Duration as a Determining Factor in Educated Edo English Rhythm Description

Abstract

Duration, the length of time used in speech production, is a significant phonetic stress-related criterion for determining rhythm, which is crucial for intelligibility in Standard British English (SBE). Some phonological studies on Yoruba, Hausa and Isoko Englishes claimed that these sub-varieties of Nigerian English (NE) tilt towards syllable-timing due to the absence of vowel reduction. Nevertheless, existing phonological studies on Educated Edo English (EEE) – a sub-variety of NE – has been on word and variable stress, while studies on its rhythm have been rare. This study, therefore, examined stressed and unstressed syllable alternation and duration of rhythm units in connected speech, to account for the description of EEE speakers as syllable-timed or stressed-timed, and the implication for NE rhythm description. Prince and Liberman‟s metrical theory, which explains the alternation of strong and weak constituents in SBE rhythm units, served as the theoretical framework. Purposive sampling technique was used to select 150 (75 males and 75 females) EEE speakers undergraduates from University of Benin and Ambrose Ali University as key informants. Two SBE speakers served as Native Baselines (NB). Speech Filing System (SFS) version 1.41 was used to record and measure informants‟ production of 35 rhythm units with anacrusis, 40 rhythm units, and 10 English words with syllabic consonants. The recordings were transcribed, subjected to perceptual analysis (frequency and percentages) and complemented with Gibbon and Gut Rhythm Ratio (RR) acoustic measures of 0-100, T-test and Mann Whitney U test at 0.05 level of significance. In rhythm units with anacrusis, perceptual analysis revealed that out of 5,250 overall expected occurrences, EEE speakers appropriately produced all the syllables with strong forms at 900 (17.1%) instances, bringing inappropriate use to 4,350 (82.9%). For rhythm units in connected speech, out of 6,000 expected occurrences, EEE speakers‟ appropriately produced 694 (11.6%), with inappropriate use at 5,306 (88.4%). Regarding the English syllabic consonants, out of 1,500 expected instances of occurrence, EEE speakers had 0.0%. From the perspective of Gibbon and Gut RR acoustic measures, out of 100RR expected absolute value, NB measured 162.97ms in rhythm units, with an overall mean of 81.5RR and a difference of 18.5RR, tilting towards stress-timing while EEE speakers‟ duration in rhythm units was 6166.25ms of 41.1RR, with a difference of 58.9RR tilting towards syllable timing. T-test revealed that there was a significant difference between EEE speakers compared to NB duration (4.571) and RR (81.5), with P-value <0.05 (0.000). Duration of males was 3057.125ms (20.4RR) while the females measured 3110.125ms (20.7RR). Mann Whitney U test showed insignificant difference for gender with P-value > 0.05, (0.482) and RR (0.293) which is greater than 0.05. The metrical analysis of EEE speakers established proliferation of Strong/Strong (S/S) juxtaposition of stressed and unstressed syllable in rhythm units, compared to the NB alternation of Weak/Strong (W/S) or Strong/Weak (S/W). Stressed and unstressed syllable alternation and duration of rhythm units in the connected speech of Educated Edo English speakers‟ do not conform to Standard British English pattern. This implies that Nigerian English rhythm tilts towards syllable-timing than stresstiming