PU *nimi / PIE *HneH3men- (was: Re: IE "Urheimat" and evidence from Uralic linguistics)

[Ante Aikio]

On Tue, 15 Feb 2000, Adam Hyllested wrote:

[I wrote:]
>> Actually, the lack of reflex of medial *h3 [in Proto-Uralic *nimi 'name'] is
>> a bit problematic. One would expect borrowing from IE *Hneh3men- to give PU
>> *nexmi / *nixmi.
(snip)

[Adam Hyllested relpied:]
> Actually, the laryngeals should be no problem for U *nime-/*nima:- as an
> IE loanword.
(snip)
> In protetic position before consonant, laryngeals are usually
> reconstructed as their consonantal variants. But the material from Uralic
> (and other language families) generally tend to speak against
> this, no matter whether you believe the look-alikes to be loans or
> cognates. So we should perhaps rather reconstruct a *@1neH3mn,
> phonetically realized as *nnoYwmn (read Y as gamma here; the consonantal
> variant of *H3 was probably phonetically realized as a voiced, labio-velar
> fricative *Yw).

But if you reconstruct *H3 phonetically as *[Yw], this should give PU *x
(which was phonetically most probably *[Y]). One can speculate that PU *xm
was perhaps phonotactically excluded, since there is no evidence of
clusters consisting of *x and a nasal. But still, one would expect that
*H3 was substituted with something (e.g. *w?)

> The development *-eH3- > *-oH3- took place already in PIE, and I find it
> very unlikely that a PIE *-oYw- should show up as *-i- in Uralic. Of
> course the borrowing could have taken place at the time of
> Pre-Proto-Indo-European, i.e. before the "colouring" of *eH3 to *oH3.

IE *o > PU *i would indeed be impossible. If PU *nimi is a loan, it is
either Pre-IE, or else it must derive from the zero grade - *(H)n(H3)men-
> PU *nimi- is phonetically sensible, given Uralic phonotaxis, which
requires roots to be of shape *(C)V(C)CV-.

> But
> the word is also found outside Uralic; it appears in Yukaghir as <niu,
> neve, nim> and in Chuvan as <nyva:>. An Indo-Uralo-Yukaghir
> reconstruction *(n)newme- seems much more probable.

It seems ad hoc to me, since no conclusive evidence of a genetic
relationship between these language families has been presented. But it
seems likely that at least the Yukaghir item is not a chance
correspondence. (Perhaps a loan U > Yukaghir? Other such borrowings have
been pointed out.)

> Critics would point out that the PIE word is formed by adding a
> derivational suffix *-men. First of all, I don't see why a stem ending in
> *-me shouldn't analogically add an *-n, if nouns are productively formed
> with a suffix *-men. Secondly, the Uralic reconstruction *-a: corresponds
> perfectly to the IE vocalic *-n (*-e doesn't).

But you can't reconstruct PU *-ä for this item: the reconstruction must be
*nimi (= traditional *nime). Second syllable *i gives regularly Finnish -i
: -e-, Saami -a and Mordvin and Proto-Samoyed zero. Thus, PU *nimi >
Finnish nimi : (oblique stem) nime-, Saami namma, Mordvin (dissim.) l´em,
Proto-Samoyed *nim. There is no evidence for 2nd syllable *ä here: it
would have been retained in Finnish and Samoyed, and changed to -i in
Saami and -e in (Erzya) Mordvin.

Regards,
Ante Aikio