"Extraordinary claims require extraordinary evidence." ~ Carl SaganSo by a happy accident I ran across some cursory notes of Blevins' Basque-PIE talk. I discussed elements of the talk here. A number of days ago, linguist Juliette Blevins made a surprising argument: Basque is related to Proto-Indo-European. Not that Basque descends from PIE, but that Basque and PIE share a common ancestor, and that Proto-Basque sound correspondences with PIE reveal a Proto-Indo-Vasconic ancestor. You can acquaint yourself with the notes here (scroll down a ways).
Because the notes are simple summaries, probably done in haste, they contain a sketch of the argument but lack any real evidence, substance. A shame because some of these proposals are so radical, they would need significant evidence to convince me. I proffer my own thoughts.
- "Five problems with this proto-inventory that Juliette Blevins is going to fix." Why are there problems with Michelena's or Lakarra's reconstructions? Why can we not be content with the reconstructed phonological inventories thus far?
- "Problem 2: VhV sequences. (where vowels are identical) e.g. behe, bihi, lahar, luhur, mahatz, zahar." In many of these lexemes, the question should not be "Why are there VhV sequences?" but rather "Why are there VnV sequences?" An intervocalic */n/ was nasalized, then lost, unless the /n/ was assimilated (assimilation occurred frequently in words with alternating vocalics). A more interesting theory could be that nasal */n/ became /h/ if bookended by identical vowels and became /ɲ/ if the vowels were different. A word like mahai would come from *manai.
The ruins of the Proto-Basque. Ancient Basque colmechs can
be found throughout the hills and valleys of Navarre. This
picture has no purpose. I just thought an image would be
a welcome break.
- "To our knowledge, no one has attempted to expalin why h is so frequent and why it has wider word-internal distribution than most of the orial stops. We will solve this problem by positing two new historical sources for h." Well, first, Michelena accounts for the presence of intervocalic /h/ as the product of a loss of the intervocalic /n/ (above). Second, it is true that aspiration is very common in some Basque dialects, and was probably very common in Proto-Basque. Because many of these h-words have forms with /g/ and /k/, I've often wondered about an earlier system of *k, *kC, *kV that can no longer be recovered (such as PB *harri and its possible relationship with European *kar- words for stone). Third, it's not true that no one has attempted to explain the frequency of aspirants; Michelena believed them to have arisen from suprasegemental accent changes.
- "We propose only a single *s in PB." Proto-Basque's phonology was likely influenced by Iberian (both lacked an /m/; both boasted laminal, apical s; identical 5-vowel inventories; etc...). To what extent the sibilants were the product of Iberian influence or were an in-house movement is unknown.
- "We [are]proposing *m." Sorry, what? You already scared me by positing no sibilant contrast. Now you're saying Basque had an /m/? I'm beginning to lose the faith, father. You're going to need some powerfully good reasons. If there was no *b > m evolution, then why are there relics of an earlier *b in toponyms (mendi "mountain" but Auzpendi).
From what we have so far, this isn't even close to enough to convince me. On the other hand, it's just a summary from a talk, so I'll still give her the benefit of the doubt. If she writes on the matter, I hope - nay, expect - a significant body of supporting evidence.