ABSTRACT

Standard analyses of the morphological structure of the verb in Chichewa, and other Bantu languages, have the verb comprising a verb root (VR) to which verbal extensions such as causative, applicative, reciprocal, passive, stative, etc. are suffixed, and to which prefixes are added. The prefixes, analyzed as clitics in some studies cf. Givon 1972, Mchombo (in press), include elements that encode information pertaining to agreement with the subject and the object, tense/aspect, negation, modality etc. The morphology of suffixation and prefixation (or cliticization) is sufficiently well established as to require no further comment. The following provides a typical example of such morphological organization:

1. Mkango u-da-omb-an-its-a alenje ndi asodzi
    3-lion 3SM-pst-hit-recip-caus-fv 2-hunter and 2-fishermen
The lion made the hunters and the fishermen hit each other.

In this example the VR –omb- ‘hit’ supports the reciprocal extension –an- and the causative extension –its-, the final vowel (fv) –a, as well as the subject marker ‘u’ and the past tense marker ‘da’. This paper will focus on the subject and object markers, which appear on the verb head. Chichewa shows both subject and object agreement in its verbal morphology. In finite verb forms the S[ubject] M[arker] is obligatory, while the single O[bject] M[arker] is optional. Bresnan & Mchombo (1986, 1987) analyzed these markers as incorporated pronominal arguments. The SM is ambiguously used for grammatical and anaphoric agreement whereas the OM is an incorporated pronominal argument. The evidence derived in part from facts about word order and the behavior the associated subject and object NPs in the presence or absence of the OM. Consider the following examples, adapted from Bresnan & Mchombo (1987):

2. a. Njuchi zi-na-lum-a alenje
   10-bees 10-SM-pst-bit-fv 2-hunters
   ‘The bees hit the hunters”

In the absence of the OM, the object NP alenje ‘hunters’ must remain in the post-verbal position, maintaining string adjacency with the verb. On the other hand, when the OM “wa” agreeing with alenje ‘hunters’ is introduced into the verbal morphology, the NPs display freedom of occurrence. Thus, all of the following sentences are grammatical and have the meaning that the bees bit the hunters.

b. SVO Njuchi zi-na-wa-lum-a alenje

c. OVS Alenje zi-na-wa-luma njuchi

d. VOS Zi-na-wa-lum-a alenje njuchi

e. SOV Njuchi alenje zi-na-wa-lum-a

Further, the presence of the SM and OM license the total omission of the actual noun phrases, as shown in (f) below:

f. Zi-na-wa-lum-a
10SM-pst-2OM-bite-fv
“They bit them”

The seminal work of Bresnan & Mchombo focused on the status of the SM and OM which are marked on the verbal head. Their effect on constituent order was confined to the relative freedom of word order that accompanied the presence of the OM, as shown above. This paper will take those observations a little further. Consider the following:

3. Njuchi izi zi-na-lum-a alenje awa
10-bees 10-prox.dem 10SM-pst-bite-fv 2-hunter 2prox.dem opusa  2SM-foolish
“These bees bit these foolish hunters.”

In these the subject and object NPs have internal constituents. Ordinarily, those constituents cannot be separated from the head noun:

4 a. *Awa njuchi izi zi-na-luma alenje opusa
         b. *Awa opusa njuchi izi zi-na-luma alenje

On the other hand, in the presence of the OM not only does the order of the constituent words of the sentences become free but, it seems that the internal components of the NPs tolerate discontinuity. :

5. a. Awa njuchi izi zi-na-wa-lum-a
2-prox-dem 10-bees 10prox.dem 10SM-pst-2OM-bite-fv alenje opusa 2-hunters 2SM-foolish

b. Alenje zi-na-wa-lum-a njuchi izi
2-hunters 10SM-pst-2OM-bite-fv 10-bees 10-prox.dem awa opusa 2-prox.dem 2SM-foolish

but note the following:

c. ?Izi awa opusa zi-na-wa-lum-a
10-prox.dem 2prox.dem 2sM-foolish 10SM-pst-2OM-bite-fv alenje njuchi 2-hunters 10-bees

The data above show possibility of discontinuity in the internal constituents of the NPs, in the presence of the OM. However, note that there appears to be an asymmetry in the behavior of the NPs in that discontinuity of the constituents of the NP linked to the SM is less preferable. This paper will investigate these discontinuous constituents, the relation between head-marking and the constituent order, and their analysis within the theory of lexical functional grammar.

REFERENCES.

Bresnan, Joan & Sam Mchombo. 1986. Grammatical and anaphoric agreement. Papers from the Parasession on Pragmatics and Grammatical Theory. Chicago Linguistic Society 22:2, 278-97. Chicago.
Bresnan, Joan & Sam Mchombo. 1987. Topic, pronoun and agreement in Chichewa. Language 63.4. 741-82
Givon, Talmy. 1971. On the verbal origins of the Bantu verbal suffixes. Studies in African Linguistics 2.2 145-63. .
Givon, Talmy.1976. Topic, pronoun, and grammatical agreement. In Li, Charles (ed) 1976. Subject and topic. New York: Academic Press 149-88
Mchombo, Sam. in press. Affixes, clitics, and Bantu morphosyntax. Perspectives on Cognitive Science 3: Language Universals and Variation. Ablex Publishing.

.

.