"We think some of these words are as ancient as 40,000 years old. The sound used to make those words would have been used by all speakers of the Indo-European languages throughout history," Professor Pagel said.Oh goodness.
Update, 9:30 am, by Joe: Pagel is a serious scholar, and he works with people who are trying hard to understand comparative linguistics and apply current computational and quantitative models from biology to the topic.
I'm not sure how this goes much beyond their paper in Nature from 2007. What looks new at this point is the stunning claim that there are Indo-European words that are 40,000 years old. That's well over 30,000 years older than any usual time depth for IE and the earliest dates there are controversial. Assuming that the ancestors of the Indo-Europeans had some linguistic continuity back 40,000 is striking.
No need for a big discussion here, but one note: The article doesn't make as clear as it might for non-specialists that we're talking about correspondences that are traceable over time, not really consistent word forms. Reconstructions differ, but let's figure for the sake of argument that Proto-Indo-European was spoken 6,000 years ago. And assume for the moment that the first person nominative singular pronoun was *Heǵ-, where the H might be some kind of fricative, the 'first laryngeal', and the last sound is a voiced palatal stop. For these purposes, that matches English I, Latvian es and Tocharian A ñuk. We might wonder what those forms look like another 30,000 or 35,000 years earlier look like.
Update 12:45, Mr V: The Log's Mark Liberman has linked the BBC audio on this and Pagel really appears to say that we're dealing with a lack of change here — that Indo-European 'cavemen' would have understood [aj] 'I', despite the fact that the IE was surely more like *Heǵ-. Likewise with our [hu:] 'who', which would have been something like *kʷós. Good lord god almighty.