Maritime Networks in the Ancient Mediterranean World

Growing out of a 2013 workshop at the University of Toronto, “Networks of Maritime Connectivity in the Ancient Mediterranean: Structure, Continuity and Change over the Longue Durée,” this volume fits within the growing body of scholarship on Mediterranean maritime networks that has appeared since the turn of the millennium. It opens with an introduction by Leidwanger and Knappett, setting out the aims, concerns, and questions that create a framework for the following chapters. Through the innovative application of formal network analysis to archaeological and historical data, their aim is to advance the use of analytical tools in network science as it applies to maritime interaction, while recognizing that this “toolbox” will likely differ depending on the scale of analysis and the genre and completeness of the data. The editors’ approach moves beyond traditional, binary notions of land and sea to conceptualize the Mediterranean as a living canvas made up of various “scapes” embodied with social memory. They bring together a group of authors who represent a range of chronological and geographical foci, broad interdisciplinary approaches, and overlapping scales of analysis.

Chapters 2 and 3 begin by introducing the reader to the vast repertoire of network models. Tim Evans (ch. 2) lays out the strengths and weaknesses of an array of spatial network models, enabling the reader to make an informed decision about the most suitable model based on their data and scale of analysis. This includes his own collaborative endeavor, ariadne, that digitally models networks and has the flexibility to allow data-specific inputs and outputs that better reflect real social systems (28–31). Ray Rivers (ch. 3) picks up these threads through a practical demonstration of the different models used for the networks before and after the Theran eruption. While other models (e.g., simple gravity and radiation) emphasize the discontinuity of post-eruption networks, the ariadne program accurately simulates the continuity visible in the archaeological record. It demonstrates that, without Thera, Late Bronze Age networks were not diminished but rather reformed to use Phylakopi as a link between the Cyclades and other regions. As a cautionary note, Rivers advises the reader that even though the data quality is not robust enough to be statistically significant, this hurdle can be overcome by choosing a model that makes as few assumptions as possible and testing various models to find one that best reflects the network as it evolves through time.

The following chapters move away from a quantitative approach to focus on the qualitative aspects of multiscalar networks. Together, Thomas Tartaron and Barbara Kowalzig demonstrate the utility of theoretical frameworks influenced by network science without explicit engagement with formal analysis models. Tartaron (ch. 4) examines the interplay between local, microregional networks that were part and parcel of quotidian life in the Bronze Age Saronic Gulf and interactions and events at larger, interregional scales. Skeptical of the disparity between the practical limitations of communication in prehistory and the concerns of models drawn from modern social network analysis, he utilizes ethnoarchaeology in Greece and India prior to the end of World War II to shed light on local practices. Kowalzig (ch. 5) similarly evaluates the reciprocal relationship between economic and religious networks within the context of “cultic cabotage” (108). Small-world dynamics, such as the shared religious practices embedded within the cults of Artemis lining the Euboean Gulf, were linked to long-distance trade through weak ties, creating cognitive maps of maritime space that shaped socioeconomic networks.

The chapter by Elizabeth Greene and that of Mark Lawall and Shawn Graham build on the theoretical approaches of Tartaron and Kowalzig by modeling qualitative data. Greene (ch. 6) uses Gephi, a social network graphing program, to model shipwreck sites and the origins of their assemblages as ego networks that represent the links between cargo, galley wares, and the findspot of the ship itself as a series of social and economic connections at various scales in the Archaic-period Mediterranean. The results of her analysis highlight the role of sociocultural connections motivating ties between communities, along with factors such as distance, sailing speed, and profit, and harkens back to one of the overarching concerns outlined in the introductory chapter: the methodological difficulties of situating shipwrecks—the representations of failed voyages—within spatial networks. Lawall and Graham (ch. 7) continue the focus on moving commodities, by using transport amphoras to create simulations of Greek maritime commercial networks in the classical and Hellenistic Aegean with the open-source tool NetLogo, using mimicry and stylistic (linguistic) change models. Rather than relying on amphora stamps and quantitative data, they take into consideration consumer shape selection, potters’ responses, and the potential impact of familiar shapes and prior social connections to determine the network structures that may be responsible for patterns in vessel shapes. Lawall and Graham, however, do not view the simulations as explanatory models, but rather, like Rivers, caution the reader that cross-comparison between the results produced from the models’ simulation runs and the archaeological record should only be used to create hypotheses about causation in antiquity (181).

In what may be the most self-consciously instructive chapter in the volume, Tom Brughmans (ch. 8) turns a critical eye toward the use or misuse of network analysis models through a case study of the changing distribution patterns of Eastern Roman tablewares. His analysis shows that just because these tools are available does not mean that they should be used; sometimes other methods are just as effective for producing the same results, and therefore, the utility of network science is to complement the limitations of traditional approaches. Nevertheless, publication of failed projects is necessary, so that others can use these as guidelines or lessons to avoid similar pitfalls.

The penultimate chapter of the volume, by Paul Arthur, Marco Leo Imperiale, and Giussepe Muci (ch. 9) revisits one of the central questions posed by Lawall and Graham: what factors, such as preference or versatility, are responsible for the adoption and geographical spread of a particular shape of transport container at the exclusion of others? Arthur, Leo Imperiale, and Muci focus on “affiliation networks,” which incorporate both sites and artifacts into an explanatory framework of the dominance of globular amphoras in the eighth century CE and the subsequent return to a more heterogeneous distribution pattern in the succeeding centuries. Visualization of the ties suggests that top-down administrative structures limited shape diversity during a period of political instability, but, once power stabilized, control was relaxed, producing a variety of forms.

In the closing chapter, Barbara Mills (ch. 10) ties together the themes and aims of the volume into a coherent whole. As a specialist in the American Southwest, she reflects on the unique aspects of Mediterranean maritime networks and comparative examples from the New World. Perhaps most importantly, she observes that the key to the future development of network science within Mediterranean research relies on the collaboration of interdisciplinary networks of scholars. Thus they, too, are microcosms of the multiscalar, overlapping connections that they seek to explore in antiquity.

This volume is thoughtfully edited and composed. One minor weakness is that the two final case studies (chs. 8 and 9) would have been better arranged in reverse order. Although they make sense in the current order chronologically, the similarity between ceramic analyses by Arthur, Leo Imperiale, and Muci and by Lawall and Graham suggest that these would be best placed together, while the importance of Brughmans’ message about the utility of failed network modeling would be best highlighted for the reader if it appeared at the end of the volume.

The editors of this volume crafted an impressive exploration of the diversity of data and unique facets that create connections in the Mediterranean basin. The book presents a transparent and conscientious overview of the wide-ranging applications of social network analysis that serves as an indispensable how-to guide for scholars new to the discipline. It also provides an in-depth illustration of interdisciplinary collaborations between specialists in archaeology, classics, ancient history, and the social and physical sciences for the reader more well-versed in network science. Readers can also find an online “manifesto” of the key ideas and insights from the volume’s discussions in “A Manifesto for the Study of Ancient Mediterranean Maritime Networks” (J. Leidwanger et al., Antiquity Project Gallery 88, 2014, 342).

Lana Radloff
Bishop’s University
lradloff@ubishops.ca