If every productive form of linguistic expression can be described by some idealized human grammar, an individuals's variable linguistic behavior (Weinreich, Labov, & Herzog, 1968) can be modeled as a statistical distribution of multiple idealized grammars. The distribution of grammars is determined by the interaction between the biological constraints on human grammar and the properties of linguistic data in the environment during the course of language acquisition. Such interaction can be formalized precisely and quantitatively in a mathematical model of language learning. Consequently, we model language change as the change in grammar distribution over time, which can be related to the statistical properties of historical linguistic data. As an empirical test, we apply the proposed model to explain the loss of the verb-second phenomenon in Old French and Old English based on corpus studies of historical texts.

This article examines the reduction of syntactic options in South Texas Spanish narrative discourse during the nineteenth century. I argue that nineteenth-century Texas Spanish made ample use of the absolute construction as an orientation strategy in narrative discourse. In the beginning of the century the absolute construction appeared quite frequently in the narratives analyzed, but by the turn of the century the construction had become virtually unknown. I argue that the loss was actualized through a series of reductive changes that were played out in the internal and external syntax of the construction. These reductive changes, I suggest, not only cut across stylistic boundaries, but also corresponded with social changes under way with the incorporation of the region into the United States. The notion of “historical generation,” as it has emerged in social theory, is invoked as a significant social variable in relation to the linguistic variation and change observed.

Investigations into dialect emergence are most often based on data from manuscripts and on comparative and internal reconstructions. Seldom does the opportunity arise to monitor the selection of competing norms during the emergence stage because the data to postulate the linguistic marketplace (and hence to know what forms were likely to have been in competition) are unavailable. The case of dialect emergence in Thyborøn, Denmark, over the past century offers just this rare opportunity. A historical demographic profile from the town's census data, dating back to its inception in the 1890s, enables a comparative analysis of the input dialects and variable linguistic forms that were in competition. It is possible to trace the linguistic and social variables at play during the emergence stage of this new dialect, following the tradition of research by Omdal (1977), Dorian (1981), Trudgill (1986), and Kerswill (1994b), among others. The findings contribute to an explanatory model of dialect emergence and transformation.

To begin, we review three theoretical problem areas in the field of research into phonological variation in children. Next, we present the results of a cross-sectional study of two groups of children, aged 6 to 7 years and 10 to 12 years, relating to the deletion of post-consonantal word-final /R/ in French (production and judgments of acceptability). In an experimental study, we then examine the mechanism involved in the learning of words with a variable /R/. Finally, the interpretation of the results within the framework of a cognitive conception of variation leads us to four conclusions: (i) children have a tendency to copy adult surface forms rather than to encode a variable rule; (ii) orthography causes the late encoding of certain variable /R/s; (iii) the establishment of linguistic factors precedes that of social factors; and (iv) age-related changes are not guided by the sociolinguistic value that groups consciously attribute to the variables.