Abstract
Phonology is the study of the sound systems of languages. Out of the very wide range of sounds that the human vocal apparatus can produce, and which are studied by phonetics, only a relatively small number are used distinctively in any one language. The sounds are organized into a system of contrasts, which signal differences of meaning within the language. The aim of phonology is to establish on what basis this contrastivity operates, to describe the patterns of distinctive sound used in a language, and to make as general statements as possible about the nature of sound systems in the languages of the world. Putting this another way, phonology is concerned with the range and function of sounds in specific languages (and is often therefore referred to as “functional phonetics”); it is also concerned with the rules which can be written to show the types of phonetic relationships that relate and contrast words and other linguistic units1.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
For introductions to phonology, see O’Connor (1973), Robins (1971: Ch. 4), Fudge (1973), Hyman (1975). “Functional phonetics” is discussed in Martinet (1949) and Haas (1968).
For the main conceptions of the phoneme, see Trubetskoy (1939), Jones (1950), Bloomfield (1933).
See Gimson (1980). For the notion of “received pronunciation”, or R.P., see that book (p. 89 ff.), and also Trudgill (1974).
The original statement was made by members of the Prague School of linguistics, especially by Trubetskoy (1939). Later developments included the approach of Jakobson and Halle (1956), which was a formative influence on generative phonology.
See Hyman (1975), Schane (1973), and for the main theoretical statement, Chomsky and Halle (1968).
See Firth (1948), Palmer (1970), Robins (1971: Ch. 4), Waterson (1971).
But it should be noted that initial, medial and final will have different meanings, depending on which notion is used: cf.
The idea of an “inventory” is not an elementary one: inventories have to be made on the basis of certain theoretical principles, and there are several ways in which phonetic inventories can be made to look very different, with phonological factors influencing decisions. For example, in a list of P’s phones, should [tJ] be regarded as a separate manner category (“affricate”) or as a type of palatal plosive (as in Cruttenden,1972); if the former, should it be placed in the inventory adjacent to plosive (cf. Connor and Stork,1972) or fricative (Grunwe11,1975)? Is [s] to be classified along with other alveolars, or as aseparate “post-alveolar” category? Is [w] labial or velar? And so on. Table 3 gives some possible inventory frameworks.
Cf. Beresford and Grady’s (1968) realization tables, e.g. slt d — k; Cruttenden’s (1972) matrix of target and realisation; Ferguson, Peizer and Weeks’ (1973) statements in a “model and replica” approach.
Cf. Hyman (1975), Schane (1973), Chomsky and Halle (1968).
For example the prosodic approach to phonological acquisition seems wholly based on a single child (Waterson 1971). An example of a factor which has been little considered in the context of phonological disability is the notion of “variable rule” (Labov 1972).
See the review in Fletcher and Garman (1979:4 ff.), and the paper by Menyuk and Menn in that volume. The traditional view is illustrated by Irwin and his colleagues (e.g. Irwin and Chen 1943). Much of the linguistic interest in babbling and speech stems from Jakobson (1941), but his discontinuity theory concerning the relationship between the two is no longer convincing, in view of the recent studies. For early (pre-babble) phonetic development in production, see Stark, Rose and McLagen (1975); on early phonetic and phonological perception, see Morse (1974).
It is based on a synthesis of Sander (1961), Templin (1957), Olmsted (1971) and Ingram ( 1976: Ch. 2); see further below.
Olmsted (1971), for example, defines acquisition as when at least half of the child’s attempts at a phone are successful. This study should also be referred to for its information about developmental trends in the distribution of phones—on the whole (apart from fricatives) initial phones medial final.
The same comments apply to studies of the development of phonemic perception, as opposed to the production studies, some of whose results are given above. Here too it is possible to establish acquisitional hypotheses concerning the order of phonemes or features. For example, Olmsted used work by Miller and Nicely (1955) to predict children’s errors in consonant discrimination. Miller and Nicely had shown that adult perceptual confusions were greatest on place of articulation, less on sounds involving a fricative component, and least on voicing and nasality, and Olmsted finds a similar predictiveness for the errors in his child sample (supported by Timm [1977] on Russian). Cf. also the progression of stops -’ voiceless fricatives voiced fricatives found by Edwards ( 1974 ). But there are many exceptions to these trends. See also Menyuk and Menn (1979).
The following is based mainly on Ingram (1974 a, 1974 b, 1979 ), Smith (1973, 1974); only the main processes are illustrated. I shall use phonemic notation for the examples, for convenience: see Smith’s studies for examples in feature notation.
See Shvachkin (1973), Garnica (1973), Edwards (1974). The lexical diffusion view below is in Ferguson and Farwell (1975), Hsieh (1972).
See Waterson (1971), who cites a parallel in Piaget’s (1926) notion of verbal syncretism; also Menyuk and Menn (1979).
See further, on the interdependence of production and perception, Eilers (1975), Eilers and 011er (1976). The Macken and Barton research (on the learning of word-initial plosives, as measured by voice-onset-time) generally provides support for the lexical diffusion model of acquisition, though aspects of their findings suggest a rapidity of acquisition which might support the across-the-board model.
However, several studies of one important variable, voice-onset-time, have taken place: for production, see Gilbert and Purves (1977), Menyuk and Klatt (1975), Gilbert (1977), Macken and Barton (1980); for perception, see Eilers, Wilson and Moore (1979). It should also be pointed out that a child may have excellent motor control, and still have phonological problems, e.g. he may have a well-formed articulation in some words, but be unable to use it contrastively. Note also distributions such as puddle [pegal], but puzzle- [pedal], which illustrates the priority of phonological over phonetic issues.
See Salus and Salus (1974). For a relevant general account of physical maturation in the infant, see Lenneberg (1967), Bosma (1975).
See Jakobson (1941), Jespersen (1926), Ladefoged ( 1975: 219–220 ).
See the discussion of phonological complexity in Hyman ( 1975: Ch. 4).
See also Shibamoto and Olmsted (1978), though they conclude that the Ferguson and Farwell lexical principle must be restricted to the first phone in a word: once this is chosen, they argue, phonological processes take over and govern the developing structure of the rest of the word.
Cf. the study by D. Barton (referred to in Menyuk and Menn [1979: 51]), who found in a discriminiation study that responses were better in the case of familiar words as opposed to nonsense words.
For example, Shibamoto and Olmsted (1978: 454) talk about individual “strategies”, Ingram (1979) of “phonological preferences” for particular articulatory patterns; and then there is Priestley’s detailed study (1977). See also the discussion in Menyuk and Menn ( 1979: 52 ).
See, e.g. Shibamoto and Olmsted (1978: 419–420). On the other hand, Leonard, Schwartz, Folger and Wilcox (1978) adopt a more optimistic attitude, and support the inclusion of imitation data in corpora of spontaneous speech—though they do advocate caution. A similar division of opinion is found in the literature on articulation testing: see, for example, the review in Johnson and Somers (1978), who conclude that there is a need to keep the two modes apart.
See Cruttenden (1972), Priestley (1977), and the general discussion in Ingram (1976) and Grunwell (1977).
For example, the 1959 terminology leaflet of the College of Speech Therapists, Powers (1957: 713), etc. Other categories (e.g. transposition) are sometimes added to this list, but these are generally reducible to types of substitution.
See further Higgs (1970), Beresford and Grady (1968), Grunwell (1975), (1977: Ch. 2), Lund and Duchan (1978).
For phonemic analysis, see Haas (1963), Hartley (1966), Beresford and Grady (1968), Beresford (1971, 1972), Connor and Stork (1972), Panagos (1974). For feature analyses, see Compton (1970), McReynolds and Huston (1971), Pollack and Rees (1972), 011er (1973), Smith (1973). For a general review, see Grunwell (1977: Ch. 2 ), Walsh (1974). The threefold division below is the main organizational principle of Grunwell.
On the general properties of phonological systems, see Greenberg (1966), Grunwell ( 1977: Ch. 3).
See Hockett ( 1958: Ch. 11), Hyman (1975: Ch. 1). The lack of this normal instability in much clinical data is noteworthy, the phonological system showing signs of developmental arrest, and thus presenting little potential for change. See Grunwell (1977: Ch. 7 ).
Items repeated during the sample are given in sequence; items with an unclear segmentation relative to the target are marked ‘F.
This profile chart is explained in detail in Crystal (1982). Several details are omitted in the present discussion.
The omissions total includes cases of one C omitted in clusters. There were no cases of clusters being omitted completely, though this may be due to ambiguity in the target, e.g. if P used [ka:] with reference to a picture of two cats, it would not be entirely clear whether the target were cat or cats.
This is hardly a real figure, in view of the small numbers. It will not be further considered in (f) and (g) below.
To be distinguished from place glottalisation above, where the whole of a segment is replaced by a glottal. In the present category, the two cases of medial clusters involving a glottal stop have not been analysed as glottalised.
For more on typology, see Ingram ( 1976: Ch. 5) and Grunwell (1977). Cf. also Lund and Duchan’s (1978) “multifaceted” approach. One of the problems of typological statements to date is that they are very restricted, through the use of only one descriptive framework. The author thus sees only what his descriptive framework lets him see, and classifications and generalisations are made accordingly. It is a permanent danger, of course, but it is as well to be aware of it.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1981 Springer-Verlag Wien
About this chapter
Cite this chapter
Crystal, D. (1981). Segmental Phonology. In: Clinical Linguistics. Disorders of Human Communication, vol 3. Springer, Vienna. https://doi.org/10.1007/978-3-7091-4001-7_2
Download citation
DOI: https://doi.org/10.1007/978-3-7091-4001-7_2
Publisher Name: Springer, Vienna
Print ISBN: 978-3-7091-4003-1
Online ISBN: 978-3-7091-4001-7
eBook Packages: Springer Book Archive