Linguistics

I remember seeing this study a few months ago. Basically: both Caucasian and Kurgan hypotheses would be correct, for Early and Late PIE respectively. The dating sounds reasonable, too - ~8000 and ~5000 years ago respectively.

For Indo-Europeanists, this means that Early and Late PIE are so distant from each other that you can't really abstract them as "the same thing" for the sake of reconstruction, as it'll cause artefacts due to the diachronic mix-and-match. And that for Early PIE it's actually useful to check the three Caucasian families (NW, NE, and Kartvelian) for potential Sprachbund features, such as:

• vertical vowel system - such as *e *o being actually [ə ä], and vowel length being a Late PIE development;
• uvulars - such as *ḱ *k being actually [k q], this should explain why *ḱ is so much more common than plain *k;
• it gives some grounds to something similar to the old glottalic theory, with what has been reconstructed as *(b) d ǵ g gʷ being actually ejectives, [(pʼ) tʼ kʼ qʼ kʷʼ]. For reference notice the absence of [pʼ] in Avar (NWC), versus early PIE *b being so uncommon.

It's not a big surprise as they are talking about two different time periods and migrations - the expansion of the Neolithic from Anatolia and the Bronze Age from the Pontic Steppe, with the latter population largely replacing the former and bringing with it the basis for most of the languages in Western Europe.