|
|
|
RENDILLE (Kenya; Afro-Asiatic, Cushitic) |
For more information about metathesis in this language, click on the following links: |
| Summary: a. Metathesis occurs in contexts in which, by vowel deletion, a pharyngeal consonant occurs before another consonant. b. When a obstruent and nasal consonant would occur in preconsonantal postion as the result of vowel deletion, metathesis occurs. |
Conditions Motivation Related information Comments References |
Deletion occurs in both verbs and nouns
with the addition of a vowel-initial suffix (Heine 1976, 1978, Oomen 1981,
Sim 1981, Zaborski 1986). When the pharyngeal fricative 'H' would occur in
preconsonantal position as the result of vowel deletion, metathesis applies
and the fricative surfaces in prevocalic position, as in (a). In sequences
in which the pharyngeal follows a consonant, no metathesis occurs, as in
(b).
B. Obstruent, Nasal/Liquid metathesis When obstruent and nasal consonants would
occur in preconsonantal position as the result of vowel deletion, as in (c)
and (d), the obstruent/nasal and the liquid metathesize and the obstruent/nasal
consonants surface in prevocalic position (Heine 1976, Oomen 1981, Sim 1981).
When the obstruent /nasal consonants would occur
in prevocalic position after vowel deletion, metathesis fails to apply,
as in (e).
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
b. Metathesis involves only tautomorphemic consonants, as evidenced by the form [umaH-te] *[umatHe] ‘you/she begin(s)’. c. Pharyngeal/C metathesis occurs when the pharyngeal fricative would otherwise follow the low vowel [a] as the result of vowel deletion, e.g. babH-o (*baHbo) ‘armpit, plur.’. It fails to apply when the preceding vowel is non-low, e.g. liHti (*litHi) ‘rock’. d. Obstruent, nasal/C metathes applies when the obstruent or the nasal would otherwise occur in preconsonatal position as the result of vowel deletion, e.g. urgo (*ugro) ‘skinbag’, Harme (*Hamre) ‘shiver’. It fails to apply when the obstruent or the nasal would occur in prevocalic position after vowel deletion, e.g. Darge (*Dagre) ‘(I/he) be full’, adalme (*adamle) ‘(I/he) be ignorant’. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Acoustic and Auditory Similarity(Hume 1997, 1998): The contiguity of a consonant to a vowel with similar perceptual cues are a motivating factor of Rendille metathesis. The similarity between the pharyngeal fricative and a low vowel is well-established in the literature. In recent work, for example, the observed patterning of pharyngeal consonants and low vowels has lead to the proposal that both be characterized by the articulator feature [pharyngeal] (see e.g. Herzallah 1991, McCarthy 1994), characterizing sounds produced with a constriction located in the pharyngeal region of the vocal tract. The common articulatory basis of the two sound types results in muted formant transitions in a pharyngeal/low vowel sequence. This is clearly seen in the spectrograms showing the voiced pharyngeal fricative preceding [a], [i] and [u] (Klatt & Stevens 1969). While marked formant transitions are evident when the pharyngeal is adjacent to a high vowel, the transitions are significantly less salient when the vowel at issue is [a]. The syntagmatic contrast between a pharyngeal consonant and adjacent low vowel with respect to place of articulation is therefore relatively weak. Perceptual Enhancement (Hume 1997, 1998): It is in a context in which the place cues to pharyngeal place in the consonant are vulnerable that metathesis applies, with the consonant shifting away from this context . That is, the realization of the pharyngeal fricative in prevocalic position, as opposed to postvocalic, preconsonantal position, strengthens the perceptibility of the consonant. b. Obstruent, Nasal/Liquid Metathesis Perceptual Enhancement:(Hume 1998) Liquids have more robust internal cues than obstruents and nasals. This may be attributed to clear formant structure throughout the duration of the liquid, which provides strong cues to both place and manner of articulation. Nasals, on the other hand, are characterized by formants as well as antiformants, the latter having the effect, among other things, of lowering the amplitude of all higher formants (Fant 1960, Johnson 1996). A nasal’s internal cues provide information concerning manner of articulation and more weakly, place of articulation. These facts suggest that the nasal has less robust internal cues than the liquid. By having no formant structurs at all, fricatives and stops have less internal cues to both place and manner than nasals and liquids, although fricatives with frication noise have stronger internal cues than stops. Therefore, shifting the obstruent or the nasal to the prevocalic position, at the expense of the liquid with stronger internal cues, can be understood as strenghtening the perceptibility of the obstruent or the nasal. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hume (1997) suggests an account in line with the perceptual cue-based analyses , drawing on the observations of Beckman & Edwards (1990) concerning word-final lengthening. Segments in word-final position are generally longer than those in word-medial position, with lengthening being even more pronounced at the end of an intonational phrase. Along these lines, it is reasonable to assume that the perceptual cues to the pharyngeal’s place in Rendille are more salient in word-final position than in postvocalic, preconsonantal position, given the possibility of geater length. |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
H = voiceless pharyngeal fricative d' = voiced post-alveolar stop D = retroflexed post-alveolar stop Last updated: 3/19/2002 |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|