Re: "b" versus "g"

--- In cybalist@..., "Richard Wordingham" <richard.wordingham@...>
wrote:

> --- In cybalist@..., "tgpedersen" <tgpedersen@...> wrote:
> > --- In cybalist@..., "Richard Wordingham"

<richard.wordingham@...>

> > wrote:
> > > --- In cybalist@..., alexmoeller@... wrote:
> > > > I have in romanian "bine"= good.
> > > > But I have in sub dialect of romanians the same word whichi is
> > > > pronounced "ghini"
> > > > And here I wonder. I should understud if tehre should have
> > > > been a "vini" a "pini" a labial one .
> > > > But ghi? How can be such kinds of divagations explained in the
> > > > same language at all?
> > >
> > > This change, particularly in voiceless consonants, has already
> been
> > > described in

http://groups.yahoo.com/group/cybalist/message/15344

> > and
> > > http://groups.yahoo.com/group/cybalist/message/15380 . Indeed

the

> > > latter describes it as a reversal of the change kw > p, though
> that
> > is
> > > not strictly true.
> > >
> > >
> > > Richard.
> >
> > That's what I'd call a shibboleth-induced retrograde development.
> ;-)
>
> Not advisable.
> Latin /kw/ > standard Romanian /p/ before o, u and sometimes a.
> Latin /kw/ > standard Romanian /tS/ before e, i.
> Standard Romanian /p/, /t/ & /k/ > Dialect /kj/ before e, i.
>
> As you see, the change is hardly ever, if at all, 'undone'. I'd

save

> you shibboleth idea for where it makes some sense.
>
>
> Richard.

But are the phonetic developments reasonable? /p/ > /kj/? /t/ > /kj/?
Nah. They are retrograde. Some of the other Romance language follow
the satem languages (before e, i) in going kW > k, so I'd start
from /k/. Thus in dialect Romanian, before e, i, /k/ > /kj/, /t/

> /tj/ > /kj/, /p/ > (by shibboleth relation) /k/ (or /k/ survived in

this dialect) > /kj/.

Torsten