jeff_l
(Jeff_L)
#1
608 10 Clitics
I have also heard the 2MaSg preverbal form as -\ha-kk with geminated kk.
The "full" forms are used in the following cases: a) when the dative
immediately follows an object clitic (in preverbal or postverbal position); b)
when the dative is postverbal and follows a high V {i u}. The "preverbal"
column shows the form taken by the clitic when they immediately follow
preverbal particles Negative waer or Future ad (which is always reduced to ά
before clitics), or a focalized constituent (e.g. mi 'who?' in questions). The
-\ha- of the full form is deleted in the lSg form, and reduced to -\ha-
elsewhere. The remaining columns show the forms in postverbal position after
a C (the h of the full form is dropped), and after α (the h is dropped and the
two α vowels contract into one).
Instead of taking underlying /-ha-/ as the lexical form of the Dative
morpheme and having the /h/ deleted in some positions, one could
alternatively take the morpheme as /-a-/ and posit an h-Insertion rule (like
that given for directional clitics in §10.2.1.1). However, the phonological
distribution of the /h/ in the datives is extensive, including some post-
consonantal positions where an h-Insertion rule would have little motivation.
This is unlike the case with Centripetal Aadd, where a minor h-Insertion rule
seems warranted.
Data from dialects other than T-ka (e.g. K-d R T-md) differ in the respects
indicated in (674).
(674) Dialectal Variants (non-T-ka)
a. no reduction of -\ha- to Aha- in preverbal forms
b. homorganic semivowel rather than h in lSg/ΙΡΙ clitic after high
vowel, hence ...w-α-... after u and ...y-α-... after i
Examples of (674.a) are below. For (674.b), compare R i-t-irdüwAa-hi 'he
believes me' with T-ka i-t-irdü-\ha-hi, and R i-t-iwly-W-naeY 'it is born for
us' with T-ka i-t-iwl-\ha-naer. In practice, given the paucity of verb forms
ending in a high V, these combinations are infrequent.
The preverbal forms are used when the clitic is hosted by a preverbal
particle, usually Negative wzer or a Future marker like ad (which reduces to a
before a clitic) Interrogatives like mi 'who?' and other fronted focalized
constituents can also be followed by a clitic. In my T-ka data, the usual -\ha-
formative is reduced to -\ha- in this position. No reduction occurs in examples
from other dialects. The lSg is -\hi in this position in all dialects.
(675) a. (T-ka) a-\ha-k 0-aenn
b. (T-md, R) a-\ha-k 0-aenn
Fut-\Dat-2MaSg 3MaSgS-say.ShImpf
'He will say to you-Sg.'
In postverbal position, the Dative formative lacks h after most C's (676).
