[kemerover]

>Uh, voice? That seems dubious. Apart from vernacular or speaking styles that are generally racial in nature, there's really nothing there.

https://www.ncbi.nlm.nih.gov/m/pubmed/17342877

>It's not the obvious cases that matter, but the borderline ones. Is that person Japanese or maybe Inuit?

First of all, I don't think it is appropriate to classify "Japanese" or "Inuit" as races. Race is a more general concept.

Anyway, Japanese and Inuit are easy to distinguish. Middle Eastern and European are better examples. If you take one European person and one Middle Eastern person it would be hard to differentiate them, if you take two groups of one hundred people it would be easy.

[VeejayRampay]

Really depends on the European though... Europe is an extremely diverse continent despite its relatively small size.

[astrodust]

It's still pretty big: https://twitter.com/mariachong/status/918553777031954432

[astrodust]

That link about vocal characteristics is only relevant when talking about huge populations. On an individual basis it's meaningless, there's just too much variation. You need samples of a thousand or more to develop a picture. Where biological differences do exist that might affect intonation in a particular dialect, but not when soemeone was raised with a different dialect or intonation.

Now if "Japanese" isn't a race then nothing is. The vast majority of Japanese are identifiable on a genetic level due to the relative isolation of the country. Same goes for Inuit who were geographically and culturally isolated.

> Middle Eastern and European are better examples.

If you're talking some stereotypically Polish person compared to some stereotypically Arabic person from Egypt, obviously, but there's cases where ordinary Turkish people look more "European" than some groups of Europeans do.

The bell curve on these things is just too wide, there's nothing but overlap in all but the most extreme fringes.

[kemerover]

>That link about vocal characteristics is only relevant when talking about huge populations. On an individual basis it's meaningless, there's just too much variation.

If we are talking about white/black voice difference, then it is meaningful.

>Now if "Japanese" isn't a race then nothing is. The vast majority of Japanese are identifiable on a genetic level due to the relative isolation of the country.

Race is just a more broad concept than an ethnic group, that's what I mean.

>If you're talking some stereotypically Polish person compared to some stereotypically Arabic person from Egypt, obviously, but there's cases where ordinary Turkish people look more "European" than some groups of Europeans do.

That's what I meant when I said that it would be easier if you take groups 100 hundred people, it averages out. And yeah, Europeans and Arabs are white, so there is little difference between them.

>The bell curve on these things is just too wide, there's nothing but overlap in all but the most extreme fringes.

Yes, "cline" is a better description than "race" but it is not as useful in daily life as "race" is.