Introduction

Preview

In this chapter, we are concerned with the processing that is involved in our understanding of utterances. Now, ‘utterances’ can be of many different kinds, ranging from marginally linguistic vocalisations of emotions, through social gambits and stereotypes, to lexical elements (which may appear as single-word utterances) and structured grammatical sequences. The concept ‘utterance’ is thus a pretheoretical one, which will defy any attempt to provide an all-embracing characterisation in terms of linguistic properties. We shall, however, characterise it for our purposes here as any signal-based implementation of the abstract language system. This provides for written as well as spoken utterances, but excludes covert use of language as in silent verbal reasoning.

In the preceding chapter, we have examined some issues in lexical access, in a way that has not really considered the role of words in (single- or multi-word) utterances: for it is one thing to gaze at letter arrays flashed briefly on a screen, in order to pronounce them, or decide whether they constitute a word, etc., but quite another matter for us to read words in written utterances. And so also for speech perception in artificial situations vs listening to utterances for their communicative potential. In this chapter, we shall attempt to review what is known about processing linguistic utterances, of the multi-word type, since they afford us evidence of lexical and structural processing proceeding in some relationship to each other (see Scholes 1978; Cutler 1983).