restricted access The fate of morphological complexity in language death: Evidence from East Sutherland Gaelic
In lieu of an abstract, here is a brief excerpt of the content:

THE FATE OF MORPHOLOGICAL COMPLEXITY IN LANGUAGE DEATH: EVIDENCE FROM EAST SUTHERLAND GAELIC Nancy C. Dorian Bryn Mawr College Simplification in structure, and also confluence between the local-language structure and the prestige-language structure, are usually predicted in language death as in pidginization. For a dying Scottish Gaelic dialect, speakers representing a broad proficiency continuum were tested in the two most excessively complex morphological structures the dialect offers, the noun plural and the gerund. The forms supplied by the semi-speakers at the bottom of the proficiency continuum show considerable simplification, as expected, but very much less than in 'classical' pidginization; confluence with English is also quite limited. Marked functional differences between this dying dialect and the typical pidgin are invoked to account for the difference in outcome.* 1. Introduction. In recent years, increasing attention has been focused on pidgin and creóle languages, in part as a kind of proving ground for both linguistic and sociolinguistic theory. Not least among the interests of recent writers have been issues of simplification and convergence as linguistic processes ; witness the very substantial third section of Hymes 1971. In its preface, Hymes writes of four '"moments" which a theory of pidgin and creóle languages must integrate'. Of these, the first two are: '(1) the universal tendencies to adapt speech, and varieties of a language, by simplification in some circumstances, expansion in others; (2) the occurrence of these tendencies in situations of language contact, so as to give rise to partial confluence of linguistic traditions'. While fully acknowledging the great value of pidginization and creolization studies in the investigation of simplification (and/or elaboration) and confluence in language use, I submit that the study of language death has much to offer in these same areas of investigation, and that so far it has been much too little tapped as a source of information in these matters. This is not to say that simplification and confluence appear in language death in the same degree, at the same points—or for the same reasons—that they do, say, in pidginization. Indeed, I hope to show in this paper that they need not. But this, I think, only makes it the more important that we include the special case of language death when we venture on the topics of simplification and confluence. It has, of course, long been recognized that dying languages characteristically show reduction of one kind or another—or, most often, of many kinds at once. Early reports tended to be quite general and to be impressionistically rendered. Thus Bloomfield 1927 characterized White-Thunder's Menomini as 'atrocious': * The research reported in this paper was supported by grants from Bryn Mawr College (1974) and the American Philosophical Society (1976). Michael Silverstein kindly read the first draft and offered a wealth of useful criticisms and suggestions, a number of which are incorporated in the present version. Several helpful suggestions offered by Suzanne Romaine are likewise incorporated; and the tentative nature of the conclusion reflects, I hope adequately, some wise cautions offered by Dell Hymes. A much abbreviated version of this paper was presented at the Celtic section of the Kentucky Foreign Language Conference in April, 1977. 590 THE FATE OF MORPHOLOGICAL COMPLEXITY IN LANGUAGE DEATH 591 'His vocabulary is small; his inflections are often barbarous; he constructs sentences of a few threadbare models.' Krauss 1963-70 offers many comments on failings in the Eyak texts he collected from a last few speakers, but most of them are general rather than specific ('inappropriate here', 'distorted', 'confused towards end', all from p. 44); and they are not systematized to show in what ways the language is suffering changes in its patterns. Miller (1971 : 1 19) offers a generalization about the terminal Shoshoni language: 'Younger speakers do not always have a complete control of the grammar and phonology, but the area which shows the greatest impoverishment is vocabulary.' Quite recently, studies have begun to appear which treat simplification and confluence in dying languages with something more nearly approaching the detail and scope with which they have been treated in pidginization studies, although there is still no full-length volume comparable to Mühlhäusler 1974. Thus Dressier...