The 2002 KRAP Archaeological Field School Overview, by William Parkinson

With generous funding from the National Science Foundation (NSF-0105851 and NSF-0139122), the Wenner-Gren Foundation for Anthropological Research (International Collaborative Research Grant), the Hungarian Academy of Sciences, and Florida State University (First Year Assistant Professor Award), we were able to continue the undergraduate field school and to continue our geophysical and gemorphological research in the Körös River Valley.

The 2002 field school was carried out from June through August and included undergraduate students from seven different universities in the United States. In addition, several graduate students ­ from Florida State University, Ohio State University, the University of Michigan, Arizona State University, and the University of Wisconsin-Madison ­ also worked on the project in various capacities. As usual, several Hungarian students from the József Attila University in Szeged worked on the project (Figure 1). We also were fortunate to receive visits from several specialists (Figure 2) including: Dr. Glen Doran (Florida State University), Dr. Alice Choyke (Aquincum Museum), Dr. Michael Galaty (Millsaps University), and Dr. Robert Schon (Stanford University). A complete list of the project participants is available on this website.

As always, we want to thank the wonderful people of Vésztő, and especially the mayor, Mr. János Kaszai. Without their continued support of our endeavors, our research would be impossible.

Geophysical Research

During the season the project geomorphologist, Dr. Tod Frolking (Denison University), with the help of undergraduate students, supervised the collection of several soil cores around several mid-Holocene sites in the Vésztő area in an effort to develop a model that will help us understand the complex paleo-hydrological processes that affected the region throughout the Holocene (Figure 3) Dr. Apostolos Sarris (Foundation for Research and Technology, Hellas, Greece) conducted magnetometric remote sensing using a flux gradiometer at Vésztő-Bikeri in an attempt to help us determine the layout of the settlement (Figure 4). Magnetometry has been conducted at other prehistoric sites on the Great Hungarian Plain (e.g., Raczky et al. 1997), and it has proven to be a very effective, and efficient, method of identifying burnt clay features as well as features that are intrusive into the sterile subsoil.

The results of Dr. Sarris' magnetometric survey helped us further define the layout of structures and other features across the settlement. Most surprising was the identification of two parallel curvilinear linear features that encircled the settlement on the north, east, and southeast (Figure 5). In an effort to 'ground-truth' the anomalies, we excavated two long (10 m x 1 m and 15 m x 1 m) blocks (Blocks 5 and 6) across the anomalies on the north and northeastern edges of the site. Upon excavation, the anomalies each corresponded to trenches that were intrusive into the sterile subsoil around the site (Figure 6). In both blocks, the two trenches that corresponded to the magnetic anomalies each surrounded another, narrower trench that ran parallel with them. Unfortunately we were unable to delineate the precise cross-sections of the trenches above the sterile subsoil in the profiles because they are not visible in the plowzone. The shapes of the features only became clearly defined when we were able to differentiate the darker, mixed, fill deposit from the yellowish, sandy, sterile subsoil.

Based both upon the regularity and consistency in form of the features in both excavation blocks, as well as their continuity as documented by the magnetometric data, we are convinced that the features are concentric circular trenches that once encircled the settlement. The magnetometric anomalies that are associated with the trenches do not extend all the way around the western edge of the settlement, but this most likely is because the features were destroyed by deep plowing, a process that seems to have been particularly destructive at the edges of the small mound where the site is locatedIt also is possible that the western portion of the trenches were eroded when the streambed where the canal is flooded in prehistory (Apostolos Sarris, personal communication, July 2002).

Another striking characteristic that emerged from the magnetometric survey was the dense concentration of anomalies concentrated at the center of the site, and their correlation with other features we excavated in the last three years at the site (Figure 5). Most notable are several rectangular anomalies that appear to be structures in a dense cluster at the center of the area encircled ­ or partially encircled ­ by the outer ditches. The area immediately surrounding the structures at the center of the settlement contains several oval features we suspect will turn out to be pits and other features such as kilns and ovens. Between the structures (and other features) at the center of the settlement and the ditches surrounding the settlement, there are relatively fewer magnetic anomalies, suggesting that this area between the ditches and the structures may have been used for other activities that did not result in the production of discrete magnetic anomalies, such as animal keeping.

The spatial distribution of the structures identified by the magnetometer augment several of the patterns we had begun to discern during the first three years of excavation at the site. In some places the rectangular features identified by the magnetometer line up in straight rows with excavated structures. This is the case in Block 2, where the northern edge of a rectangular magnetic anomaly (ca. 6 x 5 m) to the west of the block lines up precisely with the northern edge of two structures within the block, forming a line of structures lined up precisely on an east-west orientation. To the north and south of that line of structures, the distribution of other structures are not lined up directly on cardinal directions, but seem to follow the contours of the hill. In no cases do the structures cross-cut each other, suggesting that they may have been in use roughly at the same time.

This provides the first glance at the organization of a Tiszapolgár settlement on the Great Hungarian Plain. The small number and regular distribution of structures across the site reinforces the pattern indicated by other chronological indicators that habitation at the site was relatively short-lived, probably less than a few hundred years. At the same time, the structured layout of the settlement, with its concentric ditches and concentration of structures and other features, suggests that activities were organized into spatially discrete areas across the site. This s patially discrete distribution of use areas is indicated not only in the layout of the settlement as a whole, but also in the distribution of activity areas within and between the structures. Our excavations continued near the center of the site as well, where we expanded our excavations to expose more of the features that initially had been identified in Blocks 2 and 3 during the 2001 season, where our excavations have exposed the remains of three structures ­ two in Block 2 and one in Block 3. Although the structures vary dramatically in their degree of preservation, they each exhibit several similar characteristics that lead us to believe that they all may have been used for domestic purposes.

Block 2 ­ Feature 4

We initially identified the remains of a burnt structure in Block 2 during our test excavations in 2000, when a 2 x 2 m test unit came down precisely on the northeastern corner of the structure. During the 2001-2002 seasons, we were able to expose more of the structure by opening a larger (14 x 10 m) unit around the feature (Figure 7).
During the 2002 season we identified wall trenches associated with the northern and western edges of Feature 4 (Figure 8). The trenches extend approximately 85 cm below the present surface and approximately 35-50 below the edge of the daub/floor interface. They are U-shaped in cross section and contain postholes that sometimes extend into the base of the trench. The northern and western wall trenches associated with the feature meet at the northwestern corner, where they articulate with a large, bell-shaped pit (Figure 9).

Block 2 ­ Feature 14

At the end of the 2001 season we began exposing the area to the east of Feature 4, where we identified a dense daub and ceramic concentration at the easternmost edge of Block 2. We opened a 4 x 4 m extension off the northeastern corner of the block in 2002, in an attempt to define the extent of the feature (see Figure 10). It remains unclear whether this feature can be associated directly with Feature 4. It may be an extension of that structure, or it may be another structure altogether. There was a definite 1-1.5 m break in the daub concentration between the two features. We exposed a large portion of the feature during the 2002 excavation, but unfortunately the feature continued into both the eastern and southern profiles.

The quantity and relative frequencies of artifacts, bone, and daub associated with Feature 14 were very similar to those associated with Feature 4. In both cases, the floor level was characterized by several flat-laying ceramics overlaying a thin lens of slightly more compact, darker clayey silt. This thin layer is sandwiched between a thick layer of daub and a general cultural layer, or 'sheet midden' that occurs across the entire settlement.

A small group of cups were found in the southwestern corner of Feature 14, just to the northeast of the intrusive Hungarian Conquest Period burial. Roughly four meters to the east, within the same feature, several fragments of very large storage vessels were found.

The relationship between Features 14 and 4 partially were obscured by the intrusive burial, which was sunk directly into Feature 4, and which also obscured the southern portion of Feature 14. We hope our future excavations will help us clarify the relationship and extent of these two features.

Block 3 ­ Feature 5

During the test excavations in 2000 we identified another burnt daub concentration underlain by a very dense ceramic concentration in Block 3. We interpreted this as the remains of another structure, and in 2001 opened up a larger (6 x 6 m) excavation unit to try to identify the extent of the structure (Figure 11). In 2002 we extended that block further to the south to clarify the stratigraphic and spatial relationship between the structures in Block 3 and Block 2.

In contrast to Features 4 and 14 in Block 2, which exhibit a relatively clear vertical stratigraphy of wall rubble underlain by floor deposit, the structure in Block 3 is less well preserved with almost no identifiable floor deposit. By contrast, the ceramics, daub, bone, and thin compact floor layer all are much more jumbled, occurring at several different levels throughout Block 3. Nevertheless, it does exhibit similar general patterns in the relative frequencies of daub, bone, and ceramics as the structures in Block 2.

A composite drawing of the 'floor layer' ­ the level immediately below the densest concentration of daub ­ indicates that the ceramics and bone are concentrated at the southern end of the block (Figure 11). Curiously, this corresponds poorly to the distribution of daub throughout the unit, suggesting that the structure here may have been more heavily impacted by plowing and other post-depositional processes than Features 4 and 14, where the distribution of daub corresponds very well with the distribution of the floor deposits.

Despite our inability to delineate the exact layout of the structure, most of the loom weights and spindle-whorls collected so far derived from Block 3. One of these is an elongated ceramic fragment pierced all the way through, like a large loomweight, with small 'feet' (SF 91). While this block produced the majority of the evidence for textile manufacture on the site, only one bone point ­ SF 83 ­ was found in the entire block. Interestingly, this was the only bone point that was not burned and calcined, perhaps suggesting the poor preservation of the structure in Block 3 may be due to the fact that the structure was not burned as quickly or at as high a temperature as those in Block 2. While post-depositional processes, such as plowing, and other site-formation processes, such as the degree of burning, may have affected the differential preservation between the structures in Blocks 2 and 3, it also is possible these differences can be attributed to relative chronology and differential space-use within the site. More specifically, the structure in Block 3 may be slightly earlier than the structures in Block 2. In this case, the structure in Block 3 may have been used as a house, then abandoned and used as a dump for a short period before it was destroyed. But the relative frequencies of daub, ceramics, and bone all seem to suggest a pattern more similar to that of the structures in Block 2 than to the midden area. In addition, the relative fragmentation indices (weight/number) of ceramics are nearly identical between the two blocks, whereas the fragmentation index for bone is slightly lower in Block 3, suggesting that the bones there were slightly more fragmented than they were in Block 2. This, again, may be because the structure in Block 3 was not burned as quickly or as at high a temperature as the structure in Block 2. In general, a higher of the burned ceramics seemed to come from Block 2, but we have not yet had a chance to study systematically all 39,000 sherds (472 kilos) from the site.

In contrast to our perception that Early Copper Age settlements were much more ephemeral than those of the Late Neolithic, our excavations at Vésztő-Bikeri have demonstrated that despite its small size, the site retains several formalized structural features normally associated with the Late Neolithic. It represents a very high degree of labor investment in the construction of the buildings and the ditches that surround the settlement. Furthermore, the organization of the structures and the ditches within the settlement suggest that they most likely were built at roughly the same time. The distribution of what appear to be discrete activity areas across the settlement suggest that the inhabitants of the entire site operated within a single, well-integrated, social unit. This social unit itself may have been comprised of several nuclear family units that functioned together to form a household unit that in several ways paralleled its Late Neolithic precursor ­ the household cluster. But in contrast to the Late Neolithic system of pooling, which seems to have occurred within discrete sub-divisions within the settlement, during the Early Copper Age pooling seems to have occurred either across the entire settlement, which itself functioned as a single household unit. This change ­ in the scalar relationship between the primary units of production and the primary units of pooling in society ­ have important implications regarding the relative ability of individual members of society to accumulate wealth independent of the larger social unit.

While the results presented here are only preliminary, we hope our continued research in the region will allow us to gather more information to model the various changes that occurred during the transition from the Neolithic to the Copper Age. Over the next few years, we hope to continue the undergraduate field school and to continue to expose larger surface areas of the site. By continuing to add different multidisciplinary components to the project, including paleoenvironmental reconstruction and geophysical survey, we gradually will be able to build a model of Early Copper Age settlement organization that we can place within the larger regional context of our study area and the Great Hungarian Plain. Only then will we be in a position to begin to understand the various social changes that occurred here, nearly 6,500 years ago.