Tocharian A wäs, B yasa

[Ante Aikio]

The Tocharian word for 'gold' (A wäs, B yasa) has traditionally been
compared with a Uralic group of words with meanings 'metal', 'iron',
'copper'. It has often been suggested that the U words might be of IE
origin, although this assumption is problematic. I am interested in
inspecting the possibility that the Tocharian word is a Uralic /
Proto-Samoyed loan.

First, the data. The Uralic word has two forms, a front and a back vowel
variant.

The front variant can be reconstructed as *wäs´kä [> Saami veaiki
'copper', Finnish vaski id., Mordvin vis´ke 'metal wire', Proto-Samoyed
*wesä (> Nyenets yesye 'iron', Nganasan basa 'metal; iron', Selqup kësï
'iron' etc.)]. The back vowel (-a-) as well as the second syllable vowel
(-i-) in Finnish are probably secondary; there are also other examples of
sporadic *ä-ä > a-i in Finnish. Mordvin rather suggests 1st syllable *-e-
(*wes´kä); there is also an irregular dialectal variant us´ke (with *vi-
>> u-).

The back vowel variant appears in Hungarian vas 'iron' and Mari vaZ 'ore,
metal' (< *was´kV). The loss of *k in Mari is irregular.

In addition to this, there are phonologically unclear cognates in Permic
and Ob-Ugric. They seem to have undergone reductive phonological
developments, since they only appear as the last member of compound words
that are names for metals (thus, the meaning seems to have been just
'metal').

The many phonological irregularities suggest that the word is an early
"Wanderwort", and this is also compatible with the fact that PU must have
been a stone-age language. However, the word must be quite old, since it
appears in every branch of Uralic, and at least the consonant
correspondences are regular (except for Mari).

The direction of loaning can hardly have been IE / Tocharian > Uralic,
because:
a) Tocharian -s- would not give U *-s´k-. One could of course assume
Tocharian > Samoyed (this has been suggested), but this leaves the other U
words without explanation. The correspondence Samoyed *wesä ~ U *wäs´kä
/ *was´kV is hardly a coincidence, since PU *ä > Samoyed *e and PU *s´k >
Samoyed *s are regular developments.
b) It seems very unlikely that a Tocharian loan word could ever
have reached the western periphery of Uralic (Saamic, Finnic). The
Tocharian word does not seem to have certain correspondents elsewhere in
IE (IEW mentions it under *auso- 'gold': "vielleicht Toch. A wäs").

Now I have two questions:

1) What was the Proto-Tocharian form? (If it was something like *wVsV with
front vowels, it fits quite well with Proto-Samoyed *wesä. A loan Samoyed >
Tocharian is also geographically the most sensible alternative, if Tocharian
was connected with the Afansevo culture.)

2) Is there any other plausible etymology for the Tocharian word?