74 Advances in spoken discourse analysis
30 Linus: Do you want to play with me Violet?
Violet: You’re younger than me. (Shuts door)
Linus: She didn’t answer my question.
The same interpretative strategy is used with wh-elicitations: all the items
in the response slot are interpreted as attempts to provide the required
information (although in selecting an interrogative version the speaker can
mark his information as potentially unreliable).
It’s in the cupboard
31 Where’s the typewriter? Try the cupboard
Isn’t it in the cupboard
However, we must note that not all items following wh-elicits are informing
moves. There will be occasions when the second speaker chooses to produce
an eliciting move, i.e. an R/I instead of an R, which simultaneously provides
potentially unreliable information, and asks, through the meaning carried
by high termination, the original questioner to confirm whether the offered
information is in fact correct or not:
32 A: Where’s the typewriter?
B: //p ISN’T it in the CUPboard //
A: //p NO //
The limits of the exchange
In the earlier Sinclair and Coulthard version of exchange structure each
move class could only occur once and thus exchange boundaries were rarely
problematic. However, it has now been claimed that two eliciting moves can
occur in the same exchange and it will soon be suggested that two informing
moves can also co-occur. How then can one recognize an exchange boundary?
We argued earlier that the exchange is the unit concerned with negotiating
the transmission of information and that it will contain an informing move
at I or R. We now want to argue that the exchange only carries one
(potentially complex) piece of information and its polarity, and that the
information and the polarity can only be questioned and asserted once. As
just presented it looks as if we are using semantic and not structural
criteria, but in fact we can support and exemplify our claims structurally,
for the power of the exchange is that as one progresses the available
options decrease rapidly.
Before we go any further we must subdivide both eliciting and informing
moves into two subclasses:
e 1 eliciting moves which seek major information
e 2 eliciting moves which see polarity information
i 1 informing moves which assert major information
i 2 informing moves which assert polarity information