The significance of intonation in discourse 37TOWARDS AN INTERACTIONALLY MOTIVATED
DESCRIPTION OF INTONATION
The description of the way intonation functions in discourse that is outlined
below is one on which David Brazil has been working continuously since
- The most recent and comprehensive presentation is in Brazil (1992).
Brazil does not claim, by any means, to be able to handle the way in which
all paralinguistic features carry meaning (nor indeed that they all have interactional
meaning), but he does present a workable description of many pitch phenomena
which is based on sound and explicit principles.
The first principle is that features which are acoustically on a continuum
must be analysed as realizations of a small number of discrete units that
‘form a closed set, defined by their mutual oppositions’ (Labov and Fanshel,
ibid.:42). The second principle is that there is no constant relationship between
particular acoustic phenomena and particular analytic categories: it is contrasts
and not absolute values which are important. These two principles are not,
of course, novel and create no problems theoretically or practically, as analysts
of tone languages discovered long ago:
tone languages have a major characteristic in common: it is the relative
height of their tonemes, not their actual pitch which is pertinent to their
linguistic analysis...the important feature is the relative height of a syllable
in relation to preceding and following syllables. A toneme is ‘high’ only
if it is higher than its neighbours in the sentence, not if its frequency of
vibrations is high.
(Pike 1948:4)A third principle is that there is no necessary one-to-one relationship
between particular paralinguistic cues and interactional significance: on
the one hand, as Bolinger’s (1964) ‘wave’ and ‘swell’ metaphor suggests,
a given pitch choice can at the very least be simultaneously carrying both
general information about emotional state and a specific local meaning of
the kind described in detail further on in this chapter; on the other hand,
certain interactionally significant signals—for instance, request for back-
channel support—may be carried by the co-occurrence of a particular
pitch choice and a kinesic one, each of which singly conveys a different
meaning.
The final principle is to regard intonation as primarily concerned with
adding specific interactional significance to lexico-grammatical items and
thus enabling the speaker to refine and at times redefine the meaning
oppositions given by the language system. It is for this reason that Brazil
argues that the intonational divisions that speakers make in their utterances
are not grammatically motivated (though for explainable reasons intonation
unit boundaries frequently coincide with major grammatical boundaries);
rather they are motivated by the need to add moment-by-moment, situationally