The Archaeology of the Mediterranean World

Dipping my Toe in the Public History Pool

Public history is all the rage these days in history programs.  To my eye, the interest in public history is an effort within the discipline of history to connect our work with a recognized group of jobs as archivists, public historians, and at museums.  On the one hand, this is good in that it recognizes that many of our students want to stay in the field of history, but do not aspire to academic positions.  On the other hand, this is part of a larger vocational trend in the humanities that we probably need to monitor.  While there is no reason not to embrace programs which hold forth the prospect of employment for our students, history as a discipline would be weaker if preparing future public historians became the dominant goal of history departments. 

In any event, since our department is looking to get more serious about its offerings in public history, I thought I'd make an effort to see how my work in Greece and Cyprus could contribute.  So in 2010, I am going to offer a small internship program in public history.  Below is a very, very rough fact-sheet on this graduate level internship.  As you can see, many of the details are to-be-determined, but I think it captures the core of what I'd like my internship to do. 

Public History Internship
SP2010

The goal of this internship program is to provide an opportunity to gain experience with a wide range of digital and new media applications that are becoming increasingly central to public history, museum management and outreach, archival work and archaeological curation.

This work will focus on a number of ongoing and completed archaeological and historical projects and tools.

1. Topos/Chora: The Photographs of Ryan Stander. This is a series of photographs and related essays from the Pyla-Koutsopetria Archaeological Project. The photographs will appear at the Empire Theater for the month of January and then in an online gallery. They will be accompanied by a series of reflective essays written by the archaeologists who participated in the Pyla-Koutsopetria Archaeological Project. The goal here is to have something up on the web by January 15th and the gallery available for viewing by January 30th.

2. Pyla-Koutsopetria Archaeological Project: A Digital Museum. Since 2003, a team under the direction of Scott Moore, David Pettegrew, and me has been working at a site called Pyla-Koutsopetria on the south coast of Cyprus. This project has produced a vast amount of digital data ranging from video to podcast, photographs, text, descriptive data, maps, plans, illustration, quantitative data, et c. The goal of the Digital Museum is to present some subset of this data in a coherent way for the educated public. We have Omeka, online museum software, installed on a university server. This software can provide the base for our online museum.

3. Pyla-Koutsopetria Archaeological Project: Data Curation. One of the most important aspects of any archaeological or museum work is the responsible curation of all forms of data. PKAP has recorded a substantial amount of both digital and paper data over the past 7 years. This data needs to be curated. The paper data must be prepare to be deposited in the university archives and parts of the digital data, to be uploaded to Open Context for digital publication. In many ways this curation project is the flipside of the project 2.

4. Lakka Skoutara: An Early Modern Site in the Eastern Corinthia. Since 2000, David Pettegrew and I have recorded descriptive and photographic data from the early modern site of Lakka Skoutara that documented the changes at this site as a result of a whole range of abandonment practices. These photographs need to be put together with textual descriptions in a way that is useful to scholars. This archive will become the online companion piece to a published article.

5. Ohio Boeotia Project at Thisvi, Boeotia. Over the past two years, I have slowly been digitizing the results of an intensive pedestrian survey project conducted between 1979-1982 around the village of Thisvi in southeastern Boeotia. It would be excellent to report the results of this project in a transparent way or to develop an online environment where this work can be highlighted and made accessible.

Goals and Priorities: To some extent, I will let you imagine a set of priorities for this list of tasks, and I certainly don’t imagine that you’ll get all these done. On the other hand, I expect that early on, we as a team develop some sense of priorities in how we plan to attack these various projects. It is important to emphasize that public history projects are almost always collaborative. That is to say that people work together to accomplish a particular task. We are going to work together as a team to accomplish the goals listed below.

Assignments and Responsibilities: Since this is an internship, I will not have a major writing or reading assignment. You should plan to dedicate 10 hours a week to working this internship. I will insist on weekly 1 hour meetings. These will include status updates and are not optional. In addition, you will be expected to maintain a public blog detailing weekly how the various projects are progressing. The goal of the blog will be both to keep you honest (as a team) and to make the progress of the various projects underway transparent to the various stakeholders both at the university and elsewhere. I will expect each participant in the internship to contribute a single blog post a week. It might be best to blog on a Monday-Wednesday-Friday schedule. The first assignment will be to select a blogging service.

Resources:

1. Digital and New Media Laboratory. At present we have a single PC, a Linux powered laptop, and a gaggles of very powerful Macintosh computers. Time in the laboratory should be negotiated with the various other users.

2. Published and unpublished reports from Pyla-Koutsopetria Archaeological Project, Ohio Boeotia Project, and Lakka Skoutara (which was a part of the Eastern Korinthia Archaeological Survey (EKAS)).

3. People. Consider me, my various colleagues, and folks on campus potential resources. When in doubt, ask questions. Part of a successful public history project is knowing how to get the information that you need.

Some websites:

http://omeka.org/

http://www.pkap.org/
http://mediterraneanworld.typepad.com/pylakoutsopetria_graduate/
http://mediterraneanworld.typepad.com/pylakoutsopetria_season_s/
http://mediterraneanworld.typepad.com/pylakoutsopetria_undergra/

http://mediterraneanworld.typepad.com/
http://mediterraneanworld.typepad.com/the_archaeology_of_the_me/2009/10/between-sea-and-mountain-the-archaeology-of-a-20th-century-small-world-in-the-upland-basin-of-the-southeastern-korinthia.html
http://mediterraneanworld.typepad.com/the_archaeology_of_the_me/thisvikastorion_archaeological_project/

New Views on Old Data: First Draft

Find below my first effort at an AIA paper that I will be co-writing with Tim Gregory.  It's rough around the edges, but I think on the right track.

For more on this research:
Reclaiming Thisve Data
Thisve Basin, Archaeological Visualization, and Curating Digital Data
First Out: A First Draft of An Intro for New Views on Old Data
Survey Archaeology Finds as Data
More on Thisvi in Boeotia
Fine ware and Function at Boeotia Thisvi

New Views on Old Data: Reinterpreting Intensive Survey Results After 30 Years

William R. Caraher, University of North Dakota
[first draft]

It seems natural to include a paper on survey archaeology on a panel entitled “First Out”. After all, the surface assemblage is, by necessity, the first out for any excavation. At the same time, the study of surface assemblages has fit into the definition of “First out” intended by the organizers of this panel by contributing significantly to our understanding of post-Classical periods in Greece over the past four decades. In fact, the ground breaking work of many of the participants on this panel has made clear that the rigorous documentation and analysis of surface finds has expanded our notion of what constitutes an archaeological site to well beyond the built up centers of ancient polis and across every century from the end of antiquity to the modern era. Intensive surveys in Boeotia, Laconia, Messenia, and the Corinthia are rewriting both the ancient and post-Classical landscapes of these well-studied regions.

If I can continue to play with the idea of “first out”, it is also clear that this phrase could apply to the first generation of intensive, pedestrian “siteless” surveys in another way. Like the first phase of excavation at major sites across the Mediterranean, the first efforts at intensive survey often relied upon assumptions and methods that were unrefined or unsophisticated in comparison with more recent work. In fact, the constant refinement of survey techniques and the ever more robust datasets that they produce often include explicit and implicit critiques of earlier survey methods. This continuous critique has not only weakened the status of survey among a sometimes skeptical archaeological establishment, but also served as a tacit justification for neglecting the results of earlier surveys. Technological barriers, irregular recording practices, and the incomplete publication of data sets have further impaired archaeologists’ ability to redeploy data collected from the first wave of surveys for newly formed hypothesis.

While the methodological concerns associated with revisiting early “second wave” survey data prose problems, this data nevertheless preserves evidence for the ephemeral surface record in Greece. Both ever-expanding development of the Greek countryside and the irregular patterns of surface visibility, agricultural practices, and erosion patterns obscure and threaten the surface record. As Albert Ammerman famously observed based on the results of several seasons of the systematic resurvey in Italy, sites tend to blink on and off in the landscape like traffic lights. What a project documents one season may not be there the next. Consequently, intensive survey data often captures a single distinct and unique view of the landscape which is not susceptible to reproduction even using similar methods.

This paper will use the data collected from the Ohio Boeotia Expedition between 1979 and 1982 from the (modern village and) Boeotian polis of Thisvi. The results of this survey were published in a series of short articles between 1980 and 1992. While these articles provided for a broad discussion of method and a basic report on the project’s finding, they did not publish finds or quantitative data extensively. Our goal with this paper is to take the first step in re-introducing data from the OBE into the broader conversation about settlement and survey data in both in Boeotia and across Greece more broadly. To do this, we would like to first discussion briefly the process of curating the survey data produced by the OBE and then go on to analyze this data in the context of some recently published survey work from Greece.

The first step in preparing the OBE data for analysis was the keying of records preserved in a series of notebooks and binder pages. These records included counts of artifacts from survey units, which were generally 1 meter square total collection circles as well as from more robust collection procedures conducted at a number of sites across the survey area. We also keyed the finds data from both survey units and the sites into an access database. The finds tables were in turn normalized. Here it is interesting and perhaps valuable to recognize that the quality of data recorded over the course of original fieldwork was quite high, but it was hardly normalized and consequently unsuitable for systematic, quantitative analysis. The lack of normalization was perhaps, in part, the result of the novel character of desktop-level tools for quantitative analysis (e.g. SPSS-X and IBM’s iconic SQL-powered DB2 debuted the year after OBE competed its fieldwork – 1983; the Macintosh personal computer was introduced in 1984.). This is not to suggest that quantitative analysis of archaeological data did not occur prior to the early 1980s, but rather to point out that the creation of normalized practices of data-recording and well-defined hierarchies of object identification became a higher priority after desktop database and statistical tools became more common. By normalizing the robust data sets produced by intensive survey, the database became as important as the traditional artifact catalog for analyzing the chronology and function of sites across the landscape.

At the same time as we keyed data from notebooks and binder pages, we also sought to remap the location of the transects using GIS software. At some point in the 1980s [some additional historical clarity here would be helpful], the artifact counts and location of transects was entered into the Surface II software program which produced a contour map of the artifact densities across the Thivi basin. While versions of these maps were published, the data behind these maps appears to be lost. Unfortunately, at present the disappearance of this spatial data has made it difficult at this point to place the western-most transects on the ground. The written description of the locations of the western transects relies upon points of reference that are not visible on the Greek Army Mapping Service 1:5000 maps and have been destroyed on the ground as a result of the construction of a massive pipemaking factory. There is hope that we can find the location of these transects from older aerial photographs of the area.

The final step in the production of this data is recording comprehensive metadata for the all of the data that we entered. Once the keying of the data and metadata is complete we plan to make this data available to the public via the internet. This step is especially important for small projects because it distributes of digital data expands the curation process from the purview of the creator of the data to the community of users who want to make use of the data. Disseminating the data to end users, with the proper metadata, we make it possible for others to use our material and make it far more likely to be kept compatible with changes in technology.

_____

There have been significant changes in our understanding of post-Classical countryside since the Ohio Boeotia Expedition published their results in the 1980s. The work of both excavations and survey in Boeotia and elsewhere in Greece alone has produced a foundation for the reinterpretation of our survey data. Recent work by Archie Dunn and a team from the University of Birmingham has begun to document the post-Classical finds at Thisvi itself and Jonita Vroom’s study of the post-Classical ceramics from the Cambridge-Bradford Boeotia Project has shed important light on the relationship between post-ancient ceramics and settlement patterns across Boeotia.

The OBE team produced the current dataset through a number of different methods. The diversity of methods reflected the early stage in the development of field procedures and an avowedly experimental approach to documenting the landscape. The area closest to the city walls, Area A, were surveyed using a series of 11, randomly placed, 30 m radius circular survey areas from which samples were taken. The team surveyed the plain itself using a series of long transects (Areas, C, D, and E) from which they typically took 1 sq meter samples for density and diagnostic artifacts. Finally, the teams also collected samples for area of particularly high density which they designated sites. They surveyed these areas using a flexible methods best suited for documenting the extent, chronology, and function of the material on the ground. In addition to these survey areas, the OBE team also conducted intensive survey on two nearby islands in the Gulf of Corinth, Kouveli and Macronisos, which I have not included in the aggregated totals produced in the analysis below.

The survey of the mainland counted over 8700 artifacts and documented over 1700 batches of unique artifacts from the four areas of the survey.

The artifact density data from the OBE shows a decline in the number of artifacts from the units closest to the city across the central part of the Thisvi basin. This pattern, noted in the original publication of the survey, may be at least in part a product of the erosion patterns. In antiquity, an ancient barrage, described by Pausanias, controlled the flow of water and sediment into the basin. In more recent times, the lack of ability to control the flow of water may have either covered some of the sites or, at very least, discouraged habitation there. The density of artifacts, however, clearly increases on the gently sloping, stony ground the along the south side of the basin.

Against the backdrop of overall artifact density we can show the distribution of post-Classical material across the survey area. In general, the survey area is dominated by Classical to Hellenistic and Roman periods which accounted for 2/3 of the datable ceramics. There were, however, several concentrations of both Late Roman and Byzantine to Medieval pottery which represented about 10% of the overall assemblage of datable material collected from survey. Modern material and a thin scatter of pre-Classical material accounted for the other 20% or so of material from the survey.

Area A encompassed the highest density areas immediately south of the plateau upon which the ancient city and the modern village stand. The post-ancient material from this area were very focused with most of the material deriving from three units. Unit A2 contained an abundance of post-Classical material including Middle Byzantine material. It is situated immediately to the west of one of the Hellenistic fortification’s towers which appears to have undergone some modification in the post-Classical period. Units A5 and A8 produced significant quantities of Late Roman – Early Byzantine coarse wares including the ubiquitous combed ware.

The transects immediately to the south of the urban center of Thisvi, Area D, show diminishing quantities of post-Classical material with distance from the Hellenistic walls and the presumed center of post-Classical habitation. Overall only 12% of the material there was post-Classical as compared 77% of the datable material dating to the Classical-Hellenistic period. Area C, which extends south of the city walls to the west of area D, showed a similar distribution of Late Roman and post-Classical material. In fact, the only variation between Area D and C was the rich assemblage of Late Roman material collected from the habor at Vathy which fell within Area C. This collection of pottery pushed the total quantity of post-Classical material from Area C to close to 18%; without this material, the total percentage of Late Roman material was 13% or only slightly higher than found in the neighboring Area D. Unfortunately, we have not yet been able to place the some of the transects from area C. It is clear, however, that significant quantities of material came from the southern edge of the Thisvi plain where a large pipe factor stands today. The harbor area at Vathy has been completely destroyed by an industrial harbor serving that factory.

Area E to the east of the ancient city tells a similar story to areas C and D. It is notable that the overall assemblage produced by these units was smaller than either D or C (as was the overall area surveyed), and that Late Roman material accounted for close to 25% of all the material collected from this area and post-Classical material represented 27%. Much of this material, however, derived from substantial site situated along the southern edge of the basin and designated E1. Like Vathy, this single concentration of material exerted a substantial influence on the overall character of the assemblage from Area E. Without the material from this site, the overall percentage of post-Classical pottery declines to under 10%.

Since the most significant quantity of post-ancient pottery from the Thisvi basin can be dated to the Late Antiquity, it is perhaps most useful in this short paper to explore how we can reinterpret this distribution of Late Antique material in the countryside in light of the significant new analyses of material from this period in Boeotia and across Greece and with the help of more pliant dataset. It is significant, on first glance, that the distribution of material around Thisvi is similar to that recently published around the city of Thespiai to the east. The team from the Cambridge-Bradford Boeotia Project argued that the overall population of the city of Thespiai declined during Late Antiquity and, as a result, the residents of the city progressively abandoned the immediate hinterland of the city to intensive cultivation. In particular, this meant that the residents of Thespiai stopped the practice of regular manuring the fields near the city which, Bintliff and others argued, deposited ceramic material in a tell-tale halo around the urban core. In place of manuring, Late Roman farmers adopted less intensive agricultural practices and, at the same time, large tracts of land previously dedicated to feeding the urban population became part extra-urban agglomerations ranging from agricultural villas to self-sufficient hamlets.

The decline in artifact density visible for the Late Roman period in the Thisvi basin would fit well with this hypothesis as Late Roman (and more generally post-Classical) densities declining away from the city itself not simple as evidence for contracted habitation, but as the relationship between contracting populations and changing land-use patterns.

The work of the CBBP also revealed large extramural concentration of Late Roman material like those at the southeastern corner of the of the survey area, E1, and at the harbor at Vathy. The former coincides particularly well with kinds of developments documented by the Cambridge Boeotia Survey around Thespiai as villas. The assemblage from the site contained storage vessels consistent with some kind of agricultural installation as well as beehive sherds so common at Late Roman agricultural sites around Thespiai. Moreover, the site was outside the densest areas of ceramics around Thisvi even at its Classical-Hellenistic peek, and this too paralleled the findings of the work at Thespiai.

The harbor at Vathy was a more complex phenomenon. The material at this site was more diverse than a simple agricultural installation and included some of the few example of Late Roman fineware from the survey area in addition to a significant complement of transport vessels which would be expected at a coastal site. Vathy resembles more closely the assemblages present on the islands of Kouveli and Macronisos than the material present inland in the Thisvi basin or even neighboring Thespiai which likewise lacked significant quantities of Late Roman finewares. In contrast, at the Eastern Korinthia Archaeological Survey in the immediate hinterland of the important Late Roman city of Corinth some xx km to the southeast, fineware made up almost 10% of the total assemblage of Late Roman artifacts, and this is despite a collection strategy that would tend to under represent the proportion of fine ware to coarse ware.

It is the comparative context that allows us to begin to make sense of assemblage presented around Thisvi. When first documented and published in the mid-1980s the presence of Late Roman and post-Classical material in the countryside of Thisvi was worthy of remark. Now as “the busy countryside” of Late Antique Greece comes into sharper focus, the dearth of pottery present in the Boeotia countryside and its decidedly functional and non-cosmopolitan character gives pause. There is no question that southeastern Boeotian countryside continued to see investment in post-antique period with Late Antique fortifications extant at Thisvi, Thespiai, Khostia and on Mavrovouni. On the other hand, the lack of imported fine ware for Late Antiquity suggests a particular kind of investment in the countryside. The countryside around both Thisvi and Thespiai during Late Antiquity would appear to be less substantially invested in the kind of prestige rural habitation that is often associated with the concomitant decline in the urban core of the ancient world.

The title of today’s panel was “First Out” and we hope that our paper today extended the potential meaning of that phrase to include the post-Classical material from the first generation intensive pedestrian survey. Our paper today represents a point of departure for further study of both the material produced by the OBE across the Thisvi basin and the growing body of “second wave” survey material from Greece. While much second wave survey material has seen initial publication and has contributed to the present body of knowledge regarding the post-Classical landscape, we have shown the potential in returning to this material. For the Late Roman period, in particular, with think that returning to this material will allow us to move beyond the juxtaposition of rural prosperity to abandonment (a version of the old continuity or change question) and tease out indications of regional difference present in across the Late Roman landscape of Greece. The potential present in returning to the first sherds collected from the Greek landscape in an intensive and systematic way demands that we make the results of these early intensive surveys available in flexible digital formats. It seems like that a return to these survey projects will put an end to any lingering skepticism regarding the long-term archaeological significance of survey data and, in the processes, confirm the continued value of “first out”.

Fine ware and Function at Boeotia Thisvi

For those who are regular readers here, you know that I've been working on re-analyzing the data produced by the Ohio Boeotia Expedition around the site of Thisvi in southwestern Boeotia.  My focus has been on the post-antique material and the Late Roman material in particular.  This is both because the post-antique material is the focus of the panel for which I am preparing this paper and because there has been so much work done in on understanding the Late Antique landscape of Greece since the completion of the fieldwork component of the OBE in 1984.

One of the most interesting characteristics of the main assemblage produced by the OBE from the Thisvi basin is the dearth of Late Roman fine wares.  Both the transect survey and the site based survey of the Thisvi basin proper (this is the area immediately to the south of the city of Thisvi) produced virtually no fine ware.  Only at the harbor site of Vathy was any substantial concentration of fine ware found (and this area only produced a few sherds of Phocaean (LRC) Ware and a Late Roman lamp). 

The absence of any considerable quantity of Late Roman fine ware is more or less consistent with the finds of the Cambridge Boeotia Project to the east.  One of the absolutely fantastic things included in the publication of their survey around Thespiai were a series of data sets.  This data included the finds data from the sites discussed in the volume.  They easily imported into an Access database and could be queried and quantified.  The striking thing is that the villa sites around Thespiai (LSE 7, THS 2, 12, 13, 14 for those with a scorecard) likewise produced almost no imported fine wares.  Now it may be that these villas are "industrial" villas focused on agriculture rather than the luxurious rural estates often associated with the new class of Late Roman aristocrats who looked beyond participation in the local, urban unit to sources for provincial or even imperial prestige. 

The relative dearth of Late Roman fine wares from the countryside of Thisvi and Thespiai can also be compared to the conditions on the islands in the Gulf of Domvrena.  The finds from the islands of Kouveli and Macronisos produced far greater quantities of imported fine ware than the inland sites (for these see T. Gregory, DXAE 12 (1986), 287-304 and T. Gregory, BS/EB 2 (1986), 155-175).  This may well be credited to the status of these island sites as emporia or transshipment points for goods either being manufactured locally (presumably at the "industrial villas") or being imported from abroad.  It is curious, however, to see so little evidence of for the fine ware in the local landscape. 

Another useful point of contrast is the distribution of fine wares across the Late Roman landscape in the Corinthia.  David Pettegrew's recent analysis of this data (Pettegrew, Hesperia 76 (2007), 743-784) showed that close to 10% of all Late Roman material collected from the intensive survey was fine ware most of which was imported.  Likewise, Pettegrew's summary of work at rural villas in the immediate hinterland of the city of Corinth revealed sites that we luxurious in appointment with private baths, colonnaded courtyards, and mosaic floors.  These were the types of buildings likely to produce assemblages including imported fine wares.  In fact, the villa at Akra Sophia suveyed by Gregory at essentially the same time as the sites in the Thisvi basin produce both proper Phocaean (LRC) wares as well as local imitations (T. Gregory, Hesperia 54 (1985), 411-428). 

Even if we must observe some caution in assigning function to a building based on surface assemblage alone, the dearth of fine ware in both the Thespiai and Thisvi assemblages suggests that the Late Roman landscape of southwestern Boeotia is considerably different from that of the Corinthia.  The results of survey and excavation over the last 20 years has not necessary produced a Boeotia countryside that is any less busy (for a nice summary see A. Dunn, in Byzantine Style, Religion and Civilization in Honour of Sir Steven Runciman. (Cambridge 2006), 38-71).  Fortifications, possible Early Christian architecture, rural and urban installations of various types, harbor works and the distribution of Late Roman in general across the countryside point to the continued habitation and, broadly speaking, "usefulness" of the region through the 5th and 6th centuries (if not later!).  At the same time, the absence of wide spread indication of imported fine wares -- a typical and wide spread indicator of not only of prestige installations, but of domestic activities in general -- make it hard to imagine that this area is a deeply connected to the bustling world of Late Roman commerce than even the "deserted" islands found immediately offshore in the Gulf of Domvrena, much less the cosmopolitan assemblages found at our coastal site on Cyprus or the villas of the Late Roman Corinthia. 

This reading of the Late Roman countryside of southwestern Boeotia is important because it represents a more qualified reading of the prosperity characteristic of the Late Roman world in general.  This is not meant to return to the outdated notions of the Late Antiquity as a time of poverty, dissolution, and decline, but rather to demonstrate that the hallmarks of Late Roman prosperity -- namely trade, the wide distribution of prestige goods, and the continued investment in the architecture of display in domestic, urban, and ecclesiastical context -- may have been distributed unevenly across the landscape of Late Roman Greece. 

For more on this research:
Reclaiming Thisve Data
Thisve Basin, Archaeological Visualization, and Curating Digital Data
First Out: A First Draft of An Intro for New Views on Old Data
Survey Archaeology Finds as Data
More on Thisvi in Boeotia

More on Thisvi in Boeotia

The plan is to wrap up a draft of the Thisvi paper by the end of today, and it looks vaguely possible.  This weekend, I ran a bunch of queries on the finds data to attempt to determine the relationship between the ancient and post-ancient material on the site.  As our Archaeological Institute of America panel is supposed to focus on the post-Classical levels at well known sites, then it seemed better to focus on the post-Classical material from Thisvi (and ignore, mostly, the idea that surface material, no matter what the chronology is always "first out"). 

The first step to my chronological analysis was simply to look at the distribution map generated by plotting the artifacts by period across the sites and known transects.  I've put up versions of these maps  before in a slightly modified form.  The maps below include data from the more intensively collected sites (these are dots that do not appear in any survey transect) and in the circular collection units surveyed in the first year of fieldwork near the Hellenistic walls of Thisvi.  Each dot represents one artifact.  Their location within survey transect is arbitrary.

 
 

The first map (blue dots) represents Classical-Hellenistic material, the second (red dots) Late Roman material, and the third (gold dots) Byzantine-Medieval material.  They clearly show that by the Late Roman period, a significant contraction in the distribution of material occurred around the city of Thisvi.  The southern slope of the basin which were quite a busy place in the Classical to Hellenistic period appear to be used far less intensively in subsequent periods.  This seems to represent an overall contraction in the intensive activity areas of the city of Thisvi and parallels to a certain extent the results of the survey at Thespiai to the east. 

Unfortunately, the maps which appear above are incomplete.  I have not yet been able to plot several of the transects from the western most area of the survey (Area C in the map below).

While I think that there is a good chance that I'll be able to place these survey transects in the future (with the help of aerial photographs), at present the best I could do for the purpose of analysis is to compare the distribution of material in each of these sections to determine whether the distributional maps show a contraction of activity or simply a shift in the main area of activity from one part of the region to another.  These charts are based on the almost convincing assumption that the total sample of each area is roughly equivalent and thus the proportions of various types of material is roughly comparable.

A first glance it would appear that Area E and Area C produced substantial more post-ancient material than either Area A or Area D.  This is large due to two significant Late Roman sites in each of these areas.  In the case of area C, the significant concentration of Late Roman material at the beach at Vathy accounted for close to 7% of the overall percentage of Late Roman material from the area.  In Area E, the result was even more dramatic with a single site (E1) producing close to 20% of the Late Roman (and post-Classical material).  Eliminating these concentrations, however, produces a fairly even distribution of post-Classical material across the entire survey area ranging from 7% in Area A to just over 13% in Area E.  As a result, I feel comfortable stating that the distribution of what an earlier generation of survey archaeologists might call "off-site scatter" is relatively consistent across the entire survey area.  This is significant because at the site of Thespiai to the east, the survey team has argued that most of the material in the fields around Thespiai was deposited as the residents of the city spread manure (and broken bits of pottery discarded in trash piles) to fertilize crops.  Thus the distribution of "off-site material" could reflect the intensity of agricultural activity in the basin and the density of settlement at the central site of the survey area, Thisvi itself.

For more on this research:

Reclaiming Thisve Data
Thisve Basin, Archaeological Visualization, and Curating Digital Data
First Out: A First Draft of An Intro for New Views on Old Data
Survey Archaeology Finds as Data

Survey Archaeology Finds as Data

One the small arguments that I make in our paper on re-analyzing the survey finds from the Ohio Boeotia Project is that the changes in technology have influenced the way that data were recorded in archaeology. One thing that is particularly noticeable in the data is that little effort was made to normalize the finds data. This is not because the project was not imagining that their data could be analyzed quantitatively. In fact, the density or distributional data was collected on paper forms that were suitable for direct entry into a spreadsheet type program.

OBE Field Recording Sheet

The finds data on the other hand was collected in a way suitable for a text type catalog. Such catalogues have been a standard part of archaeological documentation for a century. They are typically include measurements for the object and then a textual description of the fabric, shape, decoration, and date of the object with some notes on comparanda. Such thorough textual descriptions of objects is useful for establishing the identity of fossil types in an excavation context. In other words, these objects serve to date stratigraphic layers in excavation and describing them accurate is important for establishing the validity of the identification of the object.

With the introduction of "New Archaeology" (or processural archaeology) in the early 1960s, there emerged a greater interest in quantification and quantitative methods for documenting past human activities. The tools to perform these kinds of quantitative analysis, however, were expensive and time consuming (often involving processing punch cards, expensive mainframe computer time, or even tedious and error prone hand calculations). The increasing availability of personal computers in the first years of the 1980s paralleled the development of important software packages for organizing data on desktop computers. IBM's iconic DB2 came out in 1983 as represented the first SQL driven desktop database. The same year saw the introduction of a powerful new version of the longstanding statistics package SPSS (SPSS-X). Moreover, the portability of both hardware and software made it possible to enter data in the field. This undoubtedly shed light on the practice of data collection in direct contact with data entry (if not on the fly analysis). The desktop computer, SQL driven database software, and new statistics packages put complex statistically driven archaeological research in the hands of even the smallest intensive survey project.

Apple Computer Advertisement from 1985

The Ohio Boeotia Expedition worked on the cusp of these significant changes concluding in 1982. As a result, they collected quantitative data on artifact densities (which could be easily calculated by hand), but did not collect the finds data in rigorously normalized way. This is not to say that the data was not collected systematically. In fact, the systematic and robust collection of finds data has made it possible to normalize significant parts of the finds notebooks. The results can then be projected across the transects that were remapped into our GIS.

Classical Period Finds

Hellenistic Period Finds

Late Roman Period Finds

With time and creativity these data could be translated into chronotype data. The chronotype system is the systematic recording system that we used to document finds from the Eastern Korinthia Archaeological Survey and in the Pyla-Koutsopetria Archaeological Project (as well as several other significant survey projects). We are gradually translating the context pottery from the Ohio State Excavations at Isthmia into this same system. This will create a foundation for some kinds of cross project analysis. At the same time, it will not eliminate the need for careful catalogue entries. The practice of recording careful descriptions of artifacts central to chronological and functional arguments will continue to remain central to archaeological documentation. In fact, the improved ability of desktop database software and "natural language" search engines will make these descriptions increasingly susceptible to the same kind of quantitative analysis as more standardized (and in most cases abbreviated) forms of notation.

First Out: A First Draft of An Intro for New Views on Old Data

To provide some context, this is a first draft of my introduction to a paper that Tim Gregory and I will present at this winter's Archaeological Institute of America's Annual Meeting. The paper deals with survey data produced by the Ohio Boeotia Project in the Thisvi Basin in Boeotia. You can read more on our work to re-habilitate and re-analyze this data here. The paper will be in a session called "First Out: Late Levels at Early Sites".

For our paper, I plan to tweak the meaning of first out a bit. To begin, the surface assemblage, no matter what the chronological range of the pottery present, is always first out. The systematic documentation of surface assemblages provides a means to apprehend the over all site formation process in which the surface of the ground will invariably the most recent events in the history of the site. First out, in terms of survey, data represents the boundary between the work of the archaeologist and the layer formed by "archaeological processes".

For the survey data gathered over the course of the Ohio Boeotia Project from the Thisvi Basin, first out has another meaning as well. The OBE was among the earliest "siteless" style intensive surveys typically associated with the Second Wave of intensive survey in Greece. These projects sought to document systematically the distribution of ceramics across the landscape rather than simply focusing on individual sites. The focus on the distribution of artifacts both "on site" and "off site" produced more robust and complex datasets that pushed the limits of quantitative and GIS technology of the day. The tools that we have now, on desktop computers, provide us with new ways of analyzing the data from these early siteless projects. More importantly, however, these projects capture a layer of an ephemeral landscape. Survey archaeologists have come to realize that the landscape documented by intensive pedestrian survey represents a distinct moment in the history of the surface of the ground. Changes in erosion patters, surface visibility, and cultivation practices significantly influence recovery rates which, in turn, affect the chronological character of the assemblage produced. In a bigger picture sense, the continued development of the Greek countryside (particularly in areas with easy access to the coast like the Thisvi basin) further endangers surface assemblages. Since the overall distribution of material across an area rarely warrants the status of "site", there are few protections in place to prevent the destruction of surface scatters which preserve evidence for both subsurface activity as well as more "low intensity" uses of the ancient and modern countryside.

It's important to recognize that our efforts to re-analyze the survey data from the Thisvi basin is a key step in preserving the results of an effort to document the disappearing landscape. These efforts run counter to the highly-critical stream in the methodological discourse associated with intensive survey which tends to see the results of earlier field work to be methodological unsophisticated, problematic, and potentially misleading. Each new survey purports to make significant contributions both to the survey method and the more general body of archaeological knowledge. The drive to innovate has had produced ever more robust datasets, but has also led to a relative neglect of earlier work. This in turn, has made the task of curating "early" siteless survey data less appealing: despite its potential for capturing disappearing or ephemeral landscapes, it is seen as being too methodologically problematic to reward the efforts of re-analysis.

Finally, we can address first out in the way meant by the organizers of this panel. Perhaps no area has seen greater advances through the work of siteless survey than later periods. The growth of intensive survey coincided, roughly, with John Hayes' landmark works on the chronology of Late Roman ceramics. At the same time, survey archaeology has become an important tool for documenting less visible periods at traditional sites including the Byzantine, Medieval, and Ottoman periods in Greece. Shifts in settlement pattern and the continued privileging of the antiquity in both the historical and archaeological discourse of Greece has made the work of survey archaeology in documenting post-antique levels particularly valuable.

That's my first draft of an introduction. Stay tuned for more on this project over this week as I get to writing up the analysis of the survey data...

Thisve Basin, Archaeological Visualization, and Curating Digital Data

For the past couple of weeks, I've been working on preparing the data from the Ohio Boeotia Expedition (1979-1982) for re-analysis.  This is all working toward presenting an expanded interpretation of the results from this field work at a panel put together by Kostis Kourelis and Sharon Gerstel at the 2010 Archaeological Institute of American Annual Meeting.  The panel is called "First Out: Late Levels at Early Sites".  The site of Thisve is primarily known for its relatively well preserved Hellenistic fortification wall and substantial corpus of published Greek inscriptions.  My paper will focus on the surface remains from across the broader region with particular attention to Late Roman and later material (although the surface assemblage no matter how you excavate, is the first out.)

As of this weekend, I finished keying all the density data from the survey transects that can be reasonably mapped.  There were three or four transects which I've not been able to map in accurately in the GIS.  Despite this missing data, I think that the mapped transects reveal something about the distribution of ceramics across the plain south of the ancient city of Thisve (and the modern villages of Thisvi and Doubrovna). 

One of the interesting things is that this was not the first time that this data was mapped.  Three maps of the archaeological "topography" of Thisve appeared in T. Gregory, "Archaeological Explorations in the Thisbe Basin" in the Boeotia Antiqua II: Papers on Recent Work in Boiotian Archaeology and Epigraphy (Amsterdam 1992), pp. 17-34.  The OBE team plotted these maps using Surface II software against a digitized map of the Thisve plain.  Unfortunately, from what I can gather, this original map of the basin no longer exists.  So I've had to reconstruct the location of the transects from the notebooks in which the data was originally recorded. 

The comparison of these two images and the processes that created them is a nice, small case study both for archaeological visualization and for the curation of digital media.  I think that my more recent map of artifact densities and transect presents a more accurate picture of the distribution of ceramics across the landscape.  That being said, even my plan has generalized.  The samples from most of the individual survey units, mapped as squares in my plan, were taken from a 1 m sq area.  I've extrapolated them across the entire unit (i.e. the width of the transect x the sampling interval).  The lower images, generated by Surface II plotting, have simply extrapolated the density of artifacts across the entire Thisvi plain.  I suspect that the linear arrangement of survey units (an early form of the Sydney Cyprus Survey Project's "Souvlaki method" of surveying the landscape) exposed the distribution plots to various kinds of "edge effects" as the software was asked to extrapolate artifact densities farther and farther from known or established data points.  Moreover, the jagged abstraction of these figures makes it difficult to assign the surface densities to real space on the map (note the lack of scale or even a north arrow on figure 3.6!).

The disappearance of the Surface II data is another important issue.  While it is easy to criticize a project for failing to responsibly curate their data, in fact, the field notes books and survey sheets from the project are well-maintained and organized.  The maintenance of data produced over the course of secondary analysis is a challenge for a small project like the OBE which worked in the area for only three years and published their analysis and then dispersed.  Survey projects, in particular, suffer from rather ephemeral constitutions (as opposed to the usually more permanent relationship between excavators and a particular site).  If the relatively low impact of survey archaeology on the landscape tends to attenuate the link between the archaeologist and a particular place, then the combination of paper and high tech applications ranging from relational databases to GIS mapping applications adds a layer of complexity to curating the digital data that these projects produced.  In most cases, data was (and I'd argue still is) collected from the field in paper form and then keyed and plotted into digital databases of various descriptions.  So the digital data represents the first phase of analysis rather than a primary data collection.  Perhaps this is part of the reason for failing to maintain the digital data as carefully as the paper forms and notebooks.  In recent years, a more serious approach to the practices involved in curating digital data (and survey data more generally) will undoubtedly change future practices.  Hopefully our work with the data from Thisvi will represent an important case study for the curation of digital data in the context of re-analysis.

Reclaiming Thisve Data

Long time readers of this blog will remember that some very long time ago, I was working on digitizing and ultimately publishing the data from the Ohio Boeotia Expedition.  This was a very early "second generation" intensive, siteless pedestrian survey conducted around the ancient city of Thisve in Boeotia directed by Tim Gregory.  Recently Archie Dunn at the University of Birmingham began a new project there designed to document the Late Roman and Byzantine remains from the town.  He invited us to re-examine the survey data in the context of his new work with the hope of producing a more thorough publication of the O.B.E.'s work as well as to complement the more recent work both at Thisve and throughout Boeotia.

The Thisve basin

I started work on this project about two years ago in Athens, but quickly moved it to the back burner when I was confronted by very difficult mapping and digitization issues.  In short, I had a terrible time mapping into my GIS the transects that the O.B.E. walked in the landscape.  This was primarily because I was deferring to some published maps of the project's work.  I simply could not get my maps (based on some newly collected GPS data, the Greek Army Mapping Service 1:5000s, and augmented by Google Earth images) to coincide in any recognizable way with the published drawings of the survey area and the location of the sites.  As I returned to the data this fall -- in preparation for a paper at the Archaeological Institute of America's annual meeting -- I decided to scrap the published maps and attempt to build my own new map.  That was a much easier approach and quickly demonstrated that the published maps were, for lack of a better word, wrong. 

The other decision that I had to make was to accept that I would not be able to map some of the transects.  The recorded data was simply too flawed (and the equally, but differently flawed published maps did not help) to allow me to reconstruct all the transects that the O.B.E. surveyed.  This was disappointing, but also liberating as many of the project's transects digitize quite easily.

I also began to key the finds data from the project.  The density data, as shown in the image above, produces some recognizable trends.  Artifact densities appear to be higher in the immediately vicinity of local (and presumably ancient and Byzantine) villages, decline across the open agricultural plain, and pick up again along the southern boundary of the plain.  The reasons for this might be geomorphological.  There is evidence, in fact, for an ancient water management system in the plain which Pausanias described:

Paus. 9.32.3. Nothing would prevent the plain between the mountains becoming a lake owing to the volume of the water, had they not made a strong dyke right through it. So every other year they divert the water to the farther side of the dyke, and farm the other side. Thisbe, they say, was a nymph of the country, from whom the city has received its name.

So, it may be that the sediment from the flow of water through the plain was covered ceramic evidence there (although there is some evidence for ancient material in the Thisve plain).  It could also mean that this was largely agricultural ground and did not see the kinds of activities that deposited large quantities of ceramics.  The only problem with this is that the Cambridge/Bradford Boeotia Project which along with its various spin-offs has surveyed numerous cities in ancient Boeotia, has argued that fields around Boeotian cities were often filled with ancient ceramics deposited through the process of distributing manure in the fields nearest to the town.  A more sophisticated and careful comparison of the distribution of material outside of Thisve and outside of the neighboring city of Thespiae may shed some light on this practice.

More importantly, the quality of the finds data is actually quite good.  I am confident that digitizing the finds collected and analyzed by the O.B.E. will produce a nice corpus for comparison with other survey projects both in Boeotia and in the "suburbs" of other cities in Greece.  Over the next few weeks, I'll update my progress and analysis of this data more or less regularly.

The Medieval and Post Medieval Mediterranean at the 2010 Archaeological Institute of American Annual Meeting

We got the good news last week that the panel put together by Kostis Kourelis and Sharon Gerstel for the 2010 AIA Annual Meeting in Anaheim has been accepted.  The panel is titled First Out: Late Levels at Early Sites and will feature papers by Jack Davis, Kathleen Quinn, Anne McCabe, Adam Rabinowitz, Guy Sanders, and Tim Gregory and myself.  Here's a link to the abstracts and overview statement.

Tim Gregory and I plan to re-examine the data produced by the decades old Ohio Boeotia Project around the ancient city of Thisvi.  This survey data was initially analyzed in a series of publications in the 1980s.  Since that time, digital analysis tools have become considerably more powerful and there is a growing body of work in the region, particularly associated with the Cambridge Boeotia Project and its various spin-offs, that promises to add significance to any re-examination of the OBE results.  Returning to excavation and survey results -- so called legacy data -- has taken on new importance in recent years as excavation permits have become more difficult to acquire, a vigorous ethical discourse has put pressure on project directors to make unpublished finds available, and the digital archaeology "movement" has improved our ability to make published and unpublished data alike visible and accessible to the professional public.  A recent issue of the leading electronic journal in archaeology, Internet Archaeology, has dedicated an issue to the reanalysis of "legacy data" taking advantage of the intersection of digital distribution, new technologies, and the remarkable potential of the existing pool of archaeological data to inform contemporary research questions.  We hope our paper frames not only some of the methods and procedures at stake in the re-examination of survey data, but also makes the argument that this kind of secondary analysis marks the coming of age of intensive pedestrian survey.  It marks the potential of survey data to go beyond its applicability to narrowly defined research questions and to have the kind of enduring value that excavations have nurtured by long standing methods and carefully cultivated archival practices. 

Proving that survey data is available for re-analysis is absolutely critical for its persistence as an archaeological methodology in the Mediterranean.  And the recent transformation of post-Classical landscapes from spaces seen as stagnant and unchanging to dynamic "contingent" countrysides makes the study of the post-Classical world ideally suited as a test case.

Papers, Projects, and Perspectives for Next Year...

This fall I was invited to present my research on campus in the spring. The talk will be to a general audience and sponsored by the Graduate School here at the University of North Dakota. There are relatively few opportunities for junior faculty to present their research on campus, so it's an exciting opportunity, but I need to come up with a topic.  The process of thinking about a topic gave me a chance to reflect on my ongoing research and try to find a way to bring together my various interests or at least prioritize my research so that I can have a productive spring.

So, my research agenda in no particular order...

1. Pyla-Koutsopetria Archaeological Project.  As any reader of this blog would know, I co-direct the Pyla-Koutsopetria Archaeological Project.  I am quickly learning that an archaeological project will expand to fill any time left unoccupied by other research or teaching demands (and it will often try to bully those projects to the sideline as well!).  That being said, the project also provides any number of opportunities for productive and interesting digressions ranging from methodologies, to general considerations of Cyprus in Roman Antiquity to more focused studies on Late Roman trade, ecclesiastical architecture, and the history of ancient Kition.  Moreover, my experiments with video, podcasts, and blogs have formed the basis for emerging ideas about narrative, archaeology, and the new media.  This would certainly be the most entertaining project to present to a general audience in that it could be rooted in narrative and include photos, videos, and even audio clips.

2. Early Christian Architecture in Greece.  I spent a good part of the last two years wrestling with my long moldering dissertation manuscript.  I have presented various fragments of it over the past few years to generally positive results, but as I spin parts off from it, the entire project begins to lose its conceptual cohesion.  Despite its conceptual fragmentation, I am working on a rather lengthy article which explores the role of churches in the process of Christianization in Greece.  I have posted versions of this work on this blog.  My most recent work has emphasized the role of churches as dynamic, hybrid space that served as a place of convergence for the many interests at play within Late Antique Greek society.  This work seeks to undermine the singular, hegemonic church so often portrayed in literary sources of the period and replace it with a more fluid institution which embraced ambiguity in its efforts to translate a universalizing discourse on a local level.

3. Dreams, Inventio, and the Memory of Early Christianity.  This project complements project 2 in looking at the remains of the Early Christian architecture in the archaeological record of Byzantine and post-Byzantine times.  I hold out the hope that I can somehow bring this project and project 2 together in a cohesive book manuscript.  I presented some of this research for the first time this fall at North Dakota State University and received some valuable feedback from colleagues there.  The core of this project is an effort to show that the massive number of basilica style churches built in Early Christian times had a profound impact on religious landscape of Byzantine Greece (and the entire Eastern Mediterranean).  At my most ambitious moments, I sometimes imagine that Early Christian architecture might have served as a vital filter between the remains and memory of Classical antiquity and the needs of Byzantine and even post-Byzantine society.  The frequent appearance of Early Christian spolia in Byzantine churches and their not uncommon appearance in Byzantine texts suggests that Byzantine society recognized the importance of the Early Christian period in the formation of their identity.  This challenges the more pervasive perspective that Byzantines sought primarily to establish ties to Classical antiquity.  In fact, I'd tentatively suggest that scholars' tendency to overlook Early Christian spolia speaks more to the traditional aesthetic values of Byzantine architectural historians than those of the Byzantine architects.

4. The Continuing Corinthia.  I continue to dabble in the fortifications and landscape of the Eastern Corinthia.  Most of this is in collaboration with David Pettegrew and comes from our ongoing work with the data produced over the course of the Eastern Korinthia Archaeological Survey.  Some of this involves our work documenting an Early Modern settlement at a site called Lakka Skoutara.  We have documented this site over the course of almost 10 field seasons with particular attention to archaeological formation processes in the Greek landscape.  David has also helped me continue to document the fortifications of the Eastern Corinthia (Project Fortress Corinthia).  David Pettegrew and I will present some of our recent research in this particular direction next month at the annual meeting of the Archaeological Institute of America.  I am also working with Tim Gregory to digitize and normalize the context pottery from the Ohio State Excavations at Isthmia.

5. Thivi-Kastorion Archaeological Project.  This is another archaeological reclamation project.  I am working to re-analyze survey data collected by the Ohio Boeotia Project in the vicinity of Thisvi, Boeotia with Tim Gregory.  We are collaborating with Archie Dunn who is conducting an archaeological field survey of Thisve/Kastorion, Greece.  Our hope is to produce new maps of the Thisvi basin that combine the archaeological data collected in the late 1970s and early 1980s by the Ohio Boeotia project and Archie Dunn's more recent work at the site.

So, this spring one of these projects (perhaps more) needs to come together in an engaging public lecture (at least) and ought to move forward toward a publication phase (and to be fair some of these projects have some good traction right now!).