[tied] Re: PIE Word Formation Q&A (1)

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> On 2006-04-04 17:05, Rob wrote:
>
> > Yes, "phonological" is what I meant. :) By "oral stops", do you
> > mean non-bilabial stops?
>
> No, oral stops are just "normal" stops as opposed to nasals. In
> phonetics, [m], [n], [N] etc. also count as stops, since their
> articulation involves the complete closure of the oral tract

Ah yes, you are correct. Sorry for the confusion.

> > On another note, I don't understand how such a
> > rule could have worked when there are both e.g. *plhnós and
> > *plhtós attested in descendant languages. Or do you think that
> > the latter was not native to IE, but merely a subsequent
> > innovation?
>
> *pl.h1-nó- is widely attested and certainly PIE. *pleh1-to- (sic!)
> is of limited distribution and looks like a late formation. The noun
> *pleh1-tu- > *ple:tHu- 'crowd', which apparently underlies Gk.
> ple:tHu:s and Lat. ple:b(e:)s has *-t-, but it's a separate
> formation, not a verbal adjective.

What is the distribution of *pleh-to-?

I'm not sure if I can agree with your analysis of *pleh-tu- being the
origin of Gk. _ple:thu:s_ and Lat. _ple:b(e:)s_. That is another
topic, however.

> > Unfortunately, it appears that the analysis of the origin of the
> > *-nó- and *-tó- participles strains the bounds of internal
> > reconstruction.
>
> Why? It would strain the comparative method but seems to me to be
> entirely legal as far as internal reconstruction is concerned.

I think it approaches the point where there is no decisive evidence to
prove or disprove an assertion.

> > With English, I think the difference can be attributed to the
> > placement of stress, i.e. prolónging vs. prolongátion.
>
> No, cf. <longer> with /Ng/ vs. <longing> and <longs> with /N/. The
> variation is sensitive to the nature of the morphological boundary,
> not to stress and not merely to the phonetic environment.

There is also, as Pete Gray points out, contrasts such as "singer"
/siNr=/ vs. "finger" /fiNgr=/. To be honest, I haven't thought about
such contrasts before. Do you have any idea where they came from?

> > There are also
> > some dialects of English (for example, Southern American English),
> > that pronounce <ng> as [Ng] whenever it is not word-final (and, in
> > some areas, even word-finally).
>
> Also in the British West Midlands and Middle North, roughly from
> Birmingham to southern Lancashire. But the whole point of my
> illustration is that a language may well treat different boundaries
> in different ways.

What would cause such a phenomenon, however?

- Rob