VALENCY AND SYNTACTICAL RELATION IN BIMANESS

Made Sri Satyawati1, I Ketut Artawa2, I Wayan Pastika2, I Gusti Made Sutjaja2

1Postgraduate Program, Udayana University 2Department of Linguistics, Faculty of Letters, Udyana University Email: [email protected]

ABSTRACT

This study presents the findings and descriptions of the replies to several problems that have not been completely and deeply discussed in the researches previously conducted on Bimanese. The problems are related to micro-linguistic factors, namely valency and syntactical relation in Bimanese. Both deductive and inductive approaches were applied to obtain satisfactory results. The main theory employed in this study is Role and Reference Grammar Theory (RRG) by Van Valin and J. Lapolla. It was employed to completely analyze the collected data in accordance with the problems proposed in this research, and the inductive approach was employed to analyze the data in order to get novelties.

In this study, clause structure is given the first priority to discuss, followed by the discussion on operator, voice markers, nominalizers, and definiteness. Based on the predicate category, the clause in Bimanese can be constructed with the constituents that are under the categories of verb, noun, adjective, number, and adverb (prepositional phrase). Based on the clause analysis, it has been found that in Bimanese there are several operators, each of which has different functional boundary in marking the clause meaning. One operator may only sign nucleus, core (nucleus and argument), or core and periphery. Bimanese has also been identified to have four linguistic states expressed by verbs that are made to make sense based on state (Aktisontrat), achievement, and accomplishment. RRG classifies verbs into ten instead of four. However, in this study, to make the analysis easier, verbs are classified into four. The predicate in Bimanese can be both serial verbs and secondary verbs. It has also been found that the mechanism of change in valency is marked by the attachment of markers to the verbs resulting in causativity, applicativity, and resultivity. From those syntactical constructions, the syntactical relation in Bimanese can be clearly identified. The discussion on syntactical relation starts with SUBJ and OBJ followed by word orders. The word orders found in Bimanese are SPO, adposition of PREP-N, N-G, N-Adj, N-Dem, and question word position. In addition, four voices have also been identified in Bimanese. They are active, passive, middle, and antipassive.

Key words: morphosyntax, valency, syntactical relation, operator, state, aktionsart, lexical representation, thematic relation, voice, serial verbs, secondary PRED.

  • 1.    Background

Bimanese or commonly called Nggahi Mbojo is a language natively spoken by approximately 608,679 speakers in the eastern part of Sumbawa Island including Bima Regency, Dompu Regency, and Bima City (Syaqmsuddin, 1996: 13). The eastern part of Bimanese speaking area is the Komodonese speaking area and the western part is the Sumbawanese speaking area.

Bimanese is one of the local languages existing in Indonesia. It is a language that is really interesting to explore. However, few studies pertaining to this language have been conducted. Therefore, some micro linguistic aspects such as morphology and syntax need investigating more completely and consistently. As an illustration, in the previous studies no clear explanation about markers and their functions in Bimanese has been found. In fact, their forms, functions and meanings affect the clause structure and the constituents forming the clause.

Morphologically, the discussion is not only related to the forms and functions of the verbs, but the meanings of the verbs also become the focus of the discussion. The meanings of the verbs resulting in different types of verbs may affect the verb valency. And then, the difference in valency may result in various relations between the verb and its arguments.

This morphosyntactical study is focused on morphology and syntax. Therefore, the main problems discussed in this study are as follows:

  • 1)    What are the patterns of the clause structures in Bimanese?

  • 2)    As a language with no affixes, is it possible to determine the argument thematic relation role to the subcategory of PRED in Bimanese?

  • 3)    Is there any grammatical and lexical marker for realizing the mechanism of change in valency in Bimanese?

  • 4)    How is the syntactical relation in Bimanese expressed?

  • 2.    Theoretical Framework

The main theory referred to for analyzing the phenomenon in Bimanese, as mentioned above, is Role and Reference Grammar (RRG), which is developed by Van Valin and La Polla (1997) and is intended to develop the universality of languages

viewed from the clause structure in Bimanese. In the clause structure, it views the distinctions between predicate and non-predicate elements or between noun phrase and ad position phrase (preposition and postposition). The distinctions made and the other concepts of RRG may be employed to recognize the linguistic facts existing in Bimanese. In addition, it is also employed to determine the change in valency in Bimanese syntactical relation.

Compared with other formal theories, RRG is the proper one for analyzing the morpho-syntactical factors in Bimanese. It rises as a combination of some formal theories, as some of the terms used in it are derived from the theories previously introduced. This can be exemplified by the fact that in RRG there are three grammatical relations such as Subj, Obj, and Indirect Obj., while in Functional Grammar there are only two relations, that is, Subj and Obj. Furthermore, Relational Grammar views a clause only from syntactical function and Functional Grammar views a clause from two dimensions, that is, syntactical and semantic dimensions, while RRG analyzes a clause based on syntactical, semantic, and pragmatic aspects.

  • 3.    Research Method

According to Van Valin and Laq Polla (1997), the aim of linguistic study is to explain linguistic phenomenon. This includes the explanation of a language as an individual possession and the explanation of languages that is directed to the universality of languages. From such phenomenon, it can be recognized how a language is different from others.

This study attempts to describe the linguistic phenomenon in Bimanese especially those that are concerned with valency and syntactical relation. This qualitative, explainatory and synchronic study expresees and explains the data in accordance with the language reality. This is also called a field study since the data were natural and were directly obtained from the speakers (Hyman, 2001). He stated that many researches conducted so far did not involve human participation apart from researchers themselves. In addition, according to him, field linguistic study does not only involve the first person (the researcher), but also the second person for the data elicitation and the third person for observation so that the data obtained are accurate. This means that the quality and the

quantity of the data depend on (1) the researcher, and (2) the time and the speakers’ skillfulness (Mithun, 2001).

The data collected for this study are classified into the primary data (the data that are obtained through interview) and the secondary data (the data that are obtained from texts). The technique employed to analyze the data was descriptive-analytic, and the approach applied was deductive-inductive.

  • 4.    Discussion

One of the findings show that the clause in Bimanese is made up of PRED filled with verb and non-verb. The PRED filled with verbs produce intransitive, monotransitive and bitransitive clauses, whereas the PRED filled with non-verbs result in intransitive clauses only. Bimanese does not have any affix determining any change in valency and voice. For this purpose, the concept of operator is used instead. With reference to such a concept, seven types of operator were found in Bimanese used as the indicators of the elements of nucleus, core (nucleus and arguments), and clause (core and periphery). They are aspect, negation, modality, evidential, status, illocutionary and directional. Therefore, Bimanese is considered (1) a language focusing on informational aspect proved by three types of aspect markers such as ACTOR cross reference ku-/-ku, {ma-/-na}, {na-/-na}, nominalizer {ma-},{ma-}, {ra-}, and {di-}, and lexical markers such as wunga for ‘progressive’, nee for ‘future’, and ‘past participle’; (2) a language whose marking system is determined by the role of arguments filling in the Subj (ACTOR or UNDERGOER); (3) a language whose passive construction is only marked by {ba-}or whose agent markers are attached to the PRED, that is, nee for ‘future’ and wau for ‘past participle’.

The aspects determined by the three types of aspect markers are (1) imperfective (if the cross reference of the ACTOR appears before the PRED, {di-}and nee for ‘future’ are used), perfective (if the cross reference of the ACTOR is after the PRED, {ra-} and wai for ‘past participle’ are used), habitual (if the cross reference of the ACTOR is attached to the PRED, {ma-} is used), and durative that uses the lexical marker wunga for ‘progressive’. Furthermore, the iterative marker {ku-}attached after the PRED is also found in Bimanese.

The grammatical markers indicating aspects can function as the agent markers in passive construction and as the possessive markers as well. Moreover, some can also be used as nominalizer/relativizer. The agent markers and the possessive markers are marked by the cross reference of the ACTOR, the nominalizer and relativizer are marked by {ra-}, {di-}, and {ma-}.

The other operator category is negation, which is indicated by the lexical constituent wait ‘not’, watipu ‘not yet’ and laina ‘not’. The constituent wait or {ti}’not’ is used to negate noun, verb and adjective; laina ‘not’ is used to negate noun, while watipu is used to express unfulfilled activity. The operator which is in the forms of modality, status and evidential contains illocutionary force close to modality. The lexical markers au ‘what’, cou ‘who’ and the grammatical marker {ro} are used to indicate interrogative illocutionary force, while the lexical constituent aina ‘do not’ and the grammatical markers {ja}, {put}, {ni} are used to indicate imperative one. Directional operator is also found in Bimanese. How it is used depends on geographical area and direction. The markers used to express direction are di for the object located in the west, ele for the object located in the east, da for the object located in the north, and da for the object located in the south. Furthermore, the markers used to refer to geographical area are ipa for the area isolated by a mountain, a river, and sea, ese for the upland, awa for the lowland, dei for the hinterland, and ari for outside/more developed areas.

Bimanese also has voice markers such as {ba}for oblique, weki for reflexive, angi for reciprocal, {-wea}, {kai}, and labo for an increase in valency. The markers of relative, noun and definiteness are also found in Bimanese.

In RRG, a clause is composed of core (consisting of PRED and its arguments) and periphery. PRED is expressed by nucleus which is under category of verb and in Bimanese it can express four types of state. The meaning of verb introduced in RRG is based on the types of aksionsart such as state, activity, achievement, and accomplishment. Out of ten types of verbs whose meaning is based on aksionsart, only four are used in this study. They are used to determine the lexical representation of each verb in Bimanese. To determine the thematic relation of verb, two basic types of verbs, namely state and activity are used.

Based on what has been described above, the lexical representations of the four basic types of verbs are as follows (1) be’ (x,y), (2) PRED’(x), or PRED’ (x,y), and (4) be’ (x, [PRED’(x)] for state verb; do’ (x,[PRED’(x,y)] and do’ (x,) PRED’ (x) for activity verb; INGR PRED’(x) or INGR PRED’ (x,y) for achievement verb; BECOME PRED’ (x) or BECOME PRED’ (x,y) for accomplishment verb. Furthermore, the thematic relation of verb in Bimanese indicate that the verb can involve one or more participants whose roles are PATIENT, PERCEIVER, STIMULUS, COGNIZER, CONTENT, WANTER, DESIRE, POSSESSOR, POSSESSED, EMOTER, TARGET, EFFECTOR, GOAL, MOVER, STATIC, STATIC MOVER, PERFORMER, CONSUMER, CONSUMED, CREATOR, CREATION, and LOCUS.

The PRED in Bimanese can be in the forms of serial verb and secondary verb. The serial verb is made up of two verbs, one of which is lao ‘run’ and mai ‘come’, as in lao rai ‘go running’ and mai ngaha ‘come to eat’. The secondary verb involves depective, resultive, circumstance, and complement predicate. Three of them, depictive, resultive and circumstance, are called adjuncts and their existence in a clause is optional, and the other, complement, is obligatory and is called complement.

From the analysis of verb valency, it has been found that the mechanism of alteration in valency in Bimanese is marked by the increase in valency (causative and applicative) and the decrease in valency (resultative). Causative in Bimanese is reflected lexically, morphologically, and analytically. The lexical causative takes place when the constituent of PRED is a verb with causative meaning such as gili ‘to roll’ and dunggi ‘to push’. Morphological causative is formed by attaching the causative marker (ka-) to adjective such as in makarasona ‘to clean’, to noun such as in makacina ‘to become brother’, to number such as kaduana ‘to becomed two’, and to verb such as makapaliponga ‘to make to become rolling’. Analytic causative is formed by adding the verb ndawi both to intransitive and transitive clauses containing the verb kau. In addition, there are also causative constructions with semantic parameter known as (1) pure causative and permissive causative, (2) direct causative and indirect causative.

Applicative in Bimanese is marked by the prepositions labo and kai and the marker wea. The two prepositions appear in the predicate slot and may be marked by a cross reference marker and form complex verbs such as nika labo ‘to marry with’, doho

kai ‘to sit in’. Applicativity with the marker wea is formed by adding it to the verb as in ndawi wea ‘to make for’. The complex verb with wea can also be marked by a cross reference as in the two prepositions mentioned above. In Bimanese, the process of causativity is similar to the process of applicativity in that they are formed by adding arguments. In causative, the argument added is ACTOR, while in applicative the argument added is UNDERGOER whose thematic roles are benefactive, commutative, instrument, and theme. The argument with benefactive role appears in the applicative with wea , while the commutative and instrument roles appear in applicative with labo and kai.

Resultative as a process of the decrease in valency is derived from passive. The verb in resultative is marked by a resultative marker {ra} but the agent is omitted.

There are four types of voice found in Bimanese. They are active, passive, middle, and anti-passive. The PRED of all the voices are (1) noun, verb as the basic constituent which may be marked by an aspect marker and a cross reference marker; (2) basic transitive verb, either active or passive marked by causative, applicative, and benefactive markers. Generally, the transitive verb can be recognized through the meaning of the verb itself, if there is an argument in the form of direct object (UNDERGOER) after the verb. The aspect markers attached to the verbs in active transitive construction use ACTOR cross reference markers such as {ku/-ku}, {-mu/-mu}, and {-na/na}, while the aspect markers in passive construction are {ra} for perfective and {di} for imperative. The passive aspect markers may also be used in middle voice with transitive verbs.

Theoretically, middle voice is classified into periphrastic middle, morphological middle and lexical middle. However, based on the analysis of the data, in Bimanese there are only two types of middle. They are periphrastic and lexical middle. Bimanese does not have any grammatical marker indicating morphological middle. It only has lexical markers marking reflexive middle such as weki ‘self’ or kese ‘self’ and marking reciprocal ‘angi’. In addition to the lexical markers, reflexive in Bimanese can also be expressed by reflexive verbs marked by ACTOR cross reference markers. These markers are used to express that the action conducted by the ACTOR comes and refers to the ACTOR itself.

Based on the analysis of the form of middle action, the middle voice in Bimanese is divided into (1) physical care (performance) action, (2) action performing physical change, ,(3) action showing change in physical posture, (4)action showing translational movement, (5) action showing acquisition, (6) action showing emotion, action showing cognition, and (7) action showing spontaneity.

Based on the above findings, it has been identified that Bimanese has uniqueness and universality. What is unique and universal is identified by comparing Bimanese with the other languages such as Rongganese and Sasaknese. The universality is shown by the fact that (1) PRED is filled in by verb or other categories such as noun and adjective, (2) Subj. and Obj. are filled in by N/NP, (3) the word order SVO, (4) the word pattern in which N is before Adj and Prep. The things showing that Bimanese is unique are (1) the grammatical markers ‘clitics’ bearing information on the operator, especially aspect, and (2) the grammatical markers expressing voice.

  • 5.    Novelty

Not all markers existing in Bimanese were identified in the previous studies. The markers found out were usually called affixes. In addition, how the markers were named was not consistent. Furthermore, the interaction among the markers has not been completely described so far. However, based on the theory of operator introduced in RRG, the markers in Bimanese, which are numerous, can be more easily discussed. They are {mu-/mu}, {na-/-na}, {ba}, {ra-}, {di-}, {ma-}, {ka-}, {-ku1}, { -ku2}, {-wea}, {-labo}, {-kai}, {-ro}, {angi}, {weki}, {ba}, {wunga}, {nee}, {wau}, {ja-}, {-pu}, {-pu}, {-ni}, {ndawi}, ‘buat’, and {kau} ‘suruh’.

Through those markers, it can be concluded that :

  • a.    Bimanese has two types of markers. They are grammatical and lexical markers. The grammatical marker is also called clitics featured as follows (1) it has lexical meaning, (2) it can be attached to particular constituents, and (3) it does not derive and change the valency of the attached constituents.

  • b.    In the previous studies, {ku-/ku}, {ta-/-ta}, {mu-/mu}, {na-/-na} were called pronominal clitics and verb agreement. However, in this study they are called ACTOR

cross reference since they refer to the argument serving as ACTOR. Furthermore, the clitics {ra-}, {di-}, and {ma-} are called clitics of nominalizer. In addition, {ba} is called passive marker, {ka-}, {ndawi}, {kau} are called causative markers, {ku} is called aspect marker, {ku2} is called emphasizing marker, {-wea}, {-labo}, and {kai} are called applicative markers, {ro} is called interrogative marker, {angi} is called reciprocal marker, {wunga}, {nee}, {wau} are called aspect markers, and {ja-}, {-pu}, {-ni} are called imperative markers.

  • c.    Bimanese is a language which gives priority to the information on aspects. The tree types of aspect markers such as ACTOR cross reference, nominalizer and lexical markers prove this. The marking system in Bimanese is greatly influenced by the arguments of the Subject.

  • d.    If morphological markers are used as the passive construction standard, then certainly Bimanese will not have passive construction. However, when passive voice is discussed from the syntactic point of view, Bimanese has syntactic passive as the semantic role of the Subject as UNDERGOER and the oblique agent is marked by {ba}. This can be exemplified by Foo ede radari ba ina nahu ‘Mangga itu dipotong oleh Ibu saya.’ The passive construction found in this study is different from that found by Jauhari (2000). He argues that passive construction in Bimanese is formed by PRED affixes {ra}, {di}, and {ba-}. Nevertheless, this study shows that the primary features of passive construction are (1) the agent is marked by oblique marker {ba-} and (2) the role of the Subject is as UNDERGOER.

  • 6.    Acknowledgement

I would like to express my thanks to the Rector of Udayana University, Prof. Dr. dr. I Made Bakta, Sp. PD (KHOM) and Director of Postgraduate Program of Udayana University, Prof. Dr. Ir. Dewa Ngurah Suprapta, M.Sc., and Head of Doctorate Program of Linguistics, Prof. Drs. I Ketut Artawa, M.A., Ph.D. for the opportunity and facilities provided to me in joining the Doctorate Program. My thanks are also extended to Prof. Drs. I Ketut Artawa, M.A., Ph.D., as my promoter, Prof. Dr. I Wayan Pastika,

M.S. as co-promoter I, and Prof. Dr. I Gusti Made Sutjaja, M.A. as co-promoter II for the advice and guidance provided to me during my study. My high appreciation is also addressed to the Dean of the Faculty of Letters of Udayana University, Prof. Dr. I Wayan Ardika, M.A., Dr. Ni Made Danawati, M.S. My thanks and appreciation also go to all parties who have been very helpful during the completion of my study in the Doctorate Program of Linguistics of Postgraduate Program of Udayana University whom I cannot mention personally.

10