Rhythm Types and the Speech of Working-Class Youth in a Banlieue of Paris 103
and Low’s (2002: 525) in taking as few “subjective and intuitive decisions
as possible when taking measurements.” However, since %V, ǻC, and ǻV
indices were calculated on utterances or prosodic phrases in previous studies,
the present study aimed at replicating calculations of these indices on pro-
sodic domains of similar length. Individual intonation phrases delimited by
silent pauses and major pitch movements in the text systematically overlapped
with syntactic clauses and short utterances, representing the closest possible
approximation to units over which rhythm type indices were calculated in
RNM’s (1999) and Ghazali et al.’s (2002: 332) studies. Whenever disÀ uencies
occurred, phrases shorter than ¿ ve syllables were collapsed with the shortest
surrounding phrase. The length of phrases obtained in this way ranged from
13 to 17 segments.
- Results
3.1. %V, ǻV, ǻC indices
Table 4.2 shows the total durations, the total number of intonation phrases,
vocalic and consonantal intervals, the mean articulatory rates and lengths
of phrases with standard deviations for each speaker. The total number of
measurements varied across speakers.^29 The difference between the highest
and lowest numbers of vocalic intervals was comparable to inter-speaker
differences in RNM’s 1999 study. Inter-speaker differences in the number
of consonantal interval measurements, however, were more than three times
higher than in their study. Possible reasons for this will be discussed later
in the study.
Articulatory rates varied between 10 and 13 segments per second for
most speakers. Since previous studies, including RNM’s, gave no informa-
tion about individual speakers’ articulatory rates, it is dif¿ cult to compare the
extent of these differences to previous observations. On average, speakers in
the AF group were one year older, articulated two segments per second faster,
and completed the reading task in average four seconds faster than speakers
in the EF group.^30 Laith, an AF speaker, had the longest intonation phrases
and Jacob, an EF speaker, had the shortest intonation phrases in the sample.
Average %V and ǻC values for AF and EF speakers together with average
values of various languages representing main rhythm types in RNM’s study
are represented in Figure 4.1. Since standard error values were not published by
these authors, they could not be represented on the ¿ gure.^31 AF and EF speakers
patterned closer to syllable-timed languages, such as Italian, Spanish, French,