In essence, Pargeter et al. argue that claims for the antiquity of modern San ‘cultures’ involve a misuse of analogical reasoning. In general, I agree. But, let me take issue with a few of the specifics they discuss, and argue instead that things may be more complicated than they seem.
It is perfectly true that I once claimed a greater diversity for San peoples than had previously been recognised (Barnard Reference Barnard1992). This is especially true for aspects of kinship and for settlement patterns. For other things, however, particularly religious beliefs and practices, it is less true (see Barnard Reference Barnard1988). From one end of southern Africa to the other, we do find diversity, although we also find great similarities, even in the vocabulary employed to discuss such things: words for ‘God’ and for ritual activities, for example. How old these are, we of course do not know, but the possibility exists that they are very ancient and even part of an Urkultur such as that which seems to exist both in High God concepts and in mythological systems. The latter may even exist across the continents, which means, therefore, through long periods of time (see Witzel Reference Witzel2012).
I have no idea how ancient Border Cave cultural practices may be. The authors cite Güldemann (Reference Güldemann2008), however, on the fact that the ‘San’ occupy three different language families. What they do not say is that he has characterised the Kalahari Basin as a whole as a Sprachbund, a collectivity of languages that have through time merged together, rather than having split from a common source as with Indo-European. This makes every bit of sense to me: language families can only be traced 7000 or 8000 years back, but this is not to say that there may not be a degree of continuity over a longer period. Languages diverge and converge over very long periods, and it is quite normal for hunter-gatherers to speak many languages. During my last visit to the Kalahari in 2011, I met a N!aqriaxe man who could speak languages in five different language families (the three ‘San’ ones and two others). Some of these have grammars that are among the most complex in the world. Naro (my primary fieldwork language) has 86 person-gender-number markers. /Xam has at least six verbal suffixes, 24 verbal prefixes and some 14 ways to make a plural.
How do we know the linguistic history of the subcontinent? What is to say that the situation 40000 or 50000 years ago was significantly different? In Language in prehistory (Barnard Reference Barnard2016), I discuss the probability that every Pleistocene hunter-gatherer spoke many languages, and that these languages existed for a purpose, namely the transmission of ideas through myth: myths spread far and wide and across language-family boundaries. Through genetic and linguistic evidence we know that the Naro probably spoke a Ju/’hoan-like language before they acquired a Khoe one (Pickrell et al. Reference Pickrell, Patterson, Barbieri, Berthold, Gerlach, Güldemann, Kure, Mpolaka, Nakagawa, Naumann, Lipson, Loh, Lachance, Mountain, Bustamante, Berger, Tishkoff, Henn, Stoneking, Reich and Pakendorf2012). This is also borne out by changes in Naro kinship structure, and of course the 15000 Naro alive today were not the same ‘people’ as then. Those who eventually became ‘the Naro’ were, a few hundred or thousand years ago, probably a very small, multilingual population, numbering perhaps only in the hundreds.
If we put all this together, who is to say exactly what the prehistory of Border Cave may have been. At times, I prefer reasoned speculation over either direct historical analogy or relational analogy. I leave it to archaeologists to explore this idea further.
