162 Malcah Yaeger-Dror, Tania Granadillo, Shoji Takano, and Lauren Hall-Lew
that the Social Agreement Principle is likely to be inverted in this type of
genre rather than merely neutralized.
Our results do not support the hypothesis that a language’s default posi-
tion for simple NEG has an inÀ uence on prosodic strategies (much less that it
results from such strategies), but it does support a tentative conclusion that
speakers from speci¿ c ethnic or regional backgrounds differ signi¿ cantly
from each other within each of the societies studied.
- Where do we go from here?
The present study was initiated because the data from read sentences (so com-
monly used in phonetic analysis) differ radically from what was patently obvi-
ous from analysis of conversational interaction. Even today, many of those
who create industrial applications for speech assume that reading style differs
from, say, human-computer interaction or conversations between strangers in
only minor ways. However, these researchers now need to project what people
will say (and how they will say it) in an expanding array of different social
situations. It is sociolinguists who have the expertise to collect and analyze
data from an expanding pool of interactive settings in order to isolate the rel-
evant variables for future analyses of speech.
6.1 Incorporating the social into sociophonetics
The issue of social situation is of particular interest in sociophonetic analysis
of prosodic variation, particularly when, as in this case, the different societ-
ies are purported to have radically different ways of viewing the task being
accomplished. Preliminary evidence has shown that native speakers of dif-
ferent languages do not have the same rules for emphasis on negation. One
conclusion of Yaeger-Dror (2002) is that French speakers are perceived as
confrontational by Americans partly because they do not reduce the promi-
nence on negatives in informative turns, but only in remedial turns, while, as
we see here, Americans reduce NEG in informative turns as well, if the social
occasion itself is supportive. On the other hand, the French were much more
sparing of prominence in political debate than the Americans. In the present
instance, we ¿ nd that Japanese and Spanish speakers are even more strongly
inÀ uenced by the social setting than Americans.
Even in informative footing turns in friendly phone calls, the Spanish
speakers almost never use prominence on the negative, although newscasters