jeff_l
(Jeff_L)
#1
204 4 Nominal and pronominal morphology
In most of these cases (180.a-b), the Sg is V-final (all full V's except i are
exemplified). One possibility is to argue that all non-i V's map onto the final C
position of template -CvCC- as w (V-to-C mapping). One would assume that
an i-final Sg stem would map onto the third C of -CvCC- as y (hence -CvCy-),
but there are no clear examples, since the only instance of -CvCy- (180.e) has
no stem-final V in the Sg. The dialectal cases in (180.c-d) and the unique case
in (180.e) have C-final Sg stems (in two cases the final C is geminated). Since
these patterns are unproductive, their phonology is probably less than
transparent to native speakers. For (180.c-d) one could argue that the w of the
PI is a true filler w, not corresponding to any segment in the Sg. If w is the
default third C in -CsCC-, this would obviate the need for connecting PI w to
Sg stem-final V in (180.a-b). Since most Tamashek ablaut is compositional
rather than templatic, there is little supporting evidence from other
morphological domains as to the fine points of input to output mapping.
In (181) I present several minor patterns that, like the preceding cases,
show clear ablaut vocalism.
(181) Minor Patterns of Ablauted MaPl -an
singular plural gloss
a. kin terms [with gg degeminated and lenited to w]
a-daeggal l-dawl-an 'in-law'
[also pronounced 1-dul-an]
a-lsggss l-bws-an 'brother-in-law'
[less often i-laggus-aen, i-lsggas-aen]
b. Sg x, PI i
ά-sael l-sil-αη 'day'
c. Sg a, PI u
ά-γαη 1-vun-an 'well rope'
anu ύη-αη 'well'
d. Sg e, PI 9
ae-zerh l-zsrh-an 'shoulder'
e. Sg -CCuC-
e-jrew l-jarw-an 'river'
f. Sg and PI -CuC-
ά-duf Ί-duf-an 'marrow'
'Well rope' and 'well' in (181.C) actually show regular ablaut vocalism,
with α in the stem becoming u. 'Well' shows no Accent Reduction in the PI
