It was pretty exciting to get my hands of Andrew Bevan and James Connolly’s publication of their work on the small island of Antikythera. In many ways, this book and the soon-to-be-published Michael Given et al. volume on their work in the Troodos mountains of Cyprus represent the state of the art in intensive pedestrian survey in the Eastern Mediterranean. This is important to me, of course, because David Pettegrew and I are doing final edits on the publication of the final results from our work on Pyla-Koutsopetria.
The island of Antikythera is located off the southeast coast of the island of Kythera and northwest of the island of Crete. It is small, has a tiny full-time population, and has played only a marginal role in the “great events” documented by ancient and later texts. In fact, it seems likely that the island was periodically abandoned and visited only intermittently by herds, hunters, and shipwreck victims. The famous Antikythera shipwreck demonstrates that the island sits astride major east-west travel routes linking the eastern basins of the Mediterranean sea with those to the west. (It is worth noting that the Google Streetview car has never been to Antikythera! UPDATE: This is because the Streetview car is banned in Greece…)
Rather than write a comprehensive review, which I know my more qualified colleagues will produce in good time, I’ll highlight a few things that I thought remarkable about this relatively short volume (by archaeological standards).
1. Total Coverage without Counting.
When Elizabeth Fentress asked the question “What are we counting for?” over a decade ago in fifth volume of the POPULUS project, Extracting meaning from ploughsoil assemblages, many survey archaeologists struggled to find an adequate response. Bevan and Connoly’s work is the first to almost entirely do away an approach that privileged areas with significant aggregate artifact densities. Instead, almost all pertinent analysis came from the distribution of diagnostic artifacts from specific periods with no real attention to overall patterning of artifacts. This felt like a major break though in how we think about survey.
More remarkably, of course, the Antikythera survey documented surface assemblages at close to a 10 m resolution. The field walkers are spaced at 15 m intervals and documented artifacts on the surface every 10 m. This sets a new standard for resolution in regional level survey and reflects the growing interest in intensification in field collection. As survey areas have decreased – for permit reasons as well as methodological concerns – the intensity of data collection has increased allowing archaeologists to say more using more robust assemblages over smaller areas.
2. Variable Diagnosticity
As assemblages have become more robust and survey resolution has increased, archaeologists have become better able to understand how particular periods appear in the landscape. While Bintliff’s “hidden landscapes” still persist, we can not articulate more clearly why certain landscapes are hidden. Bevan and Connolly deal with what is being called variable diagnosticity in as thorough a way as anyone this side of David Pettegrew (.pdf). This basically means that some artifacts from some periods are more visible than others. The result of this variation is that some periods are more visible than others and we have to compensate for this variation through time. The folks on the Antikythera survey do this by both allowing for ambiguity in reading of pottery by identifying some artifacts as possibly dating to a range of periods. They are also particular attentive to classes of artifacts present and absent in surface assemblages allowing them to understand which periods are more or less.
I note this aspect of the book, in particular, because much of the analysis of our survey assemblage at Pyla-Koutsopetria focuses on a similar issues and demonstrates – we hope! – that as survey assemblages become more robust, we have to be just as critical about the gaps in these assemblages.
3. Numbers and Maps.
The Antikythera project surveyed the entire island. The entire island!
And this produced, as one might expect from Andrew Bevan, absolutely stunning maps and figures. Not only did they make the relatively low density distribution of pottery clear, but he also complemented them with some well-executed charts that demonstrated how the various period assemblages coincided with certain environment conditions ranging from distance to water, to slope, soil quality, and proximity to the coast. The authors managed to keep their impressive statistical analysis transparent, but not overwhelming. In other words, I understood arguments based on statistics that I am sure I entirely grasp.
There are some other remarkable things about this survey volume. Some perhaps show the way for future intensive survey publications, whereas others are products of particular disciplinary predilections.
1. No Catalogue.
The volume included no formal artifact catalogue. To be fair, some of the finer discussions of artifacts appeared in separate publications, but it is still remarkable that there is no formal catalogue of finds to support the author’s argument.
On the other hand, as we torturously edit the catalogue in the survey volume from Pyla-Koutsopetria, we can appreciate the limited utility of a survey catalogue. Even with increasing openness toward coherent surface assemblages and horizontal stratigraphy, survey catalogues are fairly strange beasts. They contribute little to re-examining chronology or typology. Catalogues do, however, provide evidence for arguments made in the text and while the color pictures and plate are not quite sufficient to evaluate the type-fossil artifacts upon which their arguments rest.
2. Little comparison
I was also a bit bothered that there wasn’t a greater effort to compare the assemblage on Antikythera with the work done elsewhere in southern Greece, Crete, and the islands. I understand that Antikythera is a small island and, in many ways, it represented a unique case. On the other hand, it would have been revealing to understand how similar the assemblages on Antikythera were to those on Kythera itself or the Peloponnesus. This might shed some light on the how connected these communities were culturally as well as economically to the flow of objects across the Mediterranean. As survey projects become more intensive and assemblages approached more critically, it opens opportunities to compare between projects.
3. No Experiences and No Methods
Finally, this publication is – in some ways – pleasantly old-school. There is no excessive dalliance on methods or even much discussion of methodology. In fact, I’m rather curious how they collected ceramics data and counts every 10 meters without slowing the progress of their survey to a crawl! There are few arguments justify the efficacy of their approach to the landscape and, aside for the occasional remarks, very little apologia for the use of survey data to produce arguments. This is not to say that methodology or critical approaches to data are not important (see my discussion of variable diagnosticity above), but to suggest that this is a tremendously confident volume.
More disappointing, however, is the almost complete absence of any discussion of the experience of being on Antikythera. The only place where the experience of being on an small, windswept island comes clear is in the 19th century and earlier accounts documented in chapter 8. No where is the archaeologists’ experience clearly documented. I have no real idea how long it would take to walk from one part of the island to the next or how modern communication technologies enhanced or emphasized earlier senses of isolation. I recognize that not all archaeologists are comfortable with (or accept) the value of phenomenological approaches to the landscape, but since the idea of a persistent landscape was sufficiently important to appear in the book’s title, it would have been nice to get a better sense for how the persistent landscape shaped archaeological work. More importantly, with the exception of the most fleeting appearances, there is little discussion of the fragile and short term community of archaeologists working on the island.
The brutal transects that the modern archaeologist marched across the island seem to contrast with the less regular arrangement of settlements and activity areas. In other words, the archaeologists seemed to exist in a different landscape from earlier settlement. Unfortunately, I am not clear exactly his this landscape appeared.
These quibbles aside, this book is an important contribution to how we go about analyzing survey data and should fit right along side the classics in the field. It will be particularly exciting to read this volume alongside the long-awaited final publication of the Kythera Island Project.