well, prehistoric iberia was inhabited by the iberians (speaking different dialects of a non indoeuropean language), celts, phoenicia, greeks, and later romans. There have been theories on iberian being related to basque, but i remember not being convinced by the arguments and methods used. Anyway, we have testimonies of these languages, of iberian only a few, so that its unclear where that language came from and to what other language it might be related. But my point is, the first testimonies of basque are as young as the first spanish words, so we are talking about a time maybe a thousand years ago. No sign of the basque language before that point. Strabo, a historian that lived about 2000 years ago, mentions the peoples mentioned above, but again: no basques. This, I think, indicates that the basque came to iberia later. we don't know where they came from, but i can't imagine them coming from the north through the roman empire. There would surely be some historic documents. But again, I recommend reading papers on other theories on where the basques (and their language) came from.

you mentioned genetic similarities between todays basques and northern european celts. The celtic culture can be found all over europe, in spain inscriptions in celtic languages where found. So a similarity can only be a sign of celts in iberia. Also, a language and a people, are not always congruent. People tend to take over other languages under certain circumstances. So celts could be among the genetic ancestors of todays basques (i think i remeber reading, that the genetic difference between spaniards and basque ist very small. also the basques did not always separate themselves that strictly from spain).

i'm not an expert, but I remember reading about different theories on the origin of the basque language while looking into the history of early "iberian languages". I suggest using google scholar (or a library!) to do some research on that.

I guess we can be pretty sure, that the indo-european languages emerged outside of Europe. You can find a recent paper on this here: http://www.sciencemag.org/content/337/6097/957 . I'm sure you'll find some interesting references in the article as well.

Very useful response, thanks. Do you happen to have or have access to the paper that you referred to. It would be very nice to get a copy of the pdf if possible. I suppose it's understandable that there is a lot of effort put into studying the origins of Indo-European languages, not least since so many of us speak them, and it's easy to access written material. It does seem a shame, however, that very little effort seems to have been directed towards pre-Indo-European languages of Europe. Maybe there is a career for someone out there! I have read about the theory of Theo Vennemann that Basque ancestors spread through north-western Europe prior to the arrival of Indo-European-speaking peoples, and that traces of 'Vasconic' languages have been left in various Indo-European languages. It is always indicated that this theory has been 'rejected' by language scholars. To my non-educated mind, it seems obvious that something like this must have happened. We know that Cro-Magnons were in southern Europe before the LGM, and that some of these people from the Iberian Peninsula moved up through western Europe as the ice receded from 17,000 years ago onwards. Given the unique nature of the Basque language, the current positioning of Basque speakers in a well-known ice age refuge, and the genetic similarity between Basques and peoples of the north-western edges of Europe (Irish, Welsh etc), it seems logical to think that an ancient precursor to Basque would have been the tongue used by the first people to re-colonize north-western Europe after the end of the ice age. Obviously, this language was displaced by Indo-European langauges, but surely there must be remnants in languages and place names? Any thoughts anyone?

well, prehistoric iberia was inhabited by the iberians (speaking different dialects of a non indoeuropean language), celts, phoenicia, greeks, and later romans. There have been theories on iberian being related to basque, but i remember not being convinced by the arguments and methods used. Anyway, we have testimonies of these languages, of iberian only a few, so that its unclear where that language came from and to what other language it might be related. But my point is, the first testimonies of basque are as young as the first spanish words, so we are talking about a time maybe a thousand years ago. No sign of the basque language before that point. Strabo, a historian that lived about 2000 years ago, mentions the peoples mentioned above, but again: no basques. This, I think, indicates that the basque came to iberia later. we don't know where they came from, but i can't imagine them coming from the north through the roman empire. There would surely be some historic documents. But again, I recommend reading papers on other theories on where the basques (and their language) came from.

you mentioned genetic similarities between todays basques and northern european celts. The celtic culture can be found all over europe, in spain inscriptions in celtic languages where found. So a similarity can only be a sign of celts in iberia. Also, a language and a people, are not always congruent. People tend to take over other languages under certain circumstances. So celts could be among the genetic ancestors of todays basques (i think i remeber reading, that the genetic difference between spaniards and basque ist very small. also the basques did not always separate themselves that strictly from spain).

I was a little surprised by the emphasis on Basque, inasmuch as there are several non-Indo-European languages in modern Europe (Finnish, Estonian, Livonian, Lappish, and Hungarian) with substantial numbers of speakers as well as others with smaller numbers of speakers (plus those that have disappeared).

Any such language would also have to pre-date IE, whose geographic origin (i.e., where its speakers lived) is generally accepted to have been north central Europe.

The origins of the Finns and Hungarians are more clear-cut, I understand, as these languages are part of the Uralic group, with close relatives in the Ural Mountains in modern-day Russia. It's not hard to see how these could have been introduced to parts of central/northern Europe through migration.

Basques are genetically very western European, I'm sure you're correct to say there is lots of similarly to ethnic Spanish. Mitochondrial DNA and Y-chromosome evidence suggests that up to 80% of the background of peoples of north-west Europe (Irish and Welsh) is derived from the Basque refugium that provided the people for the first colonization of north-western Europe after the end of the Ice Age. With this genetic background, it is impossible to argue for a non-European origin for Basques. You sent me a very nice paper on the origins of Indo-European. So, it now looks fairly certain that its origin is Anatolia (modern-day Turkey). It seems highly likely that the introduction of Indo-European languages into western Europe coincided with the Neolithic spread of agriculture. Agriculture means more people, means more dominant societies. This all seems to be the perfect scenario for the displacement of indigenous European languagues. I am sure it will continue to be hard to prove, but it does seem like the chance of proto-Basque being one of those early indigenous languages seems is extremely high. That said, I would be very interested to look at any papers that you might have on the origins of the Basque people and their language. Do you have any of these?

DNA might be explaining the origin of the people, but not necessarily of the language. People take over other languages, other cultures. But that does not explain where there language comes from. The people we call basque might have been living in the same place for many thousand years, but still, basque might not be their "native language" (again, no sign of basque until 1000 years ago, same is true for the above mentioned iberian that suddenly is found in iberia 2400 years ago). In Lower-Saxony, Germany, 3000 year old remains were found in a cave (Lichtensteinhöhle). In a village close to that cave descendant of those 3000 year old bodies were found, still living in the same spot (there are papers and publication on this, but they all seem to be in German). But, they didn't speak German or not even a germanic language 3000 years ago. The culture changes, language changes, and language shifts (the latest language shift in that region was about a hundred years ago, giving up the language that was spoken there for ca. 100 years the shifting to High German).

These are interesting points. It is certainly correct to say that language and genetics are not directly linked. However, ethnic groups that have a technological advantage and move into areas previously occupied by other less advantaged groups very often carry their language with them. This is the story of English in North America, which has virtually wiped out (sadly) native American languages, and it is also the story in most of Latin America. More recently, there have been fewer large scale population movements. So in Africa, although English is widespread, and French too, most people have retained their African genetic identity. My suspicion is that Europe must have been similar at the start of the Neolithic, as mentioned above. In that kind of situation, there is a very high likelihood that the immigrants superior technology (agriculture) would have enabled them to increase in numbers and effectively out-compete indigenous populations.

Although English, and German too, are modern languages, we can still trace them all the way back to a common origin 10,000 or so years ago. In English, we only have to go back 1000 years and the language is no longer English and completely incomprehensible to a modern English speaker. Same must be true for German, Spanish, French, Finnish, Basque and probably most other languages. However, linguists (and I'm not one) can still pick out the relationships. So, although Basque 1,000 years ago was probably only marginally like modern Basque, a relationship could still be identified with modern Basque. Just take it a few more thousand years back, and I would assume that the same would be true. Only problem is that there are no other languages in this family to compare with. Assuming therefore, that Basques have been where they are for the last 10,000 years (highly likely from genetic and climatic evidence), and assuming the default that they are speaking an evolved form of the same language that their ancestors spoke 10,000 years ago, then it seems logical to assume that their 10,000 year-old neighbours were most likely speaking a language related to proto-Basque. Make a logical extension to this scenario and add to that evidence suggesting population movement from the Basque area up into north-western Europe roughly 10,000 years ago, and we have a perfect scenario for the widespread use of Basque-related languages across north-western Europe during this period. Now I think of it, there is an almost perfect analogy in East, Central and Southern Africa. Are you familiar with the history of the Bantu migration and the impact on Xhoisan languages?

One of the most active scholars interested in detecting traces of "Old European" languages is Theo Vennemann. Anyone interested in this question might check out some of his publications.

I was aware of this, and did get a copy of a review of his large publication on this topic. I contacted one of the authors of the review (Phil Baldi) who kindly responded, but unfortunately to tell me that he was no longer active in this area of research and did not know of anyone else who is. I would be very interested to hear if there are still people working on this area of research.

Don Ringe has a long and interesting article on the linguistic diversity of prehistoric Europe here http://languagelog.ldc.upenn.edu/nll/?p=980

Wow. Thanks Cormac. That is an extremely interesting article and the most readable piece that I have encountered, speaking as a non specialist. I very much like the way that he equates pre-historic Europe to other aboriginal geographical areas. His thesis seems to take away from the significance of present day basque, and yet it also brings that language into particular focus. Why was it that only Basque, of the dozens of pre-historic languagues that must have been used in Europe prior to the arrival of IE, survived to be spoken in the present day? Is this a consequence of the partly mountainous region occupied by its speakers in a part of Europe that is further away from the origins of IE than most. Don Ringe's piece draws my thoughts to the situation in the country where I live, which is Tanzania. Right now we see dramatic changes before our eyes in the patterns of language usage, as the 100+ languages spoken decline rapidly, being replaced with the nation state languages of Swahili and English. Casting thoughts back to the European situation, however, and thinking that IE may have worked its magic more than 3 millenia ago, it does give an indication of that language 'movement' and the power of the culture that drove it through Europe, leaving almost no trace of what was present before. It really does say something about the power of the neolithic culture and associated livestock rearing culture. I'd be interested to know if Don Ringe is contactable. Do you know him? Thanks again for the short but really helpful response.

Language, ethnicity, anthropological type are not the same things, but this popular article helps us to understand why the Basque language has its chance to be an "original European": http://www.bbc.com/news/science-environment-34175224    

Olga, many thanks for this very helpful contribution! I had a look at the BBC newslink that you provided, and what an interesting story it is. Since genetic evidence does seem to confirm that the Basques have a strong genetic background in common with early hunter-gatherer populations in that part of Europe, and that Indo-European languages came with farming groups that moved into Europe from the East, it does look like there is a very strong case for Basque being an original European language. Both genetics and language also seem to highlight the fact that this group of people have been relatively isolated through early history, which again would make it more likely that their language is ancient. Thanks again for the interesting post!

Olga and others. Do you happen to know who the authors are of the article mentioned in the BBC news story referenced above? I would very much like to request for a copy of the article in PNAS. Thanks!

Dear James, I have found out only this link http://www.pnas.org/content/early/2015/09/02/1509851112 in the article. Yes, almost all of us are immigrants from prehistorical or historical point of view.

Dear Peter, thank you very much for your contribution. I'm not an archeologist, but I'm interested in the mythology about the origin of mankind, folk cosmology, etc, so I should know not only the folk explanations. Btw, one can find a lot of important information in this journal: http://www.archaeology.nsc.ru/en/publish/journal/jrAEAE_en.aspx

Dear Olga, that link worked perfectly, so thank you very much indeed!

Dear James, you are welcome, but we should thank archeologists and specialists in genetics for their mutial work.

I mean "mutual"). Such works are very popular now in the Euroasian studies.

Maybe it shows even more links with other families. See J.Cejador "Embriogenia del lenguaje"