PROSODY
المؤلف:
John Field
المصدر:
Psycholinguistics
الجزء والصفحة:
P226
2025-09-30
285
PROSODY
Variations in the pitch, loudness, tempo and rhythm of speech. Prosody includes stress, pausing, speech rate and intonation (though not features like voice setting, which are classed as paralinguistic1).
In most human beings, prosody appears to be processed by the right side of the brain while lower-level phonetic features are handled by the left. Research has demonstrated a left ear (hence, right hemisphere) advantage for the intonation of English, whether it marks sentence type or attitude.
Prosody has been shown to fulfil a number of important functions for the listener.
Prosody marks the boundaries of Intonational Phrases (IPs), which often coincide with syntactic boundaries. This function is studied by asking listeners to respond to sentences which are ambiguous at some point in their utterance. The ambiguity may lie in the location of a syntactic boundary.
Before she washes, her hair is cut. / Before she washes her hair, it is cut.
Or it may derive from garden path situations, where probability indicates an ending different from the one that occurs.
The rescuers discovered the plane. . . had crashed (vs ‘in the jungle’)
The lawyer questioned... by the police confessed (vs ‘the witness carefully’)
In both cases, listeners have been shown to make use of prosodic cues (especially duration and changes of tempo) in order to parse the ongoing sentence syntactically and to arrive at a single correct interpretation before disambiguation occurs. However, many of the studies reported have made use of somewhat unnatural scripted sentences, read aloud on to tape. More research is needed using natural conversational speech. A further point is that, in natural speech, IP boundaries do not always coincide with syntactic ones. Here, research has shown that, when prosody conflicts with syntax, it is usually the prosodic cues that are heeded.
Pitch movement provides cues to sentence mode. In many of the world’s languages, declarative sentences are loosely associated with a pitch fall, and in around 70 per cent of languages, requests, tentativeness and certain question types are associated with a pitch rise.
Prosody provides cues to completeness. Features such as syllable lengthening, a slowing of speech rate or a marked final fall indicate to the listener that the speaker is willing to hand over the conversational turn. By contrast, an unfinished IP contour or an increase in speech rate indicate the wish to keep the turn. A new topic is often marked by a more rapid speech rate, which slows down as the topic becomes exhausted.
Tonic accent provides information focus. The placing of the tonic accent serves to highlight ‘new’ information introduced by the speaker. A study of naturalistic speech (Brown and Yule, 1983: 166) indicated that 87 per cent of new information was marked by phonological prominence (primary or secondary stress) while 98 per cent of ‘given’ information was unmarked. Note that the location of tonic accent does not correspond categorically to a ‘given/new’distinction, but reflects the speaker’s moment-to-moment assessment of the listener’s informational needs.
Tonic accent and heightened pitch movement indicate contrast and emphasis. Here, speaker decisions reflect an assessment of immediate discourse demands.
Prosody provides affective signals. Research into the relationship between speaker attitude and prosodic patterns has failed to find a systematic correlation between a given intonation contour and an emotion expressed by a speaker or attributed by a listener. Affective marking seems very variable; the impression adduced by a listener may well result from a bundle of features present in the signal rather than a single distinctive tune.
It is difficult to explain how a listener’s understanding of prosodic cues is acquired and stored. The first four functions above are to some extent systematic, and could perhaps be accounted for by the hypothesis that we have a set of typical prosodic patterns stored in our minds. However, the last two functions are very much the product of a local decision by the speaker, and introduce a degree of variability into speech.
A related issue is how and when prosody contributes to the process of planning speech. Levelt’s (1989) model of speaking includes a Prosody Generator; but it draws upon three different sources of information: the surface syntactic structure of the projected utterance, information at lexical level relating to word stress and a ‘meaning’ component reflecting the speaker’s intentions in terms of fore grounded topics and attitude. These relate to different stages of planning, and Levelt therefore argues that intonation is likely to be added incrementally into a speech plan. He takes the view that most features of prosody do not require a great deal of lookahead, and that a final form may only be determined at a late stage, immediately before articulation. Even features such as the relative duration and loudness of syllables might be finalised at this point.
See also: ‘Given/new’, Intonation, Lexical segmentation, Lexical stress
Further reading: Warren (1999)
الاكثر قراءة في Linguistics fields
اخر الاخبار
اخبار العتبة العباسية المقدسة
