GSC NOTES AND QUERIES

THE LONE PREPOSITION IN AKAN

By E. Kweku Osam
University of Ghana

A number of studies of Akan serialization (Christaller, 1875, Lord, 1973, 1982, 1989, 1993) have put forward the idea that certain verbs used in serial construction have grammaticalized into prepositions. As I have argued elsewhere (Osam, 1993, 1994a, 1994b, pp. 2-1-39), such conclusions are not warranted by the facts of the language. In my opinion, there is only one verb in Akan that has developed a prepositional meaning.

This morpheme is "wo." In simple sentences, it is used as an existential verb with a -- "be at/in," or with a possessive meaning -- "have." In negation, it is replaced by the suppletive form "nnyi" (or "nni" in the Twi dialects) as (1b) and (2b) show. In simple sentences "wo ," "be in/at" does not inflect for any tense-aspect, though in the speech of some speakers it can be marked for future tense as shown in sentence (3) which is a recording of an utterance said by a native speaker.

1a. "owo Kumasi"
3SG SUBJ-be in Kumasi
S/he is in Kumasi

1b. "O-n-nyi Kumasi"
#SG SUBJ-NEG-be in Kumasi
S/he is not in Kumasi

2a. "Ye wo sika"
1PLU SUBJ-have money
We have money

2b. "Ye-n-nyi sika"
We don't have money

3. "Hom bo-wo ho ewim bir yi?"
2PLU SUBJ FUT-Be there evening this?
Will you be home this evening?

The status of "wo" as a preposition becomes evident in a serial construction, as in (4), where it is preceded by another verb. In this context it cannot inflect for tense-aspect. More interestingly, it cannot be negated, not even by using the suppletive form used in negating it when it occurs as a verb in a simple sentence. In order to negate (4a), the negative morpheme has to occur on the initial verb only. When we try to negate "wo" in that context, it leads to an ungrammatical sentence as indicated in (4c).

4a. "Me-hyia-a no wo Kumasi"
1SG SUBJ-meet-COMPL 3SG OBJ in Kumasi
I met him in Kumasi

4b. "Me-e-n-hyia no wo Kumasi"
1SG SUBJ-COMPL-NEG-meet 3SG OBJ in Kumasi
I did not meet him/her in Kumasi

4c. "*Me-e-n-hyia no n-ni Kumasi"
1SG SIBJ-COMPL-NEG-meet 3SG OBJ NEG- in Kumasi
I did not meet him/her in Kumasi

Another fact to note about wo is that when used as a preposition it can even be deleted in some sentences. So (4a) can be rendered as (5) without any change in meaning.

5. "Me-hyia-a no Kumasi"
1SG SUBJ-meet-COMPL 3SG OBJ Kumasi
I met him/her in Kumasi

From the evidence provided by the language, it is clear that when "wo" occurs in a clause which already has a main verb it is a locative preposition; but if it occurs as the only verb in a clause then it functions as such, though it may lack most verbal properties. It is obvious that it has developed into a preposition in the context of serialization. The case of "wo" provides us with the evidence that in time, in the context of serialisation, verbs can develop into prepositions, although currently, there are not as many cases in Akan as claimed.

Based on the language evidence, it is time for teachers and students of Akan to acknowledge that there is a prepositional class among the lexical categories of the language. As of now, this class has a single member -- "wo ," "in/at."

References:

Christaller, J. (1975). A Grammar of the Asante and Fante Language Called Twi. Basel: Basel Evangelical Missionary Society.
Lord, C. (1973). "Serial Verbs in Transition." Studies in African Linguistics 4:3, 269-96.
__________. (1982). "The Development of Object Markers in Serial Verb Languages." In P. Hopper & S. Thompson (eds.), Studies in Transivity. New York: Academic Press, 277-99.
__________. (1989). Syntactic Reanalysis in the Historical Development of Serial Construction in Languages of West Africa. UCLA: PhD dissertation.
__________. (1993). Historical Change in Serial Verb Construction. Amsterdam: John Benjamins.
Osam, E.K. (1993) "Review" of C. Lord, Historical Change in Serial Verb Constructions. Journal of African Languages and Linguistics 15:2, 205-10.
__________. (1994a) "From Serial Verbs to Prepositions and the Road Between." Sprachtypologie und Universalienforschung 47:1, 16-36.
__________. (1994b) Aspects of Akan Grammar -- A Functional Perspective. University of Oregon: PhD dissertation.