On 2008-02-19 22:29, Miguel Carrasquer Vidal wrote:
> Isn't it obvious that <-skir> "father-in-law" and <-skus>
> "mother-in-law" (the - implies that a possessor prefix is
> required) are straightforwardly derived within Burushaski
> itself from <hir> "man" and <gus> "woman" (the same would
> seem to apply to <gus> => <yugus> "daughter")?
It is indeed, and Dick Grune even says so in his description of
Burushaski. With a little creative imagination it's easy to prove that
just about any language is IE, provided that one is satisfied with two
or three examples per proposed change and still admit irregularities and
a liberal dose of "special treatment". E.g. Sean's *tw/dw > lt looks
promising (though *dHw- > b-) when one looks at *dwo: > alto and
*kWetwo:r > walto; but *twe- > go- doesn't fit, so special treatment of
pronouns "after another word" is invoked (though 'thou' is <go-> also
sentence-initially). *kWe- supposedly becomes w- in <walto> but k- in
<ka> "prob. ... depending on if preceded by V". _Probably_? When there
are just two examples, each with a different outcome, and not a shred of
evidence that either of them occurs more frequently with a preceding vowel?
Piotr