R1b-L21

In origin of European haplogroups you write than R1b-L21 is Britanno-Irish.
Nevertheless we find many R1b-L21 in Germany, and others in Switzerland, in France outside Brittany and North Italy.
I think L21 is celtic. Its origin is in north of Alps and then migrates to British Islands but also in France.

Bernard

I think that you didn't read the whole section about R1b. The table with the subclade description mentions clearly (I hope) that L21 is also found in Germany and that it is Celtic.

I suppose that you are referring to the R1b chart. Even there L21 is mentioned as a branch of the Celtic S116. Britanno-Irish doesn't mean it is exclusive to Britain and Ireland but just that it is most common there (maybe due to a founder effect).

It is dangerous to read a haplogroup/subclade description thinking that it means "exclusive to ... region/country". It is never the case. We can at least indicate where it is the most common. Only a few cases of L21 have been found in Germany so far, and other kinds of R1b (S28, S21, S116 or just M269) are all more common in Germany.

Note that the main subclade of L21 (M222+, formerly known as R1b1c7) is "almost" exclusive to Britain and Ireland, with a peak in Northern Ireland and Scotland.

I tried to explain in the R1b section that R1b came from the Black Sea region, moved west along the Danube to reach southern Germany, then expanded all over Western Europe from there. Southern Germany being the starting point of this "Italo-Celto-Germanic colonisation" it is only natural that most of the R1b subclades should be found there. Only the ones that came into existence after this expansion (about 4000 years ago) are found in more specific regions (like M222).

I also want to add that haplogroups tend to leave a trail behind them. L21 is most common in the British Isles nowadays (especially the North-West, i.e. Ireland and Scotland), but it left a trail behind it showing the migration path, from the Alps and southern Germany through northern France and across to Britain.

There isn't enough Y-DNA data from Eastern Europe and the Near East yet, but if L21 is found along the Danube corridor and in Anatolia, it would mean that it already existed prior to the great Italo-Celto-Germanic migration. If not, then it must have appeared north of the Alps.

Maciamo said:

I think that you didn't read the whole section about R1b. The table with the subclade description mentions clearly (I hope) that L21 is also found in Germany and that it is Celtic.

I suppose that you are referring to the R1b chart. Even there L21 is mentioned as a branch of the Celtic S116. Britanno-Irish doesn't mean it is exclusive to Britain and Ireland but just that it is most common there (maybe due to a founder effect).

It is dangerous to read a haplogroup/subclade description thinking that it means "exclusive to ... region/country". It is never the case. We can at least indicate where it is the most common. Only a few cases of L21 have been found in Germany so far, and other kinds of R1b (S28, S21, S116 or just M269) are all more common in Germany.

Note that the main subclade of L21 (M222+, formerly known as R1b1c7) is "almost" exclusive to Britain and Ireland, with a peak in Northern Ireland and Scotland.

I tried to explain in the R1b section that R1b came from the Black Sea region, moved west along the Danube to reach southern Germany, then expanded all over Western Europe from there. Southern Germany being the starting point of this "Italo-Celto-Germanic colonisation" it is only natural that most of the R1b subclades should be found there. Only the ones that came into existence after this expansion (about 4000 years ago) are found in more specific regions (like M222).

Click to expand...

Although I like much of what I have read of your posts thus far, I have to disagree with what you have written about characterizing L21 as "Britanno-Irish". Most people, reading that L21 is "Britanno-Irish", will reasonably conclude that you mean L21 is "Britanno-Irish" everywhere it occurs. In other words, if I am Swiss, get an L21+ result, and read that L21 is "Britanno-Irish", I am going to come to the conclusion that I am somehow descended from an ancestor who came from the British Isles. I am not going to think, "That just means L21 is very common in the British Isles."

I respectfully suggest you correct the entry for L21 to read something other than "Britanno-Irish". You did not intend it to be misleading, but it is misleading.

By the way, even though testing for L21 has only been going on since October of 2008, and was not included in Family Tree DNA's Deep Clade-R test until sometime in January of 2009, we already know of at least 21 men of German and German-Swiss ancestry who are R-L21*.

Testing for U152 and U106 has been going on since 2005 and began in better economic times.

rms2 said:

Although I like much of what I have read of your posts thus far, I have to disagree with what you have written about characterizing L21 as "Britanno-Irish". Most people, reading that L21 is "Britanno-Irish", will reasonably conclude that you mean L21 is "Britanno-Irish" everywhere it occurs. In other words, if I am Swiss, get an L21+ result, and read that L21 is "Britanno-Irish", I am going to come to the conclusion that I am somehow descended from an ancestor who came from the British Isles. I am not going to think, "That just means L21 is very common in the British Isles."

I respectfully suggest you correct the entry for L21 to read something other than "Britanno-Irish". You did not intend it to be misleading, but it is misleading.

Click to expand...

Thanks for your advice. I have modified the R1b subclade table replacing the title "associated ethnicity" by "most prevalent ancient ethnic group". Most prevalent only means that it is most common in that ethnic group, not that it originated there.

I have also replaced "Britanno-Irish" in the chart by "Gallo-Britanno-Alpine". This encompass Central Europe (Alpine), France (Gallo) and the British Isles (Britanno). I will try to come up with something better once we have a clearer picture of the distribution of L21.

If everyone took the care of reading the R1b history above the table there would be no confusion. It is explained that the main subclades of R1b1b2a1b (L21, S28, M167) spread from the central European homeland (Unetice to Hallstatt cultures) to western Europe.

I even created a migration map of R1b for more visual people who don't like reading.

rms2 said:

By the way, even though testing for L21 has only been going on since October of 2008, and was not included in Family Tree DNA's Deep Clade-R test until sometime in January of 2009, we already know of at least 21 men of German and German-Swiss ancestry who are R-L21*.

Testing for U152 and U106 has been going on since 2005 and began in better economic times.

Click to expand...

Indeed. I also noticed two L21+ in Eastern Europe along the Danube. This could mean that L21 already existed before the mass migration of Proto-Italo-Celts from the Black Sea region.

It remains to be seen why L21 became to predominant in the British Isles. If it is due to a founder effect, the replacement of native haplogroups by the R1b invaders must have been absolutely massive. It would mean that a small group of L21 invaders (the smaller the group the lower the chance of other subclade being present) would have eventually replaced the majority of paternal lineages in the British Isles.

Was Bronze-Age culture really so superior or did the Indo-Europeans carry germs that wiped out most of the native population, in the same way as when Europeans reached the Americas ? Archeological records show that the Bronze-Age transition was an unusually violent time. Diseases also spread fast across continents, so I would rather go for the violent replacement hypothesis.

I also like the idea that native female population remained fairly stable (men were killed and women taken as wives by the invaders), and that this is why Western Europeans look so different from each others in Iberia, Italy, France, Germany, Britain, Ireland or Scandinavia, Local native populations were already quite diversified. It is unlikely that the phenotypes changed so much just by genetic drift over a mere 4,000 years since the Indo-European migrations.

Thank you very much! I am impressed by your open-mindedness.

Have you read the book, The History of the Celtic People, by the French archaeologist and linguist Henri Hubert? It is an older book, but still powerful. Hubert suggests that the Beaker Folk of the Middle Rhine who settled in the British Isles were Goidels who carried Q-Celtic Goidelic with them. Such an early arrival (early Bronze Age) for L21, if the Goidels/Beaker Folk were L21+, would explain a lot about the distribution of L21 in the Isles and the archaic nature of language, culture, religion, etc., in ancient Ireland.

No, I have read this particular book on Celtic history, but thank you for the recommendation.

Sorry to be a complainer again, but on your y haplogroups page, near the top, you have L21 listed as 4,000 years old and originating "in the British Isles".

I don't have a big problem with the 4,000 years part, although the estimate to the U152 and L21 MRCA is actually 3,500 years. But I do take exception to the "in the British Isles" part.

I think it VERY unlikely L21 originated in the British Isles.

Testing for L21 has only been going on since October of 2008, and L21 has only been a part of Family Tree DNA's Deep Clade-R test since sometime in January of this year. Thus we only have a very few lab batches under our belts. Yet already L21 has turned up fairly frequently on the Continent.

Recently Tim Janzen took the variance of the R-L21 Plus Project and found only 149 years difference between the British Isles group with 67 markers and the Western European group with 67 markers. Statistically, 149 years is nothing; it is not a significant difference. What is significant, however, is that Janzen had only 14 West European 67-marker haplotypes to compare and 108 British Isles haplotypes!

It's not the we have only 14 West European members; that's just the number with 67-markers tested.

Now it seems likely to me that the British Isles group, with 108 67-marker haplotypes compared, has probably just about maxed out in haplotype variance. Its variance might increase somewhat with the addition of more haplotypes, but it doesn't seem likely that it will shift much.

But the West European group is matching the British Isles in haplotype variance, and that with just 14 67-marker haplotypes! It seems likely the addition of more West European haplotypes will only push that variance higher and beyond that of the British Isles group.

Thus it seems to me the evidence indicates that L21 originated on the European Continent and not in the British Isles.

Another thing to look at is the SNP trail. Look at the numbers of R-P312* relative to R-L21* and the numbers of other P312+ subclades, which are greater on the Continent than in the British Isles. It seems likely to me that L21 originated where P312 was spinning off its other satellites around 4,000 years ago, and that was not in the British Isles.

Also recall that the time lapses between U106, P312, and P312's subclades are not great. The expansion was rapid. It doesn't seem likely that P312 reached Britain before giving rise to L21, which then had to "back migrate" to the Continent.

Besides, we know there were movements from the Continent to the British Isles. Movement from the Isles to the Continent was far more tenuous, more recent, and less extensive.

Sorry, rms2. I just forgot to update that. You will see elsewhere that I do not think (anymore) that L21 originated in the British Isles.

Maciamo said:

Sorry, rms2. I just forgot to update that. You will see elsewhere that I do not think (anymore) that L21 originated in the British Isles.

Click to expand...

Thanks!

Once I have ten posts here and can do it, I will post a link to the R-L21* European Continent Map I am maintaining.

I'm back with 10 posts under my belt!

Here is the link to my R-L21* European Continent Map (excluding the British Isles):

http://tinyurl.com/njt8n4

That represents just the R-L21* men I know about. There are many about whom I know nothing. I know for a fact, for example, that there are a number of R-L21* from the Netherlands, France, and Denmark who have not yet joined the R-L21 Plus Project. There are probably others who have not joined yet, as well.

I'm an L21 whose paper trail leads to Catholic Tipperary Ireland. However, most of all the surnames I'm related to in YSearch and Genebase are Jewish. Does anyone have any idea how L21's came from Jewish bloodlines? My deep ancestry come out of Khazachstan and before that the Iraqi Plateau, but they are much too distant to be Jewish Khazaria and Jewish Babylon.

pcusack1 said:

I'm an L21 whose paper trail leads to Catholic Tipperary Ireland. However, most of all the surnames I'm related to in YSearch and Genebase are Jewish. Does anyone have any idea how L21's came from Jewish bloodlines? My deep ancestry come out of Khazachstan and before that the Iraqi Plateau, but they are much too distant to be Jewish Khazaria and Jewish Babylon.

Click to expand...

As far as I know only a tiny fraction of L21+ are Jewish, and they may not even be "original" Jewish lineages (converted, adopted, cuckolded fathers, etc.).

Don't trust too much STR comparisons on Y-search or Genebase. Sometimes you get fairly close results outside your haplogroup or subclade, which always mean they are mistaken about the "relatedness". Your most recent common ancestor with someone from another subclade in your own haplogroup (e.g. Another subclade of R1b) is at least several thousand years ago.

This is an excellent discussion and I have learned quite a bit.

I have done deep DNA testing and can trace my heritage back to ancient Gallaecia (modern-day Galicia / northern and central-north Portugal) and my R1b markers match exactly the southern and western Irish, although there appears to be an origination point in the Aquitaine region. Are the Celtic genetic influences in these areas essentially Gaul and / or ancient Basque?

Guienne, France

My surname supposedly comes out of Guienne, Cussac, France which is in Acquatine. Then they are reported to have moved on to Normandy then Ireland in the Norman invasion.

But there are dozens and dozens of Jewish surnames that match my L21. I think there is too many for there not to be a Jewish L21 connection.

Maciamo said:

As far as I know only a tiny fraction of L21+ are Jewish, and they may not even be "original" Jewish lineages (converted, adopted, cuckolded fathers, etc.).

Don't trust too much STR comparisons on Y-search or Genebase. Sometimes you get fairly close results outside your haplogroup or subclade, which always mean they are mistaken about the "relatedness". Your most recent common ancestor with someone from another subclade in your own haplogroup (e.g. Another subclade of R1b) is at least several thousand years ago.

Click to expand...

Do you happen to know the migration pattern for that tiny fraction of Jews who are L21's ?

pcusack1 said:

Do you happen to know the migration pattern for that tiny fraction of Jews who are L21's ?

Click to expand...

No, I don't.

pcusack1 said:

Do you happen to know the migration pattern for that tiny fraction of Jews who are L21's ?

Click to expand...

I can only comment on the Jewish L21s I personally know about. I'm sure there are many more, but thus far I only know of two, and both of them trace their ancestry to Lithuania. One, Prager, has a surname that probably indicates that the surname, at least, was acquired in Prague. The other surname is Lubatinsky (anglicized to Lewis in the USA).

These two gentlemen have no haplotype neighbors in YSearch except each other and a couple of other Ashkenazim from Eastern Europe, who, as far as I know, have not had any SNP testing.

Cambria Red said:

This is an excellent discussion and I have learned quite a bit.

I have done deep DNA testing and can trace my heritage back to ancient Gallaecia (modern-day Galicia / northern and central-north Portugal) and my R1b markers match exactly the southern and western Irish, although there appears to be an origination point in the Aquitaine region. Are the Celtic genetic influences in these areas essentially Gaul and / or ancient Basque?

Click to expand...

Galicia has a Celtic heritage. The old idea that all R1b (R1b1b2 and its subclades really) is somehow Basque is obsolete.

rms2 said:

Galicia has a Celtic heritage. The old idea that all R1b (R1b1b2 and its subclades really) is somehow Basque is obsolete.

Click to expand...

Many thanks. You have been very helpful.