Publishing Partner: Cambridge University Press CUP Extra Publisher Login
amazon logo
More Info

New from Oxford University Press!


Cognitive Literary Science

Edited by Michael Burke and Emily T. Troscianko

Cognitive Literary Science "Brings together researchers in cognitive-scientific fields and with literary backgrounds for a comprehensive look at cognition and literature."

New from Cambridge University Press!


Intonation and Prosodic Structure

By Caroline Féry

Intonation and Prosodic Structure "provides a state-of-the-art survey of intonation and prosodic structure."

Review of  When Listeners Talk

Reviewer: Colin Fraser
Book Title: When Listeners Talk
Book Author: Rod Gardner
Publisher: John Benjamins
Linguistic Field(s): Pragmatics
Issue Number: 13.1808

Discuss this Review
Help on Posting

Date: Sun, 23 Jun 2002 22:56:45 +0100 (BST)
From: Colin Fraser
Subject: Gardner, When Listeners Talk

Gardner, Rod (2002) When Listeners Talk:Response tokens and listener
stance. John Benjamins, Hardback: ISBN: 1 58811 093 1, Pages: xxii, 281, Price: USD 89.00
/ Hardback: ISBN:90 272 5111 8, Price: EUR 98.00


This is a well written and thorough account of an often neglected area
of linguistic research - mono and bi-syllabic response tokens (such as
"hmm", "uh-huh") uttered by participants in dialogue.


Gardner's introductory chapter makes clear his belief that much
linguistic research has focused too much on the producer of language
and not enough on the activities of the listener of these productions.
He outlines those who he feels are the culprits and then distinguishes
them from those who, in his opinion, have championed the listener's
cause - persons from the area of sociology (Goffman),
psycholinguistics (Clark) and conversational analysis (Schegloff and
Sacks). Indeed it is the methods of this last group he himself
adopts. He goes through each of the types of activites that
participants who are in the role of listener can perform: Continuers
(like "Mm hm", which hand the floor back to the immediately prior
speaker); Acknowledgments (such as "Yeah", which claim agreement or
understanding of the prior turn); Newsmarkers (like "Oh", which mark
the prior speaker's utterance as newsworthy in some way); Change of
Activity tokens (such as "Alright" which mark a transition to a new
topic in the talk); Assessments, which evaluate the talk of the prior
speaker; Brief questions, which are attempts to clarify or seek
repair; Collaborative Completions, where one speaker completes the
previous utterance of the other speaker. It is the first four of
these (bagged together under the "semantically transparent" term of
response tokens) which are the focus of his study, principally because
these have been either generally ignored out of hand by previous
researchers as "trivial" and thus unworthy of study, or lumped
together into the catch all category of "backchannel" which neglects
to capture the observable differences in how each response token may
be used.

In the second chapter Gardner takes us through 8 common response
tokens and their common usages. He begins by setting himself a series
of open research questions in the area, emphasising that response
tokens should be interpreted according to their placement within a
sequence of talk and that prosodic features are especially important.
The function of this chapter is to introduce the reader to the various
uses response token can have within dialogue. However, due to their
flexibility, they can appear to be "semantically weak, if not empty",
making his task a difficult one. One of the ways he devises to get
round this problem is to look at intonational contour, something he
develops in later chapters of the book, offering interpretations of
various tokens according to prosodic features. He offers a series of
theses concerning the function various tokens are performing, arranged
according to the four categories mentioned in the last paragraph.

Conceding that it is impossible to provide an exhaustive
interpretation of all response tokens, Gardner decides to look at
"Mm", which becomes the token of choice for the rest of the book. His
reason for this is that it has been fairly neglected and the fact that
it has so many apparent uses. He also posits a few ideas why this
might be case - such as the fact it is the only sound in English which
has the mouth closed from onset to termination. Chapters 3 and 4
offer us eight types of "Mm". In Chapter 3, he offers us the
non-response types - such as being used for lapses in the
conversation, expressing pleasure, hesitiating, being used to intiate
repairs and being used to answer someone. In Chapter 4, Gardner tries
to distinguish "Mm" in its response token capacity from "Mm hm" and
"Yeah" which, he maintains, have slightly different usages. Whereas
"Mm hm" is what he calls a "classic continuer", in that it offers the
floor back to the speaker to whom the token is orientated, "Yeah" has
more of an affirmative role. "Mm" can have both these roles and
Gardner argues that by observing the intonation contour of a
particular token one can infer its likely response role.

The very "weakness" of "Mm" is covered in Chapter 5, where Gardner
claims that the utterance of this token does not demonstrate any
particularly strong commitment to the topic. It is thus a neutral
token in his view, and he defends this with recourse to several
examples where we can see "Mm" being used by a speakers who do not
necessarily take up the opportunity to say something about the topic
of the previous turn but do take the chance to take the turn. Indeed
when it is used in the "turn initial" position, that is at the start
of turn, its primary function is merely to demonstrate receipt of the
utterance. It rarely occurs, in Gardner's analysis, after an
agreement or an assessment of what has been said.

After his analysis in the previous chapter of "Mm" as primarily an
acknowledgement token, Gardner considers in Chapter 6 how "Mm" can
fluidly change function according to prosodic features. His analysis
focuses on the three most frequently occuring intonational mappings -
falling, fall-rising and rise-falling contour. The first of these
occurs in about 70 per cent of "Mm"s in Gardner's corpora, and it is
this which performs the acknowledging role. He claims that when it is
uttered it merely says that the turn constructional unit to which it
is oriented has been received without problems. "Mm" with a
fall-rising contour has the effect of turning the acknowledger into a
continuer, even though it retains some "acknowledging" aspects. In
this case, it seems to be making a claim that the turn constructional
unit to which it is oriented is incomplete and is urging further talk
by the other speaker. The rise-falling "Mm" has some aspects of the
falling "Mm" but demonstrates a heightened sense of involvement in the
talk, and has more features in common with an "assessment".

In his final chapter, Gardner outlines some future directions he sees
for work in this area. After his own analysis of the prosody of "Mm"
he encourages future work on the other tokens which he identifies in
his second chapter along these lines. What role do regional
differences play in the function certain tokens perform? And what
about individual differences in the usage of some tokens?


Gardner's analysis of response tokens is certainly thorough, often
stimulating and provides some facinating insights into the various
functions these can perform in conversation. He acknowledges himself
that the categories he has conjured up for his task may well be
insufficient to capture the variety of uses of each of his selected
tokens. However, his achievement in this book in all the more
admirable in that he takes the effort through detailed and very often
illuminating analysis of the conversational situation to tease out the
subtle differences in use of each token. The second chapter stands
out as an excellent review of the much neglected area of research into
response token and Gardner's thorough analysis of "Mm" will set the
standard for future work in this area.


ABOUT THE REVIEWER Colin Fraser is presently a researcher in the Division of Informatics, Edinburgh University, where he recently graduated with a Masters in Cognitive Science. His research interests are focused around different approaches to modelling dialogue, although he presently works in the rather different area of information retrieval.