A shorter version of sections 3 and

A shorter version of sections 3 and 5-7 was presented as “Regionality as an independent variable” at
Methods X, the Tenth International Conference on Methods in Dialectology, at Memorial University of
Newfoundland at St. John’s in August 1999. I benefited from comments by Joan Beal, Bill Kretzschmar,
Luanne von Schneidemesser and Walt Wolfram at the conference, and have tried to incorporate them in
this fuller version. Edgar Schneider and my associates in the Dialect Topography project (in note 3) offered
helpful criticisms of earlier drafts.
1 I am beholden to Helvi Helena Virkamäki for what I know about Finnish dialectology.
2 Von Schneidemesser (1979) devised a similar metric for a similar purpose in her survey of lexical
stability and change in Giessen, Germany. She classed her subjects in three categories of indigenousness
which she called A (equivalent to RI 1 in our terms), B (roughly RI 2-3) and C (roughly RI 4-7). Some of
her results are summarized in Barbour and Stevenson 1990: 108-09.
3 Chambers and Heisler (1999) includes a detailed discussion of the sociolinguistic setting in Quebec City
and the complex patterns of linguistic variation that are occurring there. This case study of bureau and the
third one (soft drink/pop) were first discussed there. I am grateful to Troy Heisler for his help in developing
them. This article substantively adds the statistical thresholds of the RI and the componential analysis of RI
Regionality Chambers—22
that follows from them in the final section. Christiane Richter de Poppe supplied information on the words
for carbonated drinks. I am also beholden to Gord Easson and Mary MacKeracher, research assistants in
the Dialect Topography project, for help with the analysis and interpretation.
4 In Montreal, the other major city in the province of Quebec, the avoidance of pop as the interloper word
is more clearly marked. There, the list of preferences (counted as in Table 7) is: soft drink 61.3%, pop
10.4%, soda 21.2%, sodapop 0.4%, cola 1.1%, coke 2.5%, and other brand names 2.9%. The Englishspeaking
minority in Montreal is much larger and much less fragmented than in Quebec City, with
proportionately fewer interlopers.