search this blog

Sunday, November 13, 2022

A reappraisal of Ashkenazic maternal ancestry

Kevin Brook, who occasionally comments on this blog, has published a peer-reviewed book titled The Maternal Genetic Lineages of Ashkenazic Jews.

The book focuses on 129 mitochondrial (mtDNA) haplogroups that are found in present-day Ashkenazic Jews, and reveals that these lineages can be traced back to a wide range of places, such as Israel, Italy, Poland, Germany, North Africa, and China.

Ergo, it argues that both Israelites and converts to Judaism from a variety of gentile groups made lasting contributions to the Ashkenazic maternal gene pool. In Kevin's own words, the book also:

- shows that all Ashkenazim remain genetically linked to a significant degree to other types of Jewish populations, not only paternally but maternally as well

- disproves the myth that Cossack rapists were responsible for any of the non-Israelite DNA in Ashkenazim

- presents new DNA evidence in favor of a small contribution of Khazarian and Alan converts to Judaism to the Ashkenazic gene pool.

That makes good sense based on what I've learned over the years from studying modern and ancient genome-wide Ashkenazic DNA. More information about Kevin's book is available at the website HERE.

See also...

My take on the Erfurt Jews


Andrzejewski said...

As a Pole, it strikes me as interesting that Polish Jews’ main spoken language was Yiddish, a High German based dialect. The Kna’anim who preceded the Ashkenazi migration were mainly Khazarian and they spoke a West Slavic language. In the 16th century there were many expelled Jews from the Iberian peninsula who spoke either Castilian or Portuguese making their way to Poland. The process in which the Khazarian “Knaanites” substrate and the late comer Sephardic arrivals assimilated into the Ashkenazi diaspora in Poland needs a detailed study along with a profound fine tuning.

LivoniaG said...

I wonder how much of Kevin Brook's work looks into the misty times and travels of the proto-Sephardic groups that were in pre-Roman Hiberia. Not then as a diaspora but as a part of broad trade network of market towns that reached from east Mediterranean to the Atlantic. There is some evidence that Jews were there doing business 500 years or more before the Romans arrived in Spain. That would be a very early path from Judah but maybe still traceable because of ancient genetic. It would map a rather remarkable 2500 year or so odyssey across Europe and Asia that might start and end in Israel.

Joshua Lipson said...

Stay tuned for similarly in-depth treatments of Ashkenazic Y-DNA lineages, albeit probably not in book form.

Gaska said...

@Livonia G said-"There is some evidence that Jews were there doing business 500 years or more before the Romans arrived in Spain"

Really? Proto-sephardic jews? Are you referring to the Phoenicians or to the commercial relations of the kingdom of Tartessos-Turdetania?

The Levantine genetic markers in Spain are practically non-existent, so I do not believe that the study of these hypothetical commercial relationships will be useful to trace the maternal origin of the Ashkenazis.

The first evidence of Jewish presence in the Iberian Peninsula dates back to Roman times. Some material evidence of the Jewish presence, are two trilingual Jewish inscriptions (Hebrew, Latin and Greek) found in Tarragona and Tortosa, whose dating varies according to the authors between the II-VI centuries AD - From the III century AD dates probably the sepulchral inscription found in Abdera (now Adra) of a Jewish girl, called Salomonula. The first incontestable document that proves the existence of Jewish communities in Hispania are the canons of the Council of Elvira (Iliberis-Granada) at the beginning of the 4th century AD.

StP said...

A clear distinction must be made between Ashkenazi Jews and Sephardic Jews.
There were few Sephardic or Spanish (and North African) in Poland, although Spain was the center of the Slavic slave trade.
Ashkenazi Jews, the vast majority of Jews in Poland, are more of a line of Rhine, Italic and Palestinian Jews (see Behar et al. 2017)

Steppe said...

That means the mtdna of the Ashkenazi Jews is sometimes very heterogeneous and what I saw in the book many of Indo-European origin (Indo-Iranian ....) I think, for example, the typical Ashkenazi Y-DNA lineage Z93 - Z2124- Z2122-CST 6 is definitely originated in the Middle East. This can also be related to the Khazars who adopted the Jewish faith to trade with Muslim and Christian states (extensive trade relations-China) and many in the Khazar army were of Alanic origin!

StP said...

Two dominant positions (from Wikipedia)
1)A 2013 study at the University of Huddersfield, led by Professor Martin B. Richards, concluded that 65%-81% of Ashkenazi Mt-DNA is European in origin, including all four founding mothers, and that most of the remaining lineages are also European. The results were published in Nature Communications in October 2013. The team analyzed about 2,500 complete and 28,000 partial Mt-DNA genomes of mostly non-Jews, and 836 partial Mt-DNA genomes of Ashkenazi Jews. The study claims that only 8% of Ashkenazi Mt-DNA could be identified as Middle Eastern in origin, with the origin of the rest being unclear.

2)The study was criticized by geneticist Doron Behar, who stated that while the Mt-DNA of Ashkenazi Jews is of mixed Middle Eastern and European origins, the deepest maternal roots of Ashkenazi Jews are not European.

Andrzejewski said...

I suspect Elhaik’s fingerprints are all over this paper. When I googled up “Ashkenazi” and “Khazarian”, almost of all of (the few!) studies had his signature or were quoting his name.

Andrzejewski said...

That said, I believe (and @Samuel Andrews wrote on another thread before me) that the Steppe WSH component and the Slavic-specific contribution in AJ must be much higher.

Matt said...

"disproves the myth that Cossack rapists were responsible for any of the non-Israelite DNA in Ashkenazim"

Strange idea but IDK how mtdna shows that. Mtdna in this founder effect hit population will have highly random proportions that don't correlate with admixture (besides not proving much about paternal ancestry).

Michalis Moriopoulos said...


Nah, I can understand the trepidation but Kevin Alan Brook is legit. His book is well-researched and scholarly. He's not like Elhaik at all. An interest in Khazars and their connection to Jewy is actually a very interesting topic worthy of attention. You just have to ignore the stupid conclusions some clueless or bad faith actors draw from it.

Kevin Brook said...

Andrzejewski, no, my book does not reference Elhaik's studies nor use his methods. My research team and I found actual traces from Khazaria in Ashkenazim completely independently of him, with connections to the Ural Mountains region, Central Asia, and the North Caucasus, without assuming the South Caucasus was involved. (Elhaik's Armenians and Anatolians are poor Khazarian proxies.) Furthermore, we don't exaggerate the inheritance from Khazaria.

StP, I am sure there were more than just 4 founding mothers for the Ashkenazic population, but the others' frequencies of course are lower today for whatever reasons. For example, I propose that U1b1 and L2a1l2a were also among the founding lineages. As for Richards' team's paper from 2013, we particularly dispute their claim that the major lineages K1a9 and K2a2a1 came to Ashkenazim from European converts, finding them instead rooted in the Middle East.

StP said...


I think there should be a clear steppe component in priestly lineages: kohanim Levites. Nearly all Levics and half of Kohanim carry halogroup R1a-CTS6 or R1a-Y2619 from the Asian Z93 lineage

Gaska said...


Interesting book-Have you found any mtDNA that can be clearly related to Iberia or Sephardic Jews?

I think K2a2 has been found in Georgia (1,250 BC) by Koptekin (2.022), so K2a2a1 may be Levantine. U1b1 seems to have its origin in Anatolia.

Kevin Brook said...

Matt, there's a separate section of the book dealing with Y-DNA. There are no Eastern European Ukrainian/Russian/Cossack Y-DNA lineages in Ashkenazim, and only one or two West Slavic ones. If Cossack ancestry had been present and there were male descendants, we must assume there would have been some Eastern European branches of haplogroups like R1a1, I2a, I2b, and N1c1 among Ashkenazim with Cossacks or East Slavs as the closest non-Jewish matches. I haven't verified the allegation by Wim Penninx and the Avotaynu Project that Y-DNA haplogroup N-L1027 was found in one bona fide Ashkenazic person.

Gaska, on page 142 I propose that the mtDNA haplogroups H25, I5a1b, N1b1a2, U2e1a1, and/or U6a7a1b in Ashkenazim could be descended from Sephardic women. The Ashkenazic branch of X2b7 has non-Jewish people from Spain as the closest matches.

Information I neglected to include in the book is that the Ashkenazic branch of N1b1a2 includes the mutations mutations T13608C and T4619C and that YFull's MTree innovatively calls this branch N1b1a2j1a. From Silva's 2021 study, two people from Almeria, Andalusia, Spain belong to N1b1a2j1a too and thus perhaps had a distant Sephardic matrilineal origin. These two samples' haplogroup data were verified to be accurate by Dr. Ian Logan.

Strangely, a Crypto-Jew from the Argozelo region in Bragança district from Table S1 of "Echoes from Sepharad: signatures on the maternal gene pool of crypto-Jewish descendants" belongs to a different branch of N1b1a2 that does not include T13608C or T4619C.

Also, the Chalcolithic sample I1177 from Peqi'in Cave in ancient Israel belonged to the branch MTree calls N1b1a2e, with the mutation A8443G, rather than to the Ashkenazic branch.

An Ashkenazi from Moldova and Bragança Crypto-Jews share the mutations A8014G and G13708A, representing a special branch of U2e1a1.

If there's enough interest in my book, I plan to incorporate news and extra details like these and Dilek Koptekin's September 2022 thesis with that K2a2 carrier from Georgia into a larger second edition.

Andrzejewski said...

@Gaska Given the negligible impact that Iberian Jewry, which numbered close to 1M census and has lived in Spain/Portugal for 2 millennia on the Hispanic, conquistadores population (ie Catholic natives), it should be surmised that those Jews were overwhelmingly Celtiberian converts to Judaism. By the same token, the small number (>20k) of North African Berber invaders under Tariq in 711, augmented by the fact that after 800 years there are barely any traces of them in modern pops of the peninsula, leads me to believe that the Moriscos were actually “Mozarabs” or “Muwalads”, ie Celtiberian/visigoth/Roman/Basque whose ancestors had converted to islam and then back to Catholicism.

Rob said...

@ Kevin

''There are no Eastern European Ukrainian/Russian/Cossack Y-DNA lineages in Ashkenazim, a''

what about the 'Ashkenazi cluster' within the 'Slavic I2a-P372'

Disproving a Cossack Paternal Ancestry for an Ashkenazic Lineage
by Rachel Unkefer.

Ive only skimmed through the article, but Im familiar with the phyogeny and it is correct.
It is not a major lineage, but it's there

Andrzejewski said...

@Kevin Have you read the thread on Erfurt Jews from a few months ago here?

I’m curious as to who is a “true” Ashkenazi is: allegedly, Jews who had been deported by the Romans from Jerusalem to Italy were migrating north during the Great Migration period to Lombardy. During the reign of Charlemagne Kalominus and a few other prominent Jews from the (then) Kingdom of the Longobards to settle in Alsace-Lorraine. With the split of the Frankish Kingdom, Jews who resided in France spoke Jewish-French while those in Köln spoke Yiddish. A few centuries later starting with the onslaught of the crusaders and ending with the Black Death, those “Ashkenazim” migrated to Poland, where they were welcome by the king.

This is the traditional narrative.

However, Erfurt flipped the script by showing that not only are Jews in Poland NOT exclusively descendants of Ashkenazi (original Slavic speaking Jews were called ‘Kna’anim’), GERMAN aka Ashkenazi Jews were basically partially offsprings of West Slavic and Turkic/East Asian lineages.

The question at the bottom line is: if 1. German Jews themselves were not all Ashkenazis. 2. Many German Jews were Slavic or Khazarian originally (at least in Erfurt). 3. Autochthonous Polish ‘Kna'anim’ Jews were assimilated culturally within Polish Catholic population and their Slavic dialect persisted until 16th century. 3. There was an influx of Sephardics into Eastern Europe. ——> What made the Ashkenazi prevail with their Germanic language and unique although foreign customs, over both Sephardics and Kna’anim?

Jugu said...

Hi Davidski,

Your updated Albanian average has one duplicate value. Albnorth1 and Albnorthwest1 are the same G25 coordinates, I removed it when I made my average, but I noticed your new Alb average is slightly different, I am guessing this is the reason why, because you have a repeating individual in a regional and sub regional category, duplicating it's weight on the average.

StP said...

@Kevin Brook said...
Matt, there's a separate section of the book dealing with Y-DNA. There are no Eastern European Ukrainian/Russian/Cossack Y-DNA lineages in Ashkenazim, and only one or two West Slavic ones. If Cossack ancestry had been present and there were male descendants, we must assume there would have been some Eastern European branches of haplogroups like R1a1,…

But surely you won't deny that in the Jewish community (Levites! Kohanim!), next to the indigenous Semitic lines (J1, J2, E...) there was a non-Semitic, Indo-European lineage, originating from proto-Slavic lands, like R1a1-Z645> Z93 (pre-fields of the Carpathians)>CTS6>Y2619? (Behar et al. 2017)
Most likely from this line was Melchizedek, Levi, Aaron and Moses!!

Davidski said...


Quit acting crazy.

If Z93 was from "proto-Slavic lands" then it would be an important marker in Slavs. But obviously it's not.

No one knows where Z93 is from exactly except that it's probably from somewhere in Eastern or Central Europe.

Kevin Brook said...

Rob, Davidski already disproved the idea of a Cossack element to Ashkenazic DNA in one of the autosomal models he posted to this blog in September 2016. West Slavic, yes. Cossack, no.

Andrzejewski, yes, I did carefully read the comments here on Erfurt's Jews and I've been in contact with multiple co-authors of that paper for years. I was prohibited from giving details in my book because I finalized my manuscript in April 2022 when it was sent to the typesetters whereas the Erfurt preprint wasn't out until the following month and its final version comes out in December. But I prepared the index in September so I included an index entry for "Erfurt (city in Germany)" pointing to the page where I mentioned "a certain German city".

Although certain scholars, notably including Bernard Weinryb, had rejected any continuity between Slavic-speaking Jews and later Jews in central or eastern Europe, even before high-quality population genetics studies with adequate sample sizes had been done, I was already under the impression that modern Ashkenazim = descent from the Israelite + Slavic + Khazarian ethnicities plus Sephardic + Romaniote + West Knaanic + Mizrahi communities. (All of these elements have genetic backing today, not only documentary evidence.) My early summary of Ashkenazic ethnogenesis that I wrote in 1997-1998 for the Introduction in the First Edition of "The Jews of Khazaria", published in April 1999 read: "I believe that the missing link between the Turkic-speaking Khazars and the Yiddish-speaking Jews of later times were the Slavic-speaking Jews who inhabited Kievan Rus in the period immediately following the demise of the Khazar Empire at the hands of Svyatoslav. In my opinion, these Slavic-speaking Jews were a mixture of Khazars, Slavs, and Middle Eastern Jews. These Jews eventually intermarried with Yiddish-speaking Jews from Bavaria as well as with Jews from Prague, Vienna, the Iberian peninsula, and other communities. Thus, the Ashkenazic ethnogenesis, having been formed by migrations from the east (Khazaria), west (e.g., Germany, Austria, Bohemia), and south (e.g., Greece, Mesopotamia, Khorasan), is more complex than previously envisioned."

Within Germany, Erfurt was kind of a special case as it became a gathering place for Jews from multiple places. Jews who stayed in western regions like Alsace and the Rhineland remained more "purely" Ashkenazic, without the additional Polish, Czech, and Sephardic elements. There are very few pure Ashkenazim alive today but their genetic profiles are strikingly free of Slavic DNA and score the highest in Davidski's "ASHKENAZI" element in Eurogenes' Jtest.

Something I did not anticipate at all during the 1990s was that any Berber DNA existed in Ashkenazim. I thought at the time that that was only in North African Jews.

The original Ashkenazic Jews' descendants demographically overwhelmed the other kinds of Jews so the replacement of Slavic speech and Slavic names with Yiddish ones is understandable. Alexander Beider was the best at explaining this.

As Joshua Lipson has said elsewhere, the Sephardim, who arrived in Poland only after 1492, left a legacy of less than 10 surviving Y-DNA haplogroups in modern Ashkenazim. He specifically said "low single digits". So, let's presume around 4 Y-DNA and a maximum of 5 mtDNA lines. Their autosomal imprint is nearly universal among Eastern Ashkenazim, albeit in low percentages, but their uniparentals are not so common.

StP said...


1)in Poland, "proto-slavic" means rather: "earlier than praSlavic".
2)I personally saw "Z98* " in the paper about the Glavanesti burial mound sample, published by Sirak et al. 2019 and 2020, with reference to the ENA archive.

Rob said...

@ Kevin

I’m not talking about Cossacks. I was pointing to a “Slavic” lineage in Ashkenazi Jews

Davidski said...

There is West Slavic ancestry in Ashkenazic Jews.

Andrzejewski said...

Ashkenazi Jews living in the USSR intermarried extensively with non-Jews, thus it is highly likely that Russian Jewish diaspora has extensive East Slavic markers throughout. Likewise Jews in America are very admixed with other Americans, whereas in Israel there’s a similar process taking place between Ashkenazi, Sephardic and Mizrahi. For that reason, “classical” Eastern European Ashkenazi lineages and adna are quickly becoming obsolete.

Steppe said...

Davidski @

where did the Ashkenazi Jews absorb the Z93 - Z2124 - Z2122-CST 6 lineage, I think in the Middle East ? however, this line would have to deal with Indo-Iranian migrants from Eastern Europe-South Siberia/Central Asia-Iran-Levant. I also think the Z93 Y40 was more associated with Mitannni than CST 6.

Kevin Brook said...

Andrzejewski, to an extent that is true, although not in certain Orthodox Jewish communities in the United States where most births are still 100% Ashkenazi. In a conversation with Judy Siegel-Itzkovich in January 2006, Dr. Skorecki made the observation, as she stated it, that "it was important to conduct genetic research on Jewish populations now, because the opportunity would soon disappear, with the intermarriage in Israel between Jews of Ashkenazi and Sephardi origin as well as assimilation and intermarriage of diaspora Jews with the Gentile majority outside Israel."

Miki said...

61% of Jewish marriages in the 2010s(71% excluding ultraorthodox) were indeed intermarriages.
Since 1980 that figure was hovering in the 40% range.
That means that by 2050, the Jews will no longer be a separate ethnicity.

Simon_W said...


"For that reason, “classical” Eastern European Ashkenazi lineages and adna are quickly becoming obsolete."

I agree there's a broad tendency towards this, in large part related with the secularisation process that affected Jews and gentiles alike. Anyhow, there are still orthodox Ashkenazi Jewish communities in Europe and elsewhere that stay endogamous.

Miki said...

Regarding U2e, it's an European branch

gsueso said...

Any ideas about I1c1a? Closest relatives of the Jewish branch are a Chechen and a Turk which might suggest Khazar, but the TMRCA is uncertain and it was recently found in ancient Armenia by Lazaridis et. al. so it may have been present among the Israelites (or at least post-Babylonian Jews).

Davidski said...


Yes, Middle East.

WSH said...

I think that Sephardis actually may have been the population which "loosened up" the Ashkenazi bottleneck we saw tighter in the Erfurt study. The Erfurt study estimated the non-bottlenecked mixing population as being 50,000-100,000 at the time of mixture, and the time of mixture being the 15th Century, which matches fairly well with estimates Sephardim Population size (400,000 c. 1492, half stayed in spain, maybe 100-150k moved to Nafrica + MENA, leaving an extra 50-100k in Europe.

I am also quite sure that the K lineages Ashkenazim are Italian. They might have come from the Middle East before then, but when they came into Ashkenazim they were from Romans/Italians. Seems like a serendipitous answer to this question. Ultimately, it is very interesting that so many foreign matrilineages are present in a population characterized by its matrilineal religious laws. Agamemnon on Anthrogenica (himself a patrilineal jew) has discussed the flexibility of this law and how it wasn't always agreed on, but I myself am not an expert.

I am very interested in hearing more about R1a-Z93 in Cohens and Levites. If I remember correctly there ARE some Z93 clades in Slavs. I don't understand why there is suspected Alanic influence as I have never heard of Alanic converts, but Khazar may also have had some Z93. Khazar were mostly West Eurasian in ancestry iirc. Do Cohens/Levites also have more Steppe ancestry? I doubt it, but it would be interesting. Is R1a also more common in Sephardic/Mizrahic levites? If that was the case, it wouldn't be a stretch to suggest that "Levi" or one of Levi's patrilineal ancestors was a Mitanni, Kassite, Iranic, what have you. We already have those Steppe-rich outliers from Ancient Israel, no?


There is no evidence that the "EU" samples in Erfurt were Knaanim. They had the same DNA bottleneck experienced by the Ashkenazic Jewry, so if they were Knaanic that would mean A) Knaanics were an offchute of early Ashkenazim, and B) The bottleneck happened before 900 AD, because the first Knaanic inscriptions we have are from the 800s AD. I think those were just a group of early eastern Ashkenazim, maybe they had Knaanic DNA maybe they didn't. We don't have much evidence of Knaanim in the first place.

Davidski said...


I am also quite sure that the K lineages Ashkenazim are Italian. They might have come from the Middle East before then, but when they came into Ashkenazim they were from Romans/Italians.

That'd be pretty easy to work out, unless those "Italian" lines were from very recent Middle Eastern migrants.

If I remember correctly there ARE some Z93 clades in Slavs.

A few stray cases. Not even 1% of any Slavic population.

Khazar were mostly West Eurasian in ancestry iirc

But significantly East Asian.

Andrzejewski said...

I would be curious to find out how much percentage of Moroccan Jews are converted Berbers and how much are descended from exiles from Gaska’s homeland.

On the same vein, i do recall a few years ago a thread here about Spanish genetic history, and it turns out that a common ydna marker E1b1b that used to be attributed to the Moors actually derived from the Visigoths.

Moreover, it was fascinating to discover that most of the Visigothic adna and uniparental y markers was not from Scandinavia but came from the pre-Slavic invasion , pre-Justinian Plague decimation Paleo-Balkanic substrate pop among whom the Visigothic elite had sojourned for centuries.

IIRC, that article even mentioned that the Visigoths had a rare mtdna F subclade that had been fairly common among Finno-Ugric populations in today’s Karelia until it got wiped out by invading Eastern Slavic Kievan Rus.

bellbeakerblogger said...

I have been intending to blog about one of his hypotheses, but busy with job changes at the moment.
He had proposed that Levites have a kind of pseudo-Eastern European ancestry, not as dwellers in Eastern Europe, but taking an analysis of sub-clades and TMRA that many of them (weird sub-clades of R1a and R1b) are more reminiscent of Wester Iranians (Lurs) and around the time the elite Jews were returning to Israel during the second temple period. I liked this a lot with a few exceptions.

From this I've thought a more likely scenario is that a large chunk of the Levite paternal ancestry is owed to mercenaries of Western Iran who occupied pre-Roman Israeli cities. For the same reason Romans and Egyptians employed foreigners. Cost, expertise, loyalty. I thought that is the most likely reason Captains of the military and religious elites intermarried.

Kevin Brook said...

gsueso, my book mentions the Chechen and Turkish I1c1a carriers on page 60 but I don't propose a Khazar origin for it. New information came to my attention beginning in May that the Ashkenazic branch is apparently what YFull's MTree calls I1c1a1, defined by the extra mutation G6267A. According to my research partner Leo Cooper, the two Bronze Age Armenians from Lazaridis' team's August 2022 data set are members of I1c1a1. The I1c1a carriers from Turkey and Chechnya belong instead to I1c1a2, defined by the mutation C7774T.

WSH, I agree that Sephardic arrivals helped diversify the Ashkenazic gene pool to an extent. I cannot agree with your contention that all Ashkenazic K lineages are Italian. There are no known Italian K1a9 or K1a1b1a carriers, although there are modern Italian K2a* carriers (page 76) and K2a2a1 carriers (page 77), and ancient and modern Italian K1a4a carriers (pages 73-74) and K1a1b1* was found in ancient Italy (page 70). I write that K1a4a in Ashkenazim might descend from an Italian woman (page 134).

The suspected Alan lineage is G2a-FGC1093, shared between North Ossetians and Ashkenazim (page 7). Some Alans did convert to Judaism according to the Schechter Letter translated on pages 113 and 115 of Golb and Pritsak's 1982 book "Khazarian Hebrew Documents of the Tenth Century".

The Erfurt Jews with the significantly higher European and East Eurasian components were isotopically distinct from the other Erfurt Jews who were more like what the early Rhineland Jews would have been like, so they arrived in Erfurt from different regions, but I agree that the ancestors of the "European" population cluster among them had already mixed with Ashkenazim to an extent.

Andrzejewski, there have been several different kinds of Moroccan Jews, and there's a north-to-south cline. As "Sp_loa" wrote in The Apricity based on his analysis of 23andMe estimates for Moroccan Jews, southern ones have more Berber admixture than those living in northern cities but all of them have at least some Berber.

StP said...

@WSH said...
I am very interested in hearing more about R1a-Z93 in Cohens and Levites. If I remember correctly there ARE some Z93 clades in Slavs. I don't understand why there is suspected Alanic influence as I have never heard of Alanic converts, but Khazar may also have had some Z93. Khazar were mostly West Eurasian in ancestry iirc. Do Cohens/Levites also have more Steppe ancestry? I doubt it, but it would be interesting. Is R1a also more common in Sephardic/Mizrahic levites?

Behar et al. 2017 gives us a lot of information on this topic:

Steppe said...


it's not the same Z93 , the Alans/Sarmatians have a different subclade than Ashkenazi Jews ( Z93-Z2124-Z2122-CST 6) and for example an Alanic/ Sarmatian branch Z93-Z2124-Z2125-S23592 or Z2123-Y934-CST1806...

Labayu said...

I don’t doubt that Ashkenazi Jews have a small amount of ancestry that might look Khazarian, but the Khazars probably never converted to Judaism. There really isn’t any credible evidence for it:

Absract: The view that some or all of the Khazars, a central Asian people, converted to Judaism at some point during the ninth or tenth century is widely accepted. A careful examination of the sources, however, shows that some of them are pseudepigraphic, and the rest are of questionable reliability. Many of the most reliable contemporary texts that mention Khazars say nothing about their conversion, nor is there any archaeological evidence for it. This leads to the conclusion that such a conversion never took place.

StP said...


But I do not connect the genes of the family of Levite priests in Israel with the Alans-Sarmatians!! Moreover, not this region and not this time!

Steppe said...


I think this branch came to the Middle East with Indo-Iranian migrants and was involved in the formation of the male Levite lineage Z93 CST 6 and others like Y40 with Mitanni, I mentioned above, I think there is more genetic data to come .

pnuadha said...


But significantly East Asian.

I'm confused as to why Khazar question wasnt solved long ago.

Around 2010 we could cluster modern populations and detect Siberian vs East Asian. Do Ashkenazi jews have access Asian from these old models? But I suppose that could only check for recent mixing.

Around 2013 we could test deep ancestry when Laz predicted the existence of ANE. Could we find excess Asian in Ashkenazi jews using that method?

From 2015 to present we have many ancients including basal E asian (tinyuan) and Israelites. Why isnt it simple to find the amount of Israelite (South Levant 0bc) and the amount of East Asian (proxy for Khazar). The former is probably difficult because we dont know which European population Ashkenazis jews largely derive from. There was that 2017 study that modeled jews as half levantine but they also modeled tuscans as 40% levantine. Thats a ridiculous model.

What was the point of retrieving the mideval jews in germany? I saw that some of them had east asian without much siberian. Whats the significance to that?

Do most Ashkenazi jews have some north african in them? If so what does that mean?

Finally, do we know the middle eastern source for Ashkenazi jews. Is it Israelite, something else, or a mix.

Davidski said...

Formal statistics have a very hard time with fine scale ancestry.

It's not possible to use formal stats to discriminate between Siberian and East Asian sources of gene flow given no or equal amounts of non-East Eurasian ancestry.

But Admixture can easily do that, and so can PCA.

The problem is that there's too much reliance on formal stats in ancient DNA work, and not enough cross checking between different methods. Formal stats are still very useful, but they have their limitations, and I think the good people at the David Reich Lab really need to realize this finally.

In regards to Ashkenazic genetic structure, I can't say much more for now than what I've said here...

pnuadha said...


I don’t doubt that Ashkenazi Jews have a small amount of ancestry that might look Khazarian, but the Khazars probably never converted to Judaism. There really isn’t any credible evidence for it:

The source you reference is written by an American Jew who moved to Israel. I think its fair to say there may be some bias.

Tarbagan said...

Hi David! Its offtop, but have you already seen the samples of Koban culture?:

Labayu said...


I don’t much care whether Stampher has a bias or not, as long as his arguments are solid, which they seem to be. That is of course a publication in a peer-reviewed journal, and I’ve yet to see any counter-argument made in the like. There is literally no archaeological, or epigraphic, or numismatic evidence, and there are no sources from nearby cultures that mention the Khazars being Jewish.

For the most part, there is only this:

The Khazar Correspondence is a set of documents, which are alleged to date from the 950s or 960s, and to be letters between Hasdai ibn Shaprut, foreign secretary to the Caliph of Cordoba, and Joseph Khagan of the Khazars. The Correspondence is one of only a few documents attributed to a Khazar author, and potentially one of only a small number of primary sources on Khazar history.

The authenticity of the correspondence has been challenged, on the grounds that it has little in common with the otherwise attested chronology, language, borders and economy of the Khazars at the time…

The alleged Correspondence originated with Hasdai ibn Shaprut, foreign secretary to Abd ar-Rahman III, the Umayyad Caliph of Cordoba and al-Andalus. A man of extensive contacts and virtually unlimited resources, Hasdai is said to have learned of the existence of the Khazars from Khorasani merchants. His ignorance of the Khazar state is odd, and may even have been disingenuous, either on the part of a potential forger, or Joseph's statements to the effect that there had been communications between the two communities in the past.

Hasdai's first messenger is said to have found his way to Constantinople, where Byzantine authorities refused to permit him to proceed further. Eventually Hasdai's letter is said to have been given to Jews attached to a Croat embassy, and reached Khazaria via yet another messenger, Isaac ben Eliezer of Nemetz (Germany).

Joseph's alleged reply gives an account of Khazar history and of its current (c. 960) sociopolitical and economic status. He further invites Hasdai to come to Khazaria, an invitation that Hasdai probably never accepted. The Correspondence has survived in three slightly variant versions over the centuries.

Historian Shaul Stampfer has questioned the authenticity of letter said to have been received from the Khazar King, citing numerous linguistic and geographic oddities amid a flourishing of pseudo-historiographic texts and forgeries in medieval Spain.[2] The text, written in literary Hebrew, with many Arabic touches offers a chronology hundreds of years off the actual dates of the Khazar Kingdom, borders very different from that which it held at the time and an economic and agricultural situation far removed from realities at the time.[2] Authenticity of the conversion story in the letter was dismissed by the early 12th century Talmudic scholar Rabbi Judah of Barcelona.[2] Stampfer writes that a “similar pseudo-historiographic Muslim text from the time of Hasdai ibn Shaprut is quite similar to the correspondence attributed to him in its concern with buttressing identity.”[2] Such a legend would have been of use to the emerging Jewish aristocracy to show the biblical promises of Jewish kingship were extant.[2]

To me, it appears to be a myth analogous to that of Prester John. Although, I'm open to the possibility that there may be an element of truth to it, and would certainly change my opinion if some solid evidence came to light.

Kevin Brook said...

Labayu, Stampfer's article is fringe, not accepted by most historians of Khazaria. It is also wrong. There are many independent references to Khazar Judaism and some of them pre-date the Khazar Correspondence of the 950s-960s. For example, see the accounts by Ibn Fadlan (from the 920s) and Christian Druthmar of Stavelot (from the 860s). Also, there is numismatic evidence in the form of the Khazarian series of coins declaring Moses to the messenger of God. In addition, I already genetic evidence showing links to peoples related to Turkic Khazars and North Caucasian Alans. The suspected Khazar links include the specific varieties of N9a3 and A12'23. The N9a3 lineage is particularly interesting, being one level down from Turkic-speaking Bashkirs and Mongolic-speaking Daurs. You can't find examples of other Turkic and North Caucasian populations that converted to Judaism, only the Khazars and Alans, just as the medieval documents stated.

It is true that Jewish artifacts from Khazaria are sparse but see Valery S. Flyorov's article "Iudaizm, khristianstvo, islam v khazarskom kaganate po arkheologicheskim dannym (kratkiy obzor)" (Judaism, Christianity, Islam on archaeological data in Khazaria) in volume 8 (2018) of the journal Prinosi kŭm bŭlgarskata arkheologiya on pages 139-145 regarding the image of a 7-candled menorah on a pot from Khazar-era Mariupol, and see Larisa A. Golofast's article "Rannesrednevekovaya amfora s drevneyevreyskoy nadpis'yu na svintsovoy plombe iz Fanagorii" (An Early Medieval Amphora with a Hebrew Inscription on a Lead Seal from Phanagoria) in the July-August-September 2020 issue (No. 3 of 2020) of the journal Rossiyskaya arkheologiya on pages 159-172 which provides evidence that a Hebrew-inscribed amphora was imported into the Jewish community of Khazar-era Phanagoria.

Constantine Akropolites wrote that some Khazarian Jews retained their identity into the 11th century and Abraham ibn Daud wrote that some Khazarian Jews retained their identity into the 12th century. All documentary sources for Khazar Judaism are summarized and quoted in my book "The Jews of Khazaria, Third Edition". The new genetic evidence shows that these Khazarian Jews later merged with other Jews.

rozenblatt said...

@Kevin Brook "You can't find examples of other Turkic and North Caucasian populations that converted to Judaism, only the Khazars and Alans, just as the medieval documents stated."

What about Karaites?

Kevin Brook said...

Rozenblatt, European Karaites are not a converted Turkic population. They are partly of Israelite origin and the rest of their ancestry is similarly West Eurasian. See my study "The Genetics of Crimean Karaites" in Karadeniz Araştırmaları No. 42 (Summer 2014): 69-84 and see pages 25, 131-132, and 180 in my DNA book. You can also listen to Maksym Martyn's live video lecture "The 'Khazarian Myth': The National Identity of the Crimean Karaites" on Sunday, November 27, 2022, hosted by the National Library of Israel.

Labayu, to clarify my statements above, N9a3 and A12'23 are in the Ashkenazic population. I encourage you to take a look at the samples on YFull's MTree for N9a3a1b C7810T and N9a3a1b1 T146C!

Andrzejewski said...

@Kevin Brooks @Rozenblatt “ Rozenblatt, European Karaites are not a converted Turkic population. They are partly of Israelite origin and the rest of their ancestry is similarly West Eurasian.”

Theres a difference between Ukrainian/Lithuanian Karaites and Krimchaks.

Matt said...

@Davidski, the Allen resource has been updated -

May be useful to review; I know the genotypes from Gretzinger are in there now.

Davidski said...

I'll try and have a look soon.

Would be nice though if someone sent me a Plink file with all the new samples.

Janko Raven Johnson said...

Highly recommend Kevin's Jews of Khazaria book to anyone interested in these issues - a serious, sober, but also readable resource. Also his email newsletter is great for updates.

coffeeprince said...

@Kevin Brook
I enjoyed reading your newest book. I finished it earlier today. I have a question for you - in your book you mentioned that H11b1 is either from a Polish or Czech source, but I think you thought Polish was more likely. Now that we know that H11b1 was found in the Erfurt Jews, is it more likely to be brought in from what is now Czechia instead of Poland?

Desailly said...

Sorry to be off topic, but is there any news on that paper about Old Kingdom of ancient Egypt that you mentioned was supposed to come out in the post-covid years?

Davidski said...


No idea.


I won't allow any comments here referencing My True Ancestry (MTA).

Kevin Brook said...

Replying to Coffeeprince: Davidski and I were under the impression that all of the Slavic DNA in Ashkenazim was Polish, but now I do think some of it was Czech. I didn't get to see the autosomal estimates for Erfurt's Jews until it was too late, but I was aware that they had H11b1 among them (but couldn't tell people that I was aware because the preprint wasn't out yet). The heavy Slavic admixture in some of the Erfurtians would suggest that a Czech element is present, building on the fact I noted that H11b1's closest non-Jewish match is Czech, and that the timeframe for Slavic input began earlier than the 14th century start that had been estimated by Shai Carmi's team in 2017. For this reason, I had managed to add the phrase "and possibly Czech" to page 133 in September, when it was too late to add anything more lengthy or anything that would alter pagination for any words that were already in the proofs.

There's news from Friday regarding the Ashkenazic Y-chromosomal haplogroup G2a-FGC1093: a Russian research team studying the DNA of members of the Koban culture in the North Caucasus (including North Ossetia) and Central Caucasus circa 1100-400 B.C.E. found a male (sample name lib7al_PE in project PRJNA797283) carrying G2a-FGC1093. Leo Cooper would like to independently verify it, but it seems legit to me considering how deeply rooted in the Caucasus the cladistic ancestors and relatives of G2a-FGC1093 are today.

Reliable sources tell me that the date for the release of the final Erfurt paper has evidently been moved up to November 30th.

Labayu said...

@Kevin Brook

It is obvious that you have spent a great deal of time researching this topic, whereas I have only read a bit here and there. My perception of the widespread rejection of Stampfer’s arguments is that it is akin to what I’ve seen several times in my own field (archaeology) in that old paradigms die hard, even if they are built on stacks of assumptions supported by very minimal potential evidence. That said, I will certainly read your book and reserve judgment accordingly. Likewise, I should take a look at the archaeological evidence you cited. Although, minimal evidence of the presence of some Jews in Khazaria isn’t the same as evidence of it having become the state religion nor of significant conversion. As I said, I don’t doubt that Ashkenazi Jews have some small amount Central and East Asian genetic ancestry.

To clarify, I wasn’t under the impression that the Khazar Correspondence was the source of the notion that the Khazars were Jewish, but rather that it was a pseudo-epigraphic text that was inspired by circulating rumors that about the Khazars’ Jewishness that had made it to Iberia.

Regarding Ibn Fadlan, I find this is Stampfer’s which I find persuasive:

Another relevant document on the Khazars' conversion is found in the travelogue of Ahmad ibn Fadlan, who set off in 921 on a journey that brought him to the kingdom of the Bulgars, near Khazaria. His lively description of the Khazars is therefore based on informants and possibly on literary sources but not on firsthand knowledge. We do not have the complete text of ibn Fadlan's travel report but only a large fragment, found in 1923 in Meshed (Iran), that ends in the middle of the description of the Khazars and lacks any reference to their Jewish- ness. However, a long citation of ibn Fadlan by Yaqut ibn 'Abd Allah al-Hamawi (1179-1229), who wrote long after the Khazar state had ceased to exist, contains the laconic comment that "the Khazars and their king are all Jews." This same long quotation notes, "and the king has nine judges of Jews, Christians, Muslims, and idol worship- pers. If a legal case arises for some people these [judges] adjudicate it." This might have been written by ibn Fadlan, or it might have been added by Yaqut or by the copyist of ibn Fadlan whom he relied on.

According to ibn Fadlan, the Khazar kings each had 25 wives and 60 concubines. In addition, if a king ruled for 40 years, he was put to death on the grounds that after so many years he would be confused and no longer sharp witted. Ibn Fadlan also reported that when a Khazar king died, his body was buried in a special house that was then flooded by a river; the individuals involved in the burial were killed so that the exact location of the burial could never be discovered. Almost all of this is in clear violation of Jewish law, and it also seems rather fantastic. Ibn Fadlan also makes no reference to observance of any Jewish rituals, even though Sabbath observance and other rituals might be expected to have aroused some attention.

Ibn Fadlan's fascination with the bizarre, his lack of a critical sense with regard to stories he was told, and his lack of explanation for the conversion to Judaism weaken the claim that he can be regarded as a reliable source for Khazar conversion to Judaism. It would be simpler to claim that he relied uncritically on either an earlier source or a traveler for his story.

Labayu said...

Following up on my previous comment…

Regarding Akropolites:

The earliest reference I have found in a Byzantine source to Khazar Jews is in The Life of Saint Zotikos, a hagiography written or recorded by Akropolites (c. 1250–c. 1324). Two stories at the end of the collection refer to Khazar Jews living in Constantinople. Timothy Miller, who edited the text, has suggested that the stories are much earlier, though there is no evidence for an early date. In any case, the existence of some Khazar Jews in Constantinople is not relevant to the question of whether or not the king and elite converted to Judaism.

I had forgotten about the Moses coin, but this is something Stampfer addressed:

The dirhem was a widely used coin in medieval Eastern Europe and in the Khazar lands, but it was often in short supply in Khazaria. This made trade very difficult, as the Khazars often had to pay their trading partners in familiar coins. In these circumstances, Khazars, like other outlying regimes, had no choice but to mint coins themselves. In order to make certain that they would be accepted, the minters imitated the form and the appearance of the widely accepted Islamic dirhem. Minting dirhems posed certain religious difficulties for non-Muslim rulers, however. The standard dirhem had Arabic text around the edge that included... a statement that God was one and that Muhammad was his messenger... Some Christians solved this problem by adding discreet Christian symbols or miswriting the text. For Jews, there could have been another solution: to change the name of God's prophet from Muhammad to Moses. The resulting coin would look like a dirhem, and the altered text would not attract attention, especially in lands where Arabic literacy was uncommon.

Indeed, in 837/38 an imitation dirhem was minted in Khazaria with precisely this change. The writing on the edge of the coin reads, "and Moses is the messenger of God." However, though imitation dirhems were minted in Khazaria year after year, a dirhem with this inscription was minted only once. A number of these dirhems are extant, but they are all from the same emission. Subsequent Khazar emissions featured the unaltered text referring to Muhammad. But if replacing Muhammad with Moses was such a good solution to the problem for a Jewish king, there would be no reason to do this only once. Had a ruler decided to mint coins without the Islamic content, it is difficult to understand why the decision would be reversed after a year. However, if a subordinate took the initiative to mint such a coin, assuming incorrectly that the ruler, pagan or not, would not object, it would not be difficult to understand a quick reversal. Therefore it is necessary to consider the possibility that the change was initiated by a Jewish mint master.

In the Middle Ages, Jews often served as minters for non-Jewish governments, and coins with Hebrew lettering were produced in medieval Poland and Hungary. The rulers in these states had not adopted Judaism; rather, it was the minters who were responsible for the Hebrew... al-Maqrizi reports that the early Umayyads may have employed a Jewish minter named Somayr who was responsible for minting dirhems... It is certainly possible that a Khazar ruler who needed to mint dirhems would have turned to a Jewish minter, who might have thought that he could make a change in the text without difficulties. If this was the case, it is easy to surmise that the coins provoked a negative response from Muslims, from within Khazaria or from without, leading the ruler to order a return to the standard text. However, had the initiative been an ideological decision on the part of the ruler, on theological grounds, a return would have been no simple matter. The possibility that the Moses dirhem was the initiative of a Jewish minter is speculative, but a single minting of such a coin is not evidence for the Jewishness of the Khazars or their leadership. On the contrary, the quick reversal suggests the opposite.

Desailly said...

Sorry to be off topic again, but as a historian I'm just wondering what percentage of EEF (ANF) the Neolithic/Bronze Age Maltese people had, and what percentage of ANF the Minoans had? Thanks for the answer to everyone who will provide information.

Matt said...

@Davidski, I could give a list of the new samples in the anno if it helps. If there's a tutorial online I can try separating out the samples.


Offtopic of Jewish ancestry (sorry) but quick exploration of the new England_EarlyMedieval samples in the Reich anno:

Reich Lab seem to have split samples in England dated to Early Medieval period into roughly two sets, EarlyMedieval_Saxon and EarlyMedieval.

As far as I can tell, Reich Lab's coding of these seems to reflect that England_EarlyMedieval form a clade with the England_LIA, England_MIA, England_MIA_LIA samples that they've sampled, while the EarlyMedieval_Saxon are different. (Well, at least a clade with my ancient outgroups and the HO_Public). They're the more continuous population.

Seems like there are more England_EarlyMedieval than I remembered from Gretzinger's paper, but it's generally consistent with the Saxon type samples being the majority in the sampled EarlyMedieval.

However, I didn't find that qpWave involving ancient outgroups in pright, and pleft of EarlyMedieval_Saxon and any of 4 EarlyMedieval Germany groups from Saxony worked.

In the direct f4 stats, all these early Medieval German groups seemed to have problems of some excessive and unexplainable attraction to the African outgroup. Some problem like that seemed evident in the preliminary genotypes on G25 too.


On topic I don't *think* any of the new medieval Jewish samples are in this new anno.

gsueso said...

Do you think we'll ever get dna samples from Ancient Judeans or Hellenistic Jews? Here's a quote I found in an article about a recent restoration of the Jewish catacombs of Rome:

“We explained that we were there to speak on the behalf of the people buried in the catacombs, our forefathers, and that it was important to make sure that their bones were protected and not used for scientific analysis, as well as that the tombs that were still closed would never be opened,”

Maybe Erfurt could provide a precedence?

Davidski said...


Do you think we'll ever get dna samples from Ancient Judeans or Hellenistic Jews?

They're waiting to be published.

Don't expect any surprises. They're very similar to Iron Age samples from Israel.

Davidski said...


I could give a list of the new samples in the anno if it helps. If there's a tutorial online I can try separating out the samples.

Yeah, that might speed things up for me.

Not sure if there are any tutorials, but all you have to do is to convert the files to Plink files using Convertf, then with Plink extract the relevant samples into a new set of Plink files. This might help...

Rob said...

@ Desailly

I was recently looking that. It depends on set up of analysis , but minoans are ~ 85% ANF ~ 15% CHG/Iran
So solidly ANF

StP said...

@Steppe said...
I think this branch came to the Middle East with Indo-Iranian migrants and was involved in the formation of the male Levite lineage Z93 CTS6 and others like Y40 with Mitanni, I mentioned above, I think there is more genetic data to come .
November 16, 2022 at 1:29 PM

R1a-YZ94 had three chromosomal sons or three separate filial lines: Y40, Y3 and Z2124>Z2122>F1345>CTS6 (around Iran and Palestine).
CTS6 is the chromosomal great-grandchild of Z2124, not Y40
The CTS6 man may have fathered biblical Melchizedek, who lived around 3800 BP.
Melchizedek "without father, without mother, without known lineage" (as Hebr 7,3 ews would call him) could not have been a Semite like Abraham. He was probably an Indo-European who was not interested in pedigrees.
Levi, imputed to Jacob as his son, may therefore have been the great-grandson, not of Abraham, but of Melchizedek.
Ultimately, Moses, as a descendant of Levi and an earlier Melchizedek (a man without a lineage or a stray), could call himself "the son of the erring Aramean(?)".
Thus began the Indo-European lineage of Kohanim and Levites, Y2619, among the Semites.

Matt said...

@Davidski, OK, here are the new samples to the anno compared to the last previous anno (although I think relatively few will be totally new to G25) -

I'll have a look at that link later.

Matt said...

@Davidski, I wasn't able to do it quite like you suggested (mainly because unlike people who use real OS, I'm on Windows 10 which means convertf is out), but I think I was able to do something like this using the admixtools2 function eigenstrat_to_plink

WeTransfer link:

I had to split this into several files, because this seems to be a RAM intensive process where my little laptop had some trouble doing all the new samples in one go.

This is just the HO_public though, I'm aware you're probably gonna want to the 1240k, but as I just did this one anyway as I had the files around, I thought I'd just test and check that this is what you're looking for before repeating on the 1240k.

Matt said...

@Davidski, I went and did the plink conversion for the 1240k in that style anyway. Apologies if it's not useful.

As a side note though, what I realized after doing this was that the way I did this was by converting all samples where there were population labels for the new samples. This does mean that if Reich lab added more samples under an old label, you will get the old samples in this file too so it's not strictly just the new samples necessarily. Apologies if this causes any problems.

Tigran said...


When's the Egypt paper coming out? Will southern Egypt be very low in SSA?

Davidski said...


What's the direct link to the new files at the Reich Lab?

WSH said...

What I don't understand about the Khazar Hypothesis and Alanic converts, is why were Jews proselytizing to Steppe people? Isn't Judaism non-proselytizing?


The CTS6 man may have fathered biblical Melchizedek, who lived around 3800 BP.
Melchizedek "without father, without mother, without known lineage" (as Hebr 7,3 ews would call him) could not have been a Semite like Abraham. He was probably an Indo-European who was not interested in pedigrees.
Levi, imputed to Jacob as his son, may therefore have been the great-grandson, not of Abraham, but of Melchizedek.

This is an extremely interesting theory, but seems a bit too good to be true. Do you have any further evidence that Levi is the son of Melchizedek? What is the relationship between them, if any? Also, why isn't R1a as prevalent in non-Ashkenazim if this is the case?

WSH said...


Ehh... I wouldn't bet on "pristine" Jews disappearing any time soon.

While secular "Jews" do intermix heavily, they also breed lightly. Meanwhile, Haredi are the most fertile group in the United States, even beating out the Amish. These mixed marriages not only won't be breeding at particularly high rates, they'll also probably not raise their kids to be particularly Jewish. A lot of Jewish-Americans don't even understand that Jewish is an ethnic group

Kevin Brook said...

A cladologist in Anthrogenica confirmed that the ancient Koban culture male really is in haplogroup G2a-FGC1093, just as the scientists said, and added that the sample is positive for the markers FGC1116, FGC1121, and FGC1141 but negative for the marker FGC7593.

Ashkenazic carriers of G2a-FGC1093 are positive for the marker FGC1107, whereas North Ossetian and Kumyk carriers are negative for that marker. Family Tree DNA currently calls the Ossetian-and-Kumyk level directly above Ashkenazim G2a1a1a1b1a and calls the Ashkenazic child level G2a1a1a1b1a1.

A new YFull carrier of G2a-FGC1093's parent, G-FGC1144*, is a modern Adyghe (customer # YF110845), so yet another Caucasian in this cluster of haplogroups.

WSH, there were some Jewish missionaries in ancient and early medieval times. Aside from the case of Khazaria, where the successful missionary was supposedly named Yitzhak ha-Sangari, Jewish proselytizers were active in Himyarite Yemen, Adiabene, southern Europe, and even in 18th-century Poland when it was dangerous to convert. I wrote about these cases in my Khazaria book.

The claim we sometimes read that "Jews never ever proselytized" is the same kind of sophistry as "Khazars never converted to Judaism", and in both cases contradicted by documentation and by genetics.

Aram said...

R1b -P312 in Koban? Well if not an error then this might be from Bastarnae or related Celtic people who reached Pontic Steppe.

Matt said...

@Davidski, the tarball on their site is:

StP said...

@WS said...
StP wrote: „Levi, imputed to Jacob as his son, may therefore have been the great-grandson, not of Abraham, but of Melchizedek.”

/…/This is an extremely interesting theory, but seems a bit too good to be true. Do you have any further evidence that Levi is the son of Melchizedek? /…/ 20 November 2022, 18:55

1p.) Who was the father of Levi, the progenitor of the priestly Levites?
Let's look to the biblical Book of Genesis (cap. 29 ff.)
Here is the grandson of Abraham's Pariarch and father of Levi, Jacob (alias Israel), for seven years he served to Laban, his uncle, to earn his beloved and beautiful daughter Rachel. When the wedding ceremony took place after seven years of service, on the wedding night Laban put his unloved and older daughter Leah to Jacob's bed. Jacob "came close" to her (Heb. wajjabo, meaning he had sex). But in the morning, when he sobered up, he saw that he had been deceived. Laban, conciliatory, again promised him this beloved Rachel, but for another seven years of service and finally gave him Rachel as his wife.
Over the years, Jacob was somehow (Heb. malle) with his first wife, Leah, but he probable did not "approach" her, so she remained childless, and perhaps also barren. Meanwhile, the Lord (JWWH/Adonai), knowing that Leah had been put aside in favor of Rachel, "opened her womb" and so she gave birth to Reuben, Simeon, Levi and Judah in turn.
Note that in Genesis it is not written who approached Leah, i.e. who was the father of her sons, including Levi: if not Jacob, then who? Is it possible that under the name of JHWH/Adonaj the editor hid some servant of religious rites, some sacral person? Jacob, however, somehow accepted that if they were the sons of his married wife, so he, as if the father, were assigned in the genealogy.
(to be continued)

StP said...


2p.) Jacob treated his sons unequally: he favored Rachel's sons, Joseph and Benjamin, but seemed to disregard Leah's sons, including Levi. Until his death, he remembered their transgressions, and in his will, instead of blessing, he deprived the firstborn Reuben of due priority, and deprived the families of the next two, Simeon and Levi, of the right to inheritance, announced their dispersion, denied them the right to family honor and cursed their vengeful acts against the house of Shechem, to which they had a traditional right for insulting their sister Dinah by rape (Genesis r.49).
So it is strange that despite the unclear genealogy of Levi and depriving him of the material and spiritual heritage of the Jacob family, his descendants were chosen by the biblical Lord (JHWH/Adonai) as a priestly people in the service of the Israelite sacrum as Levites and priests/kohanim for all time.
What made it so? Most likely this was due to some connection between Levi and the priesthood, initiated by Melchizedek. The figure and priesthood of Melchizedek, known from the Book of Genesis (v. 14), Psalm 109 and the Letter to the Hebrews, lived in the tradition of Israel as a figure of Christ's priesthood. The links between Levi and his grandsons, Moses and Aaron, and the priesthood of Melchizedek were probably blood ties that were associated with the action of the Lord (YHWH/Adonai).
In the Bible, Melchizedek was considered in Israel as a man "without father, without mother, without genealogy", which means he was certainly not a genetic descendant of Shem, i.e. a Semite, like eg Abraham. I imagine him rather as an earlier shaman among the Indo-Europeans from the so-called Asian line R1a-Z93….F1345>CTS6>Y2619. Melchizedek time ~1800 BC corresponds to the presence in Iran and Palestine of SNP F1345 in the MRCA stage and CTS6 in the formation stage.
He probably belonged to a group that migrated through the Central Asian Wusun (acc. to Tomenable calculator) south to India, but his line strayed west to Mesopotamia, where at the city of Ur on the Euphrates where he met Abraham at the time of his conversion. Malchizedek probably also experienced a conversion to Abrahamic monotheism and came with him to Canaan in Palestine, where, in gratitude for the victory won by Abraham, he made the first sacrifice of bread and wine in honor of the Most High God, which became a prefiguration of Christ's sacrifice.
In this way, evidence from the Bible and science can explain the presence of the Indo-European haplogroup R1a in the priestly-Levitical lineage among the Semitic Jews.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

What do you think of this?:

from this new paper:

from new study on relationship between language and DNA

For Indo-European, we consider the classic dichotomy between the old chronology / Anatolian
hypothesis at ~8,000 years ago (54) and the recent chronology / Kurgan steppe hypothesis 5,500
– 6,500 years ago (55). The genetic divergence time seems quite old overall, not fitting with the
recent chronology but exceeding the limits of the old chronology as well – while the harmonic Ne
estimates could also be included in the recent chronology time frame.

The mean genetic divergence time they have for the Proto Indo European population is 5500BCE.

Andrzejewski said...

@SIP You’re taking the bible WAY too seriously here.

The only things *I* do is attempt to find echoes of historical events in some OT stories, but you’re taking it over the top. You’re taking it too literally.

For example, trying to identify “Abraham” with some CHG/Iran chl -rich pop that migrated to Canaan and admixed into the Levant N PPNB and PPNC and gave rise to the Canaanites is legit; by the same token trying to identify Jebusites with an offshoot of Mitanni and suspecting what some bible scholar and historians think was a strong Indo-Iranian influence on fledgling Judaism (reinforced a few centuries later by strong Zoroastrian influences during the Babylonian exile) is one thing. Treating the bible like its the word of god with no critical thinking was completely different.

Rob said...

Sorry for OT
@ Aram

Which is P312?
Koban is too early for Basternae or Celts

StP said...

@Andrzejewski said...
"StP: You’re taking the bible WAY too seriously here."

I don't think you understood something. Although the Bible is treated as the Word of God by Judaism and Christianity, it is also a collection of products of various authors or editors from different times and regions, who often use various materials to convey religious and non-religious content: their or other people's idee, facts or historical traditions. customs of the time, myths, legends, parables, fairy tales, poetry, prayers… at the same time, they use various literary genres. So each part of the Bible requires individual treatment. Above all this, there is a question: what did the author or editor intend to convey to us in this fragment or in its entirety?
Thus, world science does not underestimate biblical texts (as anti-religious agitators do).
My task was to give a cold answer whether the texts of the biblical family or religious tradition somehow correlate with the results of genetic research on the Indo-European origin of the Levi patriarch and the priestly line of the Levites in the Semitic line of the Partiarchs of Israel, and whether there is any suggestion, indication of such an origin of the Levites - in the biblical records.
I am not interested in the origin of Semitic Abraham.

Labayu said...

Regarding proselytization in Judaism. Modern Judaism is certainly non-proselytizing. I don’t know if it’s known when that really became solidified, but throughout most of Jewish history proselytization wouldn’t have made sense either ideologically or politically.

Judaism grew out of the state religion of the Kingdom of Judah. This is the reason that a convert is referred to as foreigner (gêr). This word is often translated as “stranger” in English Bibles, but in Leviticus 19:33–34 for example, it is literally “And if a foreigner lives with you in your land, you shall not mistreat him. As one born among you, you shall treat the foreigner who lives among you…” In the post-exilic period when the religion began to take shape in absence of the state, one of its greatest concerns was about who had the right to the land of Israel. In this context, deliberately expanding those who could claim the land as their inheritance wouldn’t make any sense.

Judaism again became a state religion under the Hasmoneans. According to Josephus, John Hyrcanus gave the conquered Idumeans the choice of living by Jewish law or expulsion. It is not unreasonable to characterize this as a forced conversion, but that may not have been exactly how Jews at the time would have understood it. These Idumeans were people who lived in what had been the southern portion of the Kingdom of Judah, which had not been incorporated into the Babylonian province of Yehud (Judea). They would have largely been descended from the Judahite population, with some ancestry from the Edomites who had opportunistically taken control of the region in the wake of Judah’s destruction. As a consequence, they may have been seen not so much as foreigners, but rather as Jews who were in rebellion. It is notable that Hyrcanus carried out this sort of policy with other populations who could been seen as “children of Israel” in rebellion, whereas this was not done with the Hellenized Canaanites who lived on the coast. He destroyed the Samaritan Temple, and yet allowed “pagan” temples to persist on the coast.

The stated religious reason Christians and Muslims proselytize is alien to Judaism. According to the Talmud, righteous gentiles have a place in the “world to come” (Sanhedrin 105a). Although the Talmud is only relevant to Rabbinic Judaism. There was a period of time between the Jewish rebellions against Rome and the rise of Rabbinic Judaism as the dominant form of Judaism when the religion may have been more open proselytization. At least that has been argued. I know less about this period, but the conversion of the Himyarite kings falls into it.

I’m not sure how one could distinguish between genetic evidence of accepting converts plus the occasional rape versus genetic evidence of proselytization. It does seem clear that Jewish men in the diaspora married non-Jewish women. Since their descendants were considered Jewish, I suppose we can assume that these women were converted. Although, that depends on when Judaism became strictly matrilineal, and I don’t know whether or not there is textual evidence for this practice. In any case, converting women to marry is not really what I would consider to be proselytization.

Andrzejewski said...

@SIP “ I am not interested in the origin of Semitic Abraham.”

“Abraham” may have been a figment of the imagination. Some scholars link the Patriarchs with Mitanni or Nuzi, aka Hurrian/Indo-Iranians.

However, genetically this ChG or Iran Chl rich pop distanced Levantine pops from otherwise more European related populations like the blue eyed, fair skinned Hap T Peki’in cave dwellers who were overwhelmingly ANF.

Labayu said...

Following up on my previous comment regarding the Ḥimyarite conversion, it seems to have been more a case of influence than the result of proselytization:

According to the Arab-Muslim tradition, … the conversion of Yemen to Judaism occurs during the reign of… Abīkarib As‘ad (r.c.400–445), who in Arabic is given the name Tubba‘, As‘ad Tubba‘ or Abū Karin As‘ad al-Kāmil (‘the perfect’), a heroic figure who also recognised the sanctity of the Meccan shrine before being assassinated by his own people. Some of these later sources also mention that this conversion was at the initiative of two Jewish sages whom the king brought back from Yathrīb with him.

Though the latter story is most likely apocryphal, we still have to ask: why, in the mid-fourth century, do inscriptions invoking polytheistic deities suddenly give way to ones that mention the high god, Raḥmanan? …

While it is clear that the Ḥimyarite kingdom converted to some form of Judaism based on numerous Sabaic inscriptions in the area, and from external sources that attest to this, we have very little idea of the contents or contours of this ‘Judaism’. Robin notes that around 380ce inscriptions cease to reference the polytheistic deities of South Arabia, and begin to refer solely to Raḥmānān (‘the Merciful One’; al-Raḥmān in Arabic). Though these ‘post-conversion’ inscriptions refer to a monotheistic deity, it is worth noting that they nonetheless remain in the Sabaic language of South Arabia. Some have argued that Raḥmānān might be related to the Hebrew raḥamim, and/or it might function as a precursor to Allah, the high god of the North. But it is not at all clear if these, to use Beeston’s locution, ‘Ḥimyarite Raḥamanists’ in South Arabia are synonymous with ‘Jews’.

“South Arabian ‘Judaism’, Ḥimyarite Raḥmanism, and the Origins of Islam”
From the book Remapping Emergent Islam
Aaron W. Hughes

Some have argued against considering the Ḥimyarite religion to have been Judaism at all, and that it's better to think of it as a sort of proto-Islam. As far as I know, they are never mentioned in Rabbinic sources, so they may not have had any connection to what became normative Judaism. That said, some heterodox Jewish groups were eventually brought into the Rabbinic fold, so that may be irrelevant to genetic ancestry.

Andrzejewski said...

@Labayu Yemenite Jews are Himyarites

Andrzejewski said...

Speaking of Himyarites, Amhara means Himyatites, snd it refers to Semitic Ethiopian elite. They dont see themselves as African Nilotes but as Semitic. Racism in Ethiopia is just as bad as in the US Deep South. Jamaican Rasta refer to Ethiopia as the homeland of black race, but Amhara see “Shenkala” ad inferior

StP said...

@Andrzejewski said...
@StP “ I am not interested in the origin of Semitic Abraham.”
“Abraham” may have been a figment of the imagination. Some scholars link the Patriarchs with Mitanni or Nuzi, aka Hurrian/Indo-Iranians. However, genetically this ChG or Iran Chl rich pop distanced /…/

The multitude of various author's ideas about the origin of Abraham are against them.
Why torment and stress the brain so much, if the biblical text itself does not raise any doubts?

Andrzejewski said...


Labayu said...


I assume that Yemeni Jews are descended from Ḥimyarites. As far as I know, their ancestry looks mostly local, so that would be the simplest explanation, but unlike the Ḥimyarites, their religion is mainstream Rabbinic Judaism. That’s what I was alluding to at the end of my last comment. They’ve obviously been influenced by other Jewish groups since late antiquity. In contrast, Ethiopian Jews were until very recently non-Rabbinic.

StP said...

@Andrzejewski said...
StP, November 24, 2022 at 4:51 AM

I presume you gave me this material on the tablets of Nuzi for further knowledge of the ancient legal customs in the time of the patriarchs Abraham, Isaac and Jacob. Thanks!

Steppe said...

In any case, the early Israelites arose from two populations that contribute to ethnogenesis, the Shasu from the Sinai Peninsula and the Hebrews from Mesopotamia, which then merged into Caanaite territory.

Andrzejewski said...

@SIP “ I presume you gave me this material on the tablets of Nuzi for further knowledge of the ancient legal customs in the time of the patriarchs Abraham, Isaac and Jacob. Thanks!”

No. The Patriarchs *were* Hurrian, not Semitic.

Jebusites were from Mittani.

Judaism is thoroughly Indo-European-influenced.

The tripartite caste division of ancient Israelites is similar to the 4 castes in Hinduism, the pre-French Revolution stratified society made up of nobles, priests and commoners.

IA Israelites were also pastoralists.

All these traits and feats above point and allude strongly in an IE direction.

In addition, several biblical figures- Esau, King David and Jesus were described as having red hair.

Andrzejewski said...

@Rob “ When we we return to interesting topics instead of made-up Bible people. ?”

I’m sure that they are made up but we can’t deny strong IE impact on Judaism, including the name “Yahweh”.

Jews and Samaritans have strong European admixture because of Kura Araxes, Philistines, Greek, Persian and Hittites.

While I don’t regard the bible as anything of authority, I do believe to a certain degree that some writings therein reflect some historical events, including wars, migrations, exiles, cultural influences or admixture events, or at least bear some distant echos to them.

To me the story about “Abraham” may reflect some author’s memory of a founding effect by a CHG/Iran Chl immigrants who intermarried with locals starting 4,200YBP.

Steppe said...

Andrzejeawski , red/blonde hair does not necessarily have to indicate IE in Israelites because the mutation for blond hair arose independently of IE in the Levant in the Chalcolithic maybe Jesus was a half-breed what some claim , allegedly the father would have been a Roman legionnaire but speculation and no evidence

Andrzejewski said...

@Steppe “ red/blonde hair does not necessarily have to indicate IE in Israelites because the mutation for blond hair arose independently of IE in the Levant in the Chalcolithic maybe Jesus was a half-breed what some claim , allegedly the father would have been a Roman legionnaire but speculation and no evidence”

I am starting to question anything that Lazaridis and the BROAD lab in general have published (IIRC, he used to model Yamnaya as 1:1 EHG:CHG), although I do remember his peer reviewed paper fro
2017 or 2018 re: Peqi’in; the mostly ANF (PPNB/PPNC?) pop was found to have had blue eyes, red hair and Ydna Hap T (a neolithic farming marker). So there might be some grain of truth to it about ANF having hereditary recessive alleles to these pigmentation traits, along with the WSH.

StP said...


In my (and for Christians) opinion, the Bible is a collection of inspired books about God and man and our destiny to eternal life with God.
But God works in history, so there are also historical texts in the Bible, among other things.
But do you think the correlation of biblical events with extra-biblical events contradicts the truth of biblical facts?

Andrzejewski said...

I don’t believe in “god”, I believe in genetics. But even the latter is being shaken by Reich lab’s theory of PIE origin lately

Tigran said...


I believe someone on this blog said there would be a study showing 26,000 year old samples from the Caucasus had the genes for light skin, eyes and hair. Do you know anything about that study?

Steppe said...


Eastem hunter-gatherers lightintermediate skin
SLC24A5 + SLC4542 variable eye color some with blondo hair
KITLG Is a ANE Componet .

Caucasus hunter-gatherers
~ 13,000 yas
origin of blue eyes
- 42,000 ya
Intermediate & light skin
SLC2445 + SLC4542
- 22,000 - 28,000 ya

light/ intermediate skin
Anatolian SLC2445 + SLC45A2 farmers mostly brown eyes
dark hair

Steppe said...

And in addition, the Israelites later absorbed the Philistines, also known as the so-called "Sea Peoples" who mainly come from the Aegean region and partly from the Balkans, so there is also a slight genetic IE input! but I am guessing that these Aegean populations also had only a small IE heritage maybe some more ( Luwians ??? )

StP said...

@Andrzejewski said...
I don’t believe in “god”, I believe in genetics. But even the latter is being shaken by Reich lab’s theory of PIE origin lately

After such a statement, it is worth asking in what god and in what genetics you do not need to believe?
I also don't believe in some gods...and some genetics

Labayu said...

The Philistines probably contributed very little (if any) Aegean genetic ancestry to the ancient Jewish population.

The five major Philistine urban centers were Ashdod, Ashkelon, Ekron, Gath, and Gaza:

– Ashdod was completely destroyed by the Egyptians in 950 BCE and was not rebuilt until at least a century and a half later, which suggests that few survivors were left in the region. Ashdod was again completely destroyed by the Assyrians in 713 BCE, and the survivors were resettled in Media and Elam.

– Ashkelon was completely destroyed by Nebuchadnezzar II in 604 BCE, and the survivors were reportedly resettled elsewhere in the Babylonian Empire.

– Ekron was completely destroyed by Nebuchadnezzar II in 604 BCE, and was never significantly resettled.

– Gath was completely destroyed by Hazael of Aram-Damascus around 830 BCE, after which it was never significantly resettled.

– Gaza was captured by Alexander in 332 BCE. The men were killed while the women and children were taken as slaves.

It was only hundreds of years after all this had happened that Philistia finally came under the rule of the Jewish Hasmoneans during the First Century BCE. Plus, we know from Feldman et al 2019 that by the Iron IIA (c. 1000 BCE – 900 BCE) the Philistines already had very little “European-like” ancestry left.

Davidski said...


Sorry about the lack of blogging and email replies, but the World Cup is on.

Once Australia and Poland are knocked out, which no doubt will be very soon, I'll update the G25 and drop and couple of new blog posts.

Steppe said...


you are right, I have also seen the study, but new results could also come in a few years, I think the Philistines of the Aegean area were a small minority in the region, which would quickly be absorbed by the Canaanite population, also from a cultural point of view View, not all of the so-called "Sea Peoples" were foreigners, some were of local origin who had to change their way of life with the political structures and also due to climate change at the end of the Bronze Age and this also led to the partial collapse of the Hittites and the weakening of Egypt but pave the way for Phoenicians, Israelites and Greeks. There were several factors at play, without the Bronze Age apocalypse there would have been no Israel, in my opinion.

Palacista said...

New Scientist have an article on "The hunt for the lost ancestral language of Europe and southern Asia". A quick skim showed up the name Lazaridis and a reference to the Southern Arc paper. Anyone had a strong enough stomach to read it through?

Norfern-Ostrobothnian said...

Can you do these Khasi samples? It seems they are new ones separate from the Estonian Biocenter ones.

Desailly said...


Once Australia and Poland are knocked out, which no doubt will be very soon

Hehe, not so soon😉

Davidski said...


OK, hang on, I'll check it out.

Matt said...

Props to Erik(Arial) and Joshua for becoming contributing authors on the published Erfurt paper - (Carmi is one of the cooler adna authors for being engaged to listen).

Richard Rocca said...

Very interesting revelations about the upcoming Aegean paper, especially about the arrival of Steppe ancestry and the difference between the DNA in the Mycenaean ports versus interior:

Ancient DNA reveals admixture history and endogamy in the prehistoric Aegean

The Neolithic and Bronze Age were highly transformative periods for the genetic history of Europe, but for the Aegean -a region fundamental to Europe’s prehistory- the biological dimensions of cultural transitions have been elucidated only to a limited extent so far. We have analyzed newly generated genome-wide data from 102 ancient individuals from Crete, the Greek mainland and the Aegean Islands, spanning from the Neolithic to the Iron Age. We found that the early farmers from Crete shared the same ancestry as other contemporaneous Neolithic Aegeans. In contrast, the end of the Neolithic period and the following Early Bronze Age were marked by ‘eastern’ gene-flow, which was predominantly of Anatolian origin in Crete. Confirming previous findings for additional Central/Eastern European ancestry in the Greek mainland by the Middle Bronze Age, we additionally show that such genetic signatures appeared in Crete gradually from the 17th to 12th centuries BC, a period when the influence of the mainland over the island intensified. Biological and cultural connectedness within the Aegean is also supported by the novel finding of consanguineous endogamy practiced at high frequencies, unprecedented in the global ancient DNA record. Our results highlight the potential of archaeogenomic approaches in the Aegean for unraveling the interplay of genetic admixture, marital and other cultural practices.

Matt said...

@Richard, related

"On 11-13 December 2022 an international online workshop entitled Death, Rituals and Symbolism in Prehistoric Aegean will be held by the International Association for Archaeological Research in Western and Central Asia (ARWA). Further information is available at "

"P. W. Stockhammer, E. Skourtanioti, and H. Ringbauer

“Social Belonging in Bronze Age Greece from an Integrative Bioarchaeological Perspective on Burials”

Y. S. Erdal and M. M. Koruyucu (present in adna papers)

“Changes in Burial Rites Reflected on Bioarchaeological Data in Western Anatolia: A Case Study on Early Bronze Age Cemeteries at Bakla Tepe"

Andrzejewski said...

Im surprised that they have never mentioned anything about Khazar, Slavic (Kna’anim) Jews and other intermarriage events with non-Jewish pops other than to just claim that “they merged” to form modern Ashkenazi Jewish dna.

I’m still at a loss to understand how Yiddish won over Polish in Eastern Europe and what made Sephardic Jews assimilate.

Not a single word about neither Khazar nor Berber contributions.

And, since Jews were underprivileged in Christian Europe, we’d expect to find more Slavic DNA compared to maternal, which should mirror patterns prevailing among African Americans and Native Americans today, plus hispanics.

Andrzejewski said...

OT Just read a Lazaridis study (I know, I know…) from 2021 on the Medieval Slavicization of the Byzantine Empire during 500-1000AD, and one the underlying finding was a tremendous reduction of Anatolian Farmer component in favor of a huge increase in WSH one; it was written that pre-Justinian Plague SE Europe was the highest in AFS while non-Balkanic Slavs have the highest WSH and the least ANF at the same time (seems as if their WHG is much higher than Western Europeans).

Can anyone comment on these alleged findings?

(I’m aware that Balts have highest WHG - up to 33%, and that Ukrainians are an almost even/equal three way of WHG:ANF:WSH).

Rob said...

Stockhammer's talk was great.

The data from Mygdala shaft grave is interesting, 1 x G2a and 3 x Y-hg J2b2 boys linked into a broader household.

Matt said...

Re; Stockhammer, as I remember from watching that, it was a polygamous household and they were all the same J2b2's grandpatriarch's grandchildren, except G2a lad, who was a nephew by marriage; one J2b2 father married a women who migrated in, then the boy is her sister's son via a G2a? G2a man is not present at the site, perhaps because he'd died elsewhere, while the sister was, so maybe something happened with the man and the sister and her son were adopted into the household, although that's probably pushing the information.

I think the findings around cousin marriages are interesting; it's not really clear whether they dated only before or also during the steppe ancestry period, I wonder if we will generally find that these more frequently coincide with polygamy and/or rugged/insular terrain.

Rob said...

Yes it shows how different clans might have merged in developed socieities.
Additionally, the finding of J2b2 (known for its Cetina association) itself seems to be a novely in a B.A. Greek context.
Seems like all sorts of things were happening in Greece & the Aegean around ~ 2500 - 2200 bc, which would probably be missed if analysts only engage in genome-wide modelling, without the fine-tuning direction of Y-DNA and archaeology.

Bob said...

Now the Poles & Aussies are out, get moving.

Davidski said...

I reckon I caught Covid while watching a World Cup match at a sports bar a few days ago. Lots of people here are catching the new variant.

I'm fully vaccinated, so it shouldn't be a huge problem, but it's making me tear up and my nose is blocked.

Hopefully it clears up in a couple of days.

Andrzejewski said...

USA is out too. And Poland of course. Maybe next WC…

Richard Rocca said...

99% of the USA's top athletes play American Football, Basketball or Baseball. As much as I like watching European club football, the USA will never be great at it. Just the reality of it.

Gaska said...


if analysts look to Ychr, then the Balkan IE languages including Mycenaean do not have their origin in the steppes but in Anatolia and J2b-L283 may be the ultimate demonstration. Only finding R1b-Z2103 in the Mycenaean culture can keep alive the steppe origin.

Steppe said...


I had already had COVID twice, the first time it was a bit bad, the second time almost nothing and vaccinated but everyone has to know for themselves, because unfortunately there are also vaccination damages.

Get well

Davidski said...


Everyone accepts that Greek came from the steppe.

You're the only person I know who is still debating this.

Rob said...

@ Gaska

Unfortunately you lack basic understanding, even when it’s spelled out for you.
As I said, there are multiple things happening in southern Greece, but Mycenaeans aren’t protoGreeks. They’re southern, admixed Greeks
There’s steppe related I2a in Thessaly, maybe others will turn up too. Then sure there’s native G2a, Anatolian J2a , Cetina related J2b.
So there were dual Balkan movements into Greece.

Andrzejewski said...

Gaska might be a partial Sephardic 😜

Rob said...

@ Richard

Baseball requires athleticism ? It’s even slower than cricket

LivoniaG said...

Matt wrote: "it was a polygamous household and they were all the same J2b2's grandpatriarch's's not really clear whether they dated only before or also during the steppe ancestry period..."

Just a simple question. How did that J2b2 family get there? If it happened before the steppe's ancestry period, where would that J2b2 have originated from before it got there? And if it's dated during the steppes ancestry period, how would the path be different?

I think I'm a fan of Black Sea migrations, but I don't understand where these people were supposed to be before they got to Greece, Crete, the Aegean by Y-DNA, etc.

Gaska said...


Sorry I had forgotten that you know everything, next time before giving an opinion I will have to consult you. Those who according to you are southern admixed Greeks what language did they speak? Mycenaean right? and that is the oldest attested IE language in Europe right? And what male markers are there in Greek territory including the Peloponnese since Neolithic times? I tell you-G2a2a, G2a2b, J2a-L26, J2b-L283, T1a2b-S27463, L2-L595 and J1, plus R1b-PF756 in the Late Bronze Age when the Mycenaean culture had already collapsed (1. 135 BC) and one I2a2a1b in Theopetra (2.200 approx)

This is the evidence linked to the male markers to say that the Mycenaean has its origin in the steppes? me that is just wishful thinking.

J2b has its origin in the southern Arc and we have L283 in Kabardino Balkaria before Cetina and a little later than in Serbia. How do you think that lineage reached the Balkans? from the BB culture through Cetina? from Yamnaya through Catacomb or any other subculture of the steppes?- All those markers have been found in the Mycenaean culture and none have been found in the Yamnaya, CWC or BBC- I repeat only by finding Z2103 in the Mycenaean culture you can establish a paternal lineage link between that culture and the steppes.

Gaska said...


Well, years ago, I was also the "only person" who said that Yamnaya has nothing to do with L51>L151 and that P312 is western.

Lazaridis is doing mental gymnastics to connect the Mycenaean with the Yamnaya culture, but we already know that Harvard does not learn, and they are making the same mistake over and over again, there is no genetic continuity by male line between Yamnaya and the Mycenaean culture either. Am I really the only person in the world who has realized this? or is it that the rest want to believe in something without evidence for it. Just find Z2103 or even I2a-L699 in Greece (Chalcolithic and Early Bronze Age) and I will change my opinion.

Rich S. said...


"Well, years ago, I was also the "only person" who said that Yamnaya has nothing to do with L51>L151 and that P312 is western."

Yet R1b-L51>L151 has been found in very early Corded Ware, and R1b-L51>P310 has been found in Afanasievo, both loaded with Yamnaya DNA, the latter virtually identical to Yamnaya. If those folks weren't directly descended from Yamnaya, they certainly were descended from the same steppe source population as Yamnaya. That is incontrovertible.

The Eurasian steppe is a big place. The surface of it has barely been scratched in terms of ancient DNA.

As far as you being the only person who said P312 is "western", I guess that depends on what one means by western. After most others had abandoned the idea, you began some years ago by defending the erroneous notion that R1b as a whole spent the LGM in the Franco-Cantabrian LGM refuge. Then you switched and said that it was P312 that originated in Iberia. When that proved untenable, you dug in your heels and said it was DF27 that arose there. Maybe that's still your position. I don't know (and don't care).

I think Papac et al may be right, i.e., that P312 arose "somewhere between Bohemia and England, possibly in the vicinity of the Rhine". Still, R1b-U106 has been found in early Bohemian Corded Ware. It wouldn't surprise me to see its brother, P312, turn up in that neighborhood, as well (perhaps in Poland). To those of us for whom "western" is not loaded with all the baggage you bring to the word, that's not a problem: we can still see that P312 is descended from an eastern European patriline of steppe pastoralist origin.

vAsiSTha said...

"Everyone accepts that Greek came from the steppe.
You're the only person I know who is still debating this."

Heggarty does not.

As for me, I do favour a steppe origin, but must say that the steppe % is quite low, much lower than the Iran/CHG %. However, the non IE substrate in Homer's work is quite large (>60% of toponyms are non-IE as compared to ~0 for RigVeda), so its possible that non-IE speakers changed language over time.

But an Iran/CHG origin via the Anatolian region would require a 3000BCE or earlier origin for Greek because bronze age migrations by Anatolian speakers cannot be the carriers of Proto-Greek.

Andrzejewski said...

@Vasistha “ However, the non IE substrate in Homer's work is quite large (>60% of toponyms are non-IE”


vAsiSTha said...

Although Margalit Finkelberg (2001), makes the case that Linear A, attested in Crete and Mycenae, is an Anatolian language.

vAsiSTha said...



Here is my reference.

From "Quest for the origins of the Vedic culture - Edwin Bryant"

"As a point of contrast, classical Greek maintains only 40 names (from 140 toponyms in Homer) that are Greek from the point of view of the classical language (but not necessarily Indo-European, that is, they are new formations adopted after the break up of Proto-Indo-European): barely one- third of the total. The remaining two-thirds are etymologically obscure (Hainsworth 1972, 40). Such obscurity gives clear indication of a pre-Greek, non-Indo-European substratum."

epoch said...


"Margalit Finkelberg, makes the case that Linear A, attested in Crete and Mycenae, is an Anatolian language."

Yet it remains untranslated, after 20 years.

Rob said...

@ Gaska
Well you still haven’t manage to wrap your head around the large Iberian transition associated with incoming P312 and Beakers (which is far simpler), so I’ll defer to my own judgement on the Balkans instead of your trolling

Gaska said...


It would be interesting to know your opinion about the events in the Balkans and Anatolia, if your conclusions are as wrong as the ones you have about Iberia or the origin of M269 you will have to think about looking for another hobby

Richard Rocca said...

Interesting new paper with earlier EHG in Iran:

Our new genomes from Shah Tepe (n = 9; ca. 5,100 BP ), from Northeast
Iran near the Caspian Sea, could likewise be modeled as admixtures
of Zagros Neolithic-related (76%), Anatolia Neolithicrelated
(13%), and EHG-related (11%) ancestries (Figures 3
and S3). Notably, Anatolian-related ancestry was lower in
Shah Tepe relative to Zagros populations (Tepe Hissar and Hajji
Firuz), in support of a west-to-east gradient of Anatolian admixture.
Further, the Shah Tepe genomes present the earliest
indication of EHG-related ancestry in Iran
, which is consistent
with material culture records from Northeast Iran during the
Chalcolithic and BAs showing Central Asian cultural influences,
including in Shah Tepe41,42; this supports the notion of EHG
influx in Iran via Central Asia instead of the Caucasus.

Michalis Moriopoulos said...

@Richard Rocca

I assume that's just WSHG admixture typical of Turan.

Copper Axe said...

@Richard Rocca
I doubt the EHG is actually EHG proper rather than just ancestry from central asian ANE-rich populationsn, which carried EHG/EHG+CHG-like ancestry. Not taking those into account and you can make EHG or steppe pop up in most samples from southern Central Asia. Alternatively it could also be excess ANE unrelated to this like in Sarazm.

Are you eventually adding the Tashtyk samples (the one that has some coverage at least) to Global25? Don't leave my avatar hanging ^.^

Richard Rocca said...

In fairness to them, they did say it was "EHG-like"

Rob said...

Hrvatska ! What a win

Rob said...

They should have just written WSHG, as advised by one of the co-authors. There's also probably IVC-like admixture in ShahTepe too.
I'm not sure about the Y-hg O in LBA Georgia, either.

Davidski said...

So these clowns mixed up EHG with Central Asian WSHG?

Andrzejewski said...

Is Kelteminnar the “purest” WSHG culture, genetically?

rozenblatt said...

@Andrzejewski As far as I know, ancient DNA from Kelteminar was not analyzed thus far. Moreover, I cannot remember any Mesolithic DNA from Central Asia.

Andrzejewski said...

@rozenblatt We know that Kelteminnar was WSHG.

Dunno why Botai was classified as such if it was at least 45% East Asian, with all ydna being N, R1b and C.

Rob said...

Some academics seem pressured to replicate others (“experts”) findings . IMO a better approach is to challenge and come up with novel, more interesting findings, if that’s where the evidence leads
A major limitation remains with the methodology- which i call “autosomic paeudoanthropology” which consistently ignores ydna and archaeology. It’s a poor benchmark, which as we’ve seen- has brainwashed even commentators here
But it’ll change

vAsiSTha said...


"We know that Kelteminnar was WSHG."

We don't know any such thing because aDna isnt yet available. More likely than not, Iran related SC Asian ancestry was present as well.

"Dunno why Botai was classified as such if it was at least 45% East Asian, with all ydna being N, R1b and C."

Botai has less than 10% additional East Asian ancestry over WSHG (Tyumen).
For most intents and purposes, both can be clubbed together. Similar can also be said about Tarim_EMBA, although the ANE source of Tarim seems to be much more drifted.

Davidski said...


Preliminary G25 coords for the samples from the new Peltola et al. paper.

Link to the paper...

Matt said...

Thanks Davidski. I don't have normal computer access so can't deep dive into this but the SHE008 individual seems to be Western European, kind of close to some Dutch / Germans. The paper notes:

"Similarly, one individual from Shekshovo 9 (SHE008) falls closer to Central and West Europeans than the rest of the Shekshovo 9 individuals. Most strikingly, both individuals from Shekshovo 2 (SHK001 and SHK002) fell far from other individuals in our dataset, closer to East Asia and the ‘‘forest-steppe cline’’ in the Eurasian-wide PCA (Figure 1C)"

"Grave 8/2 (SHE008) - Discovered in 2017. Poorly preserved separate human skull in the northwestern corner of the burial pit of inhumation grave 8/1. The skull of a 20–35 year old adult male(?) individual originates from a disturbed burial. Dating: the 11th–12th century."

Other somewhat outliers (not including those on the Uralic cline):

KED004 - looks mixed between Europe and Caucasus. Early Modern period.
GOS003 - "Grave 4 (GOS003) - A rectangular burial pit contained poorly preserved skeletal remains of a male individual of about 30 years old lying supine, head to the southwest, right arm along the body" ... "The dating of the graves to the 13th–15th centuries is defined by their stratigraphic position.". Ukrainian or Southern Polish like person?

vAsiSTha said...

Target: BOL009:BOL009
Distance: 2.1062% / 0.02106196 | R4P
49.2 Baltic_EST_BA
24.0 RUS_Pokrovka_Sarmatian
14.6 RUS_Sintashta_MLBA
12.2 Nganassan

Target: BOL005:BOL005
Distance: 3.0124% / 0.03012356 | R4P
61.4 Baltic_EST_BA
22.6 RUS_Caucasus_Sarmatian
13.8 Nganassan
2.2 RUS_Catacomb

Target: BOL001:BOL001
Distance: 3.0933% / 0.03093262 | R4P
53.0 Baltic_EST_BA
26.0 KAZ_Saka_TianShan_IA
13.2 Corded_Ware_Baltic
7.8 Nganassan

Interesting mix of baltic, saka, CW and Uralic ancestry 200km from Moscow in the iron age ~200ce (first runs, models are not fine tuned).

vAsiSTha said...

Yo davidski

The coords of the iron age site on google maps gives this name "Davydovskoe Bol'shoe, Ivanovo Oblast, Russia, 155007". Any relation?

Davidski said...

Yes, obviously my relatives colonized the area during the Middle Ages when they moved there from somewhere far to the west.

See that's why the people there now speak Slavic, even though there were Uralic and Turkic speakers there before them.

Davidski said...


Preliminary G25 coords for the samples from the new Koptekin et al. paper.

Link to the paper...

Michalis Moriopoulos said...

Thanks as always, Dave.

For spatiotemporal Near East paper, here's quick-and-dirty West Eurasian PCA, Basic Components, and Distances runs:




ambron said...


"See that's why the people there now speak Slavic, even though there were Uralic and Turkic speakers there before them."

And Baltic speakers, as evidenced by toponyms and 50% Baltic admixture.

Rob said...

Theres a bit more detail to be gained from these data.
One small Eg. Shamaki Iron Age has Siberian-like ancestry via Lola


best coefficients: 0.755 0.245
Jackknife mean: 0.754177442 0.245822558
std. errors: 0.052 0.052

John Smith said...


Could you please remove these samples from the modern spreadsheet? They are recently mixed individuals.


vAsiSTha said...

Sarakinos, Greece 2,325–2,300 calBCE
Target: G37:G37
Distance: 2.1635% / 0.02163482 | R5P
60.2 GRC_Peloponnese_N
16.6 RUS_Catacomb
12.0 GRC_Logkas_MBA
7.0 ROU_C
4.2 ROU_Ploiesti_BA

Theopetra, Greece 2,335–2,140 calBCE

Target: G23:G23
Distance: 1.7968% / 0.01796842 | R5P
31.6 GRC_Minoan_Lassithi
26.4 ROU_C
22.8 Yamnaya_RUS_Caucasus
9.6 ROU_Targsoru_Vechi_BA
9.6 RUS_Catacomb

Catacomb/Yamnaya like ancestry in mainland Greece from 2300bce. How does that relate to IE and its first appearance there?

Rob said...

The presence of steppe ancestry in southern Greece by 2350 bc is also interesting. In proximate terms, she’s ~ 50% Balkan Yamnaya derived

alex said...

"she’s ~ 50% Balkan Yamnaya derived"

Both are males, one is J-M67 and the other is I-L701 (but L699-)

Matt said...

Potentially new data, published 18th Dec - -

Genetic structure and demographic shifts in Upper Paleolithic to Neolithic European hunter-gatherers"

"Eurasia has been populated by modern humans for over 45,000 years and for the vast majority of their time they relied on a foraging lifestyle. Our knowledge of the genetic relatedness and structure of European hunter-gatherers is, however, still limited, because dedicated studies are challenged by the scarce availability and poor molecular preservation of human remains from that period. Here we analyze 356 ancient hunter-gatherer genomes including new genomic data of 116 individuals from 14 countries in western and central Eurasia, dating between 35,000 and 5,000 years ago (35-5 ka). We identify a new genetic ancestry profile in individuals associated with the Upper Paleolithic Gravettian culture from western Europe, which is distinct from contemporaneous groups related to this archeological culture in central and southern Europe. However, it resembles that of preceding individuals associated with the Aurignacian culture and survives during the Last Glacial Maximum (LGM, ~25-19 ka) in human populations from southwestern Europe, associated with the Solutrean and later Magdalenian culture, which re-expanded northward and northeastward after the LGM. Conversely, we reveal a genetic turnover in southern Europe suggesting a local replacement of human groups during the LGM, alongside a north-to-south expansion of Epigravettian-associated populations. From at least 14 ka, a lineage related to this culture expands from the south across the rest of Europe largely replacing the Magdalenian-associated gene pool. After a period of limited admixture across the beginning of the Mesolithic, we find genetic interactions between western and eastern European hunter-gatherers, also characterized by marked differences in inherited phenotypic traits, before they are progressively absorbed into the gene pool of Neolithic farmers."


Rob said...

Yes we knew about I2a, don't know why I rememebred Sarakinos as Female. But J2a in the assimilated south makes sense

Rob said...

Another curiosity:
With a simple model focussing on central Anatolia from Buyukkaya (~ 5500 BC) to Kalehoyuk (2000 BC), we see the following


best coefficients: 0.130 0.870
Jackknife mean: 0.130159198 0.869840802
std. errors: 0.022 0.022
TP 0.31


best coefficients: 0.257 0.743
Jackknife mean: 0.257250224 0.742749776
std. errors: 0.017 0.017
TP 0.63

One can hardly speak of a 'Bronze Age CHG migration', when the change in CHG proportions are 12 % over 3000 years. This is more consistent with ongoing social contacts between central Anatolia & the Caucasus, with a pattern of 'loco-regional mobility shifts'.

In fact, as I've mentioned before, what distinguishes Armenian Kura-Araxian individuals from from their predecessors is increased Anatolia-related ancestry, consistent with the central Anatolian origin of the 'Red-Black Burnished' pattern in their ceramics.

There are many more distinctions between reality and the pseudo-narrative created by 'the experts'....

[Right pops:

right pops:

Rob said...

@ Matt
Yes these guys got it all correct. The sequence and chronology and geography.

Matt said...

New data in He Yu files above are approximately this:

Probably missing 4 or so that they have resampled from other studies but I think it's up to most of them.

They've chosen to indicate new populations/sites by a 3-letter code, so not clear what most of these are, although there seem like a lot more Karelia_HG samples in here, which would be useful for qpAdm.

vAsiSTha said...

Surprising that this dropped before the allentoft samples.

Rob said...

DOG = doggerland ?
Got some English Creswellian there
MUR could be something from Iberia (?solutrean period)

Davidski said...


Apart from the comments here, I'm also getting some emails asking me to make changes/updates to the G25 datasheets.

I'll get onto that after Christmas

But I might also need some help in picking the right samples from the new datasets that have come out, and removing the redundant ones from the G25 datasheets.

Merry Christmas!

Matt said...

Merry Xmas to you Dave.

Some other data? -

"Haploid genotype data on 1240K panel in eigenstrat format for study 'Ancient DNA reveals admixture history and endogamy in the prehistoric Aegean'. An annotation file is included as well." - Skourtanioti

Also He Yu group updated their dataset -

I can't check on a computer, but maybe their IND file is more informative about populations now.

Gaska said...


Wow, very interesting samples, I think we have R1 in Europe, although it seems to have strange affinities, we will have a lot to talk about.

Aram said...

VERT117 from Verteba cave paper is marked as I2c. Chances are quite high that it will be I2c2 (I2a2b). A haplotype found in Trialeti Vanadzor culture.

Another interesting thing from Verteba cave is the complete absence of G2-U1. So the idea of G2-U1 moving from Europe to Western Caucasus is not supported by the current evidence from Neolithic Europe. In most likelihood U1 moved from Anatolia to Pontic region and then to Western Caucasus with early farmers.

Copper Axe said...

TTK001 is a very interesting sample from that latest batch. We'll see what it is eventually but it coming from the Tutkaul site would make sense, although I'm only aware of human remains dating to the neolithic there.

Rob said...

“ New insights into the Epipaleolithic of western Central Asia: The Tutkaulian complex”

But even “Neolithic” era sample would be great

Kevin Brook said...

I learned too late that the ancient G2a-FGC1093 sample lib7al_PE was part of the data set of a published study from two years ago which was first placed online on 18 April 2020: "Mitochondrial and Y-chromosome diversity of the prehistoric Koban culture of the North Caucasus." by Eugenia Boulygina, Svetlana Tsygankova, et al. in Journal of Archaeological Science: Reports volume 31 (June 2020): article number 102357, doi:10.1016/j.jasrep.2020.102357

Matt said...

New Ringbauer group linked kinship study -


The Early Bronze Age in Europe is characterized by social and genetic transformations, starting in the early 3rd millennium BCE. New settlement and funerary structures, artifacts and techniques indicate times of change with increasing economic asymmetries and political hierarchization. Technological advances in metallurgy also played an important role, facilitating trade and exchange networks, which became tangible in higher levels of mobility and connectedness. Archeogenetic studies have revealed a substantial transformation of the genetic ancestry around this time, ultimately linked to the expansion of steppe- and forest steppe pastoralists from Eastern Europe. Evidence for emerging infectious diseases such as Yersinia pestis adds further complexity to these tumultuous and transformative times. The El Argar complex in southern Iberia marks the genetic turnover in southwestern Europe ~ 2200 BCE that accompanies profound changes in the socio-economic structure of the region. To answer the question of who was buried in the emblematic double burials of the El Argar site La Almoloya, we integrated results from biological relatedness analyses and archaeological funerary contexts and refined radiocarbon-based chronologies from 68 individuals.

We find that the El Argar society was virilocally and patrilineally organized and practiced reciprocal female exogamy, supported by pedigrees that extend up to five generations along the paternal line. Synchronously dated adult males and females from double tombs were found to be unrelated mating partners, whereby the incoming females reflect socio-political alliances among El Argar groups. In three cases these unions had common offspring, while paternal half-siblings also indicate serial monogamy or polygyny."

Matt said...

Also from an "Only in late 20th century Western academia" moment:

"Synchronously dated adult males and females from double tombs were found to be unrelated mating partners, whereby the incoming females reflect socio-political alliances among El Argar groups."

"The hegemonic model (for double tombs) since the late nineteenth century: specifically, that these graves were thought to reflect heterosexual, monogamous couples (‘marriages’) as the basis for nuclear families"

"(I)n the late 1980s (, a) statistical analysis of paired 14C dates from a sample of 23 double burials suggested a cross-generational gap for most pairs of individuals, which led to the alternative hypothesis that the relationships between adults in double tombs were genealogical rather than socio-political."

"As a result, descent/consanguinity, matrilocality and matrilineality were proposed as the main principles of Argaric kinship practices"

Rob said...

Polygynous males all L51, etc as expected . But that could be because that’s all that was on offer at the time
At a population level, not even a minor shift during the “orientalizing” period (where Aegean inspired forts and pithos burials appear), although in Olalde there was one E1b outlier (the non African one)

What I’m interested in is the crucial 2500-2000 timeframe when there was the great shift, as in elsewhere in W Europe

Gaska said...

Rob, in the orientalizing phase (new forms of burials, pottery etc...) appear some mtDNA markers strange to Iberia, for example a case of R0a that has also been documented in Sicily (aprox 2.000 BC), ergo exogamy was practiced habitually but not only with the Mediterranean, also with central Europe, the British Isles and the rest of Iberian cultures of the Bronze Age. How those markers arrived to Sicily or mainland italy and then to Iberia a millennium before the Phoenicians is a mystery but they show commercial relations much older than we think.

Regarding the period 2.500-2.200 BC archaeology relates the oldest levels of El Argar with the Bell Beaker culture (pottery, halberds, etc etc), we are talking about a small group of Proto-Iberians DF27 that settled in the ancient domains of Los Millares culture and dominated that territory in the following 2.000 years. What is absolutely surprising is that 98% of the men are P312>Z195, and even more surprising is that in the rest of the Bronze Age cultures (Cogotas, Bronce Valenciano, Motillas, Bronce Atlántico) all the men are also DF27. I think it is not unreasonable to think of a practically absolute depopulation during the Chalcolithic and the emergence of a new lineage with great reproductive success.

On the other hand, the data of the new paper on Greece are also surprising (it is a totally different situation from the Iberian one, i.e. there is not a single case of P312 or Z2103 neither in the proto-Greeks nor in the Minoan or Mycenaean culture). There is E1b-L618 and even a basal J1-P58 in Crete ?????. The genetic difference between the Balkan and Iberian peninsulas despite being Mediterranean regions are abysmal.

vAsiSTha said...

Are the Uzbek_IA samples (Rabat, Sehrkerat) from the Vikas kumar paper available in plink format?