Fabric-Based Approaches and Ceramic Petrology

Principal Investigator:
Evangelia Kiriatzi (Fitch Laboratory, BSA)

 

A sophisticated methodology, based on integrated fabric analysis, has been adopted for the study of KIP survey pottery. The aim has been for surface pottery to be treated not only as index of date, function and site size and formation, but also as evidence of past social and technological dynamics. Such an approach justifies the systematic collection of large samples of material, not just a selection of feature sherds, and further enhances our understanding of various aspects of landscape histories. In the Aegean, Jennifer Moody pioneered the use of fabric as a diagnostic for surface material (Sphakia Survey - Fabric Research), an approach also taken up by others mainly in Crete. In KIP, the study of survey pottery draws upon their work but goes further to combine this approach with the systematic application of ceramic petrology and pottery raw material prospection. In doing so, it provides a powerful tool, not only for dating, but also for understanding patterns of pottery production, supply and consumption within specific natural and cultural landscapes through time. Emphasis has been placed initially on the coarse wares, since coarse fabrics can be more safely identified and classified macroscopically than fine ones. Up to now this approach has been successfully applied to the study of prehistoric material from Kythera. The analytical method employed has been petrographic analysis with thin sections since this method fits very well the analysis of coarse wares and can be very efficiently applied to the study of technological issues, apart from compositional ones. Moreover, it can be easily and meaningfully combined with the macroscopic examination of pottery since the analytical results can be directly related to macroscopic observations. The research has evolved at three different levels, of which the first two lay the groundwork for the third.

Level 1: Creating a fabric typology for Bronze Age Kythera
The first level, prior to the survey, concerned the study and subsequent analysis of pottery from a large, excavated and at least partly stratified, control site on the island, namely Kastri. Habitation on the site spanned from the Final Neolithic to the Mycenaean period, then to Classical and Late Roman, with slight Middle Byzantine activity. Well-dated and stylistically studied pottery from the Kastri phases provided both an insight into diagnostic features liable to survive prolonged exposure to surface erosion and, of more interest in the present context, allowed the establishment of a period-specific fabric series, initially at a macroscopic level. A similar emphasis has also been placed on understanding patterns and ratios of co-existing fabrics within given chronological horizons. Up to now, the analysis has concentrated mainly on the prehistoric periods. Five main fabric types of coarse wares have been identified at Kastri:

1. Chert fabric
2. Orange Micaceous fabric
3. Sand-tempered fabric
4. Mudstone-tempered fabric
5. Red Micaceous fabric

The use of the above fabrics at Kastri is associated with specific periods (e.g. the coarse ware chronology table in the Prehistoric section) and specific types of pottery.

Level 2: Raw material prospection and replication of ancient fabrics
Apart from the definition and characterisation of period-specific fabric types, ceramic petrology also addressed the technological and provenance study of these fabrics through intensive prospection and sampling of potential local resources for pottery production throughout the island. A large number of samples were collected and analysed, including clays, soils, sands and rocks (map above left). Subsequent laboratory experimentation with the collected samples and their comparative analysis led to the replication of the main ancient fabrics. Moreover, ethnographic research has been carried out on modern concepts and uses of ceramic resources on the island by local potters and inhabitants, in order to gain an insight into their concepts of the island's landscape and the raw materials and techniques available to them in the social and historical context of the early to mid 20th century. This combined approach has enhanced significantly the understanding of the technological choices made by the ancient potters within the island's landscape. The results can be summarised in the following:

The Orange and Red Micaceous fabrics are closely associated with the red-firing, mica schist clays encountered in the northern part of the island (map above left), representing, though, different technological choices concerning the selection and processing of raw materials within the same geological area.

The Sand-tempered and Mudstone-tempered fabrics reflect similar clay paste preparation technologies, having a clay base similar to the Neogene clays of the central and southern part of the island (map above left). In both the latter fabrics, temper has been added to the fine calcareous clay base. Experimentation with locally available raw materials and comparative analysis have indicated that sources of appropriate tempering materials are located within the broader Kastri-Palaiokastro area and assisted in understanding differential technological choices concerning the above materials. All the above fabrics seem to have been produced on the island. Nevertheless, some of the techniques used in the preparation of the clay paste (i.e. sand-tempering) find close parallels in certain areas of Crete. Furthermore a small number of imports has been identified across the BA phases and have been safely attributed to specific production locations (e.g. certain areas in Crete, below right, and the Cyclades).

Level 3: The study and analysis of the survey pottery
The results of the above work have had significant implications for the study of the survey pottery enabling us to start reconstructing quite dynamic 'ceramic landscapes' for several periods of the Bronze Age. First, they provide a more reliable chronological framework for understanding settlement distributions. The fabric typology, as defined on the basis of the Kastri pottery, combined with certain types of morphological features, has proved to be a very efficient tool for dating the survey pottery even in the case of otherwise undiagnostic body sherds. Looking, for example, at the large number of Second Palace period farmsteads across the Kytheran landscape, the study of the frequency and combination of fabrics, in each case, has made possible the demonstration that not all of them had precisely the same time span. Such resolution forms the basis for Andrew Bevan's micro-analysis of rural settlement patterns (Second Palace farmsteads). Similarly, the millennium-long EBA has been broken down into discrete and more analytically useful sub-periods. Second, beyond dating, this approach provides many insights into aspects of pottery production and supply across the island through time. More specifically, on the basis of fabric and technological study and analysis of the survey pottery, it has been possible to explore ancient potters' perceptions of the island's landscape, their material selection strategies, their potting traditions (in terms of both clay recipes and vessel forming techniques), the mechanisms of distribution within the island and patterns of consumption, thereby taking the study of pottery production and circulation well beyond the simple dichotomy between the local and the imported vessels. Shifts in these patterns through time are evident, and shed light on crucial questions concerning producers' and consumers' cultural identities, changing economic practices, the issue of craft specialisation, and the definition of regions within and beyond the island.