Architecture, Urbanism, and Radiocarbon Dating at Seyitömer Höyük, Turkey
Archaeology and chronology of Seyitömer Höyük, an Early Bronze Age - Middle Bronze Age settlement in western Anatolia.
This project investigates the timing and tempo of sociopolitical changes associated with Early Bronze Age urbanism, by analyzing architectural and radiometric data from Seyitömer Höyük. A series of intensive salvage excavations carried out by Dumlupinar University and the University of Buffalo from 2009-2014 revealed five cultural layers that span the Early Bronze Age through the Roman period (ca. 2500 B.C.E. through 363 C.E.) The EBA Phase V-B settlement was selected for a detailed analysis, because it is one of the best examples of an early urban center in the region, and it dates to the poorly understood EBIII period. Further underlining the importance of research here is the site's location within an active coal mine: as of 2017, all extant remains face imminent destruction as mining commences.
The first goal of the project was to determine the chronological extent of the Early Bronze Age (EB) III period at Seyitömer Höyük. This was addressed with a Bayesian analysis of twenty-three C14 dates from the archaeologically defined phases EBA V-C, V-B, V-A, and MBA IV-C, which spans the late third millennium and early second millennium B.C.E.. Ten radiocarbon samples were analyzed at the Scottish Universities Environmental Research Center in 2014, and a further thirteen samples were analyzed at the Oxford Radiocarbon Accelerator Unit in 2015. The results were incorporated into a sequential Bayesian model built in OxCal 4.2, which incorporates prior information about the stratigraphic relationships of each sample. In general, the dates from Phase V-B and V-C conform to expectations, while the dates from Phase V-A and Phase IV-C do not agree with the model; this issue is discussed further in the Methodological Notes section below.
The second objective of this project was to develop an understanding of EBA social organization, by investigating architecture, movement, and interaction in the built environment. The EBA Phase V-B settlement contains approximately 30 tightly packed buildings located inside a circuit wall, with a single entrance to the southeast. As a fully excavated settlement, this is one of the most complete datasets of EBA architecture in western Anatolia. In order to share this data, an architecture database was built. The database incorporates qualitative observations about construction technique, features, and deposits in each architectural space, as well as quantitative space syntax values (space syntax is a type of network analysis that predicts patterns of pedestrian movement and social interaction in the built environment; Hillier and Hanson 1984). The database is linked to photos and drawings of each architectural space, as well as "Enhanced Access Graphs," which summarizes information about the architecture, features, and accessibility of buildings into a single graphic format.
Bayesian Analysis of Radiocarbon Dates
The first objective of the project was addressed by collecting articulated bone and/or seed samples from rich, primary contexts at Seyitömer Höyük, ranging from EBA Phase V-C to MBA Phase IV-C. When selecting samples for analysis, provenience and sample material was taken into account, in order to ensure that there is a minimal gap between the dated event and the archaeological “target” event. For animal bone samples, I follow Bayliss’ criteria for high-precision dating: young individuals articulated in situ (which indicates the specimen was deposited in its final resting place before soft tissue began to decay) (Bayliss, in press). Seed samples were selected from undisturbed primary deposits. Charcoal and wood samples were not dated because of the well-known “old-wood” problem. All samples were sent to internationally accredited laboratories for analysis; either the Scottish Universities Environmental Research Center (SUERC) or the Oxford Radiocarbon Accelerator Unit (ORAU). All samples were submitted along with “prior information” including the site matrix, and structural phasing, so that samples can be related to specific periods of activity and events (see Bayliss 2009:132). The Bayesian statistical modeling was subsequently carried out using OxCal 4.2 software that is available through ORAU.
The chronological extent of the EBIII period was calculated on the basis of 23 C14 determinations, with the aid of a Bayesian model built in OxCal 4.2. This model was constructed as a sequential model, and incorporates prior information about the stratigraphic relationships of each sample.
At first, all 23 dates were fed into the model, to check for outliers. A total of eleven dates (21%) showing low individual agreement (less than 60%) were eventually discarded from the model. These outliers correspond to the archaeologically defined MBA IV-C phase (Sample ID 33499, 33500, 33501, 56568); EBA V-A phase (Sample ID 56562, 56569, and 56563); and EBA V-B Phase (Sample ID 56572 and 33505).
Three of the MBA IV-C dates are from the same deposit of carbonized wheat from inside a pithos in Room 6. These samples produced dates within the expected range of EBA V-C and EBA V-A, several hundred years earlier than the expected date range of MBA IV-C (2000-1900 B.C.E.). It may be possible to account for this discrepancy because the sample’s findspot in Room 6 is in an area near the center of the site (grid square H-12 and H-13) in which intensive building activities are attested all occupation phases. In Phase V-B, this area transformed from a public street space to the interior of a building. By Phase V-A it had been converted back to a street, and again by Phase IV-C, there was a building there. It is possible that the deposit found within the pithos in Room 6 was disturbed, or was not accurately assigned to Phase MBA IV-C by the excavators, given the many renovations that took place in this location.
One of the outliers from Phase V-B (33505) was found in a deposit of carbonized wheat from inside a pithos in Room 52x. The other three dates from this context all have a high agreement index, and it is likely that this sample failed due to contamination. Sample 56572 is an articulated right metatarsal from a young cow, and it produced a date much later than expected for reasons that are not clear.
The model allows the dating of the start of Phase EBA V-C to 2459 B.C.E. and the end of the phase to 2197 B.C.E. (99.7% probability). This phase overlaps slightly with Phase EBA V-B, which starts in 2316 B.C.E. and ends in 2063 B.C.E. (95.4% probability). The one successful date from Phase EBA V-A suggests a date range from 2119 B.C.E. - 1750 B.C.E. (99.7% probability).
Of these results, the Early Bronze Age Phase V-A results are generally consistent with the expected range of 2150 - 2000 B.C.E., especially when noted that the date range reported in this study is derived from just two C14 dates. The Early Bronze Age V-B results are close to the expected range of 2250-2150 B.C.E. The dates from EBA V-C overlap with Phase B, and are consistent with the expected range of 2350 - 2250 B.C.E.
The second objective of the project was addressed by analyzing architectural and artifactual data from EBA Phase V-B at Seyitömer Höyük, using an integrative approach to architecture and social organization, modified from Fisher (2009).
High-resolution AutoCAD architectural plans exist for each phase at Seyitömer Höyük, and these plans clearly show the location of doorways and entrances for most buildings. This information is essential in order to conduct space syntax analysis, a graph-based method of architectural analysis originally developed in the field of architecture by Bill Hiller and Julienne Hanson and outlined in the seminal The Social Logic of Space (Hillier and Hanson 1984). Using this methodology, I created access graphs. The process is relatively straightforward: architectural spaces in the settlement are illustrated graphically as nodes, linked by the paths of access drawn from an entrance (the carrier space or root) to all successively accessible spaces (Hillier and Hanson 1984:148-9, VanDyke 1999:466). Access graphs make the basic syntactic properties of symmetry and distributedness much more obvious than in a basic architectural plan, and they allow for simple visual comparisons of different access patterns (Hillier and Hanson 1984:149, Moore 1996:185). Access graphs concisely express patterns of movement and interaction in space.
Access graphs are analyzed using open source DepthMap software, which was developed for space syntax analysis at University College London’s Bartlett School of Architecture. Access graphs are generated along with five numerical syntactical measures for each space: integration, relative asymmetry, control, mean depth and total depth. Each of these measures provides insight into patterns of accessibility, social interaction, and movement in space. Data about built features will then be combined with access graphs to create an “Enhanced Access Graph,” which succinctly summarizes the accessibility, construction technique, and features of all architectural in a simple and easy to read format.
Potential Applications of Data
The dataset can be used in future chronometric dating projects that investigate the late third millennium and early second millennium B.C.E. in Anatolia and surrounding regions, as well as archaeological analyses of Early Bronze Age architecture and urbanism, public and private spaces, vernacular architecture, architecture and power; and political economy.
This project has received support from the National Science Foundation (Award number 1523389), the Mark Diamond Research Fund at the University at Buffalo, the Department of Anthropology at the University at Buffalo; the Institute for European and Mediterranean Archaeology at the University at Buffalo, and the Department of Archaeology at Dumlupinar University.
Current Disposition of the Physical Collection
The radiocarbon samples were sent to Oxford, UK and Glasgow, Scotland, and were completely destroyed during the process of analysis. All additional radiocarbon material recovered from the Seyitömer Höyük excavations is stored in Kütahya. Approximately half of the material is stored in a depo at the site, and the other half is stored in a depo maintained by the Archaeology Department at Dumlupinar University.
Related Data Tables for Download
|Property or Relation||Value(s)|
[Standard: Dublin Core Terms]
[Standard: Dublin Core Terms]
[Standard: Dublin Core Terms]
Open Context editors work with data contributors to annotate datasets to shared vocabularies, ontologies, and other standards using 'Linked Open Data' (LOD) methods.
The annotations presented above approximate some of the meaning in this contributed data record to concepts defined in shared standards. These annotations are provided to help make datasets easier to understand and use with other datasets.
Managing editor reviewed
To the extent to which copyright applies, this content
carries the above license. Follow the link to understand specific permissions
Required Attribution: Citation and reference of URIs (hyperlinks)