Ancient place names in Iberia

[Taranis]

No, Taranis, I did not say that. I have say that is in western Hispania where you can find the oldest peninsular celtic languages, where derive celtiberian, and in the french pre-Alps (Rhone basin) and Alps were you can find the gaulish ancestor, and perhaps of the latin-falisco.

Well, you have been implying this due to the adherence of sound laws. If you argue that *p does not become *kw before *kw in western 'Hispano-Celtic' languages, which as I stated is a sound law that the Celtic family and the Italic family have in common, this per definition means that the Italic and Celtic languages are closer to each other than to these Western 'Celtic' languages, which in turn means the latter can, by the very definition, be not Celtic.

You also get into a very huge general problem from the linguitic perspectic if you want to define what exactly a Celtic language is and what not, if you say that neither *p > Ø nor *p > *kw before *kw are defining for a Celtic language.

About the irish language origin it is not from Spain primarily, but related with the celto-hispanic. This is in agreement with the archaeology, anthropology and genetic studies.

This is also impossible. Irish (at least archaic Irish) was closest to Proto-Celtic, in terms of conservativeness and lack of innovations.

When you mention some celtiberian inscriptions you must do the reference with K and its number (like the tartessian with the letter J), not with a summarized table of laws that derives fundamentally of what we know about the gaulish.

I mentioned the Botorrita inscriptions, which after all represent the main corpus of Celtiberian. All sound laws I have stated apply to the Botorrita inscriptions.

These laws cannot be applied as an absolute law because there are greatest divergences between the celtic dialects.

It's also kind of funny that you come around to say in the end that you admit we are not talking about a homogenous linguistic area. I still maintain that sound laws cannot be in free variation.

PS: Why do you bring up Tartessian here?

 

[Asturrulumbo]

This is a complex subject that I am just now beginning to grasp. My opinion is that, in Iberia, Celtic language influences emerged sometime in the Bronze Age and spread along the Atlantic Facade through social and commercial exchange. I suspect that the first IE language in the Iberian Peninsula may not have been Lusitanian - considered by most scholars at the moment as Para-Celtic. Celticity may well have surfaced in two different regions, Iberia to Orkney and Central Europe.

That too is what I believe. A split of the... "Celtoid" (to include Lusitanian) languages towards the Late Bronze Age, dividing into 2 main gruoups, one along the Atlantic fringe and the other around Gaul and Central Europe. Brittonic languages may belong to the latter group (or at least have a superstrate belonging to it) as a result of the Hallstatt/La Tene expansions.

 

[Taranis]

That too is what I believe. A split of the... "Celtoid" (to include Lusitanian) languages towards the Late Bronze Age, dividing into 2 main gruoups, one along the Atlantic fringe and the other around Gaul and Central Europe. Brittonic languages may belong to the latter group (or at least have a superstrate belonging to it) as a result of the Hallstatt/La Tene expansions.

Yes, this is essentially (more or less) what I too think is the case. What Burgundis brought up is the question if Lusitanian is really the whole story or if there is more, and if we may be looking to other "Celtoid" as you called them languages in ancient Gallaecia. I think we are looking more or less at this:

- Lusitanian diverges before Proto-Celtic features developed, which is why I think that it's likely that they were amongst the first IE inhabitants on the Iberian penninsula.

- I would argue for Goidelic as the most conservative branch of the Celtic languages (at least, up to Archaic Irish) that must have arrived sufficiently early in Ireland and remained relatively isolated (which is backed by archaeology as for instance, Ireland saw only peripherial La-Tene influence).

- Celtiberian as the first branch of Celtic to diverge and to develop innovations on it's own found nowhere else. The question, of course is where did the Celtiberians come from and when did they arrive. If the Celtiberians were native since the early Bronze Age, the question is why and how they could commonly develop Proto-Celtic sound features if for example Lusitanian could not do that. From that perspective, I wondered if the Celtiberians might have arrived at a later point than Lusitanian (from the British Isles?).

- Development of a 'Britanno-Gallic' branch in Central Europe with Hallstatt and La-Tene. As you said, it's also absolutely possible that this was basically a superstrate in Britain due to Hallstatt/La-Tene influence.

- Development of the typically 'Insular Celtic' features in the modern Celtic languages as result of Christianization and the existence of a sprachbund during the Dark Ages.

 

[Cambrius (The Red)]

Yes, this is essentially (more or less) what I too think is the case. What Burgundis brought up is the question if Lusitanian is really the whole story or if there is more, and if we may be looking to other "Celtoid" as you called them languages in ancient Gallaecia. I think we are looking more or less at this:

- Lusitanian diverges before Proto-Celtic features developed, which is why I think that it's likely that they were amongst the first IE inhabitants on the Iberian penninsula.

- I would argue for Goidelic as the most conservative branch of the Celtic languages (at least, up to Archaic Irish) that must have arrived sufficiently early in Ireland and remained relatively isolated (which is backed by archaeology as for instance, Ireland saw only peripherial La-Tene influence).

- Celtiberian as the first branch of Celtic to diverge and to develop innovations on it's own found nowhere else. The question, of course is where did the Celtiberians come from and when did they arrive. If the Celtiberians were native since the early Bronze Age, the question is why and how they could commonly develop Proto-Celtic sound features if for example Lusitanian could not do that. From that perspective, I wondered if the Celtiberians might have arrived at a later point than Lusitanian (from the British Isles?).

- Development of a 'Britanno-Gallic' branch in Central Europe with Hallstatt and La-Tene. As you said, it's also absolutely possible that this was basically a superstrate in Britain due to Hallstatt/La-Tene influence.

- Development of the typically 'Insular Celtic' features in the modern Celtic languages as result of Christianization and the existence of a sprachbund during the Dark Ages.

Very insightful comments. A thought: could a form of Archaic Irish have developed in western Iberia prior to the arrival of the Lusitanians?

 

[Asturrulumbo]

Yes, this is essentially (more or less) what I too think is the case. What Burgundis brought up is the question if Lusitanian is really the whole story or if there is more, and if we may be looking to other "Celtoid" as you called them languages in ancient Gallaecia. I think we are looking more or less at this:

- Lusitanian diverges before Proto-Celtic features developed, which is why I think that it's likely that they were amongst the first IE inhabitants on the Iberian penninsula.

I believe this too, however I do not think they were the only IE inhabitants of Iberia... Actually, I think all the "Celtoid" speakers of Iberia arrived around the same time; the lack of Common Celtic sound changes in Lusitanian may be ascribed to a later isolation (which led to Lusitanian not participating in them), perhaps?

I would argue for Goidelic as the most conservative branch of the Celtic languages (at least, up to Archaic Irish) that must have arrived sufficiently early in Ireland and remained relatively isolated (which is backed by archaeology as for instance, Ireland saw only peripherial La-Tene influence).

Indeed this is the case (both in Hallstatt and La Tene), and the same applies for Iberia (I attached a map of the Hallstatt culture at its peak)

Celtiberian as the first branch of Celtic to diverge and to develop innovations on it's own found nowhere else. The question, of course is where did the Celtiberians come from and when did they arrive. If the Celtiberians were native since the early Bronze Age, the question is why and how they could commonly develop Proto-Celtic sound features if for example Lusitanian could not do that. From that perspective, I wondered if the Celtiberians might have arrived at a later point than Lusitanian (from the British Isles?).

I have thought that too, but the main setback against this I think would be genetic evidence: one would expect to see a lot more L21 in Iberia if things were so. However, I do believe that both the (R1b) immigrants to the British Isles and those to Iberia came probably at the same time and from around the same place.

Development of the typically 'Insular Celtic' features in the modern Celtic languages as result of Christianization and the existence of a sprachbund during the Dark Ages.

It may be this, or it may be that the "Insular Celtic" features are substratal "relics" of a much stronger relationship, possibly genetic, among the Insular Celtic languages.

 

[Taranis]

Very insightful comments. A thought: could a form of Archaic Irish have developed in western Iberia prior to the arrival of the Lusitanians?

Honestly, I'm not quite sure why and how. Are you alluding to the Míl Espáine?

I believe this too, however I do not think they were the only IE inhabitants of Iberia... Actually, I think all the "Celtoid" speakers of Iberia arrived around the same time; the lack of Common Celtic sound changes in Lusitanian may be ascribed to a later isolation (which led to Lusitanian not participating in them), perhaps?

Well, there is that possibility. In that case one has to assume some kind of sprachbund that was crossing the Biscay.

Indeed this is the case (both in Hallstatt and La Tene), and the same applies for Iberia (I attached a map of the Hallstatt culture at its peak)
View attachment 5198

The big difference here is actually that Goidelic remained more conservative than Celtiberian (at least until the switch from Archaic to Old Irish).

I have thought that too, but the main setback against this I think would be genetic evidence: one would expect to see a lot more L21 in Iberia if things were so. However, I do believe that both the (R1b) immigrants to the British Isles and those to Iberia came probably at the same time and around the same place.

Yes, I noticed this before. There is surprisingly little L21 in Iberia, and I agree that it's likely that L21 and the Iberian R1b clades arrived around the same time. From that, I actually have come up with a very different idea which I consider somewhat speculative at this point, but which may be an alternative that may explain more what we are seeing. During the Dark Ages, we have the migration of Brythonic people to Aremorica and (to a much smaller scale) Galicia due to the Anglo-Saxon invasion in Britain:

http://upload.wikimedia.org/wikipedia/commons/f/f9/Britonia6hcentury.png

My idea is, what if something similar to this happened ~1200 years earlier? That we have a migration of Q-Celtic people from the British Isles and from Aremorica into the Iberian penninsula as a result of the (P-Celtic) Hallstatt expansion into Britain? I think that this scenario does also explain the bronze-iron age transition on the Iberian penninsula, especially the sudden increase of settlements amongst the Castro Culture. In that scenario, the "Celtoid" languages of Iberia arrived simultaneously to the Proto-Celts on the British Isles, and Celtic languages arrived only later in Iberia.

I'm not convinced myself that the scenario is viable, but it would explain why L21 is so rare on the Iberian penninsula.

It may be this, or it may be that the "Insular Celtic" features are substratal "relics" of a much stronger relationship, possibly genetic, among the Insular Celtic languages.

Well, there is that possibility. The idea that the Insular Celtic feature are actually the result of some non-IE substrate is actually a fairly old one. An Afro-Asiatic language has been often suggested, due to features such as VSO order and inflected preposition. The reason I think this is the more unlikely variant is that Archaic Irish (the language used in the Ogham inscriptions) is essentially 'Continental Celtic' and often shows identical declension forms as Gaulish. For that reason I'm more in favour of a later sprachbund as the source of the Insular Celtic features.

 

[callaeca]

1.- Have you make some comparative between Ogamic and hispano-celta? We can do it here if you want, perhaps, you are going to be surprised...Yes it is archaic, but younger tha hispano-celta (de Bernardo Stempel, 2004).
2.- The modern studies referent to indo-european languages of Hispania, suggest the terminology celto-hispanic, where you can do this división:
2.1. celtiberian
2.2. hispano-celta, includes the next dialects: cantabrian, asturian, callaecian, vaccean, vettonian and lusitanian*.
*lusitanian = place-names, personal names, god names, river names, but not the typical lusitanian texts (except the aforementioned items). I am agree with Encarnaçâo, Pena or Milcar Guerra about the posibility of a diglosical language in a primitive pre-romanic phase of evolution. The inscriptions are later (II-III a.C) and in the context of 6 centuries of the romanisation.

I have not said that celto-hispanic is not kw language, I have said that celto-hipsnic have +pento- from ie *penkto- and *perko- from ie. *perkwo-, but celto-hispanic *ekwo- from ie. *ekwo-. It is regular and indo-european, it is not submit to a nonindo-european substrat. Perhaps, the celtoid was the gaulish thas loss the indo-european *p, or can you see it in tocharian, luwian, lycian, hittite, attic, ionic, doric, greek koiné, sanskrit, iranian, latino-faliscan, sabellic, venetian, latvian, prussian, lithuanian, slavic, ilirian, noric, lepontic, celto-hispanic, etc. ...?

 

[Asturrulumbo]

Yes, I noticed this before. There is surprisingly little L21 in Iberia, and I agree that it's likely that L21 and the Iberian R1b clades arrived around the same time. From that, I actually have come up with a very different idea which I consider somewhat speculative at this point, but which may be an alternative that may explain more what we are seeing. During the Dark Ages, we have the migration of Brythonic people to Aremorica and (to a much smaller scale) Galicia due to the Anglo-Saxon invasion in Britain:

http://upload.wikimedia.org/wikipedia/commons/f/f9/Britonia6hcentury.png

My idea is, what if something similar to this happened ~1200 years earlier? That we have a migration of Q-Celtic people from the British Isles and from Aremorica into the Iberian penninsula as a result of the (P-Celtic) Hallstatt expansion into Britain? I think that this scenario does also explain the bronze-iron age transition on the Iberian penninsula, especially the sudden increase of settlements amongst the Castro Culture. In that scenario, the "Celtoid" languages of Iberia arrived simultaneously to the Proto-Celts on the British Isles, and Celtic languages arrived only later in Iberia.

I'm not convinced myself that the scenario is viable, but it would explain why L21 is so rare on the Iberian penninsula.

Hmm... That theory would need to be much more thoroughly elaborated. Are you suggesting most of the L21 and the rest of R1b came to Iberia with that (hypothetical) migration? If not, when? And do we see as much R1b-P312* in Britain (as we do in Iberia) to suggest it came from there? Also, did the Pre-Hallstatt people of Britain have L21 in your theory, or was that also brought by the Hallstatt expansion?

Well, there is that possibility. The idea that the Insular Celtic feature are actually the result of some non-IE substrate is actually a fairly old one. An Afro-Asiatic language has been often suggested, due to features such as VSO order and inflected preposition. The reason I think this is the more unlikely variant is that Archaic Irish (the language used in the Ogham inscriptions) is essentially 'Continental Celtic' and often shows identical declension forms as Gaulish. For that reason I'm more in favour of a later sprachbund as the source of the Insular Celtic features.

Good point, I hadn't thought of that... And I agree that the theory of an Afro-Asiatic substrate for IC languages is quite unlikely (especially since the proposal is to link these Afro-Asiatic peoples with megalithism, and I think genetics have very much ruled that out.)

 

[Taranis]

1.- Have you make some comparative between Ogamic and hispano-celta? We can do it here if you want, perhaps, you are going to be surprised...Yes it is archaic, but younger tha hispano-celta (de Bernardo Stempel, 2004).

I'm not going to be surprised. Yes, Oghamic Irish is younger than Hispanian Celtic, yes, but more conservative. In fact, in regard for a number of Proto-Celtic sound laws, even Old Irish is more conservative in respect to them than Celtiberian despite being younger:

Proto-Celtic *nm retained in Goidelic, but shifted to *lm (and further to *lb) in Celtiberian:
- Old Irish 'Ainm' ('name'), 'Menma' ('mind')
- Celtiberian 'Albana' (Botorrita III), 'Melmu' (Botorrita I)
(for non-Celtic cognates, compare English 'Name' and Latin 'Nomen', or Greek 'Μνήμη'.

Proto-Celtic *mn retained in Goidelic, but shifted to *un in Celtiberian (as well as, interestingly, Gaulish and Brythonic):
- Old Irish 'damnain' ('to bind'), 'follamnugad' ('ruling') (compare 'Catuvellaunos')
- Celtiberian 'taunei' (to bind, Botorrita I)

(for non-Celtic cognate, compare Latin 'damnare'.)

As you can see, even Old Irish was the more conservative language (closer to Proto-Indo-European) than Celtiberian, despite the fact that Celtiberian is attested from earlier.

2.- The modern studies referent to indo-european languages of Hispania, suggest the terminology celto-hispanic, where you can do this división:
2.1. celtiberian

2.2. hispano-celta, includes the next dialects: cantabrian, asturian, callaecian, vaccean, vettonian and lusitanian*.
*lusitanian = place-names, personal names, god names, river names, but not the typical lusitanian texts (except the aforementioned items). I am agree with Encarnaçâo, Pena or Milcar Guerra about the posibility of a diglosical language in a primitive pre-romanic phase of evolution. The inscriptions are later (II-III a.C) and in the context of 6 centuries of the romanisation.

In my opinion, despite it's late occurence Lusitanian is impossible to be a "diglossic" or pre-Romance language because it has sound laws found neither in Latin nor in the Celtic languages (for example PIE *bh > *f at intervocalic positions).

I have not said that celto-hispanic is not kw language, I have said that celto-hipsnic have +pento- from ie *penkto- and *perko- from ie. *perkwo-, but celto-hispanic *ekwo- from ie. *ekwo-. It is regular and indo-european, it is not submit to a nonindo-european substrat. Perhaps, the celtoid was the gaulish thas loss the indo-european *p, or can you see it in tocharian, luwian, lycian, hittite, attic, ionic, doric, greek koiné, sanskrit, iranian, latino-faliscan, sabellic, venetian, latvian, prussian, lithuanian, slavic, ilirian, noric, lepontic, celto-hispanic, etc. ...?

I am not talking about the loss of *p in Celtic here. I am talking about the common sound law of the Italic and Celtic languages that change *p inherited from PIE to *kw if another *kw is in the same word. The best example, which I have mentioned several times now, is this:

*penkwe > *kwenkwe (Latin 'Quinque', Old Irish 'Coic')
(compare with English 'five', Greek 'pente')

Naturally, this is a development that predates both the emergence of the Celtic and the Italic languages (therefore the description 'Italo-Celtic' is pretty accurate).

If you say that some languages in the western Iberian penninsula retained the *p as it is found *penkwe, they by definition would be non-Italo-Celtic.

 

[Taranis]

Hmm... That theory would need to be much more thoroughly elaborated. Are you suggesting most of the L21 and the rest of R1b came to Iberia with that (hypothetical) migration? If not, when? And do we see as much R1b-P312* in Britain (as we do in Iberia) to suggest it came from there? Also, did the Pre-Hallstatt people of Britain have L21 in your theory, or was that also brought by the Hallstatt expansion?

As I said, I'm not sure the scenario is that viable (but I said that from the start, remember? ) But, in this scenario, L21 would have been the dominant Haplogroup of the Bronze Age population in Britain (as well as Aremorica), and would have been carried to Iberia during that (hypothetical) migration. There is of course the possibility that this is even older (say, the Wessex Culture was already Proto-Celtic-speaking), and we did see such a spread of Proto-Celtic to the south during the Atlantic Bronze Age. In that scenario I think the pattern in respect for L21 would be sort of similar. The problem I kind of have with that is that is that the Wessex Culture seems a bit too early to be a good candidate for Proto-Celtic. However, I admit I may be mistaken on that one.

My point really is that if Celtic languages were spoken in the Atlantic region during the Bronze Age (which I find not only plausible but absolutely convincing), then the British Isles (rather than Iberia) would have been the source point. The strongest case for this in my opinion the linguistic diversity of the ancient Iberian penninsula versus the completely absence of non-Celtic languages on the British Isles. I also would argue that the pattern of L21 (and the dominance of L21 on the Bitish Isles) that we see matches this phenomenon.

Good point, I hadn't thought of that... And I agree that the theory of an Afro-Asiatic substrate for IC languages is quite unlikely (especially since the proposal is to link these Afro-Asiatic peoples with megalithism, and I think genetics have very much ruled that out.)

Indeed, genetics has prettymuch ruled out any links there.

 

[callaeca]

If you say that some languages in the western Iberian penninsula retained the *p as it is found *penkwe, they by definition would be non-Italo-Celtic.

Then with your methodology Celtiberian is not a celtic language, (and, please, you do not circumscribe the item*pente- only in the western ...it is present in all of celto-hispanic dialects: see above).

Proto-Celtic *nm retained in Goidelic, but shifted to *lm[...]Proto-Celtic *mn retained in Goidelic, but shifted to *un in Celtiberian

Yes, it was a good idea of Stifter. But again, you circumscribe the items to celtiberian language and forgetting other celto-hispanic dialects like two lusitanian personal names MELMANIVS or the galician place name Melmanzo. It is an innovation, where the archaic celto-hispanic form was MEMUNOS (K.1.3) (< *menm-ôn-), equally in gaul. MENMANDVTIS, MENMANHIAE and reduced to MEMANTUSA, MEMANDVS.

About the innovation *-mno- > *-uno- it affects to all celto-hispanic: cf. callaec. ARIOVNIS < *aryo-mno-.

This continental examples explain the marginal position of Ireland.

About the lusitanian texts, it is very unlikely that a non celtic language can survive isolated, not only at a previous celtisation, but the high degree of romanisation and its introduction in that area, from III b.C to III a.C.

There is not bh > f evolution in hispano-celta (see above), becuause this italic system afect to all aspirates and not over one especially. The regular development is *bh > b how you can see in singular dative -bo or in the widely used western personal name boudio/boudia < *bhoudh-. It is more visible from *w > f: cf. callaec. FIDUENEARUM from *widu, like in gaul. examples FRITUS-VRITUS, FLATVCIAS/VLATVS.
But it is possible an substratic efect as in old irish bréife, cuifre, fafall, Crufait, Faffand (Gearoid Mac Eoin:2007)

 

[Cambrius (The Red)]

It's not clear what the level of RL-21 really is in Iberia, as relatively few Spaniards and Portuguese have tested for it. It is found more in the north-west (Galicia, N. Portugal, Asturias). BTW, I'm RL-21* and have no direct Irish, Scottish, Welsh, etc. in my line, as far as I know.

One of the Eupedia members is an ancestry project leader for RL-21 in the British Isles, France and Iberia.

 

[callaeca]

I am RL-21*

 

[Taranis]

Then with your methodology Celtiberian is not a celtic language, (and, please, you do not circumscribe the item*pente- only in the western ...it is present in all of celto-hispanic dialects: see above).

Then the word is by definition not only non-Celtic in origin but not Italo-Celtic either, since as mentioned this is a common innovation of the Celtic languages and the Italic languages. At the same time, we do have the tribal name Quaquerni attested, which conforms to this Italo-Celtic sound law.

The problem is, you have to assume that the law *p > *kw before *kw applies in Proto-Celtic, otherwise you wouldn't be able to explain Gaulish 'Pinpetos', Welsh 'Pump', Breton 'Pemp', because instead of *kwenkw- the Proto-Celtic form would be something like *φenkw- which would be reduced further to *Øenkw- . So, we would see something like 'Oic', 'Inpetos', 'Ump' and 'Emp' in Old Irish, Gaulish, Welsh and Breton respectively.

You also get into a really big problem defining what exactly a Celtic language is and how you distinguish it from Proto-Indo-European if you say that none of the commonly held sound laws purportedly really applied. Because by your current definition I'm pretty sure that Latin or even Greek classify as Celtic language.

Yes, it was a good idea of Stifter. But again, you circumscribe the items to celtiberian language and forgetting other celto-hispanic dialects like two lusitanian personal names MELMANIVS or the galician place name Melmanzo. It is an innovation, where the archaic celto-hispanic form was MEMUNOS (K.1.3) (< *menm-ôn-), equally in gaul. MENMANDVTIS, MENMANHIAE and reduced to MEMANTUSA, MEMANDVS.

About the innovation *-mno- > *-uno- it affects to all celto-hispanic: cf. callaec. ARIOVNIS < *aryo-mno-.

This continental examples explain the marginal position of Ireland.

I don't see where you claim you are disagreeing with me? My statement that archaic Irish was more conservative than Celtic languages on the Iberian peninsula is true. If it is found in other parts of the Iberian penninsula, this circumstantiates my own statements even more.

About the lusitanian texts, it is very unlikely that a non celtic language can survive isolated, not only at a previous celtisation, but the high degree of romanisation and its introduction in that area, from III b.C to III a.C.

See, this is where I think you are making a false conjecture. You're decided that the western area of Hispania was homogenously Celtic (where I have demonstrated evidence above it might be not!), which it seems impossible for you that a non-Celtic language might survive so long. If however the area was actually not homogenously Celtic (which can be inferred from the evidence gathered) then I don't see how a language like Lusitanian could not survive that long.

It's not clear what the level of RL-21 really is in Iberia, as relatively few Spaniards and Portuguese have tested for it. It is found more in the north-west (Galicia, N. Portugal, Asturias). BTW, I'm RL-21 and have no direct Irish, Scottish, Welsh, etc. in my line, as far as I know.

One of the Eupedia members is an ancestry project leader for RL-21 in the British Isles, France and Iberia.

I'm not denying that R1b-L21 is fairly common in the northern/northwestern parts of the Iberian peninsula, but it's clear that it is rarer than in much of France and it's obviously rarer than on the British Isles. As for Irish/Scottish/Welsh ancestry, I didn't claim that in my post. I merely picked the situation during the Dark Ages where Brythonic people migrated to Galicia as an analogy for what might have happened over a millennium earlier at the start of the iron age.

My point only is that this scenario of an ancient, original presence of Proto-Celtic peoples in Iberia does not really make that much sense, both from the genetic and linguistic perspective. This is where my idea that the Celtic languages spread from the British Isles (as well as western France) to Iberia during either the Bronze Age or early Iron Age comes from. The later would be like the analogue I described above - but probably the former is far more likely. I think this scenario does explain the patterns that we see much better than the other way round.

 

[Asturrulumbo]

As I said, I'm not sure the scenario is that viable (but I said that from the start, remember? ) But, in this scenario, L21 would have been the dominant Haplogroup of the Bronze Age population in Britain (as well as Aremorica), and would have been carried to Iberia during that (hypothetical) migration. There is of course the possibility that this is even older (say, the Wessex Culture was already Proto-Celtic-speaking), and we did see such a spread of Proto-Celtic to the south during the Atlantic Bronze Age. In that scenario I think the pattern in respect for L21 would be sort of similar. The problem I kind of have with that is that is that the Wessex Culture seems a bit too early to be a good candidate for Proto-Celtic. However, I admit I may be mistaken on that one.

My point really is that if Celtic languages were spoken in the Atlantic region during the Bronze Age (which I find not only plausible but absolutely convincing), then the British Isles (rather than Iberia) would have been the source point. The strongest case for this in my opinion the linguistic diversity of the ancient Iberian penninsula versus the completely absence of non-Celtic languages on the British Isles. I also would argue that the pattern of L21 (and the dominance of L21 on the Bitish Isles) that we see matches this phenomenon.

But then if these people were L21-dominant by the time of the hypotethical migration, I don't see how it could be explained that most of the Iberian R1b is P312*.
Also, about the Wessex Culture as "Celtoid" or Celtic, I highly doubt so. It was a direct continuation of the Beaker culture, and had extensive megalithic practises. Also, consider that there does seem to have been a quite large population upheaval in Britain in the transition from the Middle (Wessex Culture) BA to the Late (Atlantic Bronze Age) BA.

 

[Asturrulumbo]

My point only is that this scenario of an ancient, original presence of Proto-Celtic peoples in Iberia does not really make that much sense, both from the genetic and linguistic perspective. This is where my idea that the Celtic languages spread from the British Isles (as well as western France) to Iberia during either the Bronze Age or early Iron Age comes from. The later would be like the analogue I described above - but probably the former is far more likely. I think this scenario does explain the patterns that we see much better than the other way round.

I too am rather skeptic about Koch's "Celtic from the West" hypothesis. But, I insist, it seems to me that the Bronze Age (hypothetical) population replacements en Iberia and Britain were parallel, not from one place from another (although in any case, a population movement from Britain to Iberia would seem much more likely to me). I do agree, however, that Western France could be a prime candidate for the origin of these population upheavals.

 

[Taranis]

But then if these people were L21-dominant by the time of the hypotethical migration, I don't see how it could be explained that most of the Iberian R1b is P312*.

Well, my idea here was that the substantial occurance of non-L21/non-U152 R1b-P312 in Iberia comes from the earlier IE wave(s?) into Iberia.

Also, about the Wessex Culture as "Celtoid" or Celtic, I highly doubt so. It was a direct continuation of the Beaker culture, and had extensive megalithic practises. Also, consider that there does seem to have been a quite large population upheaval in Britain in the transition from the Middle (Wessex Culture) BA to the Late (Atlantic Bronze Age) BA.

Well, this is a problem I have as well. From the archaeological perspectie, it's not sure what to make of Beaker-Bell, because it makes much more sense as an indigenous, natively Western European phenomenon rather than something foreign introduced. In particular due to Megalithic practices. I agree with you that the great upheavals (that occur more or less simultaneously to the Bronze Age collapse in the Eastern Mediterranean) are a much better candidate for the arrival of Proto-Celtic on the British Isles than earlier. I've seen people be favourable of an earlier date however. In particular, the Eupedia project leader rms2 is in favour of an earlier date. I also think, one conclusion that can be made here is that we may face a paradox here.

I too am rather skeptic about Koch's "Celtic from the West" hypothesis. But, I insist, it seems to me that the Bronze Age (hypothetical) population replacements en Iberia and Britain were parallel, not from one place from another (although in any case, a population movement from Britain to Iberia would seem much more likely to me). I do agree, however, that Western France could be a prime candidate for the origin of these population upheavals.

Well, the issue where Koch has a point is that Celtic languages were spoken in the Atlantic region during the Bronze Age. This is a big divergence from the 'Classicist' hypothesis if you will which sought the complete origin of the Celtic languages in the cultures of Hallstatt and La-Tene. This is prettymuch impossible, and Koch is absolutely right about that one, but he's taking it too far into the opposite direction. If we say the Celtic languages originated in the Atlantic Bronze Age, we have no real possibility to explain the presence of Celtic languages in Central Europe (especially due to the absence of a west-to-east migration movement), and neither can we explain the abundance of non-Celtic languages in Iberia. Thus, it just occured to me that the idea that the British Isles were Celticized earlier than northern Iberia (and that a migration took place from the British Isles / western France) is a possibility that nobody had considered so far, and that this would produce exactly the patterns we see.

 

[Asturrulumbo]

Well, my idea here was that the substantial occurance of non-L21/non-U152 R1b-P312 in Iberia comes from the earlier IE wave(s?) into Iberia.

From an archaeological standpoint, we face the same problem in Iberia (in the early and middle Bronze Ages) with the various cultures as we do in Britain: there seems to be a continuation with the Beaker culture, which seems to have originated in Iberia around 2900 BC and then spread east (the complete opposite of the east-to-west movement we would expect to see if it were an IE expansion.).

Well, the issue where Koch has a point is that Celtic languages were spoken in the Atlantic region during the Bronze Age. This is a big divergence from the 'Classicist' hypothesis if you will which sought the complete origin of the Celtic languages in the cultures of Hallstatt and La-Tene. This is prettymuch impossible, and Koch is absolutely right about that one, but he's taking it too far into the opposite direction. If we say the Celtic languages originated in the Atlantic Bronze Age, we have no real possibility to explain the presence of Celtic languages in Central Europe (especially due to the absence of a west-to-east migration movement), and neither can we explain the abundance of non-Celtic languages in Iberia. Thus, it just occured to me that the idea that the British Isles were Celticized earlier than northern Iberia (and that a migration took place from the British Isles / western France) is a possibility that nobody had considered so far, and that this would produce exactly the patterns we see.

Basically what I believe too. I would point, for example, to the area between the Loire and the Garonne: it coincides genetically and geographically (although I must admit, I don't know if archaeologically.)

 

[Taranis]

From an archaeological standpoint, we face the same problem in Iberia (in the early and middle Bronze Ages) with the various cultures as we do in Britain: there seems to be a continuation with the Beaker culture, which seems to have originated in Iberia around 2900 BC and then spread east (the complete opposite of the east-to-west movement we would expect to see if it were an IE expansion.).

I agree that this is a problem, one of quite a number of problems with the identification of Beaker-Bell as an Indo-European (or even Celtic) culture. Another problem is the vast extend of Beaker-Bell into areas that are later non-Celtic (notably Jutland and northern Germany) and even North Africa. From that perspective, there are strong arguments that make the Beaker-Bell Culture an unlikely candidate for being an Indo-European-speaking culture. On the other hand, I find the case that Beaker-Bell was the initial carrier of R1b much more compelling.

Another problem with Beaker-Bell is that the Celtic languages in my opinion cannot be older than the Bronze Age (due to shared Italo-Celtic vocabulary), and the Beaker-Bell Culture (or 'phenomenon' if you will) started out in the Copper Age. This is admittedly a weak argument by itself, but I think that the multitude of Italo-Celtic forms (especially due to their presence in Goidelic) cannot be dismissed as a phenomenon stemming solely from language contact and must more likely be expected to derive from a close ancestry of the two languages.

Basically what I believe too. I would point, for example, to the area between the Loire and the Garonne: it coincides genetically and geographically (although I must admit, I don't know if archaeologically.)

I agree on that one.

The interesting part is though that approximately south of the Garonne we have a non-Indo-European (Basque-Aquitanian) area in Antiquity.

 

[Asturrulumbo]

I agree that this is a problem, one of quite a number of problems with the identification of Beaker-Bell as an Indo-European (or even Celtic) culture. Another problem is the vast extend of Beaker-Bell into areas that are later non-Celtic (notably Jutland and northern Germany) and even North Africa. From that perspective, there are strong arguments that make the Beaker-Bell Culture an unlikely candidate for being an Indo-European-speaking culture. On the other hand, I find the case that Beaker-Bell was the initial carrier of R1b much more compelling.

In that you may be right, however, I do see some problems with this interpretation. First of all, how and when would R1b have reached the Beaker lands? I myself am very skeptic about the palaeolithic R1b theory. We also have tested a fair amount of Neolithic y-DNA in Europe (though is is true that none belonging to the Beaker culture), and none has been R1b-positive. Also, there is a considerable amount of I2 in many parts that belonged to the Beaker Culture (including North Africa, and although the same applies to R1b, almost all of it in North Africa is R-V88). Besides that, the east-west STR cline of R1b in Europe does not apply well with the west-east Beaker expansion.

The interesting part is though that approximately south of the Garonne we have a non-Indo-European (Basque-Aquitanian) area in Antiquity.

That is why, in my opinion, an expansion from Gaul any more to the south would not be feasible. Also, I recently thought that L21 could have originated around 3500 years ago between the upper Seine and middle Garonne and then spread west to Aremorica, from which it could have migrated to Britan and (to a lesser extent) Iberia.