basic vocabulary borrowing (was: IE "Urheimat" and evidence from Uralic linguistics)

[Ante Aikio]

On Sat, 5 Feb 2000, Miguel Carrasquer Vidal wrote:

> Ante Aikio <anaikio at mail.student.oulu.fi> wrote:

>> What is likely to be borrowed and what is not in a given
>> circumstance depends on many factors. In U languages, the numerals are
>> largely cognate and there are no known borrowings except for 'seven',
>> 'hundred' and 'thousand'.

> It has been claimed, though, that the affix -deksan (etc.) found
> in Finnic words for 8 and 9 is also of IE origin.  (I doubt this
> is true).  Permic *das "ten" is obviously of Iranian origin.  I
> can't remeber what the story is on Hungarian <ti:z>.

Well, I actually forgot about 'ten'. The Hungarian word is also an Iranian
loan. Concerning the affix *-teksä(n), this explanation has been
recently revived: it has been argued that it is a loan from Proto-Iranian
*detsa. The phonetics are flawless; there are other examples of U *ks <
Iranian *ts (The cluster *ts was illegal in U, hence the substitution).
This explanation seems more plausible to me at least than the previous
rather fabricated theory that Finnish kahdeksan and yhdeksän developed
from *kakta e-k-sä-n 'two do not exist' (i.e., "two are missing from
ten") and *ükti e-k-sä-n.

>> But even numerals can become cultural
>> items; e.g. the loan origin of Ob-Ugric/Hungarian *säptä '7' (< Aryan /
>> Iranian) and Samoyed *sejpti (< Tocharian?)

> Tocharian A has <s.pa"t> and B has <s.ukt>.  We would expect
> *septm. to give PToch *s^IptI (*s.a"pta"), which leads to the
> Toch. A form without much problems (*s^IptI > *s^IpIt > s^pIt).
> If the Toch. B form went through a stage *s^Iw(I)tI (*-p- >
> *-w-?, with -kt later by analogy from <okt> "8"), that might
> explain Nenets <s'i?~iv>, Enets <sew>.  But I can't see how
> *sejpti (based, I guess, on Nganasan <s'ayb@> etc.) might derive
> from Tocharian.  The /b/ in the Samoyed forms rather reminds one
> of Germanic *sibum.

I have to correct myself a bit. The Proto-Samoyed reconstruction should be
*sejTwE (where *T = *k or *t; *E = schwa) - this accounts for all the
Samoyed forms. The idea of borrowing from Proto-Tocharian originally
derives from Juha Janhunen, and it seems phonetically problematic, to say
the least. In addition to the *j, which is hard to account for, one has to
assume an irregular metathesis *pt > *tw in Proto-Samoyed.

>> , replacing PU *s´exs´imi '7',

> As in Permic <sizim>, Mordvin <sisem>.  Is this also agreed to be
> the prototype of Finnish <seitsema"n>?

My reconstruction *s´exs´imi is a bit problematic. It should rather be
*s´Vs´imV (the first syllable vowel cannot be reliably reconstructed, but
at any rate it was a front vowel). There is very little evidence for my
*-ex- here, although it would account for the vowel relation between Saami
and Mordvin rather nicely. The Finnish word is irregular. -än is
clearly due to analogy of kahdeksan and yhdeksän, while -its- (pro -s-) is
difficult to account for. If one assumes that the diphtong -ei- in
Finnish is not secondary (although the other cognates show no evidence
in support for this), it might even be possible to link Samoyed *seyTwE
here. *w might be the irregulary weakened reflex of PU *m, thus something
like PU *s´Vjsimi > *s´ejsEmE > *sejtEwE > *sejtwE (U *s´, *s > Samoyed
*s, *t are regular developments). In this case one would have to assume
sporadic assimilation *js > *(j)s´ in the other branches of Uralic. This
is of course quite irregular, but hardly more than the proposed IE loan
etymology.

 Ante Aikio