Appendix. NAA Analysis of Lydian Ceramics from Sectors HoB and PC

Scientific analysis of ceramics at Sardis has a long history, dating back to the 1970s and covering the long chronological range from the Early Bronze Age through the Byzantine period.¹ Among the goals were to identify local Sardian ceramic industries and their products; for although kilns and pottery workshops are not common at Sardis, the consistency of fabrics and decorative techniques in many different periods, as well as molds and other finds, clearly suggests the presence of a strong local industry that supplied the majority of pottery used at the city through most of its history. Many of these analytic studies remain incomplete and unpublished, but two in particular have produced data relevant to the material from HoB and PC. The current Report is not intended to present the results and conclusions of these analyses, which are ongoing, but a brief mention of their history is in order.

In 2004–2005 two projects began independent studies of Lydian pottery at Sardis using neutron activation analysis: the Anatolian Iron Age Project (AIA), led by Lisa Kealhofer and Peter Grave; and Michael Kerschner and Hans Mommsen, studying the pottery of western Anatolia. Fortuitously, Kerschner, Grave, and Kealhofer overlapped at the site in 2005, and so were able to coordinate and cross-sample a number of sherds to allow comparison of their analyses. In 2004 and 2005, Kealhofer and Grave took 344 samples of local sediments and of ceramics, including sixteen in the current volume, for NAA analysis, and samples of ceramics for residue analysis. The results of their NAA analyses were published in 2013, and the raw data made available on Open Context.² Based on statistical analysis of their elemental compositions, Kealhofer and Grave divided the ceramics and sediments into three major groups, A, B, and C. One group, Macrogroup A, included six subgroups, A1.1–A1.6, and was identified as local production based on chemical similarities to two local sediment samples. From an archaeological point of view, this is a generally coherent set of artifacts, most of which we would have otherwise identified (with greater or lesser certainty) as local products. The group includes most common local shapes and decorative techniques such as lydions, marbled ware, and other typical Lydian features.

A second set of ceramics, Macrogroup B, included sediments from the foothills of the Tmolus range as well as a wide variety of ceramics. While the sediments are obviously local, some of the ceramics are, from an archaeological point of view, clearly imported, such as the East Greek Geometric krater HoB 351, Corinthian pottery (including an aryballos), and bird bowls. Other ceramics in this macrogroup, however, are very probably local products. A third set of ceramics, Macrogroup C, was matched by one local sediment sample and included a diversity of different ceramic wares, some of which are identified as imports to Sardis on archaeological grounds. Kealhofer and Grave also compared their groups to the published results of Mommsen’s analyses of pottery from various East Greek sites, to try to identify the sources of some of their nonlocal groups.

Kerschner sampled seventy-seven Lydian ceramic artifacts, including thirteen in the current volume; his analysis remains in progress.

In 2020, Hans Mommsen reanalyzed Kealhofer and Graves’ raw data, and arrived at a number of different conclusions. We are still in the process of analyzing these datasets and comparing the results of the different analyses, both chemical and statistical, and this is not the place for a discussion of this ongoing research. We hope, however, to publish this analysis shortly, and are extremely grateful that the work of the authors of this volume has so clarified the typology and chronology of Lydian pottery, as this will help advance the technical studies as much as the technical studies will in turn help us understand the production of Lydian pottery at Sardis and more broadly through western Anatolia.³

• 1See for instance Jones 1997; Scott and Kamilli 1981; Rotroff et al. 2018, pp. 134–35 for analyses begun in the 1970s but not completed then. Samples of pottery from many different periods were taken for analysis at the Missouri University Research Reactor (MURR); those have not been included in this volume.
• 2Kealhofer, Grave, and Marsh 2013; data is at Open Context.
• 3On regional production of Lydian pottery see especially Gürtekin-Demir 2002; Gürtekin-Demir 2003; Gürtekin-Demir 2007; Gürtekin-Demir 2014; and Gürtekin-Demir 2021.