> It wasn't anything similar. I'm not sure what you're
> referring to, perhaps something I said about the alternation
> *gWem- ~ *gWah2- and similar ones. If I remember correctly,
> I tried to explain it as **gWamh2-V- (syllabificatioon
> *gWa-mh2a-) > *gWem- versus **gWa:mh2-C- > **gwãh2-C- >
> *gWah2- (with etymological /a/, not /e/, before *h2).
>
Then this from the same article by Pulleyblank might interest you:
"
7. GO, COME
Ch.
wâng .. EMC wuaân,' < *wàân,? "go",
yú .. EMC waâ < *wàâ < *wàG "go (to); to; at, in",
Tib.
h.on-ba 'come',
Burm. wan, "enter; go in; come in' Bodman (1980:81);
IE
*gWa:-, *gWem- "go, come"
OI
jíga:ti, gámati, gácchati, "goes"
Gk. báino: "go", `ébe: "went"
Lat.
venio: "come",
Eng.
come etc
"
He further posits PPIE *w > PIE *gW; I think Sean Whalen did too in
his reconstructions?
Perhaps it's PPIE *wax-/*wan,- > PIE *gWax-/*gWem- with labial
assimililation in the latter variant, where PPIE *wax-/*wan,- is a
loan from Chinese, in which homorganic consonant alternation in
finals is reconstructed (eg. mjag "not", mjang "not"; ngag "we, my,
our", ngang "I, me")?
Torsten