Talk:Proto-Basque language

Nasality

The *n to /h/ change is really odd, esp. considering the contrary *b to /m/. Is there any speculation how this happened, or what the intermediary steps were? kwami (talk) 11:46, 17 May 2009 (UTC)

If I remember rightly, it only applies to intervocalic n and goes something like this: VnV > VV (two nasal vowels) > VhV insertion of h to break up the vowel group > modern loss of h in Western Basque takes us back to square one. So you get *bini 'tongue' > *mini > *mĩĩ > mihi > mii/mi. Akerbeltz (talk) 00:20, 19 May 2009 (UTC)

Neato! Are the nasal vowels retained in any dialects? Is *b > m related to this, or is it unconditioned? kwami (talk) 00:32, 19 May 2009 (UTC)

Interestingly, Portuguese has gone through a very similar development. Unfortunately, while it is tantalising to suspect a historical connection between the sound change in Basque and in Portuguese, I cannot see how it could be established even in principle, seeing as the territories are separated by those of Asturian-Leonese and Castilian, which do not even exhibit anything like that as far as I'm aware. But at least, the typological parallel stands (and if the sound changes in question are entirely independent of each other, that may even be more interesting). --Florian Blaschke (talk) 03:07, 22 December 2009 (UTC)

Except that in Basque this was over by the 11th century, so they couldn't be related even in theory.. This loss of intervocalic */n/ is apparently termed "Mitxelena's law" and didn't usually operate on morpheme boundary at all, and was a part of a general trend of loss of nasality (which itself later arose by some other means) which happened in different dialects in different periods with different outputs..

The change of */b/ to */m/ is apparently a simple assimilation by nasality */bVn/ > */mVn/, and must have happened prior to the aforementioned sound change (because of *bene > *mehe example in the table). In case of *un-be > *ume we're dealing with assimilation by place of articulation *unbe > *umbe and simplification of the resulting cluster *umbe > *ume.

It's all in the Trask/Wheeler dictionary.. --Ivan Štambuk (talk) 06:24, 22 December 2009 (UTC)

Except that we don't know when exactly the loss of intervocalic /n/ in Old Portuguese happened – it could well have been contemporaneous with the loss of intervocalic /n/ in Basque, see p. 180 bottom. --Florian Blaschke (talk) 18:11, 31 October 2021 (UTC)

Loss of /n/ and "loss of nasality" are two DIFFERENT stages. The first one was due to assimilation, as surrounding vowels became nasalized. For example, proto-Basque *s'eni became *s'ei~ in the High Middle Ages. Then became loss of nasality: *s'ei~ > sehi, segi. But in some cases fro this nasal vowel a new /n/ was generated: *s'ei~ > sein. These stages are well represented in Galician-Portuguese: (1) nasal vowels in Portuguese (e.g. ma~o 'hand'), (2) complete loss of nasality in Eastern Galician (e.g. mao), and (3) regeneration of /n/ in Western (now standard) Galician (e.g. man). In Basque, stage (1) is PARTLY conservated in the Easternmost dialects, Zuberoan and Roncalese (which is now extinct).

IMHO, the common opinion that intervocalic -m- arises from a *-nb- cluster is the result of an incorrect analysis of linguistic data. Definitely, /ume/ doesn't come from **unbe but from *unne (*uNe in Mitxelena's notation), as in /ahuña, ahüñe/ (Biscayan /aume, auma/) 'goat kid' < *an-unne, the /m/ resulting from LABIALIZATION of /n/ after /u/. This is also the case of /zume/ 'osier' < *sunne (PNC *ts’s’wǝ:nHe: 'reed, cane'), which some idiot has proposed to be a compound from zu- 'wood' and mehe 'slender'. The cases of /ama/ 'mother' < *anna and /eme/ 'woman' < *enne are rather peculiar because there's no /u/ to explain labialization.

The reason for the absence of *m in Proto-Basque is rather simple, as it simply merged with *n. For example, Basque zehe 'palm (measure); line' < proto-Basque *sene (PNC *tʃ’VmħV 'span') corresponds to Roncalese (t)xeme 'span of thumb and index finger'. Talskubilos (talk) 12:26, 6 April 2010 (UTC)

Agreed, they're two different stages. The problem with your interpretation of intervocalic -m- though is that we have the attested form of Aquitanian VMBE > ume. Not suggesting it couldn't be the result of -nn- > -m- either but it's hard to argue against an attested form. What's PNC? Akerbeltz (talk) 12:51, 6 April 2010 (UTC)

Whoa, relax, I'm not having a go. Yes, you're right, the attested form is VM.ME - sorry, I was doing this from memory and seem to have gotten it mixed up in my memory with SEMBE. Still not sure what PNC is. Akerbeltz (talk) 21:11, 6 April 2010 (UTC)

PNC stands for Proto-North-Caucasian, a language reconstructed by S. Starostin & S. Nikolayev in their North Caucasian Etymological Dictionary (NCED). See North_Caucasian_languages article. Talskubilos (talk) 21:24, 6 April 2010 (UTC)

OR by some layperson based on tortuous interpretation of ambiguous evidence (where the layperson even admits that there is evidence that contradicts his hypotheses and that he himself can't explain, though omitting the even more awkward example seme) and Vasco-Caucasian fantasies, and contradicted by RS written by actual (sceptical, not credulous) experts like Trask (who has refuted attempts to link Basque with Caucasian languages himself), is irrelevant. The ancient forms SEMBE- and OMBE-, VMME are clear enough, and disprove the baseless reconstruction **uNe.

Also, mao > man makes no sense and the magical reappearance of the nasal previously lost without trace is impossible – the obvious solution is that Old Portuguese mão /mã.u/ simply yielded Modern Standard Portuguese (based on southern dialects) mão /mɐ̃w̃/, Eastern Galician mao and Western Galician man /maŋ/ directly. --Florian Blaschke (talk) 18:11, 31 October 2021 (UTC)