Archaeological Investigations of Early Glass Production at Igbo-Olokun, Ile-Ife (Nigeria)

The sacred grove at Igbo Olokun, located in Ile-Ife, Southwestern Nigeria, has long been known for the massive quantities of glass beads present on the surface and in subsurface deposits. Overshadowed by the bravura terracotta and copper alloy sculptures of human heads that characterize the so-called Classic period (12th-15th century A.D.) of Ife’s history, the glass materials at Igbo Olokun have received relatively cursory attention. Few of these materials have been described in any detail and their subsurface contexts are largely undocumented. Between 2010 and 2012, the archaeological excavations I conducted at Igbo-Olokun recovered over 12,000 glass beads as well as hundreds of glass-encrusted crucible fragments, other glass production debris, and pottery. This dissertation describes the deposits and the recovered materials in significant detail, establishing a basic framework for future comparative and analytic research at the site. Using chemical and physical analyses of the glass beads and glass production debris, the competing hypotheses of local primary glass production or re-melting of imported glass to create beads are explored in detail. Optical microscopic examination of the glass-encrusted crucibles in cross-section provides evidence for use in primary glass production rather then secondary re-melting of glass. Studies of other production debris including glass droplets, wasters, and cullet suggest the different stages in glass bead production from initial drawing of glass canes to heat treatment of snapped bead ends. The techniques used were highly sophisticated; producing mainly seed beads less than four millimeters in diameter. The thread or wire used to string the beads must have been exceptionally fine, but there is no evidence of the material used. The large quantity of material recovered from a relatively small area suggests glass production on an industrial scale. Compositional analysis of samples of the glass beads, crucibles, and other production materials using LA-ICP-MS, SEM/EDS, and SEM confirms the prevalence of glass that is very high in alumina content. High Lime High Alumina (HLHA) glass has previously been identified at Igbo-Olokun on samples of uncertain provenience. The recently excavated beads confirm that HLHA glass is very common, but Low Lime High Alumina (LLHA) glass is also present. These results expand on the work of James Lankton, Akin Ige, and Thilo Rehren who first proposed in 2006 that high alumina glass represents a glassmaking tradition unique to West Africa, and possibly unique to southern Nigeria. The compositional analysis provides additional evidence for local sourcing of the raw materials for this glass. The dissertation contextualizes this tradition with reference to the history and geography of early glass production in the Old World. The chronology of glass production at Igbo Olokun is not incontrovertibly established, since much of the material may have been redeposited and may therefor represent materials from different time periods. However, analysis of the pottery recovered from the excavations shows characteristics consistent with Classic period pottery from nearby sites in Ile-Ife that have been radiocarbon dated to the twelfth to fifteenth centuries CE. Several radiocarbon dates from the 2010-2012 excavations also fall within this time range, but there are also two much later dates, indicative of mixed deposits. Results of thermoluminescence (TL) and optically stimulated luminescence (OSL) dating of crucible samples, undertaken to clarify the chronology of glass production, have instead confused it further. The dates cluster in the early second millennium and the first millennium BCE, with some weaker evidence for the second millennium CE. The dissertation evaluates the various factors that might be at play, and identifies the additional data that will be necessary to support one or more possible scenarios for the dates of glass production at Igbo Olokun. The results of the bead classification and compositional studies have expanded the limited comparative database for distribution studies of various glass recipes and bead types in West Africa. The presence of HLHA beads in other West African archaeological sites after the ninth century CE suggests interaction between Ile-Ife and sites such as Igbo-Ukwu in southern Nigeria, Gao and Essouk in Mali, and Kissi in Burkina Faso. The dissertation concludes that Igbo-Olokun was a regional primary glass production center that specialized in large-scale production of glass beads, which then enjoyed widespread distribution in West Africa, particularly in the first half of the second millennium CE