distinctiveness (n.)

A term used in LINGUISTICS for any feature of speech (or writing) which enables a CONTRAST to be made between PHONOLOGICAL, GRAMMATICAL or SEMANTIC UNITS. Such contrasts might also be labelled ‘relevant’, FUNCTIONAL or significant. The main use of the term has been in phonology, as part of the phrase distinctive feature, where it refers to a minimal contrastive UNIT recognized by some linguists as a means of explaining how the sound SYSTEM of languages is organized. Distinctive features may be seen either as part of the definition of PHONEMES, or as an alternative to the notion of the phoneme. The first of these views is found in the approach of the PRAGUE SCHOOL, where the phoneme is seen as a BUNDLE of PHONETIC distinctive features: the English phoneme /p/, for example, can be seen as the result of the combination of the features of BILABIAL, VOICE, PLOSIVE, etc. Other phonemes will differ from /p/ in respect of at least one of these features. In distinctive feature theories of phonology, however, the phoneme is not considered to be a relevant unit of explanation: symbols such as p, b, etc., are seen simply as convenient abbreviations for particular sets of FEATURES. It is the features which are the minimal units of phonological analysis, not the phonemes. It is argued that, by substituting features for phonemes in this way, generalizations can be made about the relationships between sounds in a language, which would otherwise be missed. Moreover, because features are phonetic units, it should be possible to make inter-language (e.g. DIACHRONIC and DIALECTAL) and cross-language comparisons, and ultimately statements about phonological UNIVERSALS, more readily than by using a phonemic model of phonology.

Distinctive feature analysts claim that there are several advantages over the traditional phonetic alphabet approach to phonological description, which describes UTTERANCES as a sequence of SEGMENTS. For example, it was originally suggested that a relatively small set of abstract feature OPPOSITIONS (a dozen or so) would account for all the phonological contrasts made in languages: it would not then be necessary to recognize so much phonetic classificatory detail as exists on, say, the IPA chart, where the phonological status of the segments recognized is not indicated. In fact, it has turned out that far more features are required, as new languages come to be analyzed. Another advantage, it is suggested, is that CONSONANTS and VOWELS can be characterized using the same set of phonetic features (unlike traditional ‘two-mouth’ descriptions, where the classificatory terminology for VOWELS – HIGH, LOW, etc. – is quite different from that used for consonants – LABIAL, PALATAL, etc.).

By using a system of this kind, some quite specific predictions can be made about the sound systems of languages. For example, using the Jakobson and Halle system below enables one to distinguish phonologically two degrees of FRONT/BACK contrast in the consonant system and three degrees of vowel height. But what follows from this is a universal claim – that no languages permit more than these numbers of contrasts in their phonological systems. These are empirical claims, of course, and in recent years much effort has been spent on investigating these claims and modifying the nature of the feature inventory required.

Two major statements concerning the distinctive feature approach were influential: one by Roman Jakobson and Morris Halle, in Fundamentals of Language (1956), the other by Noam Chomsky and Morris Halle, in The Sound Pattern of English (1968). The Jakobson and Halle approach set up features in pairs, defined primarily in ACOUSTIC terms (as could be detected on a SPECTROGRAM), but with some reference to ARTICULATORY criteria. Examples of their features include VOCALIC v. non-vocalic, CONSONANTAL v. non-consonantal, COMPACT v. DIFFUSE, GRAVE v. ACUTE, NASAL v. ORAL, discontinuous v. CONTINUANT, STRIDENT v. MELLOW, FLAT v. SHARP/PLAIN and VOICED v. voiceless. The emphasis in this approach is firmly on the nature of the oppositions between the UNDERLYING features involved, rather than on the description of the range of phonetic REALIZATIONS each feature represents. In the Chomsky and Halle approach, more attention is paid to the phonetic realizations of the underlying features recognized, and a different system of feature classification is set up. Some of the earlier features are retained (e.g. voice, consonantal, tense, continuant, nasal, strident), but many are modified, and new features added, some of which overlap with the earlier approach (e.g. SONORANT v. OBSTRUENT, DELAYED v. INSTANTANEOUS RELEASE, ANTERIOR v. non-anterior, CORONAL v. non-coronal, DISTRIBUTED v. non-distributed, SYLLABIC v. non-syllabic). The application of these features to languages is not without controversy, and in recent years further suggestions have been forthcoming as to the need for additional features, such as LABIAL.

In recent phonological theory, features have become a focus of attention in their own right, and are widely viewed as the basic unit of phonological representation. The merits of unary (single-valued) as opposed to binary analyses have been presented by some models (e.g. dependency phonology). In addition to questions of feature identification and definition, however, recent research has focused on the nature of feature organization within phonological representations, as part of non-linear phonology. In particular, feature geometry looks especially at the non-linear relationship between features, and at the way they can be grouped into a hierarchical array of functional classes. Several formalisms have been devised to handle the relationships between features in particular phonological contexts, and terminology has begun to develop accordingly. For example, in the study of assimilation, a rule which spreads only features not already specified in the target is said to be operating in a feature-filling mode; if the rule applies to segments already specified for the spreading features (thereby replacing their original values), it is said to apply in a feature-changing mode.