Bryn Mawr Classical Review

Bryn Mawr Classical Review 2007.03.21

Anna Marguerite McCann, John Peter Oleson, Deep-Water Shipwrecks off Skerki Bank. The 1997 Survey. Journal of Roman Archaeology Suppl. Series, 58.   Portsmouth, RI:  Journal of Roman Archaeology, 2004.  Pp. 224; ills. 270.  ISBN 1-887829-58-X.  $99.00.  

Reviewed by Justin Leidwanger, University of Pennsylvania (
Word count: 2723 words

[Authors and titles are listed at the end of the review.]

In the late 1980s underwater archaeology entered a new phase in which interdisciplinary teams of archaeologists and engineers extended their sphere of exploration to the deep seas. The development of more refined and readily available remote sensing equipment and Remotely Operated Vehicles (ROVs) now facilitates survey for shipwrecks at all depths in the Mediterranean. At the forefront of this endeavor has been Robert Ballard, who, with a team from the Woods Hole Oceanographic Institute and collaborating institutions, initiated in 1989 an extensive survey at a depth of approximately 750 m in international waters off northwest Sicily near a reef called Skerki Bank. This initial project, a milestone for deep-water archaeology in the Mediterranean, was followed by two additional seasons on the site in 1995 and 1997, and a later visit by Ballard's team in 2003. With Anna Marguerite McCann [M] as archaeological director, the results of the 1989 investigations of a late Roman site dubbed the "Isis Wreck" were published in 1994 in volume 13 of the JRA Supplementary Series.1 Ten years later, the present volume by M and John Oleson [O], along with 16 contributing authors, offers a comprehensive, final publication of the methodologies utilized and finds uncovered during the 1997 survey of eight shipwrecks and two related cultural sites.2 Several introductory chapters and appendices frame M-O's core individual discussions of the ancient shipwrecks and their contexts. As a whole, the book presents a useful, well-executed and comprehensive treatment that is unique among such publications of ancient wrecks for its scope and research design.

Three introductory chapters provide the background for the project, its organization and area, technology, and exploratory and sampling strategies. Discussions of the survey parameters and techniques fill M-O's Chapter 1, and include reasons for choosing the site, previous work in the area, and the place of the present survey within the overall Skerki Bank Deep Sea Archaeological Project. A brief overview outlines all phases of the operation and parameters, from remote sensing and visual inspection with ROVs and a submarine on the seafloor to mapping, test excavation, artifact lifting, documentation, and conservation. In the discussion that follows, M-O raise a number of the significant legal and ethical questions surrounding this and other such projects in international waters.

Chapter 2, by D.A. Mindell and other members of the engineering team, picks up with a discrete summary of the techniques utilized in, and often developed specifically for, the site recording efforts at Skerki Bank. One major new improvement allowed the investigators to augment traditional two-dimensional photographic documentation with more precise contours ("microbathymetry") of the wreck site - an approach that facilitated better resolution at the individual artifact level and greater understanding of the overall site and its seabed context (28-29).

In Chapter, 3, one of the most original and interesting contributions of the book, O and J. Adams make significant additions to an otherwise generally poorly understood notion of site formation processes of wrecks in deep water. Although the amphoras are sometimes scattered, the authors propose that many of the vessels here and elsewhere may have landed upright on the seabed, with cargo stowed, and perhaps only an anchor and a few other unsecured items fell loose. Still, the scattered but intelligible remains from the Skerki Bank wrecks seem to indicate that the unique information deep waters may yield is not so much a product of any perfect preservation (here, wood and other organic remains are no better preserved than in shallow waters) as the lack of continuous casual looting that prevails along much of the dive-able Mediterranean coastline. Concluding the chapter is a final discussion of several applied and proposed excavation techniques for robotically removing sediment, which may foreshadow future avenues of archaeological exploration at these depths.

Chapters 4 through 10, mostly by M-O, discuss the wrecks in chronological order, including single comprehensive chapters devoted to each of the five ancient wrecks (Wrecks D, F, G, B and "Isis" in Chs. 4-8), one on a ship of medieval date (Wreck A in Ch. 9), and a short chapter by J. Adams on two modern ships of the late 19th or early 20th century (Wrecks C and E in Ch. 10).

A closer examination of Wreck D in Chapter 4 provides occasion to look at both the regular structure of these core chapters, as well as some of the advantages and limitations of interpreting this type of evidence. This medium-sized merchant vessel is dated by M-O to c. 80-50 B.C. based on the cargo amphoras. Following brief remarks on the site's location (41), M-O elaborate on the distribution pattern of the artifacts in two distinct heaps (41-43). An issue immediately arises in that, alongside some galley wares and ship's equipment, "the 70 or so amphoras visible at this site do not by themselves constitute a convincing complete cargo," (53) especially for a vessel with an estimated length of c. 28 m. Of course, this 28 m is a tentative best guess, especially since the often disjointed wrecks at Skerki Bank do not seem to present the same tall, oval mound scenario that is more clearly interpretable in the case of a wreck like the Kyrenia ship.3 Without any method of quantifying what may be a considerable component of buried cargo, it is naturally very difficult to judge not only the vessel's size, but how representative these surface finds are of the actual ship's consignment. In the case of Wreck D, the authors advance the reasonable hypothesis that the north end of the site, marked by two lead anchor stocks (with interesting depictions of knuckle-bones), represents the bow (54), since their convenient placement here is well-attested in the iconographic and archaeological record. Yet this seems hard to reconcile with the notion that the northern artifact heap was the galley (noted by the authors and indicated by the presence here of food preparation pottery, a quern, and flat ceramic tiles often used as hearth lining). This part of the ship is usually found at the stern, and in fact is often used by archaeologists to identify this end of the site.4

The careful examination of the ship and cargo segues into discussions of the chronology and "Patterns of Trade," with many wreck comparanda from across the Mediterranean. Given the number and variety of artifacts raised and identified, the suggested date of c. 80-50 B.C. seems reasonable, and in fact is as tight if not tighter than many excavated ships. The diverse origins for the 12 types of amphoras in the ship need not indicate as many disparate stops, but rather a final stop at a larger harbor with trade connections to these various regions; in this case, the authors favor Cosa. Even so, while the 27 Dressel 1B jars are certainly actual cargo, identifying the single globular amphora as ship's stores (71-72), for example, merits caution when only a small proportion of the hold may be visible. Finally, the extensive artifact analyses provide key insights into the various connections represented especially in those cargo components that are raised and individually catalogued. Likewise, the many petrographic observations help bolster some provenances. The authors clearly came to the project with a well-defined sampling strategy (36-37), but important restrictions on which artifacts can be raised often limit the conclusions that can be drawn. For Wreck D, the attribution of six amphoras to the island of Kos or its neighboring region (the farthest flung connection in the wreck) is difficult to substantiate for certain when, as the authors note, imitations were made in the central and western Mediterranean beginning in the mid-first century B.C.

The authors tackle the remaining wrecks with the same approach in Chapters 5 through 10. Wreck F (Ch. 5), of the mid-first century A.D., is another rather mixed assemblage of unprovenanced stone architectural elements alongside various western Mediterranean amphoras, for which the authors consider southern Gaul a likely destination. Wreck G (Ch. 6), a mid-sized mid-first-century A.D. ship with strong western Mediterranean connections, may have set out from Gaul or Spain before calling in North Africa and perhaps also Sicily. The late first-century A.D. Wreck B (Ch. 7) represents a diverse assemblage of far-flung amphoras from Egypt, Crete, Tripolitania and Campania, with galley ceramics indicative of a central Italian connection and perhaps origin. In Chapter 8, M-O describe anchors and other new finds from the previously published late fourth-century "Isis" wreck:5 hull remains that reaffirm the traditional "shell-first" method, and a lead patch that raises the possibility that this ship, en route perhaps from Carthage to Pontus, may have been nearing the end of its use-life. Wreck A (Ch. 9), quite different in nature and dated generally to c. A.D. 1000-1250, may be a small fishing boat operating in ballast with a crew of two perhaps from Spain, although the limited finds preclude a definitive origin. In Chapter 10, Adams rounds out this section of the book with a brief discussion of two modern vessels, Wrecks C and E, both of the late 19th century or later.

The analysis of the cultural material at Skerki Bank proceeds from the treatment of discrete wrecks to some observations, by B.P. Foley and Ballard in Chapter 11, on the distributional patterns at two large sites, named Amphora Alleys I and II. Amid widely spaced individual jars dating from the late fourth or early third century B.C. through the ninth-twelfth centuries A.D., these two long, linear patterns of amphoras were observed. Amphora Alley I was published generally as an ancient trade route in the 1994 volume;6 the discussion here focuses on additional observations made in 1995 and 1997 at the kilometer-long Amphora Alley II. From the tight pattern of the material, the authors argue that the 20 jars from southern Gaul, and perhaps too the handful of contemporaneous North African amphoras, represent a single depositional event: a desperate attempt to jettison cargo during an emergency. The feature is certainly curious and the interpretation seems reasonable, albeit without an actual wreck at the end of this trail of breadcrumbs. Nonetheless, the hypothesis raises intriguing questions about what process might account for the other tight linear clustering of more chronologically diverse finds at Amphora Alley I.

Chapter 12 provides an overview of the conservation procedures, most of which are standard field and lab practice adapted to the project's logistics. One notable new challenge D. Piechota and C. Giangrande describe is the breakdown of the ceramic temper in some artifacts buried in the sediment and exposed to the pressure and cold of the water, a situation that underscores the need to continue developing conservation techniques with such environments in mind. This technical discussion is enhanced by two appendices on the archaeobotanical remains (App. A) and glass analyses (App. B).

The conclusion, in Chapter 13, poses the formidable task of synthesizing the results of these eight shipwrecks and related sites. For M-O, the finds represent, on the one hand, a well-traversed ancient open-water trade route during an intense period of exchange in the first centuries B.C. and A.D., and on the other, significant evidence for large-scale cabotage in the western Mediterranean during Roman times. Though considerably offshore, the dangerous seas around Skerki Bank lie at the intersection of several important routes that accounted for much of the maritime trade in the western Mediterranean, including that most prominent one: between the imperial capital and Carthage. M-O propose that investigations of known trade routes elsewhere in the Mediterranean may hold the key to documenting similar concentrations of ancient wrecks that have thus far eluded archaeologists. Yet the compound cargoes of these generally medium-sized vessels point not only to the importance of cabotage, but, as the authors observe, to a more sophisticated system in which entrepots like Carthage and Puteoli played a key role in the transshipment of goods from both east and west.7

Clearly the story is a complicated one and precludes any universal model of trade even in the case of a Rome-centered world. Although reconstructions of ship dimensions without excavated hulls or even discrete, well-defined wreck mounds are problematic, the authors' estimates of 28 m and 40 m for the lengths of Wrecks D and B illustrate quite clearly that cabotage and mixed cargos were not phenomena solely of small-scale merchants operating on a restricted local level. At the same time, none of the Skerki Bank vessels fit the model of the large, state-driven apparatus of the annona, the type of exchange for which one would almost certainly expect to find evidence along this particular route.

With technology developing at a pace that is often faster than can be circulated through traditional publishing venues such as this, some of the more technical aspects of the discussion will naturally be a bit out-of-date. Still, the engineering team's directions for future research here are apt and many of the issues that arose from the Skerki Bank project are already being addressed. For example, seabed penetration with localized remote sensing (30) promises to reveal additional details of cargo composition at sites like Wreck G, where 12 amphoras on the surface of a vessel perhaps 15 m in length seem inadequate for a complete shipment. The observations by O and Adams of the wrecks' seabed contexts, and the corresponding depositional processes, should inform the research designs of other such surveys. While the authors discuss initiatives of more intrusive robotic investigation, "judicious excavation in the deep sea" (210) still seems some ways off for technological and financial reasons. Practicalities aside, the authors are certainly right to draw (overdue) attention to the ethical complexities of sites in deep and, therefore, international waters beyond the protection of any present or draft UNESCO Convention guidelines.

The present study marks an important new frontier in underwater archaeology, but also provides the first fully comprehensive publication of a multi-site survey in deep water. This is not to say that deep-water sites have not been extensively published; quite the contrary. The work at Skerki Bank had been presented and published at many venues, as has Ballard's subsequent work in the southern Black Sea and off the coast of Israel.8 What sets this book apart is, first, the extensive and diverse artifact collection and the scale of detailed analyses (typological, petrographic, palaeobotanical, etc.). Second, the authors interpret the wrecks and Amphora Alleys not only as discrete individual sites, but also collectively as an overarching diachronic site with implications for Roman imperial-era maritime trade. Of course, current technological constraints in such areas as sampling and subsurface detection limit the reliability and broader applicability of interpretations, especially in such cases as the unidentified stone that comprises the bulk cargo of Wreck F. In this light, the book stands well as a whole, and the core shipwreck chapters are also eminently readable as individual essays for those interested in a particular wreck, period or artifact type. A few typographical errors and bibliographic omissions scarcely detract from an otherwise well-organized and well-presented text, with abundant illustration in the form of site photomosaics and drawings, photographs (many in color) or both, of nearly every artifact raised.

Table of Contents

Preface and Acknowledgements

Note on the catalogued objects

1. Introduction: The genesis and significance of the 1997 survey (A.M. McCann and J.P. Oleson)

2. Precision mapping and imaging of underwater sites at Skerki Bank using robotic vehicles (D.A. Mindell, H. Singh, D.R. Yoerger, L. Whitcomb and J. Howland)

3. Formation, survey, and sampling of the wreck sites (J.P. Oleson and J. Adams)

4. Wreck D (c. 80-50 B.C.) (A.M. McCann and J.P. Oleson)

5. Wreck F (mid-1st c. A.D.) (A.M. McCann and J.P. Oleson)

6. Wreck G (c. A.D. 50) (A.M. McCann and J.P. Oleson)

7. Wreck B (last quarter of the 1st c. A.D.) (A.M. McCann and J.P. Oleson)

8. The "Isis" wreck (last quarter of the 4th c. A.D.) (A.M. McCann and J.P. Oleson)

9. Wreck A (c. A.D. 1000-1250 or later) (A.M. McCann and J.P. Oleson)

10. Wrecks C and E (19th c. A.D. or later) (J. Adams)

11. Amphora Alleys I and II (B.P. Foley and R.D. Ballard)

12. Conservation of archaeological finds (D. Piechota and C. Giangrande)

13. Conclusions (A.M. McCann and J.P. Oleson)

Appendix A: Archaeobotanical remains (C. Ward)

Appendix B: Laboratory analysis of two glass objects and a glaze sample from Wreck A (R.H. Brill, P.M. Fenn and H. Shirahata)



1.   A.M. McCann and J. Freed, Deep Water Archaeology: A Late Roman Ship from Carthage and an Ancient Trade Route near Skerki Bank off Northwest Sicily (Ann Arbor 1994).
2.   It is fortunate that the authors, although not involved in the subsequent endeavor, were able to include some preliminary remarks on a return visit to the site in 2003, when additional information was gathered and a few artifacts were raised.
3.   M.L. Katzev, "Reconstructing the Oldest Known Greek Ship," National Geographic 137.6 (1970) 841-57.
4.   F.H. van Doorninck, Jr., "The Galley," in G.F. Bass and F.H. van Doorninck, Jr., Yassi Ada, Volume I: A Seventh-Century Byzantine Shipwreck (College Station 1982) 87-120 (especially 110-20, "The Hearth").
5.   See McCann and Freed 1994 in note 1 above.
6.   See discussion of Amphora Alley I in McCann and Freed 1994, 61-89.
7.   Including, most notably, P. Horden and N. Purcell, The Corrupting Sea: A Study of Mediterranean History (Malden 2000) 137-43.
8.   R.D. Ballard, F.T. Hiebert et al., "Deepwater Archaeology of the Black Sea: The 2000 Season at Sinope, Turkey," AJA 105.4 (2001) 607-23; R.D. Ballard, L.E. Stager et al., "Iron Age Shipwrecks in Deep Water off Ashkelon, Israel," AJA 106.2 (2002) 151-68.

Read Latest
Index for 2007
Change Greek Display
Books Available for Review

HTML generated at 13:31:42, Friday, 03 April 2009