The sherds of conquistadors: a petrological study of ceramics from Graciosa Bay and Pamua, Solomon Islands

ABSTRACT We present evidence linking vessel forms with ceramic wares resulting from the petrological analysis of 33 sherds from two sixteenth century Spanish colonial sites in the Solomon Islands. Our results expand the range of fabric types previously published, and comparative literature analyses support earlier studies suggesting probable ceramic origins in the Americas and Spain. Keywords: petrology, ceramics, Spanish, Solomon Islands. INTRODUCTION In the 1970s, archaeological investigations at Graciosa Bay (Nendo) and Pamua (Makira) (Figure 1) brought to light strong material evidence of a Spanish presence in the Solomon Islands (Allen 1976; Allen & Green 1972; Kaschko 1979). Both sites were linked to the second expedition of Alvaro de Mendana y Neyra (c.1595-1596) and yielded large ceramic assemblages. Initial ceramic studies showed that the sherds were wheel-made and of the same tradition, with glazed sherds pointing to a non-Oceanic origin. A typology developed by Allen and Green (Allen & Green 1972; Green 1973) showed that many of the wares and vessel forms were consistent with sixteenth-century colonial Spanish assemblages. Further supporting the Mendana link, a petrological study indicated not only that the sherds were non-Oceanic, but that they probably originated variously in South America and Spain (Dickinson & Green 1973). In this report, we expand the previous petrological study as part of current re-analysis of the assemblages, creating a larger comparative reference set including previously unreported fabric variants. This study then addresses a question of fabric type and vessel form concerning flat-bottomed bases raised during Allen and Green's typological analysis (Allen 1976; Green 1973). Moreover, we present additional evidence confirming New and Old World origins for various wares on the basis of comparative literature. BACKGROUND Allen and Green's flat-bottomed bases Several flat-bottomed bases were recovered during excavations at Pamua in the 1970s (Figure 2a). Green (1973: 24) was originally inclined to classify the unglazed sherds as Fine Plain (here Fineware, Figure 2d,e), but the possibility remained that they were Red Earthenware (Allen 1976: 27). These flat bases typically have fewer visible inclusions than Red Earthenware and all except one have a grey core, with one base green-glazed on the interior. The later was classified as Green-glazed Ware (Figure 2f, g). Red Earthenware sherds from both sites are associated with the olive jar (botija) form, with rim profiles associated with a late sixteenth century date (Avery 1997: 108, 120, Marken 1994: 52-62) and vessel form conforming to Goggin's Type B Middle Style olive jar (Goggin 1960: 28) (Figure 2b,c). No rounded bases of this form were retrieved from Pamua, though some were found at Graciosa Bay. If the Pamua bases proved to be Red Earthenware, then they would represent examples of flat-bottomed olive jars. Since the 1970s, colonial Spanish ceramic studies have noted occurrences of flat-bottomed olive jar forms. The basal fragments from Pamua are too small to provide evidence of a concave base defined as Form III by James (1988: 54), nor those described by Mirabal (2006:51). Marken (1994: 80-9) reported flat-bottomed olive jars from early seventeenth century shipwrecks, but the Atocha and the Santa Ana Maria examples have larger base diameters (>200 mm) than the sherds from Pamua (100-140 mm) and no rims matching those from the reconstructed Atocha ceramic vessel have been found. Flat bases were also recorded by South et al. (1988: 271-83). Petrological analysis was undertaken to confirm which fabric type the flat bases from Pamua represented and to better define the associated ware form. Ceramic origins and previous analyses Early twentieth century speculation on the provenance of scattered pottery finds at Pamua favoured a Melanesian origin (Paravicini 1931: 123, 132, White 2002). …