Diagnostics of age-graded linguistic behaviour: The case of the quotative system 1 Isabelle Buchstaller Stanford University, California This article presents a cross-variety investigation of quotatives be like and go in apparent and real time. Distributional and attitudinal evidence points to a change in progress as the underlying process for the distribution of be like. However, there is also evidence of life-span change (Sankoff to appear). The patterning of go across age is much less clear-cut. It could be interpreted as age grading or as a change in progress. This paper discusses seemingly contra- dictory findings from U.S. and British English. It will be suggested that the distribution of go is due to unstable behaviour at both the individual and the community level. Furthermore, there is evidence that go has a latent presence in the linguistic repertoire and was picked up again after its frequency dipped due to the introduction of be like. This finding ties in with other reported cases of recycling of variables (Dubois and Horvath1999). KEYWORDS: Quotatives, apparent time, age grading, change in progress, adolescent trends 1. INTRODUCTION Sociolinguistics has amply demonstrated that ‘all change involves variability and heterogeneity’ (Weinreich, Labov and Herzog 1968: 188). Variability, on the other hand, while certainly having the potential to change, does not neces- sarily need to. When analyzing variable synchronic data, the researcher is thus faced with the problem of ‘how to distinguish between variability which leads to change and variability which does not’ (Cameron 2000: 249). Put differently, we have to ask ourselves whether or not the variability we witness in the data is stable across real time. Earlier work in sociolinguistics has made significant progress in devising strategies which help identify typical patterns of change that serve as a base for inductive reasoning. The postulate of apparent time, first outlined by Labov (1963), viz the use of the present to explain the past, has given sociolinguists a tool to reconstruct diachrony on the basis of synchronic evidence (Bailey 2004; Bailey, Wikle, Tillery and Sand 1991; Chambers 2004). Monotonicity in apparent time is usuallyconsidered a strong indication of change in real time (bearing in mind the possibilityof age grading, see below). Journal of Sociolinguistics 10/1, 2006: 3^30 # The author 2006 Journal compilation # Blackwell Publishing Ltd. 2006 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden MA 02148, USA.