ABSTRACT
Research into the linguistic correlates of social categories has been almost exclusively based on the study of lexical, grammatical, and segmental phonetic and phonological characteristics. What are generally referred to as ‘speech styles’, i.e. modes of speaking restricted to or primarily associated with a particular social group, are illustrated solely with reference to restricted usage of items of vocabulary and of grammatical inflections or structures, and to differences in the articulations of vowels, consonants, and vowel–consonant sequences. There is remarkably little attention paid to one other aspect of speech, which I would hold is of major importance for the linguistic definition of social categories, namely, the non-segmental phonetic and phonological characteristics of utterance. These features I shall describe in more detail below; meanwhile, it will suffice to say that they refer to vocal effects due to contrasts in pitch, loudness, and speed of utterance, or to the use of qualities of voice such as nasalization or breathiness in order to communicate specific meanings. ‘Intonation’, or ‘speech melody’, clearly comes under this heading, therefore, as does ‘rhythm’, and what is regularly, albeit vaguely, called ‘tone of voice’. Certain aspects of tone of voice are sometimes studied separately under the heading of ‘paralanguage’. More precisely, non-segmental effect would include any sound effect which cannot be described by reference to a single segment (or phoneme) in the sound system of a language, but which either continues over a stretch of utterance (minimally one syllable), or requires reference to a number of segments in different parts of an utterance that are all affected by a single ‘set’ or configuration of the vocal organs – as when velarization of certain sounds produces a cumulative impression and a semantic interpretation that affect the whole of the utterance. Goffman (1964: 133) looks at this area from a different viewpoint, referring to the expressive aspects of discourse which cannot be clearly transferred through writing to paper – he refers to them as the ‘greasy’ parts of speech! ‘It's not what you say, but the way that you say it’ summarizes the scope of this field, from the formal point of view. Functionally, it is generally agreed that non-segmental phonology provides the main method of communicating emphasis and personal attitudes (sarcasm, surprise, etc.) in language; and that it may also be used with a grammatical role, as when intonation distinguishes between restrictive and non-restrictive relative clauses in English (cf. the use of commas in writing, as between ‘My brother, who's abroad, wrote me a letter’ – one brother – and ‘My brother who's abroad wrote me a letter’ – more than one brother). I would add that non-segmental phonology is also one of the main ways of establishing the identity of social groups in speech.
