MASYARAKAT LINGUISTIK INDONESIA No. 01 Feb 2014.compressed.pdf · Keanggotaan institusional dalam negeri adalah Rp250.000 dan luar negeri US$50per tahun. Naskah dan resensiyang panduannya

MASYARAKAT LINGUISTIK INDONESIADidirikan pada tahun 1975, Masyarakat Linguistik Indonesia (MLI)

merupakan organisasi profesi yang bertujuanmengembangkan studi ilmiah mengenai bahasa.

PENGURUS MASYARAKAT LINGUISTIK INDONESIAKetua : Faizah Sari, Universitas Katolik Indonesia Atma JayaWakil Ketua : Rindu Parulian Simanjuntak, SIL InternasionalSekretaris : Siti Wachidah, Universitas Negeri JakartaBendahara : Katharina Endriati Sukamto, Universitas Katolik Indonesia Atma Jaya

DEWAN EDITORUtama : Bambang Kaswanti Purwo, Universitas Katolik Indonesia Atma JayaPendamping : Faizah Sari, Universitas Katolik Indonesia Atma JayaAnggota : Bernd Nothofer, Universitas Frankfurt, Jerman; Ellen Rafferty,University of Wisconsin, Amerika Serikat; Bernard Comrie, Max Planck Institute; TimMcKinnon, Jakarta Field Station MPI; A. Chaedar Alwasilah, Universitas PendidikanIndonesia; E. Aminudin Aziz, Universitas Pendidikan Indonesia; Siti Wachidah,Universitas Negeri Jakarta; Katharina Endriati Sukamto, Universitas Katolik IndonesiaAtma Jaya; D. Edi Subroto, Universitas Sebelas Maret; I Wayan Arka, UniversitasUdayana; A. Effendi Kadarisman, Universitas Negeri Malang; Bahren Umar Siregar,Universitas Katolik Indonesia Atma Jaya; Hasan Basri, Universitas Tadulako; YassirNasanius, Universitas Katolik Indonesia Atma Jaya; Dwi Noverini Djenar, SydneyUniversity, Australia; Mahyuni, Universitas Mataram; Patrisius Djiwandono,Universitas Ma Chung; Regina Yanti, Universitas Katolik Indonesia Atma Jaya.

JURNAL LINGUISTIK INDONESIALinguistik Indonesia diterbitkan pertama kali pada tahun 1982 dan sejak tahun 2000diterbitkan tiap bulan Februari dan Agustus. Linguistik Indonesia telah terakreditasiberdasarkan SK Dirjen Dikti No. 040/P/2014, 18 Februari 2014. Jurnal ilmiah inidibagikan secara cuma-cuma kepada para anggota MLI yang keanggotaannya umumnyamelalui Cabang MLI di pelbagai Perguruan Tinggi, tetapi dapat juga secaraperseorangan atau institusional. Iuran per tahun adalah Rp 200.000 (anggota dalamnegeri) dan US$30 (anggota luar negeri). Keanggotaan institusional dalam negeriadalah Rp 250.000 dan luar negeri US$50 per tahun.

Naskah dan resensi yang panduannya dapat dilihat di www.e-li.org dikirim keRedaksi dengan mengikuti format Pedoman Penulisan Naskah di bagian belakangsampul jurnal.

ALAMATMasyarakat Linguistik IndonesiaPusat Kajian Bahasa dan BudayaUniversitas Katolik Indonesia Atma JayaJI. Jenderal Sudirman 51, Jakarta 12930, Indonesiae-mail: [email protected], Tel./Faks.: +62 (0)21 571 9560

Daftar Isi

Forensic Linguistics: Forms and ProcessesGeorgina Heydon ............................................................................. 1

A Distributed Morphology Analysis of Indonesian ke-/-an VerbsLanny Hidajat ................................................................................... 11

Measuring Proficiency in Standard Indonesian for Enggano SpeakersRindu Parulian Simanjuntak ........................................................ 33

Akronim yang Berfonotaktik Tidak Lazim dalam Bahasa IndonesiaSariah ................................................................................................ 47

Verbs of Excretion in TabaFrederick John Bowden .................................................................. 63

The Role of Culture in the Translation Process throughThink-Aloud ProtocolsJulia Eka Rini .................................................................................... 77

Resensi:Jan Zienkowski, Jan-Ola Östman, dan Jef Verschueren

Discursive Pragmatics(Handbook of Pragmatics Highlights, volume 8)

Diresensi oleh Faizah Sari .......................................................................... 93

Jelajah Linguistik:Meninjau Ulang Metodologi Kontemporer

Faizah Sari .................................................................................................. 95

Bincang antara Kita dari Dunia Maya:Susunan Pengurus MLI dan Dewan Editor Linguistik Indonesia .... 99Deklarasi dan Ikrar Anggota MLI ................................................... 100Iuran Keanggotaan MLI dan Sumbangan Penerbitan Artikel .......... 101

Linguistik Indonesia, Februari 2014, 1-10 Volume ke-32, No. 1Copyright©2014, Masyarakat Linguistik Indonesia, ISSN: 0215-4846

FORENSIC LINGUISTICS: FORMS AND PROCESSES

Georgina Heydon*RMIT University

[email protected]

AbstractFollowing a brief introduction to the notion of forensic science and analysis, this paperwill explain the different ways in which linguistics has contributed to policeinvestigations and civil law. The paper will cover linguistic identification using spokendata and written data, and will discuss the use of discourse analysis as well as the moretraditional phonetic and syntactic analysis for forensic examinations. Other applicationsthat will be discussed include analysis for language of origin in refugee status claims,commercial applications and trademark disputes, and lie detection. Each of theseapplications will be considered critically and in relation to both the validity of thetheories underlying them, and the statistical reliability of the analysis used to attainresults.

Keywords: Forensic linguistics; police investigations; credibility assessment

AbstrakMenyusul pengenalan singkat pengertian ilmu dan analisis forensik, makalah ini akanmenjelaskan bagaimana ilmu linguistik telah memberikan kontribusi terhadappenyelidikan polisi dan hukum perdata. Makalah ini akan mencakup identifikasilinguistik menggunakan data lisan dan data tertulis, dan akan membahas penggunaananalisis wacana serta analisis fonetik dan sintaksis yang lebih tradisional untukpemeriksaan forensik. Aplikasi lain yang akan dibahas meliputi analisis untuk bahasaasal gugatan status pengungsi, aplikasi komersial dan sengketa merek dagang, dandeteksi kebohongan. Masing-masing aplikasi akan dipertimbangkan secara kritis dandalam kaitannya dengan kedua validitas teori-teori yang mendasari mereka, dankeandalan statistik analisis yang digunakan untuk mencapai hasil.

Kata kunci: Linguistik Forensik; penyelidikan polisi; penilaian kredibilitas

INTRODUCTION

When most people hear the term ‘forensics’ they immediately think of boffins in white coatsperforming almost magical scientific tests to provide that crucial piece of rock-solid evidencethat solves a complex crime. On the rare occasion that a language expert is featured in one of themany television crime dramas, the expert is invariably able to rely on some kind ofsupercomputer to identify suspects by their ‘voiceprint’, irrespective of the quality or quantity ofdata available. Sadly, such an image is strictly limited to the realm of science fiction at thisstage, as no-one has yet been able to isolate a uniquely distinguishing feature of the humanvoice upon which to base such a ‘voiceprint’. Nonetheless, the student of linguistics, excited bythe possibilities of such a real-world application of their skills in analysing language, should notbe put off by this gap between myth and reality: the truth is that forensic linguistics is a morevaried and fascinating field than could ever be imagined by a lay audience. Forensic linguistics,and the study of language and the law more broadly, requires the analyst to enlist a wide varietyof analytic tools and skills, drawing on almost every aspect of core linguistic study, fromphonetics to pragmatics, from syntax to sociolinguistics. Forensic linguistic enquiries may formpart of criminal investigations by police, or they may be initiated by the defence team on behalfof a client. They may involve trademark disputes or even non-legal concerns, such as selectingthe most appropriate brand name for a product to be launched in a specific market.

Georgina Heydon

2

The International Association of Forensic Linguists and the International Associationfor Forensic Phonetics and Acoustics were formed in the early 1990s by specialists working inthese areas and, while each organisation represents a different emphasis in the field, they areboth represented by the scholarly publication, the International Journal of Speech, Language andthe Law.

The aim and purpose of International Association of Forensic Linguists, from thewebsite www.iafl.org: ‘The purpose of the Association is to improve theadministration of the legal systems throughout the world by means of a betterunderstanding of the interaction between language and the law.’

AN EARLY EXAMPLE OF FORENSIC LINGUISTIC ANALYSIS

The following case study will give some indication of the breadth of scholarship needed by theforensic linguist in providing expert evidence.

In 1950, Timothy John Evans was hanged for the murders of his wife and child at10 Rillington Place, London. Evans had maintained his innocence, and accusedJohn Reginald Halliday Christie of committing the crimes, despite the fact thatthe police were able to produce an allegedly verbatim record of Evans confessingto the murders in his police interview. In a dramatic turn of events, Swedishlinguist Jan Svartvik showed, in 1968, that the confession used to convict Evanswas most likely the product of police influence and differed sharply in style andstructure from the remainder of Evans’s statements. Svartvik 1968 was able toshow, using linguistic analysis of the discourse structures, that the key sections ofthe statement, where Evans apparently confesses to the murders, are writtenusing a formal register, typical of police texts, but most atypical of the speech ofEvans and inconsistent with the remainder of the interview transcript. Evans wasposthumously pardoned and, when several more bodies were discovered at 10Rillington Place, Christie was hanged.

Svartvik’s analysis of the Evans case, often cited as the original forensic linguistic investigation,involved the analysis of syntactic structure, style, register, spoken language versus writtenlanguage, and knowledge of the specific police register used in statement taking. It is interestingthat this landmark case did not involve the analysis of vocal qualities, which has now becomethe archetypal forensic linguistic application.

HERE ARE SOME TYPES OF FORENSIC ANALYSIS

IdentificationThe types of analysis that are most commonly associated with the practice of forensic linguisticsare those that seek to identify the source of a message. This includes the analysis of spoken andwritten data, and can involve some level of computer or statistical analysis. The data being usedin the analysis will almost always include at least one sample of ‘known data’, where the sourcehas been reliably identified, and one sample of ‘questioned data’, where the source is unknown.There may be some rare cases where two data samples are compared to establish whether or notthey originate from the same source, without either being identified as belonging to a specificspeaker. A ‘scoresheet’ may be devised, where each segment from the questioned data is ratedagainst the known data for similarity.

Linguistik Indonesia, Volume ke-32, No. 1, Februari 2014

3

Spoken DataThere is some debate as to whether identifying the source of spoken data should be labeled‘voice identification’ or ‘speaker identification’, but either way, the process for this kind ofanalysis involves comparing samples of recorded voice data in order to establish the likelihoodthat they represent samples of the same voice, versus the likelihood that they represent samplesof different voices. Individual sounds as well as longer stretches of talk are usually isolated forcomparison across the samples. Nowadays, this is done using some form of digital sound editingsoftware that allows the analyst to isolate very small amounts of speech data, such as one vowelsound, into separate files or segments. In this way, the analyst can collect many tokens ofspecific sounds or utterances which represent different aspects of the voice(s). So for example,the analyst may choose to isolate all the instances of close front unrounded vowels (the vowel inseen), and compare them across the samples using either an automated or machine-assistedprocess or by listening with an ‘expert ear’. This process would then be repeated for a range ofindividual phonemes, as well as for longer stretches of talk.

The approach that is commonly known as forensic phonetics involves the automated orsemi-automated processing of vocal data segments according to measurable features, such asformant values (see the box below), pitch range or the rate of speech. This computer-basedanalysis may be supported by the expert phonetician’s perception of the vocal features in eachsample.

During speech production, the air in the vocal tract vibrates at many differentfrequencies at once and in the production of vowel sounds, the most dominantfrequencies combine and appear as bands on a spectrograph image of the airvibrations. Each of these bands is known as a formant and the different vowelsare typically characterised by the different combinations of the first threeformants (i.e. the three bands appearing at the lowest frequencies).

When the analysis relies solely on the expert’s ‘ear’ to determine the phonetic qualitiesof the samples, the term used by some forensic linguists is ‘aural-perceptual analysis’ (Hollien2002). The analyst considers the data in terms of various vocal parameters, such as perceivedpitch, articulation and prosody, which can be heard and compared across samples.

The aural-perceptual analysis in particular may also rely on the analyst’s knowledge ofdialectology and sociolinguistics to compare the data samples. In this kind of analysis it is thepresence of certain dialectal features that enables the analyst to distinguish the two speakers.

For the analysis of spoken data in legal cases, the known data samples are very oftendrawn from a recording of the police interview with the suspect. The unknown data samples areusually some form of covert recording obtained by tapping the suspect’s phone, or from aconcealed recording device, but in one case well known to the author, the questioned audio datasamples were extracted from a video surveillance tape (CCTV) made during armed robberiesallegedly committed by the suspect. While the video footage was of such poor quality that theperson committing the crimes could not be identified visually, the audio tracks were of highquality and provided critical evidence in the case.

Written DataThe principles of the forensic examination of written data are essentially the same as for spokendata – that is, samples of known and questioned data are compared according to a set oflinguistic parameters. In the case of written texts, the sorts of features that can be systematicallyanalysed include punctuation, sentence structure, verb types, terms of address, spelling andgrammatical errors, any idiosyncrasies of the writer, and broader patterns of discourse, such asthe development of specific themes.

Georgina Heydon

4

One of the problems facing prosecutors in legal cases that involve written data is how toobtain known data samples from the suspect. Very often, the questioned material, such as athreatening letter or ransom note, is readily available as it forms the basis for the complaintagainst the suspect. The known text, where available, often has to be drawn from personalcommunications signed or otherwise identified as written by the suspect. Email is becoming amajor source of this kind of data, but it is important that the impact of the medium ofcommunication is taken into account when comparing email with, say, a handwritten letter.

The method of analysis might also present problems for a court. In the case describedbelow, the court was unable to accept the results of a statistical analysis of punctuation data butinstead decided to rely entirely on the qualitative analysis of specific features. This highlightsthe differences between courtroom rules of evidence and academic standards of validity. Thisissue is discussed further in the Summary of Key Issues below.

Case study:A murder trial in which I gave evidence in the Bendigo Supreme Court (Victoria,Australia) involved the analysis of four anonymous letters that contained threatsagainst both the deceased and the accused. The letters were produced asevidence of the accused man’s innocence by the defence, but were deemedsuspicious by the prosecution, as explained below.The case revolved around the death by arson of a middle-aged woman living in arural part of Australia with her husband. The couple had not been living in thearea very long, and had apparently been subject to a campaign of bullying or atleast social rejection and threats from the local community, and might have hadsome specific enemies in the town. A short time after they had arrived in thetown, the wife died in suspicious circumstances as a result of a fire at thecouple’s property. During the police investigation into her death, the husbandwas arrested and charged with her murder, but he subsequently pleaded not guiltyin court.In his defence, the husband produced four short letters printed together on asingle sheet of paper. He claimed that he had copied the letters from theiroriginal source although it was unclear in the trial whether this was an emailsource or hard copy letters. In any event, the evidence consisted of a series offour letters that were alleged to have been written to the couple by a third party orgroup of people who were apparently vehemently opposed to the couple residingin their community. The defence had presented these letters as evidence thatthere was an existing threat against the wife and it was these other people thathad carried out attack that ended the life of the victim. When other evidence wasgathered that suggested the husband was the perpetrator, such as crime sceneevidence related to the fire, the police decided that the letters should be verifiedfor their authorship.At the time of my analysis I was unaware of the context surrounding theevidence. I was only provided with a copy of the four letters and a collection ofemails sent by the husband to his wife immediately prior to her death. Theseemails were used in the analysis as the ‘known data’, since their authorship wasnot in dispute.Despite my lack of knowledge about the case, I believe I did make somesuppositions, unconsciously, about the data. For instance, I recall that I had anexpectation that all the threatening letters would have been produced by the sameauthor – either the suspect or some unidentified party. In other words, although Icompared each letter to the known data set individually, I expected to find that


5

either all the letters would match the known data set or none of them wouldmatch the known data set. It was therefore an unexpected outcome that three ofthe letters did have a strong match with the known data set, but one letter, thefirst letter on the page, had a very weak match, or a negative match, with theknown data set.The analysis of data that I carried out consisted of two main phases: the first wasan analysis of syntactically classified punctuation as described in Chaski (2001).In this paper, Chaski demonstrates how punctuation marks that are quantified andclassified according to their syntactic function can be used reliably todifferentiate between authors of a number of written (English language) texts.The second phase was a comparative stylistic analysis of specific words andphrases used across the two sets of texts. There were a number of ratheridiosyncratic spelling and grammatical choices that appeared in the known dataset and the three later letters, but not in the first letter. While punctuation wasanalysed in the first phase for its syntactic function, in the second phase I notedsome patterns of punctuation that appeared in the three later letters and the emailsbut not the first letter. The pattern of first person pronouns (“I” and “we”) thatwere used in the letters also changed between the first letter and the later threeletters. Finally, I examined the use of taboo language (terms of abuse) and Ifound that there was a clear shift between the first letter and the later three lettersin the type and strength of the abusive language that appeared. English swearwords are highly sensitive to hierarchical ordering: that is, English speakerseasily able to distinguish between swear words according to their level of socialtaboo and there are clear examples in everyday usage of certain words beingbanned under certain circumstances, most obviously in the media. Therefore,when a writer uses some moderate swear words but not other more offensivewords, they are making a conscious choice to restrict the level of taboo language.In the evidence for this case, the first letter contained emphatic but onlymoderately offensive swear words. This contrasted strongly with the remainingthree letters where the swear words and taboo phrases used were of the mosthighly offensive categories, for instance, words that are almost always banned inpublic television broadcasts even for an adult audience.As is typical in forensic casework, it was about two years before I was called togive evidence in court. At the Supreme Court trial in Bendigo it became clearthat the case for the prosecution was consistent with my findings. Although thecouple had received one threatening letter from anonymous communitymembers, and may have received verbal threats as well, the accused was believedto have written the other three letters to exaggerate the threat and present theauthor/s of the original letter as the murderers. My evidence was consideredalong with other forensic evidence such as chemical analyses of the burntmaterials in the room where the wife perished to aid the identification of theaccelerants used in the fire.My evidence was presented over two days, most of which was taken up by thedefense in cross examination. Naturally, a great deal was made of the differencesidentified between the first letter and the other three, and the defense case restedmainly on the assertion that just as authors can change their style of writing overthe course of their career, an individual might change the way that they write aseries of threatening letters.It is difficult to refute an assertion like this when being cross-examined in theadversarial court system, because you are confined to answering questions. The

Georgina Heydon

6

counsel for the defense began by asking me if I’d read several works byHemingway, or EM Forster or Evelyn Waugh. There ensued a rather uglyattempt to smear my professional standing on the basis that I had not read muchof these early twentieth century writers. I suggested Mervyn Peake, but Peake’sbooks were unknown to the defense counsel, and at that point the judgeinterrupted this farcical cross-examination to insist that the defense counsel get tothe point. Counsel then tried to have me agree to the assertion that althoughauthors have a distinctive style, this can change over time, just like a letter writer.I refused to agree with this assertion, and forced Counsel to request that I explainmy refusal. This gave me the opportunity to point out that the work of famousauthors is not at all like the production of letters used in forensic linguisticanalysis most obviously because, unlike famous authors, threatening letterwriters do not have their work modified by an editor. Astonishingly, the counselfor the defense tried to claim that truly great authors never have their workmodified by editors, but even the jury laughed at this claim. It was difficult toignore the sarcasm and derision used by the defense counsel to emphasize hisclaims, but at such times it is always necessary to focus on the content andvalidity of the claims, and not the emotional impact of their delivery.As the cross-examination continued, I was able to explain that the analysis ofsyntactically classified punctuation (Chaski 2001) that I had undertaken is notjust a matter of style, but can distinguish between authors on statistical grounds.The jury were quite interested in my evidence, perhaps because it was easier tounderstand than chemical analysis and burn marks. The two types of evidencethat I gave seemed to be reasonably well understood, and although thecomparative stylistic analysis was clearly more accessible to the jury than thestatistical analysis of punctuation, like most researchers, I felt the quantitativepunctuation analysis was more reliable.The judge, however, eventually made a decision that the evidence relating to theChaski method of analysis could not be admitted, because Chaski’s article hadonly been published three or four years before I had used it in my analysis. In acourtroom, this is not considered long enough for a published method to beproperly tested and accepted as reliable.This illustrates an important feature of courts, at least in an adversarial system:the main test of whether a particular kind of evidence can be admitted as part of acase is precedence. If evidence of this kind has been admitted by a judge in ahigher court, or an equivalent court in another country, then it is deemedacceptable. What this means, however, is that whenever a completely new kindof expert evidence is presented to the court, the judge faces a great deal ofpressure in deciding whether or not to allow the novel procedure or expertevidence. If the evidence is allowed, then it sets a legal precedent, and all otherlower or equivalent courts can accept this evidence too. In the Supreme Court,the judge’s decision to allow a new type of evidence will flow on to most othercourts in the country (Supreme, County and Magistrates for example). Thus, itshould not have been surprising that the Bendigo Supreme Court judge wasrather cautious about allowing what he saw as a novel form of analysis, namelyChaski’s syntactically classified punctuation, even though from a scientificperspective, his reservations were ill-founded.


7

Over the last ten years, cases have been undertaken by forensic linguists that involved mobilephone text messages, and the distinctively individual nature of text messages is considered anadvantage in author identification. For instance, handwriting and typeface comparison are notconsidered part of forensic linguistic practice, but are often carried out by the relevant experts inconjunction with the forensic linguistic analysis.

Discourse AnalysisThe area of forensic discourse analysis is concerned with the appropriateness of interviewingtechniques, the interpretation of legal texts, suspect profiling and other applications of discourseanalysis to a legal case or to law enforcement more broadly.

An interpretation of a legal text for a Bankruptcy case in heard in the Federal Court of Australiais described below.

The expert testimony reads: The semantic interpretation of the adjectival phrasestanding to the credit of rests in the aspect of the verb form standing. It ispossible to expand the elliptical form of the phrase from the moneys standing tothe credit of to the full, implied form the moneys [that are] standing to the creditof. The structure of standing is thus TO BE + stand + ING. This form is thepresent progressive and is described by Kreidler 1998 as a temporary or boundedform. Kreidler further notes that: ‘the present progressive is used for what istemporarily true’. Thus, the activity standing to the credit of is confined to abounded period of time – it does not extend indefinitely as might be the case withthe simple present stands to the credit of, which would express a durativeactivity, something that may continue to be true. Conclusion. According to thesemantic analysis of the questioned text, the phrase the moneys standing to thecredit of is highly likely to be interpreted by the addressee to mean that themoneys are those standing to the credit of the relevant accounts at the time thatthe words were written – that is, the date of the Order or, at the latest, the date ofthe Order being read by the addressee.

In this case, an Order was made by a judge prohibiting the defendant from accessing moneyheld in various bank accounts. The forensic linguistic analysis focused on a phrase used in theOrder where it described the relevant funds as ‘moneys standing to the credit of [bank accountsheld by the defendant and his associates]’ and considered whether or not the paragraph would bereasonably interpreted to mean that the word ‘moneys’ referred only to funds in the relevantaccounts at the time that the Order was made, and not to any future funds credited to theaccounts.

Language of OriginLinguists may be called on to identify a speaker’s national or regional origin. This is done byanalyzing features of their language such as accent, vocabulary and grammatical features.Clearly, this type of analysis has applications in establishing refugee status, where it iscommonly used, but as it requires the analyst to have a thorough knowledge of both languageacquisition theory and the very specific regional dialect spoken by the subject, its legitimacy isstrongly contested by organizations such as the International Association of Forensic Linguists.

Commercial ApplicationsThere are many commercial applications of forensic linguistics, but the most commonapplications are in disputes over trademarks and copyright. Very often a company will requireexpert testimony as to the extent to which a rival company may have infringed copyright ortrademark legislation in naming or promoting a new product.

Georgina Heydon

8

A case study is provided below, but names and geographic references have been changedto protect the commercial interests of the clients.

Company A was introducing a new Scotch Whiskey onto the market, and hadchosen the name Hoch Ay: a purely made-up name that was perhaps intended toreflect the Scottish origins of the drink. However, another company, CompanyB, already had a brand of Scotch Whiskey called Hawk Eye. Company B wassuing Company A for copyright infringement on the basis that if the name of theproduct made by Company A (Hoch Ay) were to be pronounced with a Scottishaccent, it would sound identical to the name of the product made by Company B(Hawk Eye). In a crowded bar or nightclub, where the drink would be sold,Company B claimed that this would cause unacceptable brand confusion.A forensic linguistic analysis was required to show whether a) the brand nameHoch Ay would indeed sound like Hawk Eye if it were pronounced with aScottish accent and b) the target market of young Australian adults would adopt aScottish accent to pronounce the brand name Hoch Ay.The analysis cannot be discussed in any detail because the actual case did notinvolve a Scottish accent, or even an English language accent. However it maybe useful to make some general observations about the case, because it involvesvariational sociolinguistics, rather than phonetics, syntax or semantics.Part a) was a straightforward analysis of the made up words of the new brandname according to the rules of a foreign pronunciation. Part b) was a moredifficult sociolinguistic analysis, because it involved research into the socio-economic status of the target market, and then a calculation of the likelihood ofthis group using a foreign accent to pronounce a made up, but potentially foreign,brand name.At the time of writing this case has not gone to trial, but it may yet come beforethe courts at which time the results can be published in full.

Lie DetectionVarious forms of linguistic analysis are employed by those attempting to establish the veracityof written, and sometimes spoken, statements. Often the analysis includes consideration ofpersonal pronouns, tense, vocabulary items and sentence structure, but does not usually takeaccount of socio-cultural and regional variation in language use. This type of analysis is onlyloosely classed as forensic linguistics as its validity is often the focus of disagreement amongprofessionals and academics working in the area.

It is important to emphasize that commercial methods of lie detection involving theanalysis of written texts, such as Scientific Content Analysis and Criteria Based ContentAnalysis, have not been verified by laboratory testing. A great deal of psychological researchhas been conducted into verbal cues to deception (see Vrij 2008 for a comprehensiveintroduction), but far less attention has been paid to the linguistic methods purportedly used infor example Scientific Content Analysis (but see Heydon 2008). Identifying reliable ofdeception in written texts is therefore a critical area of research for scholars wishing to make avaluable contribution to forensic linguistics.

SUMMARY OF KEY ISSUES

A major concern for members of the international forensic linguistics community is thereliability of the evidence obtained through forensic linguistic analysis. There continues to bemuch controversy surrounding almost every aspect of forensic linguistics described in thispaper, primarily due to questions of methodology and expertise. In many instances, the debate


9

focuses on whether the analytic technique is statistically valid, though in some cases, such as liedetection and language of origin analysis, the main concern is that the methods are not based onsound sociolinguistic and/or language acquisition theory in the first place, but are useduncritically by government departments and law enforcement agencies. In such cases, manylinguists have considered it their professional duty to inform and educate such agencies aboutthe pitfalls of using unreliable analytic techniques.

The controversy over the statistical reliability of analytic methods, especially in the areaof voice or authorship identification, is generating intense debate within the forensic linguisticcommunity. There is a clear division between those who reject all analysis that cannot bestatistically validated, and those who argue that there is a place for qualitative analysis, such asdialectology and sociolinguistic analysis, in expert testimony.

In cases relating to the analysis of discourse structure, and especially where the disputecentres on the interpretation of a legal text (see the case study above) it is common for thelinguistic evidence to be disregarded on the basis that such interpretation is a matter for legalexperts, not linguists. In the case presented above concerning the semantic interpretation of theverb standing, the judge ruled that the matter was a point of law and therefore not the provinceof a linguist, irrespective of their expertise.

The question of whether linguistic expertise will be recognized by courts, or whetherlinguistic evidence will be allowed by judges or magistrates, is further complicated by thereliance on precedence as a method for selecting evidence for inclusion. As discussed earlier,judges and magistrates are wary of allowing new kinds of evidence or expertise into their courtsand thus setting a legal precedent. As members of an international community, forensic linguistshave a role to play in furthering the reach of our expertise. The publication of the InternationalJournal of Speech, Language and the Law is part of this process of improving our profile andchances of recognition as experts. However there is more that can be done by linguisticassociations, such as KIMLI, simply by raising awareness of linguistics as a science. All toooften, legal practitioners report that they would never think to consult a linguist about matters oflanguage, because they believe that all their experience in writing legal cases and appearing incourt makes them experts in language by default. They, like many members of the generalpublic, are blissfully unaware of the discipline of linguistics as the scientific study of languageas a system. In the bankruptcy case described above, the judge simply refused to accept thatthere was such a field as ‘linguistics’, which makes you wonder what we all do for a living!

NOTE

* I would like to thank an anonymous reviewer for very helpful comments on the earlier draft.

FURTHER READING

The International Journal of Speech Language and the Law is published by Equinox. ISSN(print) 1748-8885, ISSN (online) 1748-8893

See also Coulthard 1997; 2000; Danet 1980; Gibbons 1994; Hall and Collins 1980;Hammarström 1987; Jensen 1995; Kaplan, Green, Cunningham et al. 1995; Shuy 1993;Solan 1998.

REFERENCES

Chaski, C. 2001. Empirical evaluations of language-based author identification techniques.Forensic Linguistics: The International Journal of Speech, Language and the Law 8 (1):1-65

Georgina Heydon

10

Coulthard, Malcolm. 1997. A failed appeal. Forensic Linguistics: The International Journal ofSpeech, Language and the Law 4: 287-302.

Coulthard, Malcolm. 2000. Whose text is it? On the linguistic investigation of authorship. InDiscourse and Social Life, ed. by Malcolm Coulthard and Srikant Sarangi. Harlow:Longman. Pp. 270-87.

Danet, Brenda. 1980. Language in the Legal Process. Law and Society Review 15: 445-565.

Gibbons, John (ed.). 1994. Language and the Law. London: Longman.

Hall, M. C. and A. M. Collins. 1980. The admission of spectographic evidence: A note of Reg vGilmore. The Australian Law Journal 54: 21-24.

Hammarström, Göran. 1987. Voice identification. The Australian Journal of Forensic Sciences19 (3): 95-99.

Heydon, Georgina. 2008. The art of deception: myths about lie detection in written confessions.In L. Smets and Aldert Vrij (Eds) Cahiers Policestudies: Special InvestigativeInterviewing techniques; The use of written - and oral analyses. Brussels, Politeia pp173-186

Hollien, Harry F. 2002. Forensic Voice Identification. San Diego and London: Academic Press.

Jensen, Marie-Therese. 1995. Linguistic evidence accepted in the case of a non-native speakerof English. In Language in Evidence: Issues Confronting Aboriginal and MulticulturalAustralia, ed. by Diana Eades. Sydney: University of NSW Press. Pp. 127- 46.

Kaplan, J.P., G.M. Green, C.D. Cunningham and J.N. Levi. 1995. Bringing linguistics intojudicial decision-making: semantic analysis submitted to the Supreme Court. ForensicLinguistics: The International Journal of Speech, Language and the Law 2: 81-98.

Kreidler, Charles W. 1998. Introducing English Semantics. London: Routledge.

Shuy, Roger W. 1993. Language Crimes: The Use and Abuse of Language in the Courtroom.Oxford: Blackwell.

Solan, Lawrence M. 1998. Linguistic experts as semantic tour guides. Forensic Linguistics: TheInternational Journal of Speech, Language and the Law 5 (2): 87-106.

Svartvik, Jan. 1968. The Evans Statements: A Case for Forensic Linguistics. GothenburgStudies in English 20. Göteborg: Göteborgs Universitet.

Vrij, Aldert. 2008. Detecting Lies and Deceit: Pitfalls and Opportunities. Second Edition.Chicester: John Wiley and Sons Ltd.


A DISTRIBUTED MORPHOLOGY ANALYSISOF INDONESIAN KE-/-AN VERBSi

Lanny Hidajat*Atma Jaya Catholic [email protected]

AbstractIndonesian ke-/-an verbs have a complex argument structure. Similarly to Indonesianpassive di- verbs, ke-/-an verbs never have an agentive NP in the subject position andtheir subject NPs must be definite. However, unlike passive di- verbs, these verbsgenerally cannot be followed by an agentive prepositional phrase. In addition, when ke-/-an verbs have two arguments, the applied argument appears in the subject positioninstead of the internal one. In this study, the structure of Indonesian ke-/-an verbs isanalyzed by using the Distributed Morphology framework (Folli dan Harley, 2002;Kratzer, 1996; Marantz, 1997; among others). Based on the verbs’ distribution andinterpretation, this study argues that of ke-/-an verbs are derived by attaching the ke-/-ancircumfix, which is an overt representation of a verbalizing head, to the projection ofROOT.

Keywords: ke-/-an circumfix, verbalizing head

AbstrakVerba berimbuhan ke-/-an mempunyai struktur kalimat yang kompleks. Seperti verbapasif berawalan di-, subjek dari verba berimbuhan ke-/-an tidak pernah pelakutindakan dan bersifat takrif. Tetapi, tidak seperti verba pasif berawalan di-, verbaberimbuhan ke-/-an tidak bisa diikuti oleh pelaku tindakan di posisi objek. Selain itu,pada saat verba berimbuhan ke-/-an memiliki dua nomina, maka verba ini akandidahului objek tidak langsung dan diikuti oleh objek langsung, bukan sebaliknya.Dalam artikel ini, struktur verba berimbuhan ke-/-an akan dianalisis dengan kerangkateori Distributed Morphology (Folli dan Harley, 2002; Kratzer 1996; Marantz 1997;dan lain-lain). Menurut kerangka teori ini, verba berimbuhan ke-/-an dibentuk dariproses afiksasi antara imbuhan pembentuk verba ke-/-an dengan akar kata (ROOT)yang belum mempunyai kategori. Argumen ini diajukan sesuai dengan distribusi danmakna Verba berimbuhan ke-/-an.

Kata kunci: sirkumfiks ke-/-an, pembentuk verba

THE IDIOSYNCRASIES OF INDONESIAN KE-/-AN VERBSLittle has been said in the literature about Indonesian ke-/-an verbs, presumably because thevoice of ke-/-an verbs is less productive than the active voice, in which the verbs are optionallymarked by the prefix meN- or N-, and the passive voice, in which the verbs are obligatorilymarked by the prefix di-. In fact, the voice of ke-/-an verbs is interesting and poses problemswhich need to be solved because of its complex nature, as already noted in the previous studies(Dardjowidjojo 1966 and Sneddon 1996, 2000). The fact that this voice has an interestingargument structure is exemplified in the following two sentences, in which the theme role isassigned to an NP in a different position:

(1) Gudang itu kebakaranWarehouse that KE-burn-AN‘The warehouse was on fire.’

Lanny Hidajat

12

(2) Joni kejatuhan mangga.Joni KE-fall-AN mango‘Joni was fallen on by a mango.’

In (1), the theme role is assigned to the subject NP. On the other hand, in (2), theme is assignedto the object NP, while the subject NP is goal. Although the position of the theme NP in (1) isdifferent from the one in (2), the form of the ke-/-an verbs in both sentences in the same. This isin contrast to the verbs in active voice which undergo a change in form when the theme NP is ina different position, as exemplified in (3):

(3) a. Gambarnya udah nempel.Picture-DET already N-stick‘The picture is already stuck.’

b. Paman nempelin gambar itu.Uncle N-stick-IN picture that‘Uncle stuck the picture (to something).’

In (3a), the theme role is assigned to the subject DP; while, in (3b), it is assigned to the objectNP. Correspondingly, unlike the N- verb in (3a), the one in (3b) is also affixed by the suffix –in.

The idiosyncratic behavior of ke-/-an verbs is also indicated by the number of the NParguments that they can take. Some ke-/-an verbs, such as: kebakaran ‘to be on fire’, keguguran‘to miscarry’, and kebongkaran ‘to get broken into (referring to a house)’, can only have one NPargument. Other ke-/-an verbs, such as: kejatuhan ‘to be fallen on by’, kerobohan ‘to becollapsed on by’, and ketumpahan ‘to be spilled by’, must have two NP arguments. There arealso ke-/-an verbs that can have either one or two NP arguments. This class of ke-/-an verbs canbe categorized into three sub-classes, as follows:

i. Ke-/-an verbs with an optional NP complement: As already noted by Dardjowidjojo (1978),the NP complements of some ke-/-an verbs can be optional; especially when these verbsappear in discourse. Included in this sub-class are: kecopetan ‘to have (something) stolen bya pickpocket’, kecurian ‘to have (something) stolen’, and kebagian ‘to get a share’. The ke-/-an verbs of this sub-class may occur in a sentence without an object NP; however, nativespeakers always interpret them as if they occur with their object NP, as in (4). This factindicates that the object NPs of the ke-/-an verbs of this sub-class are actually a verbcomplement.

(4) Joni kecopetan (dompet).Joni KE-pick.pocket-AN wallet‘Joni’s wallet was stolen by a pickpocket.’

ii. Ke-/-an verbs verbs with an optional NP complement with agentive flavor: The optionalobject NPs of this sub-class of ke-/-an verbs are different from the ones of the previous sub-class because they have an agentive flavor, as exemplified in (5):

(5) Aduh! Bukunya Joni kedudukan (ama) (Bobi) nih!EXCL book.DET Joni KE-sit-AN by Bobi this‘Oh, no! Joni’s book was sat on (by Bobi)!’

The optional object NPs of this sub-class of ke-/-an verbs appear to be agentive because ofthe following reasons: i) the object NP is optional and can be animate, and ii) when theobject NP is animate, it is optionally introduced by the preposition ama ‘by’. The agentiveflavor the object NPs of this sub-class of ke-/-an verbs may mislead one to assume that theyare analogous to the agentive adjunct in passive voice marked by the prefix di-. Thisassumption implies that the argument structure of this sub-class of ke-/-an verbs is similar tothat of passive di- verbs. However, the optional object NP of this sub-class of ke-/-an verbsis actually not an agent because the ke-/-an verbs cannot be modified by subject-oriented


13

manner adverbials or instrumental phrases (see endnote v). This consequently means thatthis sub-class of ke-/-an verbs is not another type of passive voice. Included in this sub-classof are: kedudukan ‘to be sat on (by someone)’ and ketutupan ‘to be blocked (by someone)’,and kemasukan ‘to be broken into (by someone).’

iii. Ke-/-an verbs verbs with an optional NP complement with agentive flavor complement andVP complement: The following ke-/-an verbs: kelihatan ‘to be visible’ or ‘to be seen (doingsomething)’, kedengaran ‘to be audible’ or ‘to be heard’, ketahuan ‘to be found out (doingsomething)’, and kedapatan ‘to be found out/detected (doing something)’, form a sub-classbecause they are optionally followed by either an object NP or a VP or both, as representedin the following sentence. The optional object NPs of this sub-class of ke-/-an verbs alsoappear to be agentive adjunct.

(6) Dia/Joni kelihatan (ama) (Wati) (lagi marahin Sita).3sg/Joni KE-see-AN by Wati PROGRESS angry-IN Sita‘He/she/Joni was seen (by Wati) scolding Sita.’

In this study, the argument structure of ke-/-an verbs is explained under the DistributedMorphology framework (DM) (Marantz 1997, 2001; Harley and Noyer 1999, among others).Further explanation of DM and the analysis of the argument structure of ke-/-an verbs arediscussed below. Beforehand, the following sections discusses issues related to how theargument structure of ke-/-an verbs is analyzed, which are: the interpretation of ke-/-an verbs,the distribution of ke-/-an verbs in comparison to the distribution of active meN-/N-/- verbs andpassive di- verbs, and the eventivity of ke-/-an verbs.

ANALYZING THE ADVERSATIVE INTERPRETATION OF KE-/-AN VERBSA serious attempt to analyze the structure of ke-/-an verbs was made by Dardjowidjojo (1978).According to Dardjowidjojo (1978:117), ke-/-an verbs have the following semantic features: (i)the event or condition is unexpected, unpredicted, or unavoidable, and (ii) the effect isadversative. He classifies ke-/-an verbs based on syntactic-semantic criteria and then argues thatthe various derivations of ke-/-an verbs are the result of affixing the adversative feature, whichis in the form of the prefix ke- and the suffix –an, to the roots, which can be verbal, adjective, ornominal (Dardjowidjojo 1978: 117). Dardjowidjojo’s analysis of the structure of ke-/-an verbsimplies that all ke-/-an verbs have adversative interpretation. This means that the event describedby a ke-/-an verb always negatively affects its argument, in particular the subject NP. Forinstance, in (7), the event ketiduran ‘to oversleep’ negatively affects the subject NP Jonibecause it causes Joni to come late to school.

(7) Joni ketiduran jadi telat ke sekolah.Joni KE-sleep-AN so.that late to school‘Joni overslept so that he was late to school.’

Yet, not all ke-/-an verbs are adversative. For instance, the event described by the ke-/-an verb kebagian ‘to get a share of’ does not negatively affects the subject NP Joni.

(8) Joni kebagian mangga.Joni KE-share-AN mango‘Joni got a share of a mango.’

Similarly, the subject NP suaranya ‘his sound’ in (9) is not negatively affected by the ke-/-anverb kedengaran ‘to be audible’

(9) Suaranya kedengaran dari sini.Sound-DET KE-listen-AN from here‘His voice can be heard from here.’

Lanny Hidajat

14

The fact that not all ke-/-an verbs are adversative reflects that the adversativeinterpretation of ke-/-an verbs is not the result of affixing the circumfix ke-/-an to the root. Inthis paper, I argue that it actually originates in the speakers’ real-world knowledge. Thisargument is based on the fact that whether or not a ke-/-an verb is adversative depends on thecontext in which it appears, as reflected in the interpretation of kebakaran ‘to be on fire’ in thefollowing sentences:

(10) Bangunan tua itu kebakaranstructure-AN old that KE-burn-AN‘The old building was on fire.’

In (10), it is not indicated whether or not bangunan tua ‘old building’ is owned by someone;therefore, kebakaran ‘to be on fire’ is not adversative. However, as noted by Sneddon(1996:124), in (11), kebakaran has an indirect adversative interpretation because the possessorof the entity which is on fire is indicated.

(11) Rumahnya kebakaranHouse-3sg KE-burn-AN‘His/her house was on fire.’

Sneddon argues that the adversative interpretation in (11) is indirect because kebakaranadversely affects the possessor of the entity which is on fire (i.e. -nya), instead of the entityitself (i.e. rumah ‘house’). As shown in (12), kebakaran will still have an indirect adversativeinterpretation even if the possessor of the entity which is on fire is only given in the discourse.

(12) Amat kemarin beli rumah. Sekarang rumah itu kebakaran.Amat yesterday -buy house now house that KE-burn-AN‘Amat bought a house yesterday. Now, the house is on fire.’

In (12), kebakaran still adversely affects Amat, although Amat does not appear in the sentencecontaining kebakaran. This is because in the earlier sentence it is stated that Amat was theowner of the house which was on fire. Interestingly, in (13), kebakaran loses its indirectadversative interpretation although it appears with the possessor of the entity which is on fire:

(13) Jonii sangat senang waktu rumahnyai kebakaran, karena diai akan dapat uang asuransi.‘Jonii was very happy when hisi house was on fire as hei would get some money fromthe insurance.’

-nya in rumahnya ‘his house’ refers to Joni. This means that Joni should be adversely affectedby kebakaran. However, because of the context in which it appears, kebakaran even has abenefactive interpretation in (13), instead of adversely affecting Joni. The fact that kebakarandoes not always have adversative interpretation, as reflected in sentences (10) to (13), showsthat it is discourse and real-world knowledge that create the adversative interpretation of ke-anverbs. If adversative interpretation is the property of ke-/-an verbs, then kebakaran always haveadversative interpretation, even when it appears in a context such as in (13).

Similarly to kebakaran ‘to be on fire’, ketiduran ‘to oversleep’ can be eitheradversative, as in (7), or not adversative, as in (14), depending on the context in which itappears. The adversative interpretation in (7) emerges because, according to real-worldknowledge, being late to school is not a good thing. On the other hand, in (14), ketidurantriggers a benefactive interpretation because real-world knowledge tells us that escaping from anaccident is a good thing.

(14) Pagi ini pesawat Garuda mengalami kecelakaan. Untungnya, Boby ketiduran sehinggaia batal naik pesawat itu.‘Garuda airplane had an accident this morning. Fortunately, Boby overslept so that hefailed to board that airplane.’


15

To reiterate, not all ke-/-an verbs have an adversative interpretation. In addition,contexts can cancel the adversative interpretation of some ke-/-an verbs which appear to beadversative. These two facts contradict Dardjowidjojo’s proposal that the ke-/-an circumfix is anadversative feature; hence, raises the question of the function of this circumfix. In addition,there is still no clear explanation for the idiosyncratic argument structure of ke-/-an verbs. Theanswers to these two questions are discussed after discussing other issues related to how theargument structure of ke-/-an verbs is analyzed.

THE INTERPRETATION OF KE-/-AN VERBS WITH TWO NP ARGUMENTSAs revealed in the previous section, ke-/-an verbs do not always have an adversativeinterpretation. In fact, the interpretation of a ke-/-an verbs depends on the context in which itoccurs. In this section, I argue that ke-/-an verbs—in particular those with two arguments—havea directional interpretation, instead of an adversative interpretation. Directional interpretation isan interpretation in which one argument of the ke-/-an verbs with two arguments is eithermoving closer to or going away from the other argument.

The ke-/-an verb kejatuhan ‘to be fallen on by’, as in (15), is typically assumed to beadversative because it means that the fallen mango hit Joni. However, if kejatuhan is reallyadversative, then (15) will also be true in the situation in which Joni was negatively affectedbecause the fall of the mango causes him not to be able to eat the mango. The fact that (15) isonly relevant in a situation in which a mango fell on Joni shows that kejatuhan is directional,instead of adversative. I assume that kejatuhan is directional because the mango will end up onJoni’s body, which means that the mango is moving toward Joni.

(15) Joni kejatuhan (ama) mangga.Joni KE-fall-AN by/with mangoDirectional: ‘Joni was fallen on by a mango.’Adversative: *‘Joni was negatively affected by mango falling because he could not

eat it.’

Other ke-/-an verbs with two NP arguments that have go toward interpretation are kebagian ‘toget a share of (something)’ and kedudukan ‘to be sat on by(someone)’, among others.

Similarly to kejatuhan, kecopetan ‘to have (something) stolen by a pickpocket’ in (16)also appears to be adversative. However, kecopetan is actually not adversative because (16) isnot true in the situation in which Joni was negatively affected because other person’s wallet gotstolen. (16) is actually directional because it is true only in a situation in which Joni’s wallet gotstolen when it was with him, and not in a situation in which the wallet was stolen in Joni’sbedroom when Joni was away. The directional interpretation of (16) is reflected in theinterpretation in which Joni’s wallet is moving away from him.

(16) Joni kecopetan dompet.Joni KE-pick.pocket-AN walletDirectional: ‘Joni’s wallet was stolen while it was with him.’Adversative: *‘Joni was negatively affected by the stealing of other person’s wallet.’

Other ke-/-an verbs with two NP arguments that have go away from interpretation arekehilangan ‘to lose (something)’ and ketinggalan ‘to accidentally leave (something)’, amongothers.

To summarize, ke-/-an verbs with two NP arguments actually have a directionalinterpretation, instead of adversative. The fact that native speakers generally interpret ke-/-anverbs as adversative is triggered by discourse and real-world knowledge.

Lanny Hidajat

16

THE DISTRIBUTION OF KE-/-AN VERBSThis section discusses the distribution of ke-/-an verbs in contrast to the distribution of activemeN-/N-/- verbs and that of passive di- verbs. As will be revealed in the discussion below, ke-/-an verbs are different from active meN-/N-/- verbs and passive di- verbs because they arenon-agentive.

Ke-/-an verbs versus meN-/N-/- verbsAs exemplified in the following sentences, there are two characteristics of ke-/-an verbs thatsignificantly distinguish them from meN-/N-/- verbs. Firstly, the subject NP of ke-/-an verbs isnever agent or causer. Secondly, the subject of ke-/-an verbs with two arguments is either a goalor a source.

(17) Rumah itu kebakaran.Home that KE-burn-AN‘The house was on fire.’*‘The house burned (something).’

(18) Joni melompat lalu lari.Joni MEN-jump past run‘Joni jumped and then ran.’

(19) Bobi kejatuhan mangga.Bobi KE-fall-AN mango‘Bobi was fallen on by a mango.’*‘Bobi made the mango fall.’

(20) Joni jatuhin mangga itu (ke atas Bobi).Joni -fall-IN mango that to above Bobi‘Joni dropped the mango (to Bobi).’*‘(Bobi made) the mango fall on Joni.’

In (17), the NP argument in the subject position of the ke-/-an verb kebakaran ‘on fire’ cannotbe interpreted as agent. This is in contrast with the NP argument in the subject position ofmelompat ‘to jump’ in (18). In (20), the subject NP of jatuhin ‘to drop’ must be interpreted asagent and cannot be interpreted as goal or source. On the other hand, the subject NP of the ke-/-an verb kejatuhan ‘to be fallen on by’ in (19) is goal and cannot be interpreted as agent. Theabove two distinctions lead to the conclusion that ke-/-an verbs are not in the active voice andthey are not derived in the same way as meN-/N-/- verbs.

Ke-/-an verbs versus di- verbsAt a glance, ke-/-an verbs and di- verbs share some characteristics. Firstly, as shown in (21) and(22), similarly to passive di- verbs, the subject NP of ke-/-an verbs can be theme:

(21) Gudang itu kebakaran.Warehouse that KE-burn-AN‘The house was on fire.’

(22) Gudang itu dibakar.Warehouse that DI-burn‘The house was burnt.’

Secondly, the subject NP of both ke-/-an verbs and passive di- verbs can be goal, as shown in(23) and (24):



17

(24) Joni dibagiin mangga.Joni DI-share-IN mangoLit. ‘To Joni was shared the mango.’

Despite the above similarities, there are several characteristics of ke-/-an verbs whichdistinguish them from di- verbs. As shown in (25) and (26), unlike passive di- verbs, ke-/-anverbs cannot be followed by an optional agentive prepositional phrase.

(25) Gudang itu kebakaran (*ama Joni).Warehouse that KE-burn-AN by Joni‘The house was on fire (*by Joni).’

(26) Joni kebagian mangga *(ama Bobi).Joni KE-share-AN mango by Bobi‘Joni got a share of a mango (*by Bobi).’

The second difference between ke-/-an verbs and di- verbs is that the subject NP of two-argument ke-/-an verbs cannot be theme, as shown in (27). In contrast, the subject NP ofditransitive passive di- verbs can be theme, as shown in (28).

(27) *Mangga itu kebagian ke Joni.Mango that KE-share-AN to JoniLit. ‘The mango was shared (to Joni).’

(28) Mangga itu dibagiin ke Joni.Mango that DI-share-IN to Joni‘The mango was distributed to Joni.’

The above two differences reflect that ke-/-an verbs are not derived in the same way asdi- verbs. The fact that ke-/-an verbs cannot co-occur with an optional agentive prepositionalphrase, as discussed above, suggests that they are non-agentive. In contrast, passive di- verbshave an implicit subject, which means they are actually agentive (cf. Baker, Johnson, andRobert’s, 1989, argument for English passives.) The non-agentivity of ke-/-an verbs is alsoevidenced by the fact that they cannot be modified by subject-oriented manner adverbials,except tidak/gak sengaja ‘accidentally’,ii or instrumental phrases, as shown in (29) and (30),respectively (see Baker, Johnson, and Roberts 1989; Dubinsky and Simango 1996, for furtherexplanation regarding the agentivity tests.)

(29) *Joni buru-buru kebagian mangga.Joni RED.hurry KE-share-AN mango*‘Joni got a share of a mango in a hurry.’

(30) *Gudang tua itu kebakaran pake bensin.warehouse old that KE-burn-AN use gasoline*The old warehouse was on fire by using gasoline.’

In comparison, passive di- verbs can be modified by subject-oriented manner adverbials andinstrumental phrases:

(31) Joni buru-buru dibagiin mangga.Joni RED.hurry DI-share-IN mangoLit. ‘Joni was distributed the mango in a hurry.’

(32) Gudang tua itu dibakar pake bensin.warehouse old that DI-burn use gasoline‘The old warehouse was burnt by using gasoline.’

To summarize, the differences in the distribution between ke-/-an verbs and meN-/N-/-verbs and also between ke-/-an verbs and di- verbs reveal that ke-/-an verbs are not either in theactive voice or passive voice. Unlike meN-/N-/- verbs and di- verbs, ke-/-an verbs are non-

Lanny Hidajat

18

agentive. This fact implies that ke-/-an verbs are derived differently from meN-/N-/-verbs anddi- verbs.

THE EVENTIVITY OF KE-/-AN VERBSThe fact that ke-/-an verbs do not bear an implicit agent might mislead us to assume that theseverbs are stative, because non-agentivity is generally used as a diagnostic for stativity (Dowty1979, Dubinsky and Simango 1996, Katz 2003, among others). However, the results of other“stativity” tests, which were used in Katz’s (2003; drawing on Lakoff 1966) study, show thatke-/-an verbs are actually eventive despite their non-agentivity. According to Katz (2003: 206,drawing on Dowty 1979; Sag 1973; Hinrichs 1985), “state predicates are always non-agentive,temporally homogeneous, and have a present orientation.” On the other hand, eventivepredicates are agentive, can ‘move’ narrative time in discourse, and lack past orientation.

One feature of ke-/-an verbs which shows that they are non-stative is the fact that theyhave a past orientation. According to Katz (based on work by von Stechow 1995, Ogihara 1996,and Abusch 1997), one way to check whether a predicate has past or present orientation is byinserting it into a complement clause of verbs such as believe or think. The complement clauseof the matrix verb believe or think must have a present orientation with respect the matrix verbitself. Since stative verbs, such as love or know, have a present orientation, they can be theinfinitival complements of the matrix verb believe or think, as shown in (33). On the other hand,eventive verbs, such as kiss, have a past orientation; therefore, it is unnatural for eventive verbsto be the infinitival complements of believe or think, as shown in (34)

(33) Thelma believed Hans to love Lin.(34) ??Thelma believed Hans to kiss Lin.

(Katz 2003:209)

In Indonesian, when stative verbs, such as tau ‘know’, appear as an embedded verb ofyakin, which corresponds to English believe, they have a present orientation, as shown in (35):

(35) Gue yakin Joni tau jawabannya.1sg certain Joni know answer-AN.DET‘I’m sure Joni knows the answer.’*‘I’m sure Joni knew the answer.’

In contrast, when the complement clauses of yakin ‘certain’ contain eventive verbs, such asmukul ‘to hit’, they exhibit a past orientation with respect to the matrix verb, as shown in (36):

(36) Gue yakin Joni mukul temennya.1sg certain Joni N-hit friend-3sg‘I’m sure Joni hit his friend.’*‘I’m sure Joni hits his friend.’

When ke-/-an verbs are inserted into the complement clause of yakin ‘certain’, they have a pastorientation, as shown in (37):

(37) Gue yakin Joni kejatuhan mangga.1sg certain Joni KE-fall-AN mango‘I’m sure a mango fell on Joni.’*‘I’m sure a mango falls on Joni.’

The fact that ke-/-an verbs have a past orientation is one indication that they are eventive verbs.The second stativity test is that stative verbs cannot be modified by an in-adverbial

phrase, such as in an hour (Katz 2003), as shown below:

(38) ??He was away from home in an hour (Katz, 2003:10)


19

As shown in (39) and (40), respectively, similarly to active eventive verbs, such as bagi ‘toshare something to someone’, ke-/-an verbs, such as kebagian ‘to get shared something’, arecompatible with the adverbial phrase dalam waktu semenit, which corresponds to in a minute.

(39) Bobi bagiin makanannya ke Joni dalam waktu semenit.Bobi -share-IN eat-AN-3sg to Joni in time one minute‘Bobi shared his food to Joni in one minute.’

(40) Joni kebagian makanan dalam waktu semenit.Joni KE-share-AN eat-AN in time one minute‘Joni got shared food in one minute.’

In fact, Indonesian stative verbs, such as tau ‘know’, can also be modified by dalam waktusemenit ‘in a minute’. However, modifying stative verbs with an in-adverbial phrase causesthem to have an inchoative interpretation, instead of the stative interpretation, as shown in (41).

(41) Joni tahu jawabannya dalam waktu semenit.Joni know answer-AN-DET in time one minute*‘Joni is in the state of knowing the answer in one minute.’ (Stative)‘Joni came to know the answer in one minute.’ (Inchoative)

The fact that ke-/-an verbs do not become inchoative when modified by dalam waktu semenitshows that they are not stative.

To reiterate, although ke-/-an verbs are non-agentive, they have the following twocharacteristics of eventive verbs. Firstly, when they are inserted into the complement clauses ofyakin ‘certain’, they have past orientation. Secondly, ke-/-an verbs do not become inchoativeswhen modified by in a minute-type of adverbs.

THE SUFFIX –KAN VS. THE CIRCUMFIX KE-/-ANTo illustrate the effect of the circumfix ke-/-an on thematic role assignment and the argumentstructure of the based verbs to which it is affixed, I will first discuss the effect of the suffix –kanas reviewed by Cole and Son (2004). According to Cole and Son, the suffix –kan functions assyntactic licenser which serves to syntactically license an argument, which is thematicallylicensed by the verb, in the argument structure. Without the presence of a syntacticlicenser, a thematically licensed argument is unlicensed to be in an argument structure.As a syntactic licenser, the suffix –kan can increase the valence of the base verbs. As shown in(42), attaching the suffix –kan to an intransitive and adjectival base verb, such as jatuh ‘to fall’,adds an additional NP argument.

(41) Mangganya jatuh.Mango-DET fall‘The mango fell.’

(42) Bobi menjatuhkan mangga (ke atas Joni).Bobi MEN-fall-KAN mango to top Joni‘Bobi dropped the mango (on Joni).’ (i.e. ‘Bobi caused the mango to drop (on Joni).’)

In (41), the unaccusative verb jatuh ‘to fall’ already assigns the theme role to its argumentmangganya ‘the manggo’. The affixation of –kan adds the number of the NP argument of theverb. In accordance to the hierarchies of thematic relations and argument structures suggestedby Grimshaw (1990), the additional NP argument is assigned a causer theta role and becomesthe surface subject.

As a syntactic licenser, the suffix –kan also adds the valence of transitive base verbs, asreflected in (44). In (44), the suffix –kan syntactically licenses the object NP saya ‘3sg’ in theargument structure of the transitive base verb pikir ‘think’. Although the verb pikir has assigned

Lanny Hidajat

20

the theme role to the NP saya, the NP can only occur in the argument structure after the verb isaffixed by the suffix –kan. In this case, the suffix –kan becomes an object marker.

(43) *Dia tidak memikir saya.3sg not meN-think 1sg‘She does not think about me.’

(44) Dia tidak memikirkan saya.3sg not meN-think-KAN 1sg‘She does not think about me.’

(Cole and Son, 2004:351)

When transitive base verbs that syntactically license both of their NP arguments in the argumentstructure are affixed to the suffix –kan, the number of their NP argument increases. As shown in(45), the transitive base verb panggang ‘bake’ assigns the agent role to the NP saya ‘1SG’ andthe theme role to the NP bread ‘roti’ and also syntactically licenses them. However, it does notlicense and also assign a thematic role to the NP Eric. The NP Eric is an adjunct; therefore, it isoptional. However, when the verb panggang ‘bake’ is affixed by –kan, Eric becomes the NPargument of panggang ‘bake’ and receives the beneficiary role; therefore, it is no longeroptional, as shown in (46). In this case, the suffix –kan is a benefactive suffix.

(45) Saya memanggang roti (untuk Eric).1sg meN-bake bread for Eric‘I baked bread for Eric.’

(46) Saya memanggangkan roti *(untuk Eric).1sg meN-bake-KAN bread for Eric‘I baked bread for Eric.’

(Cole and Son, 2004:343)

As reflected in the above discussion, affixes can change the argument structure and thethematic role assignment of the base verbs to which they are affixed, which subsequentlychanges the valence of the verbs. What is the effect of the circumfix ke-/-an to the valence of theverbs to which it is affixed? As shown in (48), in some cases, ke-/-an can increase the valence ofintransitive base verbs by one NP argument:

(47) Joni jatuh (di tangga)Joni fall LOC stairs‘Joni fell on the stairs.’

(48) Joni kejatuhan mangga.Joni KE-fall-AN mango‘Joni was fallen on by a mango.’

However, attaching the circumfix ke-/-an to intransitive base verbs does not always add anadditional NP argument. For example, it does not increase the valence of the intransitive baseverb tidur:

(49) Joni tidur (di kamarnya).Joni -sleep LOC room-DET‘Joni is sleeping (in his room).’

(50) Joni ketiduran (di kamarnya).Joni KE-sleep-AN LOC room-DET‘Joni overslept (in his room.)’

The presence of the circumfix ke-/-an can even decrease the valence of the verbs towhich it is affixed. For example, when the ditransitive base verb bagi ‘to share’ and thetransitive base verb bakar ‘to burn’ are affixed by ke-/-an, they have one less NP argument, asexemplified in (52) and (54), respectively.


21

(51) Joni udah bagi mangga ke Bobi.Joni already -divide mango to Bobi‘Joni has already shared the mango to Bobi.’

(52) Bobi udah kebagian mangga (*oleh Joni).Bobi already KE-divide-AN mango by Joni‘Bobi has already got a share of a mango (*by Joni).’

(53) Joni bakar gudang itu.Joni -burn warehouse that‘Joni burn the warehouse.’

(54) Gudang itu kebakaran (*ama Joni).warehouse that KE-burn-AN by Joni‘The warehouse was on fire *(by Joni).’

Similarly to the suffix –kan, the affixation of ke-/-an can also affects the thematicrelations of the base verbs. As shown in (48), when the unaccusative verb ‘jatuh’ is affixed byke-/-an, the theme theta role is assigned to the NP argument in object, instead of the one insubject. The NP argument in subject received the goal theta role. Furthermore, as illustrated in(52) and (54), transitive and ditransitive base verbs cannot assign the agent theta role when theyare affixed by ke-/-an; therefore, a ke-/-an verb cannot have an agent NP as its argument.However, the circumfix ke-/-an does not always affect the thematic relations of intransitiveverbs. As exemplified in (50), the unaccusative verb ‘tidur’ still have a theme subject when it isaffixed by ke-/-an.

To reiterate, the circumfix ke-/-an affects the argument structure of the verbs to which itis affixed by either increasing or decreasing their valence. Another notable effect of thecircumfix ke-/-an is that it blocks verbs from assigning an agent theta role.

THE STRUCTURE OF KE-/-AN VERBS

In this paper, the structures of ke-/-an verbs are explained in terms of the DistributedMorphology framework (DM) (Marantz 1997, 2001; Harley and Noyer 1999; among others).This framework is an extension of the idea that some semantic aspects of a word are alsorepresented in the syntax (Hale and Keyser, 1993). According to Folli and Harley (2002:5), inthis structural approach, it is assumed that the interpretation of a verb is determined by “thefunctional/aspectual structure in which the verb is inserted” and “the syntactic positions inwhich its arguments are realized.” Another assumption of DM is that words are inserted into thesyntactic operations as category neutral components (ROOT) (Marantz 1997). ROOTs are latercategorized in accordance to their syntactic positions, or as Marantz (1997:215, based onChomsky, 1970) puts it:

When the roots are placed in a nominal environment, the result is a“nominalization”; when the roots are placed in a verbal environment, theybecome verbs.

In other words, the derivation of verbs involves inserting ROOT into the syntax. If the ROOT isgoverned by a verbalizing v head, it becomes a verb.

The Structure of One-Argument ke-/-an VerbsAccording to DM, the affixation of ke-/-an to base words is an operation in the syntax. DM alsoposits that ke-/-an verbs start as category neutral components in the syntactic operation. Thisassumption its with the fact that these verbs seem to be derived from various lexical categories,as noted by Dardjowidjojo (1978) and Sneddon (2000). In line with the notion of DM thatfunctional heads play an important role in the derivation of verbs, I argue that ke-/-an is theovert representation of a functional head. The questions which will be answered in the

Lanny Hidajat

22

discussion in this section are (i) what functional head the circumfix ke-/-an represents and (ii)where its position is in the syntax.

First, I would like to address the non-agentivity of ke-/-an verbs. In recent studies, it isgenerally assumed that the functional head which generates the external argument is differentfrom the one which generates the internal argument (Kratzer 1996, Marantz 1997, Folli andHarley 2002, among others). Kratzer (1996, based on Marantz 1984) proposes that the externalargument is introduced in the specifier position of the voice projection, which is immediatelyabove VP, in which the internal argument is generated. Following Kratzer’s proposal, I arguethat the non-agentivity of ke-/-an verbs indicates the lack of the functional head which generatesthe external argument in the structure of ke-/-an verbs. This implies that there is no voiceprojection in the structure of ke-/-an verbs, which means that the position of the functional headrepresented by the circumfix ke-/-an is below the voice head.

In this paper, I propose that the circumfix ke-/-an is an overt form of the v head thatdirectly attaches to ROOT and verbalizes it. My proposal is based on several pieces of evidence;the first being the fact that ke-/-an verbs are not productive, as pointed out by Sneddon (1996).This condition corresponds to Marantz’s proposal (2001), stating that a head which attachesimmediately to the ROOT is semi-productive because its selectional requirements must besatisfied by the idiosyncratic properties of the ROOT.

Marantz (2001) also predicts that a head which is immediately attached to the ROOTcan create an idiom. This prediction is based on the assumption that roots in combination withother elements within the locality domain may have a special meaning (Marantz 1997). Marantz(1997:8) also assumes that the functional head that projects agents is the boundary for thedomain of special meanings. The fact that there exists an idiom with a ke-/-an verbs, as shownin (55), supports the assumption that the circumfix ke-/-an is a verbalizing v head.

(55) Tingkahnya seperti orang/kambing kebakaran jenggot.act-3sg look.like person/goat KE-burn-AN beard‘He acts frantically.’ (Lit. ‘He acts like a person/a goat whose beard is on fire.’)

The fact that the subject NP of kebakaran jenggot in (55) can be either orang ‘person’ orkambing ‘goat’ without changing the meaning of the idiom shows that the subject NP is not partof the idiom. On the other hand, as shown in (56), the idiom orang kebakaran jenggot ‘to actfrantically’ loses its idiomatic interpretation when the base verb bakar ‘to burn’ appears with theprefix di-, which is a passive marker. This shows that the verb kebakaran is part of the idiom.

(56) Tingkahnya seperti orang/kambing dibakar jenggotact-3sg look.like person/goat DI-burn beard‘He acts as if someone who is being burnt by his beard.’

Additional support for the proposal that the circumfix ke-/-an is a verbalizing v headwhich directly attaches to ROOT comes from the fact that ke-/-an verbs are not stative, asalready discussed earlier. According to the structure of the verb phrase proposed by Marantz(2001), as shown in Figure 1, the stative head is nearer to the ROOT as compared to theverbalizing head. However, the eventivity of ke-/-an verbs suggests the absence of a stativehead. Consequently, in the structure of ke-/-an verbs, it is the verbalizing v head which realizesthe circumfix ke-/-an in its overt form, which immediately attaches to the ROOT.


23

Figure 1. The Structure of the Verb Phrase (Marantz 2001:5)

To reiterate, the fact that ke-/-an verbs are non-agentive postulates that there is no voiceprojection in the structure of ke-/-an verbs. Furthermore, several evidence indicate that thecircumfix ke-/-an is an overt presentation of a verbalizing v head which directly attaches toROOT. In line with these two assumptions, I propose that the structure of one-argument ke-/-anverbs, such as in rumah itu kebakaran ‘the house was on fire’, is as shown in Figure 2:

Figure 2. The Structure of One-Argument ke-/-an verbs

As shown in Figure2, the theme NP, which on the surface appears in the subjectposition, originates as the internal argument of ROOT. The assumption that ROOT can take asingle argument is suggested by Embick (2004) to account for the structures of transitive verbsas well as unaccusative verbs. According to this assumption, the internal argument of transitiveverbs and unaccusative verbs originates as the argument of ROOTs and it is interpreted as thelogical object of ROOT. Following Chomsky’s (2000, 2001) Agree operation, I assume that thecomplement of ROOT in one-argument ke-an verbs moves to the spec of TP to satisfy the EPPfeature, which is borne by T.

I assume that the subject NP of the single argument ke-/-an verbs is actually thecomplement of ROOT because it is always assigned theme, which is relevant to its status as thelogical object of the ROOT. This assumption conforms to Chomsky (1981:36)’s ThetaCriterion, which is “each argument bears one and only one theta-role, and each theta-role isassigned to one and only one argument.” The Theta Criterion postulates that each argument isassigned one thematic role at the level of D-structure. Subsequently, a thematic relation remainsthrough a derivation, even though the argument may undergo movement or raising.

A piece of evidence that supports the assumption that the subject NP of one-argumentke-/-an verbs is base-generated as the complement of ROOT and move to the spec of TP is the

Lanny Hidajat

24

fact that they must be definite. The two sentences in (57) are ungrammatical because the subjectNPs of the ke-/-an verbs are indefinite.

(57) a. *Gudang kebakaranWarehouse KE-burn-AN‘A warehouse was on fire.’

b *Anak kejatuhan mangga.child KE-fall-AN mango‘A boy was fallen on by a mango.’

The behavior of the subject NP of ke-/-an verbs is similar to that of the subject NP of passive di-verbs. As mentioned by Sneddon (1996:254), the subject NP of passive di- verbs must be definite.In line with Guilfoyle, Hung, and Travis’s (1992) analysis of Indonesian passive construction, Iassume that the restriction on the definiteness of passive subjects relates to A-movement, which isthe movement of the theme NP from the complement of V to the spec of IP in order to get case.Correspondingly, the requirement for the subject NPs of ke-/-an verbs to be definite indicates thatthey are base-generated in a position below VP and moves to the spec of IP.

The assumption that the subject NP of ke-/-an verbs undergoes A-movement to the specof IP also gets support from the fact that a sentence can have a ke-/-an verb in conjunction witha di- verb with only one subject, as illustrated in (58):

(58) Jonii kejatuhan ti tangga dan ti dikejar anjing.Joni KE-fall-AN leader and PASS-chase dog‘Joni was fallen on by a leader and chased by a do.’

The fact that Joni can be the subject for both kejatuhan ‘to be fallen on by’ and dikejar ‘to bechased’ indicates that the subject NP of this sentence originates inside VP (McCloskey 1997,drawing on McNally 1992 and Burton and Grimshaw 1992).

The Structure of Two-Argument ke-/-an verbsAccording to the structure represented in Figure 2, ke-/-an verbs cannot have more than oneargument. The fact that these verbs can actually have two arguments indicates the presence ofan applicative construction in their structure. However, where is the position of the applicativeconstruction in the argument structure of ke-/-an verbs?

As discussed earlier, I assume that the two-argument ke-/-an verbs have a directionalinterpretation because one of the two NP arguments either moves toward or away from the otherNP argument, as already reflected in (15) and (16). The directional interpretation of two-argument ke-/-an verbs indicates the presence of a low applicative construction in the structureof the two-argument ke-/-an verbs. This notion is based on Pylkkänen’s (2002, based onBresnan and Moshi 1990) proposal of the two types of applicative constructions: highapplicatives and low applicatives. The high applicative construction attaches above the verb, asshown in Figure 3; while the low applicative attaches below the verb, as shown in Figure 4.


25

Figure 3. The Structure of High Applicative Construction (Pylkkänen 2002:19)

Figure 4. The Structure of Low Applicative Construction (Pylkkänen 2002:19)

According to Pylkkänen, the difference in the syntactic positions of the high and lowapplicative head is reflected in two different interpretations. High applicatives denote a relationbetween an event and an individual, which is the applied argument, as exemplified in (59),which is an Albanian sentence with a high applicative construction. In (59), the event Agimholds my bag is related to the applied argument Drita, so that the interpretation of (59) is Agimholds my bag for the benefit of Drita.

(59) Agimi i mban Drites anten timeA.NOM DAT.CL holds Drita.DAT bag.ACC my‘Agim holds my bag for Drita.’

(Pylkkänen, 2002:25)

Unlike the high applicative construction, low applicatives denote a relation between twoindividuals, i.e. the applied argument and the internal argument. As exemplified in (60), theapplied argument John is only related to the internal argument the book in the sense that thebook ends up being possessed by John

(60) Mary bought John the book. (Pylkkänen 2002:23)

Since low applicatives denote a relation between the applied argument and the internalargument, Pylkkänen (2002) argues that low applicatives cannot appear in a structure that lacksan internal argument, such as unergative verbs.

Another piece of evidence supporting the argument that the applicative construction oftwo argument ke-/-an verbs is a low applicative comes from the fact that only the appliedargument, which is the goal/source NP, can be in the subject position of two argument ke-/-anverbs. As illustrated in (61), the internal argument, which is the theme NP, cannot be the subjectof ke-/-an verbs.

Lanny Hidajat

26

(61) *Mangga kejatuhan Joni.Mango KE-fall-AN Joni‘Joni was fallen on by a mango.’

The aforementioned evidence corresponds to McGinnis’s (2001:112-113) claim that, in astructure with a high applicative construction, when the verb is passivized, either the internalargument or the applied argument can be a subject. On the other hand, in a structure with a lowapplicative construction, only the applied argument can be a subject. This claim is based on heranalysis of the distribution of Kinyarwanda benefactives, which bear a high applicativeconstruction, and the locatives, which bear a low applicative construction. According to her,when a Kinyarwanda benefactive verb is passivized, either the theme NP or the beneficiary NPcan be in the subject. In contrast, when a Kinyarwanda locative verb is passivized, only thelocative NP can occupy the subject position (see McGinnis 2001 for the discussion of thedifference between the high and low applicative constructions with respect to the NP subject.)

To summarize, I argue that the applicative construction in the structure of two-argumentke-/-an verbs is a low applicative. My argument is based on the following two facts: firstly, thetwo-argument ke-/-an verbs have a directional interpretation; and, secondly, only the appliedargument can be the subject NP of two argument ke-/-an verbs. In line with the aboveargument, the structure of two-argument ke-/-an verbs, such as kejatuhan ‘to be fallen on by’, isas represented in Figure 5.

Figure 5. The Structure of Two-Argument ke-/-an verbs

As shown in Figure 5, ROOT in the two-argument ke-/-an verbs, unlike ROOT in the one-argument ke-/-an verbs, selects the low applicative construction as a complement. This impliesthat selection of the low applicative construction is an inherent property of ROOT. Under theassumption that the internal argument is the argument of ROOT, the low applicativeconstruction has to be part of ROOT, so that the internal argument and the applied argument aregenerated at the same level. If the low applicative construction combines with and the internalargument combines with ROOT, the applied argument and the internal argument will begenerated at different levels; consequently, there will not be a directional interpretation betweenthe applied argument and the internal argument.

I assume that the low applicative head of ke-/-an verbs bears [+to] and [+from] features(Legate 2001; Pylkkänen 2002). Pylkkänen (2002) argues that the low recipient and sourceapplicatives have the following lexical entries:


27

(62) Low-APPL-TO (Recipient applicative): x.y.f<e<s,t>>.e. f(e,x) & theme (e,x) & to-the-possession (x,y)

Low-APPL-FROM (Source applicative):x.y.f<e<s,t>>.e. f(e,x) & theme (e,x) & from-the-possession (x,y)

(Pylkkänen, 2002:22)

In line with the above argument, I argue that the [+to] and [+from] features borne by thelow applicative head represent the direction of the internal argument with respect to the appliedarguments, which are as follows: the [+to] feature means that the internal argument is headingtoward the applied argument and the [+from] feature means that the internal argument is movingaway from the applied argument. I assume that it is the semantic property of the ROOT thatcombines with the projection of the low applicative that determines which feature will be borneby the low applicative head in a particular sentence. The low applicative head bears the [+to]feature when it appears in sentences such as (63) and (64).

(63) Joni kejatuhan (ama) mangga.Joni KE-fall-AN by/with mango‘Joni was fallen on by a mango.’


In (63) and (64), the internal argument mangga ‘mango’ is moving toward the internal argumentJoni. On the other hand, in (65), the low applicative head bears the [+from] feature because theinternal argument dompet ‘wallet’ is moving away from Joni.

(65) Joni kecopetan dompet.Joni KE-steal-AN wallet‘Joni’s wallet was stolen.’

As mentioned earlier, according to Pylkkänen (2002), the low applicative constructioncannot co-occur with unergative verbs. Therefore, the argument that applicative construction ofthe two-argument ke-/-an verbs is low applicative is somewhat problematic because there aretwo argument ke-/-an verbs which seem to be derived from an unergative verb; for instance,kedudukan ‘to be sat on by’, as represented in (66):

(66) Aduh! Bukunya Joni kedudukan (ama) (Bobi) nih!EXCL book.DET Joni KE-sit-AN by/with Bobi this‘Oh no! Look, Joni’s book was accidentally sat on (by Bobi)!’

This problem actually can be accounted for by assuming that kedudukan is derived from duduk‘sit’ as the verb of spatial configuration with the simple position meaning (see Levin andRappaport Hovav 1995 for the discussion of the possible interpretations of the verb of spatialconfiguration). According to Levin and Rappaport Hovav (1995), the spatial configuration verbswith the simple position meaning are unaccusative. This means that kedudukan is actuallyderived from duduk ‘sit’ as an accusative verb, instead of unergative. Accordingly, the structureof kedudukan can contain a low applicative construction. The argument that kedudukan isderived from duduk with the simple position meaning is supported by the following three facts:

i. Goal DP, which functions as the location phrase, is mandatory in the subject position, asshown below:

(67) *Aduh! kedudukan (ama) (Bobi) nih!EXCL KE-sit-AN by/with Bobi this*‘Oh no! Look, was accidentally sat on (by Bobi)!’

Lanny Hidajat

28

ii. Subject DPs of kedudukan cannot be an agent. Therefore, (68) cannot be interpreted asBruno (the dog) was accidentally sat on by Joni.

iii. Kedudukan cannot be modified by a subject-oriented manner adverbials, such as terburu-buru ‘in a rush’, which shows that it has a non-agentive interpretation.

To summarize, this section discusses the structure of ke-/-an verbs. Ke-an verbs areeventive, non-agentive, and can take either one or two NP arguments. To account for the factthat ke-/-an verbs are non-agentive, I argue that the structure of ke-/-an verbs lacks voiceprojection. In addition, the circumfix ke-/-an is an overt form of a verbalizing v head thatdirectly attaches to ROOT and verbalizes it. The fact that there are ke-/-an verbs that take twoNP arguments indicate that it structures can contain an applicative construction. I argue that theapplicative construction in the structure of two-argument ke-/-an verbs is the low applicative forthe following two reasons: first, two-argument ke-/-an verbs have a directional interpretation. Inaddition, I assume that the low applicative head bears [+to] and [+from] features, whichrepresent the direction of the internal argument, that is, whether it is moving toward or goingaway from the applied argument. I also assume that it is the semantic properties of the ROOTcombining with the low applicative projection that determine which feature is borne by the lowapplicative head in a particular sentence.

A REMARK ON THE ACCIDENTAL INTERPRETATION OF KE-/-AN VERBSKe-/-an verbs also trigger an interpretation in which the events described by the verbs happenaccidentally. In line with the proposed structures of ke-/-an verbs, which are represented inFigure 2 and Figure 5, I assume that the accidental interpretation of ke-/-an verbs is due to thesyntactic structures of ke-/-an verbs, instead of being semantically implicated in the circumfixke-/-an. Ke-/-an verbs yield accidental interpretation because none of their arguments aregenerated above the ke-/-an v projection.

This assumption corresponds to Hale and Keyser’s (1993) view that the most salientmeaning of the inner VP is “change.” According to Hale and Keyser, a projection of a lexicalcategory is associated with a structural relation of c-command and complementation and anelementary semantic relation. Following the aforementioned argument, Hale and Keyser (1993)argue that the structural relation of a causative sentence, such as the cook thinned the gravy, isas shown in Figure 6:

Figure 6. The Structural Relation of a Causative Sentence (Hale and Keyser 1993:72)

Each of the two v heads projected by thin contributes a subpart of the verb’s meaning. Theupper v head represents a CAUSE event and the lower v head represents a BECOME event.Accordingly, the NP in the specifier of the upper v heads is regarded as the subject of the causalevent and the one in the specifier of the lower v head is the subject of a ‘predicate of change’.


29

Since the ke-/-an v head is a verbalizing head and non-agentive, I assume that the ke-/-an v head corresponds to the lower v head in Figure 6, which means that it represents aBECOME event. Correspondingly, the NP argument taken by ROOT (i.e. the internal argument)in the structures of ke-/-an verbs resembles Hale and Keyser’s subject of a predicate change, i.e.it corresponds to an entity undergoing change. The accidental interpretation subsequentlyemerges because the internal argument is affected by ke-/-an verbs, instead of causing the event.

CONCLUSIONIt has been shown that the idiosyncratic distribution of ke-/-an verbs can be accounted for byusing the Distributed Morphology framework. The fact that ke-/-an verbs are non-agentiveindicates that the structures of ke-/-an verbs lack the voice head which licenses the agent NP.Although ke-an verbs are non-agentive, they are nevertheless eventive. In the present study, Ihave argued that the circumfix ke-/-an is an overt form of a verbalizing v head which directlyattaches to ROOT. This argument can explain why the presence of the circumfix ke-/-an doesnot necessarily add a new argument to the argument structure. I attributed the fact that some ke-an verbs take only one argument while others take two arguments to the inherent properties ofthe ROOT taken by the ke-/-an v head. ROOTs that take one NP argument yield one-argumentke-an verbs; while the ones that select a low applicative construction as its complement bringforth two-argument ke-an verbs.

The postulation that the circumfix ke-/-an is a verbalizing v head that attaches directlyto ROOT can explain why ke-/-an verbs are not productive in Indonesian. According to Marantz(2001), a functional head which attaches immediately to ROOT is less productive because itsselectional restriction must be satisfied by the idiosyncratic properties of ROOT. This viewimplies that ke-/-an verbs are not productive because there are limited ROOTs that can satisfythe selectional restrictions of the ke-/-an verbalizing v head. This assumption raises a newquestion, namely, what are the inherent properties of ROOT which are required by the ke-/-an vhead. I leave this question open for now.

The above assumption also suggests that there is more than one verbalizing head. Thisbrings up many questions, such as how many types of verbalizing v head there are, what makesone verbalizing v head different from other v heads, whether all languages have more than onetype of verbalizing v head, and so forth. Further studies are required in order to answer thesequestions. The present study is only hints that the properties of a verbalizing v head can be acomponent which distinguishes one language from another.

I would also like to point out that the proposed structures of ke-/-an verbs in this studycannot account for the ke-/-an nouns, such as kesenangan ‘happiness’, kesedihan ‘sadness’,and kebersamaan ‘togetherness’. Unlike ke-/-an verbs, ke-/-an nouns are very productive inIndonesian. This fact suggests that the circumfix ke-/-an can also be an overt form of afunctional head for nominalization. This notion implies that Indonesian has twohomophonous functional heads with the circumfix ke-/-an as their overt form.

NOTE* I would like to thank an anonymous reviewer for very helpful comments on the earlier draft.** Parts of the earlier version of this paper were presented at ISMIL 10, 21-23 April 2006, in theUniversity of Delaware, Newark, DE, and AFLA XIV, 4-6 May 2007, in the McGill University,Montreal. I am indebted to Julie Anne Legate who supervised me in the process of writing this paper. Ialso would like to thank Gabriella Hermon, Peter Cole, and Benjamin Bruening for the invaluablecomments and the inspiring discussion on the topic of this talk. My gratitude also goes to my friends inthe Linguistics Dept. of the University of Delaware: Yanti, Laura Spinu, Tim McKinnon, Sean Madigan,Masahiro Yamada, and Sachie Kotani for their insightful comments and suggestions to improve thequality of this paper. I am also grateful to my friends in UNIKA Atma Jaya, Jakarta, dan the Jakarta FieldStation for helping me with the language data. All errors in facts and analyses are my own responsibility.

Lanny Hidajat

30

i The Indonesian analyzed in this paper is the colloquial register of Indonesian as spoken in Jakarta.All examples are the kind of utterances used in daily conversation by educated speakers.

ii I assume that ke-/-an verbs can co-occur with tidak/gak sengaja because ke-/-an verbs induce aninterpretation of events which happen unexpectedly.

REFERENCES

Baker, Mark, Kyle Johnson, and Ian Roberts. 1989. Passive Arguments Raised. LinguisticInquiry 20: 219–251.

Chomsky, Noam. 1981. Lectures on Government and Binding. Dordrecht: Foris.

Chomsky, Noam. 2000. Minimalist inquiries: The framework. In Roger Martin, DavidMichaels, and Juan Uriagereka (eds.) Step by step: Essays on Minimalist Syntax inHonor of Howard Lasnik, (pp. 89-155). Cambridge, MA.: MIT Press.

Chomsky, Noam. 2001. Derivation by Phase. In Michael Kenstowicz (ed.) Ken Hale: A Life inLanguage, (pp. 1-52). Cambridge, MA: MIT Press.

Cole, Peter. and Min Jeong Son. 2004. The argument structure of verbs with the suffix –kan inIndonesian. Oceanic Linguistics, Volume 43.2:339-364.

Dardjowidjojo, Soenjono. 1978. The semantic structures of the adversative ke-…-an verbs inIndonesian. In S. Udin (ed.) Spectrum: essays presented to Sutan Takdir Alisjahbana onhis seventieth birthday (pp. 107-124). Dian Rakyat, Jakarta.

Dowty, David R. 1979. Word Meaning and Montague Grammar. Dordrecht: Reidel.

Dubinsky, Stanley, and Silvester Ron Simango. 1996. Passive and Stative in Chichewa:Evidence for Modular Distinctions in Grammar. Language 72:749-781.

Embick, David. 2004. On the Structure of Resultative Participles in English. Linguistic Inquiry35.3: 355-392.

Folli, Raffaella and Heidi Harley. 2002. Consuming Results in Italian & English: Flavors of v.In P. Kempchinsky and R. Slabakova (eds.) Aspect. (pp. 1-26). Dordrecht: KluwerAcademic Publishers, (Manuscript is downloadable at http://dingo.sbs.arizona.edu/~hharley/PDFs/ FolliHarley2002Final.pdf)

Grimshaw, Jane. 1990. Argument Structure. Cambridge, MA, MIT Press.

Guilfoyle, Eithne, Henrietta Hung, and Lisa Travis. 1992. Spec of IP and Spec of VP: TwoSubjects in Austronesian Languages. Natural Language and Linguistics Theory 10:375-414.

Hale, Kenneth and Samuel Jay Keyser. 1993. On Argument Structure and the LexicalRepresentation of Semantic Relations. In S.J. Keyser & K. Hale (eds.) The view fromBuilding (pp.53-109). Cambridge, MA, MIT press.

Harley, Heidi and Rolf Noyer. 1999. Distributed Morphology. Glot International, Volume 4,Issue 4, April 1999: 3 – 9.

Katz, Graham. 2003. On the stativity of the English perfect. In A. Alexiadou, M. Rathert and A.von Stechow (eds.) Perfect Explorations (pp. 205-234). Mouton de Gruyter, Berlin.

Kratzer, Angelika. 1996. Severing the External Argument from Its Verb. In Johan Rooryck &Laurie Zaring, (eds.). Phrase Structure and lexicon, (pp. 109-137). Dordrecht: Kluwer.


31

Legate, Julie Anne. 2001. The Configurational Structure of a Nonconfigurational Language. InJ. Rooryck and P.Pica (eds.) Linguistic Variations Yearbook 1:61-104.

Levin, Beth, and Malka Rappaport Hovav. 1995. Unaccusativity: At the syntax-lexicalsemantics interface. Cambridge, Mass.: MIT Press.

Marantz, Alec P. 1997. No escape from syntax: Don’t try morphological analysis in the privacyof your own lexicon. In A. Dimitriadis, & L. Siegel, (Eds.), University of PennsylvaniaWorking Papers in Linguistics, 4.2 (pp. 201-225). Philadelphia: University ofPennsylvania Working Papers in Linguistics.

Marantz, Alec P. 2001. Words and Things. Handout, MIT, Cambridge, Mass. (http://dingo.sbs.arizona.edu/~hharley/Oxford/Marantz2.pdf)

McCloskey. 1997. Subjecthood and Subject Positions. In L. Haegeman (ed.) Elements ofGrammar, (pp. 197-235). Dordrecht: Kluwer.

McGinnis, Martha. 2001. Variation in the Phase Structure of Applicatives. In J. Rooryck andP.Pica (eds.) Linguistic Variations Yearbook 1: 105-146.

Pylkkänen, Liina. 2002. Introducing Arguments. Doctoral dissertation, MIT, Cambridge, Mass.

Sneddon, James Neil. 1996. Indonesian: A Comprehensive Grammar. London and New York:Routledge.

Sneddon, James Neil. 2000. Another Look at ke- … -an Verbs. In Bambang Kaswanti Purwo,Ed., Kajian serba linguistik: untuk Anton Moeliono, pereksa bahasa (Transl:Languages, linguistics, and civilization: festschrift in honor of Anton Moeliono,Indonesian expert in linguistics). Jakarta: Universitas Katolik Indonesia Atma Jaya,BPK Gunung Mulia.


MEASURING PROFICIENCY IN STANDARD INDONESIANFOR ENGGANO SPEAKERS

Rindu Parulian Simanjuntak*SIL International, Masyarakat Linguistik Indonesia

[email protected]

Abstract

The main purpose of study described in this paper was to measure the proficiency ofStandard Indonesian spoken in Enggano Island, Bengkulu Province, Indonesia.Enggano is spoken in six villages: Malakoni, Apoho, Meok, Ka’ana, Kayapu andBanjar. Of these, two villages — Meok and Ka’ana — were selected as research sitesfor collecting data. The instruments used were the Indonesian Sentence Repetition Test(ISRT) (Hanawalt 2008) and a set of bilingualisms questionnaire. The results show anaverage bilingual proficiency in Standard Indonesian of level 2 on the ILR scale(Interagency Language Roundtable 2007) for Enggano speakers, which indicates anability to use Indonesian limited to particular situations and domains. The analysis alsoshows that bilingual proficiency in Enggano varies on average between males andfemales and is influenced by education and age, but not by the other factorsinvestigated. This means that bilingual proficiency in Standard Indonesian for Engganospeakers is largely the result of continuing on to higher levels of education. Eventhough the average proficiency of Enggano speakers in Standard Indonesian is at level2, the attitudes of most Enggano speakers towards both the Indonesian language andthe local vernacular language are strongly positive.

Keywords: bilingual proficiency

AbstrakPenelitian ini bertujan untuk mengukur tingkat kemahiran berbahasa Indonesia standaryang dituturkan di Pulau Enggano, Propinsi Bengkulu, Indonesia. Bahasa Engganodituturkan di enam desa: Malakoni, Apoho, Meok, Ka’ana, Kayapu dan Banjar. Duadesa — Meok dan Ka’ana — dipilih sebagai tempat untuk mengambil data-data.Instrumen yang digunakan adalah Tes Pengulangan Kalimat (TPK) dalam bahasaIndonesia atau Indonesian Sentence Repetition Test (ISRT) (Hanawalt 2008) danseperangkat kuesioner kedwibahasaan. Hasil penelitian mengungkapkan, rata-ratakemahiran berbahasa Indonesia standar lisan masyarakat Enggano di level 2 padatingkat ILR (Interagency Language Roundatable 2007), mengindikasikan keterbatasankemampuan berbahasa Indonesia dalam situasi dan ranah tertentu. Hasil analisis jugamenunjukan rata-rata kemahiran kedwibahasaan antara pria dan wanita berbedakarena dipengaruhi oleh pendidikan dan usia, tetapi tidak dengan faktor-faktor lain.Artinya, hasil kemahiran kedwibahasaan bahasa Indonesia standar lisan pada penuturEnggano dapat digunakan untuk melanjutkan ke jenjang pendidikan yang lebih tinggi,dan sikap penutur Enggano pada bahasa Indonesia dan bahasa Enggano sangat positif.

Kata kunci: kemahiran berbahasa Indonesia

INTRODUCTION

In the Indonesian context, most people use Indonesian as a medium for communication informal situations, while vernacular languages are used in informal situations and isolated areas.The country currently faces the challenge of how to maintain both Indonesian and vernacularlanguages. A large number of Indonesians speak both a vernacular language (L1) andIndonesian (L2), and many also speak an international language. Thus, most people in Indonesiaare bilingual or multilingual. Limited research has been conducted to quantitatively measure the

Rindu Parulian Simanjuntak

34

Standard Indonesian (hereafter BI) proficiency level of second-language speakers of Indonesian.Even though in many parts in Indonesia it is reported that people are bilingual in, little has beendone to investigate the actual level of Indonesian proficiency throughout the nation. This meansthat more bilingualism research is necessary, especially that conducted with research tools thatcan quantitatively measure the proficiency level of BI in a certain speech community.

The Indonesia archipelago consists of more than 17,000 islands from the island ofSumatera to Papua. Among these is Enggano, an isolated island located to the south west ofSumatra Island. Administratively, Enggano has status as a sub-district of the regency of NorthBengkulu in Bengkulu Province. The population of Enggano was reported in census data at1927 persons (Bengkulu Utara, 2000). The six villages of Enggano, located from southeast tonorthwest, are Kayapu, Ka’ana, Malakoni, Apoho, Meok, and Banjarsari.

Figure 1. Map of South Sumatra and Bengkulu Province


35

According to Ethonologue (Gordon 2005:435), Enggano [eno], or Engganese (itsalternate name), is classified as Enggano, Sumatera, Malayo-Polynesian, Austronesian. It isspoken on Enggano Island and on four smaller nearby islands. No other languages are reportedto be closely related to it. Gumono reported in his paper, entitled Upaya Pemeliharaan BahasaDaerah yang terancam punah: Kondisi Kebahasaan, Sikap Bahasa, dan kebutuhan bahasamasyarakat Pulau Enggano, that Indonesian and Enggano both are spoken by Enggano speakersover 40-years-old. Gumono states that Enggano people prefer to speak Indonesian to Engganodue to negative attitude towards their vernacular language. Furthermore, they feel inferior toother cultures brought from outside of the island. Despite this, Enggano people want to keepboth Indonesian and vernacular language. Indonesian is seen as useful for increasing theirincome in the future and vernacular is seen as necessary to be maintained in order that theculture of Enggano could be passed on to the younger generation.

METHODOLOGY

A Sentence Repetition Test (SRT) is a type of bilingualism test for large-scale assessment ofsecond-language (L2) proficiency. It is used to gain an objective and general assessment of aperson’s proficiency in a second language, not an in-depth analysis of one person’s strengths orweaknesses in that L2 (Radloff 1991:8). The Indonesian Sentence Repetition Test (ISRT) scorehas been calibrated against the OPI (Oral Proficiency Interview), which is considered as one ofthe most reliable tests in measuring the bilingual proficiency (see Hanawalt 2008). The OPIwas first used by the United States Foreign Service Institute (FSI) to evaluate an individual’s L2proficiency (Radloff 1991: 47). In the OPI, an individual’s proficiency in L2 is described with arange of ILR (Interagency Language Roundtable) levels from 0 (No Performance) to 5 (MasterProfessional Performance). The correlation between ISRT score and ILR level as measured bythe OPI is shown in Table 1 below.

Table 1. Correlation Between ISRT Score and ILR Level (Hanawalt 2008:18)ISRT Total Score OPI (ILR) Level0-4 05-7 0.6 (0+)8-12 113-15 1.6 (1+)16-20 221-24 2.6 (2+)25-29 330-32 3.6 (3+)33-37 438-40 4.6 (4+)41-54 5

The Indonesian SRT consists of 15 sentences in Indonesian. Each sentence has anassigned point value. The total possible score on the ISRT is 54. The respondent being testedlistens to the sentences one by one and repeats each sentence immediately after hearing it(Radloff 1991:7) and is scored on his or her ability to repeat the sentence exactly. Beforestarting the test, the respondent is allowed to practice with 3 example sentences. If the score ofthe respondent is low up to the 5th sentence, the interviewer does not need to continue the test.SRT tests have some advantages for assessment of community L2, such as short administrationand scoring time with only a few personnel required, the fact that it can be administered even byresearchers who only have rudimentary proficiency of the language being tested (provided thathe/she has received proper training, and the fact that testing procedures are easy to understand.


36

The test also follows the natural tendency of people to repeat what they hear, and it caneliminate cultural and social constraints (for example, the subject need not to be isolated fromthe group). Once developed and calibrated, an SRT can be used without further modificationwherever bilingualism in that particular language is a question (Radloff 1991:6-7).

Survey ProceduresTwo research tools were used to gather data to fulfill the research goal: a Bilingualismquestionnaire, and the ISRT (Indonesian Sentence Repetition Test). The questionnaire containeda series of personal demographic questions for the respondent (see Appendix 1). Verbal consentwas obtained from the respondent before the questionnaire was administered. Explanations ofthe questionnaire and the SRT usually took place in Indonesian. However, if the respondent wasnot sufficiently fluent in Indonesian to understand or reply in Indonesian, a local guidetranslated the explanation into the Enggano language.

SamplingRandom sampling was applied in each level of the sampling procedures to ensure the validityand representativeness of the collected data. At the village level, stratified random sampling wasused by classifying the villages into two categories.Village SamplingThe villages that were selected as research sites in Enggano are the ones where, according to theinformation gathered prior to the research in 2008 and confirmed afterwards, at least 75% of theresidents who were native Enggano speakers. Based on this criterion, two out of six villages ofEnggano were selected. They are Meok and Ka’ana. In our larger study of bilingualism inBengkulu province, representative research site villages were randomly selected from twogroups: villages located 10km or less to the nearest high school and villages located 20km ormore to the nearest high school. This selection was made to see if the distance to the nearestschool influences bilingual proficiency. However, there were no high schools in Meok andKa’ana, so both villages selected as research sites where it located more than 20 km from thenearest high school.Household SamplingA systematic sampling method was used. Households were chosen by totaling the number ofhouses in the village of Meok and Ka’ana, and dividing that by the number of respondentsneeded. The total respondents were tested was 62, which 27 in Ka’ana, and 35 in Meok. Forexample, if there were 150 houses in a village, 150 were divided by 30 and so every 5th housewas chosen. If the residents of a household were not originally from Enggano island, the nexthouse was chosen.Respondent SamplingA respondent between the ages of 15 to 60 was selected in each household using the KISH grid(see Appendix 3). If the respondent chosen was not originally from the Enggano languagegroup (e.g. Javanese or Sundanese who married into the family), this person was not accepted asa respondent, and the next person indicated by the KISH grid was tested. A person was deemedan acceptable respondent if he/she met the following requirements: at least one parent was anative speaker of the local language, the local language was the respondent’s first language, therespondent was physically and mentally able to take the test, and the respondent was willing totake the test.


37

RESULTS

Average BI ProficiencyThe overall average OPI or ILR levels (see Appendix 2) for BI bilingual proficiency of theEnggano is on level 2 (Limited Working Performance). According to the InternationalLanguage Roundtable (2007) a person with language proficiency of level 2 is able to converseabout routine or predictable subject matter, but sometimes require repetition or clarification.

Table 2. Percentage of Respondents with BI Proficiency at ILR Level 3Language Percentage of respondents at or above ILR3 Percentage of respondents below ILR3Enggano 21% 79%

The above table shows the percentage of Enggano respondents who are at or above ILR 3 in BI.Less than one-third of the population in all clusters is at or above ILR 3 (ProfessionalPerformance). The Enggano language group is therefore not predicted to be adequatelyproficient to use language materials exclusively in BI as defined in Sanders (2004): “80% ormore of a cross-section of the language community (ages 20-45) speak a second language (L2)at a proficiency level of 3+ or above and this second language is used in a sustainable manner.”Thus, vernacular language materials may have benefit for the Enggano community.

Social and Demographic FactorsBased on information about the respondents gathered using the sociolinguistics questionnaire,ten factors were investigated as potential correlates with BI proficiency: years of education, age,exposure to literature, length of time spent outside the Enggano community, gender, distance ofthe respondent's village from school, frequency of travel, exposure to radio, contact withoutsiders, and exposure to television. All the data were entered into electronic format, and thenanalyzed using the EPI software package.

For most of the independent variables mentioned above, respondents were askedquestions about the frequency that they engage in related activities (watching television ortraveling outside of Enggano island, for example). They were then assigned by the analysts tofrequency groups in the categories “never”, “seldom”, “sometimes”, and “often”. There werefour exceptions. For years of education, respondents were asked how many years they hadspent in school and these numerical values were used. For age, we used the respondents’reported age. For gender there were two categories, male and female. For distance from school,we categorized respondents into two groups: those who live 10km or less to the nearest highschool and those who live 20km or more to the nearest of high school. As mentioned above, allof the villages on Enggano are more than 20 km away from the nearest high school. As a laststep, before statistical analysis was conducted, non-numerical data was converted to numericalcodes to facilitate the analysis.

The EPI software package was used to conduct multiple variable regression analysis tofind the correlation between the dependent variable (ISRT score) and independent variables(years of education, age, exposure to literature, length of time spent outside the community,gender, distance from school, travel frequency, exposure to radio, contact with outsiders, andexposure to television) and also to determine which of the factors investigated had an influenceon BI proficiency.

As shown in Table 3 below, the factors investigated can be grouped into two categories.If the p-value resulting from the multiple variable regression analysis is less than 0.05, the factoris considered to have a statistically significant relationship with ISRT score, and if the p-value isgreater than 0.05 the relationship between this factor and ISRT score is considered statisticallyinsignificant.


38

Table 3.Results of Multiple Variable Regression Analysis for Relationship with BI Proficiency

Variable Coefficient Std Error F-test P-ValueYears of Education 1.266 0.136 86.6471 0.000000Age -0.086 0.036 5.6390 0.018142Exposure to Literature 0.683 0.411 2.7689 0.097068Contact with 0.389 0.287 1.8315 0.176882Gender 0.503 0.828 0.3698 0.543534Distance From School -0.035 0.093 0.1409 0.707676Travel Frequency -0.101 0.465 0.0469 0.828677Exposure to Radio 0.047 0.329 0.0204 0.886526Contact with Outsiders 0.014 0.119 0.0138 0.906444Exposure to Television 0.039 0.489 0.0062 0.937037

Statistically Significant FactorsOf factors investigated, there are only two statistically significant factors which where shown toinfluence respondents' ISRT results: years of education and age.

Years of EducationThe multiple regression analysis showed a positive correlation between years of education andbilingual proficiency in Indonesia. In the following presentation of the data, level of educationis divided into four levels, as shown in Table 4 below.

Table 4. Categories for Level of EducationLevel of Education Ages Years of EducationElementary 7-12 1-6Junior High School 13-15 7-9Senior High School 16-18 10-12University 19 or more 13 or more

Table 4 shows most children finished their education until Senior High School.Averagely, the children spent three years for the level of Elementary, Junior High School, andSenior High School. Only few students continued their studies to the higher level of education,such as college or university.

Table 5. ILR Levels by Education

Level of Education EngganoILR Level

Elementary 1+Junior High 2Senior High 2+University *

Table 5 shows ILR level of Indonesian proficiency for respondents by education inEnggano. The average proficiency level in Indonesia was level 1+ for respondents witheducation up to SD, level 2 for respondents with education up to SMP, and level 2+ forrespondents with education up to SMU. This means that the bilingual proficiency level isapparently increasing when respondents have higher levels of education. For the people whohad an opportunity to continue their studies, proficiency in Indonesian was somewhathigher;,whereas proficiency in Indonesian was lower when respondents had little or noopportunity to continue their studies. Thus in the study sample, BI proficiency increases with


39

increasing education. Therefore, education is very significant, and has a strong influence onIndonesian proficiency for the respondents in the study.

AgeIn Table 6 below, the respondents were divided into two age groups: Young and Old. The Younggroup was defined as respondents between 15 and 35 years old. The Old group was defined asrespondents between 36 and 60 years old. It is likely that the difference in BI proficiencybetween the two age groups is caused by variance in level of education. In general, most youngpeople who were between ages 15 and 35 had attended school for 9 years or more, up to theSMP level (junior high school), while the older people who were between ages 36-60 hadattended school for less than 9 years or had never been to school.

Table 6. ILR Levels by AgeAge ILR levelYoung 2+Old 1+

On average individuals in the Young group are more proficient in BI than those in the Oldgroup. The Young group had an average BI proficiency of ILR level of 2+ while the Old groupshad an average BI proficiency of ILR level of 1+.. In Table 3, the coefficient value for theindependent variable age is negative, meaning that higher age correlates with lower SRT score.

Statistically Insignificant FactorsEight of the ten social and demographic factors investigated in this study were shown to bestatistically insignificant and were not proven to influence BI bilingual proficiency in the studysample. These factors were: exposure to literature, length of time spent outside community,gender, distance from school, travel frequency, exposure to radio, contact with outsiders, andexposure to television.

CONCLUSION

Education plays an essential role in the life of Enggano speakers, and also has a strong factor inincreasing bilingual Indonesian proficiency. In the study sample it appears that higher levels ofeducation increase a person’s bilingual proficiency in Indonesian. Even though the average levelof Indonesian proficiency for Enggano speakers can be described as Limited WorkingPerformance (level 2), their attitudes towards both Indonesian and the local vernacular languageare strongly positive. This can be seen from their motivation to still uses Indonesian in thedomains of education and government while using the vernacular is mostly in the domain ofhome. Additionally, for Enggano speakers, Indonesian is used for continuing higher education,while the vernaculer needs to be maintained as part of Enggano culture so that it can be passedon to the younger generation.

NOTE* I would like to thank an anonymous reviewer for very helpful comments on the earlier draft.** This research was conducted when the writer worked for SIL International Indonesia from 2006-2012.*** The writer also would like to thank to the research teams that I could not mention their names one byone. They were from different institutions that helped the writer gathering the data, creating the map anddoing data entry.


40

BIBLIOGRAPHY

Gordon, Raymond G, ed. 2005. Ethnologue, languages of the world. 15th ed. Dallas: SILInternational, 435.

Gumono. 2008. Upaya Pemeliharaan Bahasa Daerah yang Terancam Punah: Deskripsi KondisiKebahasaan, Sikap Bahasa, dan Kebutuhan Bahasa Masyarakat Pulau Enggano.. FKIPUniversitas Bengkulu (Accessed on http:fs.unitomo.ac.id/wp-content/uploads/2008/10/daftar-makalah-semna-uny-2008.do), April 4, 2009

Hanawalt, Charlie. 2008. The development of the Indonesian Sentence Repetition Test. Jakarta:SIL International-Indonesia. Unpublished draft.

http://www.ppmibengkulu.org/?p21. Post by admin | Wisata Bengkulu |, accessed on Sautrday,April 4, 2009

Interagency Language Roundtable. 2007. ILR skill level descriptions for interpretationperformance. URL: http://www.govtilr.org/Skills/interpretationSLDsapproved.htm

Sanders, Chip and IDB Academic Team. 2004. Towards an IDB policy on language projectplanning in multilingual settings: IDB Academic Team meeting. SIL International,Indonesian.

Sugondo, Dendy., Mahsun., Inyo., Kisyani., Multamia., Nadra., and Ferry, ed. 2008. Bahasa-Bahasa di Indonesia. Jakarta: Pusat Bahasa. Unpublished draft.

Radloff, Carla F. 1991. Sentence Repetition Testing for studies of community bilingualism.Dallas: Summer Institute of Linguistics and University of Texas at Arlington.


41

APPENDIX 1: SOCIOLINGUISTIC QUESTIONNAIRE

A. Kuesioner Penjajakan Dwibahasa#Responden

Tanggal: ____/____/____Lokasi: Desa_____________ Kec_____________ Kab_____________Pewawancara: _______________________Pengamat: _______________________Penerjemah BD: _______________________1. BiodataNama lengkap: ______________________Umur: ______________________Jenis kelamin: ______________________Pekerjaan: ______________________Apaka Anda sudah menikah? Y/TPendidikan formal: Y/TBerapa lama: _____ tahunSD SMP SMA D31 2 3 4 5 6 1 2 3 1 2 3

Tempat dilahirkan:Tempat dibesarkan:Bahasa apa yang Anda gunakan pertama kali waktu kecil?

Dari suku apa ayah Anda berasal?

Bahasa apa yang paling sering ayah Anda gunakan dengan Anda?

Dari suku apa ibu Anda berasal?

Bahasa apa yang paling sering ibu Anda gunakan dengan Anda?

Dari suku apa pasangan (suami/istri) Anda berasal?

Bahasa apa yang digunakan pasangan Anda dengan Anda?

Bahasa apa yang Anda gunakan sehari-hari di rumah?

Apakah Anda bisa berbahasa yang lain? Y/T1.18. Dari semua bahasa yang Anda bisa, apa bahasa yangpaling lancar: __________________kedua: __________________ketiga: __________________

2. Kontak2.1. Tempat tinggal di luar wilayah bahasaApakah Anda pernah tinggal di luar daerah ini? Y/TJika ya, berapa lama? (Catat yang lebih dari 6 bulan)_____________________________________________________________________Bahasa apa yang digunakan di daerah itu? __________________________________


42

2.2. Perjalanan (*Tulislah semua jawaban dalam tabel yang disediakan di bawah)Apakah Anda sering pergi ke luar daerah ini? Y/TKe mana Anda pergi?Berapa sering Anda pergi ke sana? (sethar beb/mng 1x/mng 2x/bln 1x/bln <1x/bln)Bahasa apa yang digunakan di daerah itu?Apa tujuan Anda biasanya pergi ke sana?b. Lokasi c. Frekuensi d. Bahasa e.Tujuan perjalanan2.3. Apakah Anda mempunyai tetangga atau teman yang tidak berasal dari [suku setempat]?

Y/TJika ya, bahasa apa yang Anda biasanya gunakan dengan mereka? ____________________Seberapa sering Anda bertemu mereka? Sering Kadang-kadang Jarang3. Media3.1. Radio/TVApakah Anda sering mendengarkan Radio? Y/TJika ya, seberapa sering? Sering Kadang-kadang JarangSiaran apa yang biasanya Anda dengarkan di radio? ________________________Siaran-siaran tersebut biasanya dalam bahasa apa? ________________________Apakah Anda mengerti siaran tersebut? ________________________Apakah Anda sering menonton TV? Y/TJika ya, seberapa sering? Sering Kadang-kadang JarangSiaran apa yang biasanya Anda dengarkan? _______________________Siaran-siaran tersebut biasanya dalam bahasa apa? _______________________Apakah Anda mengerti siaran tersebut? _______________________3.2. Buku dan BacaanApakah Anda suka membaca? Y/Ta. Jika ya, seberapa sering? Sering Kadang-kadang Jarangb. Jenis bacaan apa ada di rumah ini:Jenis bacaan Bahasa Mengerti (Y/T)Surat kabar, majalahAl’Quran/AlkitabBuku tentang agamaCerita (fiksi/nonfiksi)KomikBacaan di poster, Kalendar, dllBuku petunjukBuku kesehatanKamusSurat, pesan, email, SMSBrosur, pengumumanPeta, sketsa/gambarRekening atau fakturJumlah BI_____ BD____ dll____ Ya_____


43

APPENDIX 2: ILR LEVEL DESCRIPTORS (Interagency Language Roundtable 2007)

ILR level Attribute Description0 No Performance No functional ability to transfer information from

one language to another.0+ Memorized Performance Unable to transfer more than isolated words.1 Minimal Performance Unable to transfer more than isolated short

phrases.1+ Minimal Performance Plus Unable to transfer information reliably, even if

familiar with the subject matter.2 Limited Working

PerformanceUnable to transfer information reliably in mostinstances. May communicate some meaning whenexchanges are short, involve subject matter that isroutine or discourse that is repetitive orpredictable, but may typically require repetition orclarification. Expression in the target language isfrequently faulty.

2+ Limited WorkingPerformance Plus

Able to transfer information, not alwaysaccurately and completely, during routine,everyday, repetitive exchanges in informalsettings, but unable to perform adequately in thestandard interpretation modes. May falter,stammer, or pause, and often resort tosummarizing speech content. Idiomatic or culturalexpressions may not be rendered appropriately inmost instances. Language may be stilted orawkward.

3 Professional Performance Able to interpret consistently in the mode(simultaneous, consecutive, and sight) required bythe setting, provide renditions of informal as wellas some colloquial and formal speech withadequate accuracy, and normally meetunpredictable complications successfully. Canconvey many nuances, cultural allusions, andidioms, though expression may not always reflecttarget language conventions. Adequate delivery,with pleasant voice quality. Hesitations,repetitions or corrections may be noticeable butdo not hinder successful communication of themessage. Can handle some specialized subjectmatter with preparation. Performance reflects highstandards of professional conduct and ethics.


44

ILR level Attribute Description3+ Professional Performance

PlusAble to interpret accurately and consistently in themode (simultaneous, consecutive, and sight)required by the setting and provide generallyaccurate renditions of complex, colloquial andformal speech, conveying most details andnuances. Expression will generally reflect targetlanguage conventions. Demonstrates competencein the skills required for interpretation, includingcommand of both working languages, theircultural context, and terminology in thosespecialized fields in which the interpreter hasdeveloped expertise. Good delivery, with pleasantvoice quality, and few hesitations, repetitions, orcorrections. Performance reflects high standardsof professional conduct and ethics.

4 Advanced ProfessionalPerformance

Able to interpret in the mode (simultaneous,consecutive, and sight) required by the setting andprovide almost completely accurate renditions ofcomplex, colloquial, and idiomatic speech as wellas formal and some highly formal discourse.Conveys the meaning of the speaker faithfully,including most if not all details and nuances,reflecting the style, register, and cultural contextof the source language, without omissions,additions or embellishments. Demonstratesmastery of the skills required for interpretation,including command of both working languagesand their cultural context, expertise in somespecialized fields, and ability to prepare newspecialized topics rapidly and routinely. Very gooddelivery, with pleasant voice quality and onlyoccasional hesitations, repetitions or corrections.Performance reflects the highest standards ofprofessional conduct and ethics.


45

ILR level Attribute Description4+ Advanced Professional

Performance PlusAble to interpret in the mode (simultaneous,consecutive, and sight) required by the setting andprovide accurate renditions of informal, formal,and most highly formal discourse. Conveys themeaning of the speaker faithfully and accurately,including virtually all details and nuances,reflecting the style, register, and cultural contextof the source language, without omissions,additions or embellishments. Demonstratesmastery of the skills required for interpretation,including command of both working languagesand their cultural context, expertise in a number ofspecialized fields, and ability to prepare otherspecialized topics rapidly and routinely. Excellentdelivery, with pleasant voice quality and rarehesitations, repetitions or corrections.Performance reflects the highest standards ofprofessional conduct and ethics.

5 Master ProfessionalPerformance

Able to excel consistently at interpreting in themode (simultaneous, consecutive, and sight)required by the setting and provide accuraterenditions of informal, formal, and highly formaldiscourse. Conveys the meaning of the speakerfaithfully and accurately, including all details andnuances, reflecting the style, register, and culturalcontext of the source language, without omissions,additions or embellishments. Demonstratessuperior command of the skills required forinterpretation, including mastery of both workinglanguages and their cultural context, and wide-ranging expertise in specialized fields.Outstanding delivery, with pleasant voice qualityand without hesitations, unnecessary repetitions,and corrections. Exemplifies the highest standardsof professional conduct and ethics.


46

APPENDIX 3: KISH GRID FOR RANDOM SAMPLING

Eligible people Household1 2 3 4 5 6

1 1 1 1 1 1 12 1 2 1 2 1 23 1 2 3 1 2 34 1 2 3 4 1 25 1 2 3 4 5 36 1 2 3 4 5 67 1 2 3 4 5 68 1 2 3 4 5 69 1 2 3 4 5 6≥ 10 1 2 3 4 5 6


AKRONIM YANG BERFONOTAKTIK TIDAK LAZIMDALAM BAHASA INDONESIA

Sariah*Universitas Padjadjaran

[email protected]

AbstrakPenelitian akronim ini bertujuan mendeskripsikan akronim yang berfonotaktik tidaklazim dalam bahasa Indonesia. Ada tiga indikator yang dapat dijadikan pedoman, yaitu(1) jajaran fonem yang fonotaktik, (2) konsonan penutup suku kata yang lazim, dan (3)jumlah suku kata yang lazim dalam bahasa Indonesia. Data yang digunakan adalah datasurat kabar nasional Januari 2012—Maret 2013. Temuannya adalah akronim yangberfonotaktik tidak lazim umumnya terdapat pada bahasan penjajaran fonem, khususnyapada jajaran fonem /md/ Gakumdu, jajaran fonem /nm/ Menmud, jajaran fonem /pk/Apkasi, dan jajaran fonem /pm/ Ipmi. Akronim dengan dua konsonan berurutan yangsama terdapat pada konsonan /pp/ Bappebti, konsonan //ss/ Kopassus, /mm/ Jamman,dan /tt/ Unpatti. Akronim dengan konsonan rangkap adalah /dh/ pada Pusbadhi,konsonan /ngg/ Unpatti, dan konsonan /nd/ pada Unand. Akronim yang menggunakankonsonan yang tidak bisa menjadi penutup suku kata umumnya adalah akronimberpenutup suku kata dengan konsonan /g/ dan /d/. Konsonan penutup suku kata/c/ tidakditemukan dalam data akronim. Selain itu, ada akronim yang menggunakan konsonanpenutup suku kata /z/ dan konsonan /v/, sedangkan dalam kata serapan pun tidakditemukan konsonan /z/ dan konsonan /v/ sebagai penutup suku kata, seperti Baznas,Pemprov. Perulangan fonem yang sama dalam beberapa suku kata terdapat padariskesdas. Jumlah suku kata (silabe) terdiri atas empat suku kata atau lebih, sepertiBabinkumnas, Ditreskrimsus.

Kata kunci: akronim, fonotaktik, suku kata

AbstractThis acronym research describes some uncommon phonotactic acronyms in Indonesian.Three indicators were used as guidance: (1) the list of phonotactic phonemes, (2) the commonclosed consonants, (3) the number of common syllables in Indonesian. The data were takenfrom national news papers from January 2012 thorugh March 2013. The results indicated:(1) uncommon phonotactic acronyms were usually found in the phoneme list of /md/ forGakumdu, /nm/ for menmud, the /pk/ for Apkasi, and /pm/ for ipmi. (2) The acronyms of twoidentical sequenced consonants were found in the consonants /pp/ as in Bappepti, /ss/ inKopassus, /mm/ in jamman, and /tt/ in Unpatti. (3) The acronyms of double consonant werefound in the consonants /dh/ as in pusbadhi, /ngg/ in unpatti, and /nd/ in unand. Theacronyms containing consonanst that were not used as word-closing acronyms were theconsonants /g/ and /d/. The closing word consonant /c/ was not found in the data. Besides,acronyms using closed word consonants were found in the consonants /z/ and /v/. In borrowedwords the consonants that were not found were /z/ and /v/ as word-closing acronyms, such asin baznas, pemprov. The repetition of similar phonemes in some syllables was found inriskesdas. The number of syllables commonly found in Indonesian acronyms consisted of fouror more words, for example, babinkumnas, ditreskrimsus.

Keywords: acronym, phonotactic, syllable

PENDAHULUAN

Pembentukan akronim lazimnya mengikuti beberapa pertimbangan. Pertama, pengucapanakronim yang diciptakan mudah dilafalkan dalam bahasa yang bersangkutan. Kedua, jumlah

Sariah

48

suku kata yang lazim, mengikuti kaidah fonotaktik, yaitu kaidah yang menggambarkanpenjajaran fonem-fonem yang berlaku dalam suatu bahasa. Kaidah fonotaktik merupakankaidah yang mengatur urutan atau hubungan antara fonem-fonem suatu bahasa. Fonotaktikmempunyai pola yang terkait dengan pola penyukuan kata dan pergeseran bunyi yangmenimbulkan variasi bunyi satu fonem yang sama (Chaer, 2009:84).

Banyak bentuk akronim yang tidak mempertimbangkan kaidah fonotaktik, misalnyaakronim calhaj (calon haji). Dalam suku kata bahasa Indonesia tidak ditemukan kata yangberakhir dengan fonem /j/, kecuali dari bahasa asing. Di samping itu, akronim Unibraw(Universitas Brawijaya), Unand (Universitas Andalas) juga tidak sesuai dengan kaidahfonotaktik bahasa Indonesia karena tidak ditemukan kata bahasa Indonesia yang berakhirdengan fonem /w/ dan konsonan rangkap /nd/. Ada beberapa fonem konsonan bahasa Indonesiayang tidak dapat menjadi penutup suku kata, yaitu fonem konsonan /b/, /c/, /d/, /g/, /j/, /ny/, /w/,dan /y/ (Sudarno, 1990:31). Jika ada kata yang berakhir dengan salah satu konsonan tersebut,diduga kata itu berasal dari bahasa asing. Dengan demikian, jika ada akronim yangmenggunakan konsonan yang tidak dapat menjadi penutup suku kata, berarti akronim itu tidakmengikuti kaidah fonotaktik bahasa Indonesia. Bentuk lain yang tidak fonotaktik masih banyakditemukan, seperti caleg, penjaskes, Kopkamtib, Kemendikbud.

Saat akronim sprindik mengemuka, banyak orang bertanya apa sprindik sebuah kataIndonesia atau kata asing. Umumnya orang tidak mengetahui bentuk tersebut, kecuali orangyang setiap hari membaca surat kabar. Ternyata sprindik adalah sebuah akronim dari ‘suratperintah penyidikan’. Akronim itu terkesan asing bagi penutur bahasa Indonesia karenamenggunakan gugus konsonan /spr/ yang merupakan gugus konsonan asing atau serapan.Dengan demikian, gugus konsonan /spr/ pada kata sprei tidak dijadikan acuan untukmenentukan sebuah akronim yang fonotaktik.

Sebenarnya secara sederhana konsep fonotaktik tidak sukar dipahami. Misalnya, katabahasa asing phrase, psychology, business sukar dilafalkan dalam bahasa Indonesia. Olehkarena itu, orang Indonesia mengubah ejaannya menjadi frasa, psikologi, dan bisnis supayamudah dilafalkan. Dengan demikian, kata frasa, psikologi, dan bisnis mudah diucapkan dalampelafalan bahasa Indonesia karena susunan ejaan hurufnya (fonem) sudah sesuai dengan kaidahfonotaktik bahasa Indonesia. Sebaliknya, kata aslinya phrases, psychological, business sulitdilafalkan dengan lidah orang Indonesia karena susunan hurufnya tidak sesuai dengan kaidahtersebut. Dalam kaidah fonotaktik bahasa Indonesia tidak ada rangkaian huruf /p/ dan /h/(phrase) dan /p/, /s/, /y/ (psychology) pada awal suku kata dan konsonan /y/ pada psycgology diakhir suku kata. Konsonan /y/ diubah menjadi vokal /i/ supaya memudahkan pelafalan. Padakata asing business adaptasinya ke dalam bahasa Indonesia menjadi bisnis.

Berdasarkan paparan pada latar belakang di atas, masalah dalam tulisan ini adalahbagaimana akronim berfonotaktik tidak lazim yang terdapat dalam surat kabar nasional? Sesuaidengan masalah yang telah disebutkan di atas, tujuan penelitian ini adalah mendeskripsikan danmenganalisis akronim berfonotaktik tidak lazim yang terdapat dalam data surat kabar nasional.Manfaat penelitian dapat ditinjau secara teori dan praktik. Secara teori, penelitian ini dapatdimanfaatkan sebagai masukan dalam melahirkan akronim berfonotaktik lazim dalam bahasaIndonesia. Di samping itu, penelitian ini dapat mengungkapkan kriteria akronim berfonotaktiktidak lazim yang didasarkan pada kaidah fonotaktik bahasa Indonesia. Penelitian akronimberfonotaktik tidak lazim dengan data yang signifikan dapat menjadi informasi mengenaiakronim yang diciptakan selama ini: apakah telah sesuai atau menyimpang dari kaidahfonotaktik bahasa Indonesia. Secara praktis, penelitian ini dapat memberikan gambaranpemakaian akronim berfonotaktik tidak lazim sehingga dapat menjadi pedoman untukmenciptakan akronim baru yang mengikuti kaidah fonotaktik bahasa Indonesia, terutama bagilembaga atau sebuah kepanitiaan yang sering melahirkan akronim baru. Di samping itu,penelitian ini juga dapat menjadi informasi kebahasaan yang penting.


49

KERANGKA TEORI

Beberapa peneliti bahasa telah mengungkap pemakaian akronim. Di antaranya adalah Darheni,Irawati, Alaudin, dan Suratminto. Darheni (2000) meneliti pemakaian abreviasi yang berjudul“Penggunaan Singkatan dan Akronim Berbahasa Indonesia dalam Media Cetak: Ditinjau dariSegi Bentuk dan Makna”. Darheni mengamati keragaman bentuk singkatan dan akronim dalammedia massa cetak dan hasilnya adalah (1) proses pembuatan singkatan dan akronim yangberupa kata tunggal, gabungan kata, berafiks, berreduplikasi, pelesapan, dan penyingkatandalam akronim, (2) hasil pembentukan akronim yang meliputi perbedaan bentuk akronimdengan acuan sama, dan persamaan bentuk akronim berupa kata dan frasa, dan (3) acuanakronim dan/atau singkatan yang berupa kata tunggal dan jamak.

Penelitian senada pernah dilakukan Irawati (2004). Penelitian itu berjudul “Singkatandan Akronim dalam Media Chatting dan SMS: Analisis Komunikasi Teks dalam Internet danTelepon Seluler”. Hasilnya adalah pembentukan singkatan dan akronim pada media catting danSMS di samping menggunakan pola yang sudah ada juga menggunakan pola baru yang belumada dalam pola pembentukan singkatan dan akronim dalam bahasa Indonesia. Pola-pola barutersebut adalah singkatan dari angka sebagai pengganti huruf dan kata, singkatan dari gabunganhuruf dan angka, singkatan yang mengubah beberapa huruf, dan singkatan yang menghilangkanunsur vokal sebuah kata. Pola yang disebutkan terakhir ini yang sering muncul dalam teks SMS.

Penelitian yang berjudul “Bentuk-Bentuk Singkatan Bahasa Indonesia pada Iklan Mini(Studi Kasus pada Iklan Mini Kompas Tanggal 1—31 Agustus 2002)” pernah diteliti oleh Alaudin.Temuan pentingnya adalah bahwa tedapat singkatan pada iklan mini tersebut yang memilikibeberapa makna atau pengertian yang berbeda dari makna umum yang sudah dikenal masyarakat.

Mulyati (2009) mengamati pola-pola pembentukan akronim dan fenomena akronimdalam bahasa Indonesia. Tulisannya berjudul “Menyoroti Abreviasi: Singkatan dan Akronim”Hasil yang diperoleh dari pengamatannya itu adalah (1) pola pengakroniman dalam bahasaIndonesia tidak memiliki keajekan kaidah, (2) ledakan akronim memaksa penentu kebijakanperencana dan pembaku bahasa untuk selalu jeli dan cepat tanggap dalam menangani masalahini, paling tidak meninjau kamus secara periodik dan pengodifikasian pola-pola baru, (3)ledakan akronim juga berpengaruh terhadap pihak yang mau mempelajari bahasa Indonesia, (4)banyak yang sulit untuk dipahami, kecuali oleh kalangan tertentu sehingga bahasa Indonesiamenjadi “dialek”, dan (5) banyak akronim yang dipakai secara tumpang-tindih.

Lilie Suratminto (2010) meneliti “Abreviasi dan Akronim pada Batu Nisan Masa VOCdi Batavia”. Simpulan yang diperoleh adalah bahwa kebiasaan untuk menggunakan abreviasidan akronim sudah membudaya pada masyarakat VOC di Batavia pada abad ke-17 dan 18. Disamping itu, pembuatan plakat (pengumuman resmi dari pemerintah) menggunakan abreviasidan akronim.

Dengan demikian, beberapa penelitian sebelumnya dapat dikelompokkan berdasarkankajian dan data. Penelitian akronim berdasarkan kesamaan data terdapat pada penelitian Darheni(2000), penelitian Alaudin (2002), dan Sariah (2013). Selanjutnya, peneltian akronim berdasarkankesamaan kajian terdapat pada penelitian Darheni (2000), penelitian Irawati (2004), danpenelitian Suratminto (2010). Penelitian Sariah (2013) menggunakan teori fonotaktik. Jikadipetakan, penelitian sebelumnya dan penelitian yang akan dikaji sekarang adalah sebagai berikut.

Fonotaktik mengatur bagaimana fonem-fonem dan alofonnya tersusun secara beraturansehingga membentuk suatu kata yang bermakna. Alwi (2000:24) menegaskan bahwa fonotaktikmembahas rentetan bunyi, yaitu satu bunyi diiringi bunyi yang lain. Bunyi-bunyi itu mewakilirangkaian fonem serta alofonnya. Rangkaian fonem itu tidak bersifat acak, tetapi mengikutikaidah-kaidah tertentu. Fonem apa yang dapat mengikuti fonem yang mana ditentukanberdasarkan perjanjian oleh para pemakai bahasa itu sendiri. Kaidah yang mengatur penjajaranfonem suatu bahasa dinamakan kaidah fonotaktik.

Pengertian fonotaktik menurut Parera (1985:19) dalam Fonetik dan Fonemik adalahsuatu prosedur penemuan dan penentuan tata urut dalam tata hubung fonem-fonem dalam

Sariah

50

sebuah bahasa. Yang dibicarakan dalam fonotaktik adalah pola urutan bunyi, distribusi fonem,pola suku kata, gugus bunyi konsonan dan vokal, jenis-jenis gugus bunyi yang mungkin danyang tidak mungkin pada tingkat kata atau antarkata.

Bahasa Indonesia juga mempunyai kaidah semacam itu. Kaidah fonotaktik itulah yangmenyebabkan kita dapat merasakan secara intuitif bentuk mana yang berterima (kelihatanseperti kata Indonesia) meskipun belum pernah kita dengar/lihat sebelumnya dan mana yangtidak berterima.

Pola fonotaktik adalah kaidah pergeseran bunyi dalam pelafalan kata, baik kata dasarmaupun kata turunan, sebagai akibat pengaruh bunyi yang ada pada lingkungannya (baiksebelum dan sesudahnya). Pergeseran ini menimbulkan variasi bunyi dari satu fonem yangsama. Bahasa Indonesia mengizinkan jajaran, seperti /-nt-/ (untuk), /-rs-/ (bersih), dan /-st-/(pasti), tetapi tidak mengizinkan jajaran, seperti /-pk-/ dan /-pd-/ karena tidak ada morfembahasa Indonesia yang menjajarkan fonem seperti itu (Alwi, 2000:24). Jadi, bentuk-bentukseperti opkir dan kapdu terasa janggal dan memang tidak ada kata dengan jajaran fonem yangdemikian dalam bahasa Indonesia. Oleh karena itu, singkatan, khususnya akronim hendaknyaserasi dengan kaidah fonotaktik bahasa Indonesia.

Menurut Kentjono (2005:164) kaidah fonotaktik adalah aturan dalam merangkai fonemuntuk membentuk satuan fonologis yang lebih besar, misalnya suku kata. Deretan fonem yangterdapat dalam bahasa Indonesia mempunyai pola fonotaktik seperti halnya deret fonem bahasa-bahasa lain. Deret fonem tersebut meliputi gugus vokal, deret vokal, gugus konsonan, deretkonsonan, deret vokal dan deret konsonan dalam satu suku kata (silabel).

(1) Gugus vokal bahasa Indonesia adalah diftong /ai/, /au/, /oi/ dan /ei masing-masing padakata pulau, santai, asoi, survei.

(2) Deret vokal bahasa Indonesia adalah sebagai berikut./ae/ : daerah/ai/ : saing/au/ : kaum, mau/ea/ : beasiswa, kreasi/ea/ : seakan/ee/ : keenam/ee/ : seekor/ei/ : mei/ei/ : seikat/eo/ : feodal, beo, pameo/eu/ : seutas/ia/ : tiap, dia, giat/io/ : kios radio, biola/iu/ : tiup, nyiur/oa/ : soal, doa/ua/ : dua, suap/ue/ : kue. duet/ui/ : buih. kuil/uo/ : kuota

(3) Gugus konsonan bahasa Indonesia terdapat di bawah ini./bl/ : blanko, gamblang/br/ : brantas, ambruk/dr/ : drama, drakula/dw/ : dwi/fl/ : flu, flamboyan/fr/ : frustasi, fragmen/gl/ : gladi, global


51

/gr/ : granat, gram/kl/ : klinik, klasik/kr/ : kriminal/ks/ : ekstra, eksponen/kw : kwartet/pl/ : pleonasme, pleno/ps/ : psikologi, psikiater/sl/ : slip, slogan/sp/ : spanduk, sponsor/sr/ : sragen/sw/ : swasta, swalayan/tr/ : mitra, tragedi

(4) Deret konsonan bahasa Indonesia terdapat di bawah ini/bd/ : abdi, sabda/gm/ : dogma, magma/gn/ : signal, kognisi/hb/ : sahbandar/hd/ : syahdu, syahdan/hk/ : bahkan/hl/ : ahli, tahlil/hs/ : dahsyat/ht/ : tahta, bahtera/hw/ : bahwa/hy/ : sembahyang/kd/ : takdir, bakda/kl/ : maklum, takluk/km/ : sukma/kn/ : yakni, makna, laknat/kr/ : makruf, takrif/ks/ : paksa, siksa/kt/ : bukti, bakti/kw/ : takwa, dakwa/ky/ : rukya, rakyat/lj/ : salju, aljabar/lm/ : ilmu, gulma, palma/ls/ : palsu, balsem, pulsa/lt/ : salto, sultan/mb/ : ambil, gambar/ml/ : jumlah/mp/ : empat, pimpin/mr/ : jamrut/nc/ : kunci, lancar/nd/ : indah, pandang/ngk/ : engkau, mungkin/ngs/ : angsa, bangsa, mangsa/nj/ : banjir, janji/np/ : tanpa/ns/ : insang, insan/nt/ : ganti, untuk/pt/ : sapta, baptis/rb/ : kerbau, terbang/rc/ : percaya, karcis

Sariah

52

/rd/ : merdu, merdeka, kerdil/rg/ : pergi, harga, surga/rj/ : kerja, sarjana, terjang/rk/ : terka, perkara, murka/rl/ : kerlip, kerling, perlu/rm/ : cermin, derma/rn/ : warna, purnama, ternak/rs/ : gersang, bersih,/rt/ : arti, harta, serta/sb/ : asbak, asbes/sh/ : masyhur/sl/ : muslim/sm/ : basmi, asmara/sp/ : puspa, aspal/st/ : pasti, dustaParameter yang dijadikan patokan adalah jajaran fonem yang berasal dari bahasa

Indonesia sehingga jajaran fonem yang bergugus tiga konsonan tidak digunakan sebagai acuan.Gugus konsonan terdiri atas tiga fonem berderet dalam satu suku kata adalah /stra/ strategi,instruksi, /spr/ sprei, /skr/ skripsi, /skl/ sklerosis. Jumlah konsonan asli bahasa Indonesia adadelapan belas dengan ketentuan sebelas konsonan dapat menjadi penutup suku kata, yaitukonsonan /h/, /k/, /l/, /m/, /n/, /ng/, /ny/, /p/, /r/, /s/, /t/ dan ada tujuh konsonan bahasa Indonesiayang tidak bisa menjadi penutup suku kata, yaitu konsonan /b/, /c/, /d/, /g/, /j/, /w/. dan /y/(Sudarno, 1990:46). Konsonan penutup /ny/ dalam tulisan berupa konsonan /n/ saja. Mengapapenulis berpatokan pada penjajaran fonem asli dan konsonan penutup suku kata bahasaIndonesia karena pertimbangan bahwa banyak unsur serapan yang berterima dalam bahasaIndonesia sehingga akronim yang dilahirkan masih belum mendekati kata bahasa Indonesia,seperti akronim sprindik tadi. Di samping itu, konsonan yang tidak bisa menjadi penutup sukukata di antaranya adalah /b/, /d/, dan /j/, misalnya pada kata serapan dari bahasa Arab, sepertisebab, mikraj, sujud.

Dengan mengenal karakteristik kata bahasa Indonesia dan kaidah fonotaktiknya, adatiga indikator yang dapat dijadikan pedoman, yaitu jajaran fonem yang fonotaktik, konsonanpenutup suku kata yang lazim, dan jumlah suku kata yang lazim dalam bahasa Indonesia.

METODE PENELITIAN

Metode yang digunakan dalam penelitian ini ialah metode deskriptif. Hasil yang diperolehberupa pemerian bahasa apa adanya secara terperinci dan mendalam (Sudaryanto, 1993:62).Berdasarkan uraian tersebut, metode deskriptif dipakai untuk memaparkan hasil temuan yangdiperoleh dalam penelitian berupa penggambaran akronim yang berfonotaktik tidak lazim secarasistematik dan faktual berdasarkan data yang dikumpulkan dari Januari 2012—Maret 2013 padasuarat kabar nasional (Kompas, Media Indonesia, Republika, dan Sindo). Pendapat Sudaryantoini sejalan dengan apa yang dinyatakan oleh Djajasudarma (2010:8) bahwa penggunaan metodedeskriptif bertujuan membuat gambaran yang sistematik dan akurat mengenai data, sifat-sifat,serta hubungan fenomena-fenomena yang diteliti. Metode deskriptif dalam tulisan ini dipakaiuntuk memaparkan hasil temuan yang berupa pemakaian akronim yang berfonotaktik tidaklazim dalam bahasa Indonesia. Pengambilan data dilakukan dengan studi pustaka, yaitumengumpulkan pemakaian akronim dalam surat kabar nasional. Kemudian data tersebut dicatat,diklasifikasikan, dan dianalisis.


53

PEMBAHASAN

Berdasarkan pengamatan terhadap data, paparan dalam tulisan ini dikelompokkan menjadi (1)jajaran fonem yang tidak lazim dalam bahasa Indonesia; (2) dua konsonan berurutan yang samadalam satu kata; (3) konsonan yang tidak bisa menjadi penutup suku kata bahasa Indonesia; (4)perulangan konsonan yang sama dalam beberapa suku kata; (5) jumlah suku kata yang terdiriatas empat suku kata lebih.

Jajaran Fonem yang Tidak Lazim dalam Bahasa IndonesiaJajaran fonem bahasa Indonesia yang lazim telah disebutkan di atas. Berdasarkan telaahterhadap data, ditemukan akronim yang berfonotaktik tidak lazim. Ketidaklaziman itu mengacupada teori fonotaktik yang telah disebutkan di atas. Dengan demikian, kefonotaktikan mengacupada daftar gugus vokal (diftong), deret vokal, gugus konsonan, dan deret konsonan yang telahdisebutkan di atas.

Dalam paparan ini jajaran fonem yang berfonotaktik tidak lazim atau berbeda denganapa yang telah dirumuskan oleh para ahli bahasa tidak disebutkan seluruhnya hanya beberapacontoh saja.

Tabel 1. Jajaran Fonem yang Tidak LazimNo. Akronim Jajaran Fonem Bentuk Panjangnya1 Bulcup /bk/ Federasi Organisasi Buruh Migran Indonesia2 FOBMI /bm/ Badan Legislatif Daerah3 Balegda /gd/ Depot Logistik Daerah4 Dologda /gd/ Bulungan Cup5 FAGSAM /gs/ Front Antigerakan Aceh Merdeka6 GATBI /tb/ Gabungan Toko Buku Indonesia7 Dispendukcapil /kc/ Dinas Kependudukan dan Pencatatan Sipil8 dokcil /kc/ dokter kecil9 Amdal /md/ Analisis Dampak Lingkungan10 Gakumdu /md/ Penegakan Hukum Terpadu11 Jamda /md/ Jambore Daerah10 Kamnas /mn/ Keamanan Nasional11 Komnet /mn/ Kompor megnet12 Binmatkum /nm/ Binaan Masyarakat Taat Hukum13 Menmud /nm/ Menteri Muda14 Ranmor /nm/ Kendaraan bermotor15 Kopbumi /pb/ Koperasi Perlindungan Buruh Migran Indonesia16 Pusdalopbang /pb/ Pusat Pengendalian Operasi Pembangunan17 APDASI /pd/ Asosiasi Pedagang Daging Sapi Indonesia18 Bakopda /pd/ Badan Koordinasi Pembangunan Daerah19 Apkasi /pk/ Asosiasi Pemerintahan Kabupaten Seluruh

Indonesia20 Gapki /pk/ Gabungan Pengusaha Kelapa Sawit Indonesia)21 Kopkamtib /pk/ Komandan Operasi Keamanan dan Ketertiban22 GAPMMI /pm/ Gabungan Pengusaha Makanan dan Minuman

Indonesia23 Apmiso /pm/ Asosiasi Pedagang Bakso Indonesia24 HIPMI /pm/ Himpunan Pengusaha Muda Indonesia

Sariah

54

Jika diamati, akronim FOBMI mempunyai jajaran fonem /b/ dan /m/. fonem /b/ dan /m/berasal dari daerah artikulasi yang sama, yaitu daerah artikulasi bilabial (bibir). Di samping itu,fonem /f/ sebagai pembuka bunyi merupakan konsonan frikatif labiodental (bunyi yangdihasilkan dengan udara menggeser alat ucap). Jika konsonan frikatif labiodental /f/dipertemukan dengan konsonan hambat bilabial /b/ pada suku kata fob, bunyi yang dihasilkansulit untuk diucapkan. Dengan demikian, akronim FOBMI tidak mengikuti penjajaran fonemyang lazim dalam bahasa Indonesia. Selain itu, jajaran fonem /bm/ tidak terdapat dalam jajaranfonotaktik bahasa Indonesia yang disebutkan di atas.

Jajaran fonem /gd/ pada akronim balegda dan dologda tidak terdapat dalam jajaranfonem fonotaktik. Fonem /g/ dan fonem /d/ adalah fonem hambat bersuara meskipun berasaldari daerah artikulasi yang berbeda. Daerah artikulasi fonem /g/ adalah hambat velar (belakanglidah ditempelkan pada langit-langit lunak), sedangkan fonem /d/ adalah hambat alveolar (ujunglidah ditempelkan pada gusi). Jajaran fonem itu tidak ditemukan dalam kata-kata bahasaIndonesia.

Akronim Bulcup merupakan akronim yang dilafalkan menjadi Bulkap. Jajaran fonem/lc/ juga tidak lazim dalam kata-kata bahasa Indonesia. Jajaran fonem /gs/ pada akronimFAGSAM juga tidak ditemukan dalam daftar jajaran fonotaktik yang telah disebutkan di atas.Selain itu, fonem /g/ umumnya tidak menjadi penutup suku kata bahasa Indonesia. Bentuk lainterdapat pada akronim GATBI. Jajaran fonem /tb/ tidak terdapat dalam daftar jajaran fonemfonotaktik.

Jajaran fonem /kc/ terdapat pada akronim Dispendukcapil dan dokcil, tetapi jajaranfonem /kc/ tidak ditemukan dalam jajaran fonem fonotaktik. Fonem /k/ adalah konsonan hambatvelar yang dihasilkan dengan menempelkan belakang lidah pada langit-langit lunak. Udaradihambat pada tempat tersebut dan kemudian dilepaskan. Sebaliknya, fonem /c/ adalahkonsonan afrikat palatal yang dilafalkan dengan lidah ditempelkan pada langit-langit keras dankemudian dilepas secara perlahan sehingga udara dapat lewat dengan menimbulkan bunyi desis.Konsonan /k/ bertumpuh pada lidah dengan langit-langit lunak, sedangkan konsonan /c/bertumpuh pada lidah dengan langit-langit keras. Secara pelafalan dua konsonan itu mudahdiucapkan, tetapi tidak lazim dalam kata bahasa Indonesia.

Akronim amdal, Gakumdu, dan Jamdal dibentuk dengan jajaran fonem yang sama,yaitu /md/. Bentuk tersebut tidak terdapat dalam daftar jajaran fonem fonotaktik. Bahkan, tidakditemukan kata bahasa Indonesia (melihat kamus) yang menggunakan deret konsonan /md/.Selain itu, akronim yang dihasilkan dengan jajaran fonem tersebut terasa janggal ketikadilafalkan. Berikutnya adalah jajaran fonem /mn/ yang terdapat pada akronim Kamnas dankomnet. Jajaran fonem /mn/ tidak terdapat dalam daftar jajaran fonem fonotaktik. Setelahmelihat kamus, penulis belum menemukan kata bahasa Indonesia yang menggunakan jajaranfonem tersebut.

Binmatkum, menmud, dan ranmor adalah akronim yang terasa aneh diucapkan.Akronim Binmatkum terdiri atas tiga suku kata. Suku kata pertama ditutup dengan fonem /n/nasal alveolar, suku kata kedua ditutup dengan fonem /t/ hambat alveolar, dan suku kata ketigaditutup dengan fonem /m/ nasal bilabial. Suku kata kedua dan suku kata ketiga masing-masingdiakhiri dengan konsonan hambat sehingga terasa janggal diucapkan. Akronim menmudmempunyai dua suku kata. Suku kata pertama diakhiri dengan fonem /n/ nasal alveolar dan sukukata kedua dikahiri dengan fonem /d/ hambat alveolar. Jadi, dua suku kata diakhiri denganfonem dari daerah artikulasi yang sama, yaitu alveolar maka bunyi yang dihasilkan terasa aneh.Berikutnya adalah akronim ranmor. Akronim yang diawali dengan fonem getar alveolar /r/ dandiakhiri juga dengan fonem yang sama, awal bergetar dan akhir juga bergetar. Di antara ketigaakronim tersebut, terdapat satu persamaan, yaitu ketiga akronim menggunakan jajaran fonem/nm/. Dalam daftar jajaran fonem fonotaktik tidak ditemukan jajaran fonem /nm/.


55

Jajaran fonem /pb/ tidak lazim dalam fonotaktik bahasa Indonesia. Fonem /p/ adalahfonem hambat bilabial tak bersuara, sedangkan fonem /b/ adalah hambat bilabial bersuara.Dengan demikian, kedua fonem berasal dari daerah artikulasi yang sama. Perulangan fonem daridaerah artikulasi yang sama menimbulkan kesulitan dalam pengucapan. Akronim yangmenggunakan jajaran fonem /pb/ adalah Kopbumi dan Pusdalopbang. Kedua akronim initampak janggal terdengar.

Akronim APDASI dan Bakopda adalah akronim yang menggunakan deret konsonan/pd/. Deret konsonan tersebut tidak ditemukan dalam deret konsonan fonotaktik bahasaIndonesia. Di samping itu, tidak ditemukan kata (lema) bahasa Indonesia yang menggunakanderet konsonan /pd/, kecuali pada akronim APDASI dan Bakopda. Jika melihat daerah dan caraartikulasinya, konsonan tersebut adalah konsonan hambat. Konsonan /p/ adalah konsonanhambat bersuara yang dilafalkan dengan bibir atas dan bibir bawah terkatup rapat sehinggaudara dari paru-paru tertahan untuk sementara waktu sebelum katupan itu dilepaskan.Sebaliknya, konsonan /d/ adalah hambat bersuara yang dilafalkan dengan ujung lidahditempelkan pada gusi kemudian udara dari paru-paru dilepaskan. Tampaknya, sulit melafalkandua konsonan hambat secara bersamaan sehingga akronim dengan deret konsonan /pd/ tidaklazim digunakan penutur bahasa Indonesia.

Apkasi adalah akronim yang menggunakan deret konsonan /pk/. Adalah Akronim lainyang menggunakan deret konsonan /pk/ adalah Apkindo dan Gapki. Fonem /p/ berasal darikonsonan hambat bilabial dan fonem /k/ berasal dari konsonan hambat velar. Berarti keduafonem merupakan konsonan hambat dan hambatnya adalah hambat tak bersuara. Jajaran fonem/pk/ tidak terdapat dalam daftar jajaran fonem fonotaktik. Di samping itu, ketiga akronimtersebut, yaitu Apkasi, Apkindo, dan Gapki agak sulit diucapkan.

Jajaran fonem /pm/ terdapat pada akronim GAPMMI dan HIPMI. Fonem /p/ dan fonem/m/ berasal dari daerah artikulasi yang sama, yaitu bilabial. Jajaran fonem tersebut tidak terdapatdalam daftar jajaran fonem fonotaktik sehingga akronim GAPMMI dan HIPMI dianggapakronim yang tidak fonotaktik dalam bahasa Indonesia.

Dua Konsonan Berurutan yang SamaBahasa Indonesia tidak mengenal konsonan rangkap yang sama dalam satu kata, sedangkanbahasa Arab mengenal tasydid. Tasydid jika diindonesiakan menjadi konsonan rangkap yangjenisnya sama. Kata majalah awalnya adalah majallah, kata wasalam awalnya adalah wassalam.Dua konsonan berurutan yang sama dalam EYD tidak ditemukan, kecuali konsonan rangkap,seperti /kh/ pada akhlak, /sy/ pada syarat, /ny/ pada nyali, dan /ng/ pada ngeri (2003:3). Dengandemikian, akronim dalam bahasa Indonesia harus mengikuti fonem-fonem yang terdapat dalamEYD sehingga diharapkan bahwa akronim tidak menyimpang dari kata bahasa Indonesia.

Akronim yang menggunakan dua konsonan berurutan yang sama tedapat pada beberapacontoh berikut.

Tabel 2. Dua Konsonan Berurutan yang SamaNo. Akronim Jajaran Fonem Bentuk Panjangnya1 GAPMMI /mm/ Gabungan Pengusaha Makanan dan Minuman

Indonsia2 Jamman /mm/ Jaringan Masyarakat Mandiri3 APPI /pp/ Asosiasi Pemain Profesional Indonesia atau Asosiasi

Produsen Pupuk Indonesia4 APPMI /pp/ Asosiasi Pengusaha dan Perancang Mode Indonesia5 APPUI /pp/ Asosiasi Produsen Pakan Udang Indonesia6 Bappebti /pp/ Badan Pengawas Perdagangan Berjangka Komoditi7 Bappeda /pp/ Badan Perencanaan Pembangunan Daerah

Sariah

56

No. Akronim Jajaran Fonem Bentuk Panjangnya8 Bappenas /pp/ Badan Perencanaan Pembangunan Nasional9 Bappertal /pp/ Badan Pengawas Pelaksanaan Reformasi Total10 Formappi /pp/ Forum Masyarakat Perduli Parlemen Indonesia11 GUPPI /pp/ Gabungan Usaha Perbaikan Pendidikan Indonesia12 HIPPI /pp/ Himpunan Pengusaha Pribumi Indonesia13 Noppen /pp/ Nomor Pokok Penduduk14 Perppu /pp/ Peraturan Pemerintah Penggani undang15 Kopassus /ss/ Komando Pasukan Khusus16 Unpatti /tt/ Universitas Patimura

Dalam EYD dan dalam jajaran fonem bahasa Indonesia tidak ditemukan duakonsonanan berurutan yang sama dalam satu kata. Akronim APPI memiliki dua makna yangberbeda, yaitu pemain profesional dan produsen pupuk. Konsonan rangkap /pp/ pada akronimtersebut termasuk menyimpang dari kelaziman karena tidak ada kata dalam bahasa Indonesiayang menggunakan bentuk tersebut. Bentuk yang sama terdapat pada akronim APPMI danAPPUI. Konsosan rangkap /pp/ pada APPMI tidak langsung bertemu vokal, tetapi masih adakonsonan lain, yakni konsonan /m/ sehingga jajaran fonemnya tersusun dari tiga konsonansekaligus dan tidak termasuk dalam gugus konsonan atau kluster. Akibatnya, bentuk akronimAPPMI agak sulit dilafalkan. Akronim APPUI hanya mempunyai dua konsonan berurutan yangsama, yaitu /pp/, tetapi vokal rangkap /ui/ di akhir suku kata tidak nyaman terdengar. Bentuksenada terdapat juga pada GAPPMI, GUPPI, HIPPI, Bappebti, Bappeda, Bappenas, Bappertal,Formappi, Noppen, dan Perppu. Tampaknya, dua konsonan berurutan yang sama dalam contohakronim didominasi oleh fonem /p/. Namun, ada bentuk lain, yaitu konsonan /ss/ pada akronimKopassus, konsonan /mm/ pada akronim Jamman, konsonan /tt/ pada akronim Unpatti.

Tabel 3. Dua Konsonan Berurutan yang BerbedaNo. Akronim Jajaran Fonem Bentuk Panjangnya1 Pusbhadi /bh/ Pusat Bantuan dan Pengabdi Hukum Indonesia2 Sultengg /ngg/ Sulawesi Tenggara3 Asphurindo /ph/ Asosiasi Penyelenggara Haji Umrah dan Inbound

Indonesia4 Konphalindo /ph/ Konsorsium Perlindungan Hutan Lindung Indonesia5 Unand /nd/ Universitas Andalas6 Yapusham /sh/ Yayasan Pusat Hak Asasi Manusia

Bentuk lain yang juga masih menggunakan konsonan rangkap, tetapi dengan jenis yangberbeda adalah Pusbadhi (Pusat Bantuan dan Pengabdi Hukum Indonesia), Sultengg (SulawesiTenggara), dan Unand (Universitas Andalas). Konsonan rangkap yang dimaksud bentuknyatidak sama karena menggunakan dua konsonan yang berbeda, seperti /dh/ pada akronimPusbadhi, /ngg/ pada akronim Sultengg, dan /nd/ pada akronim Unand. Jadi, ada dua pengertiandalam bagian ini, yakni dua konsonan berurutan yang sama dalam satu kata dan dua konsonanberurutan yang berbeda dalam satu kata. Keduanya tidak berterima dalam EYD dan lazimnyakata dalam bahasa Indonesia.

Konsonan yang Tidak Bisa Menjadi Penutup Suku Kata Bahasa IndonesiaKonsonan yang tidak dapat menjadi penutup suku kata bahasa Indonesia adalah /b/, /c/, /d/, /g/,/j/, /w/, dan /y/. Fonem /b/ dalam bahasa Indonesia tidak dapat menjadi penutup suku kata padakata bahasa Indonesia asli, tetapi ada beberapa kata serapan yang menggunakan fonem /b/,seperti kata sebab, magrib. Penulis bersikap tetap menggunakan acuan tersebut untukmenentukan akronim yang berfonotaktik tidak lazim. Dalam tulisan ini hanya beberapa data sajayang diungkap, yaitu sebagai berikut.


57

Tabel 4. Konsonan yang Tidak Bisa Menjadi Penutup Suku Kata Bahasa IndonesiaNo. Akronim Jajaran Fonem Bentuk Panjangnya1 Abnon /b/ Abang None2 Pemkab /b/ Pemerintahan Kabupaten3 Seskab /b/ Sekretaris Kabinet4 jablai /b/ jarang dibelai5 Baznas /z/ Badan Amil Zakat Nasional6 Balegda /g/ Badan Legislatif Daerah7 Bulog /g/ Badan Urusan Logistik8 Caleg /g/ Calon Legislatif9 capeg /g/ calon pegawai10 Dologda /g/ Depot Logistik Daerah11 Gerag /g/ Gerakan Rakyat untuk Reformasi Agraria12 karpeg /g/ kartu pegawai13 Kemenag /g/ Kementerian Agama14 Disperindag /g/ Dinas Perindustrian dan Perdagangan15 Mayjen /y/ Mayor Jenderal16 Danlanud /d/ Komandan Pangkalan Udara17 Kostrad /d/ Komando Strategi Angkatan Darat18 Menmud /d/ Menteri Muda19 Pacad /d/ Perwira Cadangan20 Pemred /d/ Pemimpin Redaksi21 Pidsus /d/ Pidana Khusus22 Pemprov /v/ Pemerintah Provinsi23 Pomad /d/ Polisi Militer Angkatan Darat24 Puskud /d/ Pusat Koperasi Unit Desa25 Unibraw /w/ Universitas Brawijaya26 Unej /j/ Universitas Jember27 Calhaj /j/ Calon Haji

Konsonan /b/ dalam akronim abnon dan seskab termasuk bertentangan dengan lazimnyakata bahasa Indonesia yang tidak dapat berupa kononan /b/ meskipun ada beberapa kata yangdiserap menggunakan konsonan /b/ diakhir suku kata. Akronim abnon dan Seskab jika dilafalkanterasa sulit dan terdengar janggal. Jadi, meskipun konsonan tersebut ditemukan dalam kata serapan,dalam membentuk atau menciptakan akronim diupayakan untuk menghindari bentuk tersebut.

Akronim Baznas adalah akronim yang salah satu penutup suku katanya menggunakankonsonan /z/. Konsonan /z/ adalah konsonan serapan. Dalam kata bahasa Indonesia tidakditemukan konsonan penutup suku kata yang menggunankan konsonan /z/. Konsonan /z/termasuk konsonan frikatif alveolar yang dibentuk dengan cara menempelkan ujung lidah padagusi dengan pita suara bergetar. Konsonan /z/ sebagai konsonan serapan pemakaiannya hanya diawal kata, terutama kata serapan. Dalam kamus tidak ditemukan konsonan penutup suku katayang menggunakan konsonan /z/ meskipun kata itu berasal dari kata serapan. Dengan demikian,pembentukan akronim Baznas dianggap keluar dari kaidah fonotaktik bahasa Indonesia.

Konsonan /v/ sebagai konsonan serapan tidak ditemukan sebagai penutup suku katasehingga jika ada akronim pemprov, bentuk itu terasa sulit dilafalkan. Konsonan /v/ umumnyadigunakan sebagai konsonan pertama pembuka suku kata, tetapi tidak bisa menjadi konsonanpenutup suku kata. Beberapa kata yang menggunakan konsonan /v/ adalah varian, verba, vena,volume, dan sebagainya.

Sariah

58

Konsonan berikutnya yang tidak bisa menjadi penutup suku kata dalam bahasaIndonesia adalah konsonan /d/. Akan tetapi, akronim yang menggunakan konsonan /d/ di akhirsuku kata banyak ditemukan dalam data. Akronim itu antara lain adalah Menmud, paced,pemred, pidsus, Pomad, Puskud. Akronim Menmud, paced, pemred, pidsus, Pomad, dan Puskudtidak sulit untuk dilafalkan, tetapi terasa aneh terdengar. Bunyi fonem /d/ dalam kata bahasaIndonesia memang tidak lazim, kecuali kata serapan, seperti zuhud, sujud.

Akronim Balegda, Bulog, caleg, dan Disperindag termasuk akronim yangmenggunakan konsonan penutup suku kata yang tidak lazim dalam kata bahasa Indonesia.Konsonan /g/ di akhir suku kata menyimpang dari kaidah fonotaktik sehingga pencipta akronimdiharapkan mempertimbangkan kaidah fonotaktik dalam melahirkan sebuah akronim baru.Konsonan /g/ dalam bahasa Jawa tampaknya sering berubah menjadi konsonan /k/ ketikamenjadi bahasa Indonesia, misalnya kata bedug berubah menjadi beduk.

Konsonan selanjutnya yang tidak bisa menjadi penutup suku kata adalah konsonan /j/,/y/, dan /w/. Akronim yang menggunakan konsonan penutup suka kata /j/, /y/, dan /w/ tidakbanyak ditemukan dan termasuk langka dalam data. Akronim yang menggunakan konsonantersebut adalah unej, calhaj, mayjen, dan Unibraw. Namun, ada satu konsonan penutup sukukata yang tidak ditemukan dalam data, yaitu konsonan /c/. Dalam kata-kata bahasa Indonesiatidak ditemukan konsonan penutup suku kata menggunakan konsonan /c/.

Perulangan Fonem yang Sama dalam Beberapa Suku KataAkronim yang menggunakan fonem yang sama dalam beberapa suku kata terdapat dalam tabelempat berikut.

Tabel 5. Perulangan Fonem yang Sama dalam Beberapa Suku KataNo. Akronim Perulangan Fonem Bentuk Panjangnya1 Bapedalda /d/ Badan Pengendalian Dampak

Lingkungan Daerah2 Ditreskrimsus /s/ Direktorat Reserse Kriminal Khusus3 Jamkesmas /s/ Jaminan kesehatan Masyarakat4 Kopkamtib /p/, /m/, /b/ Komandan Operasi Keamanan dan

Ketertiban5 Pangkopkamtib /p/, /m/, /b/ Panglima Komando Pemulihan

Keamanan dan Ketertiban6 Pangkopkamtibda /p/, /m/, /b/ Panglima Komando Pemulihan

Keamanan dan Ketertiban Daerah7 Paspampres /p/, /s/ Pasukan Pengaman Presiden8 Perspebsi /p/, /s/ Persatuan Spesialis Bedah Saraf

Indonesia9 Puscadnas /s/ Pusat Cadangan Nasional10 Pusdalobpang /b/, /p/ Pusat Pengendalian Operasi

Pembangunan11 Puskesmas /s/ Pusat Kesehatan Masyarakat12 Puslabfor /b/, /f/ Pusat Labortorium Forensik13 Riskesdas /s/ Riset Kesehatan Dasar14 Perpres /p/, /r/ Peraturan Presiden15 Perpusnas /p/, /s/ Perpustakaan Nasional16 Propernas /p/, /s/ Program Pembangunan Nasional17 Bapepam /p/ Badan Pengawas Pasar Modal


59

Akronim Bapedalda menggunakan pengulangan suku kata dal dan da pada suka kataketiga dan suku kata keempat. Kata bahasa Indonesia yang menggunakan tipe seperti itu tidakditemukan sehingga akronim Bapedalda terdengar janggal atau terasa aneh. Bentuk yanghampir sama ditemukan pada akronim Jamkesmas, riskesdas, Prepusnas, puskesmas puscadnas,paspampres. Akronim Jamkesmas dan Prepusnas mendapat perulangan fonem /s/ pada sukukata kedua dan suku kata ketiga. Perulangan seperti itu tidak ditemukan dalam kata bahasaIndonesia. Apalagi, pengulangan fonem /s/ tiga kali berturut-turut pada akronim riskesdas danpuskesmas jelas tidak lazim dalam fonotaktik bahasa Indonesia. Sebaliknya, akronim puscadnasdan paspampres pengulangan fonem /s/ terjadi pada suku kata pertama dan suku kata kedua. Disamping itu, akronim paspampres tidak hanya mengulang fonem /s/, tetapi juga fonem /p/ ditiga tempat, yaitu suku kata pertama, suku kata kedua, dan suku kata ketiga. Dengan demikian,pengulangan atau repetisi fonem seperti itu termasuk menyimpang dari fonotaktik bahasaIndonesia.

Fonem /s/ pada suku kata pertama berulang pada suku kata ketiga dan fonem /s/ padasuku kata kedua berulang pada suku kata keempat. Bentuk itu terdapat pada akronim perspebsidan akronim Ditreskrimsus. Selain itu, kedua akronim ini sulit untuk dilafalkan. Hal itudisebabkan jajaran fonem /p/ yang berulang dan fonem penutup suku kata /b/ pada akronimperspebsi adalah sama-sama konsonan hambat bilabial, sedangkan akronim Ditreskrimsus sulitdilafalkan karena gugus konsonan /tr/ dan gugus konsonan /kr/ dipertemukan dan fonem /s/pada suku kata kedua berulang pada suku kata keempat.

Fonem /p/ pada Bapepam, Perpres, dan Propernas adalah fonem yang mengalamiperulangan. Pada akronim Bapepam terjadi perulangan fonem /p/ dan perulangan daerah dancara artikulasi, yaitu fonem /b/ dan /p/ sama-sama hambat bilabial. Akronim Perpres danPropernas berulang fonem /p/ dan fonem /r/. Ketiga akronim itu agak sulit dilafalkan.

Repetisi tidak hanya terjadi pada fonem yang sama, tetapi juga pada daerah dan caraartikulasinya. Contoh akronim yang seperti itu terdapat pada Kopkamtib, pangkopkamtib,pangkobkamtibda, pusdalobpang, puslafpor. Fonem /p/, fonem /b/, dan fonem /m/ berasal daridaerah artikulasi yang sama, yaitu bilabial dengan ketentuan fonem /p/ dan fonem /b/ hambatbilabial, sedangkan fonem /m/ nasal bilabial sehingga perulangan terjadi pada daerah artikulasibibir (bilabial). Contoh akronim yang mengikuti tipe tersebut adalah akronim Kopkamtib,Pangkopkamtib, Pangkopkamtibda, Pusdalobpang, sedangkan akronim puslafpor pengulanganterjadi pada fonem tak bersuara /f/ dan fonem /p/.

Jumlah Suku Kata yang Terdiri Atas Empat Suku Kata LebihJumlah suku kata bahasa Indonesia pada umumnya adalah dua meskipun ada yang lebih daridua. Untuk keperluan telaah akronim fonotaktik, penulis mengelompokkan akronim yang terdiriatas empat suku kata ke atas. Pertimbangan akronim empat suku kata ke atas mengingatbiasanya pelafalannya atau penyebutannya sulit. Dari telaah terhadap data ada beberapa akronimyang demikian (Tabel 6).

Tabel 6. Jumlah Suku Kata yang Terdiri Atas Empat Suku Kata LebihNo. Akronim Jumlah Suku Kata Bentuk Panjangnya1 Babinkumnas 4 Badan Pembinaan Hukum Nasional2 Bapedalda 4 Badan Pengendalian Dampak Lingkungan

Daerah3 Distanbuhut 4 Dinas Pertanian Perkebunan Kehutanan4 Ditreskrimsus 4 Direktorat Reserse Kriminal Khusus5 Mahkumjakpol 4 Mahkamah Agung, Kementerian Hukum dan

Hak Asasi Manusia, Kejaksaan Agung, danPolri

6 Pusdalobpang 4 Pusat Pengendalian Operasi Pembangunan

Sariah

60

No. Akronim Jumlah Suku Kata Bentuk Panjangnya7 Disnakertransos 5 Dinas Tenaga Kerja Transmigrasi dan Sosial8 Dispendukcapil 5 Dinas Kependudukan dan Pencatatan Sipil9 Disperindagkop 5 Dinas Perindustrian dan Perdagangan10 Ikapergabin 5 Ikatan Pekerja Galian dan Bangunan

Indonesia11 Kemenkominfo 5 Kementerian Komunikasi dan Informatika

Akronim yang terdiri atas empat suku kata ke atas dan jajaran fonem yang tidak lazimmengakibatkan akronim yang dibentuk sulit untuk dilafalkan oleh penutur bahasa Indonesia.Akronim Babinkumnas, Bapedalda, Distanbuhut, Ditreskrimsus, Mahkumjakpol, Pusdalobpangadalah akronim yang terdiri atas empat suku kata dengan penjajaran fonem yang tidak lazim.Penjelasannya akronim itu telah dipaparkan pada bagian lain, yaitu pada butir penjajaran fonemdan perulangan konsonan yang sama dalam satu akronim.

Selanjutnya, akronim yang terdiri atas lima suku kata terdapat pada Disnakertransos,Dispendukcapil, Disperindagkop, Ikapergabin, Kemenkominfo. Pelafalafan atau pengucapanakronim yang terdiri atas lima suku kata tersebut tidak luwes dan penjajaran fonemnya tidaklazim, seperti Dispendukcapil dan Kemenkominfo. Penjajaran fonem /kc/ dan fonem /nk/ padakedua contoh tersebut memang tidak terdapat dalam fonotaktik bahasa Indonesia.

Di samping itu, ada beberapa hal yang perlu ditambahkan. Akronim Kowau (KorpWanita Angkatan Udara) dan akronim Seskoau (Sekolah Staf dan Komando Angkatan Udara)adalah akronim yang sulit dilafalkan. Deret vokal /au/ yang sebelumnya didahului fonem /w/bertemu sehingga fonem /w/ diikuti deret vokal /au/ yang juga berbunyi /w/ menjadikanpelafalan akronim tersebut menjadi sulit dan aneh. Kowau yang dilafalkan menjadi Kowaw.Padahal, kata bahasa Indonesia tidak ada yang diakhiri dengan fonem /w/. Penjelasan yangsama juga berlaku untuk akronim Seskoau yang fonem akhir yang dilafalkan dengan fonem /w/.Kondisi lain terdapat pelafalan pada huruf di tengah suku kata, misalnya pada akronim Kowal.Konsonan /w/ pada akronim itu untuk kata wanita. Konsonan itu diapit oleh vokal /o/ dan vokal/a/ (o+w+a). Kenyataannya di dalam KBBI (Kamus Besar Bahasa Indonesia) tidak terdapat satupun lema kata yang mengandung konsonan /w/ diapit oleh vokal /o/ dan vokal /a/ (o+w+a) danjuga (u+w+a). Kalaupun ada, lema itu tentu berasal dari bahasa lain, atau merupakan bentuktidak baku, misalny lema sowang yang merupakan bentuk tidak baku dari soang dan lemakuwalat adalah bentuk tidak baku dari kualat. Hal ini mungkin karena lafal dalam artikulasibahasa Indonesia konsonan /w/ digolongkan sebagai semivokal atau mirip vokal, bukankonsonan murni. Akibatnya akronim Kowal dilafalkan hampir identik dengan lafal koal.Konsekuensinya konsonan /w/ pada Kowal hampir tidak dilafalkan dalam pelafalan kata bahasaIndonesia (kecuali kalau dilafalkan dengan lafal tidak baku). Kemenkeu (KementerianKeuangan) adalah akronim yang diakhiri dengan deret vokal /eu/. Lema bahasa Indonesia tidakditemukan suku kata yang diakhiri dengan deret vokal tersebut. Dengan demikian, akronimdengan deret vokal /eu/ di akhir suku kata tidak lazim dalam bahasa Indonesia.

SIMPULAN

Temuan dari akronim yang tidak fonotaktik dikelompokkan menjadi jajaran fonem yang tidaklazim; dua konsonan berurutan yang sama; konsonan yang tidak bisa menjadi penutup sukukata; perulangan konsonan yang sama dalam beberapa suku kata; jumlah suku kata yang terdiriatas empat suku kata lebih. Akronim yang berfonotaktik tidak lazim umumnya terdapat padabahasan penjajaran fonem, khususnya pada jajaran fonem /md/ pada akronim amdal, jamda,gakumdu, dsb.; jajaran fonem /nm/ pada akronim Binmatkum, menmud, ranmor, dsb.; jajaranfonem /pk/ pada akronim Apkasi, Apkindo, Gapkindo, dsb.; jajaran fonem /pm/ pada akronimApmiso, GAPMMI, IPMI, dsb. Akronim dengan dua konsonan berurutan yang sama terdapat


61

pada konsonan /pp/ Bappebti, konsonan //ss/ Kopassus, /mm/ Jamman, dan /tt/ Unpatti. Akantetapi, konsonan berurutan yang sama didominasi oleh fonem /pp/ pada data akronim. Akronimdengan konsonan rangkap adalah /dh/ pada Pusbadhi, konsonan /ngg/ pada Sultengg, dankonsonan /nd/ pada Unand.

Akronim yang menggunakan konsonan yang tidak bisa menjadi penutup suku kataumumnya adalah akronim berpenutup suku kata dengan konsonan /g/ dan /d/, seperti Balegda,FAGSAM, caleg, paced, pidsus, puskud. Akronim yang menggunakan konsonan penutup sukukata /c/ tidak ditemukan dalam data akronim. Ada akronim yang menggunakan konsonanpenutup suku kata /z/ dan konsonan penutup suku kata /v/, sedangkan dalam kata serapan puntidak ditemukan konsonan /z/ dan konsonan /v/ sebagai penutup suku kata, seperti akronimBaznas dan Pemprov. Perulangan fonem yang sama dalam beberapa suku kata tidak terdapatpada kata bahasa Indonesia, misalnya riskesdas, puskesmas, Ditreskrimsus, dan sebagainya.Jumlah suku kata (silabe) terdiri atas empat suku kata atau lebih, seperti Babinkumnas,Ditreskrimsus, Disnakertransos, Dispendukcapil.

CATATAN* Penulis berterima kasih kepada mitra bebestari yang telah memberikan saran-saran untuk perbaikanmakalah ini.

DAFTAR PUSTAKA

Alwi, Hasan et al. 2003. Tata Bahasa Baku Bahasa Indonesia. Edisi Ketiga. Jakarta: BalaiPustaka.

Alwi, Hasan et al. 2003. Kamus Besar Bahasa Indonesia. Jakarta: Balai Pustaka.

Chaer, Abdul. 2009. Morfologi Bahasa Indonesia (Pendekatan Proses). Jakarta:Rineka Cipta.

Djajasudarma, T. Fatimah. 2010. Metode Linguistik: Ancangan Metode Penelitian dan Kajian.Bandung: Refika Aditama.

Kentjono, Djoko. Ed. 2003. Dasar-Dasar Linguistik Umum. Jakarta: FSUI.

Parera, Jos Daniel. 1985. Pengantar Linguistik Umum: Fonetik dan Fonemik. Ende: Flores.

Pusat Pembinaan dan Pengembangan Bahasa. 2002. Pedoman Umum Pembentukan Istilah.Cetakan ke-4. Jakarta: Pusat Bahasa.

Pusat Pembinaan dan Pengembangan Bahasa. 2002. Pedoman Umum Ejaan Bahasa Indonesiayang Disempurnakan. Edisi Kedua. Jakarta: Pusat Bahasa.

Sudarno. 1990. Morfofonemik Bahasa Indonesia. Jakarta: Arikha Media Cipta.

Sudaryanto. 1985. Linguistik: Esai tentang Bahasa dan Pengantar ke Dalam Ilmu Bahasa.Yogyakarta: Gadjah Mada University Press.

Sudaryanto. 1993. Metode dan Teknik Analisis Bahasa. Yogyakarta: Duta Wacana UniversityPress.


VERBS OF EXCRETION IN TABA

Frederick John Bowden*Jakarta Field Station, Max Planck Institute for Evolutionary Anthropology

Universitas Katolik Indonesia Atma [email protected]

AbstractThis paper describes an unusual pattern of argument marking found in a small number ofTaba verbs, all of which have meanings relating to excretion. The verbs concerned aresio ‘to shit’, mio ‘to piss’, sito ‘to fart’, and hantolo ‘to lay eggs’. In normal usage theseoccur with the excretor argument obligatorily encoded twice, once as a proclitic in thesame way that Actor argument are encoded on other verbs, but also as an enclitic,similar to an Undergoer argument with other verbs. While no clear rationale for whythese four verbs should be treated differently from all the other verbs in the Taba lexicon,they do all express activities that the performers are only in partial control of, and theseverbs may in fact constitute something like a ‘middle voice’ in Taba.

Keywords: argument marking, excretion, proclitic, enclitic, middle voice

AbstrakMakalah ini mendeskripsikan sebuah pola pemarkahan argumen yang ditemukan dalamempat verba bahasa Taba. Semua verba yang dimaksud mempunyai arti yang berhubungandengan ekskresi, yaitu sio ‘buang air besar’, mio ‘buang air kecil’, sito ‘kentut’, dan hantolo‘bertelur’. Dalam penggunaan normal verba tersebut harus diberi dua pemarkah argumen,yaitu proklitik yang berfungsi sebagai argumen pelaku (Actor) dan enklitik sebagai argumenobjek langsung (Undergoer). Meskipun tidak ada alasan yang jelas bagaimana keempatverba tersebut mempunyai pola yang berbeda dengan verba lain dalam leksikon bahasaTaba, semua verba ini menunjukkan bahwa para pelaku hanya mengendalikan sebagian darikegiatan yang dilakukan dan keempat tersebut dapat dikategorikan sebagai middle voicedalam bahasa Taba.

Kata kunci: pemarkah argumen, ekskresi, proklitik, enclitik, middle voice

INTRODUCTIONi

In this paper I want to discuss a small topic in Taba verbal morphosyntax (and argumentstructure) that remains something of a puzzle to me years after I first encountered it. There is avery small class of verbs, the meanings of which all relate to excretion, that share an unusualpattern of morphosyntactic marking, distinct from all other Taba verbs. The verbs involved arethe roots sio ‘to shit’, mio ‘to piss’, sito ‘to fart’, and hantolo ‘to lay eggs’. In normal usagethese occur with the excretor argument obligatorily encoded twice, once as a proclitic in thesame way that Actor argument are encoded on other verbs, but also as an enclitic, similar to, butnot exactly the same in all details, as an Undergoer argument. I have made exhaustive attemptsto find other verbs that might be members of this class without any success: things such asbleeding, sweating, ejaculating, menstruating, etc. etc. are all expressed in a variety of differentways, but not like the class of excretion verbs discussed in this paper. I am afraid I am not ableto share any startling revelations about why these four verbs should behave differently from allother verbs, but I will canvas a few ideas about possible explanations at the end of the paper.I shall approach this topic by first sketching out some basic patterns of morphosyntax foundwith verbs other than those referring to excretion, and then by contrasting these to the patternsfound with the excretion verbs themselves. After we have examined the ‘basic’ cases, I willexamine some valence-affecting affixes and what these do with different kinds of verbs in an

Frederick John Bowden

64

attempt to understand the excretion verbs themselves a little better. I will begin first of all bygiving a brief introduction to the Taba language and its place in the world.

Figure 1. Maluku Utara Showing Locations of Languages


65

Taba was traditionally known in the literature as ‘East Makian’, ‘Makian Dalam’ or‘Makian Timur’. It is classified by Blust (1978) as a South-Halmahera – West New Guinealanguage, along with about 40 or so other languages spoken in the North Maluku and WestPapuan regions. According to Blust, these languages are the closest relations of the Oceanicsubgroup of the Austronesian language family, and together they form a group known as‘Eastern Malayo-Polynesian’. The location of the Taba language is shown in the map in Figureone. In the map, taken from Lewis, ed. (2009), Taba is referred to as ‘East Makian’.

Taba speakers are Muslims and they have had a relatively long period of contact withthe rest of the world. Makian (and a few other small islands off the west coast of Halmahera)were once the only source of cloves in the entire world. After Magellan’s first round the worldvoyage (when his ships visited Makian), Maluku Utara and its spices became a huge center ofinterest for would-be colonial powers out to exploit the potential spoils. Islands in this area wereat the middle of a number of conflicts between various colonial powers throughout theseventeenth century and quite a few loan words from Portuguese, Spanish and Dutch are foundin Taba and other nearby languages. Traditional power in North Maluku was shared betweenfour sultanates, but after the colonial struggles had been played out, the sultanate of Ternate(with the support of the Dutch) became the dominant force, and the Ternatan language was forsome time used as a local lingua franca alongside Malay. Ternatan and all of the languagesspoken in the northern half of Halmahera as well as on the western side of Makian island arenon-Austronesian languages. A number of typological features not common in Austronesianlanguages are found in Taba and presumably some of these have entered the language as a resultof contact with the neighboring non-Austronesian languages. All language of the region(including these days North Maluku Malay) participate in a sprachbund with many sharedfeatures. The major reference on the Taba language is Bowden (2001).

OVERVIEW OF TABA VERBAL MORPHOSYNTAX

Taba is a mixed ‘split-S’ and accusative language. It has basic AVO word-order in transitiveclauses, as illustrated in (1).

(1) Mado npe ubangMado n-pe ubangMado 3sg-make fence‘Mado is making a fence.’

Other orders are also possible. Any argument (core or otherwise) can be shifted to preverbalfocus position so long as it is represented by a full NP and not just a pronoun.

(2) Ubang ya Mado npeubang ya Mado n-pefence REC Mado 3sg-make‘That fence, Mado is making it.’

Ellipsis of readily retrievable participants is common.

(3) Npen-pe3sg-make‘He’s making it.’

Taba has a split-S system for marking the non-human core arguments of verbs. Core argumentscan be divided into either ‘Actors’ or ‘Undergoers’. Actors are cross-referenced by verbalproclitics agreeing in number and person with the actor they index.


66

Table 1. Taba Pronominal FormsFree form Bound form Free form Bound form

1 sg. yakii k- 1pl. (incl.) tit t-1pl. (excl.) am a-

2sg. au m- 2pl. meu h-3sg. i n- 3pl. si l-

(4) Actor = 1sgYak kunak doyak k-unak doI 1sg-know REAL‘I know.’

(5) Actor = 2sgAu mhan appo Tarnateau m-han ap-po Tarnateyou (sg) 2sg-go ALL-down Ternate‘You’re (singular) going to Ternate.’

(6) Actor = 1pl.exclAm atala motor laweam a-tala motor la-wewe.excl 1pl.excl-meet boat sea-ESS‘We (exclusive) met the boat by the sea.’

Verbs may be subcategorised according to how many core arguments of each kind (either Actoror Undergoer) they take.

Table 2.Sub-Classification of Taba Verbs According to Number and Type of Core Arguments

Number of Actor arguments0 1

Number of 0 - Active intransitivesUndergoer 1 Stative intransitives TransitivesArguments 2 Bivalent intransitives Ditransitives

Active Intransitive ClausesActive intransitive clauses are distinguished by the fact that they have one core argument andthat the sole argument is an Actor.

(7) Wangsi lmul do.wang=si l-mul dochild-pl 3pl-return REAL‘The children have returned.’

Stative Intransitive ClausesStative intransitive clauses are distinguished by the fact that they take a single Undergoer coreargument.

(8) Kutukutu nener dakutu-kutu nener dasmall-REDUP sp. fish MED‘These nener (fish) are small.’

Fronting of the Undergoer is quite common with these verbs. However, the Undergoer is nevercross-referenced on the verb.


67

(9) Nener da kutukutunener da kutu-kutusp.fish DIST small-REDUP‘These nener are small.’

Basic Transitive ClausesBasic transitive clauses have verbs which take both an Actor argument and an Undergoerargument.

(10) Banda not yan bakanBanda n-ot yan bakanBanda 3sg-catch fish be.big‘Banda caught a fish that’s big.’

Note the stative intransitive relative clause (yan bakan ‘big fish’) which is not marked in anyway.

Bivalent Intransitive ClausesDouble object intransitive clauses take two Undergoer arguments and no Actor arguments. Theverbs involved in these constructions are always derived applicatives, and one of the twoarguments is invariably fronted.

(11) Hamasik posak niwihamasik posa-Vk niwirice be.boiled-APPL coconut‘The rice is cooked with coconut.’

Ditransitive ClausesDitransitive verbs have three core arguments: an Actor argument and two Undergoers. Allditransitive verbs are derived by applicativization.

(12) Bib npunak kolai pedaBib n-pun-ak kolai pedaBib 3sg-kill-APPL snake machete‘Bib killed the snake with a machete.’

(13) Banda notik yak yanBanda n-ot-ik yak yanBanda 3sg-catch-APPL me fish‘Banda gave me some fish.’

VERBS OF EXCRETIONAs indicated in the introduction, excretion verbs normally occur with both proclitics and encliticscross-referencing the excretor, unlike most active verbs which occur only with proclitics. Theforms used to cross-reference the arguments of excretion verbs are listed in table 3.

Table 3. Cross-referencing of Intransitive Excretion Verbs1sg k—k / k—yak 1pl.incl t—tit

1pl.excl a—m2sg m—u 2pl h—meu3sg n—i 3pl l—si

Illustrative examples are given below.


68

(14) Buang nciwiBuang n-sio-iBuang 3sg-shit-3sg‘Buang did a shit.'

(15) Yak kmiok do / Yak kmioyak doyak k-mio-k do yak k-mio-yak do1sg 1sg-piss-1sg REAL 1sg 1sg-piss-1sg REAL‘I had a piss.’

(16) Si lcitosisi l-sito-si3pl- 3pl-fart-3pl‘They farted.’

Note the formal similarity with pronominal forms and also note that object pronouns can be (butneed not be) encliticised. Note that the forms are not exactly the same as the cliticised objectforms in all person and number combinations, but that they seem to be pretty obviously derivedfrom such forms historically.

VALENCE-AFFECTING MORPHOLOGY

A variety of affixes are available for deriving verbs of various kinds. These affixes are listedbelow. We will not examine all of these in detail, but simply look at the behavior of a few,contrasting what happens with the excretion verbs and the other verbs, and seeing if there areany clues here that might help explain the aberrant nature of the excretion verbs themselves.Let us first note that although all of the affixes listed can have valence-affecting functions, theycan also perform some other functions, such as marking ‘intensity’.

Table 4. Valence Affecting Affixesha- valence increasing

prefix - ‘causative’ derives transitive verbs from active intransitives derives active intransitive verbs from stative verbs derives active verbs from a variety of other word classes ‘intensive’ marker

-Vk valence increasingsuffix - ‘applicative’

derives ‘bivalent intransitive’ verbs from stative verbs derives transitive verbs from active intransitives derives ditransitive verbs from transitive verbs ‘intensive’ marker

-o valence increasingsuffix - ‘applicative’

derives ‘bivalent intransitive’ verbs from stative verbs derives ‘process oriented’ stative verbs from unmarked statives derives transitive verbs from active intransitives derives ditransitive verbs from transitive verbs

ta- valence decreasingprefix - ‘passivizing/detransitivizing’

derives stative verbs from transitive verbs derives stative verbs from active intransitive verbs

‘Causative’ Prefix ha-The most common functions of ha- are deriving transitive verbs from active intransitives, andderiving active intransitives from other kinds of roots.

(17) a. Paramalam nmot.paramalam n-motlamp 3sg-die‘The lamp has gone out.’ (lit. ‘the lamp has died’).


69

b. I namot paramalam.i n-ha-mot paramalam3sg 3sg-CAUS-die lamp‘He turned the lamp off’ (lit. ‘he made the lamp die’).

(18) a. Ni calana kudaPOSS trousers black‘Her trousers are black.’

b. John nakuda sama lo JulJohn n-ha-kuda sama lo JulJohn 3sg-CAUS-black same as Jul‘John is black like Jul.’

Example (18) illustrates the most common non-causative use of ha- which is to derive an Actororiented verb from an Undergoer oriented root to enable the verb to be used with a humanargument. Since excretion verbs are cross-referenced by a proclitic, the Actor orientedderivation illustrated in (18) is not possible with excretion verbs, but we might imagine a trulycausative derivation in which we might derive verbs which could describe situations such as amother telling her child to go to the toilet. Causative derivations of excretion verbs are notfound with this kind of meaning, as illustrated in (19). Example (20) without any cross-referencing cannot be used with a causative meaning either.

(19) * Ni mamasi nasiwi Buangni mama-si n-ha-sio-i Buang3sg.POSS mother-PL 3sg-CAUS-shit-3sg Buang‘His mother made Buang shit.’

(20) ? Ni mamasi nasio Buangni mama-si n-ha-sio Buang3sg.POSS mother-PL 3sg-CAUS-shit Buang* ‘His mother made Buang shit.’

Another function of ha- prefixation with active verbal roots is to indicate that an activity wasperformed with some degree of intensity.

(21) a. Tit twonga maliling yatit t-wonga maliling ya1pl.incl 1pl.incl-stay.awake.all.night night up‘We stayed awake all last night.’

b. Tit tawonga maliling yatit t-ha-wonga maliling ya1pl.incl 1pl.incl-CAUS-stay.awake.all.night night up‘We stayed awake all last night.’

The prefix can also be used with this reading with excretion verbs.

(22) Ismit nasitoiIsmit n-ha-sito-iIsmit 3sg-CAUS-fart-3sg‘Ismit did a big fart.’

The Applicative -VkWith ‘normal’ verbs, this suffix is used to add applied objects of various kinds, most commonlyinstruments, but also themes and recipients. It can derive bivalent intransitives from stativeverbs, transitives from active intransitives, and ditransitives from transitive stems. Example (23)


70

shows the derivation of a bivalent intransitive verb from an Undergoer oriented intransitiveverb.

(23) a. Ubang bulangubang bulangfence be.white‘The fence is white.’

b. Ubang bulngak cetubang bulang-k cetfence be.white-APPL paint‘The fence was whitened with paint.’

Example (24) illustrates the derivation of an instrumental ditransitive verb from a transitivestem.

(24) a. Ahmad npun kolaiAhmad n-pun kolaiAhmad 3sg-kill snake‘Ahmad killed a snake.’

b. Ahmad npunak kolai pedaAhmad n-pun-ak kolai pedaAhmad 3sg-kill-APPL snake machete‘Ahmad killed the snake with a machete’

In example (25) we can see the derivation of a simple transitive verb from an Actor orientedintransitive stem.

(25) a. i nggaleiti n-galeit3sg 3sg-burp‘He burped.’

b. i nggaleitik susui n-galeit-ik susu3sg 3sg-burp-APPL milk‘He burped up milk.’

Note that the applicative suffix also occasionally has an intensive reading, similar to that foundwith some ‘causatives’.

(26) a. Kaidisk-ha-idis1sg-CAUS-spit‘I spit (making no noise).’

b. Kaidcikk-ha-idis-k1sg-CAUS-spit-APPL‘I spit (making a lot of noise).’

Spitting publically is quite common practice amongst Taba speakers, but polite spitting isalways done silently. To make a large noise while spitting is seen as insulting, and theapplicative form is used here to signal that spitting is performed with such an insulting noise.Of course the example in (26) can be read literally as ‘spit with noise as theme or companion’.The way the applicative suffix works with excretion verbs parallels the last couple of examples:they allow us to express whatever was excreted as an object of the verb, parallel to (25) abovewhere -ak allows a theme to be introduced:


71

(27) Ksiak haluak-sio-ak halua1sg-shit-APPL halua‘I’m shitting halua.’ [halua = ‘toffee made from palm sugar’]

The Detransitivizing Prefix ta-The ‘passive’ prefix ta- most commonly derives agentless stative verbs from transitive verbs.The sole derived argument in these constructions is the root Undergoer. In this kind ofderivation it could perhaps also be called a ‘resultative’ prefix, because the effect of itsapplication is to focus semantically on the resulting state of the patient after a process of somekind has reached its end.

(28) a. underivedi nbes niwii n-bhes niwi3sg 3sg-husk coconut‘She husked the coconut.'

b. derivedniwi tabhes doniwi ta-bhes dococonut DETR-crack REAL‘The coconut has been husked.'

(29) a. underivedMale tcakal boamale t-sakal boamust 1pl.incl-smash door‘We had to smash down the door.'

b. derivedBoa tasakalboa ta-sakaldoor DETR-smash‘The door was smashed down.' (either intentionally or not)

The ta- detransitivizing prefix can also be used with the verbs of excretion, but whenever itoccurs, it must co-occur with the -Vk applicative. These constructions are used to indicate thatthe person excreting is totally incapable of not excreting, as when for instance they might havediarrhea.

(30) Tasiak yakta-sio-ak yakDETR-shit-APPL 1sg‘I’ve got the shits.’

I have never actually heard anyone mention the theme supposedly licensed by the -Vk applicativewhen this construction has been used spontaneously, but it is possible to do so, and people tell methat they understand the theme as the cause of the uncontrollable excretion as in (31).

(31) Tasiak yak niwita-sio-ak yak niwiPASS-shit-APPL 1sg coconut‘The coconut has given me the shits.’ [lit. ‘I’m uncontrollably shitting coconut’]iii

This sentence is formally a bivalent intransitive, analogous to sentence (23) above which isrepeated as (32).


72

(32) a. Ubang bulangubang bulangfence be.white‘The fence is white.’

b. Ubang bulngak cetubang bulang-k cetfence be.white-APPL paint‘The fence was whitened with paint.’

Although (31) is formally a bivalent intransitive verb, as already mentioned such sentences withexcretion verbs are rarely offered with the instrumental argument overtly coded.

The Applicative –oIn addition to the –Vk applicative discussed above, Taba also has an applicative suffix –o whichusually introduces a locative argument to the frame of the derived verb. It is illustrated in itsnormal use in (33).

(33) a. I nyogi n-yog3sg 3sg-jump‘She jumps.’

b. I nyogo meseli n-yog-o mesel3sg 3sg-jump-APPL wall‘She jumped on the wall.’

It appears that the excretion verbs can also be used with the –o applicative to derive transitiveverbs where the object of the verb is a locative goal (or where the excretia ended up). The factthat the –o applicative is attached to excretion verbs is somewhat obscured in Taba because allof the excretion verb roots end in the segment o and thus there is no overt sign of the applicativesuffix having been added part from the disappearance of the cross-referencing suffix and thepossible appearance of a locative goal. In example (34) the locative goal is overt, but in (35) it isellipsed.

(34) Buang ncio ni calanaBuang n-sio-o ni calanaBuang 3sg-shit-APPL POSS trousers‘Buang shitted his trousers.'

(35) Q: Mesel tasakal hapwe A: Iswan ncitomesel ta-sakal ha-pu-e Iswan n-sito-owall DETR-collapse CAUS-what-FOC(how) Iswan 3sg-fart-APPL‘How come the wall collapsed?’ ‘Iswan farted on it.’

One further example of the locative applicative being used with an excretion verb is suppliedbelow. In (36), the verb hantolo ‘to lay eggs’ is shown with double cross-referencing, and in(37) it is shown with the locative applicative suffix.

(36) Nantolin-han-tolo-i3sg-INCH-egg-APPL‘It lays / they lay an egg / eggs.’


73

(37) Nantolo gutn-han-tolo-o gut3sg-INCH-egg-APPL nest‘It lays / they lay an egg / eggs in the nest.’

DISCUSSIONWhy should it be that this small class of Taba verbs is marked differently from all others in thegrammar of the language? One thing that can be said about all of these verbs is that people who‘shit’, ‘piss’, and ‘fart’ as well as the creatures which ‘lay eggs’ have a certain degree of controlover the process, but to a large degree these processes are things that happen to you, and theexcretor can be seen to some degree as both volitional actor and involuntary undergoer. Thesplit-S system by which Taba clausal grammar is organized builds this semantic distinction intothe core of the grammar, so perhaps we should not find it so unusual if a few verbs whosearguments are not easily distinguished in terms of their volitionality should be marked in anunusual way.

A number of other languages use double agreement or something similar with smallclasses of verbs. Kokota, in the Solomon Islands (Palmer, 2005) has double agreement for asmall set of verbs including ‘laugh’, ‘hiccup’, ‘yawn’, ‘be angry’ and ‘’be pleased’. Palmersuggests these verbs are doubly marked because their arguments are subjected to forces overwhich they have little control. Kemmer (1993:53ff) also notes languages which have ‘middlevoice’ constructions in which ‘body action middles’ often with similar semantic elements to theTaba constructions also have peculiar properties that may derive from these same semanticelements. Finally, we can observe something similar with the standard Indonesian word tertawa‘laugh’ or its colloquial Jakarta equivalent ketawa where the initial prefixes are generally usedto signal something accidental.

Why these particular verbs and not others with similar semantics are singled out in Tabamust remain a mystery though. Perhaps some sort of cultural explanation is necessary. I wouldlike to end this paper by making a few remarks about what might be called the ‘culture ofexcretion’ Taba speakers always appeared to be both quite discrete about their own excretionand very amused by it when excretion became the topic of a joke. Many of the examples givenabove were taken from natural conversations in which the speaker was making fun of someoneelse. The following postscript is a transcription of the telling of a Taba riddle, and the answer toit that plays with this sort of humor. It also has many naturally occuring examples of theexcretion verb sio ‘shit’ and some of its derivatives. The text consists of (a) the asking of ariddle, and (b), the very detailed explanation of why the author's response (that he would elect togo to sleep first) was the wrong choice. Not transcribed are the great peals of raucous laughterfrom the audience as they listened to the detailed explanation of excretory catastrophe found inpart (b).

Postscript: A Taba Riddle(a) John ni we mhonas... nim we nalusa mhonas. Nim

John ni we mhonas... ni-m we n-alusa mhonas. ni-mJohn POSS leg sore... POSS-2sg leg 3sg-say sick POSS-2sgpappuko me nalusa 'mhonas'. Bingo namolam. Ulo nmau nhanpappuko me n-alusa mhonas. Bingo n-amolam. ulo n-mau n-hanknee well 3sg-say sick. Stomach 3sg-hungry heart 3sg-want 3sg-goPoto pope nmau nhan nciwi. Sumo nalusa 'khan'. MtoPoto po-pe n-mau n-han n-cio-i. Sumo n-alusa k-han. Mtoanus down-ESS 3sg-want 3sg-go 3sg-shit-3sg mouth 3sg-say 1sg-go eye


74

nuyak, poyo mhonas, ulo nmau nhan... mtumo e loe?n-uyak, poyo mhonas, ulo n-mau n-han, m-tumo e lo-e?3sg-tired head sick heart 3sg-want 3sg-go 2sg-follow FOC where-FOC?

‘John, your leg is sore. Your leg says 'I’m sick'. Your knee says it’s sick. Your stomach ishungry. Your heart wants to go somewhere. Your anus down there wants to go for a shit. Yourmouth says 'I'm going (I’m hungry)'. Your eyes are tired, your head is sick, your heart wants togo somewhere, which one do you follow (first)? Your anus down there is about to shit itself.Which one do you follow (first)?’

(b) Poto pope me tasiaki. John ni suka nhantuliPoto po-pe me ta-sio-ak-i. John ni suka n-han-tuliAnus down-ESS but PASS-shit-APPL-3sg John POSS desire 3sg-go.sleepsedangkan ni klolo nparenta nhan... Han akno nisedangkan ni klolo n-parenta n-han... han ak-no niwhile POSS NOM-inside 3sg-stop 3sg-go go ALL-there POSSdalawaci de ncio ni calana de namliakdawalat-si de n-sio-o ni calana de n-amlih-akgirlfriend-PL PURP 3sg-shit-APPL POSS trousers PURP 3sg-laugh-APPL 1pl.incl.tit Ncuka te. Male sio malai han. Idia. Sio okik, malaitit N-suka te. Male sio malai han. I-dia. Sio okik, malai1pl(incl.) 3sg-like NEG must shit then go DEM-DIST shit finish thennhantuli malai nhan ronda. Tasiak nit calana?n-han.tuli malai n-han ronda. Ta-sio-ak nit calana?3sg-go.sleep then 3sg-go walkabout PASS-shit-APPL 1pl.incl.POSS trousersTatés! Polo tese, cio calana, cioTatés! Polo te-se t-sio-o calana t-sio-ocrazy if NEG-POT 1pl.incl-shit-APPL trousers 1pl.incl-shit-APPL mattressjok berarti masure te.jok berarti masure te.mattress means good NEG

‘But your anus down there is shitting itself! And John you want to go to sleep while yourinsides have stopped working? If you go to your girlfriend's place you'll shit your trousers andshe'll laugh at you!!! She won't like it! You have to shit first and then you go do other things.That’s how. Once you’ve finished shitting then you go to sleep then you go for a walk. Shit ourtrousers? Crazy!!! If we don't shit our trousers, we'll shit on the mattress. It’s no good!’

NOTE*I would like to thank an anonymous reviewer for very helpful comments on the earlier draft.i The following abbreviations are used in this paper: 1 first person; 2 second person; 3 third person; ALLallative; APPL applicative; CAUS casusative; DETR detransitivizing; DIST distal; ESS essive; exclexclusive; FOC focus; INCH inchoative; incl. inclusive; MED medial; NEG negative; pl plural; PLplural; POSS possessive; POT potential; PURP purposive; REAL realis; REC reciprocal; REDUPreduplication; sg singular.ii Note that in some dialects (Mailoa, Kayoa, Peleri) the 1sg. independent pronoun is lak.iii Coconut is viewed locally as a cure for constipation.


75

REFERENCESBlust, Robert A, 1978. ‘Eastern Malayo-Polynesian: a subgrouping argument’. In S.A. Wurm and

Lois Carrington, eds. Second International Conference on Austronesian Linguistics:proceedings, 181-234. Pacific Linguistics C-61. Canberra: Pacific Linguistics.

Bowden, John, 2001. Taba: description of a South Halmahera language. Canberra: PacificLinguistics.

Kemmer, Suzanne, 1993. The Middle Voice. Typological Studies in Language, 23. Amsterdam:John Benjamins.

Lewis, M. Paul (ed.), 2009. Ethnologue: Languages of the World, Sixteenth edition. Dallas, Tex.:SIL International. Online version:http://www.ethnologue.com/.

Palmer, William Dennis, 2009. Kokota Grammar. Honolulu: University of Hawaii Press


THE ROLE OF CULTURE IN THE TRANSLATION PROCESSTHROUGH THINK-ALOUD PROTOCOLS

Julia Eka RiniUniversitas Kristen [email protected]

AbstractCultural differences have been a major focus in translation. This study investigates twoaspects; first, the translation process of culture-bound words and second, the strategiestaken by two translators. These translators are English Department students and theyhave different cultural backgrounds, Moslem Javanese and Buddhist Chinese. Each ofthem has to translate the same four texts: one text whose cultural background both ofthem are familiar with, another text whose cultural background both of them are notfamiliar with and two other texts. Out of the two texts, only one text has a familiarcultural background to one of the translators. The method used to investigate what washappening in the translators’ mind is think-aloud protocols. Two points can be concludedfrom this study. One, translators’ cultures do play a role in their consideration ofchoosing the words they use. Two, translators generally use the strategy of culturalsubstitution if the culture is nearly the same, but they use the strategy of using a neutralword or paraphrasing by related words when the culture is different.

Keywords: culture-bound words, translation strategy, cultural differences

AbstrakPerbedaan budaya telah banyak mendapat sorotan dalam penerjemahan. Yang dikajidalam penelitian ini ialah pertama, proses penerjemahan kata-kata yang disebabkanoleh perbedaan budaya dan kedua, strategi yang diambil oleh penerjemah. Keduapenerjemah ialah mahasiswa Jurusan Bahasa Inggris, dan keduanya mempunyai latarbelakang budaya yang berbeda. Yang satu berlatar budaya Jawa dan Islam dan yanglain Tionghoa dan Budha. Masing-masing menerjemahkan empat teks yang sama: satuteks mempunyai latar belakang budaya yang dikenal keduanya, satu teks lainnyamempunyai latar belakang budaya yang tidak dikenal keduanya, dan dua teks yang lainmempunyai latar belakang budaya yang hanya dikenal salah satunya oleh masing-masing penerjemah. Penelitian ini menggunakan metode analisis protokol untuk menelitiapa yang terjadi dalam pikiran penerjemah ketika sedang menerjemahkan. Ada dua halyang diperoleh dari penelitian ini. Pertama, latar belakang budaya penerjemahmemegang peranan dalam memilih kata-kata yang mereka gunakan. Kedua, strategiyang berbeda digunakan kedua penerjemah dalam menerjemahkan kata-kata yangmengandung budaya. Jika budaya hampir sama, strategi yang digunakan ialahmenggunakan kata yang mempunyai budaya yang mirip; namun, jika budaya berbeda,strategi yang digunakan adalah menggunakan kata yang netral atau memparafrasadengan kata-kata yang masih ada hubungannya dengan kata tersebut.

Kata kunci: kata-kata budaya, strategi penerjemahan, perbedaan budaya

INTRODUCTION

The term ‘translation’ can refer to translation as a process and also as a product. The researchfocus on translation product (the translation text) is already common in Indonesia, but the focusto the translation process is not that popular. However, in other countries “there has been anincreasing interest in studying the translation process since the mid-1980s” (Li 2004:301). Withthe shift of focus to the translation process, the translators also become more important in thefocus of research. The important role of translators can be seen in what experts in translation say

Julia Eka Rini

78

about it. Wills (2004:3), for example, says, “A translator is supposed to be a bridge betweenlinguistic and cultural communities, but at the same time is different from both the source-textauthor and the target-text reader(ship).” Two important aspects in Will’s statement about theimportant role of translators in translation are in the area of language and also culture. Since therole of translators cannot be ignored in translation process, this study tries to describe how thethree aspects of translation process—language, culture, and translators—are related. This studytries to describe how translators of different cultures deal with language. Culture, as defined byBeamer and Varner (2001:3), is the coherent, learned, shared view of a group of people aboutlife’s concerns that ranks what is important, furnishes attitudes about what things areappropriate, and dictates behaviour. This general definition underlies the concept of culture inthis study, but a more specific definition from Appelbaum and Chambliss (1997) is used as aguideline to choose the words or culture-bound units in this research that are taken as databecause it is more systematic and concrete. According to them, sociologists and anthropologistsdifferentiate two aspects of culture that are different but interrelated aspects of human culture,material culture and non-material culture. “Material culture includes all the physical objectsmade by the members of a particular society to help shape their lives” (1997: 63). This includestools and technologies to make goods, the goods consumed, the place of worship, the offices orstores and the cities or towns where people live. “Non-material culture consists of all thenonphysical products of human interaction, that is, the ideas shared by people in a particularsociety. This includes languages, values, beliefs, rules, institutions and organizations” (1997:64). This study deals with difficult words, terms, or expressions concerning culture and they arereferred to as culture-bound units.

Concerning culture in translation, Baker (1992) lists a number of common problems,such as: 1) the source language concepts are not lexicalized in the target language, (2) thesource-language is semantically complex, (3) the source and target languages make differentdistinctions in meaning, (4) the target language lacks a superordinate, (5) the target languagelacks a specific term, (6) differences in physical or interpersonal perspective, (7) differences inexpressive meaning, (8) differences in form, (9) differences in frequency and purpose of usingspecific forms, (10) the use of loan words in the source text, and (11) culture-specific concepts,that is, “the source language word may express a concept which is totally unknown in the targetculture. The concept in question may be abstract or concrete; it may relate to a religious belief, asocial custom, or even a type of food” (1992:21). Some common strategies to overcome theproblems are (1) translation by a more general word (superordinate), (2) translation by a moreneutral or less expressive word, (3) translation by cultural substitution, (4) translation using aloan word plus explanation, (5) translation by paraphrase using a related word, (6) translation byparaphrase using an unrelated word, (7) translation by omission, and (8) translation byillustration. Baker (1992:42) admits that the list is not limited to what she has described, andfurther studies of the strategies—how translators cope with the culture-bound units—areencouraged.

The solutions of dealing with special words which are ‘culturally bound,’ according toDuff (1981:26), may differ greatly, although the problems of all translators of literary, general,or technical texts are the same. Therefore, the strategies taken by the translators are worthdiscussing. Some common solutions, according to Duff (1981), are (1) the word is retained in itsoriginal form and no explanation is given, (2) the word is retained in its original form, witheither a literal translation in brackets, an official or accepted translation in brackets, or anexplanatory footnote, (3) the word is never mentioned in its original form, (4) the same with theprevious, but the translator expands the text in order to convey all associations, (5) differenttranslations of the same word are used, because the target language differentiates more than thesource language, (6) part of the source language is omitted, (7) a target language expression isgiven, although it is not derived from anything in the text. This study, then, tries to portray


79

more specifically, what is happening in the minds of translators of different cultures whendealing with culture-bound units in the process of translation.

What is happening in the minds of translators is worth paying attention to, sincetranslation, according to Hatim and Munday (2004:36), happens not only linguistically, but alsocognitively as the translator works on a translation. In translation theory that describes thetranslation process in stages, the stage that happens in the translators’ minds is the stage oftransfer; the meaning or message obtained from the analysis is transferred from the Source Text(ST) to the Target Text (TT). In the translation process there are four stages (Suryawinata 2003:19), which can happen quickly or slowly, and either once or repeatedly, depending on theintensity of the difficulties encountered by the translators. The following is the completediagram of translation process.

Figure 1.Translation Process Adapted from Nida and Taber Modified by Suryawinata

In the stage of analysis, before the stage of transfer, the translators analyze the text to get thetextual or contextual meaning of the text. In the stage of restructuring, translators write the TT,maintaining the equivalent content, meaning, and message of the ST. The stage of evaluationand revision is where the translators evaluate the TT (the translation) to determine whether ornot it is the same as the ST. If it is not the same, then the TT is revised and the process isrepeated from analysis. During the translation process, translators can pay attention to onesentence, one clause, one group of words or even one word, referred to as translation units. Theunits discussed in this study can be words, phrases or sentences.

One of the few available means to know what happens in the translator’s mind in thestage of transfer is verbal reports or verbal protocols or think-aloud. This method was first usedin psychology and cognitive science in the early twentieth (Brown and Rogers 2001:54). Toexplain what is verbalization or think-aloud, it is useful to use what Brown and Rogers(2001:53-54) say about it; it is like doing mental (or even written) arithmetic of 45 times 52.They explain as follows: In a simple language, verbalizing or think-aloud is saying what one isthinking so that others can hear it.

Later, this method is also used in language research. Verbal protocols ask subjects toverbalize or tell their thought processes when they are involved in processing language. McKay(2006) cites from Brown and Rogers some principles that should be given attention while

Julia Eka Rini

80

conducting verbal reports: 1) verbal reports should occur either while the activity is occurring oras soon as possible afterward; 2) because verbalization needs additional demands, subjectsshould be allowed to use their first language (p. 61); 3) researchers should be as unobtrusive aspossible; researchers should take notes on both nonverbal and verbal behaviour; and 4) verbalreports cannot be used to report automatic thought processes. Some procedures to follow inconducting verbal reports in language rsearch are as follows:

1) provide students with a practice activity;2) give simple directions;3) be as unobtrusive as possible;4) ask subjects to report their thought processes at particular points in the text after they have

read the text;5) do not ask leading questions;6) record the session;7) and pay attention to nonverbal behaviour.

Beginning in 1980’s this method is also used in translation research. The name of thismethod is then recognized as Think-aloud protocols (TAPs) as the name of the transcriptions ofverbalizations of thoughts that subjects are instructed to produce while carrying out a translationtask (Bernardini 2000). Verbalization in translation means saying everything that is happeningin one’s mind while translating. For example, when a translator encounters a difficult word, inher/his verbalization, he might say, “What does this word mean? Dictionary. Let’s see. Firstmeaning. No. It’s not that. Second, third. Ah this one, yes.” What the translator is saying isrecorded and later on, transcribed. The transcription is also called TAPs.

The two ways of doing verbal reports are introspective and retrospective. In translationresearch, the retrospective one is done immediately after translation. After a translator hasfinished translating a text, s/he is asked to describe what was going on during the translating ofthe text. What s/he says about the process of translating is recorded and later on, transcribed.This method is considered less appropriate for this research because the chance of forgettingwhat actually happens is bigger. The one chosen for this research is the introspective report orTAPs. In this way the recording is done while the translator is translating a text. Everything thatthe translator is saying that s/he is thinking is by being recorded; ideally, the thinking, sayingand the recording happens simultaneously. The recording is later transcribed and thetranscription is also called think-aloud protocols (TAPs).

However, this method, according to Bernardini (2000), has limits because it is highlyinfluenced by individual differences in terms of personality, personal history, capacity toverbalise thoughts, and attitude towards the task. Additionally, it is not yet proven whether longcomplex methods could be accessed and reported on in the same way as short problem-solvingtasks. However, since this is the most probable way to know what is going on inside someone’smind, it is one good tool to use. In this research these points are already tackled in themethodology of this research. The following points were carefully considered in themethodology: the criteria for the subjects chosen for this research, the background of thesubjects to determine their suitability for this research, the texts chosen to be translated by thesubjects, and the training given to the subjects to make them accustomed to TAPs.

According to (Bernardini (1999), the major concerns of researchers using TAPs intranslation research are translation strategies, attention units, automaticity of processing andaffective factors. In her review of translation strategies, Bernardini (1999) quotes eleven‘problem indicators’ of translation problems and translation strategies according to Krings(1986): (1) the subjects’ explicit statement of problems; (2) the use of reference books; (3) theunderlining of source-language text passages; (4) the semantic analysis of source –language textitems; (5) hesitation phenomena in the search for potential equivalents; (6) competing potential


81

equivalents; (7) the monitoring of potential equivalents; (8) specific translation principles; (9)the modification of written target-language texts; (10) the assessment of the quality of thechosen translation; and (11) para-linguistic or non-linguistic features. Early TAP studies havebeen concerned with classification of translation strategies and differences between professionaland non-professional strategies. Regarding translation units, she discusses ‘unmarkedprocessing,’ which refers to unproblematic sections of the protocols in which a subjectverbalizes fluently while reading or writing. Marked processing begins with a problem indicatorand ends with a solution to the problem or an indication that the problem is temporarilyabandoned. Those theories are used by the writer to conduct this research, in determining thedata collection and data analysis.

METHODS

Participants joining this research were students who joined the translation class of EnglishDeparment of Petra Christian University. After considering the cultural backgrounds of severalstudents, two students were chosen to be the subjects in this research based on the ethnicity andreligion. One student is a Moslem and Javanese; the other one is a Buddhist and Chinese. Thetwo students fulfilled the criteria of the subjects in this research. The first criteria was that theyhad taken three translation classes. Before joining the third class of translation, they had takenall structure, reading and writing classes. It was assumed that they had had enough practice intranslation so that it could be assumed that their difficulties in translation were not caused bylack of practice or lack of ways of dealing with difficulties concerning language (English) orstructure. Second, they were able to explain their reasons when they were asked to explain theassignments in front of the class in the previous classes of translation; this was important inchoosing the subjects because if they were not good at explaining, then it would also be difficultto verbalize their thoughts while doing the translation.

Before doing the real think-aloud protocols, both students had been given a practice oftranslating a text with verbalization. The text used in the practice was different from the textsused in the research, but the level of difficulty was nearly the same. Before they translated thetext, they were given some explanation how to do the verbalization and to record theirverbalization. When each of them was doing the practice, they were also equipped withdictionaries. After they finished translating the texts and recording, the recording was checkedto see whether they did it in the way they were expected, namely verbalizing what they werethinking. After the practices and the checking, the two students were asked whether they wantedto continue taking part in the research and they agreed.

The two students (henceforth referred to as translators) were asked to translate the samefour texts. Two stories were taken from the same book—entitled Who Is the Most Talkative ofThem All? Stories for Language Teacher Education (1996)—so that the language difficulty ismore or less the same. One text was of familiar culture for both translators, entitled “The Farmerand the Rice Plants.” Both translators are Indonesians and therefore, farmers and rice plants arenot strange for them. Another text was of unfamiliar culture for both translators, which wasentitled “How Do You Shower a Bride.” Both translators were not familiar with the culturespoken in the text (bridal shower). The other two texts were familiar to one translator and not tothe other. The Buddhist text was taken from a book entitled A Still Forest Pool: The InsightMeditation of Achaan Chah, (1985); the title of the text was “The Real Magic”. The Islamic textentitled ”She Had True Faith” was taken from http://www.batkhela.com/islam/story7.shtml.

In doing the translation of the four texts, they were also equipped with dictionaries. Thetwo translators determined their own schedules. Each did the translation and recording ondifferent days, one text at a time. The verbalization was recorded in C-90 cassettes. Eachtranslator pushed the recording button when they were ready to translate and pushed it again

Julia Eka Rini

82

when they finished translating. The C-90, not C-60, cassettes were used because one side of C-90 lasted 45 minutes; this minimized the risk of the unrecorded verbalization.

The following is an example of a transcript and how the analysis is done. The transcriptbelow is an example taken from Translator A. The underlined parts are the ones he writes down;others are just what he says. The transcript below has been divided into six parts (indicated bynumber 1-6). Verbalization can be divided into might include utterances that are not related tothe translation process of the text, but it is transcribed in the protocol (Part 1). Part 2 is thereading of the English text before the translation. The woman’s voice might be his classmategreeting him. Part 4 is the process of translating the text with dictionary checking. The examplewritten here is shortened because the complete one is very long. Part 5 is the utterances that arenot related to the translation, except when he said that he has finished translating. The last part isreading the Indonesian translation.

1) {ehmm sodok ngantuk/ tap: ndak papa/aku milih kelas 206/ b 206/ Ini aku duduk sebelahejendela/ jadi nek bosen ngeliat ijo ijo seger/ apalagi sambil bau minyak angen/ ini aku lagibau minyak angen/ jadi biasane nek aku nulis nek ndak pake minyak angen ya/ nek pakeminyak angen ngantuke ilang/ini apa ya?/ The farmer and the rice plants/sek sek sana/ thefarmer and the rice plants/ tak bacae ae sek/Mbacae sambil mba- bau minyak angin ya/}

2) {the farmer and the rice plants/there once a farmer who always wished that the rice in hisfield would grow more quickly/ the rice like any other crop/ takes time to grow and cannotbe hurried/ the farmer lost patience with waiting/ and thought of a plan to make the ricegrow more quickly/ he ran to the field and pulled every one of the rice plant /plant/ ee/just alittle bit higher/he was tired out when he came home but very pleased with himself/ what ofa day/ I’ve worked so hard he said to his family/ but at least I know that the rice plants are alittle bit higher/ when his son heard that the rice plants had/ grown taller/he ran to the field totake a look/instead of finding taller/ healthy rice plants he found that}

3) {Woman: /halo/}4) {bau... bau minyak angin enak/ iya, e/ the farmer and the rice plants. [clears throat]

ee/seorang…petani dan...tanaman.. padi..nya. There was! Once a farmer who alwayswished that the rice in his fields would grow more quickly. Pada.. suatu.. ketika.. emm.. initangan kananku megang bolpen tangan kiriku megang minyak angin, jadi.. kadang kalo akuberhenti nulis.. aku bau minyak angin ini lagi. M enak baune. Pada suatu ketika/ adaseorang/ petani/ mm/ yang selalu/ berharap/ ee/that the rice in his fields/ agar tanamanpadi/ field/ field itu apa ya?/ field itu kalo ga salah anu/ apa? lahan/ di lahannya/dapattumbuh dengan/ lebih.. cepat] ….

5) {[sighing] ya tanaman hijau menyegarkan mata/ enak kok memang duduk sebelahe jendela/nek mata sepet /sediluk/ ke luar./ yah selesai}

6) {sek sek sek sek sek/ cek lagi/ ee seorang petani dan tanaman padinya/ pada suatu ketikaada seorang petani yang selalu berharap agar tanaman padi di lahannya dapat tumbuhdengan lebih cepat/ akan tetapi padi seperti halnya tanaman budidaya dan/ lainnyamembutuhkan waktu untuk tumbuh dan tidak dapat dipercepat/ petani tersebut kehilangankesabarannya karena menunggu/ dan ia berpikir sebuah ide untuk membuat tanaman paditumbuh/ lebih cepat/ dan ia memikirkan sebuah ide untuk membuat tanaman padi agardapat/ agar dapat/ ditambahi/tumbuh lebih cepat/ petani tersebut pergi ke lahannya danmenarik setiap tanaman padinya sedikit lebih tinggi/ petani ini ke lelahan ketika ia tiba dirumahnya/ tetapi puas dengan dirinya/ dia berkata pada keluarganya/ hari yangmelelahkan/ aku telah bekerja dengan keras/ tetapi setidaknya aku tahu/ bahwa tanamanpadi telah tumbuh sedikit lebih tinggi/ karena anak laki-lakinya mendengar/ bahwatanaman padi telah tumbuh lebih tinggi/ dia pergi ke luar menuju lahan padi untuk melihat/justru bukan menemukan/ padi tersebut lebih tinggi/ dia menemukan bahwa tanaman padiyang sehat telah mulai kehilangan kekuatannya layu dan mengering/ yaah selesai/ emmmulet sek/ mulet/ mm...m/ wes mari}


83

All the process of translation of the four texts done by the two translators were recorded andtranscribed. The transcriptions were then analyzed to find the strategies taken by the twotranslators and whether the different cultures have an influence in translation.

The discussion of the findings below is organized according to the texts related to thecultures of the translators. To make it easy, the two translators are labeled as Translator A(Chinese Budhist) and Translator B (Javanese Moslem).

CULTURE KNOWN TO BOTH TRANSLATORSThe culture known to both translators was that there was a plant called rice that grew in a field.According to Baker, a type of food is culture-bound (1992:21); therefore, it could be assumedthat a type of plant was also culture-bound. The text used for translation contained this type ofplant and words related to it. The culture-bound words and the translations were as follows:

Table 1. Culture Known to Both Translatorswords translator A translator Brice padi padifield lahan ladangcrop tanaman budidaya tumbuhan

There were apparently no difficulties of culture-bound words encountered by bothtranslators. Both could directly translate rice into padi, and were not confused whether it shouldbe padi or beras. The following is the example of the translation process of translator A intranslating rice. A read first the English text and the example below is the one when he startedwriting his translation (underlined words are the translation that he wrote on his paper.)

the farmer and the rice plants [clears throat] e… seorang… petani dantanaman…padi…nya/there was/once a farmer who always wished that the ricein his fields would grow more quickly/ pada…suatu…ketika…emm/ ini tangankananku megang bolpen tangan kiriku megang minyak angin/ jadi kadang kaloaku berhenti nulis/ aku bau minyak angin ini lagi/emm/enak baune/pada suatuketika/ ada seorang…petani…emm yang selalu…berharap ee that the rice inhis fields…agar tanaman padi… field/field itu... apa ya? Field itu kalo gasalah anu… apa? Lahan/

After reading the title once again (the farmer and the rice plants), he directly wrote thetranslation of the Indonesian translation. After reading the first line of the story (there was oncea farmer who always wished that the rice in his fields would grow more quickly), although hecommented about the herbal oil that he held in his hand—this happens in TAP—he directlywrote the translation after repeating again the clause [that the rice in his fields]. Although Athought for a while about the right translation of the word field [field itu... apa ya? (field is …what is it?) Field itu kalo ga salah anu… apa? (Field is if it is not mistaken er …what?). Lahan]he could directly solve the problem. Therefore, it was not considered a difficulty.

Likewise, B also could directly translated rice as in the example below.… the farmer and the rice plant/there was once a farmer who always wishthat/the rice in his field would grow more quickly/ suatu saat/pada suatusaat/there was once/pada suatu ketika ada seorang petani yang selaluberharap bahwa…padi…di…di ladang/in his field/padi di ladangnya akantumbuh lebih cepat

The word padi came directly and he did not think about it again.Checking the dictionary is only done by A for the word crop. He checked the dictionary

because he wanted to preserve its specific meaning. Translator B did not check the dictionaryand could directly translate the words field and crop. The strategies taken by translator B wereusing the general term for crop and the specific term for field, while A used a more general termas the strategy to translate field into lahan.

Julia Eka Rini

84

However, both translators had difficulties in wither and droop that did not haveanything to do with culture. The following is the example of translator A.

had all began to wither/ and droop/ yang sehat …telah… mulai…wither ituapa? layu to… layu dan/ droop ini apa ya?Buka kamus/ini kok gatel kabeh setanganku/ tangan kiriku gatel/ guaruk garuk terus ae ket tadi /wither/ droop ehdroop seh kok wither/ droop/ abcdefghijklmno o kelewatan/opqrs/drug?jklmnoprstu/ uudoop/ 0/drop/lho?/klmnopqrs/d/op/op/droop/to sink down/hangor bent down/to lose vitality or strength/became weaken/languish/adjective/ tolose vitalityor strength/ to sink down/ hang bent down/to lose vitality orstrength/ ya to lose vitalityor strength/sip emm/ berarti ketemu/ bukan layu shetelah mulai layu dan coret sek. telah mulai kehilangan/kekuatannya … danlayu/iya/wither tu/ layu ya? sek sek buka kamus lagi (open dictionary)/kamuskamus kamus/ w/ wither/ absdefghi/ ini sampe piro ya kesete ya?(how longdoes one side of this cassette last) Cukup? (Enough?) baru dua pertiga kok/aduh kepalaku gatel lagi (my head itches) /aneh (strange)/ kok aku kok gatelgatel se ndak kelas ini (why does my head itch in this class? onok opo se?what’s wrong with this class?)/klmn/ opqrst/ wither/ mm to dry up as/ forgravish/to lose/frigil/or freshness/layu/iya/hilang kesegaran/to wither lemah/apa? kering/layu dan mengering/berarti/telah mulai kehilangan kekuatannyakoma/ini dan ndak usah/layu dan mengering [sighing]

The two words are not that easy to translate. He opened both English-English dictionary. Liketranslator A, translator B also checked the dictionary. The following is the example of translatorB when dealing with the two words.

semua tumbuhan… itu … mulai wither and droop/ wither/ wither and droop/wwwwwww wither/ mana w ini/sebentar ya/ wither wither wither wither witherwither wither wither wither/to dry up/mengering/mana wither/wither gak pakaied/to dry up/to weaken/languish/mengering/mm lemah atau mengering/bisalemah atau/droop apa?/dra droop droop/droop/to sink down/hang or bentdown/halah layu/terus/withering/down/ya layu/mulai mengering dan layu/witherand droop/mengering dan layu/OK/saatnya writing.

Both A and B needed the dictionary to translate the two words. Besides, both translators alsoasked themselves the meaning of the words using apa (what). Both the dictionary and thequestion word apa are signals of difficulty faced by both translators

Globally, the translation process is smooth in translating a text whose culture is known.Translator A translated the text without great difficulties concerning culture-bound units. Heread the text first, translated every sentence and wrote it directly. After that he read it once againto check the translation as can be seen in the given example in the method above. Translator Bstated explicitly in the protocol, “wah ini lebih gampang ini” (this is easier) compared to thefirst text he translated (How Do You Shower a Bride)

The results of A and B are more or less the same, except in translating what aday. The following is the translation of both A and B.

Table 2. The Indonesian Translation The Farmer and The Rice PlantsEnglish text A’s translation B’s translation

“What a day! I haveworked so hard, “Hesaid to his family, “butat least I know that therice plants are a littlebit higher.”

Dia berkata kepada keluarganya,“Hari yang melelahkan! Akutelah bekerja keras, tetapisetidaknya aku tahu bahwatanaman padi telah tumbuhsedikit lebih tinggi!”

“Hari yang menyenangkan!Aku telah bekerja keras,” ujarsang petani pada keluarganya.“Namun, paling tidak, aku tahubahwa tumbuhan padiku sudahsedikit lebih tinggi!”


85

All sentences were more or less translated in the same way, except “What a day!” as can be seenbelow in the process of Translator A; he associated it with the clause after it—I’ve workedvery hard.

Mm dia berkata kepada keluarganya/ tanda petik/ hari yang menye/ salahtulis/ menyenangkan/ koma/ loh tanda seru bukan koma/ aku telah bekerjadengan keras. At least I know that rice plants are little bit high/ What aday/What a day ini bukane maksude hari yang melelahkan/ I’ve worked veryhard/ o iya/What a day ini berarti disini bukan hari yang menyenangkan tapihari yang melelahkan/ ini menyenangkan salah/ coret/ jadi melelahkan/ hariyang melelahkan/

On the other hand, B’s translation is different; it has the opposite meaning. In the process, first,he used the word indah, then replaced it with menyenangkan; he associated what a day with theclause before the interjection—but very pleased with himself. Different context used can resultin different words chosen in the translation.

CULTURE KNOWN TO ONE TRANSLATOR, BUT NOT TO THE OTHERThe culture known to Translator A is the religion of Buddhism and to Translator B is thereligion of Islam. Each text used for the translation was about Islam and Buddhism. It can beseen below how culture plays a role in the translation process. In translating an Islamic text, SheHad True Faith, the culture of the text (Islam) is known to Translator B, who is a Moslem, butnot to Translator A, who is a Buddhist. The difficult culture-bound words they encounteredwere (1) true faith, (2) day of judgment, (3) games of chance, (4) idols, (5) intoxicants, (6)divining arrows.

B translated faith as keyakinan and did not use a specific word like iman. Below is howthe subject thought about the word.

Figure 2. True faithtrue faithkeyakinan yang benarkepercayaankeyakinan yang benarkeyakinan yang teguh

True faith was translated into keyakinan yang benar and later was changed into keyakinan yangteguh because of the context; this can be seen from the transcript [judulnya di atas kayaknyasalah ini...(seen from the title, this seems wrong)]. Thus, he used a more neutral strategy. Afterhe translated all the lines and checked them again, he decided that teguh was better than benar.Although faith can be considered as a culture-bound word (since it is about religion), it is a littlebit general because nearly all religions talk about faith.

A more specific culture-bound word is the day of judgment and it can be seen belowthat culture influenced the translators’ decisions of which term to use.

Figure 3. Day of Judgmentthe day of judgment judgment dayhari pembalasanhari penilaianpengadilanhari pengadilanhari penentuanhari pembalasan

In the course of thinking, culture played a role when he said, “...opo yo? (what is it) Kalau diIslam itu hari pembalasan....” (In Islam the term is hari pembalasan).

Another culture-bound word is “divining arrows.” This word was not attended by B(this is the strategy of translation by omission) because probably it is not from Javanese cultureor Islam; it is probably from another culture, Middle East, with which the translator was notfamiliar of. Other words are culture-bound, but they are not deeply embedded in the culture andreligion because all culture and religions have them, for example: intoxicating drink, games ofchance, and worship idols. The word intoxicant was translated as minuman racun, whileactually it should be minuman yang memabukkan. The word idols was translated aspertanda.The translator used a less expressive strategy.

Julia Eka Rini

86

Some other words are culture-bound, but they did not cause difficulties. B could directlytranslate them. If the Moslem’s translation of those words are compared to those of theBuddhist, it can be clearly seen in the table below that culture plays a role in choosing the term.

Table 3. The Non Difficult Culture-bound Words of She Had True Faithculture-bound words Moslem Buddhistprayer sholat doachapter surat babbelievers umat Islam orang percaya

When translating an Islamic text, the Buddhist used general words, while the Moslem usedwords related to Islam.

Culture known to the Buddhist, but not to the Moslem, in translating the Buddhist text,“The Real Magic”The difficult words are (1) magic, (2) disciples, (3) power, (4) clairvoyance, (5) samadhi, (6)dharma, (7) suffering, (8) freedom, (9) psychic power, (10) blessing, 11) charm, (12) spell, (13)path, (14) the way Buddha passed, (15) vipassana, (16) contemplation, (17) mental object, and(18) liberation. In the following table, the differences of word choice between the two subjectscan be seen.

Table 4. Difficult words of “The Real Magic”culture-bound words Moslem Buddhistmagic kekuatan sihirdisciples pengikut muridpower kemampuan kekuatanclairvoyance kemampuan untuk melihat

yang tidak kasat matakemampuan untuk melihat hal-halyang tidak dapat dilihat oleh mata

samadhi semedi meditasidharma dharma dharmasuffering penderitaan kesengsaraanfreedom kebebasan kebebasanpsychic power kekuatan fisik kekuatan tubuhblessing pemberkatan pemberkatancharm mantera jimat-jimatspell aji-ajian mantera-manteravipassana vipassana vipassanacontemplation meditasi meditasimental objects pikiran objek-objek dalam pikiranliberation kebebasan pembebasan

In translating the text of a culture which was not familiar to him, he referred to his greathealing power as kekuatannya untuk pulih kembali, but before he decided to use this translation,he referred to untuk kekebalan sesuatu. This was clearly from Javanese culture, where peoplewho are good at martial arts have power to be immune from sharp weapons. [Ini cerita tentangkemampuannya untuk pulih kembali. (This story is about the ability to recover)]

Also, for the phrase of his clairvoyance, before he decided to translate it as untukmelihat sesuatu yang tidak kasat mata, he used the term goib. [Of his clairvoyance..what isclairvoyance? I don’t know! clair.. clairvoyance hemm. Ini (this) clair clair clair clair clairclairvoyance ah kok dak ada? (it is not here) oh ini?! that are not in sight or that cannot be seenthat cannot oh.. kemampuannya untuk membaca pikiran orang… (the abilityto read people’smind) kemampuannya indra mosok indra ke enam (the ability of sixth sense) kemampuannyauntuk melihat sesuatu yang goib (the ability to see something goib) This is paraphrasing byusing related word strategy.


87

Also, with the phrase samadhi, Translator B used semedi because his culture has it anddirectly related it with bertapa (meditation and fasting) [penetrating semedi..penetrating itumemasukkan.. dan memasukkan dan melakukan semedi melakukan semedi semedi itu opo yo(what is it)? dan bertapa dan bertapa dan melakukan- dan bertapa]. In translating charm andspell, he used the word aji-ajian, which is common in Javanese culture. He used the strategy oftranslation by cultural substitution.

Magic should be translated as keajaiban not sihir, because real magic here refers to theteachings of the Dharma that can liberate the mind and put an end to suffering (English text,sentence 5). Thus, real magic should be translated keajaiban sejati, not kekuatan nyata (as Btranslated it) or sihir yang sesungguhnya (A).

Translator A referred to Samadhi as meditasi and translated contemplation as meditasikeagamaan because meditasi is a common practice in Buddhism. Translation B, however,referred to it as semedi, which is a common practice in Javanese culture. He also used meditasifor contemplation.

Culture-bound words related closely to religion were absorbed, such as dharma andvipassana by both translators. Both translators made mistakes in translating psychic power askekuatan fisik or kekuatan tubuh (physical power) because psychic is connected more with mindthan with body.

Translator A said, “Aku rasa bacaan yang kedua ini...lebih gampang dingerti ya, soalemungkin memang kepercayaan yang saya pegang dengan bacaan mungkin sama jadi mungkinsaja lebih mudah untuk…belajar untuk memahami.” (I think this second text is easier, becausethe text is the same with my belief; therefore, it is easier to comprehend). This also proves thattexts containing a familiar culture are easier than those containing one that is unfamiliar.

CULTURE UNKNOWN TO BOTH TRANSLATORS (How Do You Shower a Bride?)The text used for translation was a funny story in which the culture was unknown to bothtranslator A and translator B. The story is about the bridal shower, which is very culture-specificbecause both in Chinese and Javanese custom, the party for a bride-to-be is without gifts.However, in Javanese custom, water is used, but it is not used in Chinese custom.

The difficulties related to culture encountered by the translators in translating theEnglish text were the word shower and groups of words containing this particular word, forexample, shower a bride in the title, bridal shower in the first sentence, to the shower in thesecond sentence, shower the bride-to-be with gifts in the fifth sentence, and shower in the lastsentence. This word or group of words is difficult to translate because the word shower isusually associated with water, such as to take a shower, which is translated in Indonesian asmandi. In the context of the story, the word shower in the phrase bridal shower is culture-bound and has the meaning of giving gifts to the bride-to-be. It has nothing to do with theassociation of water. This double meaning makes the story funny because the character thoughtthat it was a swimming pool party (the character associated the program with water because ofthe word shower, but actually it does not have anything to do with water, but with gifts). So, thetranslators faced two difficult tasks. One, they needed to keep the story funny; they needed toreveal the funny point bit by bit, not all at once, and certainly by not breaking the point ofhumor from the very beginning. Second, they needed to find a word or group of words whichcontain the double meaning. Although both translators checked the dictionary and knew fromthe dictionary that bridal shower means giving gifts to the bride-to-be, which resulted in thesame translation, the association and considerations they took were different in the process, as itcan be seen in the following discussion.

Translator B’s (the Javanese) difficulties were translating the title How Do You Showera Bride and later also bridal shower (1) as can be seen below.

Julia Eka Rini

88

Figure 4. The Javanese Moslem Translator

For (1) in the diagram, the TAPs say [Opo iki a bridal shower.(What is bridal shower) Gakngerti istilah Indonesiae ini (I don’t understand the Indonesian term)... How do you shower abride. Shower ini apa.. apa ya (what is it) shower. For (2), Liat di kamus dulu (see thedictionary first). After finding the meaning about ‘shower a bride’ in the dictionary, the TAPssay

Ooo ini nomer empat ini (ooo this one number four)/ A party of- at which anumber of gifts are presented to the guest of honor… hmm, a apaan ya?(hmmwhat is it?) Siraman pengantèn/ perasaan pengantèn itu ndak pake hadiah (Ithink of the bride, gift is not used) a party at which a number of gifts,adalah…].

At this point the Javanese translator still directly associated ‘shower’ with siraman pengantin(3); he still associates it with the Javanse tradition. In Javanese culture, brides are indeedshowered with water mixed with flowers. And he said that in such kind of ritual, no gift is given(perasaan penganten ndak pakai hadiah)(5). Although the translation he wrote later on wasBagaimana cara mau memberikan kado pengantin (give a gift) (7), he had been thinking ofsiraman pengantin and also menyiram calon pengantin dengan kado a number of times. Thiswas cultural substitution strategy. Bridal shower was translated as Penyiraman kado, also acarasiraman pengantin, again, taking reference from the culture (4). Later, in the evaluation step, hewas aware that the translation of the title is not suitable and thought again of siraman (Showeritu istilah yang paling… harus bisa diartikan /di sini?/)and compared the customs in Indonesia(with water) and in the US (with gifts) (tapi kalo dikasi siraman.. orang Indonesia siraman yakaro air, siraman pengantin, bukan.. pake kado biasanya. Lek barat, kalo di Amerika pakekado.). In the last sentence, … whenever I hear the word, shower, the Javanese translatortranslated the sentence into setiap kali mendengar kata siraman, referring to the customs.

As it can be seen in the diagram below, different from the Javanese translator, theChinese one, before consulting the dictionary took the meaning of bathing (mandeni) the bride(1), and after consulting the dictionary (3), he directly took the meaning of giving gifts to thebride-to-be (4).

Figure 5. The Chinese Buddhist


89

He just took the dictionary meaning because it does not exist in his culture. However, theChinese translator thought of words like penghormatan, penganugerahan, pemberian whichhave the meaning of to give a gift. This is using less expressive word strategy.

In the 2nd sentence, for the phrase ‘… what she had to bring to the shower, the Javanesetranslator used a generic word acara, but the Chinese used acara pemberian kado and in 3rd

sentence added pada saat acara memandikan mempelai wanita (bathing the bride-to-be) and inthe evaluation he was aware of the funniness that he had to transfer (5).

Cuma ndak jadi joke malahan (no joke then). Jadi semacam kayak (it’s like)/apa namae..eee...apa ya (what is it called?)/ bukan joke lagi ya (no joke then)/ya dob- ya (it’s …)/ artikel biasa kalo mau jadi joke ya harus jadi memandikan(if the joke should be there, it should be bathing)/ tapi nek memandikan pikire(but if it’s bathing, people will think that…)/ganti ya (it’s better to change it)/bagaimana anda memberikan kado kepada seorang mempelai wanita... howdo you shower a bride shower bride/ ya emm/ dirubah ae (change it).Bagaimana anda memandikan seorang mempelai wanita ya gitu... ee how doyou shower a bride... emm.. bagimana! anda/ ini bingung ini aku (I’mconfused)

In the 5th sentence, ‘… she was supposed to “shower” the bride-to-be with gifts’ was translatedby the Javanese as ‘menyiram calon pengantin dengan kado’ while the Chinese translator stillused ‘bathing.’ The Chinese translator used ‘shower’ and did not translate it.

In comparing the translation of the Javanese and the Chinese, some important itemscould be noted. Right from the first sentence, the Javanese translator used the word siramanwhich had something to do with water. Therefore, the context was still useful when in the thirdsentence, it was said that bringing a swimming suit, the character in the story was wonderingabout the swimming pool. It was different from the Chinese translator’s translation. Right fromthe very beginning, he used ‘giving gifts,’ (which does not have anything to do with water) andso suddenly at the third sentence, the character was wondering about swimming pool (which isassociatd with water). Although he compensated it by adding ‘the bathing program’ in theIndonesian translation, it made the translation strange. It was even more strange when the word‘shower’ was not translated, because the readers of the translation did not find that word in thebeginning and suddenly it came up at the end. The Javanese was more consistent, translatingshower into siraman, although the translation of the title using ‘giving gifts’ also makes the TTnot funny.

Although the Chinese translator’s translation was not as consistent or funny as theJavanese’s translation, it did not mean—after knowing the process happening in the translator’smind—that he did not know what to do with the difficulties he encountered. He was very wellaware of it and did the steps of analyzing, transferring, restructuring and also evaluating manytimes. He was not only thinking about finding the right word, but also trying hard to maintainthe joke. This tells us that a mistake is not produced without thinking and effort to find the righttranslation.

CONCLUSION AND SUGGESTIONFrom the process of translation of the two translators, two points can be concluded. One, it isclear that translators’ culture does play a role in their consideration in picking up the rightwords. Translators will derive something from their experience of his own cultures, e.g. theJavanese translator derived something from the Javanese prewedding ceremony although he hadconsulted the dictionary beforehand. If the culture in the ST is not like their own culture, like theChinese, translators rely on information taken from the dictionary or any other written text.Two, culture-bound words creates some difficulties for the translators, but some words which donot become the difficulties for the translators are translated differently by the translators, which

Julia Eka Rini

90

is influenced by culture. However, text for culture known to both translators used in thisresearch is not that difficult because the text is not too deeply embedded in the culture.Concerning the translation strategies applied by the translators, cultural substitution strategy isgenerally used by the translation if the culture is nearly the same. When the culture is different,translators generally use the strategy of using a more neutral word or paraphrase by relatedwords.

What people see in the product (the translation printed on paper) might not convey all thethings that happened in the process (in the translator’s mind). The crucial point is what makesthe translator decide to choose one word and not the other. The problem of culture is indeeddifficult to cope with if the translator’s culture is really different from the text, moreover, if thisdifficulty occurs in a humorous text.

Another important factor is time. When the translator does not have enough time toevaluate his own work, he reads his work as a translator, not as a reader. As a translator, it isdifficult to see whether he can preserve the joke or not, while as a reader, he will be moresensitive to this. The ideal position to evaluate his own work is when the translator can positionhimself as a reader when evaluating his own work.

Suggestions for further research are as follows. Concerning the problems of theresearch, some problems can be added for the next research, for example, translation unit in theprocess of translation. Concerning the materials to be translated, texts used should be short andcontain only one or two difficulties. The difficulties should be center or global understanding.because the translation process of dealing with the difficulties could be seen in a better way. Ifthere are too many difficulties and only about local understanding, the process is not strikingbecause the strategies taken are general and translators tend to abandon difficulties. Concerningmethodology, people doing the transcript should read the English text first before transcribingso that they do not have difficulties in doing the transcription. Mentioning the time for thecassettes should be emphasized so that translators understand that it is not the time for thetranslation, because if it is not emphasized, translators will think that it is the limit for thetranslation. Combination of introspective and retrospective methods can also be used. Thismight bring clearer explanation of what is happening in translators’ minds while translating.

NOTE* I would like to thank Professor Esther Kuntjara, Ph.D., who conducted the research together with me,for her invaluable suggestion and feedback for the study. I feel very indebted to Prof. Bambang KaswantiPurwo, who gave me feedback about the article. I am also very grateful to Angie Elaine Breneman forediting this article.

REFERENCESAppelbaum, Richard P. & William J. Chambliss. 1997. Sociology. New York: Longman.

Armstrong, Nigel. 2005. Translation, Linguistics, Culture: A French-English Handbook.Clevedon: Multilingual Matters Ltd.

Baker, Mona. 1992. In Other Words: A Coursebook on Translation. London: Routledge andKegan.

Bassnett, Susan & André Lefevere. 1998. Constructing Cultures: Essays on LiteraryTranslation. Clevedon: Multilingual Matters Ltd.

Beamer, Linda & Iris Varner. 2001. Intercultural Communication in the Global Workplace.New York: McGraw-Hill.

Bernardini, Silvia. 2001. Think-Aloud Protocols in Translation Research: Achievements,Limits, Future Prospects. Target 13 (2):241-263.


91

Bernardini, Silvia. 2000. ‘Think-aloud Protocols in Translation Research: Achievements,Limits, Future Prospects’. A paper presented at the Conference of Process-orientedTranslation Research. 28 April. Accessed 1 March 2007. Available from: http://www.art.man.ac.uk/SML/ctis/events/Conference2000/process1.htm.

Bernardini, Silvia. 1999. Using Think-Aloud Protocols to Investigate the Translation Process:Methodological Aspects, in John N. Williams (Ed.). RCEAL Working papers in Englishand Applied Linguistics 6, pp. 179-199. Cambridge: University of Cambridge. Accessed6 November 2006. Available from http://www.rceal.cam.ac.uk/Publications/Working/Vol 6/Bernardini.pdf.

Brown, James Dean, & Rodgers, Theodore S. 2001. Doing Second Language Research. Oxford:Oxford University Press.

Duff, Alan. 1981. The Third Language. Oxford: Pergamon Press.

Hatim, Basil & Jeremy Munday. 2004. Translation: An Advanced Resource Book. London andNew York: Routledge.

Kornfield, Jack & Paul Bleiter (Eds.). 1985. A Still Forest Pool: The Insight Meditation ofAchaan Chah. Illinois: The Thesophical Publishing House.

Li, Defeng. 2004. “Trustworthiness of Think-Aloud Protocols in the Study of TranslationProcesses.” International Journal of applied Linguistics 14.3:301-313.

McKay, Sandra Lee. 2006. Researching Second Language Classrooms. Mahwah, New Jersey:Lawrence Erlbaum Associates.

Rajan, B.R. Sundara & George. M. Jacobs (Eds.). 1996. Who Is the Most Talkative of ThemAll?: Stories for Language Teacher Education. Singapore: SEAMEO RELC.

Suryawinata, Zuchridin & Sugeng Hariyanto. 2003. Translation: Bahasan Teori dan PenuntunPraktis Menerjemahkan. Yogyakarta: Kanisius.

‘She Had True Faith’. Accessed 12 December 2006. Available from http://www.batkhela.com/islam/story7.shtml

Wills, Wolfram. 2004. “Translation Studies-The State of the Art.” Meta, 49.4:777-785,retrieved on November 11, 2006.


Resensi Buku

Judul : Discursive Pragmatics (Handbook of Pragmatics Highlights, volume 8)ISSN : 1877-654XPenyunting : Jan Zienkowski, Jan-Ola Östman, dan Jef VerschuerenPenerbit : Amsterdam: John Benjamins B.V. 2011Tebal : 307 halaman

Faizah SariSurya University

[email protected]

Buku ini merupakan koleksi atas lima belas artikel komprehensif yang disunting secara teliti danyang membahas beraneka ragam topik menarik dalam analisis wacana dan pragmatik. Kelimabelas artikel itu dipaparkan dalam gaya bahasa yang sederhana namun jelas. Dalam Appraisal(hlm. 14-36), Peter R.R. White dari University of Birmingham membahas kajian mengenaikerangka teoretis penggunaan bahasa penilaian (appraisal) yang relatif masih baru padapenelitian linguistik umum. Dalam artikel ini dibahas parameter-parameter penilaian, sepertipengaruh (affect), keputusan (judgement), penghargaan (appreciation), cara langsung atautersirat (direct or implied modes of activation), kriteria tipologis (typological criteria), salingpengaruh antarcara bersikap (the interplay between the attitudinal modes), kedudukanintersubjektif (intersubjective stance), penilaian sikap (attitudinal assessment), dan keterlibatan(engagement).

Dalam Cohesion and Coherence (hlm. 37-49), Wolfram Bublitz dari University ofAugsburg mengkaji hubungan antara kepaduan (cohesion) dan kelekatan (coherence) dalamwacana yang membedakan sudut pandang peneliti yang lebih berorientasi pada bentuk danstruktur (form and structure oriented linguists) daripada pada fungsi (function orientedlinguists). Dalam Critical Linguistics and Critical Discourse Analysis (hlm. 50-70) Ruth Wodakdari University of Lancaster mengkaji kesimpangsiuran terminologis dan pemahamankomunitas linguistik mengenai Critical Discourse Analysis (CDA). Wodak menelaahpendekatan-pendekatan yang mampu membentuk Critical Discourse Analaysis menjadi bagiandari kajian linguistik yang benar-benar berdasarkan pemahaman intelektual sosiopolitik yangkuat. Dalam Énonciation: French Pragmatic Approaches (hlm. 71-101), Marjut Johansson danEija Suomela-Sami dari University of Turku membahas pendekatan teoretis mengenaipragmatik enunsiatif (enunciative pragmatics), yaitu produksi pengucapan penutur dalam situasitertentu, terutama dalam bahasa Prancis, dengan menjelajahi sejarah pragmatik enunsiatifbahasa Prancis dan tipe-tipe modalitas enunsiatif.

Dalam Figures of Speech (hlm. 102-118), Manfred Kienpointner dari University ofInnsbruck mengemukakan ungkapan kiasan (figures of speech/FSP) berdasar pada kerangkaretorika kuno (ancient rhetoric) dan menguraikan ikhtisar upaya kontemporer dalammendefinisikan dan mengklasifikasikan FSP dalam berbagai disiplin ilmu, misalnya, kajiankomunikasi lisan, linguistik, kritik sastra, filsafat, dan psikologi. Dalam Genre (hlm. 119-134),Anna Solin dari University of Jyväskylä mengkaji konsep genre dalam wacana. Denganmemformulasikan teori genre awal dari Bakhtin (1953, 1986), Solin menggarisbawahi genresebagai konsep penting interaksi antara tipe-tipe wacana dan antara interaksi pengguna danpenyusun teks. Dalam Humor (hlm. 135-155), Salvatore Attardo dari Texas A&M University,Commerce membahas pendekatan linguistik dalam mengkaji humor, terutama dalam humorreferesial dan lisan, semantik, prinsip-prinsip kerja sama lisan (the cooperative principles),analisis percakapan (conversation analysis), termasuk di antaranya humor berdasarkan situasi(canned jokes) atau lelucon percakapan, dan sosiolinguistik humor (the sociolinguistics of

Resensi Buku

94

humor). Dalam Intertextuality (hlm. 156-175), Stef Slembrouck dari University of Ghentmemaparkan telaah intertekstual dalan analisis percakapan melalui pandangan teoritis Bakhtin,Volosinov, dan Bordieu. Dalam Manipulation (hlm. 176-189) Paul Chitton dari University ofLancaster membahas manipulasi sebagai pengendali pemikiran (thought control) yang tidakmelekat pada struktur bahasa. Dalam Narrative (hlm. 190-207), Alexanda Georgakopoulou dariKing’s College London mengemukakan narasi sebagai alat komunikasi yang diperoleh dandipraktikkan melalui pendekatan kontekstual.

Dalam Polyphony (hlm. 208-222), Eddy Roulet dari University of Ghent mengkajidialogisme dalam wacana yang kerap dikenal dalam kajian pragmatik sebagai polyphony.Dalam Pragmatic Markers (hlm. 223-247), Karin Aijmer dan Anne-Marie Simon-Vandenbergen dari University of Gothenburg dan University of Ghent membahas pendekatankontemporer dan lintas-bahasa atas kajian pemarkah pragmatik (pragmatic markers) bahasa-bahasa yang bermunculan melalui pembahasan pragmatik (Aijmer & Simon-Vandenbergen,2006). Aijmer dan Simon-Vandenbergen juga mengulas problematika terjemahan dan semantikatas penggunaan pemarkah pragmatik dalam percakapan. Dalam Public Discourse (hlm. 248-265), Srikant Sarangi dari Cardiff University menelaah model-model dominan dalam pragmatiksosial. Dalam Text and Discourse Linguistics (hlm. 266-285), Jan-Ola Östman dan TuijaVirtanen dari University of Helsinki dan Åbo Akademi University memberikan gambaran luaspada linguistik teks dan wacana, yang di dalamnya terhimpun beraneka ragam topik yang telahdibahas secara khusus berdasarkan data individual oleh penulis-penulis sebelumnya padaHandbook of Pragmatic Highlights ini dan mengupas perihal keumuman dan kekhususan kajianyang ada pada lingkup wacana. Dalam Text Linguistics (hlm. 286-296), Robert de Beaugrandedari University of Vienna menyimpulkan bahwa linguistik teks merupakan kajian yang dinamis,di antaranya, tidak hanya mencakup unit linguistik tetapi juga interaksi. Menurutnya, penelitiankontemporer mesti berdasar pada data yang otentik, yaitu data yang berasal dari naturallyoccurring texts and discourse. Ditambah lagi, peneliti harus ‘melekat’ pada konteks wacana danmemiliki motif pengembangan hubungan antara teks dan masyarakat. Dengan demikian padaakhirnya, linguistik teks merupakan kajian yang interdisipliner.

Buku ini baik sekali digunakan sebagai bahan acuan bagi penelitian bermetodologianalisis wacana dan berdata bahasa lisan alami, Systemic Functional Linguistics, FunctionalGrammar, dan Critical Discourse Analysis. Walaupun kajian di bidang pragmatik sudah cukupbanyak, penelitian bersifat spesifik, misalnya partikel pragmatik, analisis wacana, analisispercakapan, dan sejenisnya, masih bergantung banyak pada publikasi khusus mengenai bidang-bidang ini. Jilid kedelapan Handbook of Pragmatic Highlights ini merupakan salah satupublikasi yang konsisten memaparkan dan memopulerkan kajian-kajian spesifik di bidangpragmatik yang luas namun sedang berkembang. Artikel-artikel dalam jilid ini secara eksplisitjuga menjelaskan medan penelitian yang masih dapat dikembangkan, sehingga para penelitiyang tertarik akan pendekatan linguistik ini dapat berkontribusi memperluas diskusi penting atastopik ini.

DAFTAR PUSTAKA

Aijmer, Karin & Simon-Vandenbergen, Anne-Marie (Ed.). 2006. Pragmatic markers incontrast. Studies in pragmatics 2. Amsterdam: Elsevier.

Bakhtin, Mikhail Mikhailovich. 1986. [1953]. The problem of speech genres. In Speech genresand other late essays: 60–102. (translated by V. McGee; ed. by C. Emerson & M.Holquist.) University of Texas Press.


JELAJAH LINGUISTIK

Rubrik ini membuka peluang untuk saling berbagi di antara kita tentang beberapa kemungkinan topik ini:a. pencanangan metode penelitian linguistik yang belum lazim digunakanb. daur-ulang metodologi penelitian linguistikc. persoalan data yang – meskipun barangkali belum ditemukan pemecahannya – penelusurannya

berpeluang membuka sesuatu yang baru yang belum pernah menjadi perhatian peneliti terdahulud. penerapan teori linguistik tertentu untuk menjelaskan data bahasa seperti bahasa Indonesia yang

membuat peneliti mempersoalkan teori yang bersangkutan

MENINJAU ULANG METODOLOGI KONTEMPORER

Faizah SariSurya University

[email protected]

Hampir seratus tahun yang lalu kumpulan catatan kuliah Ferdinand de Saussure pertamadipublikasikan oleh mahasiswanya dalam Course in General Linguistics (1916) dan telahmenjadi sumber rujukan bagi peneliti linguistik dan pada kelas-kelas linguistik di institusipendidikan di seluruh dunia bahkan sampai di putaran milenia baru ini. Sejak itu begitu banyakaspek-aspek bahasa dan penggunaan bahasa terutama pada konsep kalimat yang telah ditelitidan dipublikasikan. Namun, pertanyaan klasik selalu muncul setiap kali penelitian bahasadilakukan oleh seseorang: dengan mengaitkan teori A dan/atau teori B dengan data yang sayamiliki, apakah sebenarnya yang akan digali dari sistem linguistik sebagai objek penyelidikan(object of inquiry) ini? Apa yang ingin saya capai saat penuntasan penelitian melaluipembedahan suatu kalimat?

Beaugrande (2011) memberi ulasan singkat bagaimana peneliti bahasa dapat mempertajameksplorasi pada kalimat sebagai unit berstruktur (a grammatical or syntactic unit) denganmelihat secara makro ”perjalanan” kajian linguistik secara umum agar peneliti bahasakontemporer dapat menciptakan metodologi yang lebih kritis dan meluas sehingga ‘potret’penelitian linguistik dapat terlihat lebih terfokus. Beaugrande menawarkan cara melihatpembatas-pembatas (constraints) di sekeliling suatu kalimat dengan lebih jauh lagi yang berefekpada “struktur (grammar), misalnya kepaduan (cohesion), kelekatan (coherence), penderetantopik (topic progression), penarasian (narrativity) dan perubahan keadaan (situationality)”(2011, hlm. 288). Dengan kata lain, tantangan penelitian kontemporer jatuh pada kemampuanpeneliti melihat lebih dalam lagi melampaui konvensi yang biasa pada aspek-aspek kerangkawacana, proses berwacana, pendekatan sintaksis lewat pragmatik dan semantik khususnya padaaktivitas komunikasi, dan gaya bahasa. Dengan demikian kerangka teoritis yang digunakanuntuk menganalisis bentuk-bentuk kalimat.

Sampai saat ini Beaugrande berpendapat bahwa cara-cara menjaring data dapat dikelompokkanmenjadi dua polar umum, yakni “kumpulan data asli berupa rekaman tersimpan dalam korpusdan penciptaan kalimat-kalimat tak berkonteks (isolated sentences), hanya berdasarkan intuisi”(2011, hlm. 286). Namun, keduanya memiliki ruang untuk perbaikan. Pertama, menurutBeaugrande, pada saat mengkaji rekaman atau contoh bahasa yang dipakai dalam tindakkomunikasi (naturally-occurring language sample), peneliti bahasa terbiasa untukmengkontekstualisasi perkataan, sanggahan, dan ucapan yang diproduksi oleh penutur dan inijustru menjadikan pembatas bagi penyelidikan yang lebih dalam. Dalam kenyataan yangsesungguhnya “suatu teks tidak dapat menampung semua aspek komunikasi, termasukkeinginan dan strategi penutur” (hlm. 289). Kedua, pengamatan berdasarkan intuisi terhadapsuatu kalimat ciptaan sendiri telah lama diketahui sebagai hal buatan yang kurang andal sebagai

Jelajah Linguistik

96

data karena, selain tidak berdasarkan bukti empiris, suatu contoh kalimat buatan dapat sajamenyesatkan dan terpolarisasi oleh latar belakang kebahasaan dan pemahaman subjektifpeneliti, bukan dari fakta yang objektif. Ringkasnya, kedua polar di atas masih dapatdikembangkan seperti halnya metodologi yang terus-menerus berpedoman pada jalur menuju kearah komprehensif melalui kritisi yang relevan dan konstruktif. Pelajaran bagi peneliti adalahkenyataan bahwa tatanan suatu metodologi dapat terus digoyang sehingga menjadi tugas utamapeneliti untuk terus memperbarui metodologi.

Satu hal imperatif yang perlu kita perhatikan adalah bahwa bentuk pokok semua analisis danpenyelidikan ilmiah yang kita lakukan akhirnya dituangkan dalam publikasi yang dibaca sesamakolega peneliti, calon peneliti, dan pemelajar bahasa yang akan menelaah publikasi tersebutuntuk mencari contoh-contoh yang mudah dipahami. Bagi pemelajar bahasa, misalnya, contohbahasa yang sederhana dan masuk akal dapat memberikan informasi akurat mengenai strukturkalimat atau arti kata. Contoh sederhana, bergantung pada tingkat kemahiran berbahasa, seorangpemelajar bahasa Indonesia sebagai bahasa asing perlu mengetahui bahwa bahasa Indonesiamembedakan pronomina pertama jamak kami (eksklusif) dan kita (inklusif), sebagaimana yangumumnya terdapat pada buku-buku teks pelajaran bahasa Indonesia bagi penutur asing. Akantetapi, dapat dijumpai pemakaian kita di antara sesama penutur bahasa Indonesia yang dalampengertian eksklusif, yakni tidak mengikutsertakan kawan bicara (addressee), misalnya dalamkalimat kita suka nonton bola bareng. Keanekaragaman ini dapat membingungkan pemelajarpemula bahasa Indonesia sebagai bahasa asing, terutama bagi penutur asing berbahasa pertamabahasa Inggris atau bahasa-bahasa Eropa lainnya.

Masih pada pronomina, mari kita lihat satu contoh yang relevan bagi sesama kolega penelitibahasa-bahasa Indonesia tanpa memperhatikan kebangsaan atau bahasa pertama.Keanekaragaman pronomina pada bahasa-bahasa di Indonesia, representasi pronomina pertamatunggal saya dan jamak kami digunakan longgar dan tidak ada pembatas tunggal/jamak dantidak perlu inklusif, misalnya torang dan kitorang dalam bahasa Melayu Maluku; awak dalambahasa Ocu, Melayu Riau; dan urang dalam bahasa Sunda. Seluruh keanekaragaman inimemberikan cara baru melihat struktur bahasa karena pentingnya mengikutsertakan proseswacana untuk menentukan pengaruh arti dan konteks.

Lalu, bagaimana cara kita menyikapi ruang untuk perbaikan pada kedua metode ini? Aijmer danSimon-Vandenbergen (2011) dan Beaugrande (2011) merangkumkan celah yang terdapat padametodologi linguistik kontemporer, bahwa peneliti (a) menggunakan sumber data asli, bukanrekaan; (b) terus berhubungan dengan data, bukan alih-alih memisahkan diri sebagai peneliti;(c) tidak superior atas data karena struktur komunikasi merupakan hal yang kompleks danmenyangkut pengetahuan dan masalah tentang masyarakat penutur. Intinya, kemajuanmetodologi bertumpu pada kemampuan peneliti berkolaborasi, meninjau ulang, dan meluaskanjangkauan penyelidikan.

Memang halnya, tekanan yang jelas berada di depan mata tiap peneliti adalah hasil penelitianyang akurat dan dapat dipublikasikan paling tidak secara nasional. Sementara disiplin ilmubahasa di belahan “dunia pertama” telah berkembang begitu pesat sehingga peneliti bahasa dibelahan dunia mana pun cenderung berpedoman pada konsep-konsep teori bahasa dari duniapertama, antara konsep parole dan langue (Saussure, 1916), konsep linguistik historiskomparatif (Bloomfield, 1933; cf. Salverda, 1985), konsep bahasa universal sebagai strukturbahasa yang cukup deskriptif (Chomsky, 1957), sampai pada linguistik fungsional (Halliday &Hasan, 1976).

Namun, penelitian bahasa-bahasa di Indonesia pun patut dibanggakan. Sampai tahun 1988 diIndonesia terdapat 95 kamus bahasa Indonesia dan bahasa-bahasa daerah telahdidokumentasikan (Idayu, 1988). Publikasi kamus ini terutama kamus bahasa-bahasa mayoritas,misalnya kamus bahasa Indonesia (hampir 50%), bahasa Sunda (5,3%), bahasa Bali dan bahasa


97

Jawa (4%), bahasa Minang (3%), dan bahasa Batak dan bahasa Melayu-Jakarta (2%). Dari sudutgeopolitik daerah, sampai tahun 1988 ini pula upaya penulisan kamus bahasa daerah, termasukbahasa-bahasa minoritas di Indonesia telah didokumentasikan, misalnya kamus bahasa-bahasadi Aceh (3%; bahasa Aceh, Alas, dan Gayo), Sumatra Utara (24%; bahasa Batak, Deli, Melayu-Langkat, Karo, Angkola, Dairi Pakpak, Mentawai, dan Nias), Sumatra Barat-Riau-SumatraSelatan (3%; bahasa Minang, Kerinci, dan Sakai), Kalimantan Barat-Tengah (3%; MelayuBanjar dan Bakumpai), Sulawesi Utara-Barat-Tengah-Tenggara-Selatan (3%; Bahasa Manado,Gorontalo, Tondano, Mandar, Suwawa, Muna, Tolaki, Wolio, Melayu Makasar, dan Bugis),Papua (3%; bahasa Biak), Nusatenggara Barat (bahasa Bima dan Sasak), dan Timor Leste (3%;bahasa Tetun).

Pasca 1988 dapat merupakan titik penting di saat upaya pendokumentasian bahasa melaluipenelitian bahasa dan pembuatan kamus bertemu dengan revolusi teknologi. Pada titik ini duniamenyaksikan derasnya informasi mengenai bahasa-bahasa di seluruh dunia, dan Indonesia turutmenikmati sorotan atas diversifikasi bahasanya. Bentuk-bentuk publikasi terdahulu melibatkanpercetakan non-digital, misalnya stensil atau dimuat pada bahan kertas yang mudah rusakapabila tidak disimpan dengan baik. Sekarang publikasi telah bertransformasi, termasuk dalambentuk pangkalan data (database) dan dapat leluasa diakses melalui jaringan internet. PenelitiIndonesia pun tertolong dengan revolusi informasi ini dan dapat dengan cepat menjelajahinformasi mengenai bahasa-bahasa di Indonesia. Contohnya, The World Atlas of LanguageStructures (WALS) menyediakan akses hasil penelitian pada 2679 nama bahasa yangterdokumentasi di dunia berikut dengan daftar publikasi berkenaan dengan bahasa-bahasa yangada pada daftar. Contoh lain, Ethnologue juga menyediakan informasi mengenai 7105 bahasa didunia.

Lalu, sebagai peneliti Indonesia, apa yang dapat kita petik dari ulasan rangkaian kesatuanmetodologi kontemporer ini? Beaugrande menganjurkan peneliti untuk “mengkaji ulangprioritas penelitian” (hlm. 293) Kita telah biasa menghadapi permasalahan bagaimanamengaitkan unsur teori dan empiris. Kita juga telah terbiasa merumuskan kategori berdasarkandata yang kita terima dari lapangan. Akan tetapi, justru pada persimpangan milenium ini kitaperlu mempertanyakan diri kembali: apakah metodologi yang digunakan telah menyentuhsegala aspek yang diperlukan dan menggunakan sebanyak mungkin perangkat penelitian yangtersedia sehingga data dapat muncul sebagai ”potret” yang mampu menceritakan keunikan danpotensi kontribusi ilmiah bagi ilmu linguistik.

Pada akhirnya, penggalian ilmiah mengenai bahasa merupakan suatu proses yangmendewasakan kemampuan bermetodologi peneliti dan memajukan keilmuan ke arah yanglebih beranekaragam dan interdisipliner. Sebagai peneliti kita merupakan bagian dari kontinumbesar keilmuan ini, yaitu pemegang peranan penting bagi penemuan linguistik (linguisticdiscovery), advokat pelestarian bahasa, dan duta keberlanjutan ilmu bahasa sebagai bagian darikemajuan peradaban manusia. Penelitian linguistik tetap akan terus berevolusi dan menemukankembali aspek-aspek bahasa yang belum terkuak. Sama pentingnya, penelitian linguistik mestitetap meninjau konsep-konsep linguistik sebagai hipotesis yang masih harus terus dibuktikanmelalui data bahasa sebagaimana dipakai dalam tindak komunikasi dan kerangka teoritis yangkomprehensif.

DAFTAR PUSTAKABeaugrande, Robert de. 2011. Text Linguistics. Dalam Östman, Jan-Ola, Zienkowski, Jan, dan

Verschueren, Jef (ed.). 2011. Discursive Pragmatics. Amsterdam: John Benjamins.

Jelajah Linguistik

98

Cysouw, Michael. 2013. Inclusive/Exclusive Distinction in Independent Pronouns. Dalam Dryer,Matthew S. dan Haspelmath, Martin (Ed.). The World Atlas of Language StructuresOnline. Leipzig: Max Planck Institute for Evolutionary Anthropology. (Available onlineat http://wals.info/chapter/39, Accessed on 2014-03-09.)

Dryer, Matthew S. & Haspelmath, Martin (Ed.). 2013. The World Atlas of Language StructuresOnline. Diakses pada 23 Februari 2014 dari http://wals.info

Ethnologue. Diakses pada 23 Februari 2014 dari http://www.ethnologue.com

Idayu, Yayasan. 1988. Bibliografi Bahasa & Kesusastraan Indonesia dan Daerah 1945-1988.Ed. 1. Jakarta: CV Haji Masagung.

Salverda, Reinier. 1985. Leading Conceptions in Linguistic Theory: Formalist Tendencies inStructural Linguistics. Amsterdam: Vrije Universiteit te Amsterdam.


BINCANG ANTARA KITA DARI DUNIA MAYA

SUSUNAN PENGURUS MLI DANDEWAN EDITOR LINGUISTIK INDONESIA

PERIODE 2014-2016;DEKLARASI DAN IKRAR ANGGOTA MLI

Pada Rabu, 5 Maret 2014 18:09, MLI <[email protected]> menulis:

Yth. Bapak/Ibu Anggota MLI,

Dalam Musyawarah Nasional Masyarakat Linguistik Indonesia yang diselenggarakan di HotelSheraton Lampung, tanggal 20 Februari 2014 yang lalu, diadakan pemilihan Ketua MLI periode2014-2016. Sebagai Ketua MLI terpilih untuk periode 2014-2016, melalui milis ini saya inginmengumumkan susunan Pengurus MLI periode 2014-2016, sebagai berikut:

Ketua : Dr. Katharina Endriati Sukamto (Unika Atma Jaya)Wakil Ketua : Dr. Fairul Zabadi (Badan Pengembangan dan Pembinaan Bahasa)Sekretaris : Dr. Ifan Iskandar (Universitas Negeri Jakarta)Bendahara : Dr. Yanti (Unika Atma Jaya)

Untuk pengelolaan jurnal Linguistik Indonesia, Dewan Editor periode 2014-2016 adalah:

Editor Utama : Prof. Dr. Bambang Kaswanti Purwo (Unika Atma Jaya)Editor Pendamping : Dr. Lanny Hidayat (Unika Atma Jaya)Anggota : akan ditentukan kemudian

Selain kedua hal di atas, kami juga ingin menyampaikan Deklarasi KIMLI 2014 dan IkrarAnggota MLI yang dibacakan pada saat penutupan acara KIMLI tanggal 22 Februari 2014(mohon lihat lekapan). Kiranya Bapak/Ibu berkenan menyebarluaskan Deklarasi dan Ikrar inikepada rekan kerja dan bahasawan yang Bapak/Ibu kenal dan tidak hadir pada KIMLI 2014 diLampung.

Rencana kegiatan Pengurus MLI periode 2014-2016 akan kami sampaikan pada kesempatanberikutnya.

Demikian informasi ini kami sampaikan. Mari kita majukan MLI, dengan berperan serta secaraaktif dalam kegiatan-kegiatan yang dipelopori oleh MLI. Para Ketua Cabang MLI di mana pundi seluruh Indonesia, kami mengimbau untuk menggerakkan peran serta dan motivasi paraanggota di tempat masing-masing.

Terima kasih dan salam MLI,Katharina Endriati SukamtoKetua MLI

Bincang antara Kita dari Dunia Maya

100

REKOMENDASI KIMLI 2014

KIMLI 2014, yang bertemakan “Peran Bahasa Ibu dan Bahasa Nasional dalam

Pengembangan Potensi Penutur Bahasa”, dalam empat hari kongresnya di Bandar

Lampung (19-22 Februari 2014) sebagai bentuk kerja sama antara MLI Pusat dan MLI

Cabang Universitas Lampung, menghasilkan rekomendasi berikut.

Dalam beberapa puluh tahun terakhir ini pewarisan bahasa ibu ke generasi muda

mengalami “putus mata rantai”; banyak di antara generasi muda yang tidak lagi

menguasai bahasa ibunya. Padahal bahasa ibu -- bagi anak yang berkembang melalui

bahasa ini -- merupakan pintu pertama untuk mengembangkan potensi akademik serta

potensi penguasaan bahasa kedua, ketiga, dan seterusnya.

Untuk itu marilah kita, seluruh anggota Masyarakat Linguistik Indonesia, mengucapkan

ikrar sebagai berikut:

IKRAR ANGGOTA MASYARAKAT LINGUISTIK INDONESIA

Pada hari ini, Sabtu, 22 Februari 2014, kami anggota MLI berikrar untuk memberi

perhatian kepada bahasa ibu di daerah kita masing-masing, mendorong penggunaannya,

membantu proses pewarisan ke generasi penerus melalui kegiatan pendidikan,

penelitian, dan pengabdian kepada masyarakat.

Bandar Lampung, 22 Februari 2014

Tertanda

Seluruh anggota MLI


101

IURAN KEANGGOTAAN MLIDAN SUMBANGAN PENERBITAN ARTIKEL LI

Pada Kamis, 24 April 2014 14:32, MLI <[email protected]> menulis:

Bincang antara Kita dari Dunia Maya

102

FORMAT PENULISAN NASKAHNaskah diketik dengan menggunakan MS Word dikirimkan ke Redaksi melalui [email protected] atau dalam bentuk disket dan satu printout. Panjang naskah,termasuk daftar pustaka, adalah minimal 15 halaman dan maksimal 30 halaman, denganspasi tunggal dan jenis huruf Times New Roman 11 point. Naskah disertai denganabstrak sekitar 150 kata dan kata kunci (key words) maksimal tiga kata. Abstrak dankata kunci ditulis dalam dua bahasa: bahasa Indonesia dan bahasa Inggris, diletakkansetelah judul naskah dan afiliasi penulis.

Kutipan hendaknya dipadukan dalam kalimat penulis, kecuali bila panjangnyalebih dari tiga baris. Dalam hal ini, kutipan diketik dengan spasi tunggal, menjorok kedalam (indented) sepuluh karakter, letak tengah (centered), dan tanpa tanda petik. Namapenulis yang dirujuk hendaknya ditulis dengan urutan berikut: nama akhir penulis,tahun penerbitan, dan nomor halaman (bila diperlukan); misalnya, (Radford 1997),(Radford 1997:215). Catatan ditulis pada akhir naskah (endnote), tidak pada bagianbawah halaman (footnote).

Setiap rujukan baik artikel maupun buku tanpa dipilah-pilah jenisnya, diurutkanmenurut abjad berdasarkan nama akhir, tanpa diberi nomor urut. Untuk buku: (1) nama akhir, (2) koma, (3) nama pertama, (4) titik, (5) tahun pe-

nerbitan, (6) titik, (7) judul buku cetak miring, (8) titik, (9) kota penerbitan, (10) titikdua (colon), (11) nama penerbit, dan (12) titik, seperti pada contoh berikut:

Gass, Susan M. dan J. Schachter. 1990. Linguistic Perspectives on Second LanguageAcquisition. Cambridge: Cambridge University Press.

Thornbury, Scott. 2005. Beyond the Sentence: Introducing Discourse Analysis.Oxford: Macmillan.

Untuk artikel dalam jurnal: (1) nama akhir, (2) koma, (3) nama pertama, (4) titik, (5)tahun penerbitan, (6) titik, (7) tanda petik buka, (8) judul artikel, (9) titik, (10) tandapetik tutup, (11) nama jurnal cetak miring, (12) volume, (13) titik, (14) nomor (kalauada), (15) koma, (16) spasi, (17) halaman, (18) titik, seperti pada contoh berikut:

Chung, Sandra. 1976. “An Object-Creating Rule in Bahasa Indonesia.” LinguisticInquiry 7.1, 41-87.

Steinhauer, Hein. 1985.“Number in Biak. Counterevidence to Two AllegedLanguage Universals.” Bijdragen Tot De Taal-, Land- En Volkenkunde 141.4,462-485.

Untuk artikel dalam buku: (1) nama akhir, (2) koma, (3) nama pertama, (4) titik, (5)tahun penerbitan, (6) titik, (7) tanda petik buka, (8) judul artikel, (9) titik, (10) tandapetik tutup, (11) berilah kata "Dalam", (12) titik dua, (13) nama editor disusul (ed.),(14) koma, (15) halaman, (16) titik. Buku ini harus pula dirujuk secara lengkap dalamlema tersendiri, seperti pada contoh berikut:

Dardjowidjojo, Soenjono. 2007. “Derajat Keuniversalan dalam Pemerolehan Bahasa.”Dalam: Nasanius (ed.), 233-261.

Nasanius, Yassir. (ed.). 2007. PELBBA 18. Jakarta: Yayasan Obor Indonesia. Jika ada lebih dari satu artikel oleh pengarang yang sama, nama pengarangnya ditulis

ulang secara lengkap, dimulai dengan tahun terbitan yang lebih dulu, mengikuticontoh ini:

Shibatani, Masayoshi. 1977. “Grammatical Relations and Surface Cases.” Language 53,789- 809.

Shibatani, Masayoshi. 1985. “Passives and Related Constructions: A PrototypeAnalysis.” Language 61, 821-848.

MASYARAKAT LINGUISTIK INDONESIA No. 01 Feb 2014.compressed.pdf · Keanggotaan institusional dalam negeri adalah Rp250.000 dan luar negeri US$50per tahun. Naskah dan resensiyang panduannya

Documents