In lieu of an abstract, here is a brief excerpt of the content:

Introduction This is the report of a project designed to measure change and variation in the lexicon by contrasting data collected in the mid-1930s with comparable data collected in 1990. It is a broad study, spread across 62,500 square miles and encompassing 78 speakers with birthdates ranging from 1847 to 1959. Responses to one hundred and fifty different questions, or linguistic variables, are tabulated for each speaker. The result are analyzed separately for the two sets of data collected in different decades in order to reveal synchronic patterns ofvariation by the social and regional variables of age, sex, race, education, rurality, and region. These two analyses are then compared to see how such patterns have changed. Finally, the words, or variants, used in the 1930s are compared to those given in 1990 in order to determine change in the vocabulary across the 55-year time span. Linguistic change can then be checked against the language variation observed previously in order to better understand the relationship between variation and change. Several issues of theoretical interest are explored. Among these are questions about the role of social groups in linguistic change and the amount of variation attributable to each of the categorizing variables. The amount of change that may be attributed to sociolinguistic variation and the efficacy of social and regional variables in explaining language difference will be explored as well. One corollary to the former is the question of the utility of an analysis of variation across age groups for studying change. Lastly, there are questions about characteristics of the lexicon, as opposed to syntax and phonology. Previous studies have shown that particular social groups seem to lead in linguistic change while others are more conservative. By noting which groups of speakers are associated with new words and which with older words, this study may provide additional evidence for the relation- 2 Introduction ship betweeen variation and change. Since this project analyzes variation for similar populations at two distinct points in time, it may shed light on how the influence of particular non-linguistic variables can differ over time. Speaker characteristics, such as ethnicity, may change in their relative importance to language variation, reflecting an increase or decrease in social polarization along those lines. Such changes have been the topic of much heated debate, notably the convergence/divergence controversy with regard to Black Vernacular English. The testimony of cultural historians is important in explaining why certain social and regional variables are more important than others. Also, cultural factors are often linked to preferences ror older or newer words. One issue that has not been widely discussed is the amount of variation that is attributable to any ofthe standard sociolinguistic variables and whether it may increase or decrease. Many sociolinguists are inspired by the hope that they can provide evidence of systematic order for speech differences that seem to vary chaotically. The questions for this study were selected because they would each provide evidence of variation, Le. there was more than one possible response for each item. It was not known at the outset of the project whether any of this variation was linked to specific groups of speakers. The entire range of responses can thus be compared to those that do show such a link with a particular group, as evidence for what percentage of variation in the lexicon can be explained through an analysis using the most common demographic variables. This is not possible for studies which examine only those language features that are known to correlate to such variables. The question of how much language variation can be explained by groupbased differences is akin to the question of how much language change is revealed by age-based differences. A comparison of the total set of lexical changes to those suggested by apparent time results within each data set will be of interest in examining the question of which age-related differences are actually indications of linguistic change. The amount ofdifferentiation in language as a whole can also be studied using this data. By comparing the number of responses to the same questions at different times, growth and decline in specific areas of the vocabulary and in the entire set of lexical items analyzed here can be delimited. The ability to assess the amount of variation at different points in time by examining the size of the vocabulary raises questions about how the lexicon is unique. Whether differentiation proliferates in the same way for pronunciation and grammatical features...

Share