That's well interesting, but the last paragraph seems to be unfounded speculation. I don't really see where the notion of non-familiar kinshipncomes in, as gene

That's well interesting, but the last paragraph seems to be unfounded speculation. I don't really see where the notion of non-familiar kinshipncomes in, as genetic inheritence is usualy enough for a first lesson in recursion, the father of the father of the father. Although, this didn't get me very far in class, as unrolling of more involved recursive loops in an itterative declaration style is notoriously difficult and in fact associated with increased complexity. What might be more instructive for logic is negation, I suspect, as 50kya would roughly be the time of Aurignacian expansion and Cro Magnon meeting Neandertaler in Europe, though I'm not sure if that's the area they are looking at. Path recursion would surely play a role, too, for nomadic hunter gatherers, but material evidence is of course substantial. In that sense, locative expressions might suffice to generate terminology, eg. mary @ john, or master of engineering (in the gravel pit). This requires no social construct perse, only tolerance to the fact. Analysis in post hoc rationalization could thus be an emergent property. But! Bridestealing might require risk awareness, ie. predictive power, so it's a back and forth and survival ofthe fittest at play, if bride stealing played as much a role as it later did.