LINGUIST List 13.845

Wed Mar 27 2002

Review: Discourse Anal, Phonetics: Wennerstrom (2001)

Editor for this issue: Terence Langendoen <terrylinguistlist.org>


What follows is another discussion note contributed to our Book Discussion Forum. We expect these 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 discussion." (This means that the publisher has sent us a review copy.) Then contact Simin Karimi at siminlinguistlist.org or Terry Langendoen at terrylinguistlist.org. Subscribe to Blackwell's LL+ at http://www.linguistlistplus.com/ and donate 20% of your subscription to LINGUIST! You get 30% off on Blackwells books, and free shipping and postage!

Directory

  1. David Deterding, The Music of Everyday Speech: Prosody and Discourse Analysis

Message 1: The Music of Everyday Speech: Prosody and Discourse Analysis

Date: Wed, 27 Mar 2002 16:23:40 +0800
From: David Deterding <dhdeternie.edu.sg>
Subject: The Music of Everyday Speech: Prosody and Discourse Analysis

Wennerstrom, Ann (2001) The Music of Everyday Speech: Prosody and Discourse
Analysis. Oxford University Press, xix+317pp, paperback ISBN 0-19-514322-1,
$24.95, hardback ISBN 0-19-514321-3.

David Deterding: National Institute of Education, Singapore

SYNOPSIS
The central theme of this book is that prosody, particularly intonation,
should play an essential role in the analysis of spoken discourse.
Wennerstrom argues that too often prosody seems to be regarded as an
embellishment on the message carried by the words, and instead it ought to
be studied as something absolutely central to the way that the message is
conveyed.

After two introductory chapters that discuss theoretical issues regarding
the representation of intonation, stress, and rhythm, Wennerstrom devotes
one chapter each to: the choice of H and L pitch accents to maintain
coherence; the intonation of discourse markers; speech act theory;
conversational analysis; oral narratives; and second language discourse.

In each chapter, after a theoretical consideration of the issues involved,
there is detailed analysis of some data. Most of these data originate from
Wennerstrom's own speech corpora, but she also includes contributions from
guest writers, with Kathleen Ferrara analysing the intonation on the
discourse marker 'anyway', Philip Gaines describing the use of tag questions
in legal proceedings, Susan Fiksdal considering changes in tempo that occur
at uncomfortable moments during interviews, and Heidi Riggenbach
investigating the way that pauses interfere with the fluency of non-native
speakers of English.

CRITICAL EVALUATION
Wennerstrom aims to represent the intonation of conversational speech and
also academic lectures using the model devised by Pierrehumbert (1980) and
developed by Pierrehumbert and Hirschberg (1990), and she finds it necessary
to modify and augment this model in a number of ways: the use of four pitch
accents instead of the original six, echoing the doubts about the H*+L and
H+L* tones that have been noted in the ToBI system and elsewhere (Ladd,
1996:95); the combination of the phrase accent and boundary tone into a
single boundary, as she finds it hard in practice to maintain a distinction
between phrase accents that are associated with intermediate phrases and
boundary tones associated with sentences; the introduction of an additional
partially-falling pitch boundary to the basic inventory of four boundaries:
high-rising (= H-H%), low-rising (= L-H%), plateau (= H-L%), and low (=
L-L%); the use of initial key, based on the work of Brazil (1997), to
capture how the speaker's attitude relates to the previous utterance; and
the use of paratones to mark major topic switches. All of these
modifications were found necessary in order to capture the intonation of
real speech, and this is the major contribution of Wennerstrom's book: much
of Pierrehumbert's work was based on speech synthesis and the analysis of
constructed examples, and when the model is used for large-scale
transcription of real data, certain adaptations are found to be necessary.
Wennerstrom has provided a substantial contribution to the development of
the model in this respect, by analysing a wide range of data from a variety
of perspectives, and her findings are exceptionally valuable.

This book provides an impressive wealth of painstakingly transcribed data to
illustrate each of the areas. All the examples are carefully explained with
the regular provision of helpful computer-based pitch tracks. It is a pity
that the data is not available for us to listen to, and one wonders if it
might be possible for some of it to be placed on the web or provided in an
accompanying tape. However, it must be accepted that this may not be
feasible, for copyright reasons or considerations of privacy, and the
careful and clear explanation of all the points together with the detailed
transcripts and computer pitch plots do prove to be adequate even if not
ideal.

With such an impressive array of meticulously and lucidly presented data, it
might seem a little churlish to note two areas where the data analysis is
rather thin: the measurement of rhythm, and speech act theory. For the first
of these, Wennerstrom makes a strong case for rhythm contributing to the
success of conversational interaction, but in fact she only provides one
brief example of a waveform with superimposed rhythmic beats (p.54), with no
attempt to measure the durations between the beats in order to determine if
they do actually occur at roughly equal intervals. In fact, Couper-Kuehlen
(1993) has shown that the measurement of rhythmic beats is rather difficult,
as for example it is hard to decide which point in the syllable to measure
from and it is furthermore sometimes necessary to posit two separate
coincidental rhythmic chains, so it is not surprising that Wennerstrom
admits that the alignment of beats in her brief example "would not stand up
to scrutiny with a precise ruler" (p.51). Nevertheless, it is a pity that
some kind of measurement of rhythm was not attempted, even if only to
illustrate the nature of the problem. The contribution from the guest
writer, Fiksdal's analysis of interviews with foreign residents in the USA,
demonstrates clearly that changes in tempo tend to accompany awkward moments
in the conversation, but this could be interpreted simply in terms of
speaking rate rather than changes in the timing of rhythmic beats, so we
really have no extensive investigation of rhythm in the data.

In the second area where the data is somewhat lacking, the chapter on speech
act theory is markedly more theoretical than the others with very little
analysis of real data, though it is probably true that the logical and
philosophical nature of much of the background work in this field lends
itself more to the consideration of constructed examples than real
conversational data. While there is an interesting guest contribution from
Gaines on the loaded use of tag questions by the defence lawyers in the
trial of O. J. Simpson, this does not really get to the core of speech act
theory. One wonders whether an investigation of differences between the
intonation of yes-no questions and wh-questions might not be appropriate
within the domain of speech act theory, particularly as Wennerstrom pays
little attention to any distinction between these two types of question, in
contrast to Pierrehumbert and Hirschberg (1990) who clearly regard them as
quite different. Perhaps Wennerstrom has found in her analysis of
conversational data that there is no real distinction and so any suggestion
that yes-no questions tend to have rising pitch and wh-questions falling
pitch is simply a myth, but it would have been interesting to find out more
about this.

However, these suggested deficiencies are exceptionally minor in comparison
with the impressive wealth of data in such a wide range of areas, with
convincing demonstrations of the semantic roles of the different pitch
accents to introduce new information or refer back to old information and
thereby maintain coherence in conversations, the use of paratones to mark
the onset of new conversational topics, the role of the various initial keys
and final boundaries to maintain links with previous and following
utterances, the use of high pitch to represent quoted speech in oral
narratives, and the effect of hesitations and pauses in interfering with the
fluency of non-native speech. The presentation of the data in all of these
areas is excellent, though ultimately one is left wishing that a whole book
could be devoted to each one.

In summary, this book provides a wealth of fascinating data illustrating a
wide range of intonational phenomena in the analysis of conversations. The
text is always exceptionally clear and it makes a highly convincing argument
that the analysis of prosody should be central to work on discourse
analysis. This book has made an important contribution to the field,
especially as it is written in language designed to make it accessible to as
many people as possible. It is certain to become widely used and highly
valued.

REFERENCES
Brazil, D (1997) The Communicative Value of Intonation in English.
Cambridge: Cambridge University Press.

Couper-Kuhlen, E (1993) English Speech Rhythm: Form and Function in
Every-day Verbal Interaction. Amsterdam: John Benjamins.

Ladd, D R (1997) Intonational Phonology, Cambridge: Cambridge University
Press.

Pierrehumbert J (1980) The Phonology and Phonetics of English Intonation.
PhD Thesis, Massachusetts Institute of Technology, Cambridge MA.

Pierrehumbert, J & Hirschberg, J (1990) The meaning of intonational contours
in the interpretation of Discourse. In P R Cohen, J Morgan & M E Pollack
(eds) Intentions in Communication, Cambridge MA: MIT Press, 271-311.

ABOUT THE REVIEWER
David Deterding is an Associate Professor at the National Institute of
Education, Singapore, where he teaches phonetics, syntax, and translation.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue