A Grammar of Tamashek (Tuareg of Mali)

(Jeff_L) #1
7.4 Pronominal subject paradigms 435

(444) W-Contraction (Stem-Final α Plus V-Initial Subject Suffix)


conversion distribution

/a-ae/ —> e a. augment verbs: all V-initial suffixes
b. light V-final non-augment verbs: lSg and 2Sg
subject suffixes
c. V-final non-augment long imperfectives: all
subject suffixes

/a-ae/ —» as d. light V-final non-augment verbs: 2MaPl and
3MaPl (but not lSg or 2Sg) subject, and
V-initial Participial suffixes
e. heavy non-augment verbs: all subject and
Participial suffixes
[note: the as may be subsequently lengthened to α if targeted
by an ablaut lengthening formative χ-pcl or χ-f]

Examples: for (444.a), aehnasffe-n 'they-Ma moaned'; for (444.b) ajlas-n
'they-Ma went'. I argued in §3.4.9.1 that the pattern (444.a) is the direct
contraction of the two V's, with α shifting to e under the influence of /as/
(which is slightly more front in articulation than α itself). By contrast, (444.b)
reflects an intervening rule of Prefixal α-Shortening, converting /a-as/ first to
/ae-ae/ before it surfaces as ae. This shortening rule also applies before C-initial
suffixes, as in sjlae-naet 'they-Fe went'.
In the short imperfectives, we get alternations for certain verb types like
3MaSg ad 0-aeks 'he will eat', lSg ad deks-ser, 3MaPl ad aksa-n, and 3FePl
ad aksa-naet. I analyse these as V-final stems with Shlmpf /asksi/. The /i/, an
underspecified high V, is deleted word-finally by Stem-Final i/A-Deletion (29)
(§3.1.2.4), appears as a before C-initial suffix, and contracts with suffix-initial
/as/ to produce a, or ae in the cases of lSg -asr and 2MaPl Imprt (~)-set
(§3.2.3.3, §7.3.1.3).
For the same V-final light verbs, I recognize an underspecified low vowel
/A/ in the LoImpfP, e.g. -jail- 'go' for /-jallA-/. The /A/ deletes word-finally,
appears as ae before C-final suffix, and contracts with suffix-initial /as/ as as in
all cases.
I segment lSg Shlmpf asks-aer but 3MaPl akss-n. I place the morpheme
break before -aer event though VV-Contraction has occurred, since the ae
quality is determined by the suffix. By contrast, with 2Sg -aed, 2MaPl -aem,
and 3MaPl -aen, when they contract to form a surface short V, the quality of
the short V is determined by the preceding stem, so I place the hyphen after the
contracted V. The other morpheme that behaves like lSg -aer in this respect is
2MaPl Imprt suffix -aet (§7.4.3). The hyphenation is of no real significance in
phonological derivations.

Free download pdf