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

[Ante Aikio]

[I wrote]
>There is no reason to concider *-i (= your *-e) in PU *nimi 'name' a stem
>formant. *-i is no known a PU morpheme

[Fabrice Cavoto]
> It certainly is, see below.

No, it isn't - see below.

[I wrote]
>, and *nim- couldn't even
>theoretically be a PU morpheme, since all roots (except pronouns and the
>two auxilaries) had to be of shape *(C)V(C)CV-.

[Fabrice Cavoto]
> See f.ex. Decsy 1990, p. 26-35, as well as, more
> recently, Abondolo (ed) 1998, p. 6-7. The structure of Uralic *roots* is V,
> CV, CVC or CVCV. The fact that PUral. *words* ended in a vowel doesn't
> imply that the vowel was originally part of the root. Some call it
> "thematic vowel", other "stem vowel", and it is mostly seen as an element,
> or part of an element, forming stems.

Abondolo 1998: 6-7 doesn't discuss PU root structure. As for Décsy, his
view that all PU words ended in a vowel is untenable (as are many of his
other views concerning PU phonology). There is a very clear distinction
between the reflexes of PU suffixes of the shape *-C, *-Ca/-Cä and
*-Ci. Cf. e.g. PU genitive *nimi-n, accusative *nimi-m, nominative plural
*nimi-t, but locative *nimi-nä, nominative with sing.1.p. px *nimi-mi and
separative *nimi-tä.

Also, pronouns could end in a consonant: e.g. PU *mun 'I', *tun 'thou'.
Because *-i and *-a/*-ä were also present in the morphologically unmarked
nominative of nouns in PU (e.g. *nimi 'name'), there is no justification
for a segmentation like *nim-i. I can't see why a PU noun "root" should be
anything else than the nominative singular.

> Whether it is for prosodic or
> morphologic reasons is not entirely clear, but the fact that the vowel is
> not always the same (since there are a-stems, e/i-stems), and that it
> happens that the same root is attested with both, also points to the fact
> that that element, what ever we call it, is added on the root.

The fact that the vowel is not always the same, i.e. that there are
stems in *-i and stems in *-a/-ä, points to the opposite: the vowel was a
part of the root, since it is an arbitrary element - there's no way to
predict which stems have a high vowel and which a low one.

Could you specify what you mean by "the same root is attested with [*-i and
*-ä/-a]"? I can only think of the morphologically unclear pair PU *luki-
'count', *luka 'ten', which has no parallels. I did check the 700+ cognate
lexemes between Saamic and Finnic and found only two cases with irregular
correspondence of stem vowel: Finnish raina 'badly worn item' (< *rajna) =
Saami ruoidna- 'to become skinny' (< *rajni-) and Finnish sysätä 'to shove'
(< *süskä-) = Saami saskat 'to copulate (of reindeer bull)' (<
*süski-), both of which are obviously affective.

Regards,

Ante Aikio