Late Antique and Byzantine Anatolia

Last week I worked my way through John Haldon, Hugh Elton, James Newhard, Archaeology and Urban Settlement in Late Roman and Byzantine Anatolia: Euchaïta-Avkat-Beyözü and its Environment (Cambridge 2018) in preparation for my annual trek to the Eastern Mediterranean for field work. As the major field seasons for the survey phase of the Western Argolid Regional Project have concluded, we have begun to think more about what we need to do to publish our results. While I have tended to focus on the sherds on the ground (and in the project’s GIS), Haldon et al. reminded me that there was much more than just field data to producing a significant regional study. 

I don’t really write reviews here, but here are four or five thoughts on the book:

1. Low Density and Limited Collection. The area around Euchaïta-Avkat-Beyözü produced very few sherds and even fewer that were diagnostic. Moreover, they could only collect sherds from the Roman period and later, and this created a particularly challenging relationship between their study assemblages and the distribution of material on the ground. James Newhard’s clever methods for smoothing ceramic densities over different sized units, different surface conditions, and different visibilities provided a foundation for interpreting the assemblages collected and studied from the survey area. 

A bit less clear was the relationship between these artifact densities and the kinds of sites that the project asserted existed in the landscape. It was a bit hard to understand the difference between an independent structure, house, farmstead, and watchtower, for example, in the text itself, but the detailed discussion of these functional categories appeared in a later appendix. I’m still not entirely sold on this method of creating sites, but there is something compelling about the complexity of the historical, landscape, and archaeological variables considered in site definition.    

2. Climate and the Environment. I tend to look at the surface and artifacts when I think about archaeology. In a pinch, I’ll think about a building or a strata. I rarely step far enough away from the artifactual landscape to think clearly about the environment and climate as important factors in understanding how people in the past lived in their world. This is obviously a blind spot in my research focus, and as I extend my interests into more recent periods, the pressing realities of climate change, for example, and our adaptation to the changing environment in the last 50 years, has nudged me to expand how I think about the archaeological universes that I study.

Archaeology and Urban Settlement demonstrated the potential of a careful study of the ancient environment at a regional scale for understanding the development of settlement, agriculture, and land use in their region. Interestingly, their study area had rather few opportunities for sampling pollen or other scientific approaches to studying paleoenvironmental variable. Nevertheless, the team was able to draw one evidence from Ancient, Medieval and Early Modern texts as well as modern agricultural and climate date to model the ancient environment in useful ways. They demonstrated that the landscape around Avkat was not unproductive, but as relatively marginal in antiquity as it was in the 21st century with most communities surviving on the cultivation of cereals and pastoralism. Climate change appears to be just one of the variable that shaped changes in agricultural practice, settlement and life in the area. 

3. Roads and Routes. In the Western Argolid, we think constantly about roads and routes through our survey area. In fact, travel through the Inachos valley and its relationship both to neighboring Arcadia and Corinthia as well as the Argive plain to the east, was part of the original plan for the survey project from the onset. So far, we’ve written a few papers that attempted to understand settlement and movement in our landscape and have thought about the relationship between water, routes, bridges, and churches. In general, we have not used least-cost path kinds of analysis, in part because we have some ethnographic and archaeological information on movement through the valley, and in part, because the flat or gently sloping Inachos River valley bottom exerts a strong pull on any path through the area. As a result, we’ve leaned a bit more heavily on cultural factors on movement through the valley, and considered the ways and reasons for which known routes defy least-cost expectations to avoid crops and fields, to follow the line of an aqueduct, or to pass close or far from settlements.

 Archaeology and Urban Settlement does a nice job integrating historical and topographic information into mapping movement in their survey area. This not only provides context for the relationship between sites and routes, but also demonstrates the tension between persistent major routes that shaped the significance of major settlements in the region and the dynamism of smaller routes that linked settlements to their fields or rural sites to other rural sites. While such temporal variability across the landscape is hardly surprising, it is worth noting the trans regional movement on major routes likely represented a less common and regular kind of movement in a landscape. The permeability of the countryside, in contrast, might have reflected myriad, changing smaller routes that accommodated more regular traffic on a daily basis. 

4. Foodways and Ceramics. One of the more intriguing sections of the volume was Joanita Vroom’s chapter of Byzantine foodways and ceramics. Because the local ceramic typologies were relatively poorly know, it was rather difficult to identify and date the surface assemblages. Rather than create an unmoored typology or speculate too wildly on potential economic or social links between the ceramics present in the survey area and potential production sites, Vroom focused on the evidence for Byzantine foodways in the region. By compiling evidence for food, trade, and the related vessels need to provide sustenance to communities who lived in the region.

On the one hand, there is little that is specifically related to the region around Avkat, but, on the other hand, her chapter continued her effort to redefine the study of ceramics from the vessels themselves to their role in the everyday life of Late Roman and Byzantine communities. When this attention to foodways intersects with routes through the area, paleoclimate studies, and agricultural history and ethnoarchaeology, and, of course, excavated and survey ceramics, I can imagine an opportunity to connect the broadly general with the individual at the scale of the landscape, and this is an exciting proposition. 

5. Publishing Data. One particularly intriguing element of the book is that most of the maps and many images were published digitally via Open Context rather than printed in the book itself. This is useful for the digital book, where, if you’re on wifi, the image is just a click away. I was reading on my iPad, on a flight, so I lost a bit of that convenience, but back at my laptop everything worked fine. I imagine that for a reader of the paper book, this would be a bit more inconvenient. 

More promising still is the prospect that the project will publish its full datasets on Open Context in the future.  

 

Sometimes Survey Makes Sense: A View from Chelmis

The formation processes that produce the surface context studied through intensive pedestrian survey are really annoying. They hide things that you know MUST be there (like Late Roman material on a prominent coastal height overlooking a Roman to Late Roman settlement). They make visible things that have no rational explanation (like the famous one sherd of Classical black-slipped pottery on a lonesome hillside). Various formation processes scramble and smear and move and interrupt and complicate assemblages across the landscape and force survey archaeologists to keep stepping back and back and back until the picture comes into a kind of blurry focus. Outliers are noted, but major patterns become the basis for analysis.

This past week, I’ve been working on an article with Grace Erny on the houses of Chelmis in the Argolid which we are study as part of the Western Argolid Regional Proect. Grace is going to give a paper on our preliminary results at the annual Archaeological Institute of America meeting next week (check out a list of all the WARP papers here). These are “Early Modern” to “Modern” period houses around which we conducted intensive survey in 2016. To get a sense for the impact of the houses on the distribution of artifacts in their immediate vicinity, we did 10 m of very intensive documentation counting all the roof tiles, ceramic artifacts, and other objects around the houses. We grouped our survey into three units on each side: one of 2 m in width walked by a single walker and two of 4 m in width walked by 2 walks. These units produced largely consistent patter of a 65% percent drop off between the 2 m unit and the first 4 meter unit and then a 25% drop off between the first and the second 4 m unit.

Chelmis4Blog

Not every house produced this distribution, but it was consistent enough both at Chelmis and at the two other sites where we conducted intensive survey to qualify as a pattern. And this is worthy of note in a universe where the trickeration of formation processes often makes any small scale patterning of artifact densities in the landscape either rare or suspect.

Hoping that all your survey assemblages pattern predictably in the new year! 

Siteless Survey and Artifact Densities

This week, I’m continuing to work on producing a preliminary report for the Western Argolid Regional Project. One of the particular challenges in writing a report is the tension between the granularity of our survey data and the size and complexity of our survey area. As a project committed to conducting siteless, “artifact-level” survey, field walkers spaced 10 m apart collected all artifacts visible on the surface from their 2 m wide swaths. These artifacts were all analyzed and given a chronological range, a functional category, and, whenever possible, a place within existing artifact typologies. At the same time, we also counted each visible artifact using a clicker counter allowing us to have an almost instant assessment of the quantity of sherds present in each walker’s swath and each unit. We recorded this data, along with basic environmental data collected from each unit, on a survey form that we then entered into the project database and projected into our project’s GIS.

In a traditional intensive survey, artifact densities serve as an indicator of sites in the landscape. These projects then documented these sites with a different level of intensity usually through gridded collection or some other more spatially rigorous sampling strategy. This allowed survey projects to distinguish the assemblages produced by these “on site” practices from those produced by “off site” survey practices which generally involve larger transects and less intensive sampling regimens.

Artifact level survey, in contrast, tends to emphasize a consistent method for sampling the landscape, in part, because they recognize the complexities of site formation in the landscape and approach any definition of a site with skepticism. In fact, my colleagues and I argued over a decade ago that many sites in the Corinthian represented the complex interplay of assemblages from multiple periods rather than a single “multi-period” site with recognizable continuity in activity. In this context, then, overall artifact densities offer only the coarsest indication of activity in the landscape or, worse, are illusory when they occur only when the edges of unrelated, narrow-period scatters overlap. A growing awareness of geomorphological and geological processes, varying levels of artifact visibility, and changing vegetation, the presence of background disturbances and the character of the surface clast, further complicate any argument that high density sites produce meaning in the landscape. 

As this approach to intensive survey fundamentally questions the value of artifact densities as the basis for the historical analysis fo the landscape, are artifact densities meaningful in siteless survey? 

As I work on the WARP preliminary report, I argue that they are for three reasons. First, artifact densities do provide a measure of artifact recovery rates from a particular unit especially when combined with surface conditions like visibility. A unit that produces a high artifact densities in relation to visibility is probably a unit with a significant quantity of material obscured by the independent vagaries of surface conditions. Knowing this kind of information, for example, allows us to consider more carefully issues like the extent and continuity of an artifact scatter datable to a particular period. While we may not be able to control for all the site formation processes that shape a context as complex as the surface, we can sometimes control for the variables that impact artifact recovery from the plowzone.

The second advantage of collecting density data for intensive survey is that it provides a context to measure the diversity of the assemblage present in a unit. We recognized in studying data from the Eastern Korinthia Archaeological Survey that the chronological and functional diversity of an assemblage tends to increase with artifact densities. In other words, we had very few examples of units where a single period – or even artifact type (think: modern tiles) – increased artifact densities in a significant way. We then applied this realization to lower density units that often result from compromised surface visibility or other conditions. In these contexts, we discovered that some units with low artifact density tend to produce more diverse assemblages than others. These high diversity, low density units may provide windows into more complex scatters that the vagaries of surface conditions and site formation obscured. 

Finally, artifact densities do provide insights into large-scale, diachronic patterns in the landscape. Diachronic intensively survey has some chronological challenges in that artifact level survey tends to push us to fragment the landscape into the finest periods possible. In some cases, these periods are quite narrow (final decades of the 4th century) and in other cases quite broad (Medieval or Classical-Roman).  The broadest and narrowest periods create some challenges of commensurability as they tend to produce vastly different meanings in the landscape. A narrowly dated artifact, for example, might well represent a particular activity – such as a funeral practices or household activities – whereas a broadly dated artifact tend to represent broader and more persistently uniform functional categories: storage or roofing. Understanding the relationship, then, between landscapes defined by narrower chronological (and usually functional) categories and and those defined by more broad categories is difficult and always risks stripping from countryside the temporal aspect of persistent activities while pocking it with episodic behaviors tied closely to more precisely datable artifacts. Total artifact densities offer a big picture way to see human activity in all its complexity and consistency across the landscape. These densities must, of course, be read critically and contextually, but that’s true of all archaeological evidence.

 

 

Writing WARP Wednesday

Over the last week or so I’ve been working on writing a draft of the preliminary report from the Western Argolid Regional Project. Yesterday, I returned to one of the more interesting issues facing intensive pedestrian survey in the Mediterranean: the matter of intensity.

WARP was a siteless survey and we took this quite literally. We did not return to sites after our initial survey to conduct more intensive investigations or gridded collections. In fact, we tended not to talk about sites at all (outside of the commonplace naming convention associated with particular known ancient places in our landscape, like the walls of the polis of Orneai) and assumed that high density scatters in the landscape could as easily represent a single period as the overlap of a number of periods through time. (Our experience in the Eastern Korinthia had suggested that many areas of high artifact density in the landscape reflected the overlap of single period scatters that may or may not be contiguous.)

This being said, we did recognize that more intensive practices – such as total collection or intensive sampling across small grid squares or other forms of “hoovering” – produced more robust assemblages of material. We experimented a bit with this on the Pyla-Koutsopetria Archaeologcial Project and while our results are very, very boring to read, they demonstrate that careful collection with our noses to the ground produced artifact densities that we much higher than traditional field walking. At the same time, the uniformity of the assemblage at Pyla-Koutsopetria ensured that more intensive collection strategies did not produce a more diverse assemblage. In other words, on PKAP, where our assemblage was pretty uniform, doing more intensive artifact collection did not yield more nuanced results.

In the Western Argolid, the situation is a bit different. The survey area is larger and encompasses a generally wider range of environmental conditions. Moreover, our artifact scatters across the survey area tend to reflect more complex and varied historical processes than we found at our “large site survey” on Cyprus. As a result, it seemed like a good place to run another series of experiments to see whether more intensive collection strategies produced different kinds of assemblages across the survey area. In particular, we were interested in seeing whether more intensive collection strategies revealed the presence of “hidden landscapes” consisting of pottery that tends to be overlooked during traditional survey field walking.

To do this, we set down slightly over 100 2m radius total collection circles across our survey area. These resurvey units fell within our existing units and we intended to use them to compare the resurvey assemblages to the assemblages produced by traditional intensive survey.

This proved pretty challenging for a number of reasons. Some involved research design. Our resurvey units were total collection circles with a 2 m radius (i.e. 12.57 sq m) and we typically did two per unit for about a 1% sample of the survey units which averaged about 2500 sq m. This sample could then be compared to the must larger, but less intensive ~10%-20% sample collected using typical survey procedures of field walkers spaced at 10 m intervals looking 1 m to each side. This comparison, of course, isn’t all that great. First, we recognize variation in the surface assemblage across the unit so it makes sense that our resurvey circle may well capture an assemblage with a different character than the assemblage produced by field walking. Ideally, they two assemblages are similar in some way, but some variation is likely to reflect that, short of total collection of the entire survey area, surface sampling is not designed to discover an example of every kind of sherd present in every survey unit, but a representative sample of the area generally. More than that, there was a general tendency to locate the resurvey circles in areas with higher visibility than the average visibility across the survey unit in general shifting slightly the recovery conditions for material.

That being said, our preliminary analysis of the material has produced some other challenges as well. First, it’s very difficult compare complex assemblages. We have remarkable ceramicists who are capable of defining pottery in very granular ways both in terms of typology and chronology. This granularity makes it difficult to compare two assemblages because the variation inherent in how we analyze ceramic artifacts. For example when I compared the resurvey and the survey assemblages this past summer, the vast majority of assemblages showed little chronological overlap based on the specific periods assigned to each chronotype. On the one hand, this meant that our resurvey circles were producing different kinds of chronological information from the larger survey units in which they were situated, but, on the other hand, this chronological need not result in substantively more knowledge about the artifact scatter. An artifact datable to the “Late Hellenistic to Early Roman” period is different from a “Hellenistic to Early Roman” artifact, to be sure, but this kind of granularity is as likely to represent the irregular distribution of chronological knowledge across a typology than it is to represent a general pattern of artifact dates that would influence the identification of the function of the site, for example. (I.e. some kinds of pottery, say fine ware, can be dated more specifically than other kinds of pottery.)  

It is worth noting, however, that in general, standard survey units produced more artifacts with narrower chronologies than the resurvey units did. While this, in and of itself, is not meaningful (as, for example, prehistoric pottery dated quite specifically still tends to have wide chronological ranges than historical period pottery), it suggests that our standard survey practices produced assemblages that were susceptible to chronological analysis that are at least comparable and perhaps more fine grained than our total collection circles.   

To mitigate this, I started to generalize the chronological categories a bit, grouping the finely honed chronology offered by our ceramicists into “Prehistoric,” “Greek,” “Roman,” “Greek/Roman,” “Medieval,” “Modern,” and “General” for artifacts only dated to broad spans of time. This aggregation, predictably, made the comparison between our two assemblages easier. For example, it showed that over half of the assemblages produced pottery of broadly the same date. Moreover, it allowed us to observe that the smaller the assemblage the less likely overlap occurs. In other words, our smallest assemblages from either standard survey or resurvey were not just producing worn, undiagnostic pottery that we tend to aggregate into more general chronological categories, but that they also produced variation. 

It also gave us a way to see if the resurvey units had any particularly telling trends. It appears, however, that the resurvey units did not yield, as a general pattern, assemblages that could be dated to narrower periods than their standard survey collections. This tells us that our more intensive collection circles are not producing more narrowly datable pottery in general, but not necessarily that they don’t produce pottery datable to particular narrow date ranges.   As an example, the only the resurvey assemblages produced any examples of Final Neolithic and FN-Early Helladic I in those units that we resurveyed. Other units in the survey area, of course, produced material from those periods.  

 

 

 

 

Writing the Western Argolid

Over the last few days, I’ve been working on a preliminary report for the Western Argolid Regional Project. I mentioned on Monday how writing a preliminary report is always a bit of a fraught exercise, but when actually writing, it is easy enough to put that out of your head and focus on the words on the page.

As part of writing the report, I re-read some of the rather scant ancient sources on our survey area. Pausanias 2.25.4-6 discusses our survey area specifically and twice he notes that there isn’t much to see. In general, Pausanias sees the Inachos valley as an extension of Argive territory and a route between Argos and the neighboring city of Mantinea in Arcadia. This same lack of interest shaped how 19th century travelers treated the region with none that I have encountered venturing beyond the Venetian (?) period fort at the site of Skala where the Inachos valley widens out onto the Argive plain. 

Later scholars – namely Kendrick Pritchett – attempted to reconcile Pausanias’s description of the site of Lyrkeia being 60 stades from Argos and Orneai being 60 stades from Lyrkeia. This involved him poking around the sites of Melissi where the French excavated some Mycenaean chamber tombs in the early 20th century and Chelmis, where there is a substantial scatter of Classical period material around a church dedicated to the Panayia. Since Pausanias’s notes that Lyrkeia was in ruins by his day and suggested that it was destroyed before the Trojan War, and hence, was left out of the the catalogue of ships in the Iliad, Pritchett is content to identify it with something in the vicinity of the Melissi tombs rather than in the neighborhood of Chelmis. More than that, he suggested that Chelmis does not seem to be on a major route through the area so seemed to be an unlike stop for Pausanias who seemed mostly concerned with sites along the Inachos river bottom. Greek scholars, Ioannis Pikoulas and Ioannis Peppas, have explored the region a bit more thoroughly but also tend to follow the routes along the valley bottom that Pausanias’s traced in his sojourns from Argos.

The entire effect of the tradition from antiquity to modern times is that this region is peripheral to Argos and a mostly a travel corridor from the Argive plain to points west and north. Our project essentially tested this hypothesis both by exploring intensively the valley bottom and surrounding region to determine whether Pausanias’s somewhat laconic description was justified, and by considering the region in its own right to understand whether networks of settlement and movement functioned independently of the “central places” of the Ancient, Medieval, and Early Modern world.

As a hint, we have found some evidence that this was, indeed, the case and the Pausanian landscape suffered from his general (and well-documented) lack of interest in post-Classical sites, but also the tendency of central places and their political and economic networks to overwrite and obscure patterns of settlement and movement in the landscape that reflect decentralized and more local traditions. As Tom Gallant noted 25 years ago, these decentralized networks of relations supported a kind of social insurance for communities by allowing them to diversify the risk that came with overly strong ties to central places. While these networks are pretty hard to see in archaeology, there are signs that they exist throughout our survey area and not only help us understand the presence of sites that don’t conform to the Pausanian itinerary but also reflect a dynamic countryside that was more than simply the productive coda to the consumer city.

Lakka Skoutara and Abandonment: PrePrint with Pictures

Over almost 15 years, David Pettegrew and I have been revisiting the rural settlement of Lakka Skoutara in the southeastern Corinthia and documenting the changes. At first, our interest was to document site formation processes at the site and observe how abandoned buildings and houses fell down over time.

Figure 8 house7 image20 2009 d263338a90

After only one or two visits, however, we discovered that these houses were not simply left alone to collapse in the Greek countryside, but continued to be centers of a wide range of rural activities. For example, several of the houses lost their ceramic tile roofs during over the past 15 years, and others have seen regular maintenance and, in at least one case, expansion. As a result, our research shifted from a rather abstract (and naive) view of this settlement as a case study for site formation to a more dynamic and complex project designed to document the material engagement with the Greek countryside over a period of 15 years.

While it goes without saying that the history of rural Greece continues to attract attention from anthropologists, historians, geographers, as well as local antiquarians, there has been relatively less formal and systematic archaeological study of 20th century rural sites. Our work at Lakka Skoutara is not entirely unique, but it makes a useful contribution to the small number 20th century rural sites that have received systematic and sustained archaeological study in Greece.

You can download a draft of our paper here. Or read about our most recent visit to the site here.

Boeotia Project, Volume 2: The City of Thespiai

Over the last few weeks I’ve been snacking on John Bintliff, Emeri Farinetti, Božidar Slapšak, and Anthony Snodgrass’s Boeotia Project, Volume II: The City of Thespiai: Survey at a Complex Urban Site (2017). It’s a big book that is both impressively synthetic and filled with many distinct observations on the distribution of ceramics and survey methodology. The book focuses on work done in the 1980s at the long-lived urban site of Thespiai. Like my own project at Koutsopetria on Cyprus, the intensive pedestrian survey was not designed to locate the site (or small ex-ubran sites in the countryside), but to document the assemblage, distribution, and extent of material at a “complex urban site.”  The distributional analysis of the ceramic material from the survey interested me the most, although the book also brings together architectural fragments, epigraphy, Ottoman administrative documents, and ceramic analyses into a series of synthetic histories of the city.

I might venture a more thorough review of the book after I work my way through it all, but, for now, I wanted to record a few observations on chapter 3 which unpacks the distribution of ceramic material across this large urban site. There are five things that piqued my interest in these chapters.

First, the intensive survey around Thespiai was done in the mid-1980s meaning that this project drew upon a kind of legacy data. The authors were particularly up front about the challenges of these datasets, the occasional irregularity of their survey work, and the adaptation of their methods over time. It was a nice reminder that for all the methodological rigor associated with survey and for all the efforts projects make to demonstrate the systematic character of their artifact collection, intensive survey projects both adapt over the course of projects and in response to the landscape. Bintliff and his colleagues occasionally used irregularities to their advantage, such as when they compared an assemblage collected from units that were accidentally resurveyed with that from the original collection to demonstrate that larger assemblages tend to be more diverse. 

Second, for many of the survey units Bintliff’s team conducted both standard intensive survey collection at 15 m spacing over units that were around 3000 sq m, as well as more intensive samples over 300 sq m from the same 3000 sq m units. Comparing the assemblages produced by these two different methods demonstrated that the more intensive collection samples did not necessarily produce more chronological information than the less intensive transect walking. This confirms some of the experiments that we ran at the Pyla-Koutsopetria Archaeological Project and the preliminary analysis of our data from the Western Argolid Regional Project

It was a bit more striking that the density data from both collection methods produced more or less comparable. In my experience, more intensive sampling tended to produce much higher density per hectare than counting visible artifacts while walking units at 10 to 15 m intervals. The similarity in density counts and the distribution produced through the different methods is a remarkable sign that their less intensive methods were appropriately calibrated for the nature of the surface assemblage at this large urban site.

Third, over the last decade, I’ve been fairly concerned with issues of surface visibility in intensive survey. In fact, I’ve tended to think about visibility as having a particularly significant impact on the chronological and functional character of artifact assemblages. I’ve made two arguments for the role of visibility. First, as visibility decreases, assemblages tend to become less chronologically diverse and the most common artifacts, which tend to only date to broad periods, to dominate these assemblage. Second, low visibility units with particularly diverse assemblages likely represent windows into higher artifact density surfaces obscured by vegetation. 

Because I’m not particularly interested in overall artifact densities, per se, I’ve been reluctant to correct artifact densities for visibility and, instead, focused on identifying units with anomalous densities or diversity for their surface visibility.  Even if I was interested in overall artifact densities, however, I’d probably want whatever correction is applied to them unpacked more explicitly than the authors of this book provide. More than that, Bintliff’s long term interest in hidden landscapes might recommend greater attention to visibility as a key factor in obscuring and revealing periods that tend to be difficult to see on the surface.

Fourth, I was quite intrigued by the authors’ argument that earlier periods might be obscured by later period overburdens across the survey area. On the one hand, this is certainly the case particularly over the span of millennia, with evidence for prehistoric periods hidden, destroyed, or otherwise compromised by later eras that also tend to produce more materially visible marks on the landscape. I do wonder, on the other hand, whether the authors overstated their case a bit for the potential of various historical periods to obscure their earlier historical predecessors. The wide range of natural and human formation processes that shape contemporary landscapes ranging from cut terraces to erosional features, scars associated with historical (and contemporary) excavations, and the local movement of soil to level fields contribute significantly to the complexity to surface assemblages. While I don’t doubt that the area of the ancient city of Thespiai is relatively stable (or at least well understood by the surveyors), I suspect that the relationship between the plow zone and subsurface material is too tricky to make arguments for later overburdens in anything but very well understood situations.

In fact, one thing that I’ve come to appreciate from our soon-to-be-published work at Koutsopetria is how much earlier material finds its way into later buildings. For example, the annex room from the Early Christian basilica at Koutsopetria clearly stands atop a clear earlier Roman horizon, but the walls, the floor and roof collapse, and the erosional overburden across the Late Roman building featured a large quantity of Hellenistic and even earlier Iron Age (and Cypro-Classical) material. We have our doubts whether the area around the Early Christian basilica saw activity in the Iron Age or even Hellenistic period, but the material present in the Late Roman building ensured that the Hellenistic and Classical periods, nevertheless, appeared in the excavation. This led us to suspect that some of the earlier period scatter across the Koutsopetria plain, might well reflect material that entered the plow zone from later buildings. Needless to say, the famous example of the Pyrgouthi tower where a predominantly Hellenistic scatter obscured a significant phase of 7th century re-occuptation springs to mind as well. 

The significance for understanding the the relationship between material from various periods also underscores the complexity of defining the extent of the site at any period as well as interpreting the presence of features like cemeteries or ritual activities in the landscape (much less estimating population size based on the size of a site!)  

Finally, one thing that I really appreciate from this work is the authors’ willingness to bring to the fore the various archaeologists responsible for producing the assemblages from the site. This extends from the charming story of the “Mad Dogs of Thespaia” to various roles played by two generations of ceramicists who read the material in the 1980s and 1990s and re-examined it in preparation for the book. Intensive survey has, at times, embraced a kind of impersonal style that places quantitative analyses and well presented maps before the work of the individual survey teams and ceramicists who produced the data. On the one hand, this makes sense as the rigorous collection of information from the field and the ceramic assemblages produces datasets designed for quantitative analysis. On the other hand, anyone who has worked on a survey knows that time in the Greek countryside, with sherds, and in the company of other archaeologists can shape the results of a project in ways that the tidy analysis often obscures.

As I work my way through the volume, I’ll likely blog more on this book over the coming weeks. It’s an important and expansive book, so stay tuned!

Survey Archaeology and Dogs

Since I’ve been home, I’ve been working my way through some recent scholarly on survey archaeology as we begin to analyze the data from the Western Argolid Regional Project. Hopefully I’ll have time to blog more at length about articles like, Marica Cassis, Owen Doonan, Hugh Elton, James Newhard, “Evaluating Archaeological Evidence for Demographics, Abandonment, and Recovery in Late Antique and Byzantine Anatolia,” Human Ecology 46 (2018): 381–398. Cassis et al. bring together the analysis of a range of survey projects in Anatolia to demonstrate a diverse array of changes in settlement across the region during the seventh and eighth centuries. The authors argue for regional variation but also connections to climate change, the occupation of marginal lands, and varying degrees of regional engagement in larger economic and political systems. 

I’ve also started to read carefully, John Bintliff, Emeri Farinetti, Božidar Slapšak, and Anthony Snodgrass, Boeotia Project, Volume II: The City of Thespiai: Survey at a Complex Urban Site. Cambridge 2017. While there is much to unpack in this volume, I genuinely appreciated the anecdote on p. 31. 

“One recollection, shared between the notebooks and our own vivid memories, is that of the ‘Hounds of Thespiai’. In those days, when dogs in rural Greece were almost never treated as pets, allowed in the home or kept on a leash (in contrast to the gilded pooches on parade in Athens’ Kolonaki Square), their main function in the countryside was to guard houses and sheep-folds. Apart from the violent barking which was the first form of custodianship, few ventured physical aggression unless one really intended to break into private property. To these rules of behaviour, comforting for the nervous student on field survey in Greece for the first time, the Mad Dogs of Thespies were a permanent exception. Once the field teams were in place in the lowlands of the ancient city each morning, only a few minutes of suspicious calm would elapse before a distant belling from the top of Thespies village hill above us would announce our detection by the Mad Dogs. They would immediately pour down the hill-side towards us at a great pace, then charge at the two teams. There never seemed to be an intention to stop short and make fierce gestures: rather, one got the repeated impression that large pieces of student were believed to be on offer to the under-fed mongrels. Only a Classical education offered daily security against the presumed threat: forming a circle, the field teams would present their steel-tipped sets of 2-m ranging poles to their would-be attackers. Wonderfully, after ten minutes of the ensuing stand-off, the Mad Dogs would slink off, but one could never be sure that an unexpected reprise might not occur later in the morning.”  

The Site

This summer I spent a good bit of time thinking about “the site” in survey archaeology. After four seasons of intensive pedestrian survey in the Western Argolid, we have started to analyze the data from our intensive pedestrian survey. We designed our project as a siteless survey and covered nearly all the small survey units (~2500 sq. m), high intensity sampling (10 m spacing), and no systematic change in method for higher density units. As a result, we produced a distribution map of artifacts across the landscape of the Western Argolid that shows gradations of artifact densities rather than dots on the map as one would see from a site-based survey project.

IMG 2765

Despite this approach to our survey area, we have come to realize that the vast majority off our pre-modern ceramics are concentrated in about 20 “clusters” across the landscape. This causes a bit of productive intellectual tension on our project. Were these clusters of artifacts “sites” produced by our siteless survey? Where these sites real? Were they the product of unrelated and overlapping period-specific phenomena or did they actually represent significant places for people, communities, and material in the landscape? As a siteless survey project we were caught in an intellectual grey area situated between the site as an apparent reality of our distribution of material, the site as a central discursive element of Mediterranean archaeology, and the site as a methodologically constituted (and produced) result from certain archaeological practices from the gridded collection of early survey projects to excavation. In practical terms, we began to speak easily of “off site scatters” even though this kind of language tended to imply a methodological distinction between “on site” (typically gridded) and “off site” (typically produced by transect walking) that did not apply to our field work.

IMG 2462

This got us thinking about sites on our project and whether the use of the term simply represented a convenient shorthand for our evident concentrations of material or whether we should spend some serious thought about understanding how to talk about these “sites” in the landscape. As I have noted in an earlier post, we spent some time tracing period specific clusters of artifacts across the landscape and applying buffers of various sizes to produce assemblages that go beyond groups of units with particular periods present and tries to capture the larger material landscape (including surface conditions and other variable that impact artifact recovery). With this kind of analysis, our sites or concentrations of artifacts in our survey area become overlapping clusters of material shaped by past activities in the landscape and surface conditions.

IMG 2674

The careful study of the overlapping and interlacing period clusters could demonstrate, if not exactly continuity, at least general patterns in the way in which various assemblages drew upon (1) common contemporary aspects of the landscape (i.e. that impact recovery rates), (2) persistent features in the landscape (i.e. heights, resources, et c.), and (3) historical relationships through time (i.e. continuity, reuse, memory, et c.). Moreover and perhaps more importantly, I think we could integrate siteless survey with an approach that respects the discursive significance of sites in Classical archaeology by showing how our method both problematizes sites and defines them in new but commensurate ways. For WARP, sites could become space where surface conditions, historical processes, and topography, geology, geography, and other natural and cultural features intersect to produce archaeological visible and meaningful places. 

IMG 2634

In this ways, sites become true indicators of the limits of our method, windows into the diachronic use of the landscape, and spaces for problematizing interpretation rather than the functional results of interpretive processes.

Resurvey

This week on the Western Argolid Regional Project, I’ve been running a few queries that compare the data from our original survey field walking and subsequent efforts to expand the assemblages present in these survey units. We termed these later efforts “resurvey” on WARP and thought they might be useful both to expand our generally small assemblages into something a bit more susceptible to functional analysis and to calibrate our recovery rates (as David Pettegrew and I did on the Pyla-Koutsopetria Archaeological Project). 

The challenge with this kind of analysis is how do we compare two different assemblages. In general, these two assemblages did not produce the same specific types of artifacts on the basis of our narrow typologies (i.e. the odds seem small that we’d find, for instance, two examples of the same African Red Slip form or even two examples of a Classical cooking pot), so that is not a very useful way to compare things. 

To open up the potential for meaningful overlap, I did a quick comparison of our resurvey units and our initial survey units to see if they produced the same period. This involves comparing the exact periods present in the finds from our first walking of the units to those found in either re-walking the units or in total collection circles with 2 m radius. Generally speaking there was some overlap between periods from each collection type. A few units producing over 50% of the material from the same periods, but most resurvey units produced material that had much lower overlaps (10%-20%). In this context, overlapping periods represent specific chronological period overlaps, such as Classical or Early Roman. This does not account for overlaps that are more broadly defined, such as when one assemblage produced Classical pottery and the other produced Classical-Hellenistic.  This is the next step in analysis is to see if resurvey produced chronological (as opposed to simply period) overlaps. This is a more complicated query and not ideal for analysis in the field.

We also compared the artifact densities per hectare from the resurvey units and the original survey units. As we demonstrated on PKAP, looking more carefully at the ground produced significantly higher densities. The highest density resurvey unit – which consisted total collection circles with 4 m diameters – produced densities that were over 100,000 per ha, for a unit that produced a density of only 1,700 artifact per ha through standard field walking practice. (Despite this massive difference in density, the unit produce a period overlap of over 50%!). Other units showed a similarly massive increase in densities with the resurvey units often producing nearly the same amount of pottery as the original survey units which covered much larger areas.

The differences between the two densities likely reflect three trends. First – and most obviously – a team of two or three scouring a 4 m diameter total collection circle for 10 minutes is like to find more pottery than a field walker, standing upright, and scanning 1 meter to either side even at a leisurely pace. Total collection circles were also much more likely to be placed in high density areas. After all, part of the goal of resurvey was to produce more a robust assemblage of material for chronological and functional analysis. Finally, total collection resurvey circles tended to be in areas of the unit with higher surface visibility. For each survey unit we recorded the average visibility for the entire unit. We did the same for the total collection resurvey units and they generally were 20%-40% higher visibility than the original survey units.

In the end, my analysis of these units is just starting. Considering the functional character of the original and resurvey assemblages, the chronological overlap of the two sets of material, and whether they produced new information about the