Talk:Paleolithic Continuity Theory

From Wikipedia, the free encyclopedia

I have edited this page of all crap wich have bias in decision makings.It is about theory and it is equal with the oponent theories.If someone want to point critics on the theory that must be made in seperate column-like rest of the articles on Wikipedia.I wish to notice that space used for and against theory must be equal. Edited by Admin. from "Wikipedia The Free Encyclopedia".

[edit] Pseudo-science or no science

What if the paradigm of a not-so-early common origin of the Indo-European languages is met with serious inconsistencies, some of which are shown, for example, by Colin Renfrew in his book on the puzzle of Indo-European origins? Are we left with the choice of not trying any research at all or of resorting to pseudoscience? The proposal of the PCT that the dialect boundaries of today Europe go back to Mesolithic times at least might eventually turn out to be wrong, but is argued with the linguistic and archeological competence of an expert of European dialects and of an expert of European prehistory. Guparra 20:37, 9 November 2005 (UTC)

This article is self-contradictory. First it says PCT is the theory that Indo-European languages have been in place since Paleolithic. Then to contradict the theory in the last paragraph there is a statement that goes something like since language changes are now known not to be due to migrations, PCT must be pseudoscientific. Balazs 20:14, 23 November 2005 (UTC)

you misunderstand. The article is making the point that language spread is not connected to noticeable genetic shifts (i.e. 5% of people may migrate, and cause a change in the language of the other 95%), therefore the argument of the 80% genetic consistency is worthless. They may argue with expert competence on European dialects, but if you ask me, if they apply that competence to the Paleolithic, the outcome is still pseudoscientific. dab (ᛏ) 20:56, 23 November 2005 (UTC)

People have to get used to the new theory. Alinei is not some pseudoscientist to make such elementary faults. PCT is VERY VERY deep theory.

Cheers

I wish to point out some facts on the article, that make me believe that the article should be rewritten.

1. Alinei's theory is not based on arcaeogenetical studies, much less on those of Brian Sykes, who is not even quoted in the two volumes of "Origini delle lingue d'Europa"

It is based on other, linguistic and archaeological points, which I try to resume.

• From internal evidence from the European languages and dialects that shows inconsistencies in the usually assumed opinion dialects are late, Middle Age, phenomena.
• From evidence that linguistic chnge is rather due to contact than to an internal mutation clock: for example, the differences between American English or Spanish and their European counterparts are to be asciribed to the languages spoken in the regions where most emigrants came from rather than to later evolution.
• From the well known fact that geographically and culturally isolated isolated languages are conservative.
• From these, and many other facts, Alinei formulates a principle whereby languages change only if compelled to do so, because of contacts, but in absence of contacts they tend to be stable.
• So the distribution of languages can be much older than believed.
• Internal evidence shows that the common core of Indo-European language shows an Upper Paleolithic culture. Words that belong to later cultural ages are innovations coined out of a common stock, but innovations nevertheless. Therefore, common Indo-European is much older than believed.
• The borders of Mesolithic cultures are more or less well reproduced by the borders of modern European dialects.
• Genetics. The results of Cavalli Sforza and collaborators, of Sokal and others (not of Sykes) show that linguistic borders grossly correspond to "genetic borders". They are of course taken into account, as evidence to be added to the linguistic and archaeological evidence. But the PCT does not relay upon genetics.

1. If a theory about the origins of a linguistic group is deemed to be not scientific if it cannot be tested against records, no theory of the origins of IndoEuropeans, or Uralic peoples, or whatever else, is scientific. We can only maintain that Greek was spoken in Greece about 1200 BC, or German in parts of Germany about 800 AC. If instead we allow for the use of internal linguistic evidence and archaeology, used with the proper techniques of linguistics, Alinei theory is scientific. In addition, it explains facts otherwise unexplained, which is an excellent test. Of course, it might be wrong, in the same way that many other theories of hystorical linguistics and anthroploogy have turned out to be.

Oddly enough, it is the second instance in which I find criticisms of Alinei's theory motivated with statements that he did not issue. Here genetics being the basis of his theory, the other instance is to be found in the discussion of Etruscan language. Apparently, his theory is too new to be accepted easily. A review in English (his books are in Italian for now) is that og Jonathan Morris in Mothe Tongue [1]. Guparra 20:45, 28 November 2005 (UTC)

An amendment. I found Alinei's reference to Sykes' work. It is in the introduction to the Continuity Theory in the website {http://www.continuitas.com], which is linked to in the article. It is the website of the group of linguists and prehistorians who support the PCT.

• The work of Sykes quoted is printed in 2001, while the two books by Alinei were printed in 1996 and 2000.
• For sure, Sykes' work is relevant to Alinei because it gives him further arguments to criticize the Neolithic Farmers Wave Immigration, modelled by Ammermann and Cavalli Sforza and used by Renfrew. Renfrew hypothesizes that the IndoEuropeans were Neolithic farmers coming from the Levant who settled in Europe and by demic prevalence absorbed the scarce Mesolithic native population.
• It seems to me that the introduction to the PCT provided in the website, should make apparent that the theory relies on linguistics and archaeology, genetics being concurrent for some points only. Guparra 16:05, 29 November 2005 (UTC)

well, by all means fix the article then. It is still just a stub hacked together from online sources, so somebody who has actually read the book is sorely needed. dab (ᛏ) 18:22, 29 November 2005 (UTC)

what I shall gladly do. I only need some time because I am in the process of learning how to use wikiediting and, most important, I have to go through the books again. Guparra 20:38, 29 November 2005 (UTC)

I understand that language change is not neccessarily connected to genetic shifts (that is 5% genetic shift can change the language of the other 95%). But does it neccesarily contradict (provided that there is other evidence, which seems to be the case)? Alinei's website talks about another better established continuity theory (the Uralic one). There is also the issue of what is meant by pseudo-science vs. science (Feyerabend thought that neither is there a distinction nor would it be advantageous to have one). Balazs 16:31, 1 December 2005 (UTC)

I apologize that I shall be incommunicado for a few months. A comment about your comment. In Physics the border between science and pseudo-science can be very thin. Am example is the Superstring theory, which any physicist considers to be Science; nevertheless, we do not know when nor whether it can be actually tested, which would put it in the pseudo-science realm. Guparra 07:41, 22 December 2005 (UTC)

Alinei's judgements on Uralic are entirely untrustworthy. There is no established continuity theory for Uralic. Wik's belief in a Uralic substrate, for instance, have been dismissed by most Uralicists. The people whom Alinei generally cites whenever he mentions Uralic languages are a couple of archaeologists without meaningful training in historical linguistics. Alinei is a very deceitful writer and you should follow his footnotes sometimes to see how much he strays from reputable linguistics. CRCulver 19:00, 25 January 2006 (UTC)

I would appreciate some references in any language that suggest that Alinei is a quack (which is roughly what you are saying). My limited understand is quite the opposite. Alinei appears to be a respected scientist with considerable accomplishments and acknowledgement behind him. Doubtless his continuity theory is contreversial... but I am yet to see any evidence beyond random anonymous english-speaking internet strangers that he is considered a disreputable or intellectually dishonest man by any group of people or scientific organisation. --70.49.192.75 00:31, 29 March 2006 (UTC)

The greeks (judging by the wikipedia page) do not seem to consider the theory quackery either... Paleolithic Continuity Theory, translated from Greek --70.49.192.75 00:41, 29 March 2006 (UTC)

And what is that link supposed to indicate? All it shows is what one can already infer from other internet sites, etc.---that there is a portion of the Greek population that is ready to swallow up any jargon that may lend credence to the idea of proto-Greeks being autochthonous to Greece (this is not directly connected to the issue of ancient Greek->Modern Greek continuity, but a very different topic), which is a relatively popular idea among them, I have noticed. Alinei is still very deep in the fringe. He has all kinds of whacked-out ideas, like Etruscan being an early form of Hungarian; that the Slavic invasion never occured; that the ancient Thracians were Slavs (he actually states this). If those Greeks you cite realized that, I think they would think differently of Alinei. Alexander 007 22:51, 7 April 2006 (UTC)

The fact that you even resort to making such a pathetically puerile generalization as "the greeks (judging by the wikipedia page) do not seem to consider the theory quackery" indicates that your mental level is at the 12 year old stage, which explains your taking to Alinei in the first place. Alexander 007 23:04, 7 April 2006 (UTC)

I am surprised how easily a scientific theory can be branded as "pseudoscience" only because its results are in contrast with the mainstream. And without even having read the book! Linguistics is not my field of expertise, but I think I can discern science from pseudoscience, and Alineis method is not any more speculative than other linguistic studies, and his reasoning is sound. I would hence like to eliminate the sentence about the pseudoscientific character of Alineis theory, unless there are objections, or a reference can be provided. Pcassitti 12:51, 31 December 2006 (UTC)

I would object. PCT is generally not submitted to the same level of peer review as reputable theories, nor is it published in reputable venues. Indeed, much PCT "scholarship" is self-published. CRCulver 23:42, 9 January 2007 (UTC)

The lack of translation into english is indeed a problem, but papers related to the continuity theory have been published in a number of scientific magazines, and presented at international linguistic conferences. I fail to see any element which would objectively put this theory close to the "realm of pseudoscience", as the wikipedia article states, apart from the fact that the results are in contrast with established linguistic theories. But it is the method, not the result which defines science.

But apart from that, I think that branding something as "pseudoscience" in an encyclopedia needs at least a reference to substantiate the claim. Pcassitti 21:13, 12 January 2007 (UTC)

[edit] status of PCT

thanks for the 'review' link (more properly 'self-promotion' I suppose, since it is hosted on their own site and does not appear to have been published elsewhere (?)). Looking it over, I am now quite convinced that PCT can be dismissed as fringy nonsense. It appears to propose linguistic change with geological slowness :) no matter what your take on glottochronology (error margin of 50% or 200%?), I don't think any self-respecting historical linguist would endorse anything like this: Renfrew's timeframe is already borderline acceptable, but this is completely bat-shit beyond the pale. PCT appears, after all, to be the European answer to "Paleolithic Aryan" nonsense in India. It is at least reassuring to see that crackpottery knows no boundaries :) dab (ᛏ) 09:35, 14 August 2006 (UTC)

How do you derive the age of a proto language? How old is spoken language and how is the date arrived at? Is there any reading material available for non-linguist? --UB 10:59, 23 August 2006 (UTC)

try (the references at) historical linguistics and glottochronology. The accuracy of such estimates depends decisively on the age of the earliest sources. For PIE in particular, see Proto-Indo-European language and Proto-Indo-Europeans. The error margin is frequently admitted to be as high as 100% (i.e. a factor of 2). For PIE, dates between 8000 BC and 2500 BC are possible (10000-4500 BP, i.e. a factor of 2.2): 8000 BC is extremely early and 2500 is extremely late, most people will agree that a 6000-3000 BC range (factor of 1.6) still has a very high confidence. All we know with dead certainty is that the proto-language must have split up by 2000 BC, since our earliest text fragments date to shortly thereafter. Claiming paleolithic age of PIE simply amounts to rejecting wholesale all efforts at dating language change and taking an agnostic position of "prove that it isn't paleolithic". It would entail that languages stayed essentially unchanged for at least 10,000 years, over vast areas of Eurasia. All known language histories show that a language usually changes beyond comprehensibility (meaning it doesn't just 'change', it becomes a wholly different language) over 1,000 years, in rare cases of stability maybe over 2,000 years. Note that in this case, evidence for dating is not restricted to pure glottochronology. For example, since there is a very good reconstruction of PIE terms for "wheel", it seems evident that (late) PIE must post-date the invention of the wheel in around 4500 BC. The evidence for "metal" (Bronze) is less clear, it is possible that some branches had already separated before Bronze became known (after around 3300 BC): these dates dovetail perfectly with a 5000-3500 range of early to late PIE fully consistent with the (wider) glottochronological estimate. dab (ᛏ) 11:35, 23 August 2006 (UTC)

Nice to see some English-language interest and my article cited. If I could add a number of comments, as someone who has had some discussions with Alinei:

-It must be remembered that he is a distinguished academic at the end of his life (he's over 80), so age is a major factor in explaining why he has failed to discuss with certain key areas of IE (like Iberia, Persia, India). It also explains why he finds it difficult to get to grips with the intricacies of mtDNA. - At the same time, I think that the advances so far in mt/YDNA tend to bear him out. If you read Sykes book (good but lamentable for its lack of bibliography), he describes very clearly that Cavalli-Sforza violently opposed mtDNA and then seeing he was defeated, decided to start supporting it and claim the idea as his own. This marks a major change in favour of the PCT, in that if the mesolithic hunter gatherers were a tiny majority overwhelmed by a massive influx of farmers, the idea of IE speakers in Europe prior to the Neolithic would have been hard to believe. The consensus in genetics is now fairly solid that 80% of the population is pre-farming and if you study the models of diffusion advanced e.g. by Zvelebil, then you come to the clear conclusion that it was very much a piecemeal process. Hence, as Alinei points out, Renfrew has a real problem in explaining why there's no substrate in the last areas to be neolithicised e.g. Norway, why there's a long-standing linguistic boundary in N Latvia (i.e. why don't the farmers manage to impose IE on the "Estonians", etc. Furthermore, the theory is actually starting to creep in via the back door - a specific prediction of PCT is the presence of Germanic speakers in Neolithic Britain, and I see that Stephen Oppenheimer has mentioned this in his new book (unfortunately without citation). What you have to remember is that the world is Anglophone, IE studies is a sleepy field, so that anyone writing in a language other than English gets no "air time", with the possible exception of the Russians. There are some Spanish linguists doing excellent work, notably Francisco Vilar who has shown that the oldest toponyms even in Andalusia are IE - but because he doesn't write in English, no-one is even aware of his work. For those people interested in PCT, the figure to watch, and the "heir" to Alinei seems to be Xaverio Ballester. - Secondly, Alinei has a problem in that his method of linguistic archaeology only really works where you have peoples with defined territories, hence you have a paradox of someone proposing conjectures about the languages spoken during the Palaeolithic with a methodology which only really works from the Mesolithic onwards. As a result, when discussing pre-LGM, he tends to rely on other people's ideas and frankly hasn't chosen very wisely, appearing to be bogged down in a tool making equals syntactic structure equation which leads him to view Chinese as a kind of ur-language. This is the old Schlegelian bear-trap of the Monosyllabic/Agglutinative/Inflectional classification which captivated mid-19th century figures such as Haeckel and Schleicher, but had already been dismissed by e.g. Trombetti/Jespersen/Saussure in the early 20th century who realised that Chinese was the result of a long-process of simplifying an inflected language (see Classical Tibetan). Alinei seems to be obsessed by the stability of lithics in E Asia since Homo Erectus and imho is assuming without foundation that the original inhabitants of S China were Sino-Tibetan speakers. People who want to dismiss him seize on this older stuff and his claims that IE had differentiated 100,000 years ago. Indeed, the response to my Mother Tongue article was that most of the readers are interested in deep prehistory and Asia, so they assumed that what is actually a fairly marginal part of Alinei's work was the main part and dismissed all his extremely detailed linguistic archaeology relating to the mesolithic and neolithic.

In other words, I think that Renfrew and Gimbutas theories don't stand up at all, but if you modify PCT to take into account modern advances in genetics, you actually come up with a plausible theory.

Also: - I am not aware of Alinei ever suggesting that the PCT applied to India. I asked him about this and his comment was that he wasn't a Sanskritist and someone else should take up the torch. The PCT is purely about whether or not IE languages had differentiated and spread into Europe by the end of the ice age. - The comment above that the Thracians were Slavs is entirely inaccurate and I refer the person in question to pp. 222-223 of vol. 2 of Origini. What he actually says is that he thinks that Herodotus probably used the term 'Thracians' as a blanket term to refer to Slavs. He actually regards it as a third differentiated branch of a proto-Balto-Slavic family subject to influence from an Altaic elite. (20:39, 16 January 2007 (UTC)) - Jonathan Morris.

Hello Jonathan. It is nice to see someone with deeper knowledge of the subject. May i request you to please clear a few doubts about PCT to me.

1. When did PIE separated, is it 20000BCE.
2. What age is attested to Hittite language and Vedic Sanskrit.
3. According to PCT, where did PIE originated. (Is it Africa or PCT doesnt care about that.) nids(♂) 12:06, 17 January 2007 (UTC)

Jonathan, I am afraid all you really manage to convince me of is that the theory has indeed no merit at all. All the genetics points are granted, and you present a reasonable outline of a discussion of Meso- and Neolithic migration. This, however, has nothing whatsoever to do with Indo-European. To develop a theory of "Paleolithic PIE" and ignore Sanskrit is ludicrous. (well, Paleolithic PIE is ludicrous in any case, but to confine it do a discussion of European genetics and Stone Age archaeology is simply pointless). Language is a part of culture, and is not passed on genetically. Genetic analysis may serve as a tool for tracing populations that may have been the vehicle of linguistic spread, but the tacit assumption that a population needs to be replaced for the language to be replaced is just silly. What proportion of "Portuguese Genes" will you find in Brazil, or what proportion of "English" genes will you find in the US? And regarding glottochronology, the language changes observed over the past 1,000 years make clear that it must be inexact, but "inexact" here meaning to an error of maybe 200%, not 1000% (50,000 years as opposed to 5,000). And, to cut this discussion short, if PIE had diverged before the neolithic, why can we reconstruct the PIE terms for "wheel" or "metal"? The fact that genetically, "80% is pre-farming" (in India as well as Europe) only goes to show that a Bronze Age expansion is as good as a Neolithic expansion to account for the imposition of a new language, that is, the fact that the Neolithic migration wasn't so massive takes away Renfrew's main argument as to why PIE expansion cannot date to the Bronze Age. The proposition seems to be that the Paleolithic hunters essentially spoke PIE in 40,000 BC and that their language remained frozen until after 10,000 BCE until virtually all of Eurasia spoke pure PIE, before history and language change kick in for some reason in the Early Bronze Age. This doesn't strike me as a reasonable scenario. dab (𒁳) 13:44, 17 January 2007 (UTC)

Hi, everyone, unfortunately I have to go travelling but will sit down with Alinei's books next week and give you some answers. (23:04, 17 January 2007 (UTC)) Dab, you must be a professional linguist to assume a) that Alinei is automatically and comprehensively wrong on everything and b) that I'm a travelling salesman for the gospel of PCT. My motivation for writing the article was just to present an account of what seemed to me to be an interesting and radical theory but which wasn't available in the English-speaking world. Firstly, I'm not part of the PCT work group, and secondly while small, it's already very heterogeneous, including figures who are very highly regarded outside the PCT milieu (like Marcel Otte) and people who, from my conversations with academics in the field, don't appear to be highly regarded at all (e.g. Henry Harpending-and again, he's in there as a member but I'm not aware of him having made any specific pronouncements on the PCT itself). As such, you could say that there are various PCTs - Alinei has his, Marcel Otte has his paper on IE spreading from European glacial refuges - but the common axiom is that some form of IE was present in Europe by the start of the Mesolithic - which is evidently the major difference from Renfrew & the Classical/Gimbutas theories which insist on the notion that Europe was entirely free of IE speakers until at least the Neolithic. Anyway, more next week. (Jonathan Morris 23:29, 17 January 2007 (UTC))

OK, I’m back, sorry for the delay, but I have a business to run.

Nids: As you may be aware, essentially everything Alinei has to say on IE PCT is concentrated in a 2-volume work – the first one came out in 1996 and is more of a theoretical treatise. The second which came out in 2000 is the meaty linguistic archaeology tome (Continuity from the Mesolithic to the Bronze Age).

All in all, Vol. 2 is a solid piece of work which contrasts with a rather vague and fuzzy Vol. 1. Almost all of his views on the Palaeolithic are contained in Vol. 1 and in my view, many points require serious revision to account for the new advances in Genetics since then.

As I mentioned, Alinei devotes a great deal of space to lithics – which you might expect him to do, as you don’t have much else to go on pre-Aurignacian and he allows himself to be convinced of Matthew Dyer’s model which associates different kinds of toolmaking with different kinds of language, hence he argues the apparently enormous stability of E Asian lithics from Homo Erectus to the end of the ice age is correlated with the presence of monosyllabic languages there. In other words, if you buy this, you have to support the multi-regional hypothesis, which you could probably just about have done plausibly in 1996, but I think it’s now dead in the water since the consensus is for Out of Africa.

Now in my view, this doesn’t preclude H. Erectus or the Neanderthals from having language, it just means that if and when these earlier hominids came face to face with H. Sapiens Sapiens, neither would have understood a word of the other’s langauge due to hundreds of thousands of years of separation.

Returning to Alinei, he is aware of the Out of Africa vs. Multi-Regional Debate, so what he says, prudently, is basically, I really like my Schlegelian lithics-language model and that sort of tilts me in favour of the MRH but I’m aware that the Out-of-Africa supporters also have a good case, so I’m going to put forward 2 alternative versions, a long-run PCT and a short-run PCT and let time decide which is right, and in any case, if you’re only interested in the last 50,000 years, it doesn’t really matter which you choose because the end-result is the same.

In terms of dating, even for his short-run PCT, he appears to believe (and I don’t see an explicit statement), that the Nostratic phase was relatively short-lived and that PIE existed as an explicit entity in SW Asia as early as 80-100 kya, with Hittite splitting off at an early date (he explicitly agrees with Gamkrelidze/Ivanov here that the deep split in PIE is between Anatolian and non-Anatolian, and other PIE the remainder forms a Sprachbund which gradually crystallises into different families from then onwards.

This was his stance in 1996. I asked him about this a couple of years ago, and he seems to be prepared to bring his dates down but I haven’t seen him in print on this. His main argument is nevertheless that all the IE groups had differentiated from each other by the start of the Mesolithic (e.g there was already a discernable Germanic grouping different from a Celtic grouping) – based, as I cited in my article, on the fact that certain late Palaeolithic cultural innovations like burial have different words in different languages.

I would say that I support his end-point conclusion, i.e. that some form of IE was present in Europe by the Mesolithic, but evidently not his dates for the differentiation of PIE, which are way too high.

Dab, it’s you’re privilege not be convinced, but frankly I find your arguments for dismissing PCT to be seriously flawed.

Firstly, you cite examples like Brazil (actually Jamaica would be a much better example) where you have a genetically African population but they all speak English. Ergo, no link between genetics and language. But this blithely assumes that the social and demographic dynamics of the 17th-18th century can be extrapolated directly to the societies Neolithic – which seems completely ludicrous to me. On the one hand you have a highly hierarchical and sophisticated set of nation states capable of organising an intercontinental slave traffic and on the other a bunch of half-starved stone age farmers about whose models of social organisation (e.g. exogamy), we know virtually nothing. But you (and you’re by no means alone here) blithely equate the two and claim that this constitutes a plausible core assumption merely because it allows you to reject an idea which doesn’t agree with yours. Any half-serious historian would just laugh at you.

Secondly, it’s interesting that you choose Brazil as an example. You may be aware of work by Francisco Salzano who took a sample of men from the Northeast who considered themselves to be “white” [although Brazil has some very dark people who think of themselves as white]. He found that over 90% of them had European Y-chromosome lines, but 60% had African/native Indian mtDNA. I.e. the model of Portuguese men miscegenating with Indian women/slaves really does show up in the genetics. Furthermore, a similar pattern shows up at 1200 years in Iceland, where most of the women are shown to be of Celtic origin.

I suggest that divergent YDNA and mtDNA patterns are a genetic signature for miscegenation, and the bottom line is that there’s no evidence of this divergence in the European gene pool, which combined with the fact that we now know that “old” genetic lines predominate really does stack the deck against a Bronze age elite dominance model à la Gimbutas.

If you’re going to argue from the known past to unknown prehistory, then I suggest that what the last two thousand years shows is that unless an intrusive people settles in sufficient numbers to dominate an area’s economic and social infrastructure, their language tends to disappear without trace. The Romans achieved this by settling ex-legionaries, co-opting the local élite and probably massive displacement of slaves to latifundia, the Germanic tribes who didn’t do this singularly failed to impose their language on a single area which they held in Continental Europe, and in England, note that everyone is surreptitiously starting to follow Alinei and argue for a much older Germanic presence in Eastern England, simply because the genetics won’t support the view that the island was entirely Celtic before the 3rd-4th century. Even where invaders took over the country, this was not enough to ensure the triumph of their language in the long term (e.g. Arabic in Spain or Norman French in England). You could thus argue from this that the only people who managed to impose their language succcessfully were the Romans and this is precisely because they were as good as the English/Spanish/Portuguese in the 17th-18th centuries at shipping people around in the name of a grand economic design. This says to me that it’s actually pretty difficult for an elite to change the language of an indigenous people. However, the Renfrews and Gimbutas simply assume that what recorded history shows was damned difficult to achieve over the last 2,000 years with plenty of examples of plagues and marauding horsemen raping and killing sedentary farmers, was dead easy in the Neolithic/Bronze Age. But they advance no social models to explain this, there is no archaeological evidence for the rape and pillage and Renfrew has produced precisely zero linguistic evidence for his farming model (and while he’s evidently not a linguist, he has a lot of clout in academia and a big cheque book, so if it was such a great theory, he could presumably have persuaded at least one respectable linguist to provide a helping hand – and yet, he has to admit (p. 474 of Examining the farming/language dispersal hypothesis), that not only has he failed to show a connection between the spread of farming and IE, he hasn’t managed it for any language group anywhere in the world.

Because of the above, I suggest, at the risk of repeating myself, that if these models were true, you would have some genetic evidence of miscegenation. And it ain’t there.

This, for me, is a cardinal virtue of PCT, it just says that the first people into a territory tend to dictate its language, that once there, they tend to stay there, and there wasn’t even the possibility of outsiders coming in and kicking them around until the Bronze Age. It has a very simple mechanism for explaining how a given language comes to be spoken in a given area (My own model is slightly different but I’m still working on it) whereas Renfrew & Gimbutas have advanced no mechanism.

Thirdly, your claim that you can have language change without full population change is just a non-sequitur. Both Renfrew and Gimbutas claim that IE speakers intrude into an area previously occupied by non-IE speakers, and somehow, whether by the seductions of farming or force of arms, within a few generations, all traces of the non-IE language have been obliterated. But where is the causal link between your claim that this process of language extermination is possible without ethnic cleansing and the proof that this is actually what happened? You seem to be labouring under the misapprehension that proponents of the PCT are saying “we reject Renfrew and Gimbutas because we believe that their models only work with comprehensive ethnic cleansing and we don’t think that this could have happened”. No one is saying anything of the sort. Alinei is actually saying (and I agree with him on this point) that Renfrew and Gimbutas need to prove that there actually was a non-IE substrate for their theories to stand up and they can’t prove it a) because there are linguistic boundaries which are older than their invasions and b) particularly for Renfrew, there are fringe areas of Europe like the Norwegian coast where neolithicisation was very late or never happened, so according to Renfrew, these should be hold outs of the non-IE substrate peoples. The day you find non-IE fishing vocabulary in Norwegian dialect is the day that the PCT collapses, but Alinei’s point is that it’s not there, and furthermore, he’s not the first person to realise this – the absence of non-IE substrates in N Europe was perfectly clear to linguists the late 19th century. Let me further say that just as Europe fails the substrate test, India passes it. You have abundant evidence of substrate languages in the Rg-Veda (Munda, Dravidian, some unknown language) and only one real branch of Indo-European (as opposed to many in Europe). I’m currently talking about some other stuff with the Mother Tongue people which may show this more conclusively, but I want to point out that there are areas like India where the evidence does appear to support the opposite conclusion (note that Alinei has never suggested that PCT-IE applies to India, only to Europe – although he does believe in a PCT for Uralic and Altaic).

You also appear to have overlooked the fact that when you sever the link between language and genetics, you forge a double-edged sword. Hence I can turn your argument on its head and claim that you don’t need the survival of indigenous speakers to keep traces of a substrate alive. Brazil is again case in point. I’ve lived there on and off for 20 years and never met a native Tupi speaker, but Brazilian Portuguese is full of Tupi words for animals, placenames, personal names, etc. Until the Portuguese court cracked down at the start of the 19th century, it was the lingua franca everywhere in Brazil except the coastal cities (as its relative Guarani still is in Paraguay). Look at all colonies settled by Europeans (USA, Australia, Mexico, etc.) and you’ll find the same pattern of survival of indigenous languages, if only in place names, despite massive differences in technology, military force between the original inhabitants and the European intruders. Once again, however, Renfrew and Gimbutas insist on their intrusion theories without providing any evidence of a substrate [one person has tried to reconstruct a proto-nonIE langauge for Europe, Harald Sverdrup, but his work is so shoddy, particularly for his Basque-Etruscan cognates, that it really doesn’t stand up – if anything the one part which looks OK – his showing of links between Etruscan and Rhaetian, actually coincides with Alinei’s Hungarian model]. The PCT’s claim that the absence of a non-IE substrate indicates that IE was the original language family, seems to me to be absolutely logical.

As for glottochronology, the methods of calibration are so full of holes as to make the dates worthless. You can check out my post on Jess Tauber’s Amerind group if you like, but basically, the rate of divergence for modern languages is greatly exaggerated, and there are variants of the method (e.g. Starostin) which give deep dates (even if he rejects his own findings). There’s more to be said here, but it requires a full paper.

Finally, why don’t you argue that PIE broke up in the 20th century, since all the daughter languages probably have the same word for ‘computer’, etc. On what grounds can you claim that a cognate for a piece of technology isn’t a generalised borrowing and must be part of the original PIE vocabulary? Particularly when it’s probably a question of loans between related languages.

- Jonathan Morris.

I do not believe the lack of translations into English are a proof of anything (much valueable material is not yet translated into English and much valueable scholarship probably lies unknown to the English speaking audience).

But reading some of Alinei's materials on PCT I believe (as a non-professional) his theory is likely an artificial construct trying to answer some questions but raising many others. E.g. one obvious problem to his theory is that the linguistical map one gets for Balkan area in Antiquity doesn't match at all the Middle Ages map or the modern map. From Greek, Celtic, Illyrian, Thracian, Iranian we end up with Greek, Slavic, Hungarian, Eastern Romance, Albanian, Turkic (the lists are approximative). Therefore some additional equations must be built: Hungarian = Etruscan, Slavic = Thracian. And here I observe a laitmotif: the omission.

* Some ethno-linguistical realities are ignored - what happened to Celtic language in Balkans, for instance?

* An unfair perspective is given within the IE taxonomy - he notes Thracian, Baltic and Slavic are satem languages and based mostly on that he draws them together; but what about Indo-Iranian languages?

* The relevant scholarship is also missing (for Thracian language Dečev, Georgiev, Russu, Duridanov, etc.) while his position on Thracian = Slavic I find extremely thin and rather rhetoric than argumentative (from some hundreds of known Thracian words, names, roots he barely touches two on some unpersuading similarities: e.g. Thracian "diza" has much better parallels in other languages like Avestan).

* Alternative hypotheses - why the similarities he notices are not caused by common IE origins or neighbourhood?

I cannot say this is pseudo-science, but without a serious peer-review and with such shaky arguments, I don't find it trustworthy either. Daizus 13:28, 24 February 2007 (UTC)

Daizus and anyone else. You're guilty of misreading Alinei. Admittedly, his treatment of Thracian is cursory, but this is not a reason to misquote him. As you know, a key belief of PCT is that it is only in the Chalcolithic that societies advance to the point of permitting stratification/inter-ethnic dominance. Hence his model is of a Europe consisting of differentiated groups at the end of the Neolithic, with the intrusion of elites speaking other languages from the Chalcolithic/BE onwards. Evidently, if the elite is displaced, its language disappears and the language of the peasantry re-emerges. Outside their home area (W Europe and notably France) he sees the Celts as an elite - so they simply come and go. He also sees an amorphous mass in E Europe which for want of a better word, we'll call Balto-Slavic, which differentiates into an archaic periphery (Thracian and Baltic) and an innovative centre (Slavic). On p. 193, he states that this explains the affinities between Baltic and Thracian toponyms, noted by Trubacev.

On p222-223, it appears that no-one has read the page in full, which concludes as follows:

• In termini più precisi, dunque, si può ipotizzare che il Tracio fosse una lingua di transizione fra Baltico e Slavo, parlata da un gruppo periferico della Slavia meridionale, e pendant dei gruppi baltici della periferia settentrionale. A differenza dei gruppi baltici, questa geovariante, particolarmente soggetta alle vicissitudini dei gruppi elitari altaici, sarebbe stata riassorbita nel 'mare slavo' e si sarebbe estinta'.

Tr: In more precise terms, therefore, it may be hypothesised that Thracian was a transition language between Baltic and Slavic, spoken by a peripheral group of Southern Slavia, the counterpart of Baltic groups on the Northern periphery. Unlike the Baltic groups, this geovariant, particularly subject to the vicissitudes of elite Altaic groups would have been reabsorbed into the "slav sea" and would have become extinct.

This is not the same as saying the Thracians were Slavs, as my previous comment on Herodotus points out. Alinei does not assert this and merely tries to trace the Thracians to pre-Slavic cultures of the Neolithic. He then says that the Thracians were 'militarised'by their Altaic contacts, and at a later stage, established some transient hegemony over neighbouring Slavic peasantries, hence Slavs were mistaken by the classical historians for Thracians, where in fact they were under the rule of the latter.

Frankly, I find this idea that from the Bronze Age onwards different social classes occupying the same territory spoke different languages is far more sophisticated than the analysis of his critics.

While we're on the subject of the Balkans, I will say that I'm much less convinced by his theory of proto-Romance in the Balkans which could have given rise to Romanian. If this were true, then I think you'd see far more divergence between Romanian and classical Latin than between the latter and say Portuguese, and this is not the case. Indeed, I think that he underestimates just how effectively Trajan and his legions ethnically cleansed Dacia, although this is a very special situation.(Jonathan Morris2 13:18, 26 February 2007 (UTC))

PS His classification of Thracian with Balto-Slavic is not based on the fact that they all just happen to be satem languages, but on archaeological evidence and the similarities between Baltic and Thracian toponyms noted by Trubacev. Furthermore, I see nothing in the above which conflicts with Duridanov's view that "it turned out that the Thracian language is in close genetic links with the Baltic languages". This discussion would be far more productive if people posted on the basis of what Alinei actually said and not what they hope he might have said so that they can disagree with him. (Jonathan Morris2 13:59, 26 February 2007 (UTC))

I am not speaking generally of PCT. I understand Alinei is proposing a model, I'm fine with that, yet when I'm trying to apply his model to historical realities I am more familiar with, I can't see it validated by them as it should be. Hence my characterization: "artificial construct". If his model could explain the linguistic maps of Balkans as they evolved from the ancient times until today, I wouldn't have used this characterization. I haven't read his entire work, just some materials available online. So I'll just stick to what I've read and the view I can get from that. From "Interdisciplinary and linguistic evidence for Palaeolithic continuity of Indo-European, Uralic and Altaic populations in Eurasia, with an excursus on Slavic ethnogenesis" I've focused in my earlier reply on section 7.5.5, particularily on the page 37 where Alinei draws together Thracian and Slavic languages and argues for it.

He indeed calles Thracian a "Southern Slavic geo-variational group" but he also advances much more than that: "we could advance the hypothesis that the Thracians were a Slavic group [...] we could then advance the hypothesis that Thracians was the name that Herodotus gave to the Slavs, owing to the fact the Thracians were one of the most powerful and representative elites of Slavic speaking Eastern Europe". He mentions a "Thracians = Slavs" equation, he mentions a Slavic speaking Eastern Europe. This is a radical change in views ignoring the otherwise complex ethno-linguistical maps suggested by other scholars.

Therefore for the time being I must plead innocent for my guilt. I'll address the arguments on Thracians vs Balts vs Slavs and scholarship a bit later. Daizus 14:36, 26 February 2007 (UTC)

OK, I hadn't checked the English translation (Alinei's) since I work off the original Italian, but now I can see where the confusion lies - "Southern Slavic group" looks to me as if it means "a group of people who are Slav". The original "un gruppo periferico della Slavia meridionale" isn't the same thing at all. It means a peripheral group inhabiting the Southern Slav area - i.e. it is a geographical location. It is very clear from the original that Alinei doesn't think that the Thracians are just Slavs, since he writes "pendant dei gruppi baltici della periferia settentrionale. A differenza dei gruppi baltici..." - i.e. they are counterparts to the Balts, who are also not Slavs, although he thinks that all three groups, Slavs, Balts and Thracians share common origin (Jonathan Morris2 11:25, 2 March 2007 (UTC))

All I have read of Alinei can be found here - both in English and Italian, but only in English I've found a detailed presentation of his theories on Thracian language and Thracians. You're mentioning an Italian original quite different from the material I've consulted, I'm not saying it's not possible, I just want you to note that text is subtitled as "Expanded version of a paper read at the Conference Ancient Settlers in Europe, Kobarid {my note: in Slovenia}, 29-30 May 2003." and is authored "by Mario Alinei". No mention of translation, no mention of a translator (other papers from the same site have the mention "translated from Italian", so here would be an accidental and dubious omission), a paper for a conference held outside Italia - with all that you're saying there was an Italian original to it? Without further access on the paper read at that conference, or if other persons were involved in the creation of this text, I must assume Alinei fully responsible for the text and that the text reflects his views, regardless if the text was initially conceived in Italian or in English.

Now, considering also those execerpts from his texts in Italian, Slavic is (also) a linguistical group. A group (even peripherical) from "Slavia", if it's not stressed to be speaking another language, it is speaking Slavic (of course, a version of it - being a separate language or dialect from other members of the same group). I must emphasize again, in the text I invoked, he asserts a "Slavic-speaking Eastern Europe" at the time when Herodotus lived.

Let's consider another point of view. He clearly states there was no Slavic invasions (section 7.5 from "Interdisciplinary ...", starting at p. 32), he speaks of post-glacial Slavic area covering also the territories assigned tradionally to Thracians (p. 33), he speaks of Balts to be the northern neighbours of the Slavs (while Thracians are not the southern neighbours as one would have expect in the symmetry you're suggesting) (p.33), he assigns all the ancient archaeological cultures from Balkans (except Illyrian and Greek) to a Southern Slavic area (pp. 33-34), he justifies the Balkanic Sprachbound and the relative homogenity of Southern slavic languages through the presence of Slavs in Balkans from Antiquity exactly in the same territory where otherwise we know Thracian tribes to have lived (pp. 34-35). If the Balkans are from his point of view Slavic, while the Baltic territories apparently not, it's obvious we can't justify his view on Thracians with what he is saying about Balts.

Now I'll pick again on the arguments on Thracians from section 7.5.5:

• Alinei mentions an archaic Turkic influence on Thracian arguing the sica is a typical centr-Asian metallurgy?? - no reference
• Alinei conjugates Hoddinott's identification of Ottopeni-Wittenberg (Carpathian basin) culture as an early Thracian culture with the latest research (?) which argues this culture is a continuation of Baden and Vučedol cultures (identified by Alinei as 'Slavophone'), the latter being connected with Steppe cultures (he quotes Lichardus & Lichardus affirming a connection with kurgan cultures); from these he concludes (?) the Thracians must have been a southern Slavic group who underwent strong Turkic influences and that's why they extinguished (?).

This is all about archaeology. Now linguistics:

• Alinei stresses Thracian is an IE satem language, like Baltic and Slavic (and I have replied: he ignores completely the Indo-Iranian group which offers interesting parallels with Thracian)
• Alinei invokes Trubačev for a similarity between Thracian and Baltic toponyms. However Trubačev similarities are not so numerous and contested (Sorin Olteanu, for instance, suggests the closeness between Thracian and Baltic was exaggerated) and as Alinei he ignores parallels with other IE languages. As you have remarked Ivan Duridanov also supports a similarity between Thracian and Baltic toponyms. Yet what we should point out that none of these linguists support directly Alinei's hypothesis. a) there's no strong parallelism drawn between Slavic toponyms and Thracian b) some claims though similar with Alinei are in fact incompatible. For instance, Duridanov claims there are some similarities between Thracian, Dacian, Slavic, Baltic (he also claims Thracian and Dacian to be distinct languages) but they have broke up as distinct languages in 3rd millenium BC, ~3000 years before Alinei's analysis of the Slavic invasion. Duridanov also notes the Thracian had more distant relations with Greek, Italic and Celtic (as we were speaking of Celts in Eastern Europe).
• Alinei compares Thr. -dizos/-diza ("fortress") with the OSl. ziždo, zydati ("to build"), zydŭ, zidŭ ("wall"), claiming they are closer than the Baltic ones. Yet Alinei seems to ignore completely the Avestan daeza ("wall") or the Persian didā ("fortress") - see Pokorny (but also Duridanov. He also compares the Thracian Strymōn/Strymē (the former is a hydronym) with the Polish strumień ("river"). The Polish term could be related to the Germanic stroum/Strom ("river", "stream"). Duridanov discusses this term, too.

Here the linguistic arguments end.

Alinei's conclusion is "The most plausible hypothesis would be then that Thracian was a conservative type of Slavic, still preserving Baltic features and spoken by a peripheral group of Southern Slavs, somehow parallel to the Northern peripheral Balts (following the geolinguistic well-known rule, according to which the center innovates, and the periphery preserves).". I disgaree with him, as the aforementioned arguments barely have shown there could(!) be some similarities. The omission of larger perspectives within the IE group are fatal when jumping to conclusions. This is my personal opinion, so you can disregard it. I however demand stronger arguments to claim Thracian (or any other tradionally non-Slavic group) was a type of Slavic. Daizus 13:51, 2 March 2007 (UTC)

Hi Daizus, with all due respect, I don't think that you are adapting your thinking to a PCT framework. For Alinei, the Slavs have been around since the Mesolithic/Neolithic (see Vol2, Ch 5), as have the Balts, hence his long chapter about how there are cultural parallels to the Uralic/Baltic linguistic frontier dating back to the Mesolithic (this is a key point, since if everything up there was undifferentiated pre-IE, you wouldn't expect to find this cultural boundary at such an early stage) - hence when he says that Thracian is related to Slavic, the point at which let's say proto-Thracian and proto-Slavic start to diverge is probably back in the Neolithic. Your comments seem to suggest that you believe that he is saying that Thracian only differentiates from Slavic at a much later stage and this really isn't what he's saying at all. Please note, I am merely trying with all my comments to clarify what Alinei is actually saying, not whether or not he's right.

As for the 2 etymologies - frankly they're common IE words. -dizos could well be an Iranian loan, but who is claiming that there isn't any borrowing from Iranian into Slavic. If anything, the presence of Iranian speakers in S Russia would lead you to expect extensive borrowing.

I'm not sure you can draw many conclusions about strymon, except that it's very unlikely to be a borrowing from Iranian/Gk, since , there is a basic re-/ra- root which seems to mean 'flow', which is present in these 2 groups, but the st- prefix is Slavo-Germanic. (Jonathan Morris2 13:06, 17 March 2007 (UTC))

I have linked the paper I've read, I have browsed it and commented it in my earlier reply. If my understanding is wrong, please follow the same material as I did and show me where I was wrong.

The 2 etymologies are given as arguments by Alinei. I'm expecting for the one using them in such a way to show a) that Thracians indeed borrowed them from Slavs (and not from someone else) b) that these two examples are indeed meaningful for the relation between Thracian and Slavic languages (e.g. English has a lot of borrowed words from French). Daizus 14:03, 17 March 2007 (UTC)