prosody (n.)

A term used in SUPRASEGMENTAL PHONETICS and PHONOLOGY to refer collectively to variations in PITCH, LOUDNESS, TEMPO and RHYTHM. Sometimes it is used loosely as a synonym for ‘suprasegmental’, but in a narrower sense it refers only to the above variables, the remaining suprasegmental features being labelled PARALINGUISTIC. The narrow sense is close to the traditional use of the term ‘prosody’, where it referred to the characteristics and analyses of verse structure. The term prosodic features is preferred in LINGUISTICS, partly to enable a distinction to be drawn with the traditional use. In some approaches to phonology, the term sentence prosody is used to group together intonation, phrasal rhythmic patterning and more general features of prosodic phrasing. The above use treats ‘prosody’ as a mass noun.

In the theory of phonology proposed by J. R. Firth (prosodic phonology), prosody is treated as a count noun, and given special status. It is distinguished in this approach from PHONEMATIC UNIT: the latter is a SEGMENTAL unit, such as consonant or vowel, whereas prosodies are features extending over stretches of UTTERANCE (one talks of ‘sentence prosodies’, ‘syllable prosodies’, etc.) – a notion which took on a more central role in later thinking (and also the concept of ‘semantic prosody’ in lexicology). Not only would pitch, STRESS and JUNCTURE patterns be subsumed under the heading of prosody, but such features as SECONDARY ARTICULATIONS would also be included, e.g. lip-ROUNDING or NASALIZATION, when these are used to account for phonotactic restrictions, or to characterize GRAMMATICAL structure (as in the notion of ‘VOWEL harmony’). Another feature of Firth’s prosodic analysis is its POLYSYSTEMIC principle: it permits different phonological systems to be set up at different places in grammatical, lexical or phonological structure: e.g. the contrasts which occur at the beginning of a WORD may not be the same as those which occur at the end, and this fact is given special attention in this approach.

In PHONEMIC phonology, linguistically contrastive prosodic features are often referred to as prosodemes. In GENERATIVE phonology, prosodic features are considered to be one of the five main dimensions of classification of speech sounds (the others being MAJOR CLASS FEATURES, CAVITY features, MANNER-OF-ARTICULATION features and SOURCE features). Recently, the term has been applied to a model of MORPHOLOGY in which non-LINEAR phonological REPRESENTATIONS play a central role. Using notation derived from AUTOSEGMENTAL PHONOLOGY, the approach is based on the view that information about the CANONICAL pattern of SEGMENTS in a FORM (the prosodic template) is represented on a different tier from information about the kinds of segments occurring in the form. In METRICAL PHONOLOGY, one of the levels of structure in a metrical TREE is referred to as a prosodic level.

In prosodic morphology, the focus is specifically on the way in which morphological and phonological determinants of linguistic form interact, and the notion of prosody becomes more powerful, as it is seen to determine the structure of morphological templates. This approach makes reference to the prosodic morphology hypothesis (templates are defined in terms of the units in a prosodic hierarchy – MORA, SYLLABLE, FOOT and prosodic word) and the notion of prosodic circumscription (the domain to which morphological operations apply is circumscribed by prosodic as well as morphological criteria). In an alternative account, p-structure (i.e. ‘prosodic structure’) is seen as a LEVEL at which syntactic and phonological components interact, with its own hierarchical organization of four domains – phonological word, phonological phrase, intonational phrase and utterance – the properties of which are specified by prosodic hierarchy theory (‘hierarchy’ here referring to a higher level of structural organization than in the case of prosodic morphology). Some model of a prosodic hierarchy is assumed in most modern phonological frameworks.