64 Advances in spoken discourse analysis
A’s comment in the way A intended but then responded on the basis of
some assumed understanding which in fact was not accessible to A. Yet a
further possibility is that B has exercised his option not to reply to the
initiation. In a situation where both participants were fully aware of the
structural implications of their own and each other’s actions, B could simply
have decided that, before pursuing the matter of Friday’s meeting, there
were other matters to consider. His reinitiation which ignores A’s initiation
might under some circumstances be considered rude, but his would depend
on their relationship.
This brings us to the second point: if a speaker’s behaviour is heard as
deviant the deviance can be most satisfactorily characterized as deviance
from a social norm. This is popularly recognized in the use of labels such
as ‘rude’, ‘evasive’ and ‘eccentric’. It is worth noting that, when speaker A
fails to recover any coherent relationship between the two components of a
pair like
11 A: Will you come for a drink?
B: My brother’s just left for the States
his analysis will reflect, among other things, his knowledge of B’s manners,
his drinking habits, even his state of mental health. As a linguistic event the
latter’s contribution simply represents one of the set of options open to him
at this point in the discourse. What a competence/performance dichotomy
might separate out as an ‘error’ must be regarded at the level of discourse
as an event which has its own meaning, the latter being characterized not
in terms of whatever judgements A may be induced to make of B but in
terms of the prospective constraints that now apply to any rejoinder A might
make.
Thus we are not arguing that interaction has no structure, but rather that
the structural framework operates by classifying each successive discourse
event in the light of the immediately preceding one.
FURTHER OBSERVATIONS ON EXCHANGE STRUCTURE
The definition of the exchange
Sinclair et al. (1972) defined the exchange as ‘the basic unit of interaction’
and we see no reason to disagree with this. It is basic because it consists
minimally of contributions by two participants and because it combines to
form the largest unit of interaction, the transaction. Sinclair et al. further
suggested that there were three major classes of exchange, eliciting, directing
and informing, whose initial moves function respectively to request a verbal
response, to require a non-verbal response and to provide new information
(in the most general sense of information).
This description obviously makes a very powerful claim about the nature
of interaction, that there are only three basic types of exchange, a claim