444 8 Verbal derivation
(453) Simple Versus Double Causative Prefixation
gloss PerfP LoImpfP VblNsimple prefix
a. 'wash' -aess-orasd- -s-arad- a-s-lrad
b. 'make curdle' -ass-aslaey- -s-aslay- a-s-ästaydouble prefix
c. 'make similar' -2es-s-alaeh- -s-as-alah- a-s-ass-ibhThe "double" causatives usually function semantically as simple
causatives. As indicated below, double prefixation is common in causatives
based on full-V-initial underlying verbs (§8.1.5). Where the choice between
simple and double prefixation is not determined by the shape of the stem, there
is a tendency for causatives based on transitive inputs to prefer double
prefixation; see §8.1.9, below.
8.1.4 Causative stem shapes
The MAN stem-forms of causatives are similar to those of other multisyllabic
verbs with similar canonical shapes. There are, however, several important
differences.
First, the ablaut formatives that can appear on the second syllable of
underived light stems (the e of the PerfN, and the combination of χ and χ in
the Resit), do not apply in the same way to causatives. In other words, even the
shortest causatives are heavy. In the PerfN, where underived PerfP -aCCaeC-
stems become -aCCeC- in the PerfN (e.g., war i-jjes 'he did not enter'), we
get no change from PerfP to PerfN in causatives of similar shape. Thus PerfP
-ss-jasn- 'make kneel' occurs without change in negative utterances: wser
0-ass-jaen 'he did not make kneel' (not #waer 0-aes-jen with e). This is
because an underived -vCCvC- stem is light, and its second vowel is
simultaneously the first postconsonantal vowel and the final-syllable vowel, so
it can be targeted audibly by PerfN ablaut formative e-pclf (§7.2.2.3). By
contrast, causative PerfP -ses-jasn- is based on a derived stem-shape -s-vjvn-,
which qualifies as middleweight (hence heavy), and its first postconsonantal
vowel is not its final-syllable vowel, so e-pclf cannot attach audibly. Since
the PerfN of causatives is always identical to the corresponding PerfP, the
PerfN is omitted in the paradigms below. A further corollary of the fact that
-ses-jaen- is from -s-vjvn- is that the Resit is -as-ijaen- rather than #-aes-jan-.
Interestingly, -as-ijasn- does not show gemination of the s. The initial schwa is
confirmed by 3MaPl as-ljasn-aen 'they have already made kneel'.
A second difference has to do with the treatment of the onset. In the
perfective stems, we have Stem-Initial V-Insertion, associated with either