210 Advances in spoken discourse analysis
progression, in which situationally appropriate discourse is generated by
both parties to the interaction in response to the assumed communicative
need of the moment. All of which is to say that it treats spoken discourse
as a real-time ‘happening’, as a linearly organized sequence of events, the
details of whose organization beyond the moment of utterance may as yet
have not been determined by the speaker.
Now it is evident that there are problems in relating this set of concepts
to the act of reading aloud. Most obviously, we read aloud from a prepared
text, so there is no question of our being engaged in the generation of
here-and-now appropriate utterances. The fact of working with such a text
crucially modifies the open-ended nature of the speaker’s activity: far
from ‘happening’ piecemeal along the time dimension, the material we
read and convert to speech ‘exists’ as an already completed object and, in
certain acts of reading aloud at least, the reader’s apprehension of its
completeness —of the point to which everything is leading—must surely
be one of the things that determines how it is read. Even more important
for my present concerns is the more complex way in which the notion of
a speaker/hearer relationship must be applied if it is to be useful in a
discussion of reading. It is now quite usual to speak of reading as an
‘interactive’ activity, but what people seem usually to have in mind is
some kind of ‘interaction with the text’, an idea which differs importantly
from that of ‘interaction with a listener’. Arguably, preoccupation with
reading as a silent, private activity has favoured the former notion. I am
not, at present, concerned with this. My starting-point is the observation
that, in spite of the many differences between the circumstances in which
spontaneous speech and reading aloud occur, both activities utilize the
same repertoire of intonational options, and since these options correlate
with aspects of the speaker/hearer relationship in speech, there is a strong
case for exploring the nature of that relationship in reading.
It is this latter relationship that I propose to focus upon, and which I
shall make the basis of the differences I shall try to point up among the
various ways in which we can read aloud. In doing so, I shall make
spontaneous interactive speech my point of comparison. Roughly speaking,
I shall be seeking to identify different kinds of reading by asking how far
they resemble speech in those respects which affect their phonological
form. It may reasonably be objected that this procedure is flawed at the
outset, for there are at least as many different ways of speaking as there
are of reading aloud. Nevertheless, it is possible to work on the basis of
a centrally conceived phenomenon called ‘interactive speech’ in which
participants consistently and in good faith orientate to each other’s supposed
view of the relevant circumstances surrounding the communication—the
‘common ground’ of the Discourse Intonation conceptual framework. If
we call this complex of circumstances the context of interaction, then we
can call the speaker’s propensity to take those circumstances into account
his or her engagement with the context of interaction. In the unlikely—