Immigrants from one dialect region to another acquire features of the new dialect with varying degrees of proficiency. In modern societies regional mobility is commonplace, and for modern dialectology, involved as it is with variability, mechanisms of change, and adaptation, it is a rich source of hypotheses. This article postulates eight general principles by which immigrants adapt dialectologically to their new surroundings, based mainly on results of a developmental study of six Canadian youngsters in two families who moved to southern England in 1983 and 1984, with supporting evidence from several other studies. The principles provide a set of empirically testable hypotheses about the determinants of dialect acquisition.

Otto Jespersen and his work were well known and widely recognized in American linguistics during early decades of the 20th century, particularly by Leonard Bloomfield. This study investigates Bloomfield’s references to Jespersen’s works on the structure and history of English, the nature of language change, and foreign language teaching, and finds that Bloomfield’s few criticisms of Jespersen’s writings were concentrated in the brief period 1922-1927 and were focused almost entirely on issues of science and method as Bloomfield had come to define them at the time. The study further documents and offers explanations for widespread underrepresentation of Jespersen’s work by American structuralists in the 1940s and 1950s and by historians of American structural linguistics. It concludes with a brief account of the renewed interest in Jespersen during recent decades.

In a ‘double-object’ construction, which of the NPs after the verb is the real object? Contrary to standard assumptions, I shall show that it is the second NP, so the first is the (traditional) ‘indirect object’. This finding is important because it challenges the hypothesis that grammatical relations can be shown configurationally and supports the competing claim that grammatical relations are basic. The paper also suggests why judgments on some constructions are so divided; the reason is that three different grammars are all compatible with the same basic data, and differ only on the relatively rare patterns where the variation appears.

On the prevailing view of ultimate attainment in second language acquisition, native competence cannot be achieved by postpubertal learners. The present study offers convergent experimental evidence which suggests there are exceptions to this generalization. At the same time, early arrival in the host country—even if past puberty—correlates with attainment of native norms on a variety of measures. Also investigated are the loci of competence differences in the syntax of the target language (French). Contrary to findings by Coppieters 1987, experimental performance is not predicted by the status of a given linguistic variable as within or outside the theoretical domain of Universal Grammar.

The Eyak language of Alaska now has two aged speakers; Mandan has 6, Osage 5, Abenaki-Penobscot 20, and Iowa has 5 fluent speakers. According to counts in 1977, already 13 years ago, Coeur d'Alene had fewer than 20, Tuscarora fewer than 30, Menomini fewer than 50, Yokuts fewer than 10. On and on this sad litany goes, and by no means only for Native North America. Sirenikski Eskimo has two speakers, Ainu is perhaps extinct. Ubykh, the Northwest Caucasian language with the most consonants, 80-some, is nearly extinct, with perhaps only one remaining speaker. Here we might be accused of jumping the gun, prematurely announcing the extinction of a language, since—as I heard somewhere—two or three more speakers of Ubykh had reportedly been found. But what difference does it make in human history that a language became extinct in 1999 instead of 1989? What difference does it make if the youngest speaker is 90 or in fact 9? Only 81 years in the date of the inevitable extinction of the language, a mere moment in human history— though a crucial moment for linguists today, as we shall see.

Phonological underspecification for English vowels is addressed using data from language production. Patterns in speech errors induced by means of the experimental SLIPS procedure argue for radical underspecification; the underspecified vowel place features are [-high], [-low], [-back], and [-round] (as in /ε/). Similarity effects in naturally-occurring speech errors lead to the same conclusions. With respect to the question of why the underspecification of place features is consistent across languages for consonants but highly variable for vowels, it is argued that the difference between vowels and consonants is rooted in similar crosslinguistic differences in phoneme frequencies.

Nonlinear phonology has proposed several new approaches to the issue of ‘ghost segments‘—segments which have phonological effects of various kinds, but which either never surface or surface only under special circumstances. Mobile vowels in Polish, traditionally known as yers, fall into this category. They have recently been analyzed in terms of syllable-driven epenthesis, as empty vowels, and as floating vocalic matrices. The present paper examines these proposals, points to their problematic aspects, and offers a novel feature-geometric interpretation of yers. They are viewed as empty root nodes whose vocalization, determined by syllable structure, permits the full prosodification of consonants.

Current versions of feature geometry generally locate the features [consonantal] and [sonorant] as annotations on the root node, rather than as normal dependents of that or some other node. This geometry is intended to reflect the observation that, unlike other features, the major class features do not participate in phonological processes such as assimilation (spreading of a feature) or dissimilation (delinking of that feature). This article disputes the empirical observation concerning [consonantal]: several examples are provided where vowels or glides become consonantal next to consonants (assimilation) and where consonants become vowels or glides next to consonants (dissimilation). The feature geometry is modified to allow for such processes by placing [consonantal] as a daughter of the root node.

A theme running through much of the functionalist literature in linguistics is that grammatical structure, to a considerable degree, has an ‘iconic’ motivation. This theme can be distilled into three rather distinct claims: (1) iconic principles govern speakers’ choices of structurally available options in discourse; (2) structural options that reflect discourse-iconic principles become grammaticalized; (3) grammatical structure is an iconic reflection of conceptual structure.

After presenting numerous examples from the functionalist literature in support of the idea that iconicity is widespread in language, I argue that claim (1) is irrelevant to generative grammar; claim (2), if correct, poses no challenge to generative grammar, despite a widespread belief to the contrary; and claim (3) has literally been built into standard versions of generative grammar. I go on to discuss the implications of iconic relations in language for the autonomy hypothesis and, at a more speculative level, for the evolution of language.

In schools, from kindergarten through high school, the language of instruction was English. When students who had been taught in English left school, they were speaking English. When they married, they spoke English to their children. ‘Indians’ no longer spoke their native languages as their primary means of communication.

This was the perceived state of affairs in relation to the Hualapai language in the mid 1970s. Many members of the community thought that English was taking over their ancestral language and that their traditions were about to disappear. In response to this threat of rapid language decline, a long and tedious process of forming a community language team began, with the Hualapai Bilingual/Bicultural Program as its central force.

In Chris has left York, it is clear that the event in question, Chris’s leaving York, has occurred in the past, for example yesterday at ten. Why is it impossible, then, to make this event time more explicit by such an adverbial, as in * Yesterday at ten, Chris has left York? Any solution to this puzzle crucially hinges on the meaning assigned to the perfect, and the present perfect in particular. Two such solutions, a scope solution and the ‘current relevance’ solution, are discussed and shown to be inadequate. A new, strictly compositional analysis of the English perfect is suggested, and it is argued that the incompatibility of the present perfect and most past tense adverbials has neither syntactic nor semantic causes but follows from a simple pragmatic constraint, called here the POSITION-DEFINITENESS CONSTRAINT. The same constraint also makes an utterance such as At ten, Chris had left at nine pragmatically odd, even if Chris indeed had left at nine, so that the utterance is true.

A theory is proposed that unites the universal locative paradigm consisting of the predicate locative, the existential, and the ‘have’ predication. This is accomplished by deriving all three from a single underlying structure in which a preposition is the head of the predicate phrase. The existential and the ‘have’ structure, both having a derived locative subject, are differentiated from the predicate locative by the definiteness effect. ‘Have’ predications of various kinds, including those with ‘have’ copulas, are claimed to differ from existentials because their locative constituent contains a [+human] argument. Apparently disparate facts from many languages are accounted for and a general explanation is offered for the distinction between a ‘have’ copula and a ‘be’ copula.

This paper offers a critical review of a series of experimental studies of first-language acquisition of anaphora. These studies provide converging evidence that children acquiring English respect Principle C of the Binding Theory—specifically, a form of Principle C in which an r-expression is pronoun free. Principle C is respected continuously across overt syntax development (from 3 to 7 years). Although knowledge of Principle C interacts significantly with both directionality and pragmatic context, it is independent of these. Initial results on acquisition of anaphora in Japanese are also reported, suggesting that, subject to further research, early knowledge of Principle C may prove to be universal.

The scene is Nicaragua's Atlantic Coast, the eastern half of this Central American country. It is a multilingual region where, besides Spanish, English Creole and various indigenous languages are spoken by populations of speakers ranging from tens of thousands of Miskitu (Misumalpan) to thousands of Sumu (Misumalpan) to barely two dozen Rama (Chibchan) and a mere handful of Garifuna (Arawakan).

The time is the decade of the 1980s and of the Sandinista Revolution, during which the legal, educational, and social status of these coastal languages changed through the process of establishing an autonomy statute for the region, a development that is sometimes referred to as the second Sandinista Revolution. This Autonomy project was a response to the war situation that developed on the Atlantic Coast, as both indigenous and Creole people rejected the Sandinista Revolution in its original form.

This essay will deal with a special type of center devoted to the documentation and teaching of the linguistic traditions of contemporary Native American communities. In my remarks, I will concern myself with the needs that such a center would satisfy in North America—in particular the United States, with which I am most familiar, although some of what I will have to say applies in other parts of the world, of course. While centers possessing many of the qualities I wish to champion here exist in other parts of the world, an impressive example being the Centro Linguistico Francisco Marroquin in Guatemala, there is none that closely approximates this model in the United States.