Urban sound change beyond city limits: The spread of the Northern Cities Shift in Michigan.

Abstract

The Northern Cities Shift (NCS) denotes a series of vocalic changes currently affecting American English dialects across the northern U.S. The shift involves movements of six vowels: /ae/, /a/, /c/, / scI/, /$\varepsilon$/, and /$\Lambda$/. Standard descriptions suggest these movements are causally linked such that the vowels are participating in a chain shift. This study investigates the status of the NCS using two small towns in southern Michigan as research sites and a total of 32 speakers as subjects. The phonetic variation associated with each of the six vowels was examined using a system of auditory coding supplemented by acoustic measures. A quantitative account that considers aspects of both the linguistic and social distributions of this variation is presented. On the linguistic side, particular attention is paid to the evidence regarding the phonological conditioning of the changes, and the influence of factors in the environments preceding and following the vowel is examined. The sociolinguistic analysis focused on the parameters for which the sample was stratified: age (comparing adults and adolescents), sex (comparing males and females), and town (comparing speakers from each of the two towns). The results identified a number of key phonological factors involved in conditioning the vowel changes. Particularly interesting, in light of previous research, is the finding that preceding (as well as following) consonants have a significant effect in conditioning the changes. Among the results on the social distribution, sex was often identified as a significant factor, with females generally showing greater use of innovative forms than males. A significant difference between the two towns was also often found, a result which is discussed in relation to the question of how the changes are spreading. Age differences were identified for many of the variables, and in some cases these revealed an unexpected pattern of adults leading adolescents. Many of the findings are brought together in a critical reexamination of the claim that the Northern Cities Shift is a chain shift. The present results and other evidence suggest that acceptance of this claim by previous researchers may have been premature.