Patrick Ryan pisze:
> Again, absolutely right.
>
> But I think that the process often leads to overcomplication.
>
> I must say, you really know how to select out the most problematical
> examples.
>
> Your analysis above is greatly appreciated. It firmed up some things in my
> mind. I see the ultimate *CVC root as *HAyA-, 'wet', to which -*wA was
> added -> *Hyéw- -> *yéu- = 'wetted all over', referring to the state at
> birth.
The funny thing about etymology is that it turns words into fractals. As
long as you manage to "magnify" them, you see more and more internal
complexity rather than the final Lego blocks. For example, <young> [jVN]
doesn't decompose into CV-type "proto-words" but rather into roots of
the same order of surface complexity as <young> itself. The first
element of the compound, *h2ju-, is the zero grade of the neuter noun
*h2oju/*h2aju- 'vital force' (triconsonantal), the second is the
Hoffmann suffix (originally perhaps a root noun, possibly also
triconsonantal). Only the adjectival suffix *-k^o- is simpler, but who
knows what it was "originally". My microscope can't blow it up any further.
Piotr