August 11, 2009

Finally, an updated look at Y-chromosomes of Jewish priests (Hammer et al. 2009)

We had expected an update on the early Cohen Modal Haplotype work for a few years now. That work had established the distinctiveness of the Cohen Y-chromosome gene pool relative to that of other Jews, which suggested common founders to the Jewish priesthood, but did not provide sufficient phylogenetic resolution: the genetic signature of the Cohens was a 6-marker haplotype that could be found in both haplogroups J1 and J2 and in non-Jewish populations at substantial frequency.

Thus, it became necessary to find a more stringent characterization of Cohen Y-chromosomes that would represent true founder effects in that population. The new paper seems to identify at least two such lineages, one in J-P58, which is a subset of J1, and one in J2a-M410*, and estimates that they were founded 3.2 and 4.2 thousand years ago.

I will comment on this further when I get a hold of the paper and supplementary material.

UPDATE: The authors use the evolutionary mutation rate, and thus the presented ages are overestimated significantly. However, there are reasons to doubt the germline-rate estimate of about 1,000 years for the J1 lineage:
  • Demographic plausibility of growth to encompass nearly a third the Cohanim in about 1,000 years. I am not sure what the demic size of Cohanim is, but going from 1 individual to the current population size would require a consistent high growth over many generations. This seems implausible, unless there is indeed historical evidence for such a Cohen founder's descendants extraordinary success.
  • The presence of the founding lineage in both Ashkenazim and Sephardim may suggest a common ancestor before the separation of these two populations.
In short, I am skeptical of the age estimates in this paper due to the use of the evolutionary mutation rate, but the wide confidence intervals of Y-STR based estimates, coupled with the knowledge of the lineage's wide geographical dispersal and high number of descendants may make us believe that the founder in question lived at an older time than ~1kya.

Pinpointing Jewish priestly founders to specific individuals, including Biblical ones, is not easy, as the Y-STR technology does not allow for anything resembling accurate age estimation. However, the paper is a welcome new study of a much-discussed topic, and adds significantly to our understanding.

UPDATE II: On the other hand, the J-P58* haplogroup was found in 325 of 2,099 non-Jews surveyed, but the extended Cohen Modal Haplotype (eCMH) in none (Table S2). If the eCMH founder lived 3+ kya, it would be strange indeed if he left no non-Jewish descendants, as it would imply zero conversion from that lineage to other religions. The lack of non-Jewish eCMHs does support the "Jewishness" of this lineage, but on the other hand, makes a very old age more difficult to accept.

My guess is that the eCMH founder lived in Roman times. This would simultaneously allow enough time to explain the lineage's geographical and demographic growth, while also explaining its limited penetration to non-Jewish populations.

Human Genetics doi:10.1007/s00439-009-0727-5

Extended Y chromosome haplotypes resolve multiple and unique lineages of the Jewish priesthood


Michael F. Hammer et al.

Abstract

It has been known for over a decade that a majority of men who self report as members of the Jewish priesthood (Cohanim) carry a characteristic Y chromosome haplotype termed the Cohen Modal Haplotype (CMH). The CMH has since been used to trace putative Jewish ancestral origins of various populations. However, the limited number of binary and STR Y chromosome markers used previously did not provide the phylogenetic resolution needed to infer the number of independent paternal lineages that are encompassed within the Cohanim or their coalescence times. Accordingly, we have genotyped 75 binary markers and 12 Y-STRs in a sample of 215 Cohanim from diverse Jewish communities, 1,575 Jewish men from across the range of the Jewish Diaspora, and 2,099 non-Jewish men from the Near East, Europe, Central Asia, and India. While Cohanim from diverse backgrounds carry a total of 21 Y chromosome haplogroups, 5 haplogroups account for 79.5% of Cohanim Y chromosomes. The most frequent Cohanim lineage (46.1%) is marked by the recently reported P58 T->C mutation, which is prevalent in the Near East. Based on genotypes at 12 Y-STRs, we identify an extended CMH on the J-P58* background that predominates in both Ashkenazi and non-Ashkenazi Cohanim and is remarkably absent in non-Jews. The estimated divergence time of this lineage based on 17 STRs is 3,190 ± 1,090 years. Notably, the second most frequent Cohanim lineage (J-M410*, 14.4%) contains an extended modal haplotype that is also limited to Ashkenazi and non-Ashkenazi Cohanim and is estimated to be 4.2 ± 1.3 ky old. These results support the hypothesis of a common origin of the CMH in the Near East well before the dispersion of the Jewish people into separate communities, and indicate that the majority of contemporary Jewish priests descend from a limited number of paternal lineages.
Electronic supplementary material The online version of this article (doi:10.1007/s00439-009-0727-5) contains supplementary material, which is available to authorized users.

Link

112 comments:

Gioiello said...

From a first reading it seems not only that hg.R wasn't among the pre-exile Jews, but as well the Jewish Hgs. of the CMH (J-P58* and J-410) are present also among the non Jewish population, and not having the CMH, probably also the Jews who haven't the CMH are actually, probably, of non Jewish extraction, otherwise they should have had a link with them. I think it is an own goal.

Anonymous said...

I don't think the study says anything new or revelatory.

It is just making assumptions on the Near Eastern origin of J-P58 Jews based on the frequencies in certain Near Eastern populations of today and extrapolating thousands of years back to the time the Canaanites/Hebrews/Israelites/Judeans/whatever supposedly gotten from religious books. I don't believe in any religion other than my own and don't recognize the validity of any of the so called holy books. So there is a serious lack of proof as to who these Jews are, and where they came from in Eurasia. Also Middle Eastern folks are notorious for inbreeding. Inbreeding skews genetics and haplogroups hence the ridiculously high frequencies of J1 among Bedouins and Yemenis. They should be excluded from any studies on the grounds of inbreeding and lack of genetic variability. Further the fact that Turkish J1 is more genetically diverse than any Semitic language speaking J1 bearers is not taken into account. Even Italian J1 is highly diverse yet the percentage of J1 men in Italy is much smaller than in Yemen or among Bedouin. The impact of founder effects, recent immigrations and marriage practices have not been factored in before making the statement of J1 in todays Jews comes from an Near East source prior to the supposed diaspora of Jews from the Near East.

The R group among Jews is much older than any J groups. They don't comment much on that oddity.

It is strange that the 12 marker CMH has 439=12 yet no J1 in their list has a figure less than 14. Someone has erred in the supp material no.8. Personally I dislike sloppy workmanship. Sloppy workmanship seems to be inherent in these type of studies.

In short nothing new e.g anyone who is interested in J1 can see easily that Cohen Jews are 385a/b of 13/15 and non Jews like Arabs are often 13/18. It is all there to see in the J1 group of FTDNA.

Gioiello said...

Re the more ancientness of Hg. R among Jews it is easily explaning: all the people who have picked up their DNA from many sources have a more variance. It doesn't mean that it is water of their pit.

Gioiello said...

The Authors have had the decency to admit: “In contrast, our network and divergence time analyses suggest that R-M269 chromosomes entered the Cohanim population via several “migration” events, and do not represent a single Cohanim founding lineage. For example, divergence time estimates are much older for Cohanim R-M269 chromosomes (>10 kyears) than for the three Cohanim lineages in haplogroup J discussed above, and median-joining networks of Cohanim R-M269 chromosomes lack a modal haplotype and show many unrelated singleton haplotypes that are interspersed among Cohanim and non-Jewish samples (Figure S4)” (page 11).

Then I consider closed the vicissitude for which I was banned from two forums. As the Authors are using the Zhivotovsky rate, if also J-P58 and J-M410 are Jewish will see when these Authors will write the next paper.

AWood said...

Not necessarily. All the haplotypes are of the DYS393=12 variety. If you are suggesting some of these priests are converted European jews I would say the haplotype evidence contradicts this. At least post-diaspora. It would appear that the peopling of the Near East was not from a single source. Take any study which has sampled this region and you will find plenty of R-M269. There are adjacent regions in the Aegean who have similar levels of diversity, namely Greece and Turkey. I don't think anyone can attest to who the "real" Jews were but J1e* is at least a good chunk.

Gioiello said...

It isn’t true: 1) none have the DYS426=11, so diffused among the R/L23- Jews (see the ht 35 project of Vizachero): the others are above all Italians; 2) 116, 1403, 1405, 1512, 1477 have DYS393=13, exactly like the haplotype dominant in the country where those Jews lived. It seems that the Authors, to defend a possible Jewish J-P58 and J-M410, have ceased to insist on R1b1b2. A clear sign of weakness, though it isn’t certain that also those haplogroups are surely Jewish. A Jewish Historian, in Matai ve-ech humza ha-Yehudi, has supported the North African origin of those haplogroups and the Middle Age is better in line with the mutation rate used. Zhivotovsky can be good for 10/20,000 years, not certainly for a few thousands.

Maju said...

So we have two major Kohen lineages:

A. A J1e haplotype (46%) that is said to be some 3200 y.o. (4300-2100), that is: from c.1200 BCE (2300-100 BCE)

B. A J2a haplotype (14%) that is claimed to be 4200 y.o. (5500-2900), that is: from c.2200 BCE (3500-900 BCE)

It is very noticeable that while Jews overall have only minor amounts of the typically Palestinian J1 lineage (15%), Kohanim have almost 50%, what does suggest that this J1-CMH is probably of Palestinian (or Negev-Sinai Bedouin) origin, unlike the bulk of Jewish Y-DNA ancestry, which looks from a more Northernly or Northwesternly area (Anatolia possibly).

If we consider the archaeological evidence, Jews were already present at Beit Shemesh c.1300 BCE, what means that the semi-mythical Moshe and Aaron lived (if they actually existed) before that date. If we take for granted the Biblical story of Egyptian residence and these archeological dates, we can reasonably presume that Hebrews were a subgroup of the Hyksos, what narrows pretty much the window for the Exodus (and hence the hypothetical historical Aaron) to c. 1400-1300 BCE.

The paper's TRMCA for the J1e-MCH seems in this case rather accurate, suggesting that the mythological claim may have some basis. The other lineages would seem to have been incorporated in the caste by either illegitimacy or conscious incorporation of other lineages.

The J2a-MCH is curious indeed because it suggests an older founder (and not a more recent one, as would be expected by mere illegitimacy or individual "adoption"), so it suggests to me some lineage older than the Aaronite one (i.e. estabilished before the conquest of Canaan) was incorporated to the caste. This arises many possibilities:

a) That some member of some older Hebrew lineage was accepted within the Kohen caste soon after it was created

b) That some member of some indigenous Canaanite lineage was accepted within the caste also few generations after it was instituted, probably for political reasons

c) That some member of some very old lineage within the Hellenistic diaspora was incorporated much later, in historical times.

d) That the J2a-MCH is the actual Aaron lineage and that the J1e-MCH is in fact the incorporated one. This would require to push the dates some 900-1000 years ahead (still within the CI but close to the low edge) and also assume that at the time of the Sadducees, there was a massive replacement of the true Kohen lineage.

My favorite is option B but who knows?

Ebizur said...
This comment has been removed by the author.
Maju said...

@Ponto: I am not talking of religion but of mythology. And mythology often has some truth to it. This applies to Greek mythology, Sumerian mythology, Vedic mythology and surely as well to Hebrew mythology.

One thing is clear: some 46% of Kohen stem from a single common ancestor and mythology also attributes a single common ancestor. This evidence suggests (can't prove) that there is some truth to this foundational myth, though, of course, as with other mythical ancestors, many just have such no biological ascendancy.

... Southwest Asia (...) is probably the origin point of all haplogroups which track back to F including the R haplogroup.

No way! F is with all likelihood of South Asian origin: 4/7 derived haplogroups are only or mostly found in that region (and the rest: one west of it, another east of SA and the third, IJK, widespread in both directions). F is South Asian by origin, no doubt.

Proving that todays Jewry track back to the Levantine/Mesopotamian zones Southwest Asia is somewhat harder.

We're not talking the whole global Jewry but a specific priestly caste within it. We are not talking the whole ancestry but the paternal lineage only as well.

Tying in todays Bedouin or other inhabitants of the Arabian Plates landmass is rather silly. Every out of Africa movement of people has gone through that zone...

Actually that would not be the case if you ponder the (quite mainstream) rapid coastal migration model, which suggests a route via South Arabia, not the Fertile Crescent. Whatever the case the Fertile Crescent must have been depopulated almost totally after the Aterian related early H. sapiens (Skuhl, etc.) and only much later colonized from South Asia.

Whatever the case J1 is a marker almost exclusively associated to West Asian and North African peoples and reaches a high mark in Palestine or at least the Negev. All this makes more plausible the hypothesis that the J1e-CMH is what is supposed to be.

What is more important to me is the presence of R groups in Cohen Jews. The time to the common ancestor is older than that of J group Jews. Someone said it was due to migrants to the Jews' ancestors. Maybe. It could be that the R group is just older than the J group in Jewish people and that the J group were the migrants.

What's the presence of R in Palestine and other West Asian areas? I understand that the highest concentrations are in Anatolia (R1b), what makes it unlikely that it is a native Jewish or pre-Hebraic Canaanite lineage. This may be also the case of J2a, though this one is more common in Palestine.

Anyhow the CMH is not about the origin of Jews as a whole but about the origin of a priestly caste among them. The simple fact that the apportions of J1 are so different between the bulk of Jews (c.15%) and Kohanim (c.45%) makes any correlation impossible to sustain.

But guess that if you presume that high densities of farmers can be pushed around and replaced as easily as you imagine (i.e. your comparison with Australia), then everything is possible. Sadly for those who think that way (a very mythological way of thinking indeed), each time aDNA is analyzed we get data reinforcing the antiquity of the most important locally extant lineages.

Maju said...

Anyhow, Ponto may have a point.

I'm missing a graph of the alleged main J1e-CMH in relation with other non-Jewish populations. The ones in the supplementary material are quite revealing but for some odd reason there is no phylogenetic tree for the J1e haplotype.

One can easily see that R1b1b2 in Kohanim is just a random sample of non-Jewish R1b1b2, it's also quite apparent that J2a is inserted near the root of this lineage, forming at least two separate clusters (i.e. at least two founder lineages among Kohanim and a common ancestor that is not different from the CA of the whole J2a lineage - hence the TRMCA seems oddly hyper-recent.

On the other hand, most of the Kohen J2b forms a clearly distinct derived cluster, meaning some founder effect, though J2b should be of Anatolian or even European origin, not West Asian.

But you don't see anything of the like for the most important lineage: J1e. Why has this graph been withheld? This makes the whole study look more suspicious. Maybe there is no structure whatsoever in the Cohen J1e after all.

Hmmm... quite strange. The whole study looks more suspicious to me now that I look at this important "detail".

eurologist said...

Ponto,

While I am hesitant to subscribe to all what you said, you made a number of very good points that are often neglected, swept under the rug, ignored, or no longer kept in memory by younger folks...

Thank you for posting.

Not sure but my feeling is that all these Rabbi y-DNA debates are about the belief (hope?) that there is something special (educational, scientific intuition, emotional intelligence) in that group.

The problem with that misguided assumption is of course that such characteristics are typically carried within autosomal DNA - not in y-DNA...

BTW, I am saying this while my family also comes form a long line of teachers and priests, all with above average language, foreign language, teaching, and science credentials.

Also, the ability to easily read and speak multiple languages was certainly both incredibly advantageous and rare for people between 6,000 years ago to present if they were not born into an otherwise high class/caste.

So, on the other hand, there clearly is a possibility of some general genetic connection.

Kepler said...

Aaron,
So, what is this DYS393=12 variety?
(I am asking as I am J2 and have 393=12, but DYS426=11)

Maju,

I know Palestinians have a very high concentration of J1, but can this not be due to founder's effect?
Was J1 not originated in Arabia?

The Hebrew language was a Canaanite language very similar to Phoenician, I read somewhere people are not clear about at what point or points Aramaic and other languages from the North started to influence it.

Israel Finkelstein's book The Bible Unearthed is very interesting reading. One of the things he presents is the probability that Jews were a mix of people coming from Egypt (but not as the Exodus story) and the Canaanite themselves.

Perhaps around the VII to VI century BC several families, some related, some just in the general haplogroup J2 or J1, but working as priests gravitated around this Aaron belief for all of them, at the time when the priests in Jerusalem consolidated their power and displaced other beliefs and the books of Joshua, Deuteronomy, Chronicles etc. were produced.

Maju said...

I know Palestinians have a very high concentration of J1, but can this not be due to founder's effect?
Was J1 not originated in Arabia?
-

Why would anything originate in Arabia, a mostly desertic area? I presume you speak of the peninsula not all the Arab World. In general it's clear that Arabia peninsula was colonized from outside, specially the north (i.e. the Fertile Crescent).

Yemen has high apportions of J1 but low diversity, it seems. The "Bedouins" you see in all studies are Negev Bedouins from what is now Israel (i.e. Southern Palestinians), other Bedouins do not show this high concentration.

If you look at the structure of the haplogroup at the several Cruciani papers on the matter, you see that the root appears to be in West Asia and then there is larger cluster whose sub-root may be in North Africa. The lineage has therefore been dancing between West Asia and North Africa and, considering the role of Palestine in the Paleolithic and Neolithic it is very likely, IMO, that the core was there.

Sadly no diversity studies have been performed on Palestinians, who are mostly used if anything as a control group in studies on Jews or at best just another sample in wider studies.

But I do have the clear impression that J1 is important in Palestine since old and that was important among early proto-Semites in the same area if the mainstream theories on Semitic ethnogenesis are correct, i.e. Afroasiatization from Egypt of Neolithic Palestinians at the Negev and also in the agricultural zone, followed by the formation of a cultural arc of pastoralists at the semidesertic edge of the Fertile Crescent, who in turn invaded the agricultural zones in the 4th milennium, completing the process of ethno-linguistic homogenization as Semitic of the lowlands West Asia.

J1 in North Africa appears to be related to Afroasiatic, either Capsian Epipaleolithic or Capsian Neolithic, but not so much to Semitic as was claimed early on.

The only major doubt I have is the aboundant J1 of the Eastern Caucasus: how to fit it in the puzzle. Maybe it's just some sort of Neolithic founder effect, like E1b1b in Greece.

The Hebrew language was a Canaanite language very similar to Phoenician (...) the probability that Jews were a mix of people coming from Egypt (but not as the Exodus story) and the Canaanite themselves.

Sure. Phoenicians appear to be just regular Canaanites who, maybe influenced by the Sea Peoples (Greeks, etc.), and under pressure from all other sides (Assyrians, Hebrews), went naval when Greece fell in the Dark Ages. Hebrews instead seem some sort of invaders, but obviously they assimilated the natives. The twelve tribes only appear after the conquest of southern Canaan, so most of them were probably just absorbed Canaanite groups.

I guess the Hebrews were, apart of some sort of wacky sect (they remind me a lot of modern Mormons in this aspect), probably a leftover group of the Semitic Hyksos, who in turn had invaded Egypt from the area of Palestine and later vanquished. I guess that their post-Hyksos status in Egypt could be considered as some sort of "captivity", leading to their migration. But it's hard to discern that part, obviously severely hyped and distorted for propaganda reasons and normal mythological distortion.

Maju said...

(cont.)

Perhaps around the VII to VI century BC several families, some related, some just in the general haplogroup J2 or J1, but working as priests gravitated around this Aaron belief for all of them, at the time when the priests in Jerusalem consolidated their power and displaced other beliefs and the books of Joshua, Deuteronomy, Chronicles etc. were produced.

I'm more inclined to give some credibility to the ancestry myth. Just that as with all formally closed elite castes (think Brahmins for example or European Medieval aristocracy), loads of exceptions must have been made to acomodate allies - or even lucky imposters.

But I fail to see the evidence in favor of a J1e-MCH clear. This paper lacks the fundamental evidence that there is a distinct J1e-MCH which is not just part of the broader J1e with no particular structure. For some murky reason the network tree for this crucial lineage has been withheld, what casts more doubts than anything else.

AWood said...

Gioiello, any vendetta you have with the administrator of the ht35 project should be left out of this discussion. I was wrong, you were right...there are some DYS393=13 in Poland and Russia (not uncommon as both regions had high levels of Jews), as well as a single Dutch. These individuals have not been deep-clade tested, the DYS393=13 could very well be L23+ and test negative for everything downstream. Of the millions of men out there who belong in the ht35 project, we've only tested a tiny fragment of them. Of course, there are converts or NPE, but that is not my earlier point. How would you explain the native Israeli Cohen that are R1b1b2 DYS393=12? From the study, 2/5 were R1b1b2, and the other 3 were J-P58. These R1b individuals are not recent migrants to the area, which brings up the point, can we really define who the original/true Jews are?

Kepler said...

Maju,
The Hiskios story does not exclude the other. It could have been that a subset of people from the Hebrew-Canaanite lot of the VIII-VII centuries and who had been the priests in charge of the YHWH cult in Jerusalem all claimed being descendant of Aaron, whether some or part were descendants of those who came from Egypt or from other groups who were all the time in Canaan.

In any case: I read there is a group in Israel starting some research involving very old DNA material there. The project is just starting and goes for some years.

Gioiello said...

Which “vendetta” against the bigot? It will be rather a revenge, a “rivincita”, than a vengeance if my theories will end to win. In all this vicissitude I am the victim, the Jew of the situation, and others are the persecutors. I have always supported theories in a debate that should be free. It seems to me that my theories, also on this paper and on its authors, in a free forum like this, are shared by others.
Aaron, I sign my writings by my name and surname and don’t like who hides himself. Your name shows your origin and I wouldn’t be surprised if you were Sam Vass, one I have always considered a friend. Sam Vass knows very well my theories, long before I began to write on these forums, and I haven’t other answers as regards the past. Now these answers you have also from these illustrious scholars, who are Jews and cannot be accused to be against them.
I am working now on some descendants of the Jew converted Abraham Senyor (don’t forget that my far ancestor was named “Signorino” and I have always considered a possible Jewish origin of my DNA). Not having religions and myths, my research obeys only to reason and science: proofs. Also if I was a Jew in my origin, I don’t forget to be a compatriot of Galileo Galilei, who had himself a Jewish name.

Vincent said...

Maju wrote: "One can easily see that R1b1b2 in Kohanim is just a random sample of non-Jewish R1b1b2 . . . "

I'm not entirely sure what you are driving at, since all Cohanim are (in fact) Jewish. I don't think they'll let you be a Jewish priest unless you are Jewish.

However, I can say that the common assertion that all the R1b1b2 among Ashkenazim or Sephardim in general (and Cohanim in particular) is the result of admixture with or conversion of European men appears to be false. A significant portion (roughly 50% to 75%, depending on the sample) of Jewish R1b1b2 lineages are non-European in origin, having more in common with Arabian and Persian R1b1b2 men than with central or western European R1b1b2 men. These Arabian, Persian, and/or Jewish men are predominately negative for the L51 SNP, while the vast majority of European R1b1b2 men (something like 99% of them, actually) are positive for L51.

This looks like the case in the new Hammer et al. paper. The paper omits the important new SNPs downstream of R1b1b2, but we know enough about their STR patterns to make an inference. Seven of the 12 R1b1b2 Cohanim in this paper (roughly 58%) are DYS393=12. This is roughly in line with Near Eastern samples, but way more DYS393=12 than you'd find in any European R1b1b2 sample. I think the first presumption must be that about half the R1b1b2 Cohanim trace back to pre-diaspora R1b1b2 Jews, while the other half post-diaspora conversion or admixture.

VV

Dienekes said...

A significant portion (roughly 50% to 75%, depending on the sample) of Jewish R1b1b2 lineages are non-European in origin, having more in common with Arabian and Persian R1b1b2 men than with central or western European R1b1b2 men.

There are plenty of European (and West Asian) populations between "Central and Western Europe" and "Arabians and Persians".

Vincent said...

There are plenty of European (and West Asian) populations between "Central and Western Europe" and "Arabians and Persians".

Of course there are, though I'm not sure that's a particularly interesting observation. The main question raised re: the R1b1b2 Ashkenazi has been whether the R1b1b2 had its source in the Near East or in Rhineland. The conventional view appears to have been "virtually all from Rhineland", but it turns out the answer is probably "most Near Eastern".

The important data point is that we can observe a significant portion of the Jewish R1b1b2 men being more closely related to (say) Near Eastern R1b1b2 men than to (say) German R1b1b2 men.

VV

Dienekes said...

The main question raised re: the R1b1b2 Ashkenazi has been whether the R1b1b2 had its source in the Near East or in Rhineland.

Once again, that is a false question, as Jews spent a whole lot of time in the Mediterranean before ever migrating to the Rhineland, and thus had plenty of opportunity to pick up R1b1b2 from European populations other than "Rheinlanders"

Vincent said...

Once again, that is a false question, as Jews spent a whole lot of time in the Mediterranean before ever migrating to the Rhineland, and thus had plenty of opportunity to pick up R1b1b2 from European populations other than "Rheinlanders"

There were Jews in Mediterranean Europe, true, but what is the evidence that the Ashkenazi are descended from these communities instead of the Asian ones?

I suppose there may be an intersting quesiton in whether the two sources of Jewish R1b1b2 came into the Jewish gene pool separately or via some pre-mixed source (e.g. Greece).

At present, though, we can say that the R1b1b2 facet of Ashkenazi Cohanim is largely consistent with the tradition of an origin in the Near East, and not wholly consistent with an origin in Rhineland.

VV

Maju said...

Kepler said: In any case: I read there is a group in Israel starting some research involving very old DNA material there. The project is just starting and goes for some years.

Will be interesting to know when finished. :)

Vincent said: I'm not entirely sure what you are driving at, since all Cohanim are (in fact) Jewish. I don't think they'll let you be a Jewish priest unless you are Jewish.

The graph compares Kohanim and non-Jews. It does not mention no-Kohen Jews. It's from the supplementary material of this paper.

Anyhow, Jewish is before anything a religious identity, at least historically. And in Antiquity and early Middle Ages at least Judaism was proselytist (and Christianism and to a large extent Islam too are nothing but particularly successful Judaistic proselytist sects). In Canaan at the Bronze-Iron ages transition and in Antinquity within the Hellenistic World (and also in the early Middle Ages), Judaism absorbed many non-Jews, up to the point that modern Jews look more "Turkish" than Palestinian. Some of these converts may have penetrated the priestly caste in one way or another. At least it is known that some Khazars were accepted in the Levitic caste (and this seems to show up in Ashkenazi Levi genetics). Other such high caste absorptions are not known on historical data but we can imagine. Of course some may be just descendants of illegitimate children raised as legitimate, etc.

However, I can say that the common assertion that all the R1b1b2 among Ashkenazim or Sephardim in general (and Cohanim in particular) is the result of admixture with or conversion of European men appears to be false.

I never suggested Europeans. R1b is common enough in Turkey, where the Diaspora was centered already in the Hellenistic and Roman period. I am pretty sure that the bulk of modern Jewish ancestry is Anatolian (and from other Ancient Diaspora locations, like Syria or Lower Egypt, as well as of Medieval Judaist nations like Khazars, Berbers, Kurds, Yemenis...) There is an additional "final destination" layer from Europe and elsewhere but it's not larger than some 20% and probably more introduced by women than men.

I'm still awaiting for a good study that compares Jews and Turks (as well as other candidate ancestor peoples).

Vincent said...

Maju said: "The graph compares Kohanim and non-Jews. It does not mention no-Kohen Jews. It's from the supplementary material of this paper. "

I don't know what is going on with figure S4, but it doesn't make much sense.

Five of the haplotypes in the paper are very closely related (samples 2046, 1406, 1397, 2049, 1403, and 1512) and these have very close analogs in the Jewish R1b project. In the MJ network (figure S4) they are not placed near each other, so something appears off in this analysis.

VV

Gioiello said...

What are trying to say Mr Vizachero? As 1403 and 1512 have DYS393=13, but 2046, 1406, 1397 and 2049 have DYS393=12, they all are R/ht35, i.e. R/L23+. This would be in favor of a true Jewish origin.
1) All these are Ashkenazim, and have a MRCA between 750 and 900 YBP. They then descend from a single man.
2) Does he know Mr Vizachero how are they related to me (Gioiello Tognoni)? Between 1125 and 1350 YBP.
3) The question is that my old one: am I a Jew or are they Italians?
The question he hasn’t yet answered is: how many R/L23+ but L150- has he found among Jews? Among Italians so far we have one: Mr Romitti.

Vincent said...

As 1403 and 1512 have DYS393=13, but 2046, 1406, 1397 and 2049 have DYS393=12, they all are R/ht35, i.e. R/L23+.

Yes these six appear to be very closely related, and all appear to be L23+ L51-. They are also closely related to a large group of men in the Jewish R1b project at FTDNA, some of which have been SNP tested, which is how we can be so sure.

The question is that my old one: am I a Jew or are they Italians?

If you don't know whether you are Jewish or not, I'm sure I don't. None of these particular Cohanim are Italian as far as I can tell.

VV

Gioiello said...

Of course my hypothesis is that they are an introgression of some Italian when Jews, from South Italy, during the dark age of Middle Age, were migrating to the Rhine Valley. These ideas I have expressed many times in the past, and also that during the Roman Age some haplotypes like R-L23- and R-L23+ were more diffused in Italy and that Jews have conserved and multiplied these haplotypes from their bottle neck to the great expansion after the 14th century. Of course I have always considered the opposite hypothesis: that Italians are Jews (free or slaves) who established themselves in Italy after the Diaspora. But what will decide it will be my question: where all these haplotypes arose? For this my question about the presence of R-L23+ but L150- is crucial and I am waiting for the answer, from you or from whichever other.

Maju said...

Gioello: isn't a lot simpler to think that these R1b(xR1b1b2a1) lineages found in Southern Italy are just the descendants of Neolithic (or post Neolithic) migrations from the Aegean and that is also the main area where the modern Jewish ethnicity was formed in Hellenistic times before Christianity and Islam effectively banned Judaism from further proselytism?

The answer is maybe something more in the line of: "neither Italians nor Jews, all Pelasgians". ;)

Gioiello said...

If you ever read my postings, being I a Tuscan, then probably an Etruscan, I wrote many times on this: on Pelasgians, Etruscans, Sea Peoples, Philistin, etc. I said also many times that only three little towns have expressed in the past a geniality much over the common (Athens, Florence and "Ashkenazim", which we can consider a little town in to-day world). This seemed racist to someone, and I have had always problems with the forums. But all these peoples had a Pelasgian ancestry. My theory of the Italian refugium, then of the Rhaetic-Etruscan Fatherland, presupposes that all these peoples come from the Alma Mater Italia. This is my mythology, I who am fully atheist, but nourished by good readings.

Kepler said...

You always had problems in forums? How come? Just because you are saying that you, Gioiello, have a geniality much over the common? (indirectly, of course)

How often do you need to say it? I think if it is not too often people should not create problems and let you be happy.

Gioiello said...

If you remember, in that time there were books about the fact that Jews (above all Ashkenazim) have won 27% of the Nobel Prizes, a people of about ten million inhabitants, and that's true, and I wanted to remember also Athens and Florence. That's all. Only who knows the facts can judge if the racist and the persecutor was I or were others. And if you compare the means I have and those of the others (you are a Latin American, aren't you?) you can answer to your question.

Gioiello said...

Dienekes writes: “The presence of the founding lineage in both Ashkenazim and Sephardim may suggest a common ancestor before the separation of these two populations.”
We have many proofs that during Middle Age and above all after 1492 and the expulsion of Jews from the Iberian peninsula there was a mixing of Sephardim and Ashkenazim in Central and Eastern Europe, then I think that this proof has no value.
I agree better with what Dienekes says after: “My guess is that the eCMH founder lived in Roman times”. In Roman times the introgression can have happened everywhere in the Roman Empire.

Gioiello said...

As I’ve said many times, Jews, for their particular vicissitudes, have preserved and then expanded such haplotypes that have become extinct or very rare among the donors.
If R/L23- hasn’t been so diffused in Italy, where probably Jews took it, now we have only Jewish R/L23- and we’d think, like Vizachero, that this is a Jewish haplotype, that probably isn’t true.

Vincent said...

If R/L23- hasn’t been so diffused in Italy, where probably Jews took it, now we have only Jewish R/L23- and we’d think, like Vizachero, that this is a Jewish haplotype, that probably isn’t true.

First, don't presume to speak for me. You don't know what I think, as evidenced by the statement above.

Second, none of the clades of R-M269 (neither the L23- nor the L23+ group) are exclusive to any particular religious or geographic population. We can observe that the frequency of the most upstream clades (as a percent of R1b1b2 overall) is greatest in SW Asia and lowest in NW Europe. Given the histories of both Italy and Judaism, the genetic evidence so far suggests that Italians were not a significant source of Jewish R1b1b2 nor were Jews a significant source of Italian R1b1b2.

VV

Gioiello said...

Excluding all the other haplotypes (what have to do Jews with the R-M269, R-M17, I-M253, H-M69 of the Bnei Israel from India, J-M318 of Jerba etc.) and concentrating on J-P58, it shouldn’t be difficult for the authors of the paper, who have the samples, to expand them and to try in the world their close relatives.
And perhaps it would be interesting to read and to meditate the book of James D. Tabor (The Jesus Dynasty): the father, the founder, could even be a Roman soldiers but of Phoenician or other origin.
“Tib. Iul. Abdes. Pantera. Sidonia. Ann. LXII stipend. XXXX. miles. exs. coh I. sagittariorum. h.s.e.” (Corpus Inscriptionum Latinarum XIII 7514) “Tiberius Julius Abdes Pantera / of Sidon, aged 62 / a soldier of 40 years service, / of the 1st cohort of archers / lies here”.
Let we lose the vicissitudes of Jesus, but a royal (or sacerdotal) dynasty can arise also in this way.

Kepler said...

"if you compare the means I have and those of the others"
Oh, the means you personally have?
Means as personal achievements?

Yes, I am Latin American.

It is in fact interesting to know history, specially if the scope goes beyond a couple of centuries or even a couple of thousand years.

On another topic: ever read Jared Diamond's book Guns, Germs, and Steel or Peter Watson's History of Ideas?


Maju, I will ask the archaeologist about the details, he told me very off record, they just got the financing. The project will last for 5 years beginning now, so we may have to wait a wee bit.
I am sure there are lots of people who will become hysterical with the results, whatever they may point at, many want to use them for political and not scientific purposes. :-)

Gioiello said...

Kepler, what a strange a Venezualan with a German nickname! "Means" in Latin is "media", and my word was plenty of understatements. But are "means" also to have laboratories, funds to do these researches and to write these papers. And are "means" the paper, the forums, and the possibility to ban someone.
There is only one man in the world who knows me and who can write to me in this way. Are you sure to be hg. J? Have you perhaps confused J with I?

Kepler said...

Gioielli,
It is not strange at all to use a German, a Spanish, an Indian, a sub-Saharan or an English name. Why should it?

It is not "I am hg J". I happen to have J2 as Y haplogroup.
Whatever.

Maju said...

Maju, I will ask the archaeologist about the details, he told me very off record, they just got the financing. The project will last for 5 years beginning now, so we may have to wait a wee bit.
I am sure there are lots of people who will become hysterical with the results, whatever they may point at, many want to use them for political and not scientific purposes. :-)
-

Yah, that's always a risk, specially in Jewish ancestry issues due to political constrains. I just hope that the raw data is available whatever the conclusions, so we can judge on our own. Otherwise it'd be a waste of resources.

Kepler, what a strange a Venezualan with a German nickname! -

LOL. Kepler was a universal great man. Germans can be proud of him, of course, but all humans can as well. You don't need to share ethnicity in order to admire such great people. Think of Jesus, Gandhi, Marx, Leonardo, Einstein, Che, Buddha, Mandela, Plato, Sun Tzu or even Jimmy Hendrix. People from very different ethnicities identify with them.

Gioiello said...

Maju, I'd ask you to read what I usually write beyond the literal meaning. As I have suspected that Kepler was really Johan Hagman, for this I said that. He says I am wrong and I take note of it. On another forums I was accused to not respect the privacy, but I think that to know who he is in front of us is fundamental also for understanding what he does mean. For this I refused to enter again in the forums I was banned with another name. Only in one forum I use a nickname, but everyone knows that am I. Excuse me for this my oddity. The geniuses are geniuses, whichever they live, even though, if you permit, all the persons you enumerated aren't for me all geniuses, but I think that Buddha is probably the greatest, also if his followers are perhaps interpreting him on the contrary.
But certainly someone will call us to the topic, and I think he is right.

Gioiello said...

What does Vizachero (or whichever other) think about these two modals? The markers are those of the paper of Hammer et alii (2009).
15, 11-14, 12, 14-30, 24, 11, 13, 12, 12, 12, 12, 25, 19, 28, 11, 11, 15, 16, 9-9 (haplotypes 2049, 1406, 1397, 2046, 1403, 1512)
15, 11-14, 12, 13-29, 24, 10, 13, 12, 12, 12, 12, 25, 19, 29, 11, 11, 15, 16, 9-10 (Ysearch: KV7Y2 Gioiello Tognoni, XG6HX Giancarlo Tognoni, N92C5 Anonymous Brazilian).
They have a MRCA about 1400 YBP. During the 6/7th century probably a little community of Jews was running across Italy to the Rhine Valley. My question is the same: am I a Jew (genetically) or are those Jews Italians?
Otherwise we must change all our theories. Some haplotypes may coincide, but not for relatedness, but for convergence, then every theory has no value. We can never converge to a MRCA, but, as I have said many times in the past, we are picotant (would say French) around the modal (but this, as I have also said, makes impossible, or at least very difficult, anyway older) to calculate any time re the origins.
There would be the possibility of a rare SNP, whom we call “private”, but in this case very useful. I have one (S136) and I am waiting that my far relative Giancarlo Tognoni comes back from the Dolomiti to test him to begin to know when this mutation happened.

Anonymous said...

Italians are a Mediterranean Caucasoid people, excluding various Jewry who live in India, China or other exotic places, most Jewry are a Mediterranean Caucasoid people. There is bound to be a large overlap between the two peoples without any interbreeding or comparing dna. A study of the bones of indigenous Canarians, before European immigration, had a significant rate of J-M267. J1, of which J-P58 is the predominate variety, and with DYS388>15 the majority within J1, it seems J1 is ancient and had spread into Northwest Africa long before Phoenicians, Jews or Arabians. The same can be said about Mediterranean Europe. J1 has been in Europe long before Judaism was invented in the Levantine area. Cohens have a lot of J1 compared to the Israel caste of Jews. So what, and congratulations! Muhammad probably was J1 also. A good chance that Saddam Hussein and Bin Laden was and is also J1. It is not statistically significant that ethnicities with an eastern Mediterranean origin have significant J1.

Gioiello, having a J1 haplogroup does not mean you are descended from a Jew or Arabian, or Ethiopian Amharic speaker or Tunisian or Caucasus dweller or Sudanese African. There are many centres of significant J1 in Asia, Africa and Europe, central east Italy for instance. The haplogroup is ancient, and if you closely match someone with your STRs, it does not mean you or he are Jewish, just that you are related paternally many generations ago. I am J-P58 and I only match with one man whose origin is in Calabria Italy. Yes Jews were there. Where in the world haven't Jews wandered? Our J1 has not been found in Jews or Arabians or Caucasus dwellers or Ethiopians or Tunisians or Sudanese Africans or in men from Turkestan. Saying that it does not mean I don't have a recent Jewish or Arabian ancestor, just that it is improbable. Most J1 SNP values look very similar on first view but when working out genetic distance you find vast differences.

As I said before, Cohen Jewry may be from Judea or just as likely from Europe or Ethiopia or any other source of J1. Turks have significant amounts of J1, they could have got it from them.

Maju said...

^^ I think Gioello is talking of extremely close haplotypes within J1, rather than about J1 itself. Of course it could mean that he has Jewish ancestry, unless that haplotype is very common precisely in Italy (for what he has not provided any data).

Muhammad probably was J1 also.

Why? You have less than 50% chance for Saudians to be J1 (notice that Bedouins, even if artificially placed in Saudia in fig. 4 are actually from southern Palestine: Negev Bedouins). Possible is different from probable.

A good chance that Saddam Hussein and Bin Laden was and is also J1.

Hmmm... while Bin Ladin was Yemeni, with a surname that seems to mean "son of the Latin" or "Ladino" (Ladino meaning Sephardic Jew from Spain) there are a lot of chances that he's not J1 (while most Yemenis are indeed, Sephardic Jews have only some 20% of J1, per Adams'08).

terryt said...

"If we consider the archaeological evidence, Jews were already present at Beit Shemesh c.1300 BCE, what means that the semi-mythical Moshe and Aaron lived (if they actually existed) before that date".

They have to be mythical, otherwise the vast majority of Cohanim would carry a Y-chromosome haplogroup that can only have begun diversifying around that date. Aren't Moses and Aaron claimed to have been brothers? Therefore the same Y-chromosome.

"While Cohanim from diverse backgrounds carry a total of 21 Y chromosome haplogroups, 5 haplogroups account for 79.5% of Cohanim Y chromosomes".

Twenty-one Y-chromosome hapolgroups! A basis in any sort of myth is inadequate to account for that. And that's just in the Cohanim. What about the Jewish population as a whole?

"I am sure there are lots of people who will become hysterical with the results, whatever they may point at, many want to use them for political and not scientific purposes".

Isn't it amazing that in spite of the overwhelming evidence several major current belief systems are still based on the idea that both Jewish and Arab people descend on the paternal side from just one man, Abraham. On the maternal side Jews descend from his wife and Arabs from his wife's servant (I think). This belief continues to justify what I consider to be unjust political purposes. How do they get away with it?

"Religion is one thing. Reality and facts another".

Maju said...

They have to be mythical, otherwise the vast majority of Cohanim would carry a Y-chromosome haplogroup that can only have begun diversifying around that date.

That's what this paper would seem to state on first sight based on what they say about the J1-CMH. However they provide no neighbour-joining tree for this central haplotype, what has got my eyebrows quite high.

Anyhow, even if there's some truth to the myth there could be other ancestry perfectly well. Mythical ancestors are not real ancestors... even when they are real up to a point. Their main role is to forge an imaginary sense of unity, of kinship, for the group affected.

Aren't Moses and Aaron claimed to have been brothers? Therefore the same Y-chromosome.

Yes. Mythically, all Kohanim descend from Kohen an alleged descendant of Aaron via strict patrilineage. Obviously that is not true for the majority of them (too many different lineages) but could it be true for some of them?

This paper does not clarify things enough.

How do they get away with it? -

Keeping people in ignorance, indoctrinating them from childhood, etc. For a lot of people it's just too difficult to break apart from what was imbued on them in childhood, much less if it's a mainstream belief in their cultural environment.

Vincent said...

Maju and Ponto, note that Gioiello is R1b1b2. Specifically he is L23+ L51-, like the Cohanim R1b1b2 haplotypes he mentioned: and millions of other Italians, Greeks, Turks, Arabs, Persians, etc.

But I think that Gioiello has overestimated his relatedness to these haplotypes. His own haplotype (KV7Y20) is a GD of 6 or 7 away from these R1b1b2 Cohanim, which suggests a common ancestor more like 2500 years ago than 1400. And with the wide confidence intervals associated with so few markers, I see no reason to call special attention to it.

VV

Gioiello said...

Ponto, I agree completely with you when you are saying that such haplogroups are very ancient in Europe. I said this on the two forums which banned me, and you have had the same destiny. If you remember only a few years ago a Hg. J (1 or 2) in Italy was “Semitic”, a Hg. E in Italy was “Arab” etc. I have always supported the ancientness of the Hg. E (that is E1b1a not the more West Asiatic / European E1b1b) of Gabennesch from the Italian Tyrol (now I have surrendered, after 23andME, but I’d be curious to see the v3 of the Nigerian who is in the Adriano’s spreadsheet).
But my Hg. is R1b1b2a/L23+/L150+. On this we are discussing and I invite you to see the “R1blog” of Vizachero, and to exam his obstinacy. You have said which are the myths behind the peoples who based their history upon a faith, that, from what I have understood, nobody of us has, but which is the “faith” of Vizachero?

Gioiello said...

Maju and Ponto, note that Gioiello is R1b1b2. Specifically he is L23+ L51-, like the Cohanim R1b1b2 haplotypes he mentioned: and millions of other Italians, Greeks, Turks, Arabs, Persians, etc.

Vizachero writes: “But I think that Gioiello has overestimated his relatedness to these haplotypes. His own haplotype (KV7Y2) is a GD of 6 or 7 away from these R1b1b2 Cohanim, which suggests a common ancestor more like 2500 years ago than 1400. And with the wide confidence intervals associated with so few markers, I see no reason to call special attention to it”.
Another sample of the Vizachero’s faith. I have reconstructed the modal of the six Cohanim and mine (me, Giancarlo Tognoni and an Anonimous Brazilian who is on SMGF) and I have correctly compared the modals and the dated are mine.

Gioiello said...
This comment has been removed by the author.
Gioiello said...
This comment has been removed by the author.
Vincent said...

Gioiello wrote: Another sample of the Vizachero’s faith. I have reconstructed the modal of the six Cohanim and mine (me, Giancarlo Tognoni and an Anonimous Brazilian who is on SMGF) and I have correctly compared the modals and the dated are mine.

It is not a question of faith but of science: if you estimate the TMRCA for a pair of modals, you are not estimating the TMRCA of the modern men but rather the TMRCA of the ancestors.

In other words, the two modal haplotypes are being assumed to represent the two ancestors: the Tognoni ancestor in the first instance, and the Cohanim ancestor in the second. When you take the TMRCA between the two modals, what you get is the TMRCA between the two ancestors.

Assuming that estimate really is 1400 years ago, you also have to add the time at the ends of the two branches (which you removed when you made the modals) , from proto-Tognoni to Gioiello in the first instance and from proto-Cohanim to the modern Cohanim in the second instance.

Gioiello said...

Yes, I thank you. And if you calculate this do you reach 2500YBP? But the mutations are on average and one can have had mutations on the markers examined but not in others. I calculated the TMRCA with Sean Silver, on much more markers, and the time was this. If you compare these Cohanim with Giancarlo Tognoni or the Anonymous Brazilian, you have a different GD.
If you give me the Science I am glad and I'll be grateful to you.
But I would say that also the "modal" is an artifice, as we choose the haplotypes to compare, like when we compare us with someone found on a database just as it is the closest. For this I think that markers shall go with a paper trail, and now we have always more SNPs that we hadn't in the past.

Gioiello said...

There is believable to think that the 6 Cohanim have a common ancestor? Two are in the modal (1397/Poland and 2046/Belarus: 15, 11-14, 12, 14-30, 24, 11, 13, 12, 12,12, 12, 25, 19, 28, 11, 11, 15, 16, 9-9), one (1406/Latvia) has 1 mutation: DYS439=12→13, one (2049/Poland) has 1 mutation: DYS458=16→17, and two (1403/Russia and 1512/Russia) have 3 mutations: DYS389II=30→31, DYS393=12→13 and DYS439=12→11. Everything is possible, but they could belong also to different lineages.
To answer the question of Vizachero I can say that I and Giancarlo Tognoni coalesce in Antonio del Badia, who lived in the 15th century and had two sons (Piero e Michelangelo) who are our ancestors. Then if I must add to 1400 yet 600 years I arrive to the time of the First Jewish War, when Romans destroyed Jerusalem.
Also with the descendants of Abraham Senyor, a Spanish Jew converted to Christianity after 1492, I coalesce in the same time. The question is always the same: are I a descendant of Jewish slaves carried to Italy in that time or are these Jews descendants of Roman soldiers?

Katharós said...

Hi,
I am interested in the opinion of the posters here to what degree Haplogroup J2 in general can be associated with ancient Semitic speaker in the Near East? Especially in context of Bronze Age and later population movements from Asia Minor into the Fertile Crescent. Namely Hurrians, Mitannis and Hittites all of non-Semitic stock. To what degree did these populations contribute to J2 pool in the Fertile Crescent?
The Old Testament also frequently mentions Semitic and non-Semitic people attacking and settling in Canaan like them self’s. Most notable are the Hittites Indo-Europeans who are quite often named in the Old Testament. For instance the Book of Genesis mentions intermixing between Hebrews and Hittites.

Genesis 26:34: And Esau was forty years old when he took to wife Judith the daughter of Beeri the Hittite, and Bashemath the daughter of Elon the Hittite:
35: Which were a grief of mind unto Isaac and to Rebekah.

Genesis 27:46: And Rebekah said to Isaac, I am weary of my life because of the daughters of Heth: if Jacob take a wife of the daughters of Heth, such as these which are of the daughters of the land, what good shall my life do me?

Genesis 36:2: Esau took his wives of the daughters of Canaan; Adah the daughter of Elon the Hittite, and Aholibamah the daughter of Anah the daughter of Zibeon the Hivite; 3: And Bashemath Ishmael's daughter, sister of Nebajoth.

The second group are the Habiru a diffuse band of people but who have recently bin linked to Anatolian Hurrians thanks to the Tikunani Prism.

http://en.wikipedia.org/wiki/Habiru
http://en.wikipedia.org/wiki/Tikunani_Prism

Maybe a coincidence but the age of the haplotype J1e corresponds with the Egyptian Merenptah-Stele (1213 to 1203 BC) which first mentions a nomadic tribe called Israel in the land of Canaan.
http://en.wikipedia.org/wiki/Merneptah_Stele

I don’t have full access to the work of Prof. Anson Rainey - Tel Aviv University but Prof. Anson Rainey claims that ancient Israelites did not originate in Canaan but migrated there as pastoralists from east Jordan, so again J1e comes to mind.
"Early Israelites came from Transjordan." Biblical Archaeology Review 33, No. 3 (May/June, 2007), p. 78.
http://www.tau.ac.il/humanities/archaeology/directory/dir_anson_rainey.html
http://theologicalscribbles.blogspot.com/2009/01/revolting-peasant-theory-of-israelite.html

I wound not be surprised if actually the Hebrews and Israelites originally represent to different groups of people one Semitic and the other one non-Semitic in origin maybe the Habiru/Hebrew.

Maju said...

Maju and Ponto, note that Gioiello is R1b1b2. Specifically he is L23+ L51-, like the Cohanim R1b1b2 haplotypes he mentioned: and millions of other Italians, Greeks, Turks, Arabs, Persians, etc.

Indeed. I got confused with all that J1 talk, sorry.

The R1b graph does exist in this paper and in it Jews (Kohen specifically) are just random dots in a blue sea symbolizing others. So no Jewish ancestry in principle, rather Goyim ancestry for the affected Kohanim.

...

I am interested in the opinion of the posters here to what degree Haplogroup J2 in general can be associated with ancient Semitic speaker in the Near East? -

Not particularly. Northern Semitic speakers of Lebanon, Syria and Iraq do seem high in J2a but this marker is shared with non-Semitic peoples like Turks or Iranians and we can't forget that Iraq at least, among those Semitic areas, was not Semitic-speaker early on in history.

J2b in turn seems rather high towards Europe and South Asia and is low in West Asia.

IMO, J1 and E1b1b1 seem associated with the spread of Afroasiatic languages in general and to some extent J1 in West Asia could be associated with early Semites, if you agree with the Harifian and later circum-desertic pastoralist complex as the origins of Semitic ethnogenesis. The issue is not wholly clear anyhow but it's my favorite model.

Especially in context of Bronze Age and later population movements from Asia Minor into the Fertile Crescent. Namely Hurrians, Mitannis and Hittites all of non-Semitic stock.

I don't think these movements were too moved in fact. None of those really controlled much of Lowland West Asia for long. I'd rather asociate the modern spread of J2 to Neolithic processes original of the Zagros-Taurus arch (while J1 would be more associated to the Palestine area). For example, we know that proto-Sumerians appear to have originated in the Zagros area (Jarmo, later Samarra), so that would directly connect Southern Iraq to the Golbeki Tepe area and so on.

Vincent said...

Maju wrote: The R1b graph does exist in this paper and in it Jews (Kohen specifically) are just random dots in a blue sea symbolizing others. So no Jewish ancestry in principle, rather Goyim ancestry for the affected Kohanim.

I think I discussed this earlier, but the Cohanim R1b1b2 haplotypes in the paper are nothing like a random subset. Of the 13 haplotypes, one set of six are very closely related to each other and another set of three are very closely related to each other. In short, there is a tremendous excess of GD<5 among the Cohanim haplotypes.

I can't say WHY the MJ network in the paper is wrong, but I can say for sure it is not showing the real picture of how these R1b1b2 Cohanim relate to each other and to non-Jewish R1b1b2 in general.

I'm inclined to say that Fluxus-based MJ networks with so few STRs are simply worthless, but that'd probably be too broad.

VV

Katharós said...

True the Zagros area seems to be the cradle of unique pre-Semitic and pre-Indo-European people like the Sumerians, Kassites and Elam and God knows who else.
I don’t follow Anthro/Genetic sites regularly but I also noticed that J1 seems to be a loyal follower of E1b1b1 at least in the Afro-Asiatic realm, maybe reflecting a very early interaction between E1b1b1 and J1 in the Sinai Peninsula between Palestine and Egypt.

AWood said...

According to the book The Early History of the Hebrews, there were pastoralists that extended all the way from Northern Africa and into S. Asia along the coast. These people would have been commonly called Amorites, or potentially many tribes of similarly related peoples. They were documented by the Egyptians with similar phenotypes to the ancient Celts. (tall stature, red haired) Eventually these people were assimilated into other tribes along the Canaan. They may have been Semitic speakers, or possibly not. In any event, I think these men are strong candidates for early R1b we find there today.

Maju said...

I think I discussed this earlier, but the Cohanim R1b1b2 haplotypes in the paper are nothing like a random subset.

Can you direct me to the exact paragraph, graph or wherever you find that? I have just re-read the paper before posting this, just in case I missed something in my previous readings, and either it's some blind spot or there's just nothing in the whole document or supplementary material that says otherwise.

All I can see is figure S4, where all yellow dots (Kohen) are scattered in the wider network. All are singletons except for a slightly larger dot. Some of the yellow singletons are interconnected but still the graph shows at least 7 different R1b Kohen lineages, all with few members and each lineage more closely related to non-Jews than to other Kohanim.

I find no other reference to R1b haplotypes in all the document. You must have access to data I can't see.

...

I don’t follow Anthro/Genetic sites regularly but I also noticed that J1 seems to be a loyal follower of E1b1b1 at least in the Afro-Asiatic realm, maybe reflecting a very early interaction between E1b1b1 and J1 in the Sinai Peninsula between Palestine and Egypt.

I do think this has to do with the Afroasiatic (or at least Northern Afroasiatic) ethnogenesis and surely happened in Egypt or surroundings. Per Cruciani 2004, while J1 as a whole appears as West Asian, a branch including the bulk of haplotypes, both in North Africa and West Asia is linked by a North African singleton (and includes almost all North African haplotypes). While it's not conclusive, it does suggest tight interaction North Africa - West Asia, interaction that must have happened at the Mesolithic/early Neolithic period. It's surely within that Egypt-Palestine interaction that E1b1b1 migrated to West Asia as well.

Maju said...

According to the book The Early History of the Hebrews, there were pastoralists that extended all the way from Northern Africa and into S. Asia along the coast. These people would have been commonly called Amorites, or potentially many tribes of similarly related peoples.

Into South Asia? I doubt it very much. The easternmost border of historical Semitic presence is Iraq and the Persian Gulf (barring possible traders).

The term Amorite is ambiguous. Initially they seem the same as Semites in general (Amurru, Amaru), later same as Western Semites or Syrians and finally, in Jewish texts, as non-Hebrew Semites, interchangeable with Canaanite.

They were documented by the Egyptians with similar phenotypes to the ancient Celts. (tall stature, red haired).

Please show me such "documents". Checking Wikipedia only the Bible says that they were unusually tall ("giants") but no mention of "red hair" (which anyhow was surely rare among ancient Celts: it's specially a British/Irish phenomenon, rather than Central European). It seems such Nordicist ideas were elaborated in the romantic Eurocentric frenzy of the late 19th century but eventually abandoned (but for the Nazis).

In other words: your book seems just neo-Nordicist junk.

Gioiello said...

The Authors write: “Our estimate of the age
of J-M318 may be more reliable because this SNP represents
a terminal mutation within the M410 sub-clade of the
J2 branch (Fig. 1). Originally discovered in a single Libyan
Jew (Shen et al. 2004), we find the derived allele at M318
to be present in 16 individuals in this survey—13 of which
are Cohanim from Tunisia/Libya or the island of Jerba (the
remaining 3 samples come from Tunisian or Libyan Jews
who did not have information on their Cohen, Levite, or
Israelite status). The much younger estimated divergence
time for the J-M318 haplogroup (1.3 § 0.5 and
1.9 § 0.8 kyears for the 17- and 9-locus datasets, respectively)
(Table 1) suggests that either the M318 mutation (a)
arose within the Cohanim population of North Africa, (b)
expanded within this community following migration of
a founding J-M318 Cohen from another geographic location,
or (c) became incorporated into the Cohanim patriline
via conversion, adoption or non-paternity. The first of these
possibilities (a) is supported by the fact that the M318
mutation occurred on the M410 background (Fig. 1), and
median-joining network analysis links the cluster of Cohanim
J-M318 chromosomes to that of the Cohanim J-M410*
chromosomes (i.e., rather than to other J-M410* chromosomes
from North Africa and the Near East) (Figure S5).
The high frequency (»60%) of the otherwise rare J-M318
haplogroup in our sample from the island of Jerba may be
the result of an ancient founder effect in this Jewish isolate,
which is thought to be descended from one of the earliest
Diaspora communities that left the Middle East before the
destruction of the second Temple in 70 A.D. (Tessler and
Hawkins 1980)” ( pages 8-9).

But M318 is downstreem not only M410*, but also M24, M25, M26, M27, and this is
the Hg. J2 more diffused in Europe. On “Genealogy DNA” Bonnie Schrack noted this,
but had she nothing to say?

Gioiello said...

If the Authors would take care to search the modal of the Cohanim J-M318 on SMGF, they’d have easily find overall Europe its origin at the GD of 2 onwards. Among them many Italians, some of them I extracted in the past and put on Ysearch. Then they could compare also with 43 markers.

Vincent said...

Maju asked Can you direct me to the exact paragraph, graph or wherever you find that?

What I wrote is based on my examination of the data (i.e. the actual haplotypes from the paper). You are right that the authors don't treat the R1b1b2 with much detail, which is frustrating but understandable given the J-P58 focus.

All the haplotypes are extracted, if you want to analyze them:

http://spreadsheets.google.com/pub?key=thQFY_kLNMPE3v7TVSH1QVw&gid=4

You can see the graphic results of some analysis (though more is possible) here:

http://r1blog.blogspot.com/2009/08/r1b1b2-cohanim.html

The first graphic is a table of GD for the R1b from the paper. Note that the two clusters are very distinct against the remainder.

The second graphic shows the GD within the Cohen, within a random set of R1b, and the Cohen vs. the random. The chart is easier to interpret if you made it (:-) but you can see the excess of GD<5 in the Cohanim relative to a "random group of singletons".

The Cohanim R1b in this paper are picking up several sources, most likely. One source is a Near Eastern source, with properties nearly identical to those of the J-P58 "Cohen Modal" group - specifically, one or two families with a long tradition of being Cohanim reaching back 1000 to 2000 years or more. Another source is an overlay of "random" R1b, probably resulting from conversions, singleton lines, or other genetic interactions.

But I always think that we should, when possible, examine the data for ourselves. Often it tells us things the authors just didn't see.

Gioiello said...

Ponto, which necessity to speak about a ”left handed lesbian born in Wales”? You know I am not of mother tongue (moreover I can read more than a dozen languages beyond the classic ones!) and often I am not able to calculate the reach of such sentences. Other times I avoided to answer you, thinking your expressions as provocative. If you are curious about the “oversized genital appendages” of the Middle Easterners, you can read on Wikipedia the entry “penis”, and you will see that they are ridiculous.
Probably you have an identity problem, being as Maltese in the middle, then in a no land. Don’t forget that all my problems with “dna-forums” began when I said that Maltese are above all Italians (genetically I meant), and a Poland, one of those who have an oversized faith we haven’t, stirred up the polemic (in Greek the “war”). Don’t do like that Greek, I predicted his “Italianity” (for his paper trail and other), and after his 23andME ceased to write to me: he was pretty more Italian than me. The same I could tell you. Do 23andME and watch where your jack falls. You said you find the closest to you in a Calabrian. Wasn’t I to tell that Maltese are above all Calabrians and Sicilians, but a great geneticist like Cristian Capelli. If Dienekes thinks that Calabrians and Sicilians are Greeks, then Maltese are Greeks. You said that the variance of J1 in Italy is higher than the Middle Eastern one: at least J1 is in Italy from many thousands years.

Maju said...

Thanks Vincent. Those spreadsheets are very useful. Hopefully I'll be finally able to get an idea of whether the alleged J1e-CMH is such thing or just a convenient fantasy. Sadly Cohen and regular Jews are not differentiated.

But, in any case, I fail to see any sort of structure in the handful of R-269 haplotypes. Just a quick look makes them appear wildly different, just like in the graph.

Reading your blog link I just make no sense at all: you say that Kohanim R1b is less related among them than West European R1b, exactly as would be expected with any random sample of West Asian R1b, where the lineage is older and more diverse. So when you say that the common ancestor is some 4400 years old, you are essentially claiming that the common ancestor of all R1b anywhere is that recent, more or less.

Comparing with Alonso's haplotypes, Ashkenazim look mostly either the basic modal R1b1b2a or its most common Central European variant (see: http://leherensuge.blogspot.com/2009/03/european-and-anatolian-r1b-structure.html). Actually they look more European than even I expected. Instead Middle Eastern Jews look like the Anatolian most common haplotype in most cases. There are a few exceptions but they do not mend anything: they just belong to rarer haplotypes.

I'm sure now that, if you repeat your exercise using only Ashkenazim, you'll get a curve similar to that of random West Europeans.

Vincent said...

@Maju;

Sadly Cohen and regular Jews are not differentiated.

Table S3 contains only Cohanim. No haplotypes from non-Cohanim were published by Hammer et al.

Just a quick look makes them appear wildly different, just like in the graph.

Then take a longer look. Six of the haplotypes are pretty closely related to each other, and another three are closely related to each other. Much closer than if you were just randomly picking R1b1b2 people off the street

Actually they look more European than even I expected. Instead Middle Eastern Jews look like the Anatolian most common haplotype in most cases.

Not sure how you get that. It appears that 80% of these Cohanim R1b1b2 are L23+ L51-, which is less than 2% of European R1b1b2.

And I'll grant that this set of Cohanim R1b1b2 looks more Anatolian than, say, Irish. But Levantine, Arabian, or Persian R1b1b2 doesn't really look much different than Anatoalian R1b1b2. Not enough to call it one or the other on the data we have.

VV

Maju said...

I've been looking at the other spreadsheet, the one tagged "Hammer 2009 (paper)". Mostly because the DYS are sorted numerically.

Checking the S3 spreadsheet, the one tagged "FTDNA", I see no substantial differences. Using the comparable DYS 19-391-392-393

1. Dutch and one of Russian Kohanim are modal R1b (Alonso's HT1, typical of West Europe but also found in Anatolia)
2. Iranian, Kurdish and Tunisian ones are in the most common Anatolian haplotype family, that I consider the possible root.
3. The other Russian is in HT2, common in Central Europe. The Polish and Latvian ones are one marker away (a rarer clade surely but probably some Polish Ashkenazi founder effect)
4. The Palestinian ("Israeli") ones are the more different ones but still one is within the haplotype most common in Croatia by Alonso'05.

Whatever the case, they seem to belong to at least five or six different lineages, from both West Asia/SE Europe and West/North Europe types - and this only considering four markers.

Compare the Kohanim to a real sample of European and West Asian R1b and you'll see that they fit in almost "anonymously". They may have some local founder effects, specially the Lithuanian-Polish ones, but that's about it.

Vincent said...

@Maju: Checking the S3 spreadsheet, the one tagged "FTDNA", I see no substantial differences. Using the comparable DYS 19-391-392-393

If you are looking at just four STRs you are surely missing the real story.

A bunch of "random" draws from the R1b1b2 pool would produce a pairwise GD matrix (over 22 markers) with something approximating a gaussian function with an expected value of 8. That's the red distribution in the blog I pointed to earleir.

That's not at all what the Cohanim look like. Rather, with some noise due to the small sample size, it appears there are actually several distributions overlayed: possibly one with an expected value of 1, one with an expected value of 3, and a third with an expected value of 11.

And I insist that calling these L23+ L51- haplotypes "Anatolian" as opposed to something less specific is a mistake. Not only is this form of R1b1b2 not exclusive to Anatolia, it is not at all certain that Anatolia is a good guess for its origin point.

VV

VV

Maju said...

I am looking at four DYS because they are comparable with the Alonso's study of 2005, so far the largest I know of European, Berber and Anatolian R1b by haplotype.

Alonso used five DYS in fact but, for some odd reason, DYS 90 is not used by either FTDNA nor this paper it seems. Still, I have succeeded in previous attempts (see link above) in correlating this 5-DYS structure with Ht35/R1b1b2a1 duality. Per the FTDNA Ht35 project there are just no L11- with DYS-393=13, which seems to be indicator exclusively R1b1b2a1 (guess there can be exceptions but for what I've seen in real haplotypes tested for L11 it does apply universally).

All Ashkenazim with R1b1b2 are DYS-393=13. And do fit otherwise well enough with the R1b1b2a1 subhaplotypes. Iranian, Kurdish and Tunisian Jews are DYS-393=12 and fit in the root haplotype, common in Anatolia. One of the Palestinian Jews is also DYS-393=12 and the other markers suggest close affinty with a derived haplotype found specially in Croatia in Alonso'05. Only the other Palestinian Hebrew is an oddball, with DYS-393=13 but the other comparable markers different from any of the West European haplotypes... (except from a rare and porly linked one found in Iceland, from which it differs at DYS=391).

Don't make abstract curves and, specially, do not mix Jews from West Asia with Jews from Northern Europe, as they are obviously quite different. That will produce a curve that reflects total R1b1b2 distances, more or less. No wonder that your Jewish curve is more different than the one for West/North European non-Jews, all in a subset of R1b1b2 known as R1b1b2a1.

Compare the haplotype of these two subsets separately and not just with themselves but also with the control groups of their respective origins. If your maths are good (and I have no reason to doubt it) you should get that Ashkenazim are almost identical to Central European non-Jews (actually less diverse but that's logical considering both history and the size of the sample) and that Middle-East Jews are within the variability of the area.

I don't have the resources to do all that myself but the fact that this paper admits that in its NJ graph, reinforces my certainty.

And I insist that calling these L23+ L51- haplotypes "Anatolian" as opposed to something less specific is a mistake. Not only is this form of R1b1b2 not exclusive to Anatolia, it is not at all certain that Anatolia is a good guess for its origin point.

Well, "Anatolian" is a catchall term for the haplotypes that are most common among Turks and Armenians (and almost non-existent west and north of the Balcans). I use "Anatolian" instead of "West Asian" because I have no info on lowland West Asia (and for what I see here in regard to Palestinian Jews, it might be somewhat different). Whatever the case, Anatolia or somewhere not far away must have been the center of spread of R1b1b2, as it's there where the highest R1b diversity has been found, it seems (and also because it's a natural bridge, together with the Balcans, between West Asia and Europe).

Gioiello said...
This comment has been removed by the author.
Gioiello said...

Re the origin of R/L23-, for what I said above, and waiting that someone finds another R/L23+/L150- like the Italian Romitti (this is the necessary link between this haplotype and R/L23+/L150+ and with all subclades), I invite Vizachero, or whichever else, to calculate the variance between the Jews in his ht 35 project and the Italians. He can add now Ysearch N4Y6V (Cervino from Sala Consilina), who finds his closest cousin in Fiozzo at a GD of 6 over 32 markers. He has also a North African (there were many centuries of Roman Empire) and a Middle Eastern.

Gioiello said...

I thank Argiedude for having answered my question on "dna-forums":

"The Ashkenazi cluster is hard to mistake because it's variance is very tight and it has 392=14. The European cluster is more prone to errors. But I used 2 STRs that combined are almost unheard of in other R1b haplogroups. They're 393=12 together with 426=11. Even in other haplogroups in the ht35 Project, this combination is completely absent. And amongst 1200 samples from the U106 and L21 Project there wasn't a single sample that had both values simultaneously. All 12 SNP-tested M269+ L23- in the ht35 Project have both these values.

Further confirming the accuracy of the prediction is the fact that 75% of the European modal samples had 390=25 or greater, a trend already visible in the ht35 Project samples (that weren't Ashkenazi). In contrast, only 11% of R1b in the L21 and U106 Project had 390=25 or greater.

Also notice that the European modal has a very high frequency of 385=12/14. It's about 1/3 of all the samples. In the U106 and L21 Projects, only 5% of the samples have 385=12/14. [The Ashkenazi cluster has only 1 out of 60 samples with 385=12/14]

The variance of the European cluster is 3 times greater than of the Ashkenazi cluster.

Anyone who knows of a study that tested DYS426, I'd be glad to know. I already included the South Tyrol study from Pichler (2006). He found 2 M269+ L23-.

edit: There are 2 M269+ L23- from north Italy, one confirmed, the other almost certain because of his last name. There are some 200+ ysearch samples from north Italy, so that gives us a rate of 1% for north Italy. The percentages for Italy seem to be fixed at 1%, in South Tyrol, in north Italy proper (the Po valley), in the rest of mainland Italy, and in Sicily".

Anyway the percentage in Italy is higher, not only for this new N4Y6V, but for many surnames that seem Italian among Migrants, and my theory is that R-L23- comes from the Italian refugium. We shall see.

Katharós said...

@Aaron
In the cemetery site of Beni-Hassan a group of Aamu lead by a Sheykh Absha “wiki”, are pictured.
http://commons.wikimedia.org/wiki/File:Beni-Hassan-Asiatiques1.jpg
http://commons.wikimedia.org/wiki/File:Beni-Hassan-Asiatiques2.jpg

The tomb of Sebekhotep a senior treasury official of the reign of Thutmose IV (about 1400-1390 BC) also pictures Semites.
http://www.britishmuseum.org/explore/highlights/highlight_objects/aes/f/fragment_of_painted_plaster_-4.aspx

http://www.britishmuseum.org/explore/highlights/highlight_objects/aes/f/fragment_of_painted_plaster_-3.aspx

Very detailed Fragment of a battle scene showing Semites. Know idea which groups are being shown.
Thebes, Asasif, Dynasty 18, probably reign of Thutmosis IV (ca. 1400-1390 B.C.)
http://www.metmuseum.org/explore/newegypt/htm/wk_frag.htm

Mural with preserved colouring from the Sumerian/ Amorite city Mari in Syria. I don’t think these men are Semites based on their head dress.
18th century BC.
Louvre, Paris
http://img524.imageshack.us/img524/2885/syria1.jpg

Vincent said...

@Maju: I am looking at four DYS because they are comparable with the Alonso's study of 2005, so far the largest I know of European, Berber and Anatolian R1b by haplotype.

No matter the reason you use only four. If you use only four, you will not get an intelligent analysis. No need for you to repeat Alonso's mistakes.



Per the FTDNA Ht35 project there are just no L11- with DYS-393=13, which seems to be indicator exclusively R1b1b2a1.

True, but the Hammer paper gives us two of them (from Russian Ashkenazi).


All Ashkenazim with R1b1b2 are DYS-393=13.

That's not true, either of the Ashkenazim in the Hammer paper (50% are DYS393=12) or of Ashkenazim in general (see the Jewish R1b project at FTDNA).


Don't make abstract curves and, specially, do not mix Jews from West Asia with Jews from Northern Europe, as they are obviously quite different. That will produce a curve that reflects total R1b1b2 distances, more or less.

First, the curve is not "abstract". It is a real and concrete representation of the properties of the data. Science demands analysis, not merely the waving of hands. If you are going to carry on with this discussion, and I hope you will, then I implore you to really explore the data.

If you do that, you can only reach the conclusion that I did. That is that Jewish R1b1b2 is not merely a subset of central European R1b1b2. That the R1b1b2 Cohanim have (at least in part) a haplotype structure similar to the J-P58 Cohanim. And that the Hammer paper's short treatment of R1b1b2 was as much wrong as right.

VV

Gioiello said...

For another Italian R-L23- I tried to know more about him, also for his relatedness with a Brazilian with a Portuguese surname, but in vain.

Inviato: domenica 12 luglio 2009 15.02.56
A: sc_gen@yahoogrupos.com.br

Cc: rco2000@uol.com.br

Dear Ricardo (and colisteiros),
I have found on SMGF, and put on Ysearch (79UHC), a very interesting haplotype, that is probably a R/L23- (DYS393=12, DYS426=11 and DYS461=11), but with DYS390=26 and DYS464=15,15,15,16. Having DYS460=10 and DYSGATAH4.1= 11, I think is a particular cluster, like mine. It belongs to an Italian, named Filandro. This surname is very rare in Italy (about 20 families on the phone Director) and diffused in Calabria and other Southern Regions of Italy. The surname is explained by Greek: “who loves the man”. He matches at a GD of 1 over 43 markers a Brazilian, tested by SMGF he too (see now on Ysearch: RQ7K2). The most distant ancestor is Jose Antonio Rodrigues, born in 1850 in Soledade, Rio Grande do Sul. Among his ancestors he has also “Rita Ismeria De Oliveira”, born in Soledade she too.
I think that it would be interesting to know something more on the origin of this Rodrigues and on his relatedness with the Italian one, since this could throw some light on the origin of this original haplotype of the R haplogroup, we find so far above all among Italians and Jews.
With the best wishes, Gioiello Tognoni

Maju said...

True, but the Hammer paper gives us two of them (from Russian Ashkenazi).

You're right: a Latvian and two Poles in fact. I made the error (probable error) of conceiving them as derived from the 15-(24)-11-13-13 haplotype found in a Russian Ashkenazi but I did not realize that the mutation here is precisely in the crucial DYS-393 and that there is another possible (and more likely) root in the 14-(25)-10-13-12 haplotype (Ht35) found in a Palestinian Jew, the one I said as having a "Croatian" haplotype. This "Croatian" haplotype also has some presence in Turkey (per Alonso).

I thank you for this correction because it was very necessary, indeed, but also because it confirms my original idea of a Turkish (Hellenistic Anatolian) origin for a good fraction of Jewish ancestry, rather than Central European.

No need for you to repeat Alonso's mistakes.

There is need because I need something to compare with. We can't just analyze these Jewish samples in a vacuum: we need to compare Kohanim with regular Jews and both with other populations who are probable sources.

I really hate this emphasis in Jewish genetics that lacks of good comparisons. Such a complicated ethnos as Jews needs of many comparisons and, obviously, Palestinians, Turks and the various host populations, including those that converted en masse to Judaism, are necessary comparisons. If you don't do that you will reach to biased conclusions.

Vincent said...

@Maju: I really hate this emphasis in Jewish genetics that lacks of good comparisons.

I think this is not unique to Jewish studies, and a lack of good comparisons is truly rampant in published papers. It was a fatal flaw for the Semino et al. paper, and many others since then.

Today we are lucky, though, because the new paper on the Levant by Mirvat El-Sibai et al. has haplotypes for dozens and Near Eastern R1b1b2 and when added to the Zalloua ones (plus the Cinnioglu ones) we start to have a good basis for comparison. If we can segregate the R-P25* and R-M73, of course.

VV

AWood said...

Into South Asia? I doubt it very much. The easternmost border of historical Semitic presence is Iraq and the Persian Gulf (barring possible traders).

I meant to say SW Asian.

The term Amorite is ambiguous. Initially they seem the same as Semites in general (Amurru, Amaru), later same as Western Semites or Syrians and finally, in Jewish texts, as non-Hebrew Semites, interchangeable with Canaanite.

It is ambiguous, but these people apparently dwelt in the highlands and were pastoral nomads. In any event, I was not claiming that all would display these phenotypes, but that some might have. As you said there were probably comprised of many different people. We see high levels of R-P25 found among the Galilee region as well as coastal Carmel. It is clearly evident that R1b has been in the Levant longer than a millenia, and cannot be descended from Asian Turks.


Please show me such "documents". Checking Wikipedia only the Bible says that they were unusually tall ("giants") but no mention of "red hair" (which anyhow was surely rare among ancient Celts: it's specially a British/Irish phenomenon, rather than Central European). It seems such Nordicist ideas were elaborated in the romantic Eurocentric frenzy of the late 19th century but eventually abandoned (but for the Nazis).

In other words: your book seems just neo-Nordicist junk.


A. Sayce made this claim, although I have not yet seen the murals which supposedly demonstrate this, and I have been rigorously looking. I have seen murals comparing Nubians, Palestinians, Amorites, Syrians and Hittites. I am not sure they are 100% clearly labeled, but the one categorized as Hittite is clearly white skinned, and a couple others are yellowish, and the palestinian a brown colouring, nubians were black. source:http://commons.wikimedia.org/wiki/File:PalaceInlays-NubiansPhilistineAmoriteSyrianAndHittite-Compilation-MuseumOfFineArtsBoston.png

Maju said...

I think this is not unique to Jewish studies, and a lack of good comparisons is truly rampant in published papers. It was a fatal flaw for the Semino et al. paper, and many others since then.

Today we are lucky, though, because the new paper on the Levant by Mirvat El-Sibai et al. has haplotypes for dozens and Near Eastern R1b1b2 and when added to the Zalloua ones (plus the Cinnioglu ones) we start to have a good basis for comparison. If we can segregate the R-P25* and R-M73, of course
.

Do you have links? I have not read any of those, except Cinnioglu's. (Otherwise I'll make a search, I guess).

It is clearly evident that R1b has been in the Levant longer than a millenia, and cannot be descended from Asian Turks.

Of course. I just have no decent data for West Asia out of Turks/Armenians (and also Turks, their ancestors, are a likely candidate source for European R1b). In any case, we won't be able to understand R1b until we understand West Asian R1b.

Nubians, Palestinians...

Important: Philistines are not Palestinians. Philistines lived only around Gaza (and were possibly of Cretan origin - the elites at least). Palestine is a Roman name implemented long after the Philistines had vanished as a distinct people. Surely Philistines made a small apportion to historical Jewish and Palestinian blood but Palestine and therefore Palestinians appear only in Roman age, from mostly local stock (Judean, Samaritan, Galilean, etc.)

Essentially we can say with great likelihood that modern Palestinians are descendants of the Jewish commoners, who mostly converted gradually to Christianity and later Islam, plus whatever minor inputs they have received in these two millennia. In turn, these historical Jews would be mostly descendants from local Canaanites and ultimately from the Neolithic peoples of the area.

Philistines were just one of those minor inputs. Just that the name of the province was apparently inspired in them.

Vincent said...

Do you have links? I have not read any of those, except Cinnioglu's. (Otherwise I'll make a search, I guess).

The new paper is just out today (see the top of Dienekes blog).

http://www3.interscience.wiley.com/journal/122553140/abstract

The haplotypes are free in Supporting Informaton (tables S1 and S2).

VV

Gioiello said...

Argiedude has discovered a new Jewish cluster R-L23-. But:
1) Sephardic Jews are put together with a Cuban, Perez (Ysearch: R9BQA), whom nothing can link with Jews, and he has DYS389=14-28, that isn’t the same of the Sephardim 13-28: there are at least 2 mutation 13-28→14-29→14-28.
2) The Volkov are 3(but probably they are the same): Ysearch: MKGW2, 5MMP3, NDZNC, I have just put on Ysearch from SMGF, and nobody claims a Jewish ancestry.
3) The English Burney ( FP4EJ), like Crenson, has probably nothing to do with Jews.
4) Volkov, when I traced the descent from the Italian refugium based upon DYF385 (from 9-11 to 11-11 etc.), was the unique to have 11-13, then the most divergent. Then probably no Sephardic cluster, but an European one, diverged from the Italian refugium to East and West and that Sephardim has picked up elsewhere.

Gioiello said...

Perez has really DYS389=13-28. On YHRD this haplotype isn’t rare: 2 on 645 Czech Republic, 2 on 455 Central Bohemia, 1 on 200 Hungary, 1 on 40 Sweden, 1 on 62 Plzen (Czech Republic), 1 on 113 Turkey, 1 on 102 Algeria, 1 on 1239 Hispanic American.
Haplotype: DYS19=14, DYS389=13-28, DYS390=24, DYS391=10, DYS392=14, DYS393=12, DYS385=11-15. We haven’t DYS426, that certainly is important.

Gioiello said...

Similar haplotypes there are on the paper posted by Dienekes above (Coastal-inland differences in Y chromosomes of the Levant): Jordan 10AM71, 1AM60, 8AM71 and Iran 6AQ106.
The paper seems to demonstrate that hg. R expanded from Europe to North Africa and Middle East, being present on the coasts and diminishing when we go inside.

Gioiello said...

Other possible R-L23- on Ysearch are: EE3CC Weiner, G554J Stoll and BTPGJ Tischkevich (to deepen, having DYS426=11 but DYS393=13).

Gioiello said...

For the presumed Sephardi clade of Argiedude I think we must watch above all DYS460=10 and DYSH4=13 (these two markers determine probably a clade) that have Volkov, Burney, Watzke and Weiner, but no Jews.
Two markers are worth than one, and DYS426=11→10 is a random mutation that per se doesn’t determine a clade, as DYS389II=29→28 etc. It would be necessary to associate a SNP for having a clade, as who had a such SNP, if has descendants, can create a “modal” by the markers he casually had in that moment. So for DYS392=14: it isn’t a Jewish marker, being present also in Italians and others. It is that the ancestor of all those Jews had that value in that time when he gave origin to that descent. What is important is the TMRCA, ant that of Jews is always very recent.

Maju said...

Vincent: thanks for the link. I'm analyzing, as far as I can, right now.

Some of the R1b Kohen haplotypes could be Levantine (one of the Palestinians and the Latvian-Polish haplotype - but notice this last haplotype is found as far NW as Croatia anyhow). The Iran-Kurd-Tunisian haplotype seems more northernly, though has also similitudes in Syria/Lebanon (it's modal Asian R1b). The Dutch and Russian instead look European. I'm unsure re. the other Palestinian Kohen, because it's a unique haplotype within DYS-393=13 (should be R1b1b2a1).

...

Gioello: you should start your own blog. Even if your opinions are iconoclastic, so to say, it's clear that you know a lot, so it should be interesting.

Gioiello said...

Maju, I thank you for "you know a lot" and also for "iconoclastic" (this will enjoy Dienekes, being two Greek words), but I haven't the means to do so, and soon I will begin to work again, and I won't have the time too. I prefer to write on some blog of someone else, until he will ban me.
But I like only demonstrations and not ideology (two other Greek words).

Maju said...

You can have a blog or a dozen blogs for free in Blogger. You can even make money with them... so they say at least.

Just write when you have time/feel like but I mentioned because I find hard to follow your ideas without some background. Writing articles instead of just comments would allow us to get a better idea of what you mean.

Gioiello said...

Maju writes: “Just write when you have time/feel like but I mentioned because I find hard to follow your ideas without some background. Writing articles instead of just comments would allow us to get a better idea of what you mean”.

Maju, I have written books, but into my language (Italian), and about other matters: poetry, critics, Italian literature. When I was young I wrote also on philosophy, on your Marx, etc. I have studied for many years the theory of the monogenesis of language of Alfredo Trombetti, but what I wrote is unpublished. Anyway some ideas I have base themselves on those studies, and the difficulty is in the matter itself. I should deepen my knowledge of English: when I say something inconvenient, they say that they don’t understand because I am not fluent in English, but two forums have banned me: probably they understood very well what I was saying. Forums and blogs are very useful also for the papers we find on them and frequently we can download and read. I think that the work of Dienekes is praiseworthy. I feel good so.

Gioiello said...

I have added to Ysearch from SMGF other R-L23-: 2DMW6 (Riquelme Contreras from Chile) and 9DMF5 (Riquelme Gaete from Chile), closely related. It would be added also 5J3P7, and to be examined, having DYS461=11 but DYS393=13, 255VK and DN4GP, who is the only Italian in this group and, having a surname taken also from Jewish Italians, I wrote to him asking something about his origins.
If we add a group closely related from Pakistan and India (see APNEG), it is very strange that all are R-L23+, and I haven’t found anyone who is R-L23-, and also those found absolutely are very few. If Asia was the origin of R-M269, as somebody pretends, actually we don’t find a huge number of the intermediate haplotypes that generated the modern European subclades, i.e. R-L23- and R-L23+/R-L150-.
Re the so called Sephardic cluster of Argiedude, perhaps the two Chilean (of probably Spanish ancestry) can explain its origin. They have DYS390=25, DYS391= 10, DYS385=11-15 but DYS389I and II=13-29 and DYS392=13. Then the Italian 5VU5V (Fiozzo) (and see the close related N4Y6V Cervino who has DYS390=24) and NAUKN (Russo), who , having DYS390=25, DYS391=11, DYS385=11-15, DYS389I and II=12-28 and DYS392=14, are the closest to the so called Sephardic cluster, and we can hypothesize that both, the Spanish and the Sephardic, are of Roman descent. That the Sephardim examined by Argiedude are closely related and presuppose an unique ancestor within these last 2000 years must be demonstrated, having, in so few markers, many differences.

Gioiello said...

Why Argiedude hasn’t put in his spreadsheet Collins (T75CF), who matches very closely pretty all the Jews over 67 markers; he has put Stromberg, who writes on his Ysearch account "Jewish we believe" probably without any proof and with only 12 markers, but he has 67, and he hasn’t put Crenson, who is one of the two tested whom we know for R-L23- (see Adriano’s spreadsheet)?

terryt said...

"I think that the work of Dienekes is praiseworthy".

Very much so. It's easily my favourite site. And his posting relevant information about various species other than humans is also much appreciated, e.g. the recent dairy cattle post.

Gioiello said...

Argiedude is a serious person and is trying the truth. It is interesting his observation about the presence of R-L23- in a belt among Italy, Greece, Turkey, Caucasus. I think that on this we must work. As in the division between J2/J1, here is the division line: here Europe, intending Europe also the countries around Black Sea, under there is Middle East. I have demonstrated that also Asia is lacking R-L23-. To what I said above we must add some samples from Kyrgyzstan I put on Ysearch from SMGF.

Gioiello said...

The two Tyrolese (or, better, Ladins) tested by Pichler et alii (as they are from Val Venosta, but also the German-speakings are above all genetically Ladins, even though this may annoy Miss Pichler et alii) are really DYS389I and II= 12-28 and not 13-29 like Argiedude erroneously reports.
If he would have liked to find some cluster, he should have notice that among the 12 Italians of his spreadsheet (it is incomprehensible why he put Sicilians apart!) 5 are DYS389I and II=12-28, very rare among the others, being the unique exemplar UT2H8: Azios, but with DYS385=12-17 belongs probably to another haplogroup. It is the only one back mutation in the first segment, whilst we have some mutation onwards (14-30) and the Moroccan RRZDS, who has 12-29, presupposes the Italian mutation and a mutation onwards in the second segment, and this could demonstrate an ancient derivation of the North Africans R-L23- from Italy.
From this we see how characterizing is the marker, and in one cluster so ancient as R-L23-.

Gioiello said...

The Moroccan RRZDS, for the “Athey’s Y-Haplogroup predictor”, is Hg. L. He finds the closest to him in YHRD above all in Far East Asia.

pconroy said...

"I think that the work of Dienekes is praiseworthy".

I also agree very much so!

Gioiello said...

I have spoken for (and from) long time of the Rhaetian-Etruskan Fatherland. Argiedude has had the merit to take our attention on the paper of Pichler et alii on the “Rhaetian” population I spoke about largely when it was published (Genetic Structure in Contemporary South Tyrolean Isolated populations Revealed by Analysis of Y-Chromosome, mtDNA, and Alu Polymorphisms, Human Biology, 2006, v. 78, no. 4, pp. 441-464). Also if he has reported wrongly some data I corrected, I think that the other one within the R1b haplogroup are very interesting. First of all we haven’t only the two guys (number 52) reported by Argiedude as R-L23-, but also no. 8 and 47. We have also some R-L23+: 10, 16, 40, 45, and some R-L51+: 24, 27, 38. Then among 92 R1b we have 5 R-L23- ( 5,43%), 4 R-L23+ (4,34%) and 5 R-L51+ (5,43%). Then more than 15% under R-S116+. If we calculate the variance of these haplotypes it would take us well beyond the Younger Dryas.

Gioiello said...

Maliclavelli wrote on www.worldfamilies.net: “In an epoch of low population’s expansion, 7,450 YBP is the date when this population took agriculture from the Middle Eastern J2’s and G2’s and began to expand. Of course this population were the R1b1b2 of the Italian refugium, who, from 12,300 years ago, as the mitochondrial U5b3, have had a low expansion to Middle East till Mesopotamia and Iran, but was who remained in Italy or nearby who peopled Europe”.
The recent demonstration that Middle East lacks haplogroup R-L23- is easily explaining if we look at the map of the expansion from the Italian refugium of the mtDNA haplogroup U5b3: it lacks in Middle East too, then what could be understood as a following lost was probably an obstacle that prevented from the diffusion. We shouldn’t forget that Middle East was depopulated during the Younger Dryas.

Maju said...

We shouldn’t forget that Middle East was depopulated during the Younger Dryas.

Nope. The Oriental region (Zagros, Iran-Iraq) was depopulated during the LGM (not the Younger Dryas) and was repopulated, at least in part, from Eastern Europe. But the Occidental region (Levant, Anatolia) was continuously inhabited.

Gioiello said...

In fact in those countries there are both U5b3 and R-L23-.
Remenber what said Argiedude to Vizachero, that there is a difference between Turkey and Palestine, and what he reproved to him was to make pass what is from Asia Minor as it was Palestinian.It was just the Palestinian region (and morover tha Arab one) which was depopulated and then peopled after by J2 and J1 etc. from North.

Maju said...

There's no YD depopulation in Palestine: Kebaran (Lte UP), gives way to Natufian (Mesolithic), in turn succeeded by the PPNA (Neolithic). This evolution is generally considered local, though there was interaction between the various Neolithic centers of West Asia.

Only later, with PPNB, we do see some clear flow from the North but there is never an interrptution in settlement in Palestine since the earliest MP/UP transitional cultures, probably at the origin of (or at least closely related to) European UP.

Gioiello said...

I thank you, Maju. Certainly you know archeology better than me. I only remember to have read, among the thousands of books and papers I have, that there was a period of aridity in Palestine that depopulated the region. If I’ll find that paper, I’ll write to you. Anyway your theory doesn’t falsify mine, though a migration doesn’t happen either as a place is empty or too much crowded.

Maju said...

I found that there is a cold event related to the end of PPNB, c. 6200 BCE. But doesn't seem to mean depopulation and was much weaker than the Younger Drias, though harsher than the Little Ice Age.

Gioiello said...

But I think isn’t supposable a migration from Europe to Middle East in that time, which was the time of a migration on the contrary, from Asia Minor to Balkans and Europe. If a migration there was I think it happened before, just after the Younger Dryas, in the time when it is documented that mtDNA U53b migrated.

Gioiello said...

The oddity of the previous calculation of the haplotypes under R-S116+ in the Tyrolese (R-L23-=5, R-L23+=4 and R-L51+=5) made me reconsider the matter. In fact we have: R-L23- (no. 8, 47,52) = 5 (5,43% of R1b; 2,57% of the sample); R-L23+ (no. 3, 10, 16, 29, 37, 40, 45, 46)= 8 (8,69% of R1b; 4,12% of the sample); R-L51+ ( no. 18, 24, 27, 38, 44) = 9 (9,78% of R1b; 4,63% of the sample). The percentage of the haplotypes under R-S116+ is 23,9%.
Then in the calculation of Argiedude the Tyrolean R-L23- is 2,57% of the individuals tested, one of the higher percentage absolutely found.

Maju said...

Migrations from Europe to West Asia could be mainly two, AFAIK:

1. The Epipaleolithic Zarzian culture of the Zagros that is probably derived from Eastern European Epigravettian

2. Possible backmigration from the West Balcans (Italy-related?) in relation with rock art of European style in Beldibi, Turkey. This is complicated because the possibly related Belbasi instead shows material culture related to the Natufian of the Levant.

The Levant as such does not seem, AFAIK, to have suffered any direct backmigration from Europe, though would have been influenced by the highland areas in the Neolithic, specially in the PPNB phase.

Gioiello said...

I wrote above that the Italians in the Argiedude’s spreadsheet on R-L23- were more than those counted by him, not counting all those who should be added.
Two have an Italian surname (Viola 5GK22 and Caiazza 3BTDE) and are genetically Italians.
Also Ventura has an Italian surname, but, as this surname is taken also from Jews of Italian descent, I wrote to Jane Ventura to have some information on her family. Anyway his haplotype is totally different from the other Jews who are probably of Spanish origin (Sephardim): DYS390=22, DYS19=13, DYS392=14 like many Sephardim but also like a few Italians, DYS458=16, rare among Sephardim who have the modal =17, and also DYS464=15,16,17,18. If he was a Jew, we can think to an Italian Jew very different from Sephardim.
The Swiss Tarnuzzer (UY5NN), very close to Thiel (8334H), is from the Grisons/Graubuenden, and must be considered a Rhaetian in every respect. Even more Thil has the rare DYSYCAIIa/b= 17-23, like the 4 Italians (Donato, Merante, Prowting, Ferrero).
Some South Americans I added to Ysearch (2DMW6: Riquelme Contreras from Chile, and 9DMF5: Riquelme Gaete from Chile, closely related) shouldn’t probably made us think to a presence of R-L23- in Spain, but to an origin from Jews converted, who had taken their DNA somewhere during the Roman Empire.

Gioiello said...

I have had the answer that also Ventura must be considered genetically Italian:
“Mr. Tognoni,
Mr. Ventura's name has been in Colombia for more than sixty years and prior to that in France, South Africa and Turkey. The connection is that a great grandfather from Maratea, Italy, came to Colombia and married. This side of the family is still in Colombia.
If I can answer any more questions, please let me know.
Thank you and happy hunting Jane C. Ventura”.

Dienekes said...

I have had the answer that also Ventura must be considered genetically Italian:


Stay on topic.

Unknown said...

Hello,

I am very new to the DNA project but have been doing serious genealogy studies for 20 years.

I am also an armature historian with a passion for Joseph of Arimithae. I am not a nut, only out to prove a theory.

My Uncle did the DNA testing and has the 12-25-15 beginning line up of the strand. We do not match anyone from our group of R1b1b2. I have done an exhausted research this past fortnight.

I know the Jews were in Great Britain to work the tin minds. I know there is a strong resemblance between the Hebrew and Welsh languages, now why can't it be VERY possible for the Jews of BC to have married with the original celts of Briton?
Please let me know what you think?

Maju said...

Kelly, if the sequence you mention follows the order usual at Family Tree DNA (DYS 393-390-19...), then it is a rare haplotype belonging to what they call Ht35, i.e. R1b1b2(xR1b1b2a1). Or in other words: the "older" fraction of R1b1b2, which is more common in West Asia and the Balcans.

In Alonso 2005 (see supplemental material), he sampled four haplotypes with that sequence (varying at the other two DYS he used: 391 and 392). All them are rather rare and essentially found in West Asia/Balcans or (one) in Iberia.

Most likely your lineage comes from West Asia. Asia Minor was where the Jewish diaspora was concentrated in Antiquity and in my opinion a good deal of modern Jewish DNA comes from that area rather than Palestine or Khazaria.

J2hapydna said...

Those of you who know the science, can you please say, how long ago did P58* Arabs share a recent common ancestor with these Cohanim? Specifically, the large cluster of Arabs who pretty much match this 12 marker CMH except on the fast mutating dys 385b, who claim common paternal descent thru an ancestor of these Cohanim and who pretty much have similar religious practices and beliefs as OT Jews.

Thanks.