PIE and Uralic

[Ante Aikio]

(Patrick Ryan asked:)
> I have two questions.

> 1) Why do the Uralists feel that it is necessary to reconstruct a
> transitional /S/ on the way to Finnish /h/?

Finnish h in root morphemes goes always back to PU *S (with the exception
of the PU cluster *kt, which - like PU *St -  gives Finnish ht). The
change *S > Proto-Finnic *h is relatively late, and there are
Proto-Germanic loan words which predate this change (e.g. Finnish kana
'chicken' < Germ. *hanan-; Germ. *h- gives Finnish k- because there was no
h yet at the time of loaning). Thus, it would be anachronistic to assume
a direct substitution PIE *H > Pre-Finn. *h.

There are also loans in other U languages which show the same substitution
IE *H > U *S, e.g. IE *pewHeno- 'sieve' > Pre-Permic *pewSinV- > Komi
poZ(n-), Udmurt puZ(n-). (The Permic languages have retained PU *S as a
retroflex.)

> 2) Have Uralists speculated that the older responses (/k,x/) might be the
> result of the PIE "laryngal" being realized as a stop /?/ and a spirant
> /h,H,x/?

The loans words may no doubt reveal something about the phonetic value of
the laryngals, and the different substitutions (PU *k, *x, *S) might also,
to some extent, reflect different phonetic values of the laryngals in
different (P)IE dialects. Thus, my hypothesis of PU vs. Pre-U loans is of
course debatable. An interesting question, which has not yet been
touched upon, is if the Uralic substituents of *H1, *H2 and *H3 differ in
any way.

Regarding the sound values, it should be mentioned here that the phonetic
value of PU *x is not entirely clear. However, there seems to be some
evidence suggesting that it was actually a voiced velar fricative (and
could thus be more properly written as *g). There are also two loan
etymologies that support this: PU *wixi- 'take (somewhere)' < PIE *weg´h-
id., PU *mexi- 'give, sell' < PIE *mey-gw- 'sell' (in the latter one PU
*-x- instead of *-jx- because the latter would have been a phonotactically
illegal cluster).

(P.R. asked:)
> Would it be possible, in your opinion, for an alternate explanation that the
> words might have been borrowed before the *-H- root-extensions?

I'd rather let IE-ists answer that. But I'd like to suggest another
alternative explanation to the (Pre-)IE > Pre-U loaning hypothesis I put
forward in my last mail. It is possible that e.g. PIE *pelH- did not give
Pre-U *pelxi-, but rather directly PU *peli-; i.e., the laryngal was left
without a substituent because PU *-lx- was phonotactically impossible.

> I do have an axe to grind here but, mercifully, I will not grind it on this
> list. Of course, those who have been to my website know that I consider that
> a strong case can be made for ultimate common origin.

I am familiar with your web site, and thus I know that we probably
couldn't disagree more here. But I agree with you on that Nostratic
etc. discussions should not take place on this list. But I have no need
for mercy - I am not afraid to defend my views on the relatedness of U and
IE. It's just that I don't have the time nor the interest for such a
discussion at present.

  - Ante Aikio