Language Variation and Change

Research Article

Real-time evidence for age grad(ing) in late adolescence

Suzanne Evans Wagnera1

a1 Michigan State University


This study provides real-time support for the hypothesis, previously inferred from apparent time studies, that stable sociolinguistic variables are age-graded. Stable variables have been shown to exhibit a curvilinear pattern with age in which adolescents use nonstandard variants at a higher rate than adults do. An analysis of the morphophonological variable (ing) was carried out using recordings and ethnographic observations of 13 young American women during and after their final years of high school. Offering a detailed look at the late adolescent life stage, the study also explores speakers’ motivations for retaining or retreating from nonstandard variants as they prepare to enter adulthood. These are examined at both the group and the individual level. The results indicate that the degree of retreat from nonstandard variants is socially differentiated, in line with apparent time findings. Future enrollment in a locally oriented college, and alignment to a local ethnic network (Irish or Italian)—not social class—were the predictors of retention in high school.


I am grateful to Gillian Sankoff, William Labov, and Penelope Eckert for their support during the preparation of an earlier version of this study. Anthony Kroch asked many insightful questions that often redirected my thinking. Dominique Tobbell, Jane Machin, and Maya Ravindranath read all or part of the paper. My thanks go also to Katie Fraser and Christine Collins for their assistance. Finally, I must especially thank the anonymous reviewers whose thoughtful comments have led me to make (I hope) valuable improvements to the original paper. Any remaining oversights are solely mine.