ship sailing from Rhodes could reach Paphos in three days, while a return voyage might take nine days. Such journeys indicate the temporality (and unevenness) of maritime interaction, ultimately impacting our understanding of transport costs, regularity, and hence, investment.

Chapter 5 integrates contemporary ports (as nodes) and their hinterland communities into the analysis. After discussing the methodological challenges and archaeological evidence for the broad spectrum of port types and their distribution, Leidwanger presents two case studies. The ensuing examination of the “miniature archipelagos” (166) of the Datça Peninsula in southwest Turkey and south-central coast of Cyprus is extremely detailed, reflecting regions he knows well from 15 years of fieldwork. The aim is to look inward at hinterlands and local communities that were economically oriented around these ports, as much as outward at their maritime connections, or regular access to exchange. ArcGIS enables calculations of time for travel from a hinterland settlement to a port. Leidwanger concludes that there existed numerous small “ports” (anchorages and landing spots without permanent shoreline structures) that often go undetected in not only archaeological but also network studies, and that such sites extend the reach of maritime networks inland, as they facilitated delivery of the agricultural products represented in shipwreck cargoes. 

The volume’s final chapter picks up these two observations to analyze the structures and shifts of the region’s maritime networks. The economic downturn in the third century is an oft-cited explanation for the decline in the numbers of Late Antique shipwrecks in the Mediterranean, but in the northeast Mediterranean, the busiest period of activity is the late fourth to fifth centuries, and Leidwanger stresses that falling numbers do not mean shrinking networks. In this region, mid-Roman-period cargoes largely originate from central and western Mediterranean networks, while Late Antique links are more spatially restricted to the eastern Mediterranean, the Black Sea, and the Aegean. The shift in Late Antiquity is clearly due to the ascent of Constantinople: ships may not necessarily be destined for the southeast Aegean but only passing through (such as those transporting grain between Alexandria and Constantinople), representing the region’s placement in the middle of an interregional network not visible in the local archaeological record. Small vessels navigating between inconspicuous ports dominate the dataset, with their cargoes highlighting the ties between the “busy countryside” and distribution hubs. 

In tackling the complex topic of modeling the diversity and scales of the maritime movement of the Roman economy, Leidwanger admirably demonstrates the applicability and versatility of maritime archaeological data (ch. 2). Indeed, the study’s secondary aim is to integrate shipwrecks into mainstream ancient economic studies, and at the same time expand the theoretical and methodological approaches of traditional shipwreck site and quantification studies. Appendix 1, the dataset of 67 shipwrecks, is a significant contribution in itself, reflecting scrupulous reassessment (including firsthand reinvestigation of some ceramic cargoes) of the main shipwreck databases of A. Parker and the Oxford Roman Economy Project, among others. Thirty-four wrecks were previously included in existing databases, and the rest presented here can be considered new. 

This approach, however, does not neglect the informative data concerning the type and archaeological identification of “ports,” especially those Leidwanger calls “inconspicuous” (159). Unfortunately, the initial presentation of the data is placed just prior to the analysis (ch. 5); a more evenhanded treatment early on, alongside shipwrecks (ch. 1), might have helped clarify the study’s methodological approach to its datasets. This is but a minor concern in an otherwise logically structured and well-written volume. I applaud the inclusion of fishing as an example of maritime mobility (ch. 3). Considered largely invisible in the archaeological record, it is usually left out or mentioned only in regard to the trade in its secondary salted fish products. But how fishing can be included in network analyses—if it ever can be—is not realized. 

With this volume, Leidwanger makes a strong case that the dominant, urban-oriented, top-down, port-hierarchical view of maritime networks—of long-distance trade between large ports feeding short-haul routes to smaller ports—no longer holds water. Through his case studies, he convincingly demonstrates that the key for understanding the dynamics of complex maritime networks is from the bottom up, an approach that highlights geographical flexibility, temporal variabilities, and regional scales different from those that the extant Mediterranean-wide models suggest. The multiscalar methodology is certainly transferable to other regions of the Mediterranean: it is not Braudel’s one sea “sixty days long,” but “a variety of seas scaled to their own particular communities, mariners, and mechanisms” (218). 

Athena Trakadas
Environmental Archaeology and Materials Science
National Museum of Denmark
athena.lynn.trakadas@natmus.dk