7.4 Pronominal subject paradigms 433When a nonzero pronominal prefix precedes a C-initial stem, there are two
distinct situations. For adjectival C-initial verbs (and a small number of
V-initial adjectival verbs), subject prefixes are absent (§7.4.2). The regular
treatment, applicable to most V-initial verbs and to non-adjectival C-initial
verbs, is summarized in (442).
(442) Pronominal Subject Prefixes Before Non-adjectival C-initial Stems
basic form before C-initial verba. 3MaSg i- i-
b. 1P1 n- n- or na- or an-
c. 3FeSg, 2nd t- 0-There are actually rather few combinations directly relevant to (442).
Many verbs with basic lexical representations beginning with a C undergo
Stem-Initial V-Insertion (usually accompanied by either Stem-Initial Syncope
or ^-Gemination), so they end up as V-initial. The remaining C-initial verbs
include long imperfectives, whether beginning with the lexical C, or with
Lolmpf prefix -t-,. Thus, for LoImpfP -bdddaed- 'stands', we get 3MaSg
i-baddasd, 1P1 n-baddaed or na-baddaed or an-bdddaed, and 2Sg
0-baddaed-2ed. For verbs of the shape -vCi>, whose perfective-system stems
commonly drop their initial schwa, we likewise get combinations such as the
following with Resit -Id- 'have': 3MaSg i-ld, 1P1 n-ld or n-ald, 3FeSg 0-ld.
For deletion of t- subject prefix see §7.4.1.2, below.
The 1P1 variant an- is arguably just a phonetic realization of n- in the
relevant position. Example: waer e an-s-aenn 'we will not cook' (K dialect).
The main difference is between η - or an- on the one hand, and na- on the
other.
The variation between e.g. n-bdddaed and na-baddaed might be taken as
indicative of a dialectal difference in the form of the prefixes, C- versus Ca-.
In the case of n-ld versus n-ald, one possibility (suggested by my morpheme
breaks) is that the verb itself has C-initial and V-initial variants, depending on
whether a prefix is present. In this view, the prefix is always n-. In the case of
n-baddaed versus na-baddaed, there is no evidence for a V-initial variant of the
verb, so a stronger case can be made for a dialectal difference between C- and
Ca- forms of pronominal prefixes.
In the event we decide on Ca- for a particular dialect, we need to apply
VV-Contraction when this Ca- is followed by a V-initial stem. See §3.2.3.2
for discussion.
The dialectal variation between Ca- and C- prefixes is perhaps also
relevant to the status of -vCvC- verbs like -vwvt- 'hit' (§7.3.1.1). In T-ka and
several other dialects, the PerfP is -awaet-, as in lSg awaet-sev Ί hit'. In
K-area dialects, we get just waet-aer with no stem-initial V. A form like