Fijian cannibalism and mortuary ritual: bioarchaeological evidence from Vunda

Abstract

The human skeletal remains from the site of Vunda (AD 800–1600) in Fiji have been interpreted as evidence of cannibalism. The cannibalism hypothesis is tested by examining the modifications of the Vunda skeletal sample. The sample consists of human and non‐human bones from a midden context, as well as intact human burials extracted from the midden. Most modifications are more common in the medium mammal sample than in the midden human sample—burning (9% of medium mammal remains and 5% of human remains), cutmarks (4 and 1.5%), percussion pits (1 and 0%), and peeling (2 and 0%)—though these differences are not statistically significant. There are no significant differences between the human burials and the midden human sample in the rates of cutmarks, percussion pits, peeling, or in the pattern of element representation. Therefore the cannibalism hypothesis is not supported at Vunda. Most of the ‘midden’ human remains were probably part of the formal burials. The modifications and context of the Vunda sample are compared with the remains from the nearby, partially contemporaneous site of Navatu (where an inference of cannibalism is supported) to tentatively establish the characteristics of Fijian mortuary rituals versus cannibalism. Copyright © 2000 John Wiley & Sons, Ltd.

Identification of the first reported Lapita cremation in the Pacific Islands using archaeological, forensic and contemporary burning evidence

“…These include burning, cut marks, percussion pits, peeling, peri mortem fractures, anvil abrasions, an under-representation of vertebrae, similarities between other butchered fauna at the site, and dissimilarities of element representation compared with contemporary burials (Cá ceres et al, 2007;Cochrane et al, 2004;Degusta, 1999Degusta, , 2000Hurlbut, 2000;White, 1992). However, because of the difference in cooking methods and consumption preferences in the Pacific compared with other regions of the world the criteria of anvil abrasions and an under-representation of vertebrae may not be relevant for these remains (Pietrusewsky et al, 2007).…”

Cultural Modifications in an Adolescent Earth-Oven Interment from Fiji: Sorting out Mortuary Practice

“…Alternatively, the diagnostic criteria proposed by DeGusta (1999DeGusta ( , 2000 may not be applicable to this case since they were developed from assemblages significantly different from this one (e.g., MNI greater than one, habitation areas, midden context, etc.). DeGusta's diagnostic criteria do not easily accommodate the use of an earth-oven for cooking with steam rather than flame (lower frequency of burning), the use of bamboo for cutting (fewer and fainter cut marks), or the selection of meat over fat/marrow (lower rates of fragmentation, fewer cut marks).…”

“…However, several of these criteria (e.g., pot polish and missing vertebrae), are not applicable to bone assemblages in Fiji or other parts of the Pacific because of differing cooking methods (i.e., roasting in an earth-oven rather than boiling in a pot) and differing preferences (i.e., meat acquisition as opposed to marrow/fat acquisition). These differences prompted the use of a more globally useful finding: that of a dissimilarity between the bone modifications in the assemblage and bone modifications found in primary or "considerate" burials, and/or a similarity between bone modifications in the assemblage and bone modifications in contemporaneous faunal bone middens (DeGusta 2000).…”

“…DeGusta concluded that the human bone in the Navatu midden was evidence of cannibalism. In contrast, reanalysis of the Vuda material (AD 800 -1600) suggested the human midden remains were not the result of cannibalism since bone modifications seen in these remains were statistically indistinguishable from bone modifications in contemporaneous burials (DeGusta 2000). He further suggested that the modifications observed in the Vuda material could be the result of interment beneath houses.…”

“…However, early investigations lack the rigor required by modern analyses and so should be viewed with some caution. As Rieth (1998) and DeGusta (1999DeGusta ( , 2000 have shown, the application of rigorous, quantifiable, taphonomic dimensions and stringent criteria succeed in producing a much more accurate description of these complex midden assemblages as well as providing stronger evidence for cannibalism. Although the archaeological record is sparse, there is no apparent evidence for cannibalism in the early stages of Fijian prehistory, and the first strong evidence appears only after 50 BC on the island of Viti Levu, during a time of expanding population and settlement shifts.…”