Deinonysus wrote:Well sure, there could have been as few as three migrations from Siberia into the Americas. But I don't think "recent", even relatively, applies. Even if most of the indigenous languages of the Americas are ultimately descended from one language, that would be something very deep into prehistory. Even by conservative estimates of how early the first migration was, it would make Proto-Indo-European look like an absolute whipper-snapper.
I was working on an assumption of 20,000 to 14,000 years ago for the original peopling of the Americas. Compare that to when Eurasia is thought to have been initially peopled, 125,000–60,000 years ago.
What I'm wondering is why you presume there are "likely hundreds" of indigenous American language families. You agree that the indigenous American languages likely ultimately descend from a handful of language families, so doesn't it follow that further research will discover some of the obscured relationships and end up reducing the number of currently established (proto)language families? Or are you assuming that many of the currently unclassified languages will end up in "new" families?
Thanks for quoting that chapter from Mallon's book. Super interesting stuff!