4 Dennis R. Preston and Nancy Niedzielski
of language which distinguishes between near and true mergers can this
sequence of developments have occurred in the history of a single language
variety” (34).
In Chapter 2, Hay and Maclagan investigate the social and linguistic
conditioning of the intrusive /r/ (“clawring” for “clawing” or “ma’r and pa”
for “ma and pa”) in New Zealand English. They look at social factors as
well as those of the linguistic environment, particularly the emergence of
intrusive /r/ after /aݜ/ and alternative realizations of that diphthong locally
as well as the degree of “r-ness” of the variable. They ¿ nd intrusive /r/ to be
sensitive to lexical items, af¿ x identity, class, and gender and that the “r-ness”
of intrusive /r/ is sensitive to the same factors. In this chapter use is made of
the logistic regression option of the open-source statistical program R; since
logistic regression is the backbone of the VARBRUL/GOLDVARB programs
popular among sociolinguists for decades (most recently available for Macin-
tosh, PC, and Linux frameworks at http://individual.utoronto.ca/tagliamonte/
Goldvarb/GV_index.htm), it is interesting to see this established treatment in
new clothing. An R add-on making use of the full power of logistic regres-
sion, speci¿ cally written for variationists (Rbrul), is available at http://www.ling.
upenn.edu/~johnson4/Rbrul.R.
In Chapter 3 Roeder assesses the inÀ uence of one aspect of the Northern
Cities Chain Shift (NCCS)—the raising and fronting of /æ/—on a Mexican-
American community in Lansing, Michigan, USA. Such work as this brings
second language acquisition (and its reÀ exes in subsequent generations) into
the concerns of sociophonetics. Roeder’s major ¿ nding is that the NCCS is
indeed having an inÀ uence on the emerging vowel system of the commu-
nity, although details of that inÀ uence do not necessarily match up with the
surrounding community, a ¿ nding that supports Labov’s distinction between
those members of a speech community to whom a change is incrementally
transmitted over generations as opposed to those to whom it is diffused more
rapidly and usually with a loss of subtle environmental factors (e.g., Labov
2007). This chapter also distinguishes between those predictable or natural
phonetic processes that might have an inÀ uence on a system being acquired
from those that are idiosyncratic to the target system itself. Social factors
(sex, education) are also strongly correlated with the inÀ uence of the new
target norm (i.e., the NCCS).
Chapter 4 is a further examination of heritage language communities,
in this case a study of the degree to which North African (Arabic, Berber)
language backgrounds may account for rhythmic differences in the speech of
banlieue (multi-ethnic, immigrant, working-class suburbs) residents in Paris.
Fagyal begins with a careful analysis of methods used to compute cross-