September 13, 2012

Polish and German Y-chromosomes

I have often bemoaned the use of present-day populations as stand-ins for dealing with the subject of very old archaeological phenomena such as the Neolithic transition. Of course, I understand that until a few years ago, this was all we had to work with. But, this idea is now suspect, having been made so by a two-pronged attack. On the ancient DNA side, researchers have consistently discovered that for the better part of prehistory, ancient populations did not match modern ones: even if the constituent elements of later evolution could be identified, they were still in polarized non-admixed form as in the case of the Neolithic Swedes. On the recent side, researchers have used surnames or even toponyms to show that ethnic admixture in the recent historical past has shifted Y-chromosome frequencies around.

A new paper in EJHG follows on this tradition by comparing pre- and post-WWII patterns of Y chromosome variation in Germany and Poland. Y-haplogroup frequencies can be seen on top left. From the caption: "Phylogenetic relationship and frequencies of Y-chromosomal haplogroups in the studied populations. Ka Kaszuby; Ko Kociewie; Ku Kurpie; Lu Lusatia; Sl Slovakia; Me Mecklenburg; Ba Bavaria."

It is important to note that the researchers were able to study pre-war populations, because most everybody knows where their patrilineal ancestor lived less than 100 years ago. But, European history consists of many events whose effects on the current population are less known, because they occurred at an older time. In some cases, populations may have migrated (such as the Germans of eastern Europe following WWII), in others populations that have once existed there have almost disappeared, or become much less numerically significant (such as the Ashkenazi Jews due to persecution during WWII and after it through migration to Israel and elsewhere, or various Christian and Jewish communities that once flourished throughout the Middle East). Other, less known groups, such as the Old Prussians or the Jassic speakers of Hungary have been presumably absorbed by surrounding majorities, or through a process of elite dominance.

In many cases, the available information, in the form of linguistic, genealogical, or historical evidence, can be used to remove layers of admixture, migration, and extinction in history; but, the gap between the deep prehistoric past and the recent, historical one cannot be bridged by these methods alone. Ultimately, ancient DNA researchers must close in on the present by targeting more recent populations for analysis. As the realization of genetic change continues to amass on both sides of the divide, I suspect that this will come naturally, although I do expect some reticence to the findings as they begin to touch upon the most cherished origins traditions of the multitude of extant European nations.


European Journal of Human Genetics advance online publication 12 September 2012; doi: 10.1038/ejhg.2012.190

Contemporary paternal genetic landscape of Polish and German populations: from early medieval Slavic expansion to post-World War II resettlements

Krzysztof Rebala et al.

Abstract

Homogeneous Proto-Slavic genetic substrate and/or extensive mixing after World War II were suggested to explain homogeneity of contemporary Polish paternal lineages. Alternatively, Polish local populations might have displayed pre-war genetic heterogeneity owing to genetic drift and/or gene flow with neighbouring populations. Although sharp genetic discontinuity along the political border between Poland and Germany indisputably results from war-mediated resettlements and homogenisation, it remained unknown whether Y-chromosomal diversity in ethnically/linguistically defined populations was clinal or discontinuous before the war. In order to answer these questions and elucidate early Slavic migrations, 1156 individuals from several Slavic and German populations were analysed, including Polish pre-war regional populations and an autochthonous Slavic population from Germany. Y chromosomes were assigned to 39 haplogroups and genotyped for 19 STRs. Genetic distances revealed similar degree of differentiation of Slavic-speaking pre-war populations from German populations irrespective of duration and intensity of contacts with German speakers. Admixture estimates showed minor Slavic paternal ancestry (~20%) in modern eastern Germans and hardly detectable German paternal ancestry in Slavs neighbouring German populations for centuries. BATWING analysis of isolated Slavic populations revealed that their divergence was preceded by rapid demographic growth, undermining theory that Slavic expansion was primarily linguistic rather than population spread. Polish pre-war regional populations showed within-group heterogeneity and lower STR variation within R-M17 subclades compared with modern populations, which might have been homogenised by war resettlements. Our results suggest that genetic studies on early human history in the Vistula and Oder basins should rely on reconstructed pre-war rather than modern populations.

Link

17 comments:

Slumbery said...

"Other, less known groups, such as the Old Prussians or the Jassic speakers of Hungary have been presumably absorbed by surrounding majorities, or through a process of elite dominance."

We did not have a Hungarian elite placed upon us, the Jassic area had a special legal status until the 19. century. For example there were no feudal lords. The language loss, after earlier slow assimilation, finished during and directly after the Turkish wars that caused huge migrations in the area, partly breaking down the old communities and partly introducing non-Jassic migrants.

This is a good example of what you talking about in general actually. Because of the Turkish invasion, most of the population of the area of the current Hungary was replaced. I am quite sure that the population of (current) Central and Eastern Hungary around 1800 has less than 50% ancestry from the population of the same area around 1500.

Most of the Jassic settlements survived though, but not with their language. Not a surprise, we talk about an 1000 km2 area with a dozen villages that were not isolated geographically at all (no natural borders so far).

Sorry for the low relevance, I tend to be sentimental about my home village. :)

J said...

Since I was born in Jaszbereny, I can say that the Jasz and Kun (Petcheneg) populations lost their language but since they occupied small villages in inaccessible places, they maintained their ethnic identity. The Jaszsag, for example, is built on low sand hills in the middle of an enormous swamp (which now has been drained). The last Jassic speaking Hungarian died a hundred years ago.

formerjerseyboy said...

"I am quite sure that the population of (current) Central and Eastern Hungary around 1800 has less than 50% ancestry from the population of the same area around 1500."
Could you mention a source for futher information on this? It sounds quite similar to the demographic shifts across the Iberian peninsula, during the Spanish "Reconquista."

andrew said...

Slumbery's on the ground familiarity with the situation is surely more credible than a tertiary source, but it is worth noting that the Wikipedia article on the history of Hungary generally confirms his conclusion about demographic shift in Hungary, albeit with less specificity, as does the Wikipedia article on the Hungarian People, with considerably more specificity, citing:

* Hungary. (2009). In Encyclopædia Britannica. Retrieved May 11, 2009, from Encyclopædia Britannica Online
* A Country Study: Hungary. Federal Research Division, Library of Congress. Retrieved 2009-03-06.
* "International Boundary Study – No. 47 – April 15, 1965 – Hungary – Romania (Rumania) Boundary". US Bureau of Intelligence and Research.
* Historical World Atlas. With the commendation of the Royal Geographical Society. Carthographia, Budapest, Hungary, 2005. ISBN 963-352-002-9CM
* Steven W. Sowards. "Twenty-Five Lectures on Modern Balkan History (The Balkans in the Age of Nationalism), Lecture 4: Hungary and the limits of Habsburg authority". Michigan State University Libraries. Retrieved 2009-05-11.
* A. MaCartney D. Litt. (1962). Hungary A Short History. Edinburgh University Press.
* Robert A. Kann (1980). A history of the Habsburg Empire, 1526-1918. University of California Press. p. 205. ISBN 0-520-04206-9, 9780520042063.

I have also read the conclusion that he makes in a couple of more scholarly accounts setting forth the demographic shift in Hungary in that time period that I have read in the last few years while I was researching the disconnect between Hungary's Uralic linguistic affinities and non-Uralic population genetic profile while writing blog posts and comments that touched on that point.

But, I honestly cannot recall precisely which couple of articles pointed this out and I have been sloppy about keeping good indexes of these articles.

I think that one of them may have been in the discussion found in one of the fairly recent journal article that assigned haplogroups to some samples of Medieval Hungarian DNA, but I'm not entirely sure of that.

Onur Dincer said...

What are the main sources of information on the demographics and ethnic composition of Hungary for the centuries preceding the 19th century (according to Wikipedia, the first accurate measurements of the population of Hungary including ethnic composition were carried out in 1850–51)? Church registers? How far back in time do they go?

Slumbery said...

formerjerseyboy

I do not have linkable sources. My knowledge of the topic has two main sources:
- written Hungarian books, included my school textbooks.
- the knowledge about the origin of family names in the wider area (so not only the Jassic area).

Generally it is a common knowledge in Hungary (for those who are interested).

Just a tiny example of the first: there is a before-after map in one of my history atlases that show the settlements of the closer central Hungary before the Turkish war and after it. The "after" part is colour coded based on the ethnicity of each settlement. There is no comment in the atlas, but this pair of maps show that about 80% of the settlements were downright obliterated (especially the smaller ones) and in the 18th century Hungarians were clearly a minority in the area where the modern Budapest is. In this particular area most of the new settlers were Germans but Slovakians and Serbians are also significant. Southwest from this area Croatian elements start to dominate, mixed with Germans.

About the second: there are plenty of people even in the Jassic area that have family names with German, Slovakian or Croatian origin or family names that are linguistically Hungarian but refer to those origins. Also there is an other group that has family names connecting them to Geographical places outside of the current Hungary. They are descendants of the Hungarians from Transylvania and the Highland (Slovakia). (The use of family names among commoners started to be widespread in than time, so many people got family names referring to where they came.)

I work for a small firm in Budapest. 12 people. Everybody is Hungarian. 3 people have downright German family name, 1 told me that the family changed its German name after the II.WW., one has a family name referring to a city in current west Turkey (either Turkish or Greek family origin), another one with a family name that refer to French origin. Two of them have family names that refer to Highland settlements, in an area that was mostly outside the Turkish control/reach. One of them has a family name that clearly shows Transylvanian origin (not a Geographical name, but virtually everybody with that name comes from Transylvania). That makes 25% of male lineages that may existed in the region before the Turkish war or may not. (Well, maybe 33%, the French family could be a post Mongolian invasion settler, many French came in that time, during the same re-population when the Jassic and the Cumans came.)


J: you overemphasize that swamp-isolation thing. There were swamps in the South and East(just like all along the river Tisza in the Carpatian basin, it was like the Nile before the regulation), but all the settlements were accessible, and actually only one Jassic village exists that does not have direct Hungarian neighbour.

Onur Dincer said...

- written Hungarian books, included my school textbooks.

Just a tiny example of the first: there is a before-after map in one of my history atlases that show the settlements of the closer central Hungary before the Turkish war and after it. The "after" part is colour coded based on the ethnicity of each settlement. There is no comment in the atlas, but this pair of maps show that about 80% of the settlements were downright obliterated (especially the smaller ones) and in the 18th century Hungarians were clearly a minority in the area where the modern Budapest is. In this particular area most of the new settlers were Germans but Slovakians and Serbians are also significant. Southwest from this area Croatian elements start to dominate, mixed with Germans.


But what are the sources of information on the demographics and ethnic composition of Hungary that those books use for the centuries preceding the 19th century (according to Wikipedia, the first accurate measurements of the population of Hungary including ethnic composition were carried out in 1850–51)? Church registers? But for most parts of Hungary church registers begin only after the end of the Ottoman rule:

https://www.familysearch.org/learn/wiki/en/Hungary_Church_Records

Slumbery said...

Onur

You do need accurate measurements to notice that a whole old settlement stopped to exist or a new was founded.

During the initial re-population of the most devastated areas a significant part of the new settlements were ethnically almost pure (because the settlers were not spontaneous but organized) and in this part I talked about the ratio between the settlements themselves. You do not need accurate measurements on personal level for this. The foundation of the new settlements and the origin of the people that founded a given settlement was actually well documented and these documents are accessible in the archives.

So this is the source of that map.

Of course this is not accurate for the overall ethnic ratio, because we do not know the exact population of each settlements and they were still mixed a bit for sure. But I never said anything like "41,27% of the people were German settlers".

This settlement level ethnicity is only true for the most devastated areas (like the closer central Hungary/the area of Budapest), but even in other areas we have the names of the people from tax and estate records and the documents of local lawsuits. This also accessible in the archives.

This is also not accurate, but you do not need the accuracy of a census to say something like "more than half of the population was replaced".

Onur Dincer said...

Slumbery,

The church registers and apparently the sources you cite are almost totally from after the Ottoman rule. What sources can you cite for the Ottoman rule era and before? If Ottomans really triggered a partial but significant ethnic/demographic change in Hungary, we should first know the ethnic composition and demographics of Hungary before and during the Ottoman rule to be able to know that.

BTW, are you Jassic?

Slumbery said...

Onur

Before I answer this I must start with a disclaimer. I am not an expert of the area and not even interested enough to ever look directly into the original sources. I can only write about what I know and there will be some speculation in my answer, but I try my best.

First, this is a rather well researched topic in Hungary and the scenario of partial/mayor population replacement seems to be the consensus of the experts. They may very well have sources I do not know about, simply because did not devote enough time to the topic.

Now, we have separate the two questions we are talking about, the population replacement itself and the ethnic composition of the pre-Ottoman central Hungary.

As for the population replacement, the post war data alone are sufficient. The sources of the new population are documented both directly (settlement foundation documents)and indirectly (names and other documents) and the pre-war ethnic composition is irrelevant for the fact of population replacement itself. We do not need to know what was replaced.

Also since an immigrant population that had exactly the same ethnic composition as the earlier is rather unlikely, it is more than plausible to assume that some change in the ethnic composition happened, whatever would be the pre-war composition.


Now, the pre-war ethnic composition of the area is a totally different question. We have less data from this time, but still much more than nothing. To make a long story short: the place and other geographical names as well as the names of the people named in remained (indeed, not very numerous, but still existing) documents do not show significant/widespread non-Hungarian presence in the area from the XV. century. (Other than the well known cases of gradually assimilated XIII. century (post-Mongolian) settlers, like Jassic.)At least nothing like after the war.

I emphasize that I talk about parts of the current Hungary.

"BTW, are you Jassic?"

Yes, however this does not mean too much nowadays. We do not even have a dialect, also our appearance is pretty much within the "mainstream" Hungarian variety. It is stopped to be an actual ethnicity a long time ago.

Onur Dincer said...

Slumbery,

Actually, the Ottoman archives of Hungary, including tax registers, are still available, so we have some demographic information about Hungary during the Ottoman rule. But I doubt they contain enough information about the ethnic or linguistic composition of the Ottoman Hungary, because Ottomans were not interested in ethnicity or language of their subjects and counted households based on religion (for taxation purposes) instead and as a rule did not record names of their subjects. For the pre-Ottoman Hungary I guess we have even less data.

Justin said...


Slumbery, could you tell me about the surname Sabo/ Szabo? I am suspicious of it being a Magyar word, and when I've looked it up I've gotten a few trails other than the typical "Tailor" - one of which being the name Sabastian - and found some similarities to words of surounding countries. Might it be a borrowing or corruption?

Slumbery said...

Justin

There is no mystery, you have already found the meaning. It literally means "tailor" and a very common Hungarian family name (the fourth most common in Hungary).

There is no connection to the name Sabastian/Sebastian (Sebestyén is the Hungarian version of that name).

I can't judge similar words in the neighboring coutries, because I do not know what did you see exactly, but my etimology dictionary says the core of the word (the verb "szab") is an old word in the Hungarian, so those words are either borrowings or just accidentaly similar words with no actual connection or possibly names of families with Hungarian origin.

I do not have any idea what do you mean by "corruption" though.

By the way, Szabó [Tailor] as the fourth most frequent name is sandwiched on the frequency list by Tóth [Slovak] and Horváth [Croatian]...

Justin said...


What I meant by corruption is, say it was an original Magyar word (or Turkic word), that through association with another Hungarian ethnic group, was manipulated from the original pronounciation or developed the surname - and might also be tied to non-Magyar ancestry potentially.

Also, what is number 1, Attila? ;)

Slumbery said...

Justin

We start to be seriously off topic there, so if you have further questions, we should move the discussion to private channels. I am going to fill my profile for that...

I hope Dienekes will let this last comment go alive.
----------------------------------
I can imagine that families outside Hungary use the "Sabo" form, especially if the Hungarian identity was lost.

Attila is a "first name" (it is actually a last name in the Hungarian name order, but the English term is "first name"), while Szabó is a family name. They are not on the same list. The most common family name is Nagy [Big]. Quite simple one. :)

Also Attila is not a particularly popular name, it is not in the top ten. It was more popular 20-30 years ago, but even then never made into the top 3 and actually it used to be always expressly rare before became fashionable in the second half of the 20. century.

Onur Dincer said...

Slumbery,

I know this is a personal question, but have you or any other full Jassic you know ever tested your/his/her DNA through any DNA testing company and/or any online DNA project? I am asking this because I am curious about how a 100% Jassic would genetically show up and how similar or dissimilar to other Hungarians and other ethnic groups.

This is my last post to you on this thread and I will post no more off topics.

Unknown said...

It would be interesting to read some research on other eastern German groups as well, like on Pomeranians, Silesians, East and West Prussians... They can't be all reduced to Mecklenburg Germans. But, understandably this Polish research team had its primary focus on the Slavic/Polish groups.