Harold B. Allen

Sex-Linked Variation in the Responses of Dialect Informants Part III: Grammar:

In this, the third of a three-part series concerned with the influence of an informant’s sex upon the response to an inquiring field investigator, attention is given to what a number of researchers in the wide area of women’s speech have customarily considered standard English usage. The accepted generalization, backed by a long history of popular feeling, is that more women than men use &dquo;standard English&dquo;. They have &dquo;better grammar&dquo;. A caveat, however, comes from Philip Smith, who, in his Lon.fua.fe, the Seses and Society (1985 :80), wrote: &dquo;The generalization about the relative use of the standard features by women and men is premature in several ways, and these should be discussed in order to pre-empt the misguided expense of time and resources that the aforementioned endeavour [to enquire into the masculine and feminine connotations of standard and nonstandard speech variables] would entail.&dquo; There is indeed some risk involved in setting up an arbitrary standard before undertaking such research. But that risk can be avoided if a quantitative study is based upon the classification of the subjects by non-linguistic criteria. Fortunately, the material for such a study is available in the field records of the several American dialect atlas projects. The materials of one atlas, that of the Upper Midwest, are at present the only ones immediately accessible, either in the published second volume of the Linfuijiic ~4tlas olllle llpper Midwest (1975) or, for a few items, in the files themselves. As was described in the first article of this series, the data were obtained between 1947 and 1953 by fieldworkers interviewing 208 lifelong residents of their communities in Minnesota, Iowa, North Dakota, South Dakota, and Nebraska. On the basis of their age, amount of schooling, and social life and status they are placed in the three categories familiar to students of dialectology: Type I, seventy years of age or older with no more than grade-school training; Type II, middle-aged, with a high-school education or equivalent and a somewhat wider social background; Type III, with a regionally obtained college or university education and with fairly wide social contacts.

Sex-Linked Variation in the Responses of Dialect Informants Part 2: Pronunciation

categories of lexicon and grammar, the f indings in the data corpus used in the present study of sex-linked variation significantly strengthen the position of those who argue for the existence of a so-called women’s language. The word &dquo;so-called&dquo; is not intended to be pejorative, I must add. I use it because in this sense the locution &dquo;women’s language&dquo; has won out over such competitors as &dquo;sexist language&dquo; and because no definition of llaDluagel that I am familiar with in any branch of linguistics embraces what is essentially a set of variations within a given language. &dquo;Women’s speech* might be preferable, indeed. But I yield to current usage. Several sociolinguists have drawn attention to pronunciation as an area in which the reality of women’s language can be demonstrated. But in the absence of a supportive quantitative study in depth they have relied upon a few familiar illustrative examples. There seems to have been no awareness of the existence of a great body of meticulously recorded data in the files of the various linguistic atlas projects, data which, although admittedly collected for a quite different purpose, the description of regional differences, happily were taken from the speech of both men and

Writings on Canadian English, 1792-1975: An Annotated Bibliography. Walter S. Avis and A. M. Kinloch, pp. xi, 155. Toronto: Fitzhenry and Whiteside, 1979

piled by Walter Avis and A. M. Kinloch, is a curious enterprise. A bibliography can’t be treated like a work of fiction, for, although it has a tremendous cast of characters, it has no discernible plot or any continuity except alphabetical sequence, and the compilers cannot indulge in imaginative deviations from the hard facts of publication. Nor can it be treated like a work of non-fiction. It has no sustained and coherent content building up to a conclusion; indeed, it has no organization at all aside from that alphabetical order. But hold! t I recant. On second thought I must admit that this bibliography does build up to an implied conclusion; it may even be said to have a message. That message is simply what Avis has been declaring for many years, that the English language in Canada has its own distinctiveness and that, as Kinloch joins him in affirming, it demands the serious study and research which for too long it has been denied but which is beginning to appear. Briefly described, the publication consists of a brief preface, a list of

Dialects of American English. Carroll E. Reed. Pp. vii + 119. Cleveland and New York : World Publishing Co., 1967:

cans spoke alike. Nearly two decades before the founding of the American Dialect Society M. Schele de Vere’s Americanisms (1872) treated in detail and with many examples the variations to be found in many parts of the country. George Philip Krapp’s scholarly tomes, The English Language in America, a two-volume work published for The Modern Language Association by the Century Company in 1925, revealed an academic interest in regional speech that was matched on the popular level by the documented concern in the successive editions of H. L. Mencken’s The American Language (1919-1948), recently revised in a best-selling one-volume edition by Raven I. McDavid, Jr. (1964). But for classroom use teachers have had little to turn to. Thomas Pyles’s s