228 Notes
information units may be stored in what she calls the ‘heteromorphic distributed
lexicon’ as a unitary element.(^7) Levelt (1989: 23) argues against the existence of any single unit of talk, including
information units. However, his objection that ‘different processing components
have their own characteristic processing units’ does not seem to confl ict with the
assumption that information units are one of the processing units of increments
as are of course elements from the lexicogrammar such as nouns and verbs.
(^8) The chunking of the message into tone units, the selection of prominence and
tone, etc.
(^9) For expository purposes, this discussion of increments is restricted to telling
increments. Asking increments are discussed in Chapter 2 Section 2.2.5.
(^10) For an alternate view see Sinclair and Mauranen (2007) who code their corpus
exclusively in terms of M elements which increment the speakers’ ideational
mess age and O elements which are used to organize the textual and inter personal
interaction of the discourse. However, unlike this book, Sinclair and Mauranen
assumed the pre-theoretical existence of chunks of speech, possibly equivalent
to tone units, and were unconcerned with formally notating relations between
the chunks.
Chapter 2
(^1) Unless expressly stated in this chapter all page numbers refer to Brazil (1995).
(^2) For a similar view see Bourdieu (1991: 55) who argues, contra Chomsky, that
linguistic competence does not consist of the ability to generate an infi nite num-
ber of sentences but is the ability to generate communicatively appropriate
sentences.
(^3) Of course, as speech is not a one hundred per cent secure means of transmitting
information, speakers may misjudge the situation and fail to complete their
message to their hearers’ satisfaction.
(^4) With the advent of the computer age there has been an increase in process forms
of writing such as computer instant messaging which are not interpreted as
a unitary text but rather turn by turn; with each turn serving to help fulfi l a com-
municative purpose.
(^5) The symbol # notates an increment ending. An explanation of all transcription
conventions is printed on pp. x–xi. While the optional response oh is part of the
telling exchange it is not part of speaker A’s telling increment.
(^6) Sinclair and Mauranen (2007: 136) label the intermediate state achieved after
production of the initial tone unit ‘a completion’ and the target state achieved at
the end of the utterance ‘a fi nishing’. Completions are points reached which
signal potential units of meaning while fi nishings signal, in the context in which
they were uttered, actual achievements of target state. However, as this book
is interesting in describing actual rather than potential meaning it does not
distinguish between completions and fi nishings.
(^7) Init State, Inter State and Tar State refer to Initial State, Intermediate State and
Target State respectively.
(^8) Brazil notates suspensive elements in lowercase.