The palatal sham :) (Re: [tied] Re: Albanian (1))

> >Your parody of what I'm proposing is silly alright. I'll be

checking

> >for matching roots containing 'plain velars' in Møller and

Bomhard,

> >not "vowels or consonants".
>
> Are plain velars not consonants? What would happen if you were to

check,

> say, for matching roots containing /m/? Plenty of them in Bomhard

and

> Møller. Would that prove that /m/ was borrowed into PIE?

No, but I was not going for proof. I was saying "might have been".

> >You said (something like):
> >
> >"The existence of reconstructed PIE *kap- "take" proves that PIE

had

> >plain velars"
>
> Indeed it does.
>

Reconstructions don't 'exist'. They are reconstructions.

> >Bomhard 242:
> >PN *k{h][a|&]p[h]- "to take, to seize; hand"
> >PIE *k{h]ap[h]- "to take, to seize"
> >PAA *k[h][a|&]p[h]- "TO TAKE, TO SEIZE; HAND"
> >PFU *kappe- "to take, to seize, to grasp"
> >*käppä- "hand, paw"
> >PD *kapp- "to touch, to feel"
> >PA *kap- "to grasp, to seize"
> >
> >
> >In the face of that, one might propose one of two mutually

exclusive

> >alternatives:
> >
> >1) PIE *kap- is a loanword from some other language
> >2) PIE *kap- is not loaned from anywhere
>
> The alternatives are not mutually exclusive.
>
> The root *kap- is PIE, as it occurs in more than enough branches,

showing

> regular correspondences.

Correct, but only if you assume PIE has plain velars.

>It's definitely not a later borrowing into only a
> part of IE.
>

No? Why not?

> In PIE itself, the word may be inherited (from an ancestor

language, which

> we'll call pre-PIE), or it may be a borrowing. If it's inherited

from

> pre-PIE, then again in pre-PIE it may have been borrowed, or

inherited from

> pre-pre-PIE (let's call that Nostratic). The fact that similar

roots occur

> in other Nostratic languages suggests that the root is inherited

from

> Nostratic,

Or that it is a wanderword. If you believed in the sanctity of these
reconstructions, you would have kept your hands off 'seven'. Once you
start cherry-picking from the heap, as one my learned colleagues
called it, you can't tell other people not to do the same.

>unless you can prove there are phonetic irregularities which
> indicate borrowing from one branch to another instead of

inheritance. You

> can't do that, however, because we know too little about the

*regular*

> correspondences between the Nostratic languages, so any

irregularity you

> come up with may turn out to be regular (and any apparent regular
> correspondence may in fact turn out to be irregular).

I'm not sure you know yourself what you are saying, but it adds up to
something like "all words are inherited".

> If we limit ourselves to PIE, there are plenty of native PIE roots

and

> affixes containing the plain velars *k, *g and *gh.

'are', my foot. Those roots are reconstructed using plain velars.
Their 'existence', as you call it, by categorial mistake, is
ephemeral. Using their 'existence', which rests on the assumption of
plain velars in PIE, to prove the existence of plain velars in PIE is
as circular as an argument will ever get.

>It's completely
> irrelevant whether some of them, or most of them, or all of them,

were

> borrowed by some earlier stage of PIE from languages unknown and

most

> likely unknowable. The only relevant fact for the study of PIE is

that the

> PIE phonological system had a full series of "plain"

velars, "palatal"

> velars, and labio-velars.
>

And now you're just being dogmatic. They exist because I say so. Or
we say so. Or we have always said so.


Torsten