April 22, 2008

Kalevi Wiik's paper "Where did European Men Come From?"

Some notes on the paper (last update Apr 22):
  • p. 67: points (1) and (2) both refer to a north-south gradient of haplogroup R1b in the Balkans; one of them probably should refer to a different haplogroup (R1a?).
  • Table 12 gives the frequency of haplogroups in Macedonians, identified in Maps 64-65 by the former Yugoslav Republic of Macedonia (fYROM). However, the ultimate source of the data is Semino et al. "The Genetic Legacy of Paleolithic Homo sapiens sapiens in Extant Europeans: A Y Chromosome Perspective", SCIENCE VOL 290 10 NOVEMBER 2000 where the sample is described as "20 Macedonians from northern Greece". Thus, the identification of the sample as Slavic and representing the fYROM is incorrect (cf. Pericic et al. Mol. Biol. Evol. 22(10):1964–1975.) where the "Macedonians" of Semino et al. (2000) are clearly listed as "Macedonian (Greek)" and their language is listed as "IE (Greek)".

    A source of data from the fYROM is: Bosch et al. "Paternal and maternal lineages in the Balkans show a homogeneous landscape over linguistic barriers, except for the isolated Aromuns", Ann Hum Genet. 2006 Jul;70(Pt 4):459-87.
  • Map 1 shows the expansion of N3 into Europe as taking place 12,000 years ago. According to Derenko et al. (J Hum Genet (2007) 52:763–770) there is substructure within N3, and the oldest N3 cluster expanded 8kya into Europe.
Kalevi Wiik, "Where did European Men Come From?", Journal of Genetic Genealogy 4:35-85. (pdf)


  1. A substandard article to put it mildly.

  2. Some fascinating maps though. I notice that the authors are able to form their theory by, as usual, ignoring any evidence from East, South or Southeast Asia. Consequently they are able to write that F split directly from E. Hardly possible. What about C and D which split from F and E respectively. Accepting that would alter their perspective on early splits.

  3. The J2 map is pathetic. Whose grandson drew that? I am J1 and may be rare in Europe but it is definitely more than 1% of the Y haplogroups found in Italy and other Southern European countries and Anatolia. J1 is typically ignored. Also I and J branched off IJ, so that should mean that I and J are as old as the other and developed in the same location. So J is just as likely to be in Europe prior to the farming cultural change as I. Most of the dates are incorrect especially for the R1 haplogroups. Old Europeans indeed! More likely Old Asians.

  4. I am unhappy with him bringing languages in. Most philologists would put the ancestor of IndoEuropean languages far later than he obviously does.

  5. What I found fascinating was the fact that the Croatian LGM refugium, has signature Y-DNA I-P37 (now I2a*), and this is associated with I2a1 from Sardinia - but what I didn't know was that in central Spain about 33% of men carry I2a1 - this is fascinating, as it would seem to be the oldest Y-DNA in Europe...

  6. Haplogroup I1b1 (now I2a) is regularly found in about 15% to 30% of males in north(east)ern Spain (mainly Castile, Aragon, and Catalonia) as well as in southern France (especially Bearn, Gascony, Aquitaine - e.g. areas that have traditionally contained speakers of Basque or Basque-related languages or at least have some elements of Basque culture). However, some people refuse to associate this haplogroup with the Basques and insist that the "true Basque haplogroup" is R1b1c. I don't know how to convince these people that the original Basque haplogroup was more likely I2a (or even more specifically, its subclade I2a1), despite the fact that this would explain the apparently Basque-related elements in the language of Sardinia.

  7. ebizur,

    I totally agree, this haplogroup would link all relict European populations.

    It's interesting that Wiik mentions that Ossetians - Iranic speakers from North Caucusus - have high R1b, but of the Western European type, as do the Yighurs of the Tarim basin...

    It's beginning to seem that R1b1c may be an invasive haplotype in Western Europe?!

  8. It's interesting that Wiik mentions that Ossetians - Iranic speakers from North Caucusus - have high R1b, but of the Western European type, as do the Yighurs of the Tarim basin...

    There really is no evidence for a significant presence of R1b in the Uyghurs. This is a clear case of a rumor that has gotten out of hand, and by mistakenly equating the old HG1 or P*(xR1a) with haplogroup R1b.

    Also, Wiik doesn't distinguish between varieties of R1b in his paper.

  9. Although Dienekes is correct that there is not really any evidence for a significant presence of haplogroup R1b among Uyghurs, there definitely is evidence for a presence of some R1b among Uyghurs. The reason it is not significant is because it is not significantly higher than the frequency of haplogroup R1b among neighboring Turkic and Mongolic populations, such as Uzbek, Salar, Sarta, etc. All of these populations have slightly more or slightly less than 10% R1b. This is because of Paleolithic Basque colonization of the area of the Taklamakan Desert prior to its desiccation, of course. ;) (I hope we are allowed to be sarcastic on here once in a while.)

  10. Dienekes,

    Yes Wiik doesn't distinguish between R1b varieties and that's a great pity - though maybe the data just isn't there.

    However he does state that both Ossetians and Yighur have the Western type - by which I take it that he means that they are NOT like the R1b found in Anatolia, Iraq and neighboring areas - if this is correct then that's major news - to me at least!

  11. I think he calls R1b "Western European" in general in his simplified scheme. Anyway, perhaps there is data that it is one type or another, but I haven't seen it.


Stay on topic. Be polite. Use facts and arguments. Be Brief. Do not post back to back comments in the same thread, unless you absolutely have to. Don't quote excessively. Google before you ask.