. f ugh, {m 51.25%..429v. “‘3 ”er! .1.» ’k : 5:; .15" 53: ~ “1.3.1 1 x '{r J. {g “3‘ fa I 9.7.,- .‘. 'r. 53 V w , /( >5? 2-, illlllllllNHIIIHIHWIIHIHIIllllUIIHIHHIJIUIHIJIW thesis 301402 7282 LIBRARY Michigan State University This is to certify that the thesis entitled DISCOURSE CONDITIONS GOVERNING ASPECT: 1 THE USE OF SUDAH AND TELAH IN INDONESIAN . I presented by Ruth Louise Abbott has been accepted towards fulfillment of the requirements for M.A. degree in Linguistics Major professor Date March 13, 1995 0—7639 MS U is an Affirmative Action/Equal Opportunity Institution a.“ ... — - ._77 - u, big—,WAV- .—.- -77 _ - _ . _—— __ _I PLACE It RETURN BOX to romovo this checkout from your record. TO AVOID FINES rotum on or baton date due. D‘ !,E f _ a. 5 DATE DUE DATE DUE stag L__JC:_II__J - [—Tj « -L__[__J f—TT—W—W MSU IoAn Afflrmotlvo Action/Equal Opportunity Intuition DISCOURSE CONDITIONS GOVERNING ASPECT: THE USE OF SUDAH AND TELAH IN INDONESIAN BY Ruth Louise Abbott A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Linguistics and Germanic, Slavic, Asian, and African Languages 1995 Dr. Dennis R. Preston ABSTRACT DISCOURSE CONDITIONS GOVERNING ASPECT: THE USE OF SUDAH AND TELAH IN INDONESIAN BY Ruth Louise Abbott Indonesian, an Austronesian language has many pairs of aspect markers which have not been delineated. This thesis examined three published short stories of Putu Wijaya analyzing the pair sudah/telah. Quantitative ‘Discourse analysis shows that telah is primarily used for durative situations while sudah his used for punctual and iterative (non-durative) situations. Hopper' and 'Lhompson's (1980) transitivity' scale was applied and a statistically significant difference was found between the transitivity of the two markers. Comparing the occurrences of each marker with the inherent aspect of the verb showed in more than 50% of the cases the choice of marker agreed with the inherent aspect of the verb. The remainder of the cases showed sensitivity to the larger discourse features of: story time lapse, referred-to action, and character point of View. Wijaya, Putu. 1988. Bomb. Madison, WI: Center for Southeast Asian Studies. Hopper, Paul, and Sandra Thompson. 1980. Transitivity in grammar and discourse. Language 56.2:251-299. Copyright by RUTH LOUISE ABBOTT 1995 For my creator, redeemer, sustainer, Lord and friend. iv ACKNOWLEDGMENTS I would like to thank Dr. Dennis Preston in particular for all his constructive comments, helpful suggestions and advocacy at every point. I would also like to thank the other members of my committee Dr. Mary Bresnahan and Dr. Matsuko Endo- Hudson for their expertise and criticism which helped create a tighter paper. I deeply appreciate Liana and Hans Dulimarta for their patience and help with the Indonesian glosses. I would like tn) thank those supporters vdu: made this mission a reality. Finally, I would like to thank Gary, Christopher, and Stephen Abbott for all their patience, help, and cooperation throughout this process. TABLE OF CONTENTS LIST OF TABLES .................................... ix LIST OF FIGURES ................................... x LIST OF ABBREVIATIONS ............................. Xi INTRODUCTION 1.1 Purpose ....................................... l 1.2 Defining Aspect ............................... 3 1.2.1 Aspect vs. Inherent Aspect .................. 7 1.2.2 Durative/ Non—durative inherent aspect ...... 8 1.3 The Data ...................................... 10 1.4 Method ........................................ 12 1.5 Outline ....................................... 14 ASPECT AND INDONESIAN NNNNNNNN NNNNNNN Introduction .................................. 17 Characteristics of Perfect Aspect ............. 20 .1 Comrie ...................................... 20 .1.1 Perfect of result ......................... 21 .L 2 Perfect of a persistent situation ......... 21 .1. 3 Perfect of recent past .................... 22 .1. 4 Experiential perfect ...................... 22 .2 R. Ross Macdonald on 'sudah' and 'telah' ................................... 22 2.3 Soebardi on 'sudah' and 'telah' ............. 23 2.4 Dahl ........................................ 23 2.5 Lloyd Anderson (1982) ....................... 26 .3 Indonesian as an aspect rich language ......... 27 .4 'Sudah' and 'Telah' as aspectual .............. 30 .4.1 Comrie's categories in Indonesian ........... 34 .5 Summary ....................................... 36 vi HOPPER AND THOMPSON TRANSITIVITY SCALE wwwwwwwwwwwwwww wWWNNNNNNNNNNNi-J meow (Dub-U) Introduction .................................. 4O Hopper and Thompson Transitivity Scale ........ 41 .1 Participants ................................ 42 2 Kinesis ..................................... 42 .3 Telicity (Aspect) ........................... 43 .4 Punctuality ................................. 44 .5 Volitionality ............................... 45 .6 Affirmation ................................. 45 7 Mode ........................................ 46 8 Agency ...................................... 46 .9 Affectedness of Object ...................... 47 .10 Individuation of Object .................... 48 Applying Hopper and Thompson to Indonesian....49 .1 Method ...................................... 49 .1.1 Specific difficulties with application of Hopper and Thompson transitivity scale ................................... 50 .2 Results ..................................... 56 Anomalous Cases ............................... 57 Summary ....................................... 61 DISCOURSE CONDITIONS GOVERN CHOICE OF SUDAH/TELAH 4.1 Introduction ................................ 66 4.2 Formulaic uses .............................. 67 4.3 Non—formulaic uses which conform to inherent aspect of verb .................... 69 4.3.1 Agreement with inherent aspect .............. 70 4.3.2 Inherent aspect as an incomplete explanation ............................... 76 4.3.3 Extra-clausal reference ..................... 80 4.3.4 Duplicates .................................. 80 4.4 Conditioning factors in the discourse ........ 82 4.4.1 Story time lapse ............................ 82 4.4.2 Referred to action .......................... 85 4.4.3 Character point of view ..................... 87 4.5 Summary ...................................... 90 CONCLUSION 5.1 Summary of findings ........................... 92 5.2 Suggestions for further study ................. 93 5.3 Theoretical Implications ...................... 94 vii 5.3.1 Hopper and Thompson ......................... 94 5.3.2 Inherent aspect vs. aspect .................. 95 5.3.3 Lexical theory .............................. 96 APPENDICES APPENDIX A BISMA ................................. 98 APPENDIX B MENOLEH KE BELAKANG ................... 108 APPENDIX C PELACUR ............................... 111 APPENDIX D DALLAS DATA ........................... 116 .APPENDIX E Se- AFFIX IS DEFINITE ................. 117 APPENDIX F ANDERSON SENTENCE EXAMPLES ............ 120 BIBLIOGRAPHY ...................................... 122 viii Table Table Table Table Table Table Table Ulb \10 LIST OF TABLES Comparison of aspect markers A summary of Anderson's semantic area of the Perfect Use of markers with durative aspect verbs Uses of sudah and telah Verbal agreement with Inherent aspect theory Sudah and verb aspects Telah and verb aspects ix 24 27 71 73 74 77 78 LIST OF FIGURES Figure 1 Axis of Durativity Aux BENE C-R DAT DEF(Df) Dur Excl EXPER For Fut IA IDSM IND Jav N N-Dur N—F NEG NM NOM O P PASS PFCT Pl PROC PROG PT Punc R.Cl Sun TMA Tot TR LIST OF ABBREVIATIONS recent past V first person w/o second person X third person agent auxilliary benefactive current relevance dative definite Durative exclusive Experiential perfect formulaic future inherent aspect indirect speech marker indeterminable Javanese noun non-durative non-formulaic negative not mentioned nominative object person passive voice perfect plural process progressive past tense punctual is a relative singular Sundanese tense, mood, total transitive clause marker aspect xi verb without meaning verb INTRODUCTION 1.1 Purpose Indonesian is aa language with the capability of making fine aspectual distinctions since there are at 1 The two studied here are least ten aspectual markers. 'sudah' and 'telah'. Speakers of Indonesian are unable to provide objective criteria by which the choice between the two can be made. In fact, when directly asked, they cannot explain any difference between the two terms. More than once, I have been told by Indonesian speakers that the words 'sudah' and 'telah' are interchangeable and that speakers choose one or the other for no reason they can articulate. Such comments obviously do not arise from a lack of fluency in the language. The discourse requirements for the choice of one form over another are simply below the level of consciousness. In most scholarly works, both are glossed "already," and even Indonesian grammarians have found it difficult to determine the differences. The premodifier telah also indicates completeness, and. is parallel in. meaning to sudah, from which it differs chiefly by being more formal. It. is therefore :more commonly found in written and formal spoken material; some authors prefer telah, others sudah, others use both in the same text seemingly interchangeably (Macdonald 1967:162). As can be seen from Macdonald, these forms have an aspectual meaning. The Echols and Shadily An Indonesian—English Dictionary third edition (1989:529,561) claims that 'sudah' is used for past states and 'telah' for past actions. This distinction, however, is oversimplified as will be demonstrated later (in chapter 2.) I will. clainu however, that. words are discrete units of information——no two are exactly alike. Bolinger amusingly commented: the thesis I am going to sustain is not one that would surprise the man on the street. Tell him that if two ways of saying something differ in their words or their arrangement they will also differ in meaning, and he will show as much surprise as if you told him that walking in the rain is conducive to getting wet. Only a scientist can wrap himself up in enough SOphistication to keep dry under these circumstances (1977:1). Although the pair 'sudah' and 'telah' have long been described as very similar in meaning, I believe the complexity involved in distinguishing between them is twofold : 1) the category being examined is the particularly complex one of aspect, and 2) the key to their distribution is discoursal. Recent advancements in discourse analysis, however, have made it possible to examine the contrast between these two words in ways which were not possible before. This thesis will discuss the discourse requirements which dictate the choice of one over another, using principallly quantitative analysis as a tool. 1.2 Defining Aspect Aspect has come to tm2<1f particular interest in the last decade as linguists work to decide several key issues: 1) the attempt to classify aspect in one domain of linguistics (n: another; 2n the standardization of the definition, and 3) the determination of the range of aspectual phenomena. Perhaps the most cited recent author on aspect is Bernard Comrie, whose AapegLL_ij;UnuthufijJHL;ml_Lhe SindmeerhaleuectJnLRelateLfimhlems—LLSIQL, differentiates tense from aspect in the following way: "Tense locates the time of a situation relative to the situation of the utterance" (2). There are of course, in individual languages, finer gradations in time than simply 'before' and 'after', but, according to Comrie, these are time only necessary tenses. Aspect CH1 the other hand, is defined by Comrie as "differing ways of viewing the internal temporal constituency of a situation" (3). This View of aspect suggests that it gives some internal order (or information about that order) to the situation. This View greatly contrasts with Anderson (1973), who asserts that, Aspect, I suggest, is concerned with the relation of an event or state to a particular reference point: it is located before (retrospective), after (prospective), around (progressive) or simply at (AORIST) a particular point in time (39). Basically, for Anderson, tense creates reference points and aspect delineates the relation of a state or event to one of those points. Another definition of aspect comes from Rafferty (1982), one which concentrates on the speaker's perception. "Aspect is a verbal category that defines the perspective from which the speaker views a state/ event/ activity" (66). This definition. seems too broad; in fact it seems broad enough to include tense (e.g. a speaker's view of a past event) or almost any other temporally related diectic expression. Rafferty's definition, however, captures such possibilities as 'punctual' and 'iterative' which Anderson's cannot, precisely because these notions are not directly related to a particular reference point. 5 Beedham (1982) combines what I consider to be the best of Comrie, Anderson, and Rafferty: Aspect, (as opposed to tense) is non-deictic and is concerned. with the internal temporal structure of a situation. It refers to the way in. which. the event is seen. to ‘pass through time, i.e. whether the event is conceived of as imperfective (implies duration), perfective (event viewed. as 21 single whole), iterative (implies repetition), inceptive (implies beginning of an event), and so on (1982:84). Beedham's definition is valuable in that it combines the notion of an internal temporal structure with a speaker's perception. (If this temporal structure. However, Beedham explicitly states that aspectual meaning is non—deictic, while it may be argued that the opposite is true. In context, aspectual meanings are non— referential, or relational, in time sense that they do not necessarily reflect the actual objective duration or boundedness of a state/event /activity in the real world, but rather reflect the exaluation of the speaker concerning the relationship of one event /state /activity to other events /states / activities in the discourse . . . . Aspect in this sense is deictic because it defines the speaker's spatio-temporal perspective for viewing the event /state/activity (Rafferty 1982:66). Givon also noted the overlap between aspect and tense, in the semantic space of aspect, nearly always some element of tense is also involved, in terms of establishing a point-of—reference along sequential time (1984:272). 6 Givon would appear to be in general agreement with Beedham on this point. Dahl has written one of the most comprehensive volumes on aspect, and his definition is nearly equivalent to Comrie's. Tenses are typically deictic categories, in that they relate time points tx> the moment of speech. Aspects on the other hand, are non- deictic categories (Dahl 1985:25). Dahl, then, sides with Beedham rather than Givon, preferring to consider aspects as typically non- deictic. Clearly, even among linguists who have focused on this problem, there is disagreement as to the :nature: of aspect, particularly' in the realnt of deixis. For this thesis, aspect will be treated according to Beedham's definition, (Hue which captures both the internal temporal structure and the importance of the speaker's understanding (n? the event. This places aspect at least partially in the realm of pragmatics, because the aspectual meanings often. make reference to known facts within the discourse. This paper will give evidence tx> support the rmfljrn1 that, "aspectual meaning of any one utterance must be derived from the whole, the context, not from an isolated sentence" (Rafferty 1982:67). 1.2.1 Aspect vs. Inherent Aspect In the linguistic literature, there have been a great number of pages devoted to the distinction between inherent aspect and external aspect. Inherent aspect (or W) is the necessary aspect that the verb carries with it as part of the nature of the action (i.e. 'sleep' jjsznecessarily durative, 'knock' is necessarily non-durative). External aspect ii; an interpretation imposed (n1 a real series CM? events. (E.g., the ongoing interpretation of 'Jack is sleeping' indicates its progressive nature while that of 'Esther 115 asleep' indicates that ea certain state has been achieved.) Another distinction found in literature on aspect (although largely ignored by Comrie) is that between aspect and aktionsart. In Brinton (1988), aspect is described as '%a:matter of the speaker's viewpoint or perspective on a situation" (3). For example, a speaker may choose to describe a situation as completed, ongoing, or just beginning. Aktionsart, on the other hand, is described as "an indication of the intrinsic temporal qualities of a situation"(3). Aktionsart, then is time description of the intrinsic nature of the action described by a verb. For example, in the verb 'knock' the aktionsart is punctual because the intrinsic nature of the action is that of 'happening in an instant'. The external aspect, however, could be iterative, as in knocking on a door, or habitual, as in the knocking on a door done by someone who visits often. Brinton's work argues that this distinction between aspect and aktionsart is crucial and that interpretation of the aspectual meaning of the sentence is dependent on the interaction between the two categories. This research also will provide evidence that there should. be a separation between the two notions. 1.2.2. Durative/ Non—durative inherent aspect Comrie combines both 'durative' and 'punctual' into 'inherent aspect.‘ Durative is "(a reference) to the fact that the given situation lasts for a certain period of time (1976:41)." Punctual is defined as "the quality of a situation which does not last in time (42)." Among other things, durative differs from punctual in the assignment of the iterative events; non—durative and non—punctual both include those things which are iterative. While most linguists who have investigated aspect speak (Hf the punctual-nonpunctual distinction, they ck) not rule (nu: the possibility of the distinction breaking along durative-nondurative lines. Bickerton, for example,(who found that the punctual-nonpunctual distinction is learned by children before the past-nonpast distinction) obviously allows for alternative divisions. A TMA (tense, mood, aspect) system may be compared to a cake, a cake that is always the same size, same semantic area. . . But a cake may be split up into five, or eight, or ten. among five, or eight, or ten TMA markers and/or combinations of markers. . . How much, and exactly what, is contained in each slice will be largely determined. by the number of slices (1981:90-91). As Bickerton suggests, in Indonesian, there are many aspect markers, so the semantic area of each marker is small, and the distinctions the language can make are finer than those of English. For example, along the axis of durativity, the split into durative and non durative does not exactly overlap with the punctual and non-punctual distinction (see Figure 1). Punctual Iterative Durative [ Non—Durative ][ Durative ] [ Punctual ][ Non-Punctual ] Figure 1. Axis of Durativity Durative includes only those events which last in time-— which are not instantaneous. For example, the act of running cannot be instantaneous; it must have occurred over some time. Non-durative includes iteratives which by themselves do not persist in time; however, due to the continuing nature (M? the events, they ck) cover some 10 span of time. An example of this would be the act of hitting; each blow does not take a long time, but taken together, the 'event' (or 'group of acts') may last quite a bit of time. In Indonesian, these iteratives are included III the non-durative semantic area. An Indonesian example of this is found in sentence P0052 which begins with the introduction: 1. Sudah sering aku bilang pada dia . . . already often I said to him I have often said to him In this sentence, the action is clearly iterative; the narrator did not speak to the person just once, but repeatedly. The action is marked by the 'sudah' form of the perfect which is used to mark punctual inherent aspect. In this sentence, however, it marks iteratives, so the distinction in Indonesian is not punctual/ non— punctual, but durative/ non—durative, since the iteratives are grouped with the punctuals. 1.3 The Data3 The texts are naturally occurring Indonesian data which were published in the form of short stories. In 1988, Putu Wijaya published a book of short stories edited by Ellen Rafferty, Bomb, as the third monograph in a series published by the University of Wisconsin 11 Center for Southeast Asian Studies. These stories were chosen because they were not a translation of some western work which might be given to extensive borrowing of forms and content. These are stories on Indonesian topics, and the author is a particularly good subject for study as he has spent much of his time trying to recapture Indonesian and particularly Javanese concerns in his works. His style (as described in Ellen Rafferty's introduction) reflects the three basic principles found in traditional Indonesian art: the multiple points of focus, the snapshot descriptions of events which are not obviously related, and the juxtaposition of the real world and the imagined world without any distinction made between the two. These are common practices 1J1 non—verbal arts, particularly paintings, of the area as well. Wijaya studied under W.S. Rendra from 1967 to 1969 where he developed his own style of incorporating traditional material with modern forms. SDI the early 1970's, Wijaya became involved vnifll the Jakarta Arts Council allowing his further development as an artist. Although primarily known for his dramatic contributions, he has held.:many' positions for non— dramatic activities: Sports Editor for Ekspres (1969- 1970), Board of Editors of Tempo (1971-1979), Managing Editor of Zaman (1979-1985). 12 He has also written many articles, reviews, essays, and short stories for such publications as: Sinanjarapan, Aktuii, Komnas, Hansen, W, Eemina, Kartini, Hannah and Lelaki. Because of his many publications and positions in the publication world, the assumption is made that Wijaya is considered by his culture to be a good storyteller. In this edition, the stories are translated into English, and these translations have undergone extensive editing. These translations were found to be helpful in determining the general direction of the story, but to accurately study the forms, a nmrpheme- by-morpheme gloss was completed, followed by ea second free translation attempting to 1x3 as faithful tx> the author as possible. The gloss was the primary tool used 1J1 this investigation since Indonesian sentences are not equivalent to English in their structure. (Some Indonesian sentences would be considered fragments in English.) In retrospect, after having found the basis for discrimination between the two forms, the study could have been accomplished at any of the three levels—-— Indonesian, gloss or translation because of the contextual nature of the difference. 1.4 Method Since its inception linguistics has depended on pattern matching and intuition as a basis to exert its 13 claims. Recently, however, the importance of mathematical validatbmn has risen 1J1 importance, and rightly so. Previously linguistics subverted the scientific method by simply proposing a hypothesis, finding a pattern which seemed to confirm the findings and. describing it 1J1 standard. terminology (although there is some question about linguists' efficacy on this last point). Sound scientific procedure requires a standard terminology, a relationship to study, and a control group. Linguistics, until recently, has operated without a control group. First in socio-linguistics and now in ever- widening areas (If linguistics linguists aux; beginning to re—incorporate the control group back into their studies. Myhill (1992) has finally shown how discourse linguists can use the notion of control group to improve the quality of their study. His book is not revolutionary for suggesting parameters-—even objective parameters (Hopper and Thompson 1980)-- but in suggesting that the non-occurrences be counted. This added the element of a control group. This study relies heavily on the methodology proposed by Myhill-— applying‘ it tx> various jparameters and. attempting to explain those instances which fall (nu: of the statistical "norm". 14 1.5 Outline This thesis has five chapters. The first has briefly outlined the need for research on the Indonesian aspectual system, in particular the telah/sudah markers. Additionally, a basic definition of aspect has been addressed here. The following chapter will give 51 brief overview of aspect in Indonesian and describe the research in that area that has already been conducted. In the third chaptery Hopper and Thompson's (1980) transitivity scale will be explained in detail, with special attention given to the application of Hopper and Thompson's scale to the data. The fourth chapter will argue that discourse conditions govern the choice of 'sudah' or 'telah' in a particular context. Specifically it examines the affect of durative/ non-durative inherent aspect on the choice of markers. In general the difference between the two forms may' be summarized as an instance of perfect aspect being sensitive to the contextual aspect distinction (n5 durative/ non-durative. Contextually significant contributing elements will 1x3 shown to tme the story time lapse, the action referred to, and the narrator's point of view. Finally the fifth chapter will concentrate on the theoretical implications of this work, including comments on the aspect/aktionsart distinction and 15 Hopper and Thompson's transitivity scale. Suggestions for further study are also given. l6 1. If irrealis is included, there are at least twelve words with aspectual meaning. Some are in diads and some in triads. (See section 2.2.4 for a list.) 2. Sentence references are given according' to the story the sentence was found in and the sequential number of the sentence in the story. For example, the first sentence in the story Pelacur would be marked P001. Those sentences from Bisma are marked B, and those from Menoleh are marked M. Occasionally, in chapter three sentences are used from a story collected in Dallas TX. These will be marked (D004, D011) and are included in Appendix D. See the following note. 3. All of the data used for statistical analysis is from the work by Wijaya; however, for clarity in chapter 3, sentences were occasionally used which were collected in Dallas by the author. These were elicited orally from two Indonesian speakers-- one from Jakarta, Java and one from 'Ujung Pandang, Sulawesi and are provided in Appendix D. ASPECT AND INDONESIAN 2.1 Introduction In order to understand the similarities and differences between the two items of the aspectual pair 'sudah' and 'telah', the reader must have a basic knowledge of the development of Indonesian and an understanding of aspect, the perfect in particular. Indonesian is an Austronesian language, now the national language of Indonesia» It has always been a lingua franca for the area and is spoken as a first language by a large urban population1 in that country. The population of speakers increases as time number of intergroup marriages increases time need for ea common spoken language in the home. In 1942, the Dutch relinquished control of Indonesia to the Japanese. In that tumultuous time, the schools and free it from definitions and categories imposed by western linguistics. This thesis attempts to add to the current body of descriptive work by considering 20 Indonesian on the basis of actual data and by not forcing them into western structures. 2.2 Characteristics of Perfect Aspect Although the Indonesian pair 'sudah' and 'telah' will be examined in light of the perfect aspect, their differences cannot be attributed solely to aspect. Unfortunately, linguists do not agree on many matters pertaining to the perfect. First of all, many do not agree that the perfect is an aspect at all (McCoard, Binnick, Comrie). However, since it exists in some world between tense and aspect, linguists have tended to include it in the aspect category for the sake of simplicity. This section will examine three prominent linguists' views of the perfect. 2.2.1 Comrie Comrie includes the perfect in his book on aspect, although he admits that it is not an aspect in the usual sense. Traditionally, in works that make a distinction between tense and aspect, the perfect has usually, but not always, been considered an aspect, although :U: is doubtful whether the definition of aspect given above can be interpreted to include the perfect as an aspect. However, the perfect is equally not just a tense, since it differs in meaning from the various tense forms (Comrie 1976z6). 21 In a later chapter on the perfect, Comrie gives a working definition of the perfect as, "the continuing relevance of a previous situation" (56) and goes on to describe four types of perfect. 2.2.1.1 Perfect of result Comrie describes the perfect of result as that condition in which "a present state is referred to as being a result of some past situation" (56). In English, for example 'David has been insulted' indicates that a current state (anger) is the result of a past situation (that of being insulted). The perfect 'of result is documented by many linguists as one of the sub-types of the perfect. 2.2.1.2 Perfect of a persistent situation (called continuous in some literature) The perfect of a persistent situation is used to describe situations that have begun at a particular time and then persist into the future. Examples of this are common in English-— "I've been studying here for three years." The act of studying at a particular place began in the past (three years ago) persists to the present time and into the future. This perfect is also quite common in Indonesian. 22 2.2.1.3 Perfect of recent past The perfect of recent past has many names in the literature (including 'lxn: news' (Anderson. 1982:228) and 'perfect of an unprepared mind' (ibid:233)). It is used to relate new information, e.g., "Martha has just entered college." This use of perfect is quite common in English and used to report a situation or event which has just occurred. 2.2.1.4 Experiential Perfect2 The experiential perfect indicates that a situation has happened at least once in the past. An example of this would be, 'Lauri has been to Disney World'. Here, it is true that the speaker has at least one time in the past, been to Disney World. Comrie feels very comfortable classifying the prefect. Indonesianists, however, ‘take 51 different approach-- they are more likely to describe the words attached to the perfect (meaning) than to label them as such. 2.2.2 R. Ross Macdonald on 'sudah' and 'telah' Macdonald (1967) mentions thirteen aspectual markers in Indonesian. He does not, however, use the word 'aspect' to describe them. They are classified as 'premodifiers.‘ He often mentions 'sudah' and 'telah' in time same sentence, equating them. (See quotation 23 page 2). 'Sudah' is called the 'past completive' while 'telah' is only defined by translation-~"already." 2.2.3 Soebardi on 'sudah' and 'telah' Soebardi's grammar, written in 1973, ch) not deal in any general way with the category of aspect. It is possible, however, tx> construct a: definition through careful searching through these books, meant to teach Indonesian to speakers of English. He also mentions thirteen of the aspectual markers found in Indonesian. (All but one of which duplicates of Macdonald's list.) (See table 1 for a complete accounting.) His translation of 'sudah' is "to have+ past participle", and his translation of 'telah' is exactally the same. 2.2.4 Dahl Dahl's work (1985) is a major accomplishment in that it lists the tenses and aspects found in many languages and language families. In order to accomplish this, Dahl constructed a survey which gave sentences in context and asked for translations. His survey deals with the major meanings (or 'uses') of the particular aspect or tense involved. As a result, the survey' misses many of the finer distinctions. In Indonesian, Dahl actually misses a great deal. Dahl examines four Austronesian languages: Cebuano, Indonesian, Javanese, and Sundanese. In 24 Indonesian, he finds four Tense, Mood, Aspect (TMA) markers: 'akan', 'pernah', 'sudah', and 'sedang'(l60). In Javanese, rue also finds 'lagi' and 1J1 Sundanese, 'tadi'. A closer look at Dahl's statistics show that even though he lists 'sudah' as Perfect (PFCT), his statistics show that 'sudah' is never used in what he considers 'core meanings'. This is curious and causes one to question what the core meaning of 'sudah' is, if it is not PFCT. Furthermore, Dahl lists 181 (If 197 sentences as unmarked for aspect in Indonesian. This raises further questions about. the depth. of time study 'under consideration. A further indication of problems of Dahl's work on Indonesian, is the fact that those who have studied Indonesian have listed many other words as related to aspectual meanings. Table 1 compares Dahl's study' with the grammars of two Indonesianists (Macdonald and Soebardi). Marker Macdonald Soebardi Dahl akan 'involving' Aux: will FUT denotes futurity baru NM 'just+PT of V NM belum NEG for sudah 'not yet' NM bukan NEG for N and NOM 'it isn't' NM lagi 'again, more' 'to be+V+ing' JavaROG iterative 25 masih continuation 'to be still+ NM V+ing' pernah takes place) once 'once,ever' EXPER sedang continuation 'to be+V+ing' PROC sudah past completive 'to have+ PFCT past part' tadi 'earlier today' 'just now, a Sun:< PAST little while ago' tahu 'know' NM JaszXPER telah 'already' 'to have+ NM past part' tengah continuation 'to be+V+ing' NM tidak NEG for 'no,not NM predicatives acting' Table 1. Comparison of Aspect Markers Although Macdonald and Soebardi do not specifically use aspectual terms, it is fairly clear that the words carry aspectual meanings (particularly those words which mean continuative: 'tengah' and 'masih'). Other words which should be added to Dahl's list are 'telah,' 'lagi' (listed kn/ Dahl. under' Javanese) and 'tadi' (listed by Dahl under Sundanese). Dahl's work, however, should not be discounted, since it is abroad survey of languages, and. perhaps quite successful. For my purpose, the finer semantic distinction between the two similar markers, it will not help at all. 26 2.2.5 Lloyd Anderson (1982) Since Dahl was unable to find the core meaning of 'sudah,‘ I undertook a brief investigation of the possible uses of perfect aspect. Anderson (1982) was useful in that it treated the perfect as a semantic area that has different uses by different languages. Anderson maps the semantic area much like Labov (1973) maps the range of the use of the word 'cup'.3 His map includes many uses that may be included in a particular language. In the center' of his map are the four 'current relevance' (C-R) meanings: C—R. of .Anterior (intransitive), C-R (m3 Anterior‘ (transitive), C-R .a past action since it "has been done." In sentence M027, the connection is direct. 'Sudah' is cflrectly related ii) the verb 'menoleh' "to look back". 2. Saya tidak merasa tidak merdeka hanya I NEG TR-feel NEG TR-free only I don't feel enslaved just karena sudah men-oleh! because already TR— glance! because I glance! Likewise, 'telah' (Echol and Shadily's 'past action') can be used for both actions and states. In the data studied here, these two uses were even found in the same sentence, 8177. 3. Jalanan telah ramai dan tukang bubur street already crowded and vendor soup The streets were already crowded and the ayam di kelokan jalan telah mem-buka chicken at corner street already TR- open chicken soup vendor on the street corner already 33 dagangan-nya. stall his opened his stall. The first 'telah' refers to the 'crowded street.’ 'Crowded' is not an action, but a description of the state of the street. In the coordinated clause, 'telah' refers to the action of the vendor having opened his soup stall for business. Here, in one sentence, 'telah' is used both ways. The situation (as is common in Indonesian) is just not simple at all. No easy solution has been found for 'sudah' and 'telah'. Yet, since both Dahl and Macdonald claim 'sudah' to be aspectual, and since Macdonald is willing to place 'telah' on a parallel footing with 'sudah', one can at least tentatively conclude that both 'sudah' and 'telah' are aspectual. Ebr’rmww it is sufficient to know that they are similar in aspect. Later, it will be shown that they can be differentiated. Finally the nature of this differentiation will be explained. Comrie, although not working specifically in Indonesian, provides definitions which can kme applied to the data in this study to see if they fit his definition of perfect at all. As the next section shows, there is little difficulty placing the uses of 'sudah' and 'telah' into Comrie's scheme. 34 2.4.1 Comrie's categories in Indonesian Comrie ‘used time Indo-European. languages fill his survey of aspects, but he always allowed latitude for languages to treat various aspects differently. Using Comrie's definitions of the perfect aspect, 'sudah' and 'telah' can both be shown to correspond to use of the perfect of result. There are 47 occurrences of 'sudah' in the texts; of these 42 (89.4%) are perfect of result. A good example of this is sentence M027: 4. Saya tidak merasa tidak merdeka hanya I NEG TR-feel NEG TR-free only I don't feel enslaved just karena sudah men-oleh! because already TR- glance! because I glance! In this example, the speaker is claiming that her ten year habit of glancing behind herself has not resulted in her presently feeling enslaved-- contrary to the suggestion of her husband. The other uses of 'sudah' include both the 'continuous' and 'recent past' (6.4% and 4.2% respectively). 5. Jadi lihatlah, sudah 10 tahun saya finish see already ten years I So you see, for ten years I 35 di — ajarkan oleh tubuh saya untuk ber- hati-hati. PASS train by body my for to be careful. have been trained by my body to be careful.(M041) 6. Tapi pada suatu hari, waktu wanita itu but at one day time woman that But one day, when that woman men-oleh ke belakang sebelum tidur --ia TR- glance to back not yet sleep --she glanced back before sleeping-- she sudah berbaring, lalu tiba-tiba ter- ingat already lay down then suddenly x-ed remember had already lay down, then suddenly remembered untuk men-oleh -- ia meng-angkat badan- nya to TR- glance—-she TR- lifted up body her to glance(back) —- she sat up lalu men-oleh ke atas kasur. (M029) then TR -glanced at on top bed then looked on top of her bed. The uses of 'telah' pattern similarly; 93% (27 of 29 instances) of the occurrences are the perfect of result, typified by sentence P054: 7. Aku telah menjual kelihaianku men-y(s)usun kata. I already TR-sold cunning -my TR -compile word I had already sold my ability to write. The rest are continuous or recent past (3.45% each). 8. Lembaran yang seyogya—nya men-jadi lembar page R.CL obvious-it TR been page This page which obviously would have been the pertama perubahan pidato itu, tapi tak first revised speech that, but not first page of the revised speech, but 36 mampu kulanjutkan karena merasa telah capable continue because TR—feel already I was unable to continue because I had felt di-perbudak.(P165) PASS-slave enslaved. 9. Kenapa sekarang aku baru di- tengtang, why now I begun PASS-opposed, Why now am I beginning to be opposed, setelah aku tidak bisa lagi men-jawab.(P022) after I NEG able PROC TR- respond. after I am not capable of responding. All these examples show that 'sudah' and 'telah' both carry a range of perfect meanings in Indonesian. 2.5 Summary Indonesian is ea language rich 1J1 aspect markers and aspectual distinctions. It has a number of aspect pairs which, at first glance, seem to carry the same meaning. The perfect aspect has many different uses, four of which Comrie has carefully delineated: perfect of result, perfect of a persistant situation, perfect of recent past and experiential perfect. Dahl has done a comprehensive study of tense and aspect across many languages, including Indonesian, but he has not achieved the level of specificity I seek here. Anderson mapped many of the common uses across languages and has helped describe these different uses. Indonesian has many aspects with subtle differences 37 between them; in. particular, the words 'sudah' and 'telah' carry perfect aspectual meaning. The difference between the two is subtle but important. 38 1. The Summer Institute (n5 Linguistics reported in Dec. 1991 that: Multilingualism is common throughout the Indonesian archipelago. Most Indonesians are conversant III the regional dialects (If Malay, as a lingua_franoa, and in at least one local language as well. Bahasa Indonesia (a form of Malay developed as the Indonesian national language) is used by all formally-educated people. It is the lingua_iranaa of all but the most remote ethnic subgroups . However, very few people outside the urban areas are fluent at speaking or reading Bahasa Indonesia . In larger language groups, instruction in the mother tongue is allowed during the first three years of primary school. .After that, Indonesian is used exclusively. In the smaller language groups, Bahasa Indonesia is used from the first grade (SIL 199lz5). 2. This section is added in order to comprehensively survey Comrie's work. In Indonesian, this function is covered by the token 'pernah' (Dahl 1985:160). 3. Labov (1973) varied the dimensions of various bowl/cup shaped objects and asked speakers which dimensions would be considered cups. In much the same way, Anderson has developed a map by comparing many languages use of the perfect and by this has been able to map the uses employed by a certain languages' perfect aspect. 4. See Appendix F 39 5. If irrealis is included as aspect, there are at least a dozen words with aspectual meaning. IX reason to include irrealis markers is that they also reflect aspectual meanings. For example, 'belum' means 'not yet', 'bukan' means 'although this outcome can usually be expected, it is not true now.‘ Tidak means 'no it will not happen (it will never be the case)’ Personal conversation with Dr. Mary Bresnahan Dec. 14, 1994. HOPPER AND THOMPSON TRANSITIVITY SCALE 3.1 Introduction This study hopes to determine the discourse variables which determine the use of one form of the perfect over another~~ specifically, to determine why some sentences require the use of 'sudah' and others the use of 'telah'. Unfortunately, neither Indonesian speakers nor Indonesian grammarians are able to indicate what in the discourse conditions the choice of one over another. This chapter shows that there is a significant difference between the transitivity of clauses using sudah and those using telah based on Transitivity scores as determined according to Hopper and Thompson's scale (1980). On this foundation, chapter four Mull examine discourse contexts and. present. a Zhypothesis about the relevance of these contexts to form selection. Thompson and Hopper (1980) used a scale to evaluate the transitivity of sentences. They claim that, Transitivity is a crucial relationship in language, having a number of universally predictable consequences in grammar, and that the defining properties of Transitivity are discourse- determined (251). 40 41 This scale has the ability to make very fine distinctions between sentences. This study uses the Hopper and Thompson scale to determine if there is a significant difference in the transitivity of sentences which use 'sudah' as opposed to those which use 'telah', providing an empirical basis for the claim that the two are not identical or interchangeable. 3.2 Hopper and Thompson Transitivity Scale In order to interpret the data in a quantitative manner, this study subjected each clause containing the tokens involved to the transitivity scale developed in Hopper and Thompson, 1980. They identified and isolated. ten components of ‘transitivity' which allow clauses ti) be ranked CH1 a gradient from taint to low transitivity. Their ten parameters are: A. Number of Participants Kinesis Telicity 1 Punctuality Volitionality Affirmation Mode :13 C) '1'] [31 U 0 {D Agency I. Affectedness of Object J. Individuation of Object (252) 42 This study will describe these parameters lJlta manner consistent with Hopper and Thompson's, but where finer distinctions need to be made, these are noted. 3.2.1 Participants A clause is said to be more highly transitive if it contains two or more participants (usually an actor and an object). The participants do not need to be animate. The object may be inanimate, as can be seen by the following sentence: "Fritz knocked his coffee cup off of the table." The actor does not necessarily need. to Ibe animate either, as can be seen in the following sentence: "The wind blew Christopher off of the ladder." In the first sentence, the object is an inanimate cup of coffee and :hi the second sentence, the agent, "the wind", is generally considered inanimate. Yet, both these sentences are more highly transitive timun for example, time sentence, "Stephen was awake," in which there is only one participant. Additionally, Hopper and Thompson would claim that these first two sentences are also more transitive than one sentence which has an action and an actor but no second participant, e.g., "Sadie ate." 3.2.2 Kinesis A clause containing an action is more highly transitive than one which does not, since 'states of 43 being' are not transferable from one participant to another. Actions are potentially transferable (at the very' least). The active sentence, "Lander' ate the candy" is more transitive than the sentence "Katy wanted the candy" because the action of the first sentence is more completely transferred to the candy than the mental state of wanting it in the second sentence. In fact, the second sentence does not transfer anything to the candy at all. The candy remains unchanged for all of Katy's wanting, whereas Lander's eating the candy changes it greatly. 3.2.3 Telicity (Aspect) Telic actions are the method by which "The discourse imposes ea perfective interpretation on foregrounded events" (Hopper and Thompson 1980:286). Hopper has twice before associated foregrounded events with perfectivity (Hopper 1977; 1979). Actions which are perfective in nature are more highly transitive than those which describe chronological simultaneity or overlapping. Perfective (telic) actions follow strict chronological sequencing, are viewed as a.vdmflte (and maintain identity of subject.) Non-perfective (atelic) actions present a situation which does not necessarily require completion for the performance of a subsequent action to occur and in which subjects frequently change.2 This distinction it; clarified iii the 44 following sentences. "Del ate dinner before he washed the dishes." Here the action of washing the dishes is subsequent to eating dinner. In contrast is the following sentence which describes a simultaneous action on the part of the participants. "Kim fed the hens while Mariah was watching for the mail." In the first sentence, the actions are chronological; each is viewed as a totality, and each maintains identity of subject. In the second sentence, the actions are not crucially ordered; they could happen one after another, simultaneously, or in 'reverse order'. And the first action mentioned does not have to be completed before the second may occur. The distinction here is aspectual; time first sentence describes a1 perfective sequence of actions while the second does not. 3.2.4 Punctuality An action which is punctual in nature is one that "does not last in time, one that takes place momentarily . . . (they) do not have any duration, not even a very short period" (Comrie 1976:42). These are more transitive than ones which are ongoing or lasting in nature. For example, "Jim saw a deer," is more transitive than "Mary watched television" simply because the action of seeing is thought of as instantaneous while the action of hatching implies a duration of time. 45 3.2.5 Volitionality Clauses in which actions are carried out intentionally by an agent are more highly transitive than those in which agents are accidental performers. In Indonesian, this distinction is made relatively simple since some verbs are morpho—syntactically encoded as being unintentional in nature. 1. Dia ter- perangkap (Adapted from D0043) he accidentally trapped He was accidentally trapped. 2. Dia meng- perankap—kan kijang itu he TRANS trapped purpose kijang the He trapped the kijang(on purpose).(D011, adapted) Therefore, the first sentence, where the object is accidentally trapped, is less transitive than the following sentence, where the agent is acting volitionally. (See section 3.3.1.1, Parameter E, for a more detailed discussion.) 3.2.6 Affirmation References to real world events in clauses are more transitive than those stated negatively. When the sentence "Josh ate squid" is compared to the sentence "Elizabeth did not eat squid," the higher transitivity of the former is no surprise. 46 3.2.7 Mode Likewise, clauses which refer to real world events are more transitive than those in which events are said to occur in some alternative 'non-real' world. The contrast here is simple. Compare "Dennis heard the opera and he liked it" to "If Tucker had heard the opera, lme would have liked it." The first actually asserts that. an action. took: place 'while the second merely presents possibility. 3.2.8 Agency Agency is the first component for which Hopper and Thompson suggest that a sub-scale might have relevance. They note that high transitivity corresponds to sentences in which agents are more animate. The agency hierarchy was proposed by Michael Silverstein in 1976. The hierarchy links the mention of a participant to the pmobability that that participant will be an agent. For example, when a participant is referred to in the first person, the participant is very likely to serve as agent. If the participant is referred to in the second person, it is a little less likely to be the agent. Inanimate objects are the least likely agents. The hierarchy is structured as follows: 47 lst P> 2nd P> 3rd P> Pr Name> Human> Animate> Inanimate (Hopper and Thompson 1980:273). So, "Gary had a cookie" is higher in transitivity than "The box had a cookie" because "Gary" is a proper name and "the box" is inanimate. Even though Hopper and Thompson suggest that this hierarchy has validity, they do not suggest any particular point at which a participant should be regarded as lrwv in agency.4 With the hierarchy as a guideline, the investigator is left to decide what constitutes EH1 agent high.:h1 potency. Occasionally, context may demand that one assign an inanimate object high agency, although these cases are difficult to imagine. More often than not this scale simply becomes a distinction between animate and inanimate. 3.2.9 Affectedness of Object In clauses where the patient is completely affected, the action is said to be more highly transferred; therefore, the clause is more transitive. This is often a difficult judgment to make. An example of this in Indonesian is: 3. Ia lupa me-laku-kan-nya she forgot TR-do -DAT-it She forgot to do it. (adapted from M005) 48 In this sentence, is the object -nya completely affected by being forgotten or, conversely, not at all affected? The judgment must be made according to the wider context. of the jpassage. 1&1 this case, -nya refers to an 'errand,‘ so while the errand was unaffected, the person was affected by forgetting the errand. This is not simply a confusion over the negative thrust: of the 'Verb 'forget,‘ as one could easily say, "she remembered to do it" and still raise the question of what is being affected— the agent or the patient (in this case the errand), and I believe that the conclusion must be the same, that, although the A is affected, the O is not. 3.2.10 Individuation of Object Hopper and Thompson suggest a concrete method of determining whether the O is individuated. They give six parameters to score and suggest that the last be given double weight. The parameters are characteristics of the participant involved, they include: 1)proper, 2)human (animate), 3)concrete, 4)singular, 5)count, and 6)referential. and. definite. Objects score a point for each of the first five and one point for being either referential or definite and another for being both. On a practical level, it makes no difference in scoring whether the last category is considered as a unit or broken into two separate 49 categories. Fbr example, "Bryce lifted Alexandra" is more transitive than, "Nikki lifted sand" because .Alexandra. is proper, human, concrete, singular, referential and definite; while sand is only concrete. 3.3 Applying Hopper and Thompson to Indonesian Inherent in replication of another's method are the difficulties of interpretation and consistency. Some of these problems are simply borderline cases which may be obvious to those who conceived the method but not quite so clear when used by another particularly' in a language other than the one the method was based on. 3.3.1 Method The data were examined by using the Hopper and Thompson transitivity scale enmi granting time clauses one point for each transitive parameter they exhibited. A transitivity score was determined for each clause. The clauses were then separated into those using 'sudah' and those using 'telahi. .A mathematical mean was taken for each group, and a t-test was carried out on the two means to assure that the differences found could not be attributed to sampling errors. 50 3.3.1.1 Specific difficulties with application of Hopper and Thompson transitivity scale One of the first determinations which needed to be made was in the consideration of participants, Parameter A. In Indonesian, as in English, a mob may be the actor in a sentence. This is perhaps a small point, but, as in English, the mob (when acting in concord) was considered as one participant. The difficulty with telicity, Parameter C, came from the nature of the data used. In this study, the data came from Indonesian stories which where intended for entertainment. Hopper' and Thompson claim, "an action viewed from its endpoint, i.e. telic action, is more effectively transferred to a patient than one not provided with such an endpoint" (252). And, The ‘transitivity' feature ch.Aspect refers to the telicity' of the predicate's action” A predicate which specifies ani endpoint or conceptual boundary is said to be telic, while one which does not is atelic (286 emphasis mine). However, in real world data, not all predicates have verbs which are actions. A decision needs to be made regarding' stative enmi intransitive ‘verbs. Each. of these verbs, action or not, needs to be analyzed for some endpoint. Some stative verbs expressed an activated state and so were counted as telic. The 51 mental state of 'know,‘ for example, was considered an endpoint and counted as such. Parameter D. Punctuality is based on the fact that those actions which do not have a transitional phase between inception and completion have a :more marked effect on their patients than actions which are more inherently on-going. For example, compare these two sentences: "Matt hit the car" and "Kristi drove the car". The action of hitting in the first clause has a more marked effect on the car than the action of driving in the second clause. If the predicate is semantically a state, then it seems that punctuality is ruled out on the basis that states are hardly ever momentary and those which are ephemeral, 'blink-of—the-eye states' rarely bear mentioning in narrative. Givon notes that, Most commonly, predicates that are semantically states are already durative by definition and thus do not take the durative aspect per se (1984:275). Experiential verbs in particular have to be judged according to context; sometimes a feeling is a punctual, brief situation. In other contexts, a feeling may endure, indicating in fact, a long lasting change. Parameter E. A volitional agent affects a patient more than (Mme which is 1mm: acting purposefully. In 52 Indonesian, it is fairly simple to determine volitional versus accidental action because verbs are sometimes coded to mark intentionally. In this example, one character accidentally trapped and then, after being released, purposefully traps his rescuer. The contrastive sentences are: 4. Dan dia datang ke kijang itu, kijang itu sedang and he came to kijang that, kijang that PROG And he came to the kijang, the kijang was ter- perangkap (D004) accidentally trapped accidentally trapped. Notice the prefix 'ter-' attached ti) the verb 'trapped.’ This particular morpheme indicates and accidental action. Wouk noted that, "The common meaning of 'ter-' in all these uses is nonintentionality" (1979:86). Compare example 4 to the next sentence, where the trapping was on purpose. 5. Dia meng-angkat kayu itu dan memerangkap sang he TR- lift wood the and TR- trap the He (the alligator) lifted the wood and trapped the kijang di antara dua belah pohon itu akhirnya kijang at between two half tree the last kijang between the two halves of the tree juga (D011) anyway anyway. In this sentence, the prefix meng— (in the form mem-) was added to the verb to show that the action was not 53 an. accident. Unfortunately, rmfl: all sentences ‘were marked. so obviously, and judgments had to be made concerning the volitionality of the agent, if an agent was present. (See section 3.2.5 for a summary.) Some of the more difficult questions which arose concerned passive sentences involving the body 'training' the person. The passive marker 'di—' is more volitional than the 'ter-' choice. This morpheme choice highlights the volitionality of time action of the agent; however, pragmatically, it is odd for a body to train its person, particularly in a volitional manner; more often the case is that the person volitionally trains or 'commands' the body. 6. Jadi lihatlah, sudah 10 (sic) tahun saya finish see already ten years I So you see, for ten years I have been diajarkan oleh tubuh saya untuk berhati-hati PASS-train by body my for to be careful trained by my body to be careful.(M041) For this sentence, the parameter was thought to be more transitive, since 'training' it; usually 51 volitional action regardless of the actor. Another question arose when dealing with mental processes; in particular, the extent of control that a person has over mental processes are at issue. In the sentence which follows the previous example in the text, the claim is made that the body 'knew' something. 54 Ignoring for the moment the semantic problem of the body 'knowing' something, the problem of determing volitionality remains. Should 'knowing' be considered volitional? 7....tubuh saya sudah mengetahui akan kedatangannya, body my already TR-know-PROC will come- it, My body already knew it would come, lalu ia mempersiapkan diri.(M042 embedded clause) then it TR-prepare self then it (my body) prepared itself. In this sentence, and perhaps for all sentences, the mental processes are not volitional, so for this parameter, the sentences which showed processes like 'remembering', 'forgetting,‘ and the like were scored zero because at least these processes are thought to be done by an actor without intentionality. Parameter F is the jparameter dealing with the statement of the sentence in the positive or negative. This parameter was unproblematic in Indonesian. Mode, parameter (L is important ti) transitivity because an action which: did not occur or which is presented as occurring in 21 nonrreal world, is obviously less effective than one whose occurrence is actually asserted as corresponding directly with ea real event (Hopper and Thompson 1980:252). Here, future worlds were counted zero. Otherwise there was little confusion over proper scoring. 55 Parameter Ii. Agency. Participants which can effectively transfer an action are more transitive than those which cannot. If the clause has no agent recoverable from the wider context, then the clause must be judged as zero in this parameter. Some active clauses contained rm) agent III that particular clause and yet had an agent in the surrounding context. Sentence 8069 is a good example: 8. Se- sudah itu membuka busana- nya (B069) DEF already this TR-open costume-his After this (he) opened his costume. The agent 'dia' he is missing from the sentence completely (to be supplied by the context), but the verb is marked for transitivity. Parameters I and J. Affectedness of O and Individuation of O. If a sentence has no object, (i.e. is stative or intransitive) then I and J must be zero since there is rm) object ti) be affected. or individuated. (hi a wider frame, predicates of embedded clauses sometimes contained the token being examined. This created a question about the scoring of embedded and independent relative clauses. These clauses were scored separately of the main clause when the token itself occurred in the embedded clause or independent relative clause. 56 Despite a few difficulties in scoring, based primarily (H1 sentences viflrii are imperative or interrogative, the application of Hopper and Thompson's transitivity scale to Indonesian narrative was relatively straightforward. Their system appears to have been able to avoid the pitfall of being too language family specific. (That is ti) say, they have created something which applies to all languages with almost equal facility.) Hopper and Thompson's parameters are quite clear cut and fairly useful even in a language quite different from Indo—European. 3.3.2 Results The results of applying the Hopper and Thompson transitivity scale to the Indonesian data were 1) that the word 'sudah' including the form se-sudah (the less marked choice) had a transitivity mean of 5.277 and 2) 'Telah', the more marked choice, had a transitivity mean of 7.483. These means were subjected to a t-test to determine: if they' are statistically significant. Using the hypothesis that they are the same, the two means were analyzed, and the mathematical results suggest that this hypothesis should be rejected on the basis that the means are significantly different at a .001 level of significance. renal a T-score of 4.535 and standard deviations of 2.06 for 'sudah' and 2.08 57 for 'telah'. Therefore, the means of the two choices are statistically different. 3.4 Anomalous Cases There were 29 total incidences of 'telah' and 47 incidences of 'sudah' in the data examined. These totals include incidences of kai1 the root. and. the forms prefixed with 'se-' (see appendix E). Although 27 of 29 clauses using 'telah' fell in the 4-10 range for transitivity and 44 of 47 of the clauses containing 'sudah' fell in the 2-8 transitivity range, there were a few sentences which fell beyond the upper and lower levels which will be discussed. The anomalous cases may" be broken into two types: those anomalous for methodological considerations and those which are anomalous because CH? contextual considerations. The methodological anomalies will be discussed here, while the contextual cases will be used to exemplify the contextual nature of the choice of 'sudah' or 'telah' in the next chapter. Returning then ti) actual transitivity scores, a four stood out as anomalous: P022, P145, B069, M005. One of those was the sentence P022 which has a transitivity score of Hi. This sentence is interesting because it includes the word 'setelah' which occurs in only seven of the 29 'telah' sentences. 58 The 'se-' prefix indicates that the event referred to is definite. (See Appendix E) 9. Kenapa sekarang aku baru ditengtang, why now I begun PASS-oppose Why now am I beginning to be opposed, setelah aku tidak bisa lagi menjawab Df—after I NEG able PROC TR-respond after I am not capable of responding. The second clause in this sentence is the one which was scored since it contains the word 'telah'. The decision to score the embedded clause may have resulted in this transitivity score. The second clause has only one participant, 1PS (I) 'aku,‘ so parameter A, Number of participants, is scored as zero. The clause does not contain a kinetic action, since 'respond' is not physical. This has the effect of making parameter B, kinesis, also zero. Furthermore, telicity, punctuality, volitionality, and totally affected object, are all zero. Mode was scored as one because the speakers inability to respond is presented in the narrative as a real situation, not some conditional (n: unreal situation. Affirmation is scored as zero because the subordinate clause contains the negative 'tidak.‘ Agency, parameter H, raised the philosophical question: can a dead person be a powerful agent? The text seems to suggest no. If he were able to respond, he (the 59 dead leader in the sentence) definitely would, so this was also scored zero. Since the transitivity level does not indicate the 'telah' choice, then something else in the context is dictating the choice. Indeed, the contextual situation is unusual in that a dead official appears in a dream to the main character expressing frustration because he cannot clear his name from the grave. Chapter four will provide the contextual explanation for this unusual use of 'telah'. The second anomalous methodological case is that of sentence P145 where 'sudah' is found in a relative clause attached ti) a transitive verb vnii1 the abnormally high score of nine. 10. Me-nanti bagaimana se-seorang yang kecewa TR-next how Df.man who disappointed I waited to see how the man who was disappointed dan merasa sudah mem-bayar akan menyemprot- and TR-feel already TR- paid will TR- attack and having already paid, would attack ku. me me. In this case, the main clause was scored as a matter of course. The transitivity score, however, probably should have been based on the relative clause in this instance. Had the relative clause been scored, the clause would have fallen within the parameters set by 6O presenting a score of six, which is much more in line with the typical 'sudah' score. A time related clause, B069, is anomalous in that it contains an unusually high transitivity score for the use of 'sudah' (i.e. nine). 11. Se-sudah itu mem-buka busana- nya. Df—already this TR- open costume his. After this he opened his costume. There are many possible explanations for this discrepancy. The most obvious is that the transitivity score is based, in part, on a previous sentence. The agent in this case is not present in the sentence and so was interpolated from a previous sentence. If the agent-related transitivity points were eliminated, the sentence would be much lower in transitivity. I do not believe that this is the proper perspective. The agent is intended in the sentence and would be supplied by context in a reading of the passage (see section 4.2, 'Formulaic uses of Sudah and Telah'). The last anomalous case to be discussed in this section is an exception in a different way. In this case, the token 'sudah' is found in a sentence with a very low transitivity score, (one). The clause, in many ways, confirms that 'sudah' is normally found in sentences of low transitivity. However, since the 61 score is abnormally low, the clause necessitates some mention in this section. 13. Kalau ia lupa me-laku- kan- nya maka ia If she forget TR-conduct BENE it then she If she forgot to do it, then she se—akan—akan sudah kekurangan garam esok Df— will already less salt tomorrow would (feel) [dehydrated] tomorrow hari-nya. (M005) day. day. This sentence is a conditional, so the normally automatic point granted for affirmation could not be given because the action occurred in a non-real world. There is a question of punctuality here, but it seems that this feeling of salt depletion should be characterized.ams durative rather than punctual. The sentence is otherwise unremarkable, except to note that it is absent of transitivity features since it has neither agent nor object. 3.5 Summary The Hopper and Thompson scale gives ten parameters which, when applied to sentences, measure the transitivity of the sentence on a gradient. In Indonesian, the parameters were usually easily applicable ti) the data. In particular, mode, punctuality, and affirmation were uncomplicated and simple to apply. Perhaps this is because they are 62 binary oppositions which were present in Indonesian as in other languages. Those parameters which. were difficult to apply were telicity, volitionality, and agency. The difficulty here is that these parameters represent differences between Indonesian and English semantics or pragmatics. Volitionality and agency were particularly difficult since Indonesians tend to allow agents which are lower on the agency hierarchy to serve as subjects. They also allow volitionality in places where English speakers find it pragmatically odd. The telicity' parameter' was difficult for other reasons. In Hopper and Thompson, this parameter is closely related to perfectivity, or completion of action. In traditional Indonesian discourse, however, one of the three basic tenants of the style is to describe the actions in a fragmented dreamlike (or nightmare-like) :manner. This style CH? presentation causes a great deal of difficulty in determining the perfectivity of an action, even.cn1 a discourse level. Ike in all instances of using another's system, there were difficulties, particularly where sentence structures differ greatly. Overall, though, the transitivity scale was a useful and valid tool for analyzing Indonesian discourse. In this study, it is necessary to prove in some empirical manner that the two words 'sudah' and 'telah' 63 are different. The Hopper and Thompson transitivity scale is very important, since it gives an external measure of what many linguists call intuition. By using the transitivity scale as a :measure, and by performing statistical analysis on the results of measuring each sentence with either 'sudah' or 'telah,' empirical evidence can be seen fbr claiming that the two words are different at some level. The next chapter vflJJ. discuss specifically lmnv the tim) words differ in context. 64 Notes 1. Hopper and Thompson called this parameter Aspect (1980:252) and then defined it as telicity later on that same page. In order to avoid confusion, this paper will only call this parameter Telicity and thus hope to avoid confusing the reader with two different uses of the word 'aspect'. Aspect in this paper will refer to the broad grammatical category. .2. The information from this section was taken from Hopper 1977. It is possible that this connection was not intended by Hopper. But I believe he had this previous paper in mind when he collaborated with Thompson. 3. Dallas data. These sentences are adapted from those elicited in Dallas. (See Chapter 1, Footnote 2). 4. Hopper and Thompson imply through their analysis of the following example that the distinction be made at least as high as proper name: *3 a. Jerry likes beer b. Jerry knocked Sam down Hopper and Thompson claim that 3b has a higher transitivity than 3a since it contains the following high transitivity markers: kinesis, telicity, punctuality, affectedness of (L and individuation CL 65 They clearly omit, however, the characteristic "A high in potency." To account for the omission of this characteristic, there are two perspectives which need to be considered. The first is that Hopper and Thompson are only marking those parameters which make the sentences differ from each other and since both sentences have a third person agent this parameter is omitted because it is not a distinction. A second possibility is that Hopper and Thompson do not consider third person agents to be high in potency. DISCOURSE CONDITIONS GOVERN CHOICE OF SUDAH/TELAH 4.1 Introduction This chapter will show that the differences between 'sudah' and 'telah' are conditioned. by the discourse. Chapter" two Ihas shown. that 'sudah' and 'telah' are both markers of perfect aspect, indicating there is no great semantic gap between the two. Through the use of the Hopper and Thompson transitivity scale, however, it 1mm; been shown that 'sudah' and 'telah' are used in sentences with significantly different. levels (If transitivity. These IA= Inherent Aspect * 52 different verbs since two are used with both sudah and telah Table 5 Verbal agreement with IA theory The inherent aspect explanation is, (H) the surface, satisfying in that it explains nearly two thirds (64.81%) of the verbs used. with 'sudah' or 75 'telah' and yields nearly statistically significant differences between the two. There are, however, some obvious problems with this solution as a comprehensive explanation. Primarily, it is an incomplete argument since it leaves :more than a) third. (35.19%) of the verbal pairings unexplained. Secondarily, if inherent aspect of the verb were all that was involved, one would expect no duplications of the verb. Yet there are two verbs in this study, and certainly more than these in the larger realm. of Indonesian discourse, which need explanation. Finally, there are 11 incidences (14.47%) of 'sudah' and 'telah' which have no reference verb at all. If one strictly adhered to the inherent aspect theory, they would be forced to call these uses essentially meaningless. Another explanation needs to be provided. Inherent aspect of the context, whether provided through the verb or the discourse, provides the motivation to choose the "unexpected" choice despite other factors. An example where the inherent aspect of the verb overrides other considerations can be found in P022 where 'telah' is used to indicate a permanent change despite an abnormally low transitivity score of '1.‘ 8. Kenapa sekarang aku baru di- tengtang why now(at this point) I begun PASS- opposed Why now am I beginning to be opposed, after 76 setelah aku tidak bisa lagi men-jawab. Df-after I NEG able PROC TR- respond that I am incapable of responding. This situation is a good example of 'telah' being used to indicate a permanent change of status (where 'sudah' may indicate a temporary change). The key is that the official has changed from being alive to dead, and that state is permanent. In this instance, the discourse condition of a permanent change overrides the tendency to assign 'telah' to sentences which usually contain higher transitivity scores. 4.3.2 Inherent aspect as an incomplete explanation If the inherent aspect were a complete argument, then the inherent aspect of the verb should. match exactly with the choice of 'sudah' or 'telah'. In too many instances this is not the case. The inherent aspect of the verbs is key to understanding the choice of 'sudah' or 'telah' in many, but not all, situations. This section contains a table of the verbs and their categorization as EMrative or Non-Durative (including indeterminable). Occasionally the inherent aspect is «clear. 13) many' situations, though, the meaning seems to lack inherent aspect-- leaving the determination of durativity (and other inherent aspects) to the context. This table 77 represents a1 "closest approximation" ti) the inherent aspect of the verb devoid of context. Sudah with Indonesian Verbs Aspect Verb Gloss I.A. sudah ajarkan teach Dur bangun wake up Punc sudah baring lie down Punc bayar pay Punc sudah bilang each time Iter burgulir fall down Punc sudah enal know Dur hapus eliminate Punc sudah hilang disappear Punc ikut join Punc sudah jadi happen Indet kembali come back Punc sudah kerja work Dur laku done Punc sudah langsung direct Indet lupa forget Indet sudah oleh glance Punc rasa seems to be Dur sudah rupa has been Dur sampai to the point Indet sudah selesai finish 3. Punc siap prepare Dur sudah suap mouthful Indet susun compose Dur sudah tahu know Unmkd tempuh attack Punc sudah terima accept Punc tinggal live, stay Dur sudah tipu trick, deceit Punct tulis write Dur sudah tunggu waiting Dur 'sudah' Table 6 Sudah verb aspects section there are eleven verbs which are directly contradictory to the inherent aspect theory and must be described as being conditioned by 78 something else. I will argue that the discourse conditions these occurrences. The most obvious of these situations are those which clearly cannot be both durative and non-durative. 'Ikut' (B165) for example, means "join." The inherent aspect of 'ikut' cannot be both durative (to suit the 'telah' marker) and non—durative (to snfli: the 'sudah' marker. See section 4.4.1 for further explanation. Telah with Indonesian Verbs Aspect Verb Gloss I. A. telah amat exceedingly Dur angis crying Dur telah atur roll in time Dur bangkotan aging Dur budak slave Dur telah buka open Punc bunuh kill Punc telah dapat able Indet ikut join Punc telah jadi happen Indet jalan perform Dur telah jawab reply Indet jual sell, sale Punc telah lampau past Dur nodai stained Dur telah paska postgrad work Dur pikul carry Dur telah ramai crowded, busy Dur satu united Dur telah serah surrendered Dur tiba arrived Punc telah tukur change dress Dur ucap express Dur Table 7 Telah verb aspects 79 On the surface, there are eight exceptions in the 'telah' data which are either punctual or indeterminable. As will be seen, discourse context mitigates some of these in that, although they may happen 1J1 an instant, the results of time action are permanent so they carry some durative force. Three good examples from the above list are 'bunuh,' "to kill"; it happens in an instant, but has initiates a lasting state of affairs. .Another is 'jual,‘ "to sell." 9. Aku telah menjual kelihaianku men-y(s)usun kata. I already TR-sold cunning -my TR -compile word I had already sold my ability to write. (P054) Here the author sells his ability to write and his respect. The lasting consequences are that lme has no self-respect anymore. Finally, there is 'tiba,‘ which means arrived (B178). Again, actual arrival occurs in only an instant, but the discourse may mitigate again since the length of the stay can last a long time (even an eternity). The inherent aspect explanation would allow no such exceptions and would allow no use of context to explain the choice of one marker over another. While many of the verbs naturally agree with a particular aspect marker, not all do. Many exceptions to the rule 80 that 'telah' occurs with durative and 'sudah' with non- durative and unmarked verbs exist. 4.3.3 Extra-clausal reference When only the inherent aspect of the verb is considered, there is difficulty explaining those sentences which make reference to events or states not in the sentence but in the context. These sentences are often formulaic in grammar, but rather unpredictable in their discourse distribution. Many begin "Se- (sudah or telah) itu," 'After this,‘ where there is no verb in the clause to predict the proper choice. 10. Sesudah itu mem-buka busana- nya. (B069) Df—after this TR-open costume his After this (he) opened his costume. Occasionally, these tokens represent ea permanent change as determined by context rather than durative action. In these cases there is clearly a necessity to examine the inherent aspect of the whole situation. 4.3.4 Duplicates Occurrences of a verb which pairs with both 'sudah' and 'telah' require further explanation, 81 especially since the inherent aspect of a verb is thought ti) be lexical. Binnick (1991) describes the difference: But .Aristotelian aspect and..Aktionsarten are lexical, pertaining to classes of verbs or larger expressions. The difference between durative and perfective-durative, . . . is not a grammatical obligatory one; it is simply the case that a verb or collocation fits into the one or the other category (148). One verb in the data was morphologically marked to indicate duration. When so marked, this verb was used with 'telah'; when unmarked, it was used with 'sudah.' This duplication would be suspicious if the inherent aspect of the verb were the complete explanation, (as the internal aspect would link the verb to a particular aspect marker despite the context). When the inherent aspect does not match the choice of 'sudah' or 'telah', there is some contextual factor conditioning the choice. In the data sampled, there were three instances where 'sudah' and 'telah' were used. with the same predicates. They are the verb 'ikut' "join" which seems to be unequivocally punctual; the verb 'jadi' "become, or happen" which. might be either durative or non- durative depending on the context; and sentences where the markers were not followed by any verb at all. All three instances will be examined in the next section as 82 each introduces a different contextual conditioning factor. 4.4 Conditioning factors in the discourse While :most (Hf the choices ‘between 'sudah' and 'telah' can be made based on the inherent aspect of the accompanying verb, some verbs represent events which could conceivably be taken as either durative or non- durative. In some of these cases, the inherent aspect is in direct contrast to the expected choice of 'sudah/telah'. The differing interpretations of these verbs depends upon the story time lapse, the extra- sentential referred—to action, or the author's particular view of the event. 4.4.1 Story time Lapse The story time lapse creates a need to examine the inherent aspect of the discourse in order to understand why a particular marker was chosen. When the sentences are not formulaic, those containing 'telah' are sometimes marked for process with the -i suffix. When there is an overt time reference, 'sudah' marked sentences always involved the shorter duration. The nature of the larger context has some influence on the marking of the event. 83 The sentences containing the verb 'ikut' showed an interesting distinction. The 'sudah' sentence is unmarked: 11. Aku tidak tahu apakah mereka benar-benar tidur I NEG know if they really sleep I don't know if they were truly asleep or atau sudah ikut bersandiwara berpura-pura or already join drama pretending if they followed a script pretending not to see tidak melihat kekonydan - ku, untuk menjaga NEG TR-see foolishness my for TR-guard my foolishness to protect my feelings. perasaan-ku. (B165) feeling -my. In this sentence, the verb 'ikut' is not marked as a process. The action of the family joining a drama is easily characterized punctual. There is no indication that any time lapsed in the story (not even close to an eternity). For example, the referents involved could have simply uttered one line of the script and still be consistent with the interpretation of this sentence. In contrast, the sentence containing the 'telah' (durative) marker was also marked as a process (morpho- logically "-ing" in English) with the -i suffix indi- cating a longer time frame. 12. Saya sendiri telah meng—ikut -i sayembara I self-my already TR join PROC prize contest I already entered a prize contest 84 untuk mendapatkan tiga putri jelita for TR-obtain—BENE three graceful princess to win three beautiful princesses for untuk kedua adik saya. for two younger (brothers) my my two younger (brothers). (B023) In this instance, where the same verb is used, inherent aspect is Imfl: a sufficient explanation in itself the -i suffix helps mitigate the punctual inherent aspect. Morphological and contextual considerations need to be made as well. The difference between the two uses of 'ikut' rests in the process marking. In the first sentence, the verb is not marked for process and so takes 'sudah,‘ the non-durative choice. The phrase 'telah mengikuti' is roughly' equivalent. to saying "already competed." The second sentence is therefore marked as taking time and employs the durative choice 'telah.‘ As a further consideration, there is also the miti- gating context that the contest referred to occurred a million years ago. The discourse indicates that the contest was an event which the eldest brother fought to secure mates for his brothers. He mentions this event only because his masculinity is being questioned. This story time lapse of a mdllion years is a significant factor in the choice of marker. The inherent aspect of 'ikut' is clearly inadequate to fully explain the use of 'telah' in this 85 sentence since the verb is used with both markers. Perhaps the contextual fact that the competition was held. a million years ago sheds some light on the situation, because the story time lapse between the competition and the story-present is a million years. 4.4.2 Referred to Action Another contextual factor conditioning the use of 'sudah' and 'telah' fir; the extra-clausal referred-to action already briefly mentioned in section 4.3.3 which gives sentence B069 as an example of an extra-clausal reference. In. this sentence, the author' wishes to convey that the extra-clausal action was immediately followed by the character disrobing: 13. Sesudah itu mem-buka busana— nya. (B069) Df-after this TR—open costume his After this (he) opened his costume. In fact, the argument is contextually based. The actions happened so fast that the high transitivity token 'telah' would be obviously inappropriate to the context of the story. Even though this sentence is highly transitive, the actions happen in such close succession that they seem to be one and the same.2 Another example where extra-clausal referents condition the choice of markers can be found in 86 sentences B164 and B155. In these sentences, 'sudah' and 'telah' are both used with the verb 'jadi' which means "happen" . (In this case there is no morphological marker to distinguish between the two.) Here it is literally the context of 'what's happening' that determines the choice of 'sudah' or 'telah'. When 'jadi' is used in reference to the play or drama described in the context, the choice is 'sudah,‘ since no drama is permanent; even the characters are transitory--alive only for the time of the performance. 14. Keluarga kami, kelurga sendiri, masih tetap family we(excl) family my own still fixed My family, my own family, still snoring, mendengkur, tak tahu sandiwara apa yang sudah TR-snoring, NEG know drama what R.Cl already did not know the drama that had already ter-jadi (B164) Xed happen. happened. This sentence describes the narrator's family as unconscious of the change which had just occurred-- there is no indication of duration of the change (stated or unstated), so 'sudah' is the proper choice. However, when 'jadi' refers to turning into a mythological character for eternity, the choice is 'telah': 15. Pada saat yang sama dengan cara-cara yang sama at time R.Cl same with ways R.Cl same At that time, in the same manner, 87 kami yang telah men-jadi Bisma-Bisma kecil we R.CL already TR become Bismas small we who had become little Bismas meronta dan bangun terhuyung-huyung. (B155) TR-struggle and get up staggering strove to get up staggering. The choice of 'sudah' or 'telah' in these sentences is clearly context dependent, since the referent event is not found in the same clause as the token. Yet this is ultimately still a matter of whether the referent event is durative in nature or either indeterminable or non-durative. 4.4.3 Character point of view The third and most interesting discourse conditioning factor is point of view of the character. As :mentioned :hi Rafferty' (1982), aspectual :meanings reflect the evaluation (M? the speaker concerning the relationship of one event /state /activity to other events /states /activities in the discourse (66). Beedham also notes this factor: It refers to the way in which the event is seen to pass through time, i.e. whether the event is conceived of as imperfective. . ., perfective . ., iterative . . ., inceptive . . ., and so on (1982:84). 88 These definitions of aspect are not too liberal in allowing for speaker perception to play a crucial role in aspect, as the evidence presented in this section will show. An anomalously low Transitivity score (4) was found for sentence P085. This score cannot be explained by inherent aspect since 'dapat' "able" seems indeterminable since posessing an ability is not an action» Likewise, the score can not.kme explained by story time lapse since only a few minutes (certainly not millennia) have passed. So, in this case a 'se-' prefixed 'telah' was used in a sentence with relatively low transitivity score. Since transitivity scores of 5 were not considered anomalous, the low score here is not as problematic. The time frame is, however, of particular interest. Here, the main character is awaiting a critique from a man who commissioned him to write a speech. The speech was to reflect the changing political atmosphere, and. the writer felt bad for writing a speech with which he did not agree. When the man who commissioned the speech returns, reads the speech, and is ready to comment on the work, he hesitates. The speech writer realizes that something is amiss and is expecting to be insulted. He records the man's response as: 89 16. "Tapi begini saudara," katanya kemudian setelah but like this brother said-he then Df—after "But its like this brother," he said after men-dapat jalan untuk me-lanjut- kan. (P085) TR able way to TR-advance BENE. being able to continue. Obviously no more than a few minutes have passed in the discourse, yet the 'telah' form indicating 'ages' is used. The proper form should be 'sudah' to show that it was in reality a relatively short period of time (and to better correlate with the low transitivity). The time, however, was subjectively very long, so the token used was 'telah' to show that the time seemed to last an eternity for the speech writer. Instead of discarding the inherent aspect explanation of why 'sudah' or 'telah' is used, the definition of inherent aspect ought to be expanded to include the time lapsed on the story time line, the inherent aspect of the referred to action (even extra- clausally) and, finally the perception of elapsed time as recorded. by some character (or narrator)3. In short, instead of concentrating on the inherent aspect of the verb in the clause, the inherent aspect of the discourse needs to be considered to account for the choice of 'sudah' of 'telah' in a given situation. 90 4.5 Summary The marker 'telah' is used for perfect aspect when the verb indicates a) durative action. Likewise, the marker 'sudah' is used for perfect aspect when the context indicates EH1 action either indeterminable or punctual in nature which is now being brought into prominence in the story line. This study has shown that (contrary to many Indonesian grammars) 'sudah' and 'telah' are different entities which derive their differences from the inherent aspect of the verb involved in the phrase, or alternatively from the inherent aspect of the context. If the situation in the narrative is a durative one -- almost to the point of eternity--the choice is 'telah.‘ If the situation in the narrative is not durative, the choice is 'sudah.' 91 1. Although closely related, the verbal agreement with the inherent aspect theory refers to the actual verb (e.g. 'bayar' "to pay") while incidence (or use) refers to the total number of times a durative or non-durative verb was used with a marker. A verb may be used many times with a particular marker. 2. The English translation -although irrelevant to the thesis-- interestingly enough translates actions in the same clause tied by a conjunction. 3. There may be a possible hierarchical relationship here: A. With the inherent aspect of the verb handling most of the cases. (65%) B. Inherent aspect of the action referred to C. Story time lapse D. Character point of view However, I hesitate to posit such a sweeping continuum on the basis of a small written data base. CONCLUSION 5.1 Summary of findings This study has shown that the difference between the two Indonesian words, 'sudah' and 'telah' is that the former is used in non-durative contexts and the latter is used in durative contexts. These contexts may be as local as the verb in the same clause but are not limited to the sentence grammar. In some cases the choice of one marker over another is dependent upon the discourse context. Specific factors in the discourse which condition the choice of one token over another include the time passed on the story line, the durativity or non- durativity of the action referred to by the token, or the point of View of the narrator (or character) in the storyl. The time passed on the story line must be nearly eternal to require the use of 'telah,' whereas 'sudah' was found with time spans ranging between the immediate instant in) to ten years. .All the explicit time references using 'telah' were literally millions of years. In some cases, the 'sudah' or 'telah' in one sentence referred to an action or event mentioned in a previous sentence (or in the discourse in general). The durativity or non-durativity of the referred to action was key in these cases. Where there was a disparity between the story time lapse and the choice 92 93 of 'sudah' or 'telah', the point of View of the narrator (or character) was found to influence the choice. For example, when characters felt that the time passed slowly, they indicated that feeling' by using 'telah' where 'sudah' would 8150' have been appropriate. Likewise, even ii) highly transitive sentences (which usually take 'telah') if (Mme action happened immediately after another, the choice was 'sudah'. 5.2 Suggestions for further study This study addresses limited data and ought not to be taken as comprehensively addressing the discourse conditioning factors. One possibility for further study would be to explore discourse conditioning factors for 'sudah' and 'telah' in another set of data. Further use could also be made of Hopper and Thompson since this study did not address the question of story line. .An interesting adjunct ti) this study would be to determine which of these sentences fell on the story line of the narrative and which were in fact background. Perhaps it will be found that one or another of these markers serves as the story line aspect, while the other is pmimarily used for background information. There are more than ten aspect markers in Indonesian (see Chapter 1), and they are often found in 94 pairs (see Chapter 2). A further study might contrast more of these pairs of markers, differentiating one from another. Perhaps the durative/non-durative distinction (or some other inherent aspect) is a factor in some of these other pairs of markers as well. 5.3 Theoretical Implications Although this has been a short study of Indonesian discourse and the conditioning factors relevant to the choice between only one pair, this study suggests a few thoughts about current linguistic theory. 5.3.1 Hopper and Thompson In Hopper and Thompson, those events said to kme punctual are higher in transitivity (and more commonly found III the fereground cflf the story). This study found that those sentences associated with higher transitivity were of the durative type, not the non— durative (including punctual) type. Hopper and Thompson's universal theory' of ‘transitivity' predicts that those items which are associated with one element of high transitivity will be found to be associated with others (where they are evident). If two clauses (a) and (b) in a language differ in that (a) is higher in Transitivity according to any of the features A-J, then, if a concomitant grammatical of semantic difference appears elsewhere in the clause, that 95 difference will also show (a) ti) be higher in Transitivity. (1980:255) In other words, punctual verbs will be found in higher transitivity sentences. In this data, however, 'telah' associates higher transitivity with durativity while 'sudah' associates lower transitivity with punctuality. Predicting two different outcomes for foregrounded material. If transitivity scores are primary, 'telah' will be the more foregrounded because it has a higher average transitivity score. If punctuality ire more basic, then 'sudah' clauses will be foregrounded because they are associated with punctuality. The universal principle of transitivity espoused by Hopper and Thompson seems not to hold in this case. 5.3.2 Inherent aspect vs. aspect This thesis found that the markers of the perfect aspect in Indonesian are sensitive to the inherent aspectual distinction (n3 durative/non-durative. This provides evidence (at least in one language) to put to rest the question of whether these two similar semantic constructs are the same. Since the aspect is sensitive to the inherent aspect, the conclusion must be that they are different. 96 5.3.3 Lexical theory Finally, this thesis also jprovides quantitative evidence that two words, no matter how closely related, are indeed individual entities containing at least some semantic or pragmatic distinctions. In this case, the difference is the length of time (or perceived length of time) that an event lasted or since when an event occurred. 97 1. See Chapter 4, Footnote 3. APPENDICES APPENDIX A BISMA B001 Bisma bangkit dari tanah, udara dan air, yang Bisma arose from soil, air and water R.Cl Bisma arose from soil, air and water, that melebur jasadnya se-telah jutaan tahun yang destroyed body-his Df—already million year R.Cl destroyed his body a million years ago lalu pralaya dalam perang Bharatayuda. then ?fought in war Bharatayuda after he fought in the Bharatayuda war. B003 Mukanya yang dihiasi brewok dan cambang B004 B011 face his R.Cl PASS—dress up wiskers and sideburns His face, which was dressed up with white whiskers and sideburns putih sudah kisut akan tetapi masih tetap white already wrinkled FUT but still still already wrinkled but still memancarkan sinar yang jernih. radiate radience R.Cl clear radiating a clear light. Resi yang telah memikul pengorbanan Resi R.Cl already carry on shoulder sacrifice Resi who already carried that horrifying sacrifice yang dahsyat itu tiba-tiba mencul di Pasar R.Cl horrifying that suddenly appear at Mall suddenly appeared at Senen Mall. Senen. Senen Saudara-saudara sekalian, Bapak-Bapak, Ibu-Ibu, Siblings intense fathers mothers Siblings, fathers, mothers, anak-anak muda yang sudah mengenal maupun children young R.Cl already be familiar want and young children who are already familiar 98 8013 B015 B018 B020 99 yang belum dunia pewayangan jangan terkejut, R.Cl not yet world shadow puppet if not frighten and those who are not yet do not be frightened jangan. if not. if not. Nama saya Dewabrata, setelah bangkotan seperti Name my Dewabrata, Df—already advanced like my name is Dewabrata, after aging like this ini disebut Resi Bisma. this PASS-call Resi Bisma. I am now called Resi Bisma. (A) Ini sudah merupakan karma pala saya, this already TR-shape karma head my, This was shaped by my karma, (B) karena pada waktu muda saya dengan tidak because at that time young I with no because at the time of my youth I without sengaja telah menbunuh seorang wanita yang purpose already TR-kill person female R.Cl intention had killed a woman who membuntuti saya karena cinta. TR-followed me because love. followed me because of love. Maaf bagi mereka yang sudah tahu riwayat saya. Pardon they R.Cl already know story my. Pardon me, those who already know my story. Tetapi karena ayah saya jatuh cinta kepada but because father my to fall love to person But because my father fell in love with a seorang wanita dan wanita itu tidak mau a person woman and woman this NEG want woman and this woman did not want menjadi istrinya kalau anaknya nanti tidak to become wife-his if son-her next NEG to become his wife if her son were not next B023 8045 8051 100 mewarisi takhta, saya dengan sukarela telah TR-heir crown I with voluntary already heir to the crown, I voluntarily menyerahkan kedudukan saya pada adik-adik TR-surrender-BENE seat my to younger surrendered my seat to my younger step brother. tiri saya. step my. Saya sendiri telah mengikuti sayembara I myself already TR—join-PROC prize contest I already entered a prize contest to untuk mendapatkan tiga putri jelita untuk to TR-obtain-BENE three princess graceful for obtian three graceful princesses for my two kedua adik saya. two younger my. step brothers. Bahkan bukan dengan keyakinan tinggi bahwa furthermore NEG with confidence high IDSM Furthermore without much confidence that dengan itu saya telah menjalankan sebagian dari with that I already TR-perform divide from with that I had managed to make a good choice. kebenaran correctly Seandainya pilihan itu datang saat ini, discuss choice that to come moment this, Discussing the choice that was to come at this setelah jutaan tahun berlalu dengan kondisi after million years then with conditions moment, after a million years under conditions dan situasi yang berbeda, mungkin sekali and situations R.Cl difference,possible very and situations which are different, very possibly 8069 8073 8078 8091 8104 101 saya akan bertindak lain setidak-tidaknya I FUT to take steps other do not do not I would take other steps next time. berpikir dua kali. opinion two times. Sesudah itu membuka busananya. Df-after this TR-open costume-his. After this (he) opened his costume. Setelah itu ia mengambil sebuah kantung plastik after this he TR-get one pocket plastic After this he took out this plastic bag itu yang sejak tadi dicangkingnya. this R.Cl since just now PASS-carry something since he had been carrying it. "Dengan kata lain," sambung Bisma sambil with said other continued Bisma concurrently "In other words," Bisma continued while berkacak pinggang, "nama Bisma pun rasanya palms on waist, name Bisma also feel standing akimbo, "the name Bisma already feels sudah terlalu jinak di jaman ini untuk aku already then tame at era this for I too tame at this time for me to wear. pakai. wear. Lalu tiba-tiba saja Bhasmy sudah hilang dalam then suddenly I? Bhasmy already disappear in Then suddenly Bhasmy had disappeared in the kerumuman, tak kelihatan jejaknya. to-swarm no see trail-his crowd, his trail could not be seen. Sementara Bagawan Dorna juga sudah bersiap-siap while Bagawan Dorna also already ready While Bagawan Dorna also got ready menempuh hal yang serupa. to face matter R.Cl similar to face a similar matter. 102 B114 Kami melihat dinding yang rapuh runtuh we (excl) TR-see wall R.Cl weak collapse We saw a wall which was weak collapse perlahan-lahan dengan gaya slow-motion, lalu slow-slow with strength slow-motion, then slowly, slowly, as in slow motion, then tampak sebuah lubang tembus menuju gelap apparent one hole emerged toward darkness apparently a hole emerged toward darkness yang rupanya sudah menunggu di luar rumah R.Cl seems have been waiting at out house that seemed to have been waiting outside the house setiap malam. every evening. every evening. 8126 "Anakku, cucu-cucuku, children—my, grandchildren-my, "My children, my grandchildren, cicit-cicitku, kalian semua yang great-grandchildren you(pl) all R.Cl my great grandchildren, all you who memanggilku kembali, keisengan apa yang TR-call me come back boredom what R.Cl call me to come back, is it boredom that menyebabkan kamu menbangunkan aku dari TR-cause you TR-wake up me from caused you to wake me up from tidur abadi yang telah lampau. sleep eternal R.Cl already past my eternal sleep which has already past. B127a Apa yang telah terjadi dalam dirimu, what R.Cl already happened in yourself, What has happened to you, keluargamu, orang-orang di sekitarmu, your family, people at around you, your family, the people who surround you, 103 apa yang merisaukanmu dalam kehidupan ini, what R.Cl bother you in life this, what bothers you in this life, B127b apa yang telah menodai bendera yang 8130 8131 what R.Cl already TR—stain flag R.Cl What has stained the flag which berkibar di pekarangan rumahmu. flutters at yard house—you flutters in your front yard. Tidak mungkin kamu tidak bisa Not possible you not able It is impossible that you cannot mengatasinya, karena besok masih TR-overcome it, because tomorrow still overcome it, because tomorrow still akan ada kesempatan-kesempatan yang tak will be chances R.Cl not brings chances which are not terduga, lihat saja koran, baca Pos Kota, predictable see I newspaper, read Pos Kota, predictable, I see the newspaper, read Pos Kota, setiap hari berbagi nasib berguling pada every day various destiny rolling in every day destinies are rolling in time saatnya karena semua itu telah diatur. time because all that already arranged because it is all already arranged. Bersabarlah, kalian telah mengangis atau patient, you(pl)already TR-crying or Be patient, you have already been crying or have menahan tangis selama ini, kenapa tidak TR-hold back tear long this, why no held back tears this long, why not diteruskan selangkah lagi, selangkah lagi dan PASS—then step again, step again and wait step by step and 8133 8136 8137 104 rodamu yang terbenam dalam lumpur akan wheel-your R.Cl disappear inside mud will your wheel which has disappeared in the mud will melonjak ke titik kulminasi. TR-leap at point culmination leap to the culmination. Kamu tahu sendiri aku pun telah menukar you know yourself I also already TR-change You yourself know I also already had changed busanaku dan meninggalkannya di Proyek Senen, costume-my and TR-left-it at Mall Senen, my costume and left it at the Senen Mall, merubah namaku menjadi Bhasmy, karena TR-change name-my TR-finish Bhasmy, because my name is now changed to Bhasmy, because mereka, orang banyak, saudara-saudaramu yang they peOple many brothers-my R.Cl those, many peOple, my brothers who lain yang lebih banyak, lebih menyukai other R.Cl more than many more TR-happiness are more than many, want aku meninggalkan citra yang lama dan I TR-stay image R.Cl point in time and my image to stay current and menyanyikan lagu Rolling Stones. TR-sing like Rolling Stones sing like the Rolling Stones. Begitu banyak, begitu banyak citra yang sudah so many, so many image R.Cl already So many, so many images have already bergulir di dalam blantika edan di masa ini. fall down at in world crazy at time this fallen down in these crazy times. Jangan memanggilku lagi, jangan menyebutku don't TR summon me again, don't TR-call me Don't summon me again, don't call me 105 Bisma lagi, kau tahu sendiri Bisma sudah Bisma again, you know yourself Bisma already Bisma again, you yourself know that Bisma has dihapiskan, sekarang yang ada hanya Bhasmy. PASS-eliminate now R.Cl exist only Bhasmy. already been eliminated all that exists is Bhasmy. B155 Pada saat yang sama dengan cara-cara yang sama at time R.Cl same with ways R.Cl same At the same time and in the same way, kami yang telah menjadi Bisma-Bisma kecil we R.Cl already TR-become Bismas small we who had become small Bismas meronta dan bangun terhuyung—huyung. TR-struggle and get up staggering. struggled and got up staggering. B161 Aku sudah di sini! I already at here I have been here! B162 Aku sudah kembali! I already come back I've already come back! (B164 originally missed in the translation and so, not counted -- included for accuracy to the text) B164 Keluarga kami, keluarga sendiri, masih tetap family we(excl) family my own still fixed My family, my own family, still snoring, mendengkur, tak tahu sandiwara apa yang sudah TR-snoring NEG know drama what R.Cl already didn't know the drama that had already ter- jadi. x-ed happened. happened. B165 Aku tidak tahu apakah mereka benar—benar tidur I NEG know if they really sleep I didn't know if they were really asleep hni'v'. 106 atau sudah ikut bersandiwara berpura-pura or already join drama pretending or if they already joined the drama pretending tidak melihat kekonyolanku, untuk menjaga NEG TR-see foolishness-my for TR-guard not to see my foolishness to guard my feelings. perasaanku. feeling—my. B169 Aku merasa ribuan mata memandang rumah kami 8170 8173 8174 I feel thousands eyes TR-gazing house our I feel thousands of eyes gazing at our dark house yang gelap, tahu apa yang telah terjadi. R.Cl dark, know what R.Cl already happened knowing what had happened. Tahu di dalam rumah kami seorang telah know at in house our person already Knowing inside our house a person already memaksakan dirinya menjadi Bisma. TR-forced himself TR-become Bisma had forced himself to become Bisma. Sambil menarik nafas panjang, aku kenakan while TR-take breath deep, I wear While taking a deep breath, I put busana wayang itu di atas baju dril dan costume puppet that at on top clothes cloth and that puppet costume on top of my clothes and celana jean-ku yang sudah belel. pant jeans-my R.Cl already bell bell bottomed jeans. Mulai saat ini aku akan menjadi tokoh beginning moment this I will TR-become role Beginning this moment I became the wayang di masa lampau dengan citra yang puppet at time past with image R.Cl puppet in the past with the image that 8177 8178 8197 107 barangkali saja sudah ketinggalan jaman. maybe I already stay era was already out of date. Jalanan telah ramai dan tukang bubur Street already crowded and vendor soup The streets were already crowded and the ayam di kelokan jalan telah membuka chicken at corner street already TR-open Chicken soup vendor on the street corner already dagangannya. stall his opened his stall. Aku berdiri karena waktunya telah tiba I stand up because at that time already arrive I stood up because the time had come for me to untuk pulang, supaya semua ini tidak for go home, so that all this NEG go home, so that all this will not perlu kehilang hikmahnya karena menjadi necessarily loose meaning its because TR-become necessarily looses its meaning just becoming tontonan. entertainment entertainment. Semua itu telah bersatu dengan baju dril All that already mixed with clothes cloth All that had already mixed with my clothes dan celana jean-ku, bahkan juga lengket tak and trousers jean my furthermore also stick not and jeans, furthermore it also became inseparable terpisahkan dari kulit dan tulang-tulangku. separable from skin and bones -my. from my skin and bones. ‘m- nus-rm: xxx. v51§u|1 APPENDIX B MENOLEH KE BELAKANG M001 M002 M005 M021 M024 Sudah sepuluh tahun. Already ten years. It's already been ten years. Sudah sepuluh tahun ini terjadi. Already ten years this happened. Already ten years have passed since this happened. Kalau ia lupa melakukannya maka ia If she forget TR-conduct-it then she If she forgot to do it, then she seakan-akan sudah kekurangan garam Df will already less salt would (feel) [dehydrated] esok harinya. tomorrow day the next day. Tapi kalau itu sudah dilakukan but if that already PASS-done But if this has already been done sepuluh tahun, barangkali boleh dicurigai ten years perhaps may PASS—suspicious for ten years, it might perhaps seem suspicious sebagai sesuatu kelianan. a type a certain quirk a certain kind of quirk. Tapi kalau sudah sampai setagihan, but if already to the point of addiction But if it is already an addiction, artinya kalau tidak merokok, lalu meaning if NEG TR-smoke then then if you can't smoke, then 108 M027 M029 M041 M042 109 badan jadi nggak nafsu kerja, itu yang harus body become not want work that R.Cl must your body loses its ambition, that must dicegah. PASS-prevent be prevented. Saya tidak merasa tidak merdeka hanya karena I NEG TR-feel NEG TR—free only because I do not feel enslaved just because sudah menoleh! already TR-glance! I glance! Tapi pada suatu hari, waktu wanita itu but at one day, time-completed woman that But one day, when that woman menoleh ke belakang sebelum tidur-- Tr-glance at back before sleep (who) glanced back before sleeping-- ia sudah berbaring, lalu tiba-tiba teringat she already lie down, then suddenly remembered she had already lay down, then suddenly remembered untuk menoleh- ia mengangkat badannya lalu for TR—glance she TR-lifted up body-her then to glance back-- she sat up then menoleh ke atas kasur. TR-glance at on top bed glanced at the top of her bed. Jadi lihatlah, sudah 10(sic) tahun saya finish see already ten years I So you see, for ten years I have been trained diajarkan oleh tubuh saya untuk berhati-hati. PASS-train by body my for to be careful by my body to be careful. Jadi barangkali sejak kalajengking itu masih finish perhaps since scorpion that still Probably in the time since that scorpion was still 110 kecil atau belum ada, tubuh saya sudah small or not yet exist body my already small or not even born, my body already mengetahui akan kedatangannya, lalu ia TR-know-PROC will come it then it knew it would come, then it (my body) mempersiapkan diri. TR—prepare self prepared itself. M064 Semuanya merasa bahwa itu sudah wajar. all it TR-feel that it already natural Everyone felt that it was natural. M080 Jadi sudah sepuluh tahun juga finish already ten years also So for ten years her body tubuhnya mempersiapkan dirinya body-her TR-ready self—it also prepared itself untuk ditabrak? for PASS-hit to be hit? APPENDIX C PELACUR P001 Pada suatu kali, aku dapat pesanan membuat at one time I able commission TR-make P005 P012 P022 P023 Once, I was commissioned to make sebuah teks pidato untuk menghina seorang a text speech for TR-humble person pemimpin yang sudah meninggal. leader R.Cl already TR-dead leader who was already dead. Sudah sering aku bilang pada dia bahwa apa already often I said to him that what I have already often said to him that what yang ditempuhnya tidak benar, tapi dia R.Cl PASS-responsible NEG correct but he he was responsible for was not right, but he terus juga keras kepala. then too hard head was too stubborn. Sudah tiga gelas kopi kupakai menyuap already three glass coffee my use TR-mouth I already had three cups of coffee. kepalaku. head-my Kenapa sekarang aku baru ditengtang, why now I begun PASS-opposed, Why now am I beginning to be opposed, setelah aku tidak bisa lagi menjawab. after I NEG able PROC TR-respond after I am not capable of responding. Karena kalau aku masih hidup sampai sekarang, because if I still alive until now, Because if I were alive now, 111 P024 P028 P045 P046 P049 P051 112 bisa jadi banyak hal lain yang sudah able finish many matter other R.Cl already I would be able to finish many other matters which kulakukan. I-conduct was conducting. Semua yang sudah kukerjakan kan hanya proses all R.Cl already my-work will only process All that I have been working on is still only in untuk satu tujuan yang lain. for one purpose R.Cl other process will be for others. Ketika teringat apa yang terjadi di dalam mimpi, now remember what R.Cl happened at in dream, I remember what happened in my dream, aku kira aku sudah menipu diriku sendiri. I guess I already TR-trick self-my one self I guess I had tricked myself. Waktu itu hari sudah pukul empat. time that day already hour four. It was already four o'clock. Tetanggaku yang jual tempe sudah neighbor-my R.Cl business soybean cake already My neighbor who sells soybean cakes was already bangun. awake awake. Tak ada kebahagiaan apa-apa, meskipun sudah No is satisfaction what, although already There was no satisfaction, eventhough menyelesaikan sebuah pekerjaan. TR-ready one work the work was already done. Sudah itu melupakan apa yang baru aku tulis. Already that TR-forget what R.Cl begin I write I had already forgotten what I had written. P054 PO60 P065 P071 P075 P076 P085 113 Aku telah menjual kelihaianku menyusun kata. I already TR-sold cunning -my TR-compile words I had already sold my ability to write. Aku tak sempat mengucapkan kata basa-basi, I no opportunity TR-express word pleasantries I didn't even have a chance to say 'hi,‘ ia sudah langsung meraih kertas. he already go on gather paper he already had gathered the paper. Kukira rancangan pidato itu sudah tepat. my-guess organization speech that already exact My organization of the speech was exact. Segalanya sudah disusun dengan baik, all- it already PASS compile with good, It had all already been written well, taktis dan cukup punya orientasi sehingga tact and sufficient have orientation until with tact and its orientation having just the mempunyai jangkauan yang jauh. TR-have reach out R.Cl far right reach. Berapa lama anda sudah menulis? how long you already TR-write How long have you written? Apa saja yang sudah anda tulis? What just R.Cl already you write What else have you written? "Tapi begini saudara," katanya but like this brother," said-he "But its like this brother," he said kemudian setelah mendapat jalan untuk then after TR-able way to after he was able to melanjutkan. TR-advance continue. P111 P135 P138 P140 P145 P157 P162 114 Sudah ya nanti saya ambil. already ok next I take OK, I'll take it later. Aku telah cair dalam permainan orang lain. I already liquid in play other person I was already liquid being played with by another. Ketika pintu terbuka dan pemesan pidato at point door open and buyer speech When the door was opened and the buyer of itu masuk, aku sudah siap untuk menerima that go in, I already ready for acceptance came in, I was ready to accept penghinaan dan kemarahan. humble and anger abuse. "Selamat siang," katanya dangan gembira, "sudah save day, said-he with glad already "Hello," he said gladly, "already finished?" selesai?" ready Menanti bagaimana seseorang yang kecewa next how Df man R.Cl disappointed I waited to see how the man who was disappointed dan merasa sudah membayar akan menyemprotku. and TR-feel already TR-paid will TR-attack me. and having already paid would attack me. Tepat sekali!" serunya setelah menamatkan exact very cried-he after TR-finished Perfect!" he cried after he had finished halaman terakhir. page last. the last page. Lalu bagaikan terbang keluar rumah, then kind fly out house, Then he flew out of the house, P165 115 karena waktunya sudah sangat mendesak. because time-it already very TR-press because the time had already come. Lembaran yang seyogyanya menjadi lembar page R.Cl obvious-it TR—been page This page which was obviously would have been pertama perubahan pidato itu, tapi tak mampu first different speech that but not capable the first page of the revised speech, but I was kulanjutkan karena merasa telah diperbudak. continue because TR-feel already PASS-slave unable to continue because I had felt enslaved. APPENDIX D DALLAS DATA D004 D011 Dan dia datang ke kijang itu, kijang itu sedang and he came to kijang that kijang that PROG And he came to that kijang, the kijang ter- perangkap. accident-trapped was accidently trapped. Dia mengangkat kayu itu dan memerangkap sang he TR- lift wood that and TR- trap the He (the alligator) lifted the wood and trapped kijang di antara dua belah pohon itu akhirnya kijang at between two half tree that end- his the kijang between the two halves of the tree juga. too that was his end as well. 116 APPENDIX E 'Se-' AFFIX IS DEFINITE This thesis deals with the difference between 'sudah' and 'telah'. This