A functional description of questions 99to syntactic form rather than expected response. The three major classes of
questions that they propose are in fact based on surface form. Even when
they do look at the expected response, it is often the form of the response
that is being attended to rather than the function or the communicative
choice realized by the response.
‘QUESTION’ AS ‘ILLOCUTIONARY ACT’
Let us now look at the characterization of ‘questions’ as illocutionary acts.
Lyons (1977) characterizes ‘question’ as an utterance with a particular
illocutionary force. He asserts that the difference between a question and a
statement is that the former contains a feature of doubt and that one of its
felicity conditions is that the speaker should not know the answer to his
question. He asserts that although questions are normally associated with
the expectation of an answer from the addressee, this association is conventional
and is independent of the illocutionary force of the question. He argues that
this analysis of questions enables us to subsume various kinds of rhetorical
questions instead of having to treat them as abnormal or parasitic upon
information-seeking questions (see p. 755). The inconsistency of this
characterization of ‘question’ can be seen from two objections that I shall
raise below.
Firstly, if the expectation of an answer is independent of the illocutionary
force of a ‘question’, then there is no need to differentiate the following
two sentences:
27 Is the door open?
28 The door is open, isn’t it?
In both sentences, the speaker expresses doubt as to whether the door is
open. Yet, Lyons distinguishes between the two by pointing out that a sentence
like (28) ‘puts to the addressee the positive proposition p (which the speaker
is inclined to believe to be true and assumes the addressee will accept), but
at the same time explicitly admits in the tag the possibility of its rejection’
(p. 765) and that the function of the checking tag is ‘expressly to solicit the
addressee’s acceptance or rejection of the proposition that is presented to
him’ (ibid.). A sentence like (27), however, is
neutral with respect to any indication of the speaker’s beliefs as to the truth
value of p and when they are asked of an addressee, unless they are given
a particular prosodic or paralinguistic modulation, they convey no information
to the addressee that the speaker expects him to accept or reject p.
(ibid.)This means that one of the crucial differences between (27) and (28) lies in
the different answers expected of the addressee. By differentiating the two.
Lyons is taking the expected answer into consideration.