Dating the final IE unity

[JoatSimeon at aol.com]

>mcv at wxs.nl writes:

>As far as I know, the Hittite word for "wheel" is <hurkis>.  No relation to
>the *kwel->words. >>

-- 'hurkis' is derived from PIE *hwergh, and cognate to TocharianA 'warkant'
(wheel) and TocharianB 'yerkwanto'.

The agreement between Hittite and Tocharian -- very widely separated IE
languages -- would suggest PIE status for this word as well.

There's a broad overlap in the 4 PIE words for wheel:

1. *kwekwlom -- Germanic, Phrygian, Greek, Indo-Iranian, Tocharian,
Balto-slavic, and closely related terms in Celtic

This probably referred to the wheels in a two-wheeled cart, given the dual
form in Old Irish ('cul', from *kwolo via *kwolos).  "The two roundy-roundy
things".

2. *Hwergh -- Tocharian, Hittite

3. *dhroghos -- Celtic, Greek, Armenian

4. *roto -- Celtic, Latin, Germanic, Balto-Slavic, Albanian, Indo-Iranian,
and possibly Tocharian.

So while none of the 4 occurs in all the groups -- one wouldn't expect that,
eh? -- every single one occurs in at least _two_ of the groups.  Eg.,
Tocharian and Hittite share cognates derived from *hwergh, Tocharian has
derivatives of *kwekwlom and possibly *roto, etc.

They're all fairly transparent, too:  "the round thing", "the thing that goes
round and round", "the runner", and so forth.

This is what you'd expect if proto-Indo-European speakers invented the wheel,
by the way -- otherwise there should be at least one loan-word for "wheel",
one that isn't resolvable into a PIE root.