Re: History of French

From: Nathan Sanders (nsanders.DIE.SPAM_at_wso.williams.edu)
Date: 09/21/04 

Date: Tue, 21 Sep 2004 00:44:08 GMT

In article <414F0156.6B93@worldnet.att.net>,
 "Peter T. Daniels" <grammatim@worldnet.att.net> wrote:

> Nathan Sanders wrote:
> >
> > > >In article <414EE7F2.257D@worldnet.att.net>,
> > > > "Peter T. Daniels" <grammatim@worldnet.att.net> wrote:
> > > >
> > > >> What's the physiological explanation for Arabic having [k] and [x] and
> > > >> [G] but no [g], and [f] and [b] but no [p] or [v]?
> > > >
> > > >(I don't know the facts for the historical development of Arabic, so
> > > >I'm making guesses purely on articulatory and acoustic concerns.)
>
> Jakobson did a lot better in markedness terms!

Feature-based markedness is one area where I diverge from traditional
phonology. It's fine for purely descriptive work, but it's not
grounded in actual measureable reality enough for my taste when doing
analysis.

I find that the feature-based definition of markedness has a number of
problems, primarily that it is neither sufficiently explanatory nor
sufficiently predictive.

For example, [t] is less marked than [c], so in Jakobsonian terms, [t]
must have fewer formal phonological features. But we can't measure
these features in the real world; they aren't present in the mouth or
the speech signal. They're figments of linguists' imaginations.
That's fine if you simply want a notation for describing patterns, but
it doesn't help *explain* those patterns.

Further, [c] very often becomes an affricate or a fricative, while [t]
seems to be under no particular obligation to do so.

This is predicted by an analysis grounded in the real world, because
during the release of a palatal stop, it is harder to move the bulky,
slow-moving body of the tongue fast enough to prevent a fricated
release. In comparison, preventing such a fricated release is much
easier with the more nimble tip/blade of the tongue used for [t].
Thus, the real world predicts that palatal stops will preferentially
fricate over coronal stops.

Jakobsonian feature-based markedness can't fully predict this outcome.
It's questionable whether [t] actually has fewer features than [c],
because it really depends on how the features are defined. If place
of articulation is binary rather than private, then [t] is [+coronal,
-dorsal], while [c] is [+coronal, +dorsal], which is the same number
of features, and thus, equal markedness.

There is also no obvious inherent relationship between the
phonological place features that define palatals and thephonological
manner features that define affricates. Why should palatal stops
become affricates, when there is nothing in the feature system that
unites them to the exclusion of, say, coronals and velars?

Even worse, affricates are usually treated as more marked/complex than
stops, so it's not clear why stops would ever become affricates at all
if you ignore phonetics.

> > > In Arabic, old /g/ became a voiced palatal stop, later an
> > > affricate or fricative. It's usually transcribed as <j>.
> >
> > Ah. That development makes sense, as a voiced palatal stop eases the
> > pressure problem of a voiced velar stop (and the further development
> > also makes sense, as palatal stops are hard to keep sufficiently
> > closed to avoid frication, due to their larger area of closure).
>
> Yet Gmc. [g] doesn't become [dZ]. (And Ar. [k] doesn't become [tS].)

This is just a phonological conspiracy: multiple solutions for a
single problem. The high air pressure behind voiced velar stops could
be releived in multiple ways: allowing air to leak out orally (g > G),
allowing air to leak out nasally (g > N or Ng), increasing the size of
the post-lingual cavity (g -> j), or devoicing (g > k).

Nathan 

-- 
Nathan Sanders
Linguistics Program       nsanders@wso.williams.edu                           
Williams College          http://wso.williams.edu/~nsanders
Williamstown, MA 01267

Relevant Pages

  • Re: History of French
    ... > Feature-based markedness is one area where I diverge from traditional ... > phonology. ... > must have fewer formal phonological features. ... > fricate over coronal stops. ...
    (sci.lang)
  • Re: help wanted: phonology
    ... Changing the phonological features of a segment, ... does he suggest that assimilation of the native ... in coronals in English ... between "lexical" and "post-lexical" phonology. ...
    (sci.lang)
  • Re: Phonemes
    ... Then that is usually covered in modern phonology under "distinctive ... You can't determine which features are distinctive if you ... that have a third high vowel prefer a front round vowel rather than a ... The only times new languages had to get ...
    (sci.lang)
  • Re: Whats so archaic about Lithuania?
    ... phonology, morphology, and lexicon which are quite radical. ... > pure unaltered Elizabethan English!!) ... respect to certain features and subsystems, ...
    (sci.lang)
  • i bought 2007 why wont it convert?
    ... i've tried upgrading from the trial offer, after purchasing a hard copy. ... key code was accepted, and that's where everything stops. ... features and when i save, i am still saving in '2003' not '2007'. ...
    (microsoft.public.officeupdate)
Quantcast