Re: Portuguese, Spanish bode "buck"

 

--- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy <bhrihstlobhrouzghdhroy@...> wrote:
>

> 2013/4/2, dgkilday57 <dgkilday57@...>:

> >
> > --- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
> > <bhrihstlobhrouzghdhroy@> wrote:
> >>

> >> 2013/3/29, dgkilday57 <dgkilday57@>:

> >> >
> >> > --- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
> >> > <bhrihstlobhrouzghdhroy@> wrote:
> >> >>
> >> (...)
> >> >> For beccus I had already proposed, following the suggestions by
> >> >> Delamarre 2003: 70 and 80, a root *bek- (or maybe *gWek-) 'sting',
> >> >> unless *bekko-s < *gWet-ko-s (cf. *gWet- 'bulge', Pokorny 1959: 481).
> >> >
(...)
> > DGK:
> > A more serious objection is that if *-ko- was productive with bare /e/-grade
> > roots, there should be no shortage of *-ko-formations with such roots having
> > different auslauts, not just those in *-k- or other stops expected to
> > assimilate to a suffixal *k-.
> >
> > Matasovic' refers Celt. *balko- to PIE *bel-, which I do not follow. It
> > seems to me that he implicitly assumes a laryngeal root-extension (and I
> > have no problem with *-h1 or *-h2) and zero-grade. Of course, *bHelh1/2- or
> > *gWelh1/2- would work equally well. But for your desired /e/-grade we must
> > manufacture a root *bh2/4el-, *bHh2/4el-, or *gWh2/4el-.
>
> *Bhr.: You know I prefer long /o/ grade with Osthoff's shortening

Yes, but your only argument in favor of /o:/-grade in such words was that I could not disprove it.

 

 *Bhr.: it's the received view as well; anyway, can You explain *balko- with a better (please note "better") reconstruction?

 

> >
> > (...)
> > Anyhow, your theory of Celtic tenues geminatae needs to be checked for
> > plausibility against the frequency of parallel formations from roots which
> > do not produce geminates with the same suffixes.
> >
> > DGK
> >
> *Bhr.: Handbooks give Russian poperek 'transversal' (Old Church
> Slavonic pre:kU) < *per-ko-s and Greek-Indic isogloss *dheh1-ko-,
> -kah2 'receptacle' > Old Indic dha:k'a-s, Greek th'e:ke: (unless You
> prefer a lengthened grade *dhe:h1-ko-!).

The Slavic words could just as well continue *per-kWo-s.

 

 *Bhr.: this implies You admit that *-kWo- suffixes can be added to e-grade roots and that only *-ko- cannot, doesn't it?

 

 

Latin _facio:_, _fe:ci:_, _factum_ shows that *dHeh1- took a /k/-extension, and the Greek-Indic noun could just as well be built on *dHeh1k-.

 *Bhr.: is there a sequence where a suffix *-ko- can be recognized as such instead of either a possible root-enlargement or part of a suffix conglomerate?

 

(...)
> An instance beyond any doubt is represented by Old Icelandic brj'oskr
> 'cartilage', Swiss German briesch 'colostrum', maybe Albanian breshkE
> f. 'tortoise' (unless one accepts */ou/ > /e/ instead of /a/ in
> non-metaphonetic environments) from root *bhreus- 1 'swell' (Pokorny
> 1959: 170-171).

Koebler cites only a strong nt. _brj'osk_ 'Knorpel', Gmc. *breu(t)skam.

 

 *Bhr.: right, my fault in copying; I'm very sorry and beg Your pardon. On the other side, */t/ is by no way assured

 

 

The root could be the simplex *bHreu- 'spriessen, schwellen' (Pok. 169), making the IE formation *bHreu-skom, not *bHreus-kom.

 

 

  Verb stem *brūsana-n suggests indeed root *bhreus-

 

> I think one single uncontroversial case is sufficient to prove that a
> R(e)-ko- scheme is possible (...)

(...)
As for uncontroversial examples of *R(e)-ko-, there may be some, but I had in mind a statistical analysis of such purported forms within Celtic. Something may be possible but highly improbable.

 *Bhr.: if You have checked the whole material, please let us know it; if You don't, "highly improbable" only expresses an impression of Yours

 

 

> Note also that this case has been made for *gWet-ko-s, which I added
> as a merely prudential alternative to *bek-n'o-s with Stokes-Zupitza's
> Law; this means that if I weren't be able to justify R(e)-ko- my
> reconstruction *bek-n'o-s would become relatively stronger

Assuming you could justify *bek- in the first place, as well as Zupitza's formulation of Stokes' Law from a sufficient number of other examples.

DGK

 

 

 *Bhr.: I can't understand how You explain preserved sequences of media and nasal in Celtic if Your "MacBain's Law" predicts medias geminatas if the accent was before the cluster and Your version of Stokes' Law tenues geminatas if the accent was after the cluster