> > > Anyway, I wonder what those prefixes meant or what
> > > their functions were (e.g. in *galakt- and *melg).
> > >
> >
> > I'll get back on that.
>
> OK. I'm looking forward to it.
>
Laurent Sagart ('The roots of Old Chinese') on the prefixes:
Prefix N-
(equivalent to Pulleyblank's *a-, in Written Tibetan H-, voiced
laryngeal)
intransitivizes tranitive verbs (probably not applicable)
Prefix m-
in verbs
"The function of the prefix in verbs is difficult to establish. Most
verbs reconstructible with m- appear to be action verbs (however ...
are stative verbs). If we restrict ourselves to the relatively
secure instances of verbs where prefix m- precedes a liquid root
initial [applicable with 'milk', TP], m- verbs express controlled
actions by volitional agents [OK too, if *melg/k- is primarily a
verb]"
in nouns
1) corresponding to m- in verbs: agents (BTW cf Semitic m-)
2) names of small animals
which a cow ain't.
Prefix k-
"Like other prefixes. *k- is most conspicuous with liquid root
initials..."
in verbs of action
Jin dialects: *k-V "to V a little" or "to V for a while" or "to keep
V-ing"
in stative verbs
*k- also derives stative vers (probably not applicable)
in nouns
"Nouns prefixed with kV- in living Chinese dialects refer to
animated as well as non-animated objects. Yet they are always, or
almost always, concrete nouns (as opposed to abstract nouns) and
count nouns (as opposed to mass nouns)"
...
"It is tempting to regard the functions of *k- in verbs and nouns as
being fundamentally one: *k- would serve for actions and objects
that are well-delimited in time and space, and hence usually
concrete and countable."
The countable-noun part doesn't look good for assuming *galakt-
contains this prefix. But cf. Danish
øllet (n.) "the (non-countable quantity, eg. a puddle of) beer"
øllen (common gender) "the (countable quantity, eg. a bottle of)
beer"
In that case *galakt- would mean "a quantity of milk" (the result of
one milking? a sheepskin of?).
>
> > > I'm not sure what semantic shifts could lead to
> > > "milk", but let's consider the following
> > expressions,
> > > more or less connected to the concept of "milk",
> > from
> > > languages that did have linguistic contacts with
> > > Indoeuropean languages, as well as, perhaps, PIE
> > > itself:
> > >
> > > Kartvelian comparanda:
> > >
> > > PKartvelian *.qwel- "cheese"
> > > PKartvelian *lok.- "to lick"
> > > PKartvelian *loq.- "insipid, sweet"
> >
> > > Uralic comparanda:
> > >
> > > Saam (Lapp): lak'ca^ -âvc- (N) "cream; thick sour
> > > cream" (a loan? from where?)
> >
> > Arabic has a root for "lick" something like l-s^-,
> > OC something like
> > *lVk- (by memory).
This may be relevant (from Sagart, p 79)
"Prefix m- ...
shu2 ... *(b)mlut > zywit "glutinous millet" : Proto-Hmong (Wang
1979): *mbl&D "glutinous (rice)"
she2 ... *(b)m-lat > zyet "tongue" : Proto-Hmong (Wang 1979):
*mbleiD "tongue"
shi2 ... *(b)m-lit > zyit "fruit" : Siamese let, m-let, malet "grain"
Schuessler also compared shi2 ... *(b)m-lïk > zyik "to eat" with PTB
(Benedict 1972) *m-lyak "to lick"."
[(b) = superscript b, syllable type indication, D = subscript D)
Hm. Now suppose that *l-k- gloss isn't "lick", but "suckle",
starting as a terminus technicus of bovine domestication? That would
explain its dispersal and also support the idea of its involvement
in the derivation of the "milk" words *ga-lak(t)-, *lak(t)-, *me-lk-.
Torsten