A Grammar of Tamashek (Tuareg of Mali)

(Jeff_L) #1
124 3 Phonology

(97) Perfective Verb (Surface Melody, no Stem-Initial Syncope)


lexical form PerfP

a. underived heavy (middleweight)
-dubvn-
-siwvj-

b. derived
-m-iru-
-m-uhvr-/-m-ihvr-
-s-irvd-

-aedobaen-
-aesewasj-

-aem-era-
-semm-ohaer-
-sess-oraed-

gloss

marry
'gesture'

'be open'
'be held in common'
'wash'

c. underived light stems
-uru- -ora- 'open'
-ulwu- -olwa- 'be spacious'
-usvl- -osael- 'run'

There is an asymmetry between the cases with e and those with ο in the
perfective. With rare underived exceptions like 'gesture', e is confined to
Mediopassive derivatives with -m-. Especially in the causatives, we often get
perfective ο even when the VblN and other forms point to lexical i. Leaving
the choice between e and ο aside, the question here is how to account for the
mid-height V's in the context of ablaut mechanics.
Synchronically, the most straightforward interpretation is that the mid-
height V's reflect a mix of Η and L features. The fact that the mid-height V is
flanked by two ae's in the trisyllabic perfectives suggests that a stem-wide
melody is at hand. This would make sense if we assume a lexical high V.
In other contexts the melodic L or Η trumps lexical vocalic features. For
example, -bvlulw- 'gape' with medial u has PerfP -aeblalaer-, which
preserves no trace of the rounding feature. By contrast, the perfectives in (97)
combine /u/ with L to give o, and HI with L to give e. In other words, the
surface mid-height V is a compromise, and rounding and backness values are
preserved.
For historical speculations see the end of the preceding section.


(98) V-Height Compromise


After reconfiguration (see below), middleweight verbs of the
shape -CuCvC-, and V-initial light verbs (e.g. -uCu-), whose first
V is a full high vowel {i u}, combine with an overlaid <L>
melody as jeo), respectively. In other words, they compromise
on vowel-height, while preserving the lexical rounding and
backness value.
Free download pdf