LINGUIST List 3.522

Tue 23 Jun 1992

Disc: Things Phonological

Editor for this issue: <>

Directory

1. "Norval Smith, RE: 3.494 Natural Phonology, English Stress, Macintalk
2. Joe Stemberger, Re: 3.510 Natural Phonology, Comparatives, Predicates

Message 1: RE: 3.494 Natural Phonology, English Stress, Macintalk

English Stress

Who says that all speakers have penultimate stress on words in -ory or -ary.
All Americans maybe, but there are a lot of other English speakers in the
world too.

DYsent'ry ORdin'ry DICtion'ry conSERvat'ry congratuLAtry conserVAtry
 Noun Adjective

RP speakers (are supposed to) pronounce these words:

DYsentry ORdinry DICtionry etc.

The addition of -ly adds an extra unstressed syllable, so your normal RP
speaker is actually supposed to say:

conGRATyultry & conGRATyultrily with pre-pre-antepenultimate stress.

In my (Scottish Standard English) dialect you can have stress-shifting to a
limited extent with -ly, so:

ORdin'rily or ordiNArily

NORval SMITH

Message 2: Re: 3.510 Natural Phonology, Comparatives, Predicates

In 3.510, Rick Wojcik states:

>>...in the generative world, ... theory is grounded in intuitions about
>>well-formed linguistic structure. Natural Phonology grounds itself
>>in actual speech production.

It is a reasonable observation about most syntactic theories that they are
based on intuitions. I don't think that this has ever been true about
phonology. Phonological theory has OFTEN dealt with low-level phonological
processes about which naive speakers (and even many trained ones) don't
have clear intuitions, such as the leftward spread of nasality to vowels
and approximants in a word like LAWRENCE. In our methods for getting
phonological data from languages other than our own, we mostly ask our
consultants to say particular words or sentences. There is SOME place for
asking "can you say this", but it is a relatively minor part, and not all
studies include it.

Phonologists often don't work with low-level phenomena because they are
very difficult to collect data on, and often include things of such low
amplitude or short duration that it is hard to figure out what is going on.
Interestingly, Stampe is one of those phonologists who stand out in this
regard, because he believes (or at least did so in the 1970's) that we
should use our intutions about well-formed fast-speech/extremely colloquial
variants to pronunciation, and that these intuitions are reliable. Many
of us think that these variants really need to be studied instrumentally,
if we want reliable data. But I think that most phonologists would accept
instrumental data about what is going on (acoustic, electropalatographic,
etc.) and feel responsible for accounting for it, AS LONG AS they were
convinced it was phonological in nature, and not e.g. motor.

---joe stemberger