Aspects of the Acoustic Analysis of Imitation 383- An exemplary study
My own work on imitation has explored the ability of a member of the general
public (e.g., not a professional impersonator) to imitate another dialect. This
case study^2 involves a 29-year-old psychologist called Noah born in Morgan-
town, West Virginia, an area near the border of the North and South Midlands
dialect areas of the US (Kurath 1949: Figure 3). He lived there until the age
of 23, but his parents were natives of Detroit, a decidedly northern speech
area. Because the respondent had adequate input of both “South Midlands”
(or, more exactly, “native Morgantown”) and “non-South Midlands” speech,
there is good reason to believe that he could have control over both varieties.
It is not unpredictable that his “usual” vernacular speech reÀ ects a conserva-
tive northern system, as will be shown subsequently, given the social status of
South Midlands speech in the community of Morgantown, the home of West
Virginia University. The question addressed in this study is how accurate
his command of those elements of South Midlands speech that do not appear
in his “usual” speech is. The respondent read a word list and a reading pas-
sage with items deliberately selected to reÀ ect differences in “Northern” and
“Southern” US speech. After some discussion, he was asked to re-read the
word list and reading passage in the “West Virginia” style of the people from
his home town. This word list/reading passage format allowed Noah to use his
mental representation of that dialect but prevented him from assuming a per-
sona such as “good ole boy sheriff” or using a catch phrase like “y’all come
back now.” It has been observed by Preston (1993, 1996) that a respondent’s
performance was improved by the use of a catch phrase or persona: “’non-
performed’ performances reÀ ect a rather more systematic knowledge of the
variety” (1996: 63).
Noah’s performances were compared ¿ rst to a “conservative” Northern
vowel system, presumably not unlike his parents’ and not unlike one he would
have learned as a “university-oriented” child in Morgantown, in spite of sur-
rounding Midland speech. Of course, his Detroit parents might have shown
some incipient tendency towards the “Northern Cities Shift” (e.g., Labov
1994: 177 et passim) and that possibility is taken into consideration here. Fig-
ure 16.1 shows a conservative vowel system for Untied States English, with
no inÀ uence of either the Northern Cities or Southern Vowel Shift (e.g., Labov
1994: 211 et passim).
Noah’s vowels are then compared to the con¿ guration of the so-called
“Southern Vowel Shift” (Figure 16.2). If his vowels in both his normal and
imitation systems correspond to the conservative system, we may con-
clude that there is little low-level skill in imitation (the claim of most earlier