82 Rebecca Roeder
Figure 3.2 is the vowel chart of Solana, one of the two young women whose
pre-nasal /æ/ is both fronted and raised above non-pre-nasal /æ/. Solana’s
mean for /æ/ before nasals is very high in her vowel space, near both [e] and
[Ԍ]. Figure 3.3 shows individual tokens of /æ/ in Solana’s speech, as graphed
in the software Plotnik.
Except for the word dad, which contains /æ/ between voiced alveolar
stops—an environment that may cause fronting and raising because the
constriction involved in the articulation of /d/ causes a lowered F1 and a
raised F2—Solana’s pre-nasal tokens of /æ/ are the highest /æ/ tokens in her
vowel system.
A dramatically raised position for tautosyllabic pre-nasal /æ/ is common
in NCS speech (Labov 1994: 266). To explain this, Labov and others have pos-
ited that raising of /æ/ before nasals is a default feature in American English
that is simply exploited to a more extreme degree in the Northern Cities Shift.
However, Thomas (2001) found that some of the Mexican American speak-
ers of English he interviewed in Texas did not have an /æ/ that was raised at
all before nasals,^6 suggesting that there are dialects of American English in
Figure 3.2 Vowel chart for Solana, age 22, third generation. Pre-nasal /æ/ = æN.