Re: PIE *a -- a preliminary checklist

At 4:51:57 PM on Thursday, February 14, 2008, Richard
Wordingham wrote:

> --- In cybalist@yahoogroups.com, "Brian M. Scott"
> <BMScott@...> wrote:

>>> --- "Brian M. Scott" <BMScott@...> wrote:

>>>> Because the German glottal stop is predictable.

>> Word stems with initial vowel are automatically preceded
>> by a glottal stop: <aus> [?aUs] 'out', <beirren>
>> [b@'?iR@...] 'disconcert', <enteisen> [Ent'?aIz@...] 'de-ice,
>> defrost' (where <be-> and <ent-> are prefixes). It's
>> fully predictable synchronically; some analyses don't
>> even count it as a phoneme in standard German, for just
>> that reason.

> The prediction depends on identifying morpheme or syllable
> boundaries.

Which is part of why I wrote 'synchronically'.

> What happens when morpheme boundaries become obscured as
> words go out of use?

Or when a compound is no longer perceived as such. To the
best of my knowledge, and as your later post seems to
confirm, the glottal stop is lost. It isn't the principle
that changes, but rather the data on which it operates. And
this makes the glottal stop even less plausible as a trace
of old laryngeals.

Brian