The Psychological Foundations of the Grammar 73
his statements had their strict and literal meaning but as he assumed he had
no audience he was not engaging in a communicative act. In other words,
he was apparently producing lexical items without making any assumptions
as to the state of speaker/hearer convergence. Yet, had Chomsky’s speech
been surreptitiously recorded and replayed at a later date to a sympathetic
audience, regardless of his initial non-communicative intention, it seems
likely that the audience would have perceived his speech as a communic-
ative act.
Pickering (2001) describes an almost opposite situation: where non-native
speaking international teaching assistants (ITAs) at a North American
University, who intended to communicate with their audiences, produced
language which failed to label correctly their assumptions of the state of
speaker/hearer convergence. It seems intuitively odd to suggest that com-
munication, no matter how degraded, failed to occur. Both Chomsky and
Pickering’s ITAs produced strings of lexical elements which a grammarian
could encode according to the chaining rules. However, intonationally
Pickering’s ITAs, and perhaps Chomsky too, appear to have produced tone
movements which failed to match the speakers’ assumptions of the state of
convergence between the speakers and their audiences.
Brazil (1997: 133) introduces the term oblique orientation and defi nes it as
speakers presenting their utterances as specimens of language, i.e. speakers
do not attempt to label their message as more than an uninterpreted entity.
They opt out of evaluating the state of speaker/hearer convergence prior
to producing their utterance. Brazil suggests two reasons why speakers
produce oblique utterances, namely: production of formulaic or ritualistic
language (ibid. 137); and diffi culties in utterance planning (ibid. 139).
Speakers may be forced to focus their attention on assembling their utter-
ance and so produce instances of pause fi llers and short tone units with
level tone,^24 or they may read words aloud which they fail to understand, or
they may be unconcerned with the potential communicative implications
of their words (Brazil 1992: 213). Brazil (1997: 135) notes that oblique
utterances are completed by proclaiming tones which operate not only
in used language where they tell but also in oblique language where they
signal a potential end.
Cauldwell and Schourup (1988: 424), in their investigation of Yeats’s
readings of his own poetry, report that he chose a preponderance of level
tones in order to label his readings as specimens of language which, they
claim, resulted in the highlighting of the aesthetics of his poems. Cauldwell
(1999: 44) asserts that Yeats foregrounded the poetic at the expense of the
communicative properties of his poems. Tench (1997: 10), similarly, reports