The primary mode of faunal exploitation by Neandertals and early modern humans remains
a debated topic. Binford (1981, 1984, 1985, 1988) has argued for an obligate scavenging
mode, Stiner (1991a, 1991b, 1991c, 1993, 1994) for a more opportunistic scavenging
mode, while other researchers (Chase, 1986, 1988, 1989; Klein, 1989, 1994, 1995; Klein
& Cruz-Uribe, 1996) deny the importance of scavenging as a faunal exploitation tactic.
The scavenging interpretations rely primarily on several patterns in the faunal remains:
the presence of a skeletal element pattern dominated by heads or head and foot parts,
the presence of carnivore tooth marks on bone fragments, and infrequent cut marks
that typically are not located on shaft regions of long bones or on fleshy bones.
Five sites have been used to argue for scavenging: Klasies River Mouth, Combe Grenal,
Grotta Guattari, Grotta dei Moscerini, and Grotte Vaufrey. The former four of the
five sites are biased samples in that long bone shafts and other difficult to identify
fragments were discarded at excavation. The analysis of Grotte Vaufrey included only
those shafts identifiable to species or genus, thus excluding the vast majority of
shaft specimens. This bias systematically shapes the skeletal element and surface
modification patterning in ways that make the assemblages appear to fit a model of
scavenging, when in fact the main determinant of the pattern is the bias in the flawed
samples. This problem is illustrated with two unbiased faunal assemblages (Kobeh Cave
and Die Kelders Layer 10). Skeletal element abundance is calculated in a way that
mimics the bias in the sites listed above by excluding the shafts. Using this procedure,
both Kobeh and Die Kelders have a head and foot skeletal element pattern and thus
appear scavenged. Both assemblages are then analyzed in their entirety and a new pattern,
consistent with hunting, is revealed. Taphonomic data on bone survival and destruction
provide an explanation for this result. Excluding shaft fragments from the analysis
also biases the surface modification patterning in such a way as to produce a pattern
more consistent with scavenging. The conclusion is that there is no reliable evidence
for scavenging by Neandertals or early modern humans.