The identification of the provenance of ancient pottery is an important topic in archaeological research, [1] as the reliable provenance of an object can be used to reconstruct distribution patterns and trade networks. Moreover, the origin of ceramics allows us to trace cultural contacts between communities. Additionally, the detailed characterization of ceramic objects, including their typology and material properties, enables an understanding of their production technology, which is closely linked to the broader socio-economic history of the archaeological site. [2] The origin of some ceramic classes, such as decorated fine ware, can be easily determined through the stylistic analysis of their decoration. However, the bulk of any assemblage consists of undecorated pottery whose provenance cannot be determined by traditional stylistic examination, as the same shapes may have been produced over a long period and in several workshops. To address this issue, archaeologists have developed a method based on ‘ceramic fabric analysis’ that describes the main features of pottery, including the inclusions and color of the ceramic body. [3] These features (or characteristics) are typically described using a binocular microscope. Pottery with the same characteristics belongs to the same ceramic fabric and, therefore, is supposed to have the same origin.
FACEM (= Fabrics of Ancient Ceramics in the Mediterranean) adopts the method of fabric analysis for ceramics from the Mediterranean area. Many of the published fabrics have been defined in combination with archaeometric investigations (petrography, sometimes integrated with chemical or heavy mineral analysis). For the publication of these ceramic fabrics, an online database was chosen, as it can be quickly updated and expanded and complies with Open Access policies.
The definition and description of ceramic fabrics was developed by David Peacock in the 1970s and drew largely on the geological sciences. This was followed in the 1990s, first by Clive Orton, Paul Tyers, and Alan Vince, and then by Roberta Tomber and John Dore. [4] Since then, fabric descriptions have become well-established in pottery studies.
A fabric depends on the natural composition of the clay, how the potter handles the clay, and the temperature and atmosphere during firing. These factors determine the characteristics of the fired clay (=sherd), including its texture, as well as the sorting, size, and number of different inclusions and voids that, if not affected by post-depositional conditions, are specific to a certain production center. These features are visible on the fresh breaks of ceramic fragments under a binocular microscope. They can be used to analyze and compare ceramic material, identifying fragments with the same fabric, indicating the same provenance. To consistently and transparently define and document ceramic fabrics for further studies on ceramic production, consumption, and trade, FACEM follows a standardized procedure.
FACEM examines fresh breaks of ceramic fragments with a binocular microscope - a low-cost and easily and widely applicable approach for processing high numbers of samples. It adopts a standardized fabric description, following the method proposed by Peacock (1977) and Orton et al. (1993). In defining and describing fabrics, FACEM focuses only on compositional and petrographic features, excluding surface treatments. Its standardization is based on a predefined vocabulary, as well as on comparison charts for recording the form, frequency, sphericity, [5] and color [6] of inclusions and their sorting and voids.
A selected number of samples is also analysed archaeometrically, using mainly, but not exclusively, thin-section petrography. Ceramic petrographic analysis was pioneered by Anna Shepard. [7] The method for describing ceramic thin sections was developed by Ian Whitbread, [8] and later improved by Patrick Quinn. [9]
On FACEM, a detailed and standardized description is given for each fabric type. It generally distinguishes between the sherd’s texture and matrix, its inclusions and voids. The matrix defines the body of the sherd, which may contain grains that are smaller than 0.01 mm and are thus not clearly distinguishable or identifiable even under a microscope. Grains larger than 0.01 mm are either added by the potter as temper or, in most cases, naturally part of the raw material. Voids are caused by air bubbles in the clay, organic material that has disappeared during firing, or grains that have disappeared.
When describing a fabric, the percentages of voids and inclusions are estimated with the help of comparison charts (Courty et al. 1989). The maximum and minimum measurements (in mm) are quoted as an indication of the size range of voids and grains and their overall sorting. Voids are further described by their form.
The description for each type of inclusion includes its frequency, according to a five-step scale (singular, infrequent, frequent, very frequent, riddled with), the shape of the grain and its sphericity, and its minimum and maximum length. In determining the type of inclusion, however, the method of fabric description is limited since archaeologists, unlike trained geologists or ceramic scientists, might not be able to correctly denominate or distinguish the individual particles through optical microscopy. For instance, quartz and feldspar, two of the most common particles in ceramic fabrics, are frequently indistinguishable and thus often summarized under quartz.
Other clearly identifiable inclusions are mica, iron concretions, calcium carbonate, and microfossils. Mica occurs in white (muscovite, silver glittering) and dark (golden glittering, biotite) colors. Red and black iron oxide concretions, which frequently appear in clays, may exhibit a rusty, opaque appearance. If their identification or that of any other particle is not certain, they are described by color, e.g., as reddish or black inclusions. Calcium carbonate particles are usually discernible by their softness (easy to scratch or even destroy with a needle). Carbonatepseudomorphs are carbonate particles whose core has burnt out during the firing process, leaving usually well-rounded inclusions resembling white-rimmed voids. Microfossils are calcareous as well, being the shells of protozoa found frequently in marine clays. Their presence in ceramic fabrics may provide important evidence about the provenance of the clay. The presence of volcanic inclusions is also relevant for mapping clay sources.
Each fabric is labelled with a tripartite alphanumeric code, which can be used to bookmark or cite specific ceramic fabrics. It includes information about the production site or region (e.g., BNAP = Bay of Naples) and the pottery ware (e.g., Amphora = A, Ceramic Building Materials = CMB, Glazed Wares = G) to which the sample was assigned. Since different ceramic wares could have been produced with the same or similar raw materials and production techniques, a functional and interpretative differentiation is not always possible from the fabric alone. For instance, fabric codes with the letter C refer to both Coarse and Cooking Wares. If several fabric types of the same ware produced at a certain site have been recorded, they are differentiated by consecutive numbers (e.g., BNAP-A-1-BNAP-A-12).
Some fabrics of traditional wares or special shapes regarded as (common) imports at discovery sites, are also included in the database, even if their provenance has not yet been determined. This unknown provenance is preliminarily coded as “IG(NOTUM)”, sometimes with the (chronological) ware designation as further information. For instance, “IG-Pun-A-x” is a fabric of a Punic amphora of unknown provenance.
Images of ceramic sherds are captured with a Leica MZ6 stereomicroscope and a Leica MC190 HD camera. Each sample can be photographed at three to four standardized magnifications, considering that higher magnifications zoom into a smaller image of the sherd. The 8× magnification provides an overview of the entire sample and is comparable to an image seen through a magnifying glass. The 16× and 25× magnifications allow detailed examination of individual inclusions, while the 40× magnification is especially useful for analyzing small-grained fine wares. Given the varying sizes, each image (in JPG or PNG format) includes a scale.
The editors appreciate the further integration of ceramics from new production sites, production areas, and their fabrics into the FACEM database. Therefore, they invite interested colleagues to contact them at backoffice@facem.at (see also Impressum).
