Fwd: <OT> Dissertation update

[Elizabeth J. Pyatt]

Hi everyone,

I have posted an update of my May 1997 dissertation on the Rutgers
Optimality Archive at ROA-196-0597.  Some people had difficulty with
the fonts in the original version; that should no longer be a problem
as this is a PDF file with embedded fonts.  Also, the original version
did not include the map graphic showing the location of the Goidelic
dialects; this new version does include the map.  Otherwise, only a few
minor typos have been corrected, but nothing of content has been
changed.

To go there directly, click on
http://roa.rutgers.edu/view.php3?roa=196

The original abstract follows:

This dissertation is an examination of the prosodic structure of the
closely related Goidelic languages: Irish, Scots Gaelic, and Manx.
Several important claims about the prosodic hierarchy are made, using
facts of stress placement, weight-to-stress effects, and
syllabification. Evidence from non-Goidelic languages is brought to
bear as well.

The approach is both synchronic and diachronic; the theoretical
underpinnings are those of prosodic phonology and Optimality Theory. A
theory of how phonological change is to be captured in an Optimality
Theoretic framework is presented: it is argued that a phonological
change happens when a constraint against a marked phonological pattern
is promoted above other constraints. Further, it is shown that paradigm
leveling can be accounted for within OT by means of faithfulness
constraints governing related output forms.

The continuing role of the Weight-to-Stress Principle (WSP) in the
history of the Goidelic languages is examined. It is shown that the WSP
has had a recurring effect on the prosodic development of Old Irish
from Proto-Insular Celtic and on the evolution of Old Irish into Middle
and Early Modern Irish, and thence to the modern Goidelic languages.

It is further argued that a prosodic constituent called the colon must
be included in the prosodic hierarchy between the prosodic word and the
foot, with evidence from both Goidelic and non-Goidelic languages that
certain facts of stress and prosodic size cannot be explained
adequately without reference to the colon. In particular, it is shown
that the so-called 'forward stress' pattern of Munster Irish, East Mayo
Irish, and Manx are most insightfully explained with the colon.
Finally, syllabification of consonants and consonant clusters is
reviewed, with an argument that a requirement that stressed short
vowels be in close contact with a consonant results in ambisyllabicity
in Irish. The syllabification of rising-sonority consonant clusters is
examined, and it is shown that shallower rises in sonority are
permitted only at higher levels on the prosodic hierarchy; also
examined is epenthesis in Irish and Scots Gaelic into clusters of
falling sonority.

Yours,
Antony Green

========================================================================
Antony Dubach Green                            green at ling.uni-potsdam.de
Universität Potsdam
Institut für Linguistik                        Tel. +49 331 9772936
Karl-Liebknecht-Str. 24-25, Haus 35
14476 Golm                                     Fax  +49 331 9772087
Germany

              http://www.beepworld.de/members4/antonygreen/




--