Search Results

The chapter offers background demographic details on New York's Latino population and an initial illustration of the variable use of Spanish subject personal pronouns. The difference between tendencies and categorical differences is stressed, as is the orderly heterogeneity of pronoun use. The concepts of performance grammar, linguistic variable, reference lects, bilingual lects, and envelope of variation are introduced. An extensive review is presented of the literature on dialectal leveling and language contact, with emphasis on the permeability of grammar as a theory‐dependent notion. The prediction that pronominal usage in New York will be shaped by language contact and dialectal leveling is made. The need for explanation in variationist research is stressed. Initial details on the sample and method for obtaining and handling interviews are sketched out.

Pronoun rates are especially high among those born in New York, those who came as young children, and those who have lived in the city for a long time. Popular and scholarly assumptions regarding limited proficiency and incomplete acquisition of a simplified Spanish by heritage speakers are considered. Coherence of the concept of incomplete acquisition is challenged. Incomplete acquisition characterizes all intergenerational transmission, even in monolingual settings, and simplification is seen as part of normal language change. Simplification in the history of Spanish is reviewed. Simplification is widespread in New York bilingual lects, but there is no connection between simplification and reduced proficiency or incomplete acquisition. The highest users of pronouns in New York are studied. No convincing evidence of reduced proficiency is detected.

Variable hierarchies and constraint hierarchies are first studied in baseline reference lects in both regions. Variable hierarchies are the same in both regions and remain the same cross‐regionally and cross‐generationally, providing evidence of continuity. A prediction about constraint hierarchies is made on the basis of dialectal leveling. Constraint hierarchies should undergo changes in ranking of constraints and in reductions of constraint range differences when the regions are compared cross‐generationally. Predictions are confirmed for most of the internal variables, but not for all. Strong evidence for dialect leveling is seen in the predicted widespread reduction of range differences and in rank changes in some variables, especially in the variable related to person‐number inflections and in the constraints related to second‐ and third‐person verbs.

Predictions are made regarding groups that should differ in pronoun occurrence rates if language contact is shaping Spanish in New York. Distinctions are made between partially overlapping but distinct groups: speakers of the reference and the bilingual lects, the first and second generations, those with greater or lesser exposure (newcomers, established immigrants, and the New York raised). Higher pronoun rates are found, as predicted, in the bilingual lects, the second generation, and those with greater English proficiency and with greater exposure. These predictions hold true as well within regional and SES subsamples.

The study of internal (grammatical and discourse‐communicative) independent variables is distinguished from external social variables. Multivariate analysis is now logistic regression. Explanations are offered for observed ranking of variables and constraints in reference lects. Variable hierarchies remain the same in bilingual lects, even into the second generation. Prediction is made and confirmed regarding constraint hierarchies keeping same ranking but undergoing reductions of range in the second generation. Rank differences in reference lects and rank changes in the second generation are studied in the variable based on person‐number inflections of the verb. Special attention is paid to ranking of second‐ and third‐person singular verbs. Results are analyzed in terms of evidence for continuity and contact‐induced change.