Judgements of attitudinal meanings in isolation and in context

Richard T. Cauldwell

Department of English, The University of Birmingham

R.T.Cauldwell@bham.ac.uk

This is an abbreviated report of a demonstration of the fact that contextual factors have a considerable impact on judgements of attitudinal meanings. I extracted the utterance What do you mean? (falling tone on mean) from a recording. I then, without giving any contextual information whatsoever, asked groups of native and non-native speakers of English (students of Applied Linguistics) to make judgements about attitudinal meanings. The judgement requested was a binary one: is the speaker angry/irritated or not angry/irritated?. The utterance was played three times, and students recorded their judgements by ticking boxes in a table.

I then asked them to listen to the utterance with contextual information supplied: that this was a parent-child interaction (Matthew aged 3 and a half and Dad aged 43) during a dinner table conversation. For this stage in the procedure, the informants had a script of the interaction in front of them so they could the more easily locate the utterance in the recording. They heard the utterance situated in a 45 second extract twice. They were then asked to make the same judgement (irritated/angry or not) as they had for the utterance in isolation. The immediate context for the utterance was:

Matthew: hey...hey...when it...when...when when when it when it snows...we can...drive and our car got stuck

Dad: our c wha wha what...what do you mean

Matthew: our car got stuck in the snow

The results of the two sets of judgements are shown in Table 1.

Table 1: Judgements (n = 30) of attitudinal meanings in 'What do you mean?'

The table shows a major shift in perception between judgements made in isolation, and judgements in context. When heard in isolation 66% of informants stated that they perceived the speaker to be angry/irritated, 27% selected not angry/irritated and 7% did not want to make a judgement. On hearing the utterance in context, the percentage of informants continuing to hear the speaker as angry/irritated dropped dramatically by 56%; and there was a corresponding rise in the percentage of informants indicating that the speaker was not angry/irritated.

The group of 30 informants consisted of a mixture of native (n = 17) and non-native speakers (n=13) of English. Table 2 gives the results for the native speakers.

Table 2: Native speaker judgements (n = 17) of attitudinal meanings in 'What do you mean?'

Table 2 shows that there was a remarkable 77% swing between native speaker in isolation judgements, and their in context judgements: a swing from 8% making a not irritated/angry judgement in isolation to 85% making this judgement in context.

Two questions: (1) How is it possible for there to be such a marked difference between in isolation and in context judgements? (2) How is it possible that the native speaker group exhibit a more dramatic shift in perception? There is not space here to explore these issues fully (see Cauldwell, 1997 for a view of the relationship between attitudinal meanings and prosody) but here are speculative answers to these questions.

Heard in isolation, Dad's 'What do you mean' has characteristics of speed and tenseness which mark it as outside the 'normal range' for an average adult male voice: it has a 'rarity value' (Cauldwell, 1997) appropriate to a cueing of anger/irritation. It is this 'normal range' for an adult male speaking English, and the associated 'rarity values', to which native speakers are more highly attuned than non-native speakers (this is the answer to question 2).

However, the 'normal range' [defined as 'not cueing attitudinal/emotional meaning'] itself varies with context. The normal range varies with the time of day, the nature of accompanying activities, and who the addressees are. Hearers in context instantaneously compensate for these variations in the normal range, and will only attend to marked excursions from the normal to evaluate whether or not these excursions are cueing the presence of attitudinal/emotional meanings. When hearers have little or no contextual information, when they have no prior experience of the speaker's voice, they do not have sufficient time to attune to the speaker's normal range, and are thus likely to make judgements based on their past experience of similar voices in familiar contexts (this is the answer to question 1).

If the answer to question 1 is correct, then we have to be extremely careful with the generalisations we make about the relationship between prosody and attitude/emotions. Followers of Crystal's writings on this topic (1969, 1975, 1995) will know of the increasing importance he assigns to the role of context. Using digitally stored recordings of speech navigated in ways discussed by Cauldwell (1996) we can explore further the relationship between context and the role of prosody, in communicating attitudinal/emotional meanings.

References

Cauldwell, R. (1996) Direct Encounters with fast speech on CD Audio to teach listening. System, 24/4, pp. 521-528

Cauldwell, R. (1997) Tones, Attitudinal Meanings, and Context. Speak Out! Newsletter of the IATEFL Pronunciation Special Interest Group, No 21, December 1997, pp. 30-35

Crystal, D. (1969) Prosodic systems and intonation in English. Cambridge: CUP

Crystal, D. (1975) The English Tone of Voice. London: Edward Arnold.

Crystal, D. (1995) The Cambridge encyclopaedia of the English language. Cambridge: CUP.