The long tail of language change

In our new paper in Canadian Journal of Linguistics, “The long tail of language change: A trend and panel study of Québecois French futures“, Gillian Sankoff and I return to the future temporal reference data that we initially analyzed in Wagner and Sankoff (2011). The earlier study tracked a panel of 59 individuals’ use of inflected future (IF) forms, such as je chanterai ‘I will sing’ in sociolinguistic interviews recorded in 1971 and again in 1984. We found that overall, IF forms occurred proportionally more frequently in the speakers’ 1984 interviews than the 1971 interviews, at the expense of periphrastic future (PF) forms, such as je vais chanter ‘I’m going to sing’. This shift in preference was most marked for high socioprofessional status speakers who were already older (58+) in 1971. Since IF has been disappearing from French over the centuries, we referred to this individual pattern as ‘retrograde change’, i.e. linguistic behavior over the adult lifespan that is in opposition to the tide of community linguistic change.

In the new study, we re-analyze the panel speakers alongside a set of trend speakers: a cross-section of 34 individuals recorded 1971 compared with a different (but demographically matched) 34 individuals recorded in 1984. The trend sample analysis shows that the community use of IF remained stable over the 13 year period. So we reconsider our earlier diagnosis of ‘retrograde change’, and observe that these results are more consistent with individual ‘age grading’: age-associated shifts in language use that occur against a backdrop of community stability, and (potentially) repeat in every generation (Wagner 2012).

Additionally, although the re-analysis of the panel sample confirmed the Wagner and Sankoff (2011) results, this time we included contextual style as a predictor. We found that IF was more likely to occur in careful style (unsurprising), but only for those same older and higher status speakers. Overall it looks as if age grading is a reasonable interpretation: IF is a stylistic marker associated with older age and higher status. We conclude with some thoughts about how the retention of an old-fashioned form at a low level for stylistic purposes can contribute to the ‘long tail’ of its disappearance from the language over time. We also pose some questions, of which my favorite is this one:

…Montreal francophones inherit the remnants of a change in progress as if it were an effectively stable sociolinguistic variable, akin to the (ing) and (t,d) alternations of English (Labov 1989). It seems that they then reify these linguistic and social associations. In so doing, we could argue that they are exhibiting age grading – a regular, generationally cyclic relationship between age and a sociolinguistic variable (Wagner 2012) – rather than individual retrograde change. This inspires further questions: Should our nomenclature for the relationship between individual and community language change be expressed from the point of view of the language speaker or of the analyst? How slow must a change be for it to be subject to evaluation by speakers as a stable variable?

Sankoff & Wagner (2020: 267-268)

Labov, William. 1989. The child as linguistic historian. Language Variation and Change 1(1): 85–97.

Sankoff, Gillian and Suzanne Evans Wagner. 2020. The long tail of language change: A trend and panel study of Québecois French futuresCanadian Journal of Linguistics 65(2): 246-275.

Wagner, Suzanne Evans. 2012. Age grading in sociolinguistic theory. Language and Linguistics Compass 6(6): 371-382.

Wagner, Suzanne Evans and Gillian Sankoff. 2011. Age grading in the Montréal French future tense. Language Variation and Change 23 (3): 275-313.