[FIGURE 1 OMITTED]
We are not confident we know who was the first person to suggest that the 'eyes', 'noses', and 'faces' sometimes seen drawn on Lapita pottery sherds and whole vessels are those of human beings, i.e. that they are anthropomorphic in appearance and were intended to be regarded as such by their makers. More to the point, nobody to our knowledge has explained why this is so.
In some instances, the faces do have a human-like form, but not always, as Green (1973: 334, 1979a: 21-22, 1979b) pointed out years ago. Furthermore, it does not take much familiarity with Oceanic material culture to realize that even the most seemingly human-like masks, carvings, and other constructions made of wood, clay, stone, bone, wickerwork, and the like in the Pacific may not be intended as representations of human beings. Spirits, demons, and the like do not have to be portrayed as anatomically strange or non-human to be considered non-human, although the well-known spirit boards of the Papuan Gulf, so often identified by art dealers as 'ancestor boards', show how difficult it can be for foreign eyes to grasp local sensibilities (Welsch 2006: 6, 22). Nor is there much about the shape or appearance of most of the anatomical features drawn on Lapita vessels that is undeniably human-like (Figure 1).
Yet in now-classic papers, for instance, Matthew Spriggs (1990a, 1993) described Lapita iconography as anthropomorphic, and ever since then others (e.g. Best 2002; Chiu 2005, 2007; Kirch 1997: 132-140, 2000; Sand 2007) have apparently accepted this appraisal despite Spriggs' own disclaimer on one occasion (1993: 14, n. 1) that he was using this term as a matter of convenience. In his words, they 'may have had totally different meanings to the makers and users of the Lapita pots themselves'. Nonetheless, for example, Simon Best a decade later answered the question How much of the Lapita design system represents the human face? with a confident 'a significant amount' (Best 2002: 50).
Elsewhere we have offered several congruent lines of archaeological and ethnographic evidence (Terrell and Schechter 2007) leading to the alternative inference that what was probably being portrayed or referenced in at least some instances was something held to be important about sea turtles (Figure 2). After our report came out in 2007, Scarlett Chiu (2007) and Christophe Sand (2007) published additional information on the chronology, geography, and variability of Lapita iconography. We can now generalize further on our previous observations. Formerly we were unsure how common turtle symbolism is on decorated Lapita pottery. It is now possible to suggest that turtle iconography in one form or another may have been the major symbolic allusion, or 'message', being expressed in this fashion.
What difference does it make how human or non-human we take the eyes and faces drawn on Lapita pots to be? There is the obvious concern for historical accuracy. In addition, these potsherds and occasional whole vessels are often used to support ideas about the peopling of the Pacific (e.g. Sand 2007: 265). In our opinion, doing so may be pushing this elusive evidence too far.
We are not naive enough to think that evidence, archaeological or otherwise, can stand on its own. How anybody interprets things or events is always to some degree dependent on the contexts viewed as making things and events worthy of note. It is not unreasonable, however, to worry that too little can be made of things and events, and too much may be made of the contexts into which people may see them as fitting. For this reason, we will distinguish here between the meaning of Lapita motifs, on one hand, and their importance, on the other. For the people who made and used these pottery containers of various shapes and sizes, meaning and importance may have been one and the same. Not so for archaeologists today who must infer both from the surviving evidence at hand (Ten'ell 2001). …