New R1a map

[zanipolo]

The trouble with all that is there is really nothing in the distribution of the ancient Celtic homelands relative to the modern distribution of R1a to justify it, except the assumption that R1a is somehow responsible for the origin of all Indo-European languages.

R1b clearly dominates in the old homelands of the Celts; further than that, R-P312 and its subclades dominate in the old homelands of the Celts. R1a is scarce. What there is of it can often be accounted for historically, e.g., Norwegian Vikings in Scotland, Slavs in central and southern Europe.

I am still wondering upon what kind of sample size from what study or studies we are basing our excitement over this French aberration (small as it is) of around 5% R1a.

Notice that R1a is so scarce in western Europe (the old stomping ground of the Celts) that a finding of around 5% in one rather de-populated area of France gets everyone in a tizzy of "Kurgan" ecstasy.

what is the homeland of the celts?

do you mean piedmont and south swiss area for its R1b or its mix with the illyrians in noricum and its R1a?

[rms2]

what is the homeland of the celts?

do you mean piedmont and south swiss area for its R1b or its mix with the illyrians in noricum and its R1a?

No, I don't have in mind those old 19th and early 20th century maps showing an oval blob in Central Europe with arrows sticking out of it marking the supposed "invasion" paths of the Iron Age Celts. Are there any archaeologists who still subscribe to such a thing? I doubt it.

What I had in mind when I mentioned the ancient homelands of the Celts was pretty much all of western Europe, from the Rhine-Danube region to the Pillars of Hercules and on to Ireland and northern Scotland.

The formation of the Celts as an ethno-linguistic group probably took a thousand years or more and was a complex process involving trade and the need to communicate over a wide area. The idea that it was imposed from above by some sort of predominantly R1a "elite" just doesn't make much sense, given the evidence.

[rms2]

The frequency of R1b P312 in Austria is quite problematic if we assume that R1b P312 spread from Central Europe to western Europe.


The high frequency of R1b P312 in former non IE area (Basque country, Aquitania, Iberia...) is quite problematic too




There are distributions of R1a in western Europe (Northern Spain, île de France) that aren't explained by Germanic or Slavic migrations. A celtic origin is just a solution among other. Can we discuss it on this thread?

It is very likely the Basques, etc., were once predominantly G2a and have only become mostly R1b via admixture, as they have come to resemble their predominantly R1b neighbors. We see a similar phenomenon among the Ossetians, who speak an Iranian language and were probably once mostly R1a, but who are now mostly G2a like their Caucasian neighbors.

You can't really use the frequency of a y haplogroup in a given area to judge whether or not it came from there or was once very frequent there. One has to look at haplotype variance and the SNP trail. Judging strictly by frequency, one would expect all kinds of R1b to be showing up at Neolithic sites in western Europe, yet that is clearly not happening. The oldest R1b yet found in Europe dates to about 1,000 BC (Bronze Age).

The tiny percentages of R1a found in western Europe are mostly explicable by historic references. Why strain credulity by explaining them as "kurgan elites"?

The Romans, with their advanced civilization and advanced administrative, political, and military systems were unable to impose their language on all of their empire. The Germanic barbarians, the Vikings, and the Normans were even less able to impose their languages on the people they conquered. How likely then is an "elite dominance" model for the transmission of early Indo-European languages? Not very.

Ever notice how horse-riding steppe nomad incursions throughout history fizzle - linguistically, genetically and otherwise - at the Hungarian Plain?

[Taranis]

No, I don't have in mind those old 19th and early 20th century maps showing an oval blob in Central Europe with arrows sticking out of it marking the supposed "invasion" paths of the Iron Age Celts. Are there any archaeologists who still subscribe to such a thing? I doubt it.

What I had in mind when I mentioned the ancient homelands of the Celts was pretty much all of western Europe, from the Rhine-Danube region to the Pillars of Hercules and on to Ireland and northern Scotland.

The Classical 19th/20th century of idea of the origin of the Celts in the cultures of Hallstatt and La-Tene comes from the correlation between archaeological culture of the Gauls and the Celtic languages. That these people of Hallstatt and La-Tene spoke a Celtic language is beyond doubt (it is a favourite question I often ask, if Hallstatt was not Celtic, what was it then? Germanic? Etruscan? Slavic? none of this makes any sense).

I think, a part of the "Celtic" problem is the detour that early 2000s genetics brought into the discussion. By saying that R1b was indigenous to Western Europe, and by saying that the large-scale genetic makeup of Western Europe was basically unchanged since the end of the last ice age except for a few minor additions, it was to argue that Indo-European languages had to spread without large-scale immigration, and that a spread from the culture of Hallstatt and La-Tene was without any basis.

The formation of the Celts as an ethno-linguistic group probably took a thousand years or more and was a complex process involving trade and the need to communicate over a wide area. The idea that it was imposed from above by some sort of predominantly R1a "elite" just doesn't make much sense, given the evidence.

If we however acknowledge that R1b is not native to Western Europe, which it clearly isn't, we must argue for some kind of large scale population replacement, either through invasion or immigration, in Western Europe, and that point the idea of an invasion is not that far-fetched. Also, there is enough archaeology to argue for discontinuities in Western Europe, especially when and how the Beaker-Bell Culture came to an end.

I mean, I know of the so-called "stelae people" hypothesis, but does it really make sense? Does it make sense for people from Crimea to migrate by sea all the way to western Portugal, without leaving much evidence, and then spreading the Proto-Celtic language from there across all of Western Europe? Does it really make sense to assume Proto-Celts or Proto-Indo-Europeans were in North Africa and Sardinia?

[Taranis]

It is very likely the Basques, etc., were once predominantly G2a and have only become mostly R1b via admixture, as they have come to resemble their predominantly R1b neighbors. We see a similar phenomenon among the Ossetians, who speak an Iranian language and were probably once mostly R1a, but who are now mostly G2a like their Caucasian neighbors.

You can't really use the frequency of a y haplogroup in a given area to judge whether or not it came from there or was once very frequent there. One has to look at haplotype variance and the SNP trail. Judging strictly by frequency, one would expect all kinds of R1b to be showing up at Neolithic sites in western Europe, yet that is clearly not happening. The oldest R1b yet found in Europe dates to about 1,000 BC (Bronze Age).

The tiny percentages of R1a found in western Europe are mostly explicable by historic references. Why strain credulity by explaining them as "kurgan elites"?

The Romans, with their advanced civilization and advanced administrative, political, and military systems were unable to impose their language on all of their empire. The Germanic barbarians, the Vikings, and the Normans were even less able to impose their languages on the people they conquered. How likely then is an "elite dominance" model for the transmission of early Indo-European languages? Not very.

Ever notice how horse-riding steppe nomad incursions throughout history fizzle - linguistically, genetically and otherwise - at the Hungarian Plain?

I have two issues here: the Basques have almost no G2a, less than the average Iberians or average French. The second is that we know that R1a was in Europe since the Copper Age. What - other than a magic barrier of some kind at the Rhine - should have prevented people carrying it from moving to Western Europe within the past 4000 or so years?

[rms2]

I think that the case for R1a being connected indeed with the Proto-Indo-Europeans is fairly solid. At least, I would say that the connection with Corded Ware and R1a is pretty obvious: it wasn't in Europe before the Corded Ware period. Which is why I want to ask: what alternative is there realistically to the Kurgan hypothesis? I don't see one, at least if you realize that PIE must have been a language of the Copper Age.

I think R1a has some connection to the eastern spread of Indo-European, but I think it is overly simplistic and premature to make it the y-dna source of all of PIE, especially in the West. There is too much we do not know yet and too much that simply does not make sense if Indo-European is attributed solely to R1a.

The prevalence of R1b in Armenia and eastern Anatolia, which Gamkrelidze and Ivanov argue is the Urheimat of Indo-European (whether or not one agrees with them), and the new Case for Euphratic are factors that should be considered.

It is very possible that the R1a steppe population was Indo-Europeanized under R1b influence and not the other way around.

Aside from all that, I think it highly unlikely that any of the R1a showing up on Maciamo's map in France is Bronze Age Celtic. I just don't see why it should be when it could easily be a combination of Germans, Slavs, White Russian emigres, Jews, etc.

[rms2]

I have two issues here: the Basques have almost no G2a, less than the average Iberians or average French. The second is that we know that R1a was in Europe since the Copper Age. What - other than a magic barrier of some kind at the Rhine - should have prevented people carrying it from moving to Western Europe within the past 4000 or so years?

No magic barriers, but perhaps there were human barriers. R1a has had lots of time since the Copper Age to move into western Europe in a big way, yet it never managed it.

As for the Basques, I think they have their R1b predominance by becoming genetically admixed. I think it likely they were once mostly G2a, but perhaps they were once mostly I2 of some kind. Either way, they are a small minority and should not be regarded as the measure of western European R1b, which is overwhelmingly Indo-European speaking.

We aren't still laboring under the illusion that the Basques are Paleolithic survivors, are we?

[razor]

It is very possible that the R1a steppe population was Indo-Europeanized under R1b influence and not the other way around.

I'd be curious to know how you explain this archaeologically. The theory that Maikop victoriously spread to the steppes just doesn't wash. So what's your version?

[Taranis]

I think R1a has some connection to the eastern spread of Indo-European, but I think it is overly simplistic and premature to make it the y-dna source of all of PIE, especially in the West. There is too much we do not know yet and too much that simply does not make sense if Indo-European is attributed solely to R1a.

The prevalence of R1b in Armenia and eastern Anatolia, which Gamkrelidze and Ivanov argue is the Urheimat of Indo-European (whether or not one agrees with them), and the new Case for Euphratic are factors that should be considered.

It is very possible that the R1a steppe population was Indo-Europeanized under R1b influence and not the other way around.

Aside from all that, I think it highly unlikely that any of the R1a showing up on Maciamo's map in France is Bronze Age Celtic. I just don't see why it should be when it could easily be a combination of Germans, Slavs, White Russian emigres, Jews, etc.

The problem is this: the eastern/western distinction that you see in Indo-European doesn't really exist in that way, especially if you consider language families like Germanic, Greek and Tocharian which all criss-cross the "eastern/western" distinction. My opinion is that any origin of the Indo-European languages in Anatolia requires that Proto-Indo-European is not a language of the Chalcolithic but of the Neolithic, which is a quite futile thing to argue in my opinion if you consider that there are common words for metals and metal-working in PIE.

EDIT: I would be also *highly* sceptical of the paper, because there no evidence for sound correspondence in the purported loans given in the paper.

No magic barriers, but perhaps there were human barriers. R1a has had lots of time since the Copper Age to move into western Europe in a big way, yet it never managed it.

As for the Basques, I think they have their R1b predominance by becoming genetically admixed. I think it likely they were once mostly G2a, but perhaps they were once mostly I2 of some kind. Either way, they are a small minority and should not be regarded as the measure of western European R1b, which is overwhelmingly Indo-European speaking.

We aren't still laboring under the illusion that the Basques are Paleolithic survivors, are we?

Basque is clearly not a Paleolithic language, no, but this is exactly my point. The "native" (non-IE loanwords) vocabulary is that of a language of the Copper Age, which includes agricultural terms, words for domesticated animals and most importantly also metals and metal-working. The only archaeological culture that matches such a "package" is the Beaker-Bell Culture. If you say that Beaker-Bell was already Indo-European, you have a huge problem explaining where the Basques go their metal terms from.

Note that I am not necessarily saying that Basque was the language of Beaker-Bell (there is also the Iberians to be considered), but I think that a strong case can be made that Beaker-Bell was a non-Indo-European culture.

EDIT: What should be also added is, as the late vascologist R.L. Trask pointed out, that the number of Celtic loanwords into Basque is extremely few. It is not what we would expect if the Basques were surrounded for thousands of years by Celts or Indo-Europeans.

[Diviacus]

...but I think that a strong case can be made that Beaker-Bell was a non-Indo-European culture.

However, some renowned historians think it was an Indo-European culture !

[Taranis]

However, some renowned historians think it was an Indo-European culture !

Then, how do you explain Basque terms for metal-working and metals?

[Diviacus]

Then, how do you explain Basque terms for metal-working and metals?

I will not answer for them, but why would not these terms been borrowed from their neighbours ?

[spongetaro]

The prevalence of R1b in Armenia and eastern Anatolia

Which subclades of R1b prevail in Armenia? Because Near eastern countries like Jordan or Syria have a lots of R1b too. Pharaon Tut was also R1b. However I'm not sure that those West and south west asian R1b are linked to Proto IE people.

[spongetaro]

The problem is this: the eastern/western distinction that you see in Indo-European doesn't really exist

The association of R1b with proto Celtic speakers only relies on the assumption that this so called "eastern/western" distinction among IE languages matches the R1b/R1a distribution.

Archeology and the Kurgan hypotesis doesn't make such a distinction between two "ethnic" entities within the Kurgan as Maciamo does (South Yamna R1b vs North Yamna R1a).

[razor]

Archeology and the Kurgan hypotesis doesn't make such a distinction between two "ethnic" entities within the Kurgan as Maciamo does (South Yamna R1b vs North Yamna R1a).

Exactly. There is nothing in the archaeology of the steppes which would justify such a neat division. One may, of course, argue that both R1a and R1b (and other haplogroups) were present in the mother cultures of Yamna (if the history of R1b as reconstructed would allow it). Otherwise, one may examine the Jean Manco theory (viz., that R1b was a later addition to the expanding IE hosts, at first minor, then more successful once they reached Central Europe). Other theories seem less plausible at the moment.

[Taranis]

I will not answer for them, but why would not these terms been borrowed from their neighbours ?

Well, from which neighbours? I will give you a list of the words below:

hammer - gabi
forge - sutegi
lead - beruna
smith - (h)arotz
blacksmith - olagizon
iron - burdina

These words today are unique to Basque and have no cognates in Indo-European languages. Had the Beaker-Bell people been Indo-Europeans and spread metal-working across Western Europe, why and how would the Basques have native terms? We would expect them to have borrowed Indo-European loanwords for metalworking, as for instance, the Finnic languages did.

The association of R1b with proto Celtic speakers only relies on the assumption that this so called "eastern/western" distinction among IE languages matches the R1b/R1a distribution.

Archeology and the Kurgan hypotesis doesn't make such a distinction between two "ethnic" entities within the Kurgan as Maciamo does (South Yamna R1b vs North Yamna R1a).

Well, to pick up my examples:

- The Germanic peoples today clearly have substantial amounts of R1a, and this is to be expected if we assume that the Battle Axe Culture is an offshot of Corded Ware Culture, which is known from graves (Eulau) to have been carriers of R1a. From the linguistic perspective, the Germanic languages have an ancient connection with Balto-Slavic (predating the Centum/Satem split), but became Centumized later on.

- The Tocharians are known to be R1a. You might argue that they are R1b-M73 instead (going from the present-day Uighurs, which are known to have conquered and absorbed the Tocharians when they invaded the Tarim basin), but if you consider the tree of R1b then this is far too early and far too ancient to be seriously considered. My opinion is that R1b-M73 is more likely to be Turkic, and not Indo-European. From the linguistic perspective, you must either argue that Tocharian is one of the most ancient (after Anatolian) branches to diverge, or, which is the alternative that has also some backup, assume that Tocharian was close with Proto-Italo-Celtic.

- With the Greeks, it's a bit dicy, but in my opinion the ancient Greeks were absolutely carriers of R1a (at least, in part), especially when you take a look at southern Italy and Libya, both areas which were colonized by the Greeks in Antiquity.

All three examples clearly "violate" the R1a/R1b distinction. And if we look at things, it makes sense that only R1b-L11 (in Western Europe) really got part of the Indo-European "host".

[spongetaro]

Well, from which neighbours? I will give you a list of the words below:

hammer - gabi
forge - sutegi
lead - beruna
smith - (h)arotz
blacksmith - olagizon
iron - burdina

These words today are unique to Basque and have no cognates in Indo-European languages. Had the Beaker-Bell people been Indo-Europeans and spread metal-working across Western Europe, why and how would the Basques have native terms? We would expect them to have borrowed Indo-European loanwords for metalworking, as for instance, the Finnic languages did.



Well, to pick up my examples:

- The Germanic peoples today clearly have substantial amounts of R1a, and this is to be expected if we assume that the Battle Axe Culture is an offshot of Corded Ware Culture, which is known from graves (Eulau) to have been carriers of R1a. From the linguistic perspective, the Germanic languages have an ancient connection with Balto-Slavic (predating the Centum/Satem split), but became Centumized later on.

- The Tocharians are known to be R1a. You might argue that they are R1b-M73 instead (going from the present-day Uighurs, which are known to have conquered and absorbed the Tocharians when they invaded the Tarim basin), but if you consider the tree of R1b then this is far too early and far too ancient to be seriously considered. My opinion is that R1b-M73 is more likely to be Turkic, and not Indo-European. From the linguistic perspective, you must either argue that Tocharian is one of the most ancient (after Anatolian) branches to diverge, or, which is the alternative that has also some backup, assume that Tocharian was close with Proto-Italo-Celtic.

- With the Greeks, it's a bit dicy, but in my opinion the ancient Greeks were absolutely carriers of R1a (at least, in part), especially when you take a look at southern Italy and Libya, both areas which were colonized by the Greeks in Antiquity.

All three examples clearly "violate" the R1a/R1b distinction. And if we look at things, it makes sense that only R1b-L11 (in Western Europe) really got part of the Indo-European "host".

Thank you. R1b-M173 may indeed be of Turkic origin. Also, what's your opinion about R1b L23? I don't think that it is linked with the Proto IE people.

[rms2]

I'd be curious to know how you explain this archaeologically. The theory that Maikop victoriously spread to the steppes just doesn't wash. So what's your version?

I'm not going to try to argue with ten guys at once. Please use the Kurgan Theory to explain how R1a managed to spread Indo-European to the Atlantic coast. How does that "wash"? Not well.

I don't have all the answers, but, funny thing, neither do you.

[spongetaro]

The tiny percentages of R1a found in western Europe are mostly explicable by historic references. Why strain credulity by explaining them as "kurgan elites"?

Well, Iranian people speak an Iranic language and only have 15% of R1a. Turkish people (from Turkey) don't have a majority of "Turkic" haplogroup too. The Greeks have low R1b and R1a frequencies.You gave the example of the Ossetians.

As you see, you don't need to have a massive influx of haplogroup to impose a language

[spongetaro]

I'm not going to try to argue with ten guys at once. Please use the Kurgan Theory to explain how R1a managed to spread Indo-European to the Atlantic coast. How does that "wash"? Not well.

R1a spread Iranic to Southern Iran. There are only 15% of R1a in Iran. Half of it could be attributed to Turkish invasions.
The Persian genetic legacy might no exceed 10% of the Iran population.

Same for Turkey. only 16% of R1b and 7,5% of R1a (including Turkic R1b and R1a) yet Anatolia was fully Indo European speaking until the Turkic invasion.

the 8,5% of R1a in Northern Spain might be alone responsible for the spread of Celtic in Central Spain.

[rms2]

The problem is this: the eastern/western distinction that you see in Indo-European doesn't really exist in that way, especially if you consider language families like Germanic, Greek and Tocharian which all criss-cross the "eastern/western" distinction. My opinion is that any origin of the Indo-European languages in Anatolia requires that Proto-Indo-European is not a language of the Chalcolithic but of the Neolithic, which is a quite futile thing to argue in my opinion if you consider that there are common words for metals and metal-working in PIE.

The centum/satem division does in fact exist, and unless you are prepared to come forth and announce what it is that makes you an expert and gives you the power to abrogate it, it is what it is.

Besides that, there is the mere geographical fact that Indo-European languages are overwhelmingly spoken in the West, where R1a is scarce.

I don't have all the answers. I am amazed when I run into so many folks who think they do.

The early Anatolian IE languages are especially archaic, as Anthony indicates in his book. If Euphratic in fact exists, that pushes Indo-European in that region still farther back in time.

EDIT: I would be also *highly* sceptical of the paper, because there no evidence for sound correspondence in the purported loans given in the paper.

Great. Hmmm. You are a guy who posts at Eupedia's Y-DNA forum. Gordon Whittaker is a linguist and professor at the University of Göttingen.

Your skepticism of Whittaker's work is duly noted.

Basque is clearly not a Paleolithic language, no, but this is exactly my point. The "native" (non-IE loanwords) vocabulary is that of a language of the Copper Age, which includes agricultural terms, words for domesticated animals and most importantly also metals and metal-working. The only archaeological culture that matches such a "package" is the Beaker-Bell Culture. If you say that Beaker-Bell was already Indo-European, you have a huge problem explaining where the Basques go their metal terms from.

Note that I am not necessarily saying that Basque was the language of Beaker-Bell (there is also the Iberians to be considered), but I think that a strong case can be made that Beaker-Bell was a non-Indo-European culture.

EDIT: What should be also added is, as the late vascologist R.L. Trask pointed out, that the number of Celtic loanwords into Basque is extremely few. It is not what we would expect if the Basques were surrounded for thousands of years by Celts or Indo-Europeans.

But the Basques clearly have been surrounded for millennia by Celts and other Indo-Europeans! That much is obvious. Are you attempting to deny that?

If Basque marital tradition was matrilocal, which I believe it was, then you have the perfect scenario for the introduction of outsider y-dna and the retention of the maternal language, since, in a matrilocal society, the groom goes to live with the bride's family. The male children would carry their father's y-dna but would speak their mother's language.

A similar scenario must surely explain how the Ossetians have become predominantly G2a over the centuries.

I think it likely the Beaker Folk spoke an early form of Celtic. How that happened exactly I cannot explain, but I don't feel the need for complete and tidy explanations of all facets of a phenomenon.

The Kurgan Theory has plenty of problems. All other explanations of the spread of Indo-European have their problems, too.

But I definitely do not accept the "R1a is ultimately responsible for all Indo-European languages" idea. It just doesn't make sense to me.

[rms2]

R1a spread Iranic to Southern Iran. There are only 15% of R1a in Iran. Half of it could be attributed to Turkish invasions.
The Persian genetic legacy might no exceed 10% of the Iran population.

Same for Turkey. only 16% of R1b and 7,5% of R1a (including Turkic R1b and R1a) yet Anatolia was fully Indo European speaking until the Turkic invasion.

the 8,5% of R1a in Northern Spain might be alone responsible for the spread of Celtic in Central Spain.

What 8.5% R1a in northern Spain? Where are you getting such a figure?

Armenia's largest y haplogroup is R1b, and R1b is more frequent in Anatolia than is R1a.

Besides that, your arguments can be distilled down to this: if there is even the slightest hint of R1a anywhere, it is responsible for whatever version or versions of Indo-European are spoken in that region. You produce no real evidence, you simply assert that as the case.

Who can argue with that? Who needs to?

[spongetaro]

What 8.5% R1a in northern Spain? Where are you getting such a figure?

Armenia's largest y haplogroup is R1b, and R1b is more frequent in Anatolia than is R1a.

Besides that, your arguments can be distilled down to this: if there is even the slightest hint of R1a anywhere, it is responsible for whatever version or versions of Indo-European are spoken in that region. You produce no real evidence, you simply assert that as the case.

Who can argue with that? Who needs to?

That was Cantabria! Sorry I should have precised.
Turkey have only 16% of R1b. A big part of it must be of the R1b-M173 and R1b L23 subclades whose the association with IE is doubtful

[spongetaro]

You produce no real evidence, you simply assert that as the case.

Who can argue with that? Who needs to?

LOL you're an amateur just like me and everyone in this forum. No one has "evidences" until we have ancient Y dna from Proto Celtic people.

BTW the Urnfield ancient DNA from Germany showed more R1a than R1b.


I just tried to show that little amount of R1a can be responsible for the spread of an IE language.

[rms2]

The Classical 19th/20th century of idea of the origin of the Celts in the cultures of Hallstatt and La-Tene comes from the correlation between archaeological culture of the Gauls and the Celtic languages. That these people of Hallstatt and La-Tene spoke a Celtic language is beyond doubt (it is a favourite question I often ask, if Hallstatt was not Celtic, what was it then? Germanic? Etruscan? Slavic? none of this makes any sense).

I was not attempting to deny that Hallstatt and La Tene were Celtic, just the notion that their people were necessarily the original Celts and spread Celtic languages by means of Iron Age invasions from a Central European Celtic Urheimat.

I think, a part of the "Celtic" problem is the detour that early 2000s genetics brought into the discussion. By saying that R1b was indigenous to Western Europe, and by saying that the large-scale genetic makeup of Western Europe was basically unchanged since the end of the last ice age except for a few minor additions, it was to argue that Indo-European languages had to spread without large-scale immigration, and that a spread from the culture of Hallstatt and La-Tene was without any basis.

Hallstatt and La Tene are too late, it seems to me, to explain the spread of the Celtic languages. So, I do think there is no basis to see them as the ultimate source of the Celtic languages.

There is no evidence of large-scale Hallstatt or La Tene invasions.

If we however acknowledge that R1b is not native to Western Europe, which it clearly isn't, we must argue for some kind of large scale population replacement, either through invasion or immigration, in Western Europe, and that point the idea of an invasion is not that far-fetched. Also, there is enough archaeology to argue for discontinuities in Western Europe, especially when and how the Beaker-Bell Culture came to an end.

I agree, but that is much earlier than Hallstatt and La Tene.

I mean, I know of the so-called "stelae people" hypothesis, but does it really make sense? Does it make sense for people from Crimea to migrate by sea all the way to western Portugal, without leaving much evidence, and then spreading the Proto-Celtic language from there across all of Western Europe? Does it really make sense to assume Proto-Celts or Proto-Indo-Europeans were in North Africa and Sardinia?

Why not? I don't know. It makes as much or sense as the idea that Indo-European was spread all the way to the Atlantic coast by R1a from the Pontic-Caspian steppes.