Volume 33, Issue 2

The paper upholds the existence of a phonological unit smaller than the phoneme, serving as the component of the latter. This is the linguistic correlate of articulatory movements together with their auditory reflexes, i.e. Baudouin de Courtenay’ kinakeme. The acceptance of the kinakeme as the ultimate phonological unit makes the notion of distinctive phonemic features redundant in linguistic theory. The kinakemes of any given language form a paradigmatic system and combine syntagmatically in phonemes.

This eighth report on the ‘Paris Project’ deals with a computer-assisted analysis of the frequency of elision among 10,891 mute-e’s in natural conversations as a function of phonetic context (number and nature of consonants contiguous to the /Ó™/, position in utterance, presence of other mute-e’s in contiguous syllables) and selected paralinguistic factors (sex, age, occupation, loudness, articulatory intensity, syllabic rate, degree of formality, attitude and posture vis-à-vis interlocutor, and subject matter). A separate analysis is made for coupled liquid and mute-e elision in cases such as I’autre jour /lot₃ur/ and pane que /paskÓ™/. Finally, the frequency of occurrence of the hesitation vowel euh is calculated in the three demographic categories indicated above. The results of the investigation reveal a number of new details and indicate some modifications in the traditional rules for its maintenance and elision. The paralinguistic correlates confirm an important discovery that is emerging from our long-range research, namely that in the dialect in question the most diachronically advanced speakers belong to the central age groups (30–59 years), the youngest (20–29) and the oldest (60–69) manifesting marked conservative tendencies.

This study investigates the perceivability of coarticulated lip rounding in French. Nine utterances containing the clusters /kstr/, /rstr/, and /rskr/ followed by one of the vowels /i/, /y/, or /u/ in all possible combinations, were truncated at 4 different points before the vowel. Test items in each of the 4 groups therefore contained different amounts of information regarding the nature of the following vowel, due to coarticulatory influences of the vowel on the preceding consonants. Subjects were asked to predict the identity of the missing vowel. Results show that when segments up to and including at least half of the final consonant of the cluster are present, subjects correctly identify the missing vowel well above chance level. Several individuals were able to identify the vowel even when presented with shorter versions of the utterances. No significant difference in performance was found between French and English subjects, nor between subjects with and without phonetic training. Perceivability of individual features of the missing vowel is discussed. It is concluded that coarticulatory effects due to lip rounding (as well as to horizontal place of articulation) provide perceivable information at a level significantly above chance, and that this information may be used by the perceptual mechanism as an aid in speech sound identification.