Deep-Water Shipwrecks Off Skerki Bank: The 1997 Survey

Skerki Bank designates a series of shallow reefs located approximately 60 km north of Tunisia and 80 km west of Sicily. Coauthors McCann and Oleson and 16 contributors have searched the vicinity of Skerki Bank for ancient shipwrecks since 1989. This volume presents the results of the 1997 season in which the investigators studied eight wrecks: five Roman merchant ships, one Early Medieval fishing vessel, and two late 19th- to early 20th-century sailing ships. It goes well beyond the previous volume on research at Skerki Bank (JRA Suppl. 13), which considered one late fourth-century C.E. vessel examined in the 1989 season.

The volume’s introduction contextualizes the project’s major breakthrough: scientific documentation and excavation of artifacts on the deep-sea floor. Next, the authors describe the innovative methods by which they located, mapped, and excavated the eight shipwrecks at depths of 750 to 800 m. Most of the volume presents the shipwrecks individually. While the presentation of the two 19th- to early 20th-century shipwrecks is limited, the other six each receive a separate chapter in which the authors discuss the wreck’s location, artifact distribution, formation process, cargo and gear, chronology, and trading route. An illustrated catalogue of the ceramic, wood, metal, and glass artifacts recovered or observed completes each chapter. Two “amphora alleys” about 1 km long and not associated with these shipwrecks are also discussed. The volume ends with chapters on conservation, botanical remains, glass, and overall conclusions. Throughout the volume, extensive color and black-and-white photographs as well as line drawings allow readers superb views of the research and finds.

The Skerki Bank project will undoubtedly be long remembered for initiating deep-sea excavation using a nuclear submarine and a remotely operated vehicle (ROV). As one would expect, the description of new techniques such as “microbathymetric mapping” and “photomosaicing” is exciting. To produce the former, the ROV crisscrossed the wreck at slow speed measuring the artifacts and sea floor with sonar and transponders, creating a three-dimensional map accurate to within centimeters. For the latter, it photographed the site many times; these images were then digitally combined. The difficulty of taking measurements and photographs at 800 m below sea level notwithstanding, the resulting pictures of the wreck sites are extraordinarily clear, revealing how well preserved each Skerki Bank wreck is. The high standards of recording are also impressive, and the maps are nearly equivalent to those of a well-published terrestrial excavation. The ROV also excavated with a specially designed plastic trowel and used cushioned hands to expose and collect artifacts. Its arm operated deftly enough to preserve fragile artifacts and organic materials, although the procedures were very slow and need refinement.

What evidence do the wrecks provide? Oleson suggests that many of the ships sank intact, landed in an upright position on the sea floor, and did not suffer damage from trawling vessels. Thus they preserved nonorganic materials well, making it possible to identify and date their contents. At Wreck D (80–50 B.C.E.) the authors noted 70 amphoras of 12 types, including 27 Cosan, six Koan, and five Tripolitanian. Black-glaze pottery from Campania was also found. Wreck F (mid first century C.E.) contained blocks and columns of either marble or granite, North African, Spanish, and Sicilian amphoras, as well as North African and Italian cookwares. Gaps between concentrations of artifacts at both wrecks led the authors to speculate that they also contained organic items, such as wheat, which have now disappeared. Wreck G (ca. 50 C.E.) revealed Spanish, Tunisian, and Sicilian amphoras, while Wreck B (last quarter of the first century C.E.) contained amphoras from Campania, Egypt, Crete, and Tripolitania. The “Isis” Wreck (last quarter of the fourth century C.E.) yielded amphoras from Tunisia, Calabria, and western Asia Minor, as well as grains of wheat and barley. The most interesting feature of these five Roman-period cargoes may be their mixed nature.

The Skerki Bank ships provide some of the best evidence for cabotage, the relatively unstructured and small-scale movement of goods that scholars have argued had greater importance to Mediterranean commerce than large-scale ventures organized by major centers. While the authors do make this point in their conclusions, they do not go as far as they might to integrate the Skerki Bank wrecks into discussions of the ancient economy.

Another important issue involves trading routes. The first Skerki Bank volume suggested that ships sailing between Carthage and Rome passed through this area, but the evidence from the additional wrecks indicates that ships traveling the western Mediterranean on other routes (e.g., from Spain, Sicily, or Campania) frequented it, too. And the discovery of these wrecks in deep water, along with others currently being investigated off Greece, Israel, and elsewhere, may alter the established view that ancient sailors commonly avoided deep waters.

Readers should be warned that the authors’ identification of amphora contents is not always convincing. Wreck F contained at least three amphoras of a type so rare that it had been found only twice before, including at Pompeii, where a dipinto labeled its contents as lomentum. Yet the authors assert that Wreck F’s containers held lomentum, although this is far from certain. Not only have residue analyses shown that specific amphora types may not always have had the same contents, but one of the amphoras from Wreck F was lined with pitch while two others were not, suggesting a difference in contents. Further, ancient sources refer to lomentum as both a blue pigment made from azurite and a bean meal used in cosmetics, medications, and bread, making the identification problematic for a different reason. The contents of other amphora types found (e.g., Keay 35A, van der Werff 1, and van der Werff 3) are also open to debate.

With its pioneering research, the Skerki Bank project has also raised new ethical questions. Do modern countries (Tunisia and Italy in the case of Skerki Bank) have the right to superintend archaeological remains in nearby waters? The UNESCO Convention on the Protection of Underwater Cultural Heritage, adopted in 2001, indicates that deep-sea investigators must collaborate more closely with local authorities in the future. Disappointingly, only 15 countries have ratified this agreement to date.

Overall, this is an excellent volume, and Roman and maritime archaeologists should not overlook the valuable material it contains. But this research is also important for all archaeologists as an excellent case study of high-tech applications for the recovery of the past, and a demonstration of the relevance of discoveries in classical archaeology for the rest of the archaeological community.

David L. Stone
Department of Classics
Florida State University
Tallahassee, Florida 32306-1510
dstone@mailer.fsu.edu