phoneme (n.)

The minimal unit in the sound SYSTEM of a LANGUAGE, according to traditional PHONOLOGICAL theories. The original motivation for the concept stemmed from the concern to establish patterns of organization within the indefinitely large range of sounds heard in languages. The PHONETIC specifications of the sounds (or PHONES) heard in speech, it was realized, contain far more detail than is needed to identify the way languages make CONTRASTS in MEANING. The notion of the phoneme allowed linguists to group together sets of phonetically similar phones as VARIANTS, or ‘members’, of the same underlying unit. The phones were said to be REALIZATIONS of the phonemes, and the variants were referred to as allophones of the phonemes. Each language can be shown to operate with a relatively small number of phonemes; some languages have as few as fifteen phonemes; others as many as eighty. An analysis in these terms will display a language’s phonemic inventory/structure/system. No two languages have the same phonemic system.

Sounds are considered to be members of the same phoneme if they are phonetically similar, and do not occur in the same ENVIRONMENT (i.e. they are in COMPLEMENTARY DISTRIBUTION) – or, if they do, the substitution of one sound for the other does not cause a change in meaning (i.e. they are in FREE VARIATION). A sound is considered ‘phonemic’, on the other hand, if its substitution in a word does cause a change in meaning. In a phonemic transcription, only the phonemes are given symbols (compared with phonetic TRANSCRIPTIONS, where different degrees of allophonic detail are introduced, depending on one’s purpose). Phonemic symbols are written between oblique brackets, compared with square brackets used for phonetic transcriptions; e.g. the phoneme /d/ has the allophones [d] (i.e. an ALVEOLAR VOICED variant),  (i.e. an alveolar devoiced variant),  (i.e. a dental variant) in various complementary positions in English words. Putting this another way, it is not possible to find a pair of words in English which contrast in meaning solely on account of the difference between these features (though such contrasts may exist in other languages). The emphasis on transcription found in early phonemic studies is summarized in the subtitle of one book on the subject: ‘a technique for reducing languages to writing’. The extent to which the relationship between the phonemes and the GRAPHEMES of a language is regular is called the ‘phoneme–grapheme correspondence’.

On this general basis, several approaches to phonemic analysis, or phonemics, have developed. The PRAGUE SCHOOL defined the phoneme as a BUNDLE of abstract DISTINCTIVE FEATURES, or OPPOSITIONS between sounds (such as VOICING, NASALITY), an approach which was developed later by Jakobson and Halle, and GENERATIVE phonology. The approach of the British phonetician Daniel Jones (1881–1967) viewed the phoneme as a ‘family’ of related sounds, and not as oppositions. American linguists in the 1940s also emphasized the phonetic reality of phonemes, in their concern to devise PROCEDURES of analysis, paying particular attention to the DISTRIBUTION of sounds in an UTTERANCE. Apart from the question of definition, if the view is taken that all aspects of the sound system of a language can be analyzed in terms of phonemes – that is, the SUPRASEGMENTAL as well as the SEGMENTAL features – then ‘phonemics’ becomes equivalent to phonology (= phonemic phonology). This view was particularly common in later developments of the American STRUCTURALIST tradition of linguistic analysis, where linguists adopting this ‘phonemic principle’ were called phonemicists. Many phonologists, however (particularly in the British tradition), prefer not to analyze suprasegmental features in terms of phonemes, and have developed approaches which do without the phoneme altogether (‘non-phonemic phonology’, as in PROSODIC and DISTINCTIVE FEATURE theories).

The term phonemic clause has been used primarily in PSYCHOLINGUISTIC research into the distribution and function of PAUSES: it refers to a GRAMMATICAL structure produced within a single INTONATION CONTOUR, and bounded by JUNCTURES. The term phonemic tier is often used in AUTOSEGMENTAL PHONOLOGY for the TIER containing segments specified for the features that identify CONSONANTS and VOWELS (other than [±syllabic], which is specified on the SKELETAL TIER); also called the segmental tier.