Multifunctional Morphemes jə- and -m in Ezha: An Ethio-Semitic Language

This paper is meant to indicate the different functional tasks covered by the two multifunctional bound morphemes: jәand -m in Ezha. The former is found to mark accusative, genitive and dative cases. The latter, on the other hand, renders the functions of contrastive focus marking, converb marking, declarative clause marking in the affirmative perfective, and coordination. After a thorough description of their functions, it is attempted to propose as to whether the various roles rendered by the two forms relate to distinctive homophonous morphemes or to a single morpheme. The required data for this study were collected from native speakers primarily through elicitation, and this method was supplemented by recording free narratives.


Introduction
Gurage languages are spoken in the Gurage zone of the Southern Nations, Nationalities and Peoples Region (SNNPR) of Ethiopia. The languages are also spoken in some other parts of the country by those Gurage and their descendants who left their homeland for different reasons like trading and education. Ezha is one of the different Gurage languages spoken typically in and around the town of Agenna. The area where Ezha is spoken is surrounded to the north-east by Muher, to the south-west by Chaha and to the east by Gumer.
The Ezha language belongs to the western group of Gurage languages as can be depicted by the following genealogical tree of the Ethio-Semitic language family adopted from Hetzron (1977).

Multifunctional morphemes jә-and -m in Ezha
Ezha is found to be a language which expresses multiple grammatical functions by employing a single morpheme (cf. Assefa, 2014). In this paper, two of such multifunctional bound morphemes, jәand -m, are discussed from a purely descriptive point of view. The issue of employing the morphemes jәand -m to mark different grammatical notions in Gurage languages has been addressed by different scholars such as Hetzron (1977), Fekede (2002) and Meyer (2011). However, these scholars simply indicated that different grammatical values can be signaled by the same morphological form; they did not explain as to whether these grammatical notions really relate to different underlying morphemes or to homophonous forms of the same morpheme. Besides, they did not address the morphemes in light of a single language with a considerable level of depth. This article entertains the issue under discussion in detail by taking Ezha as a target language, hence, a point of departure from the earlier works.

The morpheme jә-
The morpheme jәin Ezha is found to mark three cases. These are accusative, dative, and genitive.
Accusative case specifies the grammatical function of a noun as a direct object within a sentence. In Ezha, this case can be marked on nouns by the prefix jә-as demonstrated by the following examples.
(2) a. * ərɨddʒ-we jə-ʒəbb-ɨwe k ' ətt'ər-ə-n-m (* ərɨddʒɨwe jəʒəbbɨwe k' w ətt'ərənɨm) The expression (2a) is ill-formed because the accusative prefix is attached to the direct object noun which does not belong to the category [+HUMAN] though it is marked for definiteness, whereas the example in (2b) is ungrammatical as the direct object noun attaches the accusative morpheme without being specified for the feature [+DEFINITE]. Finally, the construction in (2c) is not acceptable because of the fact that the direct object noun takes the accusative marker without being [+HUMAN] and [+DEFINITE]. An expression with a direct object noun which is [-HUMAN] and [+DEFINITE] is grammatical without attaching the accusative prefix as in the constructions (2 d&f). In such cases, the direct object noun is determined by an object agreement affix which is obligatorily attached to the verb.
The accusative morpheme can be freely attached to proper names and personal pronouns as they are inherently definite. Hence, the expressions given below are well formed in the language.
'He killed the two black girls.' The accusative case morpheme marks the whole nominal phrase; hence, it is treated as a phrasal marker as it is attached to the left peripheral element within the phrase serving as a direct object.
The morpheme jә-also marks dative case in Ezha which relates to the indirect object of a sentence as it indicates entities that are indirectly affected by an action.
(6) a. abza jə-dəsta x w ett bɨrr ab-ə-n-m (abza jədəsta x w ett bɨrr ab w ənɨm) In these examples, the nouns to which the dative morpheme is attached refer to the entities which are indirectly affected by the action. In (6a), for example, the action of giving is performed on (two) Birr. This action indirectly affects Desta thereby making him the beneficiary.
The verbs in (6) are not marked by applicatives which indicate the notion of beneficiary; they bear only affixes that are co-referential with the nouns to whose advantage the actions are done. In some verbs, however, such structures possess applicatives together with affixes referring to the indirect object nouns as in the following example.
(7) s'əxaj jə-adot-əxita dəbdabbe nax-əttʃ-ɨr-a-m 'Tsehay sent a letter to her mother.' In this construction, -r of the verb is the benefactive marker, while -a refers to the indirect object noun adotәxita 'her mother' which itself is marked for dative case by jә-.
In situations where both a [+HUMAN] definite direct object noun phrase and an indirect object noun phrase co-occur in a single clause, both of the nouns can be marked for case. But, the verb can only be marked by an applicative followed by an affix referring to the indirect object noun; the direct object agreement marker never appears on the verb in such structures.
'Tsehay hit the boy for the benefit of her mother.' The same prefix, i.e. jә-, signifies genitive case which signals the idea that a noun is subordinated to another noun which it modifies. There are of course nouns which modify another noun without indicating a genitive reading. If the modifying noun is in genitive, it must be marked by the morpheme jә-; hence, the modification clearly shows a possessor-possessed relationship between the two nouns as demonstrated below. The genitive prefix disappears when other case markers are attached to the noun hosting the prefix. Here, the proper noun Desta is the possessor of the head noun әrɨddʒ 'boy'. However, the genitive marker does not surface in the structure since the possessor noun is marked by the ablative morpheme tә-.
In some constructions, in addition to the genitive prefix, possessive suffixes, which agree in person, number and gender with the possessor that they imply, can be attached to the possessor noun as in the following example. The possessive suffixes can co-exist with the genitive prefix usually if the noun to which they attach shows some kind of social relationship, be it kinship or another kind of affiliation. The co-occurrence of the genitive and the possessive affixes yields a connotation of double possession. In (11), for example, the possessive suffix expresses that gobbe 'brother' belongs to the possessor ɨjja 'I' on the one hand, and the genitive prefix jә-signals that bet 'house' is possessed by gobbe 'brother' on the other hand.
If a proper noun or a personal pronoun is the possessor of a head noun, the possessive suffix does not appear (cf. 9 a&b). The absence of possessive suffixes in such expressions pertains to the universal assumption that proper noun and pronominal possessors can only be marked once: either implied by the possessive suffixes attached to the head noun, the possessor noun being covert, or jә-is prefixed to the possessor without the addition of the possessive suffixes on both the possessor and the head nouns, but not both.
If a noun attaches a possessive suffix without the involvement of the genitive marker, the structure renders no genitive reading, but possession, as shown in (12). The fact that accusative, dative and genitive cases are signaled by the prefix jә-relates to case syncretism. The three distinct case values are expressed by a single form, and this is accounted for by the idea that two or more cases can be syncretic so that they tend to be marked by a single morpheme (cf. Baerman, 2008).

The morpheme -m
The other multifunctional morpheme attested in Ezha is the suffix -m. This morpheme marks focus, declarative clause in the affirmative perfective, converb and coordination.
The notion of focus could be classified into two general domainsː information or assertive focus and contrastive focus. The former pertains to the non-presupposed part of a sentence or new information, whereas the latter represents an element which is encompassed in a set of contextually given elements for which the predicate phrase can potentially hold (Kiss, 1995).
The suffix -m marks contrastive focus in Ezha; it can be attached to nouns, pronouns, numerals, adverbs and verbs of subordinate clauses. The focus marker with nouns and personal pronouns assigns contrastive focus by emphasizing an entity selected out of other possible alternative entities. The addition of -m to personal pronouns as in (13a) and nouns as in (13b) strengthens the restriction imposed on the agent to perform the action or on the object to be affected by the state of affairs implied by the predicate, hence, restrictive focus. (13a), for example, can indicate that nobody else just other than the already stated agent is supposed to accomplish the action denoted by the verb. If we change the structure in (13a) above to the form ɨjja әsɨjәnte 'I will buy it', there is room for possibility that the action of buying could be performed by an individual other than the speaker, or the speaker might accomplish the buying together with somebody else.
If a definite noun is to be focused, the definite morpheme -we takes precedence over the focus marker as demonstrated below. In genitive constructions, the focus marker attaches to the possessed noun and not to the possessor. In such structures, the focus marker has scope over the whole genitive phrase. If possession is expressed by employing possessive suffixes, the focus marker follows the possessive suffixes. Both the proximal and distal demonstratives attract the focus morpheme -m when they co-occur with a focused head noun. If the demonstratives are followed by the suffix -ota or -әxita to express exclusiveness, the focus marker follows this suffix. In situations where numerals modify a head noun, the focus morpheme is suffixed to the modifying numeral but not to the noun as shown below. The focus marker can also be attached to adverbs in order to indicate contrastive focus by emphasizing an adverbial constituent associated with the salient information, in contrast to others. Finally, -m can be suffixed to subordinate clauses. Here, -m attaches to the right periphery of the verb following agreement suffixes as can be inferred from the text presented below. The clauses written in bold possess the focus marker suffixed to the respective verbs following the 3MPL subject agreement suffix -o.
The second function which -m renders is converb formation. Haspelmath (1995) defines a converb as a nonfinite verb form whose main function is to mark adverbial subordination. Ezha expresses sequential events by conjoining a succession of sentences. In doing so, only the verb in the last sentence of the conjoined chain is a main verb, while the verb of the preceding sentence is marked as a converb. In other words, Ezha converbs, unlike main verbs, cannot be inflected for some grammatical notions such as negation, future tense and declarative clause marking.
The morpheme -m as a converb marker attaches to perfective, imperfective and jussive/imperative verb forms. The governing verb is mostly in the same base as with the converb. The examples illustrate that the converb marker, -m, attaches to verbs in any aspectual form thereby entailing the respective governing verbs to assume the same base form. The perfective converb form is usually used to express an action that is performed ahead of the action denoted by the governing verb (cf. 20a). However, it can also be used to express an action simultaneous with the action denoted by the governing verb as in the example below. The other grammatical function of -m is declarative clause marking in the perfective affirmative. The morpheme does not occur with other aspectual forms as well as in the negative. Note that the suffix -m attaches to the right most periphery of the verb following agreement markers. The suffix is non-existent in (22b) and (22c) due to the fact that these structures are in the imperfective and in the negative respectively. Thus, -m is considered to be an affirmative declarative clause marker in the perfective aspect.
Finally, -m serves as a coordinative conjunction. It is used to connect two or more words or phrases that have equal grammatical status as shown in the following examples. The coordinative conjunction -m is suffixed to both of the components that are meant to be connected together. In (23 a&b), it connects two noun phrases (noun-noun and pronoun-pronoun coordination), while in (23c) it coordinates two adpositional phrases. The structure in (23d) is ill-formed because it violates the principle of parallelism by virtue of involving an adpositional phrase and a noun as conjuncts. Attaching the coordinator to only one of the conjoined elements also results in an ungrammatical construction.
The morpheme -m also connects adjectival or adverbial phrases together. In this case, too, the conjunctive must be suffixed to both of the conjoined items. 'Desta is a kind and handsome boy.'

Conclusion
As discussed in the preceding sections, the Ezha language morphologically exhibits the expression of several grammatical functions by the use of a single bound morpheme. Among the morphemes of this kind, only jәand -m are discussed in this paper from the perspective of descriptive linguistics.
The morpheme jәmarks three cases: accusative, dative and genitive. One would pose a question like ''Do these grammatical functions relate to a single morpheme? or Do they correlate to different but homophonous morphemes?''. In this paper, this question is accounted for by the concept of case syncretism as stipulated by Baerman (2008). By tracing Baerman's assumption, the morpheme jәis viewed to be a single form which combines multiple case values which happen to be syncretic.
As regards the other multi-functional morpheme addressed in this article, i.e. -m, it marks contrastive focus, converb, declarative clause in the affirmative perfective, and coordination. This morpheme is considered to be underlyingly a single form which relates to focus marking. This single morpheme seems to render the rest of the aforementioned grammatical functions by way of signaling different readings in different contexts.
The rationale behind claiming that the morpheme -m is underlyingly associated with focus can be strengthened by the fact that there is some kind of emphasis involved in expressing the other grammatical functions other than focus. For instance, while marking declarative clause in the perfective affirmative, there is the notion of emphasizing the idea that the state of affairs expressed by the verb is completed as opposed to the case where we use the imperfective verb in which -m is not attached to the verb.