Seminar CCXXVIII: a new method for analysing Mediterranean connectivity

The seminar report backlog now takes us back to Birmingham, where on the 5th February 2015 Dr Matthew Harpster was addressing the General Seminar of the Centre for Byzantine, Ottoman and Modern Greek Studies. Matthew is one of the friends I hope to keep from Birmingham; we had friends in common when I arrived and then someone gave me the office opposite him, so I had quite a lot of contact with him, but still I didn’t actually see him talk about his stuff until I was working at the Barber Institute of Fine Arts. This then was that occasion, and it sparked off quite a lot of subsequent thought and action. His title was “Refashioning a Maritime Past in the Eastern Mediterranean”.

Nautical archæology under way at the Bozburun shipwreck site off the Turkish coast in 1996

Nautical archæology under way at the Bozburun shipwreck site off the Turkish coast in 1996; it probably isn’t Matthew in that wetsuit, but it could be!

It had been Matthew’s doing that this same seminar had earlier been addressed by Rebecca Ingram on the subject of shipwrecks, because Matthew too is a maritime archæologist who once worked at the Theodosian Harbour in Istanbul with her, and like her he also had a particular shipwreck with which he was concerned, a ninth-century one off Turkey.1 This seems likely to have been a Byzantine one but as Matthew had poked at this he had become less and less sure that we have any solid methodology for making such judgements: does one go from the cargo, the personal effects of the crew, the location, the building style, or some or all of these? All of these things could easily be out of place as we understand them. Matthew had gone so far as to assemble all the 254 attempts to identify shipwrecks in the International Journal of Nautical Archaeology from 1972 to 2012 as he counted them, and found no consistent practice.2 At that point his project became the one that then brought him to Birmingham, to database as many shipwrecks as possible from the ancient world and try and pattern-spot in such a way as might underpin such a consistent methodology for identification.

Cover of Parker's Ancient Shipwrecks of the Mediterranean

Cover of Matthew’s foundational text, Parker’s Ancient Shipwrecks of the Mediterranean

There is at least an easy place to start, an inventory of 730 ancient shipwrecks assembled in the early 1990s to which Matthew was able to add 120 more; the standard of record is of course variable but it’s a start.3 From there Matthew had used the cargo, fittings and personal items recorded for each wreck to work out route profiles for each vessel, assigning each of its items a point of origin and using those points as plots for a polygon that represented that shipwreck’s notional catchment area. Of course, this relies on others’ identifications of the goods and archæology being able to assign them correctly to places of origin, and as Morn Capper (present) pointed out, it is also tracking the finds, not the ships, and if those finds had moved in several ships in turn, not one all the way, the polygon of the one that actually sank would be considerably larger than that ship’s own sea area.

Map of ancient shipwrecks from the Benthos project

This is, sadly, not Matthew’s work but someone else’s attempt to do something similar, mapping the ancient Mediterranean’s nautical archæology, but only where it now rests, not where it had come from… The project concerned is called Benthos, and looks interesting but doesn’t seem to have progressed beyond its preliminary phase as of three years ago, alas.

But, this is something that one can do, it’s still telling one about a species of connection and the results are impressive, the more so because of the work Matthew had done with Dr Henry Chapman to load them into GIS and process them. I wish I could show you one of the resulting visualisations, as they are not only fascinating but things of psychedelic beauty, but Matthew seems not to have put any of them on the web where I can steal them, goshdarnit. In any case, in so far as what they show can be summarised, that summary might be:

  1. The weight of maritime activity shifted eastwards in the Mediterranean between the fourth century B. C. and the fourth century A. D., with more and more material travelling in the eastern half of the Mediterranean and less in the western one.
  2. Throughout that period, however, there is a visible separation of the two halves, either side of a zone including Sicily, Malta and the northern tip of Africa, which seems to have been a zone busy with transshipment but across the whole of which relatively little passed without stopping.
  3. In the fifth century A. D. this trend changes, with the Eastern Mediterranean dropping off in importance and goods from the West beginning to travel much further. Pirenne would have been worried!
  4. Pirenne would, however, probably have taken refuge in the fact that there is much less data from the late period, and in fact almost nothing for the eighth to tenth centuries, but the real peak is in the first centuries B. C. and A. D., not as one might have expected the height of the Roman Empire, and any conclusions for what was going on outside that period are based on dangerously small samples. Was sailing just safer under Hadrian or something? In any case, moving on… 

Matthew’s main point was that, within the limits of the evidence, his method could be used to measure and display change over time in the much-vaunted connectivity of the Mediterranean, but in discussion, predictably, the gathering set to trying to work out what else it told us or might do if extended.4 Archie Dunn wondered how journeys recorded in texts would map using such a method, Rebecca Darley offered military campaigns, as well as coins of course, and I wondered about inscriptions and diplomatic formulae. It seemed to me, and I said out loud, that all these things might well map out differently and result in an even more complex and textured picture of how people moved around the Mediterranean. And at that point Professor Leslie Brubaker said, “Funding bid!” and well, somehow from this seminar came a research proposal involving seven people, including myself, Rebecca, Matthew, Henry and Leslie, and it’s currently under review after making it to the second round of the European Research Council’s Advanced Grant competition, so I guess we shall see what a great fire a little matter may yet kindle; I’m still quite excited about the prospects it raises. But whatever comes of it, Matthew started it, by giving this excellent paper to an audience who thought of useful questions, and that is really how all this is supposed to work, isn’t it?

Divers over an ancient shipwreck off the coast of Sicily

Who knows what we may find? Though I at least won’t have to get wet for my portion of the material if it all comes together… This is a recent excavation off the coast of Sicily of a ‘2,000-year-old ship’ about which I can tell you no more, but it’s a good image to close with!

1. And indeed he has published on it: see M. Harpster, “Designing the 9th-Century-AD Vessel from Bozburun, Turkey” in International Journal of Nautical Archaeology Vol. 38 (Oxford 2009), pp. 297-313, DOI: 10.1111/j.1095-9270.2009.00226.x.

2. See Harpster, “Shipwreck Identity, Methodology, and Nautical Archaeology” in Journal of Archaeological Method and Theory Vol. 20 (Heidelberg 2013), pp. 588-622, DOI: 10.1007/s10816-012-9131-x.

3. A. J. Parker, Ancient Shipwrecks of the Mediterranean and Roman Provinces, British Archaeological Reports (International Series) 580 (Oxford 1992).

4. The vaunting is primarily to be found in Peregrine Horden & Nicholas Purcell, The Corrupting Sea: a study of Mediterranean history (Oxford 2000), on which see Paolo Squatriti, “Mohammed, the Early Medieval Mediterranean, and Charlemagne” in Early Medieval Europe Vol. 11 (Oxford 2003), pp. 263-279, DOI: 10.1046/j.0963-9462.2002.00111.x.

4 responses to “Seminar CCXXVIII: a new method for analysing Mediterranean connectivity

  1. I’d love to see the ERC Advanced Grant come together. Good luck!

  2. It has been a while since I read the first chapter of Jonathon Pryor’s excellent book on Medieval Mediterranean shipping, but my memory is that prevailing winds, currents and a relative lack of harbours along the North African coast mean that, if (say) you are leaving Carthage bound for the West, what you don’t do is sail West – you sail North and go up the coast of Italy and then turn West. Which is to say that the relative lack of shipwrecks in the Western Med doesn’t necessarily mean that western destinations weren’t being sailed to, just that in the age of sail people had to go the long way around to get there. This may (partly) explain what looks like the bottleneck going through the Strait of Messina.

    I wonder if it is also fair to say that the waters of Morocco, Algeria and Tunisia have been relatively less explored in terms of underwater archaeology than those of other countries? That may be entirely wrong – more a question than a statement.

    • The latter point seems very fair, though I don’t know the answer to what is still a good question. The former I have to think harder about. The Straits of Messina are notoriously dangerous for shipping, of course, and so one would expect goods from all over to crop up there for that reason and the one you suggest, but then those goods would not reach the other end of the Mediterranean. But if people were actually sending such goods, some of them must have made it past the Straits, and should thus be available for sinking closer to destination too. Yes, the sample is small, but then it’s really too small for Scylla and Charybdis to be a steady factor too…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s