A NYTimes article on Friday considers a recent computer model of language divergence that places proto-Indo-European (PIE) speakers in Anatolia 8,000-9,500 years ago in small agricultural villages. The model thus stands in opposition to the argument for PIE arising on the Pontic Steppe sometime after 3500 BC advanced most systematically by David Anthony.
Indo-European Languages Originated in Anatolia, Biologists Say – NYTimes.com.
I leave the linguistic assumptions built into the model for others to critique. But there is an archaeological question which the model will find hard to address: what social dynamic does the early Anatolian model provide for the expansion of Indo-European? Small agricultural villages do not have built into their social dynamics an obvious mechanism for wide-scale expansion. One clear advantage of the steppe origins model is the clear socio-technical apparatus for rapid expansion provided by horse riding and chariots/wagons. The Neolithic Anatolian village provides no such mechanism.
This is not to argue that extension would be impossible–obviously farming and its technologies diffused widely. But the hoe and the horse are not equivalent technologies of dissemination. The hoe is a scale-narrowing object–one that yokes the land to human production by tying farmers to very local places (as opposed, for example, to the wider ranges of foragers). The horse, in contrast, is a scale expanding technology, one that encourages a wide-ranging sense of place. Missing from the Neolithic Anatolian model then is a sense of how language dispersal could have been so dramatically scaled up even as the lives of its putative speakers was scaling down.
On a parallel note (and based on very limited research): Renfrew seems to be the source for geneticists’ imaginings of linguistic histories. In their 1994 magnum opus, The History and Geography of Human Genes, Cavalli-Sforza et al. relied heavily on Renfrew’s work. However, in Cavalli-Sforza’s more recent popular book, he advocated a Gimbutas + Renfrew model. Again, he does not work through any archaeological analyses, but in a “comparison” of “results” finds the linguistic model that best matches his own genetic model. Also interesting are the underlying perceptions of archaeology by other “scientists,” archaeology does not seem to be a theoretical, analytic, or methodological approach, but prehistory.