search this blog

Thursday, March 3, 2016

Irano-Turko-Slavic roots of Ashkenazi Jews?


As far as I've been able to discern, Ashkenazi Jews are very similar to Sephardic Jews, except with minor admixture from Central and Eastern Europe, and perhaps Central Asia (via the Silk Road). So the hypothesis presented in this new paper at Genome Biology and Evolution doesn't work for me:

The Yiddish language is over one thousand years old and incorporates German, Slavic, and Hebrew elements. The prevalent view claims Yiddish has a German origin, whereas the opposing view posits a Slavic origin with strong Iranian and weak Turkic substrata. One of the major difficulties in deciding between these hypotheses is the unknown geographical origin of Yiddish speaking Ashkenazic Jews (AJs). An analysis of 393 Ashkenazic, Iranian, and mountain Jews and over 600 non-Jewish genomes demonstrated that Greeks, Romans, Iranians, and Turks exhibit the highest genetic similarity with AJs. The Geographic Population Structure (GPS) analysis localized most AJs along major primeval trade routes in northeastern Turkey adjacent to primeval villages with names that may be derived from "Ashkenaz." Iranian and mountain Jews were localized along trade routes on the Turkey's eastern border. Loss of maternal haplogroups was evident in non-Yiddish speaking AJs. Our results suggest that AJs originated from a Slavo-Iranian confederation, which the Jews call "Ashkenazic" (i.e., "Scythian"), though these Jews probably spoke Persian and/or Ossete. This is compatible with linguistic evidence suggesting that Yiddish is a Slavic language created by Irano-Turko-Slavic Jewish merchants along the Silk Roads as a cryptic trade language, spoken only by its originators to gain an advantage in trade. Later, in the 9th century, Yiddish underwent relexification by adopting a new vocabulary that consists of a minority of German and Hebrew and a majority of newly coined Germanoid and Hebroid elements that replaced most of the original Eastern Slavic and Sorbian vocabularies, while keeping the original grammars intact.

Das et al., Localizing Ashkenazic Jews to primeval villages in the ancient Iranian lands of Ashkenaz, Genome Biol Evol (2016), doi: 10.1093/gbe/evw046

See also...

Khazar shmazar

Khazar shmazar #2

35 comments:

Fanty said...

well, dont know. A language can belong to a language group, even if a minority of words come from this group.

I just recall German.

German is rated as a Western Germanic language even through that only 30% of the words in German are of Germanic origin while 60% of the words in modern German language are ones that are borrowed from Romance languages (Latin, Italian and French basicly) during the last 2000 years.

And then there is 10% "other".

Simon_W said...

There an example of Yiddish:

http://www3.germanistik.uni-halle.de/prinz/sprachen/085.htm

I don't speak any Slavic or Iranian language, my native language is German. Yet I can easily understand the gist of this text, and I see lots of familiar words, to me the familiarity is comparable to Dutch. Therefore, even if there is supposed to be a Slavic and Iranian grammatical structure underneath the thick layer of Germanoid words, I can't see this language as anything other than Germanic, or at most a Creole language.

Simon_W said...

I mean, just compare for yourself:

http://www3.germanistik.uni-halle.de/prinz/sprachen/043.htm

http://www3.germanistik.uni-halle.de/prinz/sprachen/039.htm

http://www3.germanistik.uni-halle.de/prinz/sprachen/099.htm

Simon_W said...

Perhaps, if the Ashkenazim really had spoken a Slavic-Iranian mix prior to adopting the Germanoid influence, it would be more apt to call this a substrate language, which left some traces of substrate influence in Yiddish. But historically the reverse would make a lot of sense, as they were first in western Germany along the Rhine and later moved to east-central and eastern Europe where they conceivably may have picked up some Slavic influence.

Labayu said...

I guess we shouldn’t be surprised Eran Elhaik was involved in this – the guy who thinks similarity between Ashkenazi Jews and Armenians proves the Khazar hypothesis and that similarity between Ashkenazi Jews and Druze can only mean that the Druze are Turkic in origin. Paul Wexler is also kind of a nutjob with his own awkward convoluted hypothesis. I believe the “opposing view” mentioned is just his.

By the way, the Khazars probably never converted to Judaism: http://muse.jhu.edu/journals/jewish_social_studies/v019/19.3.stampfer.html

Fanty said...

Hmmmmm.
I listened to the soundfile of that text and it just recalled me, to something I stumbled over before. I can understand most of what she says (German here), the spoken even better then the written (latin letters version). But what this article here actually reminds me of, is... that I often missinterpret an ACCENT, in a movie with German dubbed voice, that is meant to be a Jewish accent for beeing the accent of someone Slavic of a sort.

Maybe I am just not very good in recognizing accents. I also missinterpreted the accent of an US-American, talking German, as Dutch. :-D

PF said...

The most idiotic population genetics paper I've ever read. Embarrassing!

PF said...

@Simon_W

Right, Dutch sounds fairly close to Yiddish. Also, when I go to the Amish markets in Pennsylvania, whatever language they speak sounds like Yiddish too (with an American accent, haha). It's obviously just another West Germanic language that was developed generally around the Rhineland a millennium ago, as expected.

Grey said...

One of my pet theories is J was originally rooted in the hills/mountains along the edge of the steppe and drifted or got pushed south later but so far anyway it doesn't sound like this paper helps that.

Fanty said...

"Dutch sounds fairly close to Yiddish."
"generally around the Rhineland"

Hmmmmm

Thats kind of strange however.

The German dialects are usualy explained as 3 branches: Nether-German (Nothern Germany), Middle-German and Upper-German (southern Germany + Switzerland + Austria).

The Rhineland speaks Middle-German dialects.

this map here:
https://upload.wikimedia.org/wikipedia/commons/0/00/Deutsche_Mundarten.png

claims, that the largest part of the Netherland is speaking a language that is connected to Nether-German, only a tiny fraction in southern Netherland is rated as connected to Middle German dialects.

Why would the Jews from the Rhineland develop a language that sounds like something more northern?

truth said...

There is very little to none, of slavic in Ashkenazies.

PF said...

@Fanty

Definitely there are Germanic dialects that are linguistically closer related to Yiddish than Dutch -- specifically the Central German ones, as you say. (Apparently the closest German dialect to Yiddish is Badisch, which makes sense.)

I was just agreeing that Dutch generally sounds like Yiddish. I guess what adds to that effect is the very guttural Dutch G (don't know the correct term). Luxembourgish, for example, should sound more like Yiddish, and perhaps it does, but to my ears it sounds softer, almost more "French." Of course we're talking about language so a ton has changed in a 1,000 years.

Btw, the English wikipedia version of the map you linked to is annotated and a little more clear: https://en.wikipedia.org/wiki/High_German_languages#/media/File:Continental_West_Germanic_languages.png

Fanty said...

Hm. didnt know yet, that the dialect of the most northern part of the Rhineland is rated as a "Dutch" dialect. cO

Hm. Where I grew up, 6 is (was) spoken.
I virtually only understand some parts of it, because I learned English because the words are closer to their English (and most likely Dutch) variants than to the official High German version.

But its dying. I know, if old (about 70+) people meet, they talk to each others in this dialect. People in the age of 60 can understand it and possibly speak a few sentences (and that badly with lots of errors) and people in the age of 50 have never learned it.

So these maps actually show something that does not exist anymore. Except for some rare regions in wich people take pride in their dialects like... Bavarians or people in Cologne (horrible dialect. so ugly, but they are so proud of it)

Robert Hartmann said...

There is a political conflict since long. I assume we should concentrate on empirical findings.

Unknown said...

I made this comment on ABF, Davidski's already seen it, I'll post it here:

"That paper is unmitigated bullshit.

1) Yiddish is not 'minorly German,' it is so German as to be close to mutually intelligible according to many speakers of both languages. The 'linguistic hypothesis' they're defending is exceptionally minor.

2) Argument via false etymology: Ashkenaz may have once referred to someplace in Anatolia. In the dark ages, it referred to Germany.

3) What genomes did they use? Because saying 'they're close to Italians, Greeks, Turks and Iranians' in a paper trying to prove kinship with Iranians suggests that they're more closely related to Italians and Greeks than Iranians, which supports preexisting genetic hypotheses. Certainly anyone with an eye to IBD sharing, YDNAs, or just autosomal runs can dismiss this paper out of hand.

4) It suggests the AJs picked up non-'Ashkenazi' MTDNAs when they eventually moved to Europe but their MTDNAs hint strongly at an Italian origin around 1800-2000 years ago, if you look at nesting and TMRCAs, which flies against this idea. Isn't it simpler to suggest, as has already been suggested, that Mountain Jews, already believed to be an offshoot of Persian Jews, share YDNA with other Jews because of common Judaean patrilineage?

5) No Sephardics mentioned, hm. Sephardics are basically the massive stone in the shoe of Khazar theorists and I guess these idiots too, if objective genetic, historical and linguistic evidence weren't enough. Basically: AJs and Sephardics are genetically incredibly similar. Yet no one would suggest Sephardics are from the Steppe or wherever. What gives, theorists?

In summation it can be stored neatly with Elhaik's work, and then set on fire."

Gioiello said...

Of course I don't say anything about the linguistic question, and German speaking people have said enough. Perhaps the thesis of Elhaik could be reversed: origin in the Rhine Valley of people coming from South and later many introgressions from Black Sea (Khazars and others), but Elhaik is adversed because he, like me, says that Jews are above all introgressed and perhaps a few from Old Israel (my links of Jewish even J1 with European origin are many). I make links of the Y and mt and use a little the autosome, subdue above all to interpretations. About these samples I'd want to say only one thing. There is a person mt H41a. This haplogroup was found from Irene Pichler in many Hutterites and she (from Alto Adige, thus she believes to be of German origin) searched for the closest haplotypes in Central Europe, but H41a is the haplogroup of my paternal grandmother and was of my father's and my cousin's and her daughter's… I wrote to Irene Pichler to find in Tuscany, but he didn't reply to me.

Arch Hades said...

The genetics pretty much shows that Ashkenazi Jews are just a group of Sephardi Jews that headed North and picked up ~10% Central-East European admixture. What do you get when you mix a 90% Sephardi Jew with 10% Central-East European, an Ashkenazi. So Sephardis will have nearly all the same components Ashkenazi Jews do. The authors of these papers should always included European Sephardis when they are coming up with their hypothesis on Ashkenazi origins. It's very lazy and stupid not to.

Gioiello said...

About that 10% introgressed we should discuss a lot. Costa et al. said (so long after me) that the mt is for more than 90% introgressed. I study above all the hg. R1b (that is mine) and I've written a lot about R-V88 (the origin in Sardinia/Italy is more and more likely). About R-M269* there is a Jewish line but with a MRCA perhaps not more than 2000 years ago (we'll see next if the first 19 SNPs in common are rooted in Middle East or elsewhere: R-M269* is above all Italian so far: see the tree of Sergey Malyshev). Also of R-L277 there is only a line only Jewish so far, and the most part are Italian so far, even though L277 is an “Eastern haplogroup” and not pretty Western European as others, not only after R-L51 but already downstream CTS7556. We are waiting for other tests and above all for other aDNA...

Nirjhar007 said...

Gioiello,

What do you understand from this?.
http://barradeferro.blogs.sapo.pt/of-shulaveri-shomu-and-r1bs-and-34071?thread=47383

Gioiello said...

Nirjhar, I thank for your link, but for answering it I should publish all my 10000 letters written so far. I know only one Brazilian which believes that his J1* came from Southern Caucasus, he is Ricardo Costa de Oliveira, but he did a Full Genome and his data are deeper. To compare U4 or U5 (which are many tens of thousands of years old) and to say that R-M343* was born in the Caucasus is very ingenue. I demonstrated tha your India has R-L389- and Italy R-L389+ and all the subclades and they separated pretty 20000 years ago. That the R-L23-Z2105* found in Samara have no meaning if your don't find there both the upstream subclades and above all the downstream L51 I said from so long… and that many old subclades are from many thousands of years both in the Caucasus in the Alpine region I am sayng from so long too. Perhaps a recent paper with the oldest LGM refugia may you understand what really happened. My “Italian Refugium” may have been the around the Mediterranean refugium. South Asia had another, completely different… and very likely R1a* (amongst other haplogroups) was here and not there.

Nirjhar007 said...

Thank you Gioiello , your remarks are important and professional!.

I have only question for you now, as you already know, the Varna Cultures genome are coming very soon ( probably this upcoming week?) , Do you think there will be R1b-M269?.

Gioiello said...

What to say with these presuppositions? “The discontinuity of the Varna, Karanovo, Vinča and Lengyel cultures in their main territories and the large scale population shifts to the north and northwest are indirect evidence of a catastrophe of such proportions that cannot be explained by possible climatic change, desertification, or epidemics. Direct evidence of the incursion of horse-riding warriors is found, not only in single burials of males under barrows, but in the emergence of a whole complex of Indo-European cultural traits. The original term for: 'Castes' in India was: 'Varna', meaning: 'Color', cognate to French: 'Vernis', and Spanish: 'Barniz' (both= Varnish)”.
I thank you for the news, I didn't know. Thus we'll wait for the results.
The possible link between IE *werno/ehₐ- with alder (1) alder (2) and Skt varna varaṇa- could shed some light also on the origin of the IE languages.

Nirjhar007 said...

Caro amigo,

The Word varnish has nothing to do with the Sanskrit term, that suggestion is most likely made by some folk etymologist;) .
http://www.etymonline.com/index.php?term=varnish&allowed_in_frame=0
https://en.wiktionary.org/wiki/varnish#Etymology

But if indeed Varna culture had strong Kurgan Influence then, its not bad to assume we will see R1b !

Gioiello said...

Nirjhar, caro amico (amigo is Spanish from Latin amicu(m)), the link between Old French vernis and Skt varna is done from Wikipedia not by me. That Medieval Latin derives from “Late Greek verenike” is only an hypothesis, possible but undemonstrated. I did mean other, i.e. that could be a link between the Skt varna and varaṇa if both from IE *werno/ehₐ- (the link is done from Mallory&Adams, The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World, at list for varaṇa) if the same would be happened in the meaning of English “alder” in the two meanings. When a folk etymology was done from old peoples it becomes interesting. But we are not discussing here of linguistics but of genetics, and, just for what you said, also those results have to be interpreted, because a possible R1b will be able to be interpreted in opposite ways, either come with the invasors or due the invaded.

Nirjhar007 said...

Ok amico (it was a typo) :) , yes , varna and varaṇa can come from *werno/ehₐ- but for Varnish OED suggests Latin vernix "odorous resin," of uncertain origin, perhaps from Late Greek verenike, from Greek Berenike, name of an ancient city in Libya (modern Bengasi) credited with the first use of varnishes..

I think its quite possible.

Yes, if R1b from Varna is found it will be vital to see how related they will be with Yamnaya types, but it will be huge as the culture is older!.

Nirjhar007 said...

also Oxford is in parallel position with OED.
http://www.oxforddictionaries.com/definition/english/varnish

Ryan said...

"Irano-Turko-Slavic roots of Ashkenazi Jews?"

Shouldn't that be the Irano-Turko-Slavic roots of Yiddish?

Simon_W said...

@ Fanty

I think you may be overestimating the importance of the sound of certain dialects. Different variants of Low German, or High German for that matter, can sound quite different. Bavarian sounds very different from Swabian, yet they're both HIgh German. There is even quite some difference between Dutch and Flemish, although they're essentially the same language:
http://www3.germanistik.uni-halle.de/prinz/sprachen/130.htm

http://www3.germanistik.uni-halle.de/prinz/sprachen/068.htm

@PF

Baden makes sense. In fact the very name "Yiddish" looks like derived from the Baden dialect word for "Jewish", standard German "Jüdisch". The change from ü to i is typical in some parts of Baden and the Alsace, and also used to be typical for the Swiss city of Basel.

Joshua Lipson said...

If anyone wants a quantitative sense of how marginal any Turkic influence on Ashkenazim is, check out this spreadsheet of East Asian component values (MDLP and Dodecad; I'm sure Eurogenes would second them) among Jewish, Mediterranean, Caucasian, and Eastern European populations.

https://docs.google.com/spreadsheets/d/17rLOHTcQpws2Yit_Ka_6HdUfkeVAl-wIPHtr1ul6FSU/edit#gid=0

Yes, Ashkenazim are the most East Asian population in the Central/East Mediterranean. But by not even a hair.

Joshua Lipson said...

(that is, with the exception of Turkish-speaking ones)

Onur Dincer said...

@Joshua Lipson

The Ancestral Altaic component of MDLP K23b is not an East Eurasian component, but a West Eurasian one. It peaks in the MA1 individual from the Paleolithic Altai and was named after him. As you know, MA1 is West Eurasian by genetics. So no wonder Khanty have 47% Ancestral Altaic and 47% East Siberian in MDLP K23b since Khanty are almost an even mix of West and East Eurasians:

http://3.bp.blogspot.com/_Ish7688voT0/TC5B1k8sB6I/AAAAAAAACf4/RfaPQgIsLPw/s1600/gr1.jpg

BTW, you confused the MDLP K23b results of Ossetians and North Ossetians in your table.

Joshua Lipson said...

Onur: Good catch re: the 2 Ossetian samples. Corrected that. As for ancestral Altaic: I'm making no claims about its Paleolithic affinities, but I think it's potentially relevant as an indicator of elevated Siberian/Central Asian ancestry (which it turns out not to be in the Ashkenazi case).

Pavel Flegontov said...

Appearance of this paper just shows that the review system is deeply flawed in biology-focused journals, in case of interdisciplinary papers. Any decent linguist or historian would have dismissed this bullshit as pseudo-science upon reading the abstract. 'Yiddish is a Slavic language' - that's enough...

Unknown said...

My DNA is exactly what he says & I didn't know I was Ashkenazi Jewish, Iranian, or Japanese & I am still 98% European

H6 group

Unknown said...

My DNA says his theory isn't horrible. It's H6 from my Mom's side. #4 is most likely.