First, I deliberately imitated Manansala's sloppy practice and did
not analyse my own "matches" etymologically. I wouldn't be that
stupid when offering a serious etymological proposal. I could show
you that, e.g. <gumi> and <dagana> are not related to each other or
to *dHg^Hom- (it so happens that their origin is known quite well).
Secondly, the game is easy to play with any language families of your
choice. Regrettably, I don't happen to have a Bantu or Algonquin
dictionary at home, but other people have done the demonstration for
me (will Quechua/Semitic suffice?):
http://zompist.com/chance.htm
Piotr
--- In cybalist@..., tgpedersen@... wrote:
> Actually, what you have shown here is how easy it is to come up
with
> a match for IE in Austronesian. That does'nt demonstrate what you
> purport to show: that it is easy to come up with a match in *any*
> language. Austronesian is hardly *any* language here. So, if it's
not
> too much bother, could I ask you to repeat the demonstration with
> Bantu or Algonquin? Preferably with a genetic link too?