1 See Lado, R., Linguistics across Cultures, Ann Arbor: University of Michigan Press, 1957, p. 7 Google Scholar: “The most effective materials are those that are based upon a scientific description of the language to be learned, carefully compared with a parallel description of the native language of the learner.” Haugen, E., “Problems of bilingual description,” Georgetown Monograph no.7, Washington, D.C. : Georgetown University Press, 1954, pp. 9-19 Google Scholar, also discusses the need for a wise choice of systems and symbols and insists that in studies involving bilingualism decisions as to systems and symbols should be motivated in part by the practical consequences.

2 See Lado, Linguistics, p. 13: “Spanish does not have /v/ as in vote.”Google Scholar

3 Lehn, W. and Slager, W. R., “A contrastive study of Egyptian Arabic and American English: the segmental phonemes,” Language Learning, 9 (1959), pp. 25-33 CrossRefGoogle Scholar, use this type of layout. Samarajiwa, C. and Abeysekera, R. M., “Some pronunciation difficulties of Sinhalese learners of English as a foreign language,” Language Learning, 14 (1964), pp. 45-50 CrossRefGoogle Scholar, also use this type of layout and do so most naively. In an excellent discussion of this particular article and the approach it uses, Kandiah, T., “The teaching of English in Ceylon: some problems in contrastive statement,” Language Learning, 15 (1965), pp. 147-65 CrossRefGoogle Scholar, severely criticizes the approach and its results and in turn proposes an approach which is like the second approach discussed in this paper.

4 Weinreich, U., Languages in Contact, The Hague: Mouton, 1966 edition, p. 7 Google Scholar and section 2.2, points out that the phonemes of one language are by definition incommensurable with those of another language; however he allows himself to say, p. 20: “Neither French nor Russian have phonemes.” Haugen, “Problems,” p. 10, is also plainly on record as being in agreement with Weinreich’s views on this point.

5 Lado, Linguistics, p. 12, and Shen, Y., “Phonemic charts alone are not enough,” Language Learning, 5 (1955), pp. 122-9 Google Scholar, propose such refinements.

6 Haugen, , “Problems,” pp. 11-2 Google Scholar, uses this term.

7 Weinreich, Languages, pp. 14-8 and passim, discusses phonic interference between Romansh and Schwyzertütsch and uses this approach; however, there is little delicacy apparent nor are the problems discussed and classified very systematically.

8 Haugen, , “Problems,” p. 12 Google Scholar, offers a classification which employs diaphonie relationships of various kinds: simple, compound (convergent and divergent), complex, and complex-compound.

9 See Sumukti, R. H., “Some examples of Sundanese and Javanese phonic interference in relation to learning English,” Language Learning, 8 (1958), pp. 37-48 CrossRefGoogle Scholar.

10 See Ellis, C. D., Cree, Spoken, Part 1, Toronto: Department of Missions, Anglican Church of Canada, n.d., p. 9 Google Scholar. Ellis actually gives only the Cree vowel system in this way, but there is an implied contrast with an English vowel system representable in the same way.

11 Lado, Linguistics, refers repeatedly to these problems in his second chapter entitled “How to compare two sound systems,” but the treatment is extremely sketchy and suggestive rather than explicit and securely motivatedGoogle Scholar.

12 Weinreich, Languages, pp. 18-9, offers one such classification.

13 Lado, Linguistics, pp. 25-7, discusses examples of predictive difficulty from English to Japanese, Thai and Tagalog. Note also the conflict of views between Weinrich and Haugen over an article by Reed, D. W., Lado, R. and Shen, Y., “The importance of the native language in foreign language learning,” Language Learning, 1 (1948), pp. 17-23 CrossRefGoogle Scholar. Weinreich, Languages, p. 20, says that they “were able ... to predict with remarkable accuracy,” but Haugen, “Problems,” p. 11, is “struck by the lack of predictive correlation between the charts and the difficulties.”

14 This approach is very much the same as the one which results when Moulton, W. G., “Toward a classification of pronunciation errors,” Modern Language Journal, 46 (1962), pp. 101-9 CrossRefGoogle Scholar, combines what he calls two different approaches (the first a collection of errors and the second a predictive classification). In this paper Moulton’s first approach is rejected out of hand because such error-collecting may be independent of any concern with a systemic view of language, and throughout we insist on such a view.

15 As in Catford, J. C., A Linguistic Theory of Translation, London: Oxford University Press, 1965, ch. 8 Google Scholar.

16 Weinreich, Languages, pp. 7-8.

17 Ibid., pp. 21-2.

18 See Weinreich, U., “On the description of phonic interference,” Word, 13 (1957), pp. 1-11 Google Scholar.

19 A supplementary chapter by Martin, J. W. in Stockwell, R. P. and Bowen, J. D., The Sounds of English and Spanish, Chicago: University of Chicago Press, 1965, pp. 139-63 Google Scholar, is one of the best examples of this approach. Martin uses a distinctive features analysis of the phonemes of English and Spanish and takes distribution into account rather well.

20 Stockwell and Bowen, Sounds, ch. 2, propose a classification which also makes use of what they call obligatory and optional choices. Moulton, “Toward a Classification,” also offers a useful classification.

21 See Halle, M., “On the basis of phonology,” in Fodor, J. A. and Katz, J. J., eds., The Structure of Language, Readings in the Philosophy of Language, Engiewood Cliffs : Prentice-Hall, 1964, pp. 324-33 Google Scholar, for such a suggestion.

22 Martin’s chapter in Stockwell and Bowen, Sounds, is contained in an overall study which in its syntactic contrasts attempts, though not very successfully, to use transformational-generative theory, but which uses “traditional” phonemics in its phonological contrasts. Martin’s distinctive features treatment is also a non-generative surface one.

23 See Chomsky, N., Current Issues in Linguistic Theory, The Hague: Mouton, 1964 Google Scholar, eh. 4, for a fuller discussion of the phonological component of a transformational-generative grammar.

24 Chomsky, Current, p. 96.

25 Chomsky, N. and Halle, M., “Some controversial questions in phonological theory,” Journal of Linguistics, 1. 2 (1965), pp. 97-138 CrossRefGoogle Scholar, say, p. 136, in an Appendix to their paper that “we see nothing strange about the conclusion that an underlying base form may be ‘perceived’ (or internally represented in some way in the process of interpreting an utterance) although it corresponds to no identifiable physical part of the sound stimulus. And we see no difficulty in accepting the hypothesis . . . that a systematic phonemic distinction can persist (because of its role in the system of phonological processes) despite the fact that there is no one-to-one correlation of the systematic phonemes to identifiable physical phones.”