Hostname: page-component-6898cfcb55-w5hgh Total loading time: 0 Render date: 2026-11-11T08:18:33.828Z Has data issue: false hasContentIssue false
  • English
  • Français

Criteria for phonetic similarity

Published online by Cambridge University Press:  22 April 2026

William M. Austin
William M. Austin*
Affiliation:
Georgetown University
Get accessRights & Permissions [Opens in a new window]

Extract

The classical definition of the phoneme runs something like this: a set of sounds in complementary (noncontrastive) distribution and phonetically similar. To this one may append Trager's statement, ‘The criterion of complementary versus contrasting distribution is first, then that of phonetic similarity.‘ While it may be generally admitted that the criterion of phonetic similarity is of secondary importance, the question can be raised: is it of any importance at all? Here, as in everything else, the answer depends upon the level of analysis. The phoneme qua phoneme exists only by virtue of its distribution on a higher level, in the morpheme. On the phonetic level the phoneme is only a metaphysical possibility, on the morphemic level it is reality. Phonemes are phonemes because of their function, their distribution, not because of their phonetic similarity. Most linguists are arbitrary and ad hoc about the physical limits of phonetic similarity.

Information

Type
Research Article
Information
Language , Volume 33 , Issue 4-Part1: October-December , December 1957, pp. 538 - 544
Copyright
Copyright © 1957 Linguistic Society of America

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

1 G. L. Trager, Encyclopaedia britannica (ed. of 1956) s.v. Linguistics. Cf. also B. Bloch: ‘A phoneme is a class of non-eontrastive and phonetically similar phones’, Lg. 26.90 (1950) ; E. H. Sturtevant: ‘A phoneme, then, is either a single speech sound or a group of similar speech sounds, which in a given language function in the same way’, An introduction to linguistic science 15–6 (New Haven, 1947). There are many similar definitions.

The substance of this paper was presented to the Washington Linguistic Club on 25 January 1955.

2 Lg. 26.108 (1950).

3 Cf. L. Bloomfield, Language 365 (New York, 1933).

4 Prinzipien der Sprachgeschichte 5, Chapter 3 (Halle a. S., 1920); referred to by Sturtevant, op.cit. 75.

5 H. M. Hoenigswald, The principal step in comparative grammar, Lg. 26.359 (1950).

6 Bloch, Lg. 26.100.

7 H. Wolff, Osage I: Phonemes and historical phonology, IJAL 18.64 (1952).

8 W. E. Weimers, Notes on two languages of the Senufo group, I. Senadi, Lg. 26.129 (1950).

9 C. H. Matthews, Phonemic analysis of a Dakota dialect, IJAL 21.57 (1955).

10 H. Hoijer and E. P. Dozier, The phonemes of Tewa, Santa Clara dialect, IJAL 15.141 (1949).

11 H. Wolff, Comparative Siouan I, IJAL 16.62 (1950), referring to Z. S. Harris and R. H. Lowie, Hidatsa texts = Prehistory research series of the Indiana Historical Society 6:1 (1939).

12 Wolff, ibid., referring to R. H. Lowie, The Crow language = UCPAAE 39 (1941).

13 J. Faublée, Les langues malayo-polynésiennes, in Meillet and Cohen, Les langues du monde 2 671 (Paris, 1952).

14 Ibid.

15 Sturtevant, op.cit. 67.

16 Ibid.

17 M. Räsänen, Materialien zur Lautgeschichte der türkischen Sprachen, Studia orientalia 15.113 (1949).

18 Id. 22.

19 Wolff, IJAL 16.115.

20 IJAL 16.118.

21 D. L. Olmsted, Palaihnihan and Shasta I: Labial stops, Lg. 32.75–7 (1956).

22 G. van Bulck, Les langues bantoues, in Meillet and Cohen, op.cit. 871.

23 Ibid.

24 W. M. Austin, A brief outline of Dagur grammar, SIL 10.65 (1952).

25 F.-K. Li, Siamese wan and waan, Lg. 32.81 (1956), has noted what he calls a ‘curious correspondence’ between Siamese wan ‘day’ and the closely related Dioi ŋon ‘day’, and posits a Proto-Thai *ŋw-. This is not necessary on grounds of phonetic similarity (cf. example 16 above), though it may be on other grounds, such as pattern congruence.

Wolff, IJAL 16.113, tentatively sets up a Proto-Siouan phoneme */?/ with postvelar spirant and glottal stop as allophones. This seems impossible; but if such a situation did exist, the velar and faucal columns would have to be merged.

26 The question may be asked, what if [tš] was an allophone when the borrowing took place? This is possible but irrelevant, since [tš] would have to be a separate phoneme before it could acquire an allophone [š].