Ancient Chemistry Reveals Hidden Workshops Behind Islamic Sicily’s Glazed Pottery Revolution

When the Aghlabid armies of Ifriqiya crossed from North Africa and seized Byzantine Sicily in 827 AD, they set in motion a transformation that would reshape the island’s economy, cities, and daily life for centuries. Among the most visible markers of that change was a quiet technological revolution in the kitchen and on the dinner table: glazed ceramics. Now, the first large-scale integrated archaeometric investigation of Islamic-period glazed wares from Sicily has traced exactly where these coveted tablewares were made, revealing a surprisingly complex map of production that stretched from the capital Palermo to workshops across the island, and clarifying long-standing confusion between Sicilian products and their North African counterparts.

The study, published in Archaeological and Anthropological Sciences by Viva Sacco, Patrick Sean Quinn, and Corisande Fenwick of University College London’s Institute of Archaeology, analysed 188 ceramic samples from five Sicilian sites: Palermo, Castello della Pietra, Piazza Armerina, Agrigento, and Syracuse. By combining three complementary techniques, macroscopic fabric analysis, thin-section petrography, and instrumental neutron activation analysis, the team tested and refined earlier classifications that had relied largely on visual inspection. The result is the most detailed compositional fingerprint yet of Sicily’s glazed ceramic industry between the late ninth and twelfth centuries.

Glaze technology arrived in Palermo in the late ninth century, almost certainly carried by mobile artisans or through technology transfer from nearby Ifriqiya, the region of modern Tunisia and eastern Algeria that had recently conquered the island. For nearly a hundred years, Palermo remained Sicily’s sole centre of glazed ware production, supplying a modest local market that only began to expand in the mid-tenth century. Then, by the late tenth century, new workshops emerged across the island, breaking Palermo’s monopoly and shifting the industry from a single-centre model to a polycentric one. This Sicilian trajectory contrasts sharply with Ifriqiya, where multiple production centres, led by Kairouan, operated from an early stage and glazed tableware was already widely distributed by the late ninth century.

The analytical challenge facing the researchers was considerable. Sicilian and Ifriqiyan glazed wares share such strong similarities in vessel forms and decoration that scholars have long lumped them together under the vague label ‘Siculo-Maghrebin’, obscuring internal developments and regional specificities. Compounding the problem, no kilns dating from the late ninth to the eleventh centuries have ever been excavated in Sicily. Production centres have instead been identified through scattered finds of wasters, kiln rods, and recurring petrographic groups, at Palermo, Mazara del Vallo, Agrigento, Syracuse, Paternò, and the rural settlement of Piazza Armerina.

The clearest and most coherent group identified in the new study corresponds to wares produced in Palermo. Petrographic Fabric 1, the so-called Ficarazzi fabric, accounts for 114 of the analysed samples and is defined by a non-calcareous matrix dominated by monocrystalline quartz and frequent micritic limestone. This fabric matches the famous ‘Ficarazzi clay’, a grey-blue alluvial deposit of Lower Pleistocene age located roughly one kilometre east of the medieval city, near the mouth of the Eleuterio River. Remarkably, this clay source had been exploited since at least the Punic period and remained in use into the early twentieth century. Geochemical analysis confirmed the picture: a compact compositional cluster of 89 samples, characterised by elevated concentrations of cobalt, caesium, manganese, and rubidium, corresponds tightly to the Palermitan material, demonstrating a highly consistent paste recipe and the exploitation of a restricted range of raw materials.

Beyond Palermo, the study characterised the products of several other workshops that began operating from the late tenth century. At Syracuse, Petrographic Fabric 2 contains abundant carbonate inclusions and volcanic rock fragments, features consistent with the volcanic geology of eastern Sicily, and was tempered with limestone and weathered mafic material. At Piazza Armerina, Fabric 4 is a biotite-rich calcareous sandy quartz fabric whose mineralogical variability matches the complex local sedimentary formations around the Villa del Casale, where kiln bars have been discovered. At Agrigento, Fabric 6, sampled from sherds recovered near the kilns and production waste of the Fornaci S. Lucia artisanal district, corresponds well with the local geology of calcareous tuffs, fossiliferous marine clays, and shell breccias.

Not every attribution proved straightforward. The fabrics tentatively assigned to Mazara del Vallo, represented by samples from the rural hilltop site of Castello della Pietra, displayed a bimodal quartz distribution and calcareous matrices that overlapped chemically and petrographically with North African products. These samples scattered across multiple geochemical clusters, underscoring the genuine difficulty of distinguishing some Sicilian and Ifriqiyan calcareous fabrics. The researchers suggest this may reflect technological convergence, shared production traditions, or the exploitation of geologically comparable clay sources on both sides of the Strait of Sicily, though the smaller sample sizes for these groups also limit the statistical robustness of the groupings.

The geochemical work itself demanded unusual care. Because the entire assemblage consisted of glazed ceramics, the team had to assess whether lead-rich glazes, containing 40 to 65 weight percent lead oxide, had contaminated the bulk compositions of the ceramic bodies. Scanning electron microscopy with energy-dispersive spectroscopy of 120 samples revealed that most specimens showed either no glaze-body interaction layer or only a very thin one. The ceramics had also undergone a double-firing process, in which the body was bisque-fired before glazing, fixing the body’s composition before the glaze was applied. Statistical tests showed that elements associated with glaze colourants, such as manganese and iron, did not influence the multivariate structure of the dataset, confirming that the compositional groups reflect the ceramic fabrics themselves rather than glaze contamination.

Perhaps the most evocative technological finding concerns a shared craft practice. Nearly all the Sicilian and North African samples displayed a light-coloured vitrified layer beneath the glaze, between 0.25 and 1.5 millimetres thick, formed when salt deliberately mixed into the clay migrated to the surface during drying and locally fluxed the clay minerals, producing a ‘salt-white’ surface treatment. This technique, employed by Sicilian and North African artisans since at least the Punic period and still used by potters in Tunisia today, along with the consistently high firing temperatures evident from the low optical activity of the clay matrices, points to a body of common technical knowledge circulating between production centres in Sicily and Ifriqiya, likely transmitted through the mobility of artisans and specialised networks of technological exchange.

The study also yielded a surprise: a single sample from Palermo, classified as Fabric 12, proved to be a clear outlier whose petrographic characteristics and splashed-ware typology point to an Egyptian origin, hinting at trade connections extending beyond the immediate Sicily-Ifriqiya axis. By establishing preliminary petrographic and geochemical reference groups for Sicilian glazed wares, the research provides a framework for future provenance studies and for identifying Sicilian exports at consumption sites across the medieval Mediterranean. The authors emphasise that expanding the comparative dataset, particularly for North African products and underrepresented Sicilian workshops such as Paternò, will be essential for refining these groups and reconstructing the full pattern of production, exchange, and technological interaction that accompanied Islamisation in the central Mediterranean.

Subject of Research: Archaeometric characterisation of Islamic-period glazed ware production centres in Sicily

Article Title: Tracing glazed ware production in Islamic Sicily: macroscopic, petrographic, and geochemical evidence

Article References: Sacco, V., Quinn, P. S., & Fenwick, C. (2026). Tracing glazed ware production in Islamic Sicily: macroscopic, petrographic, and geochemical evidence. Archaeological and Anthropological Sciences, 18(9), Article 195. https://doi.org/10.1007/s12520-026-02550-1

Image Credits: AI Generated

DOI: 10.1007/s12520-026-02550-1

Keywords: Islamic Sicily, glazed pottery, ceramic petrography, geochemistry, Palermo, Ifriqiya, medieval ceramics, archaeometry, Ficarazzi clay, thin-section petrography, INAA, cultural heritage