Apparent time, real time and the critical period:

longitudinal evidence in Montreal French

The real-time study of Montreal French afforded by the existence of three corpora in which the same speakers were recorded in 1971, 1984 and 1995 has begun to reveal results relevant to the extent to which speakers are able to modify linguistic patterns acquired in childhood. In addition, it is possible to distinguish changes from above from changes from below. This paper examines two real-time phonological changes in Montreal French: one from above and two from below, as set out in Table 1.

Table 1. Three real-time changes in Montreal French, 1971 – 1995.

The change from above has been differentially adopted by members of the cohort of speakers followed across the 24-year span of their lives that we have access to. Previous research has examined the replacement of tongue-tip, alveolar [r] with the "posterior" (velar or uvular) [R] (Clermont & Cedergren 1979; Cedergren 1987, Sankoff, Blondeau & Charity 2001). Of the 60 Montrealers recorded in both 1971 and 1984, most upper-middle and upper class speakers under the age of 25 in 1971 were users of the innovative [R], but many lower middle class and working class speakers still used alveolar [r]. Re-examining the same people13 years later, several upwardly mobile individuals had switched over to [R] in their twenties and thirties, but most socially stable speakers had retained the older form.

In the case of the two changes from below, although we are convinced that the "change in progress" interpretation made by scholars who studied the 1971 corpus is correct, we are now in a position to examine how individual speakers have reacted to these two changes: the diphthongization of the vowels /o/ and of /eu/ followed by /r/ as in the words chose ‘thing’ and beurre ‘butter’. Cedergren et al (1981) extrapolated from the strong age effects evident among 52 speakers of the 1971 corpus to identify a change in progress in /o/ that would have begun between 1940 and 1945, and in the case of /eu/, at least a decade earlier/. For both variables, they observed greatly increased diphthongization among younger speakers.

Preliminary results indicate that the differential adoption of changes in progress already observed in the change from above is also characteristic of changes from below. However, given the different association of each type of change with social class and education, we find different speakers in the vanguard of change. Dramatic changes in some speakers are counterbalanced by stability in others. Against the backdrop of the 12 speakers followed across the 24-year span, two speakers are considered in more detail: one who experienced considerable upward social mobility and the other whose fortunes were much grimmer. It is suggested that the window of opportunity for linguistic modification in later life may be expanded when linguistic variables take on social significance, but that the critical period is firmly in place for most speakers in most cases.