Hacker News Reader

Comment by empath-nirvana | Hacker News Reader

I think the headline is a little bit too assertive for a paper that is very tentative, about a topic for which people should be very skeptical. Tracing languages features back 10,000-20,000+ years is a fairly iffy proposition. Proto Indo European is _only_ 5000ish years old in comparison.

View on HN

> Proto Indo European is _only_ 5000ish years old in comparison.

While I agree with that it's hard to make definitive statements about language relationships past 10K years, it's pretty well accepted that Proto Indo European existed around 7K years ago.

By 5K years ago, it's primary subgroups, i.e Hellenic, Slavic, Germanic, Italic, and Indo-Iranian were already forming.

> By 5K years ago, it's primary subgroups, i.e Hellenic, Slavic, Germanic, Italic, and Indo-Iranian were already forming

Didn't this process last a very long time though? IIRC the current consensus seems to be that Anatolian split off very early but Baltic/German/Slavic staid on the same branch until quite late (maybe even as late as 1000 BC) with other groups somewhere in between.

The article is very skimpy on the details, but it is worth mentioning that this is not the first time Siberian languages have been linked to indigenous North American languages.

Na-Dene and Yeniseian languages are theorized to be part of a common family and the evidence is pretty good; they come respectively from America and Siberia. See https://www.science.org/content/article/land-bridge-connects... for some more information.

It is interesting that this work places those languages entering only ~5kya, many millennia after the other language families were long established. Their success establishing residence particularly in the Pacific Northwest is remarkable. How they achieved it might never be known, but maybe some complex of technological innovations they brought enabled it.

Ignoring the 10k-20k years part, which I agree it's a little iffy, by definition the first languages of North America originated in Siberia and in the present Russian Far-East, for the simple reason that there's from where the first populations of North Americans came from (I think that the Trans-Pacific theory is not widely accepted).

Unless those populations only started using language once they had set foot in North America, which would be an even greater thing to push.

Am not a comparative linguist by trade but thought the Dene-Yenesian hypothesis was pretty well established at this point?

Edit: wiki link https://en.wikipedia.org/wiki/Den%C3%A9%E2%80%93Yeniseian_la...

First link was proposed by Merrit Ruhlen in 98. People have opinions about him but he and his mentor Greenberg did very well received work on African comparative ling

Unfortunately, Ruhlen, Greenberg, and Starostin's work relies on mass comparison and is neither rigorous nor noteworthy. Prefixing languages are rare and most have been suggested to be related to one another at some point on that basis alone, starting with Trombetti's unscientific speculation in the 1920s.

Ed Vajda brought rigor to the hypothesis in series of publications which established Dene-Yenesian through the comparative method starting in 2008 and continuing to the present. Nichols briefly and favorably reviews this bibliography in her paper. As a senior researcher in the field, she was one of Vajda's original advocates when his proposal made its splashy debut.

Reply on news.ycombinator.com