--- In
cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
>
> On 2009-06-03 11:56, alexandru_mg3 wrote:
>
> > We have three words
> >
> > > > 'to know' --> 'family, clan' in Gothic?
> > > > 'to know' --> 'kinsman, close relative' in Sanskrit?
> > > > even
> > > > 'to know' --> 'son-in-law' in Latvian?
> >
> > and you continously has talked only about the semantism of 'son-in-
> > -law'
>
> Because the rest is unproblematic. 'The known /recognised ones' covers
> both affinity and sonsanguinity. 'The begotten ones' would cover only
> the latter. Old English <cu:þ> (*g^n.h3-tó-) meant 'known' but also
> 'familiar, related' as contrasted with <elþe:odig> 'strange, foreign'.
> Compare also the semantic development of Old English <cy:þ> from
> 'knowledge' to 'homeland' and finally to Modern English <kith> (<... and
> kin>).
>
> Piotr
>
I think that you a joking here....
Their 'dogs' are also "'well' known and 'very familiar'" (so very closed) for that people, Piotr....and they didn't arrived to belong to the root g^neh3- ...and g^neh3- 'to know'-semantism is not applicable only to humans...
Next let's take finally the 'son-in-low' word because it seeems that you forget the g^enh1- links of the other forms ==> you will see at least "g^enh1- influence" :)
'All' the forms are quoted under the word *g^omo-/g^eme- (-hter) deduced based on Skt and Greek forms
But Latin gener HAS AN n- there ('probably under the influence of g^enh1-' (Lubotsky)(sic! :)).
Why not under the influence of g^neh3- Piotr? If 'your' semantism is 'so obvious' and 'mine' is 'so poor'???
Albanian dh&ndh&rr HAS AGAIN AN -n- and I don't remember that -
mt- goes to -nt- in Albanian: Do you?
(and of course a lot of etymologies has been proposed, indicating the root g^enh1- )
Lith zentas HAS AGAIN an n- ('probably on an Analogy with g^enh1-" (again Lubotsky) (sic-2!). Why not under 'the analogy' of g^neh3-, Piotr, if that root is the choosen one, from where 'son-in-law' words could be generated so easy ?
Finally Latv. znuots 'son-in-law' cand be derived with no problems from g^noh1- but again is derived from g^noh3- and is placed "under the influence of Lith zentas" which on his turn is "under 'an analogical' influence of the PIE g^enh1- "
To tell you more directly:
Albanian, Latin, and Lith. forms for 'son-in-law' can goes without any phonetic problem and with no issue regarding their semantism to the root g^enh1- (but in place "currently" each word has "an analogic -n-" and this 'in three distinct languages' (final-sic!))
Marius
P.S. And I will come next to this supposed 'impossibility' CERH/CREH and I will show you a flexion pattern of the same word showing this alternance.