Rob:
> Why would */d/ necessarily survive in derivatives such as *dkmtóm,
> *-dkmtáx?
But it doesn't so why do we even have to ask such a hypothetical
question? We find *kmtom and *-kmtax.
Rob:
> How are *yugóm and *pedóm collections? I thought *yugóm meant "a
> yoke"?
Yes, it does. I shouldn't be calling them "collectives" I guess. More
like "singular" entities out of an uncountable collective perhaps, if
not the uncountable collective itself. A similar situation exists with
the concept of "100" which, given the inevitable semantics intended by
*kmtom, must have been originally considered an unimaginable and
uncountable sum -- "tens of tens", "many tens".
Rob:
> Perhaps we should try to find out how much we agree on right now.
> The constituent morphemes, for example. Were they like this:
>
> *(d)km-t-om
We see *kmtom, not **dkmtom, but the latter is presumed despite my
insignificant dismay simply because of the singular *dekm. My take
on it is that the *t here does _not_ have the same function as the
*t in *dekm-t. It is in my view a plural fossilized since before
the Proto-IndoTyrrhenian stage. The *t is coincidentally found here
whenever we use the plural of the word: *-kom-t-ax, *-km-t-ih,
*km-t-om.
In all, I think that IEists should give up the pursuit of **dkmtom
and accept that *dekm is a suppletive root with a quite different form
for its non-singular, *komt- (zerograde *kmt-). This in turn would
simply be derived from Proto-Steppe *k&m-it (yes, I think I will keep
my "fourth vowel" *& here, rather than my former reconstruction *kumit),
the singular form being just *k&m or alternatively *t?u-k&m.
> One would think that there would be more evidence of a prefixal *de-
> than just in the word for "ten."
Yes, one would. Hence, Etruscan /tHu/ "one". I also wonder about the
origin of IE *dus- which means "bad". If from an aorist verb *deu-s-
meaning "to abandon", we may semantically derive "bad" from "abandoned".
If this is permitted, a connection between *dus- and the numeral for
"one" is possible.
> Furthermore, are you saying here that the *-t in *dekm(t) is different
> from the *-t in *(d)kmtóm?
Yes, the *-t in *dekmt is indicating a collection, a decad. The *t in
*kmtom has no morphological function in IE itself but, as I say above, it
is probably the fossilized plural and related to *-es (which is
sibilantized in IndoTyrrhenian from former *-it because it is found in
final position).
= gLeN