1.l. On two occasions I have discussed in American Speech certain phenomena connected with the pronunciation of 'short a' in my English (5:396-400 [1930]; 9:313-31*5 [1934]). Both of these discussions have been in effect attempts to explain the occurrences in phonemic terms in such a manner that the two sounds described could be assigned to one phoneme. A re-examination of the whole matter leads to the conclusion that such an explanation is inadequate, that there are two phonemic entities involved, and that it is still possible to recapture the procedure by which the two entities have resulted from the one entity which is found in most English dialects. The situation described is quite common in northern New Jersey, in New York City, in Philadelphia, and has been noted in the speech of many in widely scattered parts of the United States, especially in cities. 1.2. The facts are these: for the sound usually known as 'short a' and transcribed by the symbol [Z], I have in my English two sounds: one is a low (but not lowest) front unrounded lax vowel, characteristically short, but somewhat longer before voiced sounds, which is the most usual American sound in cat, and may be transcribed by the usual symbol [z]; the other sound is slightly higher, tense, always long, and is like the sound used by most Americans in such a word as bare, but without the r, so that bad differs from bared only in the absence of r; I will here use the symbol [Z1] for this second sound. The sound [z] is found, with main or secondary stress, before final [p], [t], Lrk], [tjl, [q], [1], before the same sounds followed by one or more other consonants, and before the same sounds followed by a vowel. Before final [b] [d] [9], [d], [m], [n], [f], [0], [s], Lf1 [v], [z], and before these sounds plus other final consonants, the sound usually found is [z1]; before these consonants followed by vowels, or by one other consonant, [z1] is found regularly when the following vowel is one of a regular formative or paradigmatic suffix, but [z] is most usual in other cases; [ae] is always the sound before intervocalic [61 and [;S]. There are no cases of [z1] before
No takes yet. Share an insight, caveat, or question.
George L. Trager (1940) studied this question.