LINGUIST List 16.1400
|
Mon May 02 2005
Review: Phonology/Phonetics: Hayes, Kirchner & Steriade
Editor for this issue: Naomi Ogasawara
<naomi linguistlist.org>
|
What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Dooley at collberglinguistlist.org.
|
Directory
1. Geoffrey
Morrison,
Phonetically-Based Phonology
Message 1: Phonetically-Based Phonology
|
Date: 02-May-2005
From: Geoffrey Morrison <gsm2ualberta.ca>
Subject: Phonetically-Based Phonology
EDITORS: Hayes, Bruce; Kirchner, Robert; Steriade, Donca
TITLE: Phonetically Based Phonology
PUBLISHER: Cambridge University Press
YEAR: 2004
Announced at http://linguistlist.org/issues/15/15-2908.html
Geoffrey Stewart Morrison, University of Alberta
SUMMARY
Phonetically Based Phonology is a collection of papers which could be
taken as representative of the state of the art at the turn of the
century. A number of the authors completed their doctorates in the
1990s and therefore can be viewed as the new generation of scholars
working in this field. The influence of 1990s mainstream phonology is
apparent in that most of the papers are couched in Optimality Theory
(OT) frameworks, but most of the papers are non-mainstream in that
their approach to phonology could be classified as functionalist.
Ch. 1 Introduction: the phonetic bases of phonological markedness,
Bruce Hayes & Donca Steriade
Hayes & Steriade present an introduction to the book as a whole.
They state: "The hypothesis shared by many writers in this volume is
that phonological constraints can be rooted in phonetic knowledge...,
the speaker's partial understanding of the physical conditions under
which speech is produced and perceived." [p1] Hayes & Steriade go
on to discuss the usefulness of this hypothesis, a key factor in their
opinion being that "the OT search for the right constraint can be
speeded up by the view that markedness is phonetically based". After
some general discussion of OT markedness and previous approaches
to phonetically-based markedness, they discuss the specific example
of stop voicing, relating this to the work of other contributors to the
volume where appropriate. They go on to briefly discuss the other
papers in the volume, in terms of markedness scales of perceptibility,
markedness scales of articulatory effort, ideas of segment licencing,
and contrast-based constraints.
Ch. 2 A review of perceptual cues and cue robustness, Richard Wright
Wright reviews literature on the perceptibility of acoustic cues,
including the behaviour of the peripheral auditory system. This leads
to an explanation of why, for example, consonant place cues in CV
transitions are more perceptually robust than place cues in VC
transitions, and thus why CV syllables are common. The traditional
sonority hierarchy (based more on distributional observations than
inherent properties, and thus descriptive rather than predictive) can
be replaced by a theory based on the intrinsic perceptibility of different
types of segments, considering both internal and transitional cues to
segment identity.
Ch. 3 Place assimilation, Jongho Jun
Jun picks up on the acoustic-cue perceptibility theories presented by
Wright and combines them with theories of articulatory gestural
overlap. Place assimilation targets segments that have weak
perceptible cues in the target context, e.g., nasals preceding stops. If
inherently slow gestures (e.g., velar and labial compared to coronal)
are the target then gestural overlap with the trigger is less, and
perceptual cues are more likely to be preserved. If the trigger is an
inherently slow gesture then gestural overlap with the target will be
greater and perceptual cues for the target are less likely to be
preserved.
Ch. 4 The topology of rounding harmony, Abigail R. Kaun
Whereas Jun discussed cases in which perceptual difficulty resulted in
the neutralisation of contrast, Kaun examines cases in which
perceptual difficulty leads speakers to find other ways of enhancing
and thereby maintaining contrast. The triggers in rounding harmony
are vowels for which the rounding contrast is difficult to perceive (non-
high, front, and short vowels), and the targets are vowels for which
the rounding contrast is easy to perceive (high and back vowels).
Uniformity of lip posture across intervening segments leads to
harmony for vowels of the same height.
Ch. 5 The evolution of metathesis, Juliette Blevins & Andrew Garrett
Blevins & Garrett specifically address the question of whether
phonetically natural phonological patterns are the result of phonetic
optimisation in individual speakers' grammars, or whether they emerge
as a result of diachronic change -- a kind of evolutionary phonology.
They contrast four approaches: synchronic + non-functionalist
(traditional mainstream phonology), synchronic + functionalist (the
approach implied in most other contributions to the volume),
diachronic + functionalist, and diachronic + non-functionalist (the
approach in favour of which Blevins & Garrett argue). They argue that
metathesis can be the result of diachronic reinterpretation of
perceptual cues to segments, as a result of articulatory gesture
overlap and perceptual factors; but that metathesis is not teleological.
It is not an attempt by speakers to rearrange segments so that the
perceptual cues to the segments are more easily perceived.
Ch. 6 The role of contrast-specific and language-specific phonetics in
contour tone, Jie Zhang
Zhang examines the contexts in which contour tomes can occur, and
concludes that contour tones are not licenced by abstract
phonological structures. Rather, in order to be produced and
perceived, contour tones require a minimum duration and degree of
sonority in the rhyme. It is a physical measure of duration and
maintenance of formant structure that determines whether the rhyme
is able to carry a contour tone.
Ch. 7 Vowel reduction, Catherine M. Crosswhite
/a/ can be reduced to a schwa, or other vowels can be reduced to /a/.
Crosswhite argues that there are two types of vowel reduction: In a
language such as Belarussian, mid vowels /e/ and /o/ reduce to /a/ in
an unstressed context. Contrasts which exist in stressed context are
neutralised, but the perceptual difference between the set of
unstressed vowels /i/, /u/, and /a/ is large, larger than if the non-high
vowels had reduced to schwa. Crosswhite calls this "contrast-
enhancing reduction". In a language like Bulgarian, /a/ reduces to
schwa (neutralising with a phonemic schwa which does not undergo
reduction by raising) and /e/ and /o/ reduce to /i/ and /u/ respectively.
Crosswhite argues that this is due to a "desire to avoid particularly
long or otherwise perceptually salient vowel qualities in unstressed
position." [p204] She calls this "prominence reduction". Russian and
other languages have both contrast-enhancing reduction and
prominence reduction associated with different contexts.
Ch 8. Contrast and perceptual distinctiveness, Edward Flemming
Flemming presents a theory in which perceptual markedness is a
property of contrasts rather than of individual sounds. Whether the
identity of a speech sound is easy or hard to perceive does not
depend on the isolated acoustic cues available for that sound, but on
the degree of perceptual difference between those cues and cues for
other speech sounds. Flemming's Dispersion Theory of contrast is
based on three functional goals:
i. Maximise the distinctiveness of contrasts.
ii. Minimise articulatory effort.
iii. Maximise the number of contrasts.
This requires a combination of syntagmatic and paradigmatic
constraints, and a different approach to assessing well-formedness to
that of standard OT.
Ch. 9 Syllable weight, Matthew Gordon
Gordon argues "that the divergent weight criteria observed in stress
and tone systems largely follow from differences in the phonetic
implementation of stress and tone. ... [However] languages employ
weight distinctions that operate over phonologically symmetrical
classes of syllables, even if this means not exploiting the phonetically
most efficient weight distinction(s)." [p277] The fundamental frequency
of a sound is easier to perceive in sounds that have a harmonic
structure (see also Zhang above), thus vowels are good carriers for
tonal information, resonations are moderately good carriers, and
voiced obstruents are poor carriers. A tonal weight ranking for voiced
segments is therefore V > R > O > zero. Stress was measured in
terms of duration and amplitude, and the phonetic effectiveness of the
weight distinction was assessed as the difference in duration and
amplitude between light and heavy syllables. Combination of phonetic
effectiveness and phonological simplicity resulted in different
constraint rankings for different weight-sensitive phenomena.
Ch. 10 Consonant lenition, Robert Kirchner
Kirchner presents a model of articulatory effort which leads to a
unified theory of consonant lenition phenomena. Certain gestures are
inherently more effortful than other gestures; e.g., making a stop
between two vowels requires greater movement of the articulators and
therefore greater effort than making a non-sibilant fricative between
two vowels. The effort minimisation constraints therefore have an
immutable ranking, and different patterns of lenition emerge by the
ranking of faithfulness constraints relative to the effort constraint
hierarchy. For example, in a formal speaking style the faithfulness
constraints would be more highly ranked relative to the effort
constraints than in an informal speaking style, resulting in less lenition
in the formal style. The inherent effort-based ranking motivates
typological predictions of possible lenition patterns.
Ch. 11 Language processing and the segmental OCP, Stefan A. Frisch
Frisch presents a functional language processing model that accounts
for why languages avoid sequences of similar segments. In neural
network activation/competition models of language processing,
repetitive patterns result in a conflict: "For a segment that has already
been encoded, the node in the network corresponding to that
segment has fired, and that node must be inhibited so that it does not
continue to fire. Simultaneously, for a segment that is soon to be
encoded, the node corresponding to that segment must be excited so
that it is ready to fire at the proper time." [p347]
CRITICAL EVALUATION
I will shirk the task of a critical evaluation of each paper in the volume
and restrict myself to more general comments. The book is
mechanically well produced with few typos, etc.
One of the strengths of the volume is that it provides a survey of
recent approaches to phonetically-based phonology. It is not,
however, a historical survey of phonetically-based phonology. Whilst
earlier work on phonetically-based phonology by researchers such as
Lindblom and Ohala is referenced, this work is not discussed in great
detail. It is also incomplete as a survey of recent approaches since it
lacks a contribution from Paul Boersma, a researcher whose work one
would expect to find in such a survey. In fairness the volume does not
claim to be a comprehensive survey of the field, but it is probably the
closest there is so far.
Both a strength and a weakness is that the volume is a collection of
papers. This means that any individual paper can be read
independently without the need to be aware of the contents of the
remaining papers. However, if one does try to read the volume from
cover to cover, one encounters a certain amount of exasperating
repetition: chapter 3 covers a lot of the same ground as covered in
chapter 2, and chapter 4 covers a lot of the same ground as covered
in chapter 3.
Several of the contributions are complementary and the authors
themselves point out links with one anothers' work. At the same time
there are a variety of approaches taken, Blevins & Garrett standing
out for taking an explicitly different diachronic non-functionalist
position with respect to how phonetics influences phonology. The
majority of authors appear to take a synchronic functionalist approach
but this is assumed rather than explicitly argued. Apart from Blevins &
Garrett's paper and Hayes & Steriade's introduction, the book does
not debate the underlying philosophy of phonetically-based
phonology. The paper by Frisch stand out as not in fact being a paper
on phonetically-based phonology! It is a paper on functionalist
phonology, but it is psycholinguistic language-processing based rather
than phonetically based.
I come to the volume with a stronger background in phonetics than in
phonology, and I believe the book will be of interest to phoneticians;
however, the intended audience appears to be phonologists. The
questions addressed are often new approaches to established
problems in phonology, and relevant data are often secondary data
from a range of different languages. Whilst the theories are well
argued, a phonetician might be hoping for more empirically testable
hypotheses with reports of experiments conducted. Several papers
present detailed arguments with respect to how the phonology is
based on articulatory or perceptual factors, then follow this up with an
implementation couched in OT mechanics. I felt that the latter was
generally unnecessary. The important arguments had been made and
the OT tableaux added nothing to my understanding. They seemed to
be an abstraction which actually took me further away from the
phonetically-based reality that the authors espoused.
ABOUT THE REVIEWER
Geoff Morrison is a PhD student in the Department of Linguistics at
the University of Alberta. His primary research areas are perception of
English and Spanish, and modelling cross-language speech
perception. He is also interested in phonetic and functional
approaches to phonology.
Respond to list|Read more issues|LINGUIST home page|Top of issue
|
|
Please report any bad links or misclassified data
LINGUIST Homepage | Read
LINGUIST | Contact us
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed
on its pages, it cannot vouch for their contents.
|
|
