New map of haplogroup R1b-L21 (S145)

Here is what Myles Dillon and Nora Chadwick say in their book, The Celtic Realms (p. 214):

Sorry, but 1800 BC is absolutely too early for P-Celtic. You have to consider that the *kw > *p shift was an innovation that was conducted probably simultaneously by P-Celtic, Osco-Umbrian (Italic) and Greek (Mycenean Greek might be considered 'Q-Greek' even though that term is never used).

Classial Greek is attested from circa the 8th century BC, Umbrian is attested from the 7th century BC, and the oldest P-Celtic languages (Lepontic) is attested from the 6th century BC. Mycenean Greek is attested up to the 12th century BC. This means the *kw to *p shift must have occured some time in between, ie the 12th and 8th century BC.
 
It is not absolutely necessary for Proto-Celtic to have existed at the very beginning of the Beaker period for it to have developed among and to have been spread by Beaker Folk. I don't know that it was spread by them. I am not speaking dogmatically.

How large are the R1a "peaks" you speak of in Auvergne and Cantabria? Once one accepts the dictum that R1a=Indo-European, then its presence anywhere can be trotted out to fuse the torque of circular argument. Since R1a=Indo-European, its presence in a Celtic-speaking region, even at a few per cent, must signify the descendants of the original Celtic settlers. Why? Because R1a=Indo-European!

The R1a among the Corded Ware people consisted of two bodies, as I recall, a father and son. That's it then: all Corded Ware folk were R1a.

The distribution of R1a in Eastern Europe is compelling evidence, perhaps, for a connection between R1a and Slavic, or R1a and the satem languages, but it is fairly sparse in the West, where IE languages are even more prevalent.
 
Sorry, but 1800 BC is absolutely too early for P-Celtic. You have to consider that the *kw > *p shift was an innovation that was conducted probably simultaneously by P-Celtic, Osco-Umbrian (Italic) and Greek (Mycenean Greek might be considered 'Q-Greek' even though that term is never used).

Classial Greek is attested from circa the 8th century BC, Umbrian is attested from the 7th century BC, and the oldest P-Celtic languages (Lepontic) is attested from the 6th century BC. Mycenean Greek is attested up to the 12th century BC. This means the *kw to *p shift must have occured some time in between, ie the 12th and 8th century BC.

Reread the quote. It doesn't say P-Celtic came to the British Isles c. 1800 BC. It says Q-Celtic ("Goidelic") did.

1350 BC is estimated to the latest possible date for the separation of Proto-Celtic from PIE. Estimates like that have a pretty wide margin of error.
 
It is not absolutely necessary for Proto-Celtic to have existed at the very beginning of the Beaker period for it to have developed among and to have been spread by Beaker Folk. I don't know that it was spread by them. I am not speaking dogmatically.

Yet you seem to thoughtfully avoiding any connections between Proto-Celtic and other branches of Indo-European.

How large are the R1a "peaks" you speak of in Auvergne and Cantabria? Once one accepts the dictum that R1a=Indo-European, then its presence anywhere can be trotted out to fuse the torque of circular argument. Since R1a=Indo-European, its presence in a Celtic-speaking region, even at a few per cent, must signify the descendants of the original Celtic settlers. Why? Because R1a=Indo-European!

6-8% as opposed to 1 or <1%? I think that's quite a peak.

The R1a among the Corded Ware people consisted of two bodies, as I recall, a father and son. That's it then: all Corded Ware folk were R1a.

The distribution of R1a in Eastern Europe is compelling evidence, perhaps, for a connection between R1a and Slavic, or R1a and the satem languages, but it is fairly sparse in the West, where IE languages are even more prevalent.

R1a is obviously not exclusively Indo-European, nowadays. However, where it occurs, you can often tie it. For instance the Finnic languages have a lot of borrowing from Proto-Indo-European (or close to it). And well, it's not exactly suprising you find 10% R1a amongst the Finns if they intermingled.

The distribution of R1a in Eastern Europe is compelling evidence, perhaps, for a connection between R1a and Slavic, or R1a and the satem languages, but it is fairly sparse in the West, where IE languages are even more prevalent.

There is also ~20% R1a in Scandinavia, and southern Scandinavia saw the Battle Axe Culture, which was an offshot of Corded Ware.
 
Well, we disagree. I'll have to take your word for the 6-8% R1a in Auvergne and Cantabria. At any rate, that doesn't strike me as compelling evidence, given all the genetic water that has passed under the historical bridge in those regions. It might be more compelling, if one could make the case for elevated levels of R1a in all or most of the Celtic-speaking regions. Besides, even though Celts no doubt lived in Auvergne and Cantabria, they are not known as places of central importance to the spread of Celtic languages. In other words, there is nothing so quintessentially Celtic about those two spots that their y-dna profiles - down to levels of haplogroups under 10% - make them worthy of special note.

Finland is in the East. There's some R1a there. It would be really surprising if there wasn't any.

Uralic and IE are said to have a number of close correspondences. Attributing them to R1a is, once again, based upon acceptance of the dictum, R1a=Indo-European.

Anyway, can we cut this short? I don't want to argue "Indo-Europeans". I've done it before, and it's no fun. You seem to fancy yourself an expert - and perhaps you are - and pepper your posts with linguistic "facts" I don't have the time or motivation to check.

This thread is supposed to be about Maciamo's L21 map.
 
Well, we disagree. I'll have to take your word for the 6-8% R1a in Auvergne and Cantabria. At any rate, that doesn't strike me as compelling evidence, given all the genetic water that has passed under the historical bridge in those regions. It might be more compelling, if one could make the case for elevated levels of R1a in all or most of the Celtic-speaking regions. Besides, even though Celts no doubt lived in Auvergne and Cantabria, they are not known as places of central importance to the spread of Celtic languages. In other words, there is nothing so quintessentially Celtic about those two spots that their y-dna profiles - down to levels of haplogroups under 10% - make them worthy of special note.

Finland is in the East. There's some R1a there. It would be really surprising if there wasn't any.

Uralic and IE are said to have a number of close correspondences. Attributing them to R1a is, once again, based upon acceptance of the dictum, R1a=Indo-European.

Anyway, can we cut this short? I don't want to argue "Indo-Europeans". I've done it before, and it's no fun. You seem to fancy yourself an expert - and perhaps you are - and pepper your posts with linguistic "facts" I don't have the time or motivation to check.

This thread is supposed to be about Maciamo's L21 map.

Yes, sorry. You're right. We should leave it at here. Let me say this. I agree that L21 matches the northern portions of the Atlantic Bronze Age, and I also agree with you that Celtic languages were spoken in the Atlantic Bronze Age (in particular the British Isles). Can we leave it at that?
 
Sure. That sounds good. I apologize for my tone. It is hard to make e-text sound sufficiently polite. I am sorry if I have offended you.
 
Sure. That sounds good. I apologize for my tone. It is hard to make e-text sound sufficiently polite. I am sorry if I have offended you.

That's allright. Don't worry. In retrospect I feel that my language may have sounded a tad harsh as well. :embarassed:
 
One thing is certain: we have R1b-P312, with its 5 Subclades, M65, Z196, U152 and L21:
-M65 is most frequent in the Basque Country (around 2-3%) and the Central&Eastern Pyrenees (atround 2-3%), and in much lower frequencies in Spain in general
-Z196 has 2 subclades:
--M153, which is found among Basques (around 2%, but much higher around the Pyrenees) and in much lower frequencies in the rest of Iberia
--L176.2 has 2 subclades, which I explain here: http://www.eupedia.com/forum/showthread.php?26793-L176.2-When-and-where-did-it-originate
-U152, which is constantly discussed and has a map
-L21, which is currently being discussed (and has a map)
-L238, which appears mostly in Scandinavia
Somewhere, sometime, R1b-P312 (and each of its subclades) originated. Where, when? Maybe a map of R1b-P312 would help (especially considering there's a quite large amount of P312*)...
 
It would be nice to see a map that showed all P312+ across Western Europe. It would obviously subsume the L21 and U152 maps as well as show all the P312*.

Regarding R1a, from looking in the Ireland yDNA project on FTDNA (which I'm a coadmin) the 30 or so R1a people less then a handfull have native Irish names. The rest of names appear mostly Scottish or English in origin. (30 out of over 4k members).

In general many regard R1a as a "Viking marker" in Ireland and Britain.

Personally I don't think "Proto-Celtic" developed in the west. It doesn't make much sense when you take into account connections to other IE families (Italic for one) and also the fact that we follow the R1b SNP's the direction of movement appears to be from East to West.
 
Regarding a west-to-east movement for Celtic, I am doubtful myself, but Koch and Cunliffe seem to think it a reasonable idea, so it is worthy of consideration. It is possible that Jean Manco's theory of the "Stelae People" is correct, and a sort of back door movement of Indo-Europeans occurred via coasting along the Mediterranean and coming up through Iberia.

But the idea of invading kurgan horsemen with the rising sun at their backs is an old one and colors a lot of the discussion of Indo-European origins.
 
Personally I don't think "Proto-Celtic" developed in the west. It doesn't make much sense when you take into account connections to other IE families (Italic for one) and also the fact that we follow the R1b SNP's the direction of movement appears to be from East to West.
Oh, neither do I, I think Proto-Celtic developed around the upper Danube (southern Germany, Austria and the Czech Republic). However, I do think the Atlantic Bronze Age was probably Celtic linguistically.
 
Updates of tables in the R1b - L21 main article

Hi

I've been looking through the forum in order not to ask what others may have been asking before, but I haven't been able to find an anwer still, so I ask myself.

The new european map of R1b-L21 was very useful. thanks for that. However it seems to me that the subclade tree and also parts of the table with the defining mutations is not fully updated with the latest nomenclature, which makes it quite confusing for us non-experts in the field.

Will there be an update of also these otherwise very informative tables in the main article?
 
I've created a "draft" L21 tree earlier today. Mainly as I was interested in getting an overview of DF21 under L21. I've based it what's currently in isogg tree along with DF21 project, along with couple extras thrown in (Z253,z254,z255). It doesn't include all the L21 SNP's as there are quite a number that appear to be private or whose position isn't know with regards to all know clades.

l21-draft2.png


URL: http://compsoc.nuigalway.ie/~dubhthach/l21-draft2.png
 
It turns out that L459 is not equivalent to L21 but under it. A L459- result came in there today. Most of the major clades of L21 are L459+, potentially a major development.
 
One thing is certain: we have R1b-P312, with its 5 Subclades, M65, Z196, U152 and L21:
-M65 is most frequent in the Basque Country (around 2-3%) and the Central&Eastern Pyrenees (atround 2-3%), and in much lower frequencies in Spain in general
-Z196 has 2 subclades:
--M153, which is found among Basques (around 2%, but much higher around the Pyrenees) and in much lower frequencies in the rest of Iberia
--L176.2 has 2 subclades, which I explain here: http://www.eupedia.com/forum/showthread.php?26793-L176.2-When-and-where-did-it-originate
-U152, which is constantly discussed and has a map
-L21, which is currently being discussed (and has a map)
-L238, which appears mostly in Scandinavia
Somewhere, sometime, R1b-P312 (and each of its subclades) originated. Where, when? Maybe a map of R1b-P312 would help (especially considering there's a quite large amount of P312*)...

everything would justesi vousne doing no big grseule error:

M65 marker Celtic Iberians is nonexistent in Basque proven, the peak of M65 is located in Asturias and Galicia and Portugal center then no cases among French and Basque Pyrenees rarely in the east or Spanish.
The M153 marker (subclade of M167) very rare almost entirely in English and Basque but also occasionally found among the Catalan, Irish and French West enigmatic Austria ..
M65 is a branch of the group P312 itself a branch of P311 which is Western European. This is the exclusive label to the east of the Iberian Peninsula and the smallest of the four branches of P312 classified by demographics.
U152,; DF27; L21. M65 and P312 * (not rated).
 
everything would justesi vousne doing no big grseule error:

M65 marker Celtic Iberians is nonexistent in Basque proven, the peak of M65 is located in Asturias and Galicia and Portugal center then no cases among French and Basque Pyrenees rarely in the east or Spanish.
The M153 marker (subclade of M167) very rare almost entirely in English and Basque but also occasionally found among the Catalan, Irish and French West enigmatic Austria ..
M65 is a branch of the group P312 itself a branch of P311 which is Western European. This is the exclusive label to the east of the Iberian Peninsula and the smallest of the four branches of P312 classified by demographics.
U152,; DF27; L21. M65 and P312 * (not rated).


https://www.familytreedna.com/public/atlantic-r1b1c/default.aspx?section=results
 
If by 'old' you mean relatively conservative, then yes. But, the same might be said about the Italic languages, with which the Celtic languages share a number of sound laws. If you disregard the fact that they are Satem languages, the Baltic languages might be also considered conservative, however.

Anyways, Celtic shares with Germanic, Greek and Italic to have merged the so-called 'palatovelar' series of PIE (k´, g´ and g´h - which become s, z and ʒ-like sounds in the Satem languages) with the plain velar sounds (k, g, and gh).

k´ > k
g´ > g
g´h > gh

This is known to have occured after the early 3rd millennium BC (common Germanic-Baltoslavic forms which predate the Centum/Satem split and must have arisen in Corded Ware times), but before the mid-2nd millennium BC (attestation of Mycenean Greek, the oldest attested Centum language).

One common sound law is the occurence of *p to *kw before *kw. For example:

PIE 'Penk´we'
Latin 'Quinque'
Old Irish 'Coic'

Note that this is an innovation that the Celtic language share not only with the Italic languages, but also with Venetic. Also note that, of course, the P-Celtic languages essentially 'hidden' that feature because they have shifted *kw to *p, which is why the word is 'Pimpetos' in Gaulish, 'Pump' in Welsh and 'Pemp' in Breton.

Another common Italo-Celtic shape is the existence of -ī as a genitive form:

Latin - Dominus; Dominī
Gaulish - Mapos; Mapī
Archaic Irish - Maqqos; Maqqī

(it's interesting that this is attested in Archaic Irish, because this is generally a language, despite attested very late, that is rather close to Proto-Celtic)

There are also common words for gold, silver and tin in the Italic and Celtic languages, but not for iron. Conversely, the word for iron is shared between the Celtic and the Germanic languages, which matches the archaeological fact that iron working arrived in the (Proto-Germanic) Jastorf Culture through contact with the (Celtic) Hallstatt Culture.

It's really one of the main problems which I have with the Atlantic hypothesis is really that it somehow assumes the Celtic languages come out of thin air, woefully ignoring the interrelationship they have with the other branches of Indo-European. If the Celtic languages were really completely distinct from other branches of IE. Such can be seen with the Anatolian languages, which by the way, have retained some features of PIE lost elsewhere, such as the laryngal sounds.



I think you are wrong here. As I said, if you look at the distribution of R1a in western Europe, there are peaks in Auvergne and Cantabria. Are these coincidential? I don't really think they are. As for the connection between R1a and the IE languages I think the distribution in Eastern Europe is fairly compelling. For one, R1a has been found (exclusively) in graves of the Corded Ware Culture, and the distribution of R1a in Central-Eastern Europe (including Scandinavia) matches very well to the extend of Corded Ware. There is also a number of shared vocabulary shared by Germanic and Balto-Slavic languages which predates the Centum/Satem split (for example the word for gold). The Corded Ware Culture is very much fitting exactly these conditions for a common pattern.

The high percentages of Y-R1b-L21andwithin it L222 in W-Norway is intriguing me -

I find it very surprising slaves malesof Vikings ages should have been sent to Norway in so a huge numberand allowed to reproduce themselves (with the help of some females, Iconcede) to produce a 12-15% of some norwegian districts !
So I look for other events to explainit : Bronze Age ? The maritime distribution of Y-R-L21 inNW-Europe and its apparent today association(statistical) with mt-H1 and H3, frequent from Iberia toScandinavia-Finland push me to believe in a possible old enoughsettling in W-Europe ; not to say R1b-L21 came from somewherewith mt-H1+H3, but that it found it at some stage and moved with somebearers of it between some places after that... - I'm not sure yetL21 were not in W-Europe before Bronze (we have almost no Y-DNA inNW-Europe between middle Neolithic and metal ages, do correct meif I mistake ; – If I red well even L222 would not beborn in Ireland but only knew a founder effect in this isle (thissaid because the mtH1+H3 are not so dense in Ireland of today) -
I know some scholars said mtDNA had alot of chances to be partly natural pressure selected but at littlescale of time it does not seem to drastic, because H1+H3 seem strongin countries as far away the one from the other as Iberia andFinland... -
concerning Celts and Y-R1a I'm temptedto associate western R1a to Corded I consider as a proto-satemspeaking group – I see the Celts evolving in regions were BBs had astrong influence where later were found rathermore Y-R1b and maybe Y-I2a2 (ex I1c ex I2b), without I think Celts =BBs at all ; I think BBs evolved with time, from a wellorganized federation of litlle groups of metals prospectors (I agreehere with the old traditional view) from central-south-east Europe(Croatia?) ultimately from South the Black Sea ? (according totheories concerning the bronze melting technics replacing the oldways of Balkans-Carpathians) – their skills permitted rapidprogresses in the regions they setlled – BBs I think had enoughskills and selfdefense power to be respected, not enough to invade orpopulate large countries – it's very possible that the last BBsaspects were more the result of accultured people, accultured byother accultured ones, were the first BBs had some descendants, moreand more crossed with the local populations as acculturation andassimilation were running on, these last in western Europe being forthe most celtic speaking ?...
to come back to Y-R1b-L21, they couldhave known two cultures, a non-I-Ean one and a I-Ean one, turned intoceltic early enough : the Qw- Celts ? I see them here a bitbefore S-Britain BBs settled from the Netherlands-Westphalen aboutthe 2500 BC -
& :'metal ages' is confusing : we can imagine metallurgists tookfoot in West before the all-around accepted datation for copper andbronze as collective well spred phenomenon in West) and BBs weremaybe not the first ones ?



  • for Auvergne we have limited sample, with, I think, the weight of an industrial town as Clermont-Ferrand, so I have some mistrust in this lonesome survey – and do not forget East-Auvergne is in contact with Forez which was a terminal of Burgundian invaders as Lyonnais and Dauphiné (for Y-R1a) – I don't believe Celts had ever been strong for Y-R1a what does not exclude some rare SNP's... - but I confess the very low level of Y-I1 in Auvergne does not seem checking a northern germanic population : maybe are we building too much theories based on too little data here ?


Notice my pronostics are all bets andthat I want more ancient DNA to make a sound opinion, it's safer,evidently !!! – (my brain works better in analysis than insynthesis) – I need to tickle my brain to search possible ways ofarriving for R-L51 descendants in West – Mediterranea is maybe nota so ridiculous hypothesis for the P312 group ??? wellseparated from the ancestors of Y-R1b6U106 ? - some celticlegends involving E-Mediterranean and North Africa could have a partof truth ??? or they mix the true celtic history with the BBshistory, making a medley, what is not impossible at all knowing howlegends and myths work ?
 
I have updated the R1b-L21 map using the more accurate data for Iberia from Valverde et al. (2016) and for France from Lucotte et al (2015). The new data allowed me to add L21 in Corsica, Sardinia and around Rome, and to increase it in north-western France and around Paris. I have also revised down the frequencies in England, Lowland Scotland, Leinster and Ulster.

Haplogroup-R1b-L21.png
 

This thread has been viewed 105658 times.

Back
Top