SEMANTIC PROSODY OF WORDS OF EFFECTS IN INDONESIAN

Indonesian Journal of Applied Linguistics, Vol. 5 No. 1, July 2015, pp. 157-172

SEMANTIC PROSODY OF WORDS OF EFFECTS ININDONESIAN

PrihantoroUniversitas Diponegoro

email: [email protected]

Abstract: In a cause-and-effect type sentence, the choice oflexis and grammar are of crucial importance. This paperfocuses on five near synonymous Indonesian lemmas indicatingeffect, which are: <hasil, efek, konsekuensi, dampak and akibat>. InKamus Besar Bahasa Indonesia1 (the online version of the Indonesianreference dictionary used in this study), these lemmas aredescribed without any feature of semantic prosody. Does thismean that the prosody of these words is not important? Mystudy has shown otherwise. I, here, have extracted cause-and-effect sentences from the PAN Localization Corpus2 (thereference corpus employed in this study). The collocates andgrammatical constructions show that the semantic prosody ofhasil is flexible. However, discussion of my finding shows thatthe prosody for the rest of the lemmas tends to be negative.This can be seen from statistics showing lexical preferencesfor words with negative associations and negative grammaticalconstructions where the effects are negative or unexpected.This holds true the four text types in the corpus (economy,sport, science and international affairs). For this reason, Irecommend that the KBBI development team should incorporatethis feature in forthcoming versions of the dictionary.

Keywords: corpus, concordance, collocation, words of effect,semantic prosody

The focus of my study here ison words of effect (as in

‘cause and effect’) in theIndonesian language. The data

1 www.daring.kbbi.co.id2 http://www.panl10n.net/indonesia/

1

Prihantoro, Semantic prosody of words of effects in indonesian

is obtained from the PANLocalization Project Corpus3

(Adriani & Hamam, 2008). Someexamples from the corpus data(1), (2) and (3) are shown asfollows (effect is underlinedin each case):

(1) keterbelakangan diakibatkanoleh sifat-sifat tradisional‘backwardness is caused bytraditionalcharacteristics’

(2) perubahan-perubahan yangdiakibatkan oleh mutasi‘the changes affected bymutations’

(3) kematiannya diakibatkanoleh kegiatan mata-mata bagiAmerika Serikat‘death was a result ofspying for Americans’

The stem akibat ‘effect’ isderived by the {di-kan} confix,resulting in the verbal formdiakibatkan ‘be caused bysomething’. Notice howketerbelakangan ‘backwardness’ inline 1 and kematian ‘death’ inline 3 have the feature ofnegative meaning. Thereduplicated form perubahan-

3 A localization project undertaken by the Indonesian Agency for the Assessment & Applications of Technology, known as BPPT (Badan Pengkajian dan Penerapan Teknologi)

perubahan (the plural form of‘change’) in (2) is actuallyneutral in this isolated co-text4, but the expandedcontext suggests that it isalso negative in meaning. See(4):

(4) Perubahan-perubahan yang diakibatkan oleh mutasihanya akan serupa dengan apayang dialami pendudukHiroshima, Nagasaki, danChernobyl.‘The changes affected bymutations can only belike those experienced bypeople in Hiroshima,Nagasaki, and Chernobyl’.

‘Changes’ in (2) refers tothe extreme side effect(s) ofnuclear radiation. This effectis neither something positivein meaning, nor expected.Those three samples haveillustrated that the word of

4 I here use the term co-text to refer to lexis and grammar contexts,which follows the tradition of corpus linguist. The term ‘context’ is also possible to use, but it might also cover situational contexts such as culture or pragmatics. The use of ‘co-text’ restricts the context to lexis and grammar. The title of this paper itself is self explanatory that the discussion will not go beyond semantics.

2


effect diakibatkan license thepresence of words orconstructions that have thefeature of negativeassociation. The term thatLouw (1993) has proposed forthis phenomeon is ‘semanticprosody’: a word may attractcollocates that have aparticular semantic property(usually characterized bynegative or positive) and bedetermined either by lexis orgrammar. I will review thisconcept later. As for now, itis important to understandthat besides diakibatkan, othernegative semantic prosodysuspects are also present inIndonesian.

In this paper, Iinvestigate five lemmas ofwords of effect that are near-synonyms: <hasil, akibat,konsekuensi, dampak and efek>. Theaims of this study are tounderstand, 1) how the wordsof effect are used in thecorpus, 2) the semanticprosody of words of effect viaits collocates or grammaticalconstruction, 3) how nativeIndonesians use the words ofeffect in sentence context.

Discussion to the findingsin this paper are presented bythe combination of statistics,concordance, collocates and

qualitative description. Ialso conducted a productiontest (with Indonesian nativespeakers as the respondents).The cross check between theproduction test and thefindings from the corpus is acrucial importance beforeconcluding this study. Beforediscussing this further, thereare some previous studiesabout semantic prosodies thatI want to review.

Prosody is a term thatoriginates from the field ofphonetics and phonology, butit is also used in the fieldof corpus linguistics thesedays. While prosody inphonology refers tosuprasegmental features suchas stress, rhythm andintonation (Ladefoged &Johnson, 2011), prosody incorpus linguistics refers tothe tendency of a word topositive or negative polarity(McEnery & Hardie, 2012).Consistent series ofcollocates to this kind ofword are clues to determineits prosody. The attributive‘semantic’ is often attachedto ‘prosody’ as 1) it reliesheavily on sentence context,2) the co-selection isstrongly related to the

3


semantic properties of theword and its collocates.

Through a series ofstudies, Sinclair (2003)concluded that verbs such as‘to cause’, ‘to budge’ and ‘tohappen’ have negative semanticprosody. They tend to co-select words with negativefeatures such as ‘failure’,‘burden’, or ‘expensive’. Thenegativity can also beexpressed by negation such as‘not successful’, ‘has nevermanaged to’ or ‘cannotsurvive’. Sinclair suggestedthat semantic prosody isincluded in headworddescription of StandardEnglish dictionaries5.

Hunston (2007) revisitedthe concept of semanticprosody, and she suggestedthat the prosody might bebound to register. Her findingis also important as registeris also another crucialinstrumentation of meaning.One of her criticisms toSinclair’s analysis is aboutthe semantic prosody of ‘tocause’. She believed that whenit refers to [-HUMAN]features, the prosody is notnecessarily negative.

5 See Sinclair’s proposal for headword BUDGE (Sinclair, 2003, p. 90)

Whitsitt (2005) alsocriticized this concept a bitearlier before Hunston.Although many loopholes arefound in his criticisms(McEnery & Hardie, 2012), oneof his finding seemsimportant. He believes thatsemantic prosody might changeovertime. This might hold trueas diachronic linguisticsstudies have validated thatmening is always negotiated,and after some time, withsocial convention, the meaningitself may shift or evenchange.

The findings of Sinclair,Whitsitt and Hunston areinstrumental to my study. Thedata of the corpus I use hereare from four differentsections of Indonesiannewspapers: economy,international, science andsport. It is interesting toconfirm whether the text typeis instrumental in determiningthe prosody as Hunston (2007)suggested. As for Whitsitt’sfinding (2005) aboutdiachronic change, I need tostate that my focus here is onwords of effect inpresent/modern Indonesian. Buthis finding raise importantconcern for further studies,which is to look over how the

4


use of words of effect maychange from time to time.

I here follow thesystematic investigationmethod of Sinclair (2003) todetermine the semanticprosody, which relies heavilyon both lexis and grammar. Inthe methodology section, Iwill show how some conceptslike co-selection,collocation, and co-text,applies in determining thesemantic prosody of words ofeffect in the corpus I amstudying.

METHODRaw data in this paper isobtained by the automaticretrieval of the derivationsof five lemmas in the corpus:<hasil, efek, konsekuensi, dampak andakibat>. The retrieval isperformed by a corpusprocessing tool, Unitex(Paumier, 2008). As thelexical resource forIndonesian is not yetavailable, I use morphologicalstem based retrieval. Forinstance, the retrieval ofstem <<hasil>>, has resulted onthe concordance lines with thefollowing keywords(derivations) in contexts:menghasilkan, hasilnya, dihasilkan andetc. They are derived by means

of affix(es)6, such as, me-kanconfix (menghasilkan ‘to resulton sth), ber- prefix (berdampak‘to have impact’), -nya suffix(konsekuensinya ‘theconsequence’) and etc. Of theretrieval of those five stems,there are 39 derivations thatcontain strings, where thesurface forms are exactly thesame as the lemma.

One feature of derivationis that it may change themeaning. It means some wordsthat do not fall to words ofeffect category might also beincluded in the result of theretrieval. As the focus ofthis paper is on the words ofeffect, then words that fallbeyond this classificationmust be removed. Words such aspenghasilan ‘income’, keberhasilan‘success’ or efektif ‘effective’are removed as they do notfall to this category.

To investigate the semanticprosody of the words ofeffect, I apply the followingprocedure. First, I extractedconcordances for eachderivational token. Theseconcordances are important asthey provide co-text, wherethe token is used. The part of6 For the description of Indonesian affix(es), see Alwi, et al (1998), Sneddon (2000), Kridalaksana (2007),and Chaer (2008).

5


co-text that is also essentialin determining the semanticprosody, is the collocate(s)of the words of effects. Ithen assigned prosody for eachtoken, particularly evaluativewords, and presented them bycross tabulation. This datawas then statisticallymeasured to identify anyprosodic tendency. Anotherstatistical measure isperformed to investigate thedistribution of words ofeffect across text types.Here, I use AntConc (Anthony,2006) to obtain thedistribution plot and convertthe data, also into a crosstabulation where text typesand words of effect are thevariables. The aim is to showto what extent text types aresignificant to the preferenceof words of effect.

The last stage of thisresearch was a productiontest. Ten Indonesian nativespeakers were randomlyselected to fill out an onlinequestionnaire which I haddevised. In thisquestionnaire, they were asked

to compose one sentence foreach derivational form. I theninvestigated the semanticprosodies, and crosscheckedthem with the prosodies I havepreviously assigned on thecorpus data.

One important thing that Ineed to declare before I beginpresenting my data is aboutthe translation. We willfrequently encounter examplesin Indonesian followed by thetranslation in English. Theidiomatic translation of eachdata is obtained also from thecorpus as the corpus isalready aligned to its Englishtranslation (it is a parallelcorpus).

RESULTS AND DISCUSSIONAs I have commentedpreviously, the raw data hasto go through the filteringbefore it is analyzed further.Several words that are notunder the category of words ofeffect are removed. The resultof the filtering is summarizedin Tables 1 and 2:

Table 1. Non-affix derivationsToken Alignment Samplen.hasil

‘result’Saat ini kami masih menunggu hasil penyelidikan dan pengecekan ulang tata batas areal yang dilakukan Departemen Kehutanan,

6


Now we are waiting for the result of the governmental investigation and the rechecking on the forest borders.

n.konsekuensi

‘consequence’

Anggota senior partai Demokrat di Kongres, Rabu, mencela keputusan Pentagon untuk memperpanjang penggelaran tentara di Irak sebagai konsekuensi kegagalan kebijakan Gedung Putih.Top Democrats in Congress denounced Wednesday the Pentagon's decision to extend troop deployments in Iraq as the consequence of White House policy failures.

n.akibat‘as a

result of’

membentuk tim khusus dari antardepartemen pada tahun 2008 sebagai akibat tingginya harga minyak dunia yang menyentuh 100 dolar AS per barel.establishing a special team in the 2008 as a result of world oil price to reach US$100 per barrel.

n.dampak‘impact’

kami mengharapkan kondisi pasar keuangan global kembali pulih sehingga turut memberikan dampak positif terhadap pasar keuangan domestik, katanya.we hope global financial market conditions would recover and have a positive impact on the domestic market, he said.

n.efek‘effect’

Adukan ini kedap-air dan tahan terhadap efek korosif air.This mortar is water-resistant and is very firm against the corrosive effect of water.

Table 1 shows the result ofthe derivation without theinvolvement of any affix. Ofthose five words of effect ontable 1, you see some of themare similar to English such asefek ‘effect’ and konsekuensi‘consequence’. Badudu (2009)believed that these two wordsare loans from Latin. Jones(2007) on the other hand

believed that they are Dutchloan. Although they disagreedon the origin of efek andkonsekuensi, they both agreedthat hasil and akibat are Arabicloans. However, neither ofthese scholars mentionsanything about dampak. Ibelieve that this is the onlyIndonesian original word, orat least Malay origin. The

7


retrieval of dampak on MalayConcordance Project7 Corpusresults on six occurrences ofdampak including berdampak,berdampaklah and terdampak (noneof the last two words occurredin the corpus used in thispaper) from texts between 1843to 1940. Unfortunately, Ifound no previous works thatrelate to the words of effectin the donor language (Englishor Arabic). Such studies wouldbe useful as the Indonesianprosody might be a form ofinterference of the donorlanguage prosody.

The reference of can belocated on the left or righthand side of the word ofeffect. However, sometimes, weneed to observe the extended8

co-text as well. Considerexample (5), (6) and (7):

(5) Berikut di bawah ini adalahdampak inflasi menurut paraekonom.

7 http://mcp.anu.edu.au8 The reference is not present on phrase or sentence level, but is available on discourse level such asanaphoric, or cataphoric reference of pronoun.

‘The followings are impactsof inflation according toeconomist.’

(6) harga logam timah dunia cenderung menguat sebagaidampak dilakukannya upayapenegakan hukum‘world tin prices had

strengthened since the fourthquarter of 2006 as a result oflaw enforcement’

(7) inflasi yang rendah jugamembawa dampak turunnyapenerimaanlow inflation had also

caused a decline in value-addedtax receipts

In example (5), expandedcontext is required as inflasi isnot the impacts. The impactreference itself is cataphoric(available on the extendedcontext) as shown by thephrase di bawah ini ‘thefollowings’. In example (6),dampak is predicative, andsubject harga logam timah duniacenderung meningkat ‘thestrengthening of world tinprices’ is the effect.. Inexample (7), the effect is onthe right hand side turunnnya

8


penerimaan ‘decline in valueadded taxt receipt’. Now, seetable 2 that shows the

derivations formed byaffix(es) and the location ofthe effects in my description.

Table 2. Derivations by affix(es)Token Alignment Sample

v+tr.menghasilkanto produce

sth’v+pass.dihasilkan‘be resulted

from’

Secara bersama, KPC dan Arutmin menghasilkan lebih dari 53 MT batubara pada tahun 2006Jointly KPC and Arutmin produced more than 53 metric tons of coal in 2006

v.dihasilkannya‘be resulted

by the(+agent)’

Jeruk yang dihasilkannya sangat berkhasiat bagi tubuh, sangat lezat dan harum saat dimakan.The result it yields is extremely useful for the human body, extremely tasty and fragrant when consumed by humans.

conj.alhasil‘as a result’

alhasil, ukuran populasi … dibatasi oleh faktor-faktorlingkungan‘as a result, the population is limited byenvironmental factors

n+def.konsekuensinya ‘the

consequence’

Konsekuensinya pemberdayaan dan pengembangan LKM menjadi salah satu syarat dasar.Consequently, empowerment and development of MFI becomes one of basic requirement.

v+tr.mengakibatkan

‘to cause sth’v+pass.diakibatkan‘be caused by

sth’

Hal ini menurut dia karena krisis keuangan telah mengakibatkan beban biaya fiskal bagi pemerinatah.They had done so considering that past crises had caused their governments to bear huge fiscal cost burdens.

n+def.akibatnya‘the result

Menurut sumber keamanan Palestina, satu orang Palestina terluka sebagai akibatnya.According to a Palestinian security source, one Palestinian was wounded as a result.

v+tr.berakibat‘to impact on

sth’

Menurunnya kerapatan pigmen macular berakibat pada menurunnya kemampuan retina untuk menyaring cahaya yang masuk.

9


Reduction of macular pigment density impacts to decreasing retina ability to filter entering light.

v+tr.berdampak‘to haveimpact on

sth/to affect’

menginggat itu berdampak pada tingkat bunga, nilai tukar dan variable lain, katanya.it could affect the interest rate, currency exchange rate and other variables.

n+def.dampaknya‘the impact’

saya kira dalam dua bulan mendatang dampaknya masih terasa di pasar keuangan, katanya.I think the impact on the financial market will remain in the next two months, he said.

n+def.efeknya‘the effect’

Efeknya mungkin baru terlihat puluhan tahun mendatang.The effect might take place in several decadesto come.

n+intr.berefek‘to affect on

sth’

Polusi aerosol dari Asia Timur berefek besar bagi belahan dunia ini karena adanya transportasi jarak jauh.Aerosol pollution from Asia has a huge impact on this part of the world because it transports over long distances.

Table 2 shows us somederivations that takeaffix(es). Some affixed wordsof the same lemma are putunder the same box such asmeN9-kan and di-kan(mengakibatkan, diakibatkan,

9 N is capitalized here, as it represents the abstract form (phoneme). The surface forms might take n, ng, and m as the orthographical forms, which correspond to [n], [ŋ], and [m] respectively as the allophonic variants of N. This convention has been used widely by Indonesian linguists. See ‘morphophonemic’ in Alwi, et al (1998), Chaer (2008) or Kridalaksana (2007).

menghasilkan, dihasilkan). Prefixdi- indicates passive voice,while meN- indicates activevoice. The stem also takessuffix –kan . The presence ofthis suffix means that theverb is transitive and thepresence of an object (in thiscase the effect) is necessary.See the effects (underlined)in (8) and (9):

(8) krisis keuangan telahmengakibatkan beban biaya fiskalbagi pemerinatah.

10


past crises had causedtheir governments to bear hugefiscal cost burdens.

(9) Zat ini hanya dihasilkan ditempat yang terluka. This substance is produced

only at the location of thewound.

The effect of active voice(meN-) is located on the righthand side. See beban ‘burden’in (8). Here huge fiscal costis the effect of past crises.And, the object of passivevoice is situated on the lefthand side as shown by zat‘substance’. The presence ofthis substance is the effectof the wound.

Another marker of activevoice is prefix beR10- as inberdampak or berakibat. Theeffect is usually located onthe right hand side. However,often the evaluative words ofthe effect is used without thetextual presence of the head.See (10):

10 Similar to N, R also represents the abstract form of orthographical variants r and l , which corresponds to the laterals [r] and [l] as the phonetic forms.

(10) kejadian yang dapatberakibat fatal.‘An incident that maycause something lethal’

The Indonesian version doesnot mention anything about thereference. In the Englishtranslation, ‘something’ isinserted to make the referenceexplicit. However, thepresence of the Indonesianequivalence of ‘something’ isnot important for Indonesianspeakers as they are quitesure that the effect doesexist and is lethal. Thismight be the effect of theword formation as beR- prefixattaches to noun stem whichmeans ‘to have N’.

This phenomenon alsohappens for some words wherethe grammar class is marked by+def (definite) feature. Inexample (11), the nounreference is omitted, and onlythe evaluation ’10 timesworse’ is present:

(11) kalau masyarakat internasionalgagal melakukan itu,

11


konsekuensinya akan 10 kali lebihburuk.‘if the international

community failed to do that theconsequences would be 10 times

worse.’

Here, the presence of nounis unnecessary as konsekuensiitself is the noun, and theevaluative phrase 10 kali lebihburuk ’10 times worse’ is itsmodifier.

Up to now, we havediscussed how to identify theeffect reference from thestructural point of view. Whenwe did so on the previousdiscussion, we know that someof the effect reference or theevaluative word (usuallyadjective) or phrase may bringthe impression that the effectis either something expected,or unexpected depending on thepositive or negativeevaluation that it conceives.

Semantic prosodyI assign prosodies to theextracted data in the corpusby three types: positive,negative and neutral. Positiveprosody means the effect is

positive or comes as expected,while negative prosody meansthat the result is negative orunexpected. Neutral indicatesthat the prosody is neitherpositive nor negative. Thedecision of assigning prosodycomes after observingcollocates of the words ofeffect. Consider example 12 to14:

(12) sistem kehidupan lainnya yangmenghasilkan gerak mekanik

‘all other organicsystems that producemechanical motion’

(13) melepas cadangan devisamengakibatkan rupiah menguat

‘released funds from itsforeign exchange reserves,causing the rupiah to strengthen’

(14) sebagai konsekuensinya,perusahaan penerbangan Nepaltersebut memecat seorang pegawai

‘As a consequence, theNepalese airline fired anemployee’

Example (12) is obtainedfrom a scientific text. Theeffect of the verb menghasilkanis gerak mekanik ‘mechanical

12


motion’. The collocate,motion, is the obligatoryargument as the nature of {-kan} suffix. This collocatesemantically indicates thatthe prosody of menghasilkan isneither positive nor negative.It is simply a resultdeclaration of organicsystems.

As for example (13), thephrase rupiah menguat ‘rupiah tostrengthen’ is positive inprosody. Unlike example (12),in example (13), the wordmengakibatkan (also takes –kansuffix) introduces an objectclause consisting of the nounrupiah (Indonesian currency)and the verb menguat‘strengthen’ where the verb isa positive evaluation, whichmeans that the benefactor,rupiah, strengthens as a result

of releasing the foreignexchange reserve on the lefthand side of mengakibatkan.

In example 14, the word ofeffect is a noun konsekuensithat takes a definite markersuffix-nya (which makes thepresence of effectobligatory). The verb memecat‘to fire’ is negative inmeaning. This collocate makesthe prosody negative in (14).

Collocates here play animportant role in determiningpolarity. Often, the list ofthe collocates of a word ofeffect may point out to whatpolarity the prosody leads..Consider table 3, which is thelist of 10 evaluativecollocates of the nounkonsekuensi. The list isobtained by T-score (performedby AntConc):

Table 3. Ten collocates of hasil by AntConcPositive Collocate Negative Collocatebaik ‘good’menarik ‘interesting’positif ‘positive’gembira ‘happy’laba ‘(bank) interest’puas ‘satisified’sesuai ‘proper’

mengecewakan ‘disapointing’terpuruk ‘went down’

13


Table 3 shows that thecollocates of hasil are mostlypositive. Only two collocatesare negatives, and the resteight collocates give positiveevaluation. Although thetendency of the prosody hereis positive, but I may saythat the prosody of the stemhasil is flexible. This propertyis assigned to words of effect

where they may co-select wordswith either positive ornegative evaluation.

Besides flexible, anotherproperty of the prosody is‘absolute’. ‘Absloute’ is theproperty of words of effect,where all the collocatessuggest either positive ornegative evaluation. Considerconcordance 1:

Concordance 1. konsekuensi1 karena anda memahami bahwa konsekuensi dari kegagalan akan menyedihkan ‘because you understand that the consequences offailure would be grievous…’2 mengalami kenaikan, yang membawa konsekuensi terhadap beban suku bunga ‘were relatively high ….this had affected the interest burdenin the state budget.’3 tapi saya percaya konsekuensi bagi keamanan Amerika akanmenghancurkan ‘but I believe the consequences for Americansecurity would be devastating.4 Hanya konsekuensi tragis akan mengikuti jika kita ‘Only tragic consequences will follow if we…‘5 ekonomi rendah karena memang konsekuensi dari kenaikan harga minyak ‘economic growth …low as a result of the increase in theprices of fuel oils… ‘6 lahir berbagai konsekuensi buruk. Dalam hal tingkat pendapatan ‘emerge various poor consequences. In term ofincome level’7 karena kami memahami konsekuensi kegagalan. Kalau koalisi kami mundur ‘because we understand the consequences of failure. Ifour coalition withdrew…’

14


8 tetap kita lakukan dengan konsekuensi biaya yang lebih mahal ‘We will continue to11 even at highercost 9 mencatat kemungkinan konsekuensi-konsekuensi negatif atas setiap aksi noted several possible negative consequences ofany military action, ‘10 lembaga Kristen yang anti perang. Konsekuensi-konsekuensi dari kemungkinan ‘Christian anti-war charity. The consequences ofany possible…’

The prosody of konsekuensi isnegative for all, as shown by10 concordance lines shown byconcordance 1. See line onemenyedihkan ‘sad’ and kegagalan‘failure’ (also repeated inline 7), beban ‘burden’ inline 3, menghancurkan ‘todestroy’ in line 3, tragis‘tragic’ in line 4, buruk‘bad’ in line 6, mahal‘expensive’ in line 8, andnegatif ‘negative’ in line 9.We find kenaikan ‘raise’ in line5. This word is interesting asit may be positive ornegative. But here, itconfigures itself as themodifier of harga minyak ‘oilprice’. Where the raise ofsalary is expected, the raise

of oil price is unexpected asit may trigger the price ofother items to rise as well.It is therefore negative. Tounderstand why konsekuensi-konsekuensi ‘the plural form ofconsequence’ is negative inline 10, we need to observethe expanded context. Seeexample (15):

(15) Konsekuensi-konsekuensi darikemungkinan serangan militermendatang sama sekali kontra

produktif serta sangat berbahaya.‘The consequences of anypossible future militaryaction could be whollycounter-productive aswell as highly dangerous’

11 As I have commented previously, the idiomatic translation is obtained also from the corpus. Here, konsekuensi is not translated. Comments and evaluation of this translation technique will be discussed in translation studies, which are related but not the focus of this paper.

15


The cause of effect inexample (15) is serangan militer‘military action’. Here, themeaning is already negativethat almost everyone believesthat peaceful resolution isalways prioritized andmilitary action is one lastoption to consider. However,what determines the negativeprosody of konsekuensi-konsekuensiis the evaluation marked bythe predicative compound kontraproduktif ‘counter-productive’and adjective berbahaya‘dangerous’. The adjective iseven modified by anintensifier sangat ‘highly’ toindicate that the dangerbehind the military attack isserious.

So far we have come tothree prosodies; positive,negative and neutral. Besidesthe polarities, anotherproperty of the prosody ofwords of effect here is‘flexible’ or ‘absolute’. Thedata shows that the prosody ofkonsekuensi is negative for alldata. It means that it is

konsekuensi that co-selects itscollocates and not the otherway round. Cognitively,Indonesian speakers would usewords or constructions thatlead to negative effect as thecollocates of konsekuensi or itsderivates.

Unlike ‘absolute’, wordsthat fall to the category of‘flexible’ are not definite inprosody. The collocatesdetermine whether they arepositive or negative (orneutral, as I have commentedpreviously). However, thestatistics showed that theterm ‘flexible’ must beelaborated somehow, as thetendency either may extremelygoes toward one side, positiveor negative. Table 3 has shownthat hasil can be used when theresult is negative orpositive. But the dataindicates that hasil has theextreme tendency towardspositive prosody. On the otherside, dampak seems to have andextreme tendency to negativeprosody:

Concordance 2. dampak1 lebih dari 100 persen, dampak berupa inflasi yang tinggi baru mereda

16


‘…the high inflation rate as the impact ..percentabated only …’2 terpusat pada satu titik tunggal. Dampak kerugian nyata lainnya yakni ‘... all focus onto a single point.The side impact of the eventwas …’3 mengenai manfaat atau dampak buruk dari pemberian oksigen untuk ‘…how useful or harmful oxygen administration is during theresuscitation of newborns’4 dilanjutkan akan mengakibatkan dampak hukum yang serius bagi Antam sebagai ‘continuing … would make Antam as the concession-holder liable tolegal sanctions’5 Menurut Paskah, soal penanganan dampak kerugian Lumpur Lapindo masihdibicarakan ‘Paskah… discussing the material losses caused by the Lapindomudflow…’6 prinsip kehati-hatian atas dampak krisis subprime mortgage ‘still carefully considering the impact of thesubprime mortgage turmoil …’7 atas 10 persen bisa memberikan dampak negatif yang lebih besar above 10 percent will result on worse impact…’8 mortgage', akan memberikan dampak positif terhadap nilai tukar rupiah ‘…mortgage … will … have a positive impact on therupiah's exchange rate..’9 terkendali sehingga akan memberikan dampak positif bagi perekonomian ‘can be kept in check … have a positive impact on theeconomy….10 yang turun akan memberikan dampak positif bagi APBN di mana belanjasubsidi ‘…the drop … would have a positive impact on the statebudget… would reduce… subsidy’.

In concordance 2, theprosody of dampak ‘impact’from line 1-7 is negative.Only three are positive inprosody (they are marked by

attributive adjective positif‘positive’). I believe thatunless they are marked bypositive evaluation word, theprosody of dampak is negative.

17


As I have commentedpreviously, collocates arenot the only clue indetermining the prosody.Consider 16 where theevaluation word separah (fromparah ‘bad’) stands as thecollocate. Here, the collocateseems to be negative, but theprosody is actually positiveas the grammaticalconstruction in the co-textshows that it is negated bytidak ‘not’. The whole phrasetidak separah ‘not as bad as’indicates that the impact ofthe raise of oil price is notas bad as the previous one.

(16) dampak kenaikan BBM tahun initidak separah tahun 2005

‘the impact of fuel oilprice hikes would not beas bad as those in 2005’

These two importantfindings have confirmed thebelief that certain collocatesand particular grammaticalconstructions are helpful indetermining the semanticprosody. Now have a look atalhasil . It is derived fromlemma hasil, but in the

standard Indonesiandictionary, it is a separatelemma with conjunction as itspart of speech (POS) category.This word is hapax legomena: aword that only appears once inthe corpus:

(17) alhasil, ukuran populasi …dibatasi oleh faktor-faktorlingkungan‘as a result, the

population is limited byenvironmental factors

Here, on one timeoccurrence, the prosody isnegative. I cannot decide nowthat the prosody is extremelyor absolutely negative as itonly occurs once in thecorpus. However, my recentsearch by using Google searchhas retrieved sentences wherethe prosody is all negative.Here are the top threesentences that I obtained:

(18) Alhasil, pemerintah masihmemberikan subsidi sebesar Rp1.500 per liter.

‘as a result, thegovernment still has tosubsidize 1.500 per liter’

18


(19) Alhasil, Juventus hanya mampumeraih satu poin dalam laga ini

‘as a result, Juventusonly managed to get one pointfrom this match’

(20) Alhasil, Jakarta harus puasberada di posisi ke-34 dari 40 kotayang dinilai

‘as a result, Jakartamust be satisfied by justbeing the 34th over 40 assessedcities’

The word alhasil, is alwayssentence initial, followed bya comma and the effect (thecause is on the previoussentence). This pattern

(alhasil + , + EFFECT) appliesfor the data either by theGoogle search, or the corpus.We can say here that alhasilintroduces a negative impact.The context in all of thesample sentences I obtainedsuggests that the prosody isnegative. However, since theyare not the part of thecorpus, I decide to excludethis from this research andstudy this word in moredetails for future research.As for now, the details thatwe have previously discussedare summarized by table 4(alhasil is excluded):

Table 4. Prosodic categorization of words of effectToken Prosody absolu

teFlexible

[+] [-] Extremely[+]

Extremely[-]

hasil ‘result’

menghasilkan ‘to result’dihasilkan ‘be resulted’dihasilkannya ‘the resulted’

konsekuensi ‘consequence’

konsekuensinya ‘the consequence’

akibat ‘cause’

mengakibatkan ‘tocause’diakibatkan ‘be caused’

akibatnya ‘the impact’

berakibat ‘to have impact’

dampak ‘impact’

berdampak ‘to have impact’

19


dampaknya ‘the impact’

efek ‘effect’

efeknya ‘the effect’

berefek ‘to have effect’

Table 4 summarized thedetails that we have discussedso far. Among the words ofeffects, there are only twowords where the data showsonly negative prosody;konsekuensi and konsekuensinya.Based on the data, thesemantic prosody of these twowords are absolutely negative.

The prosody of the rest ofthe words is both negative andpositive, which means they areflexible. But there is atendency toward positive ornegative. Words from lemmaakibat, dampak and efek areextremely negative. Thecomparison may reach 70%(negative) – 30% (positive).The frequency of theoccurrence of these words inpositive prosody is scarce inthe corpus. Interestingly,none of the words of effecthas tendency towards positiveprosody.

The words of effect thatare derived from lemma hasil are

more positive in prosodyalthough the gap is relativelysmall as compared to others.The comparison is 65%(positive) and 35% (negative).Besides used for effects under‘neutral’ (neither positivenor negative) category, thiswords of effect is mostly usedfor the effect that is not yetobtained. Consider example 21.

(21) Saat ini kami masih menungguhasil penyelidikan

‘Now we are waiting forthe result of thegovernmentalinvestigation’

The result in example 21 isfuture in aspect; it is notyet known whether it isnegative or positive orneutral. This is the ‘safest’words of effect to use,avoiding writer’s tone to lookobvious. When the words ofeffect of this lemma <hasil>are added together, they rankfirst among other words of

20


effect that are derived from other lemmas. Table 5 confirmsthis:

Table 5. Distribution of words of effects across text typesToken Text Types

Economy

International

Science

Sport

Hasil 'result’ 65 63 112 117berhasil 'successful

'62 40 66 70

menghasilkandihasilkan

dihasilkannya

'to result’'be

resulted’'the

resulted’

1160

2050

97454

2041

konsekuensi 'consequence’

5 13 1 1

konsekuensinya 'theconsequence

’

1 2 0 0

akibat 'cause’ 129 110 57 9mengakibatkan

diakibatkan'to cause’'be caused’

282

563

328

00

akibatnya 'theimpact’

16 4 8 0

berakibat 'to haveimpact’

4 1 4 0

dampak 'impact’ 82 37 25 0berdampak 'to have

impact’19 6 3 0

dampaknya 'theimpact’

7 1 3 0

efek 'effect’ 136 23 0 0efeknya 'the

effect’2 2 0 0

berefek 'to haveeffect’

2 0 0 0

Total 577 386 465 222

21


Table 5 indicates that thederivation by non-affix is themost productive forms for eachlemma. The distribution acrosstext types has shown that allderivations are used exceptfor sport texts. In this typeof text, the only onederivations of extremenegative prosody are usedkonsekuensi (1 hit) and akibat (9hit). Other derivations, whichare extremely negative inprosody are not used. Thewords of effect used in thiscorpus are mostly derived fromstem hasil. As a result, itranks the lowest among othersin terms of the use of wordsof effect.

Validation from ProductionTestThe discussion on corpus datahas shown that the onlyabsolute prosody is forkonsekuensi and konsekuensinya(negative prosody); theprosody for rest of the wordsof effect is flexible. Tovalidate this, I conducted aproduction test. Tenrespondents (Indonesian nativespeakers) participated in thistest by filling out thequestionnaire I devised. It iscomposed of a list of words ofeffect and the respondents arerequired to respond by writinga sentence using the word.Table 6 is the result ofproduction test for one of therespondents:

Table 6. Validation sample from subject 2Token Sentence Production Prosodyhasil 'result’ hasil yang diperoleh adalah meningkatnya

jumlah penderita ebolathe result obtained is the increase on ebola patients

+

menghasilkan|dihasilkan|dihasilkannya

'to result’'be resulted’'the resulted’

pupuk alami menghasilkan panen lebih besarnatural fertilizer results on better crops

+

konsekuensi 'consequen Sebagai konsekuensi, kamu tidak lulus ujian. -

22


ce’ as a consequence, you failedkonsekuensinya

'the consequence’

konsekuensinya adalah turunnya minat belajar.the consequence is low enthusiasm to study

-

akibat 'cause’ Tiga orang tewas sebagai akibat tabrak-lari itu.three died as a result of the hit and run

-

mengakibatkan|diakibatkan

'to cause’'be caused’

meningkatnya kriminalitas diakibatkan oleh naiknya harga BBMthe incrase of cream is resulted by the raise of oil price

-

akibatnya 'the impact’

Akibatnya, pabrik minuman keras itu disegel oleh polisi.as a result, the liquor factory was sealed by the police

-

berakibat 'to have impact’

Lemahnya semangat belajar berakibat pada turunnya nilai.the weak motivation to study has resulted on low grades

-

dampak 'impact’ Banyak desa tenggelam sebagai dampak semburan lumpurmany villages were drown as the impact of mudflood

-

berdampak 'to have impact’

Naiknya harga BBM berdampak pada langkanya pasokanthe raise of oil price impacted onlow supply

-

dampaknya 'the impact’

Dampaknya adalah turunnya perolehan suara.the impact is low vote

-

efek 'effect’ Efek samping yang ditimbulkan cukup membahayakan.the side effect that it causes is dangerous

-

efeknya 'the effect’

Efeknya, tiga orang pemuda meninggal duniathe effect is, three died

-

berefek 'to have effect’

Murahnya harga berefek pada persaingan tak sehatthe low price has an effect of unhealthy competition

-

23


The data in table 6 isobtained from respondent numbertwo as a sample. However, eachrespondent shows the sametendency, which is to useevaluative words as a clue to

prosody. From the productiontest, the result is binary:positive or negative. Neutralprosody is not found. In table 7,I summarized the result for allten respondents:

Table 7. Validation from production testToken Respondents

1 2 3 4 5 6 7 8 9 10

hasil 'result’ + + + + + + + + + +menghasilkan

dihasilkandihasilkannya

'to result’'be

resulted’'the

resulted’

+++

+++

+++

+++

-++

+++

+++

+-+

++-

-++

konsekuensi 'consequence’

- - - - - - - - - -

konsekuensinya 'theconsequence

’

- - - - - - - - - -

akibat 'cause’ - - - - - - - - - +mengakibatkan

diakibatkan'to cause’'be caused’

- - - - - - - - - -

akibatnya 'theimpact’

- - - - - - - - - +

berakibat 'to haveimpact’

- - - - - - - - - +

dampak 'impact’ - - - - - + - - - -berdampak 'to have

impact’- - - - - - - - - -

dampaknya 'theimpact’

- - - - - + - - - -

efek 'effect’ - - - + + - - - - -efeknya 'the

effect’+ - + + - - - - - -

berefek 'to have + - + - + - - - - -

24


effect’

The result presented bytable 7 is similar to some ofthe finding; konsekuensi andkonsekuensinya are alwaysnegative in prosody. Theresult for other words isalmost similar, especially forwords with extremeness towardnegative prosody (akibat, dampak,efek and etc). A littlevariation is observed for thederivations of hasil where theyare mostly positive in prosody(more than 80%). Although thegap between positive andnegative prosody of hasil in thecorpus is not significant, butit complies with the result onthe production test that thepositive prosody of hasil ismore than the negative ones.

CONCLUSIONFindings in this paper hasshown that among five lemmasof words of effect, only lemmahasil that has the orientationtowards positive prosody.Derivations from lemmakonsekuensi are absolutelynegative in prosody, whichmeans they only co-select

effects that are negative inmeaning, poor in impression orunexpected. The rest of thederivations (akibat, dampak, efek)are used for both positive andnegative, but have tendencyover negative prosody as well.The distribution of words ofeffect across text typesindicate that the derivationsof lemma hasil (hasil, menghasilkanand dihasilkan) are the mostfrequently used words ofeffect. Reasons are: 1) it canbe used for negative orpositive effect, 2) the effectis neutral, 3) the effect isnot yet known.

These findings areconfirmed by the nativespeakers’ intuition viaproduction test. Aninteresting result is hasil ismostly used for effects thatare positive. Although thecorpus finding also showssimilar tendency, the gapbetween negative and positiveeffect is not as high as inthe production test. As forfuture studies, it isrecommended that there is a

25


comparison with anothercorpus, or the enrichment ofthe words of effect itself.The main recommendation ofthis study is for thedeveloper of KBBI toincorporate the feature ofsemantic prosody in thedictionary entry ordescription.

REFERENCESAdriani, M., & Hamam, R. (2008). Research report phase 2.1: Final design report

on statistical machine translation network. Jakarta: Badan Pengkajian dan Penerapan Teknologi (BPPT).

Alwi, H., Dardjowidjojo, S., Lapoliwa, H., & Moeliono, M. (1998). Tata bahasa baku bahasa Indonesia (3rd Edition). Jakarta: Balai Pustaka.

Anthony, L. (2006). Concordancing with AntConc: An introduction to tools and techniques in corpus linguistics. JACET Newsletter, pp. 155-185.

Badudu, J.-S. (2009). Kamus kata serapan asing dalam bahasa Indonesia. Jakarta: Penerbit Kompas.

Chaer, A. (2008). Morfologi bahasa Indonesia. Bandung: Rhinneka Cipta.Hunston, S. (2007). Semantic Prosody Revisited. International Journal of

Corpus Linguistics 12(2) , pp. 249-268.Jones, R. (2007). Loan-words in Indonesian and Malay. Leiden: KITLV Press.Kridalaksana, H. (2007). Kelas kata dalam bahasa Indonesia edisi ke lima.

Jakarta: Gramedia Pustaka Utama.Ladefoged, P., & Johnson, K. (2011). A course on phonetics 6th Edition. New

York: Wadsworth Cengage Learning.Louw, W.-E. (1993). Irony in the text or insincerity in the writer?

The diagnostic potential of semantic prosodies. In M. Baker, G. Francis, & E. Tognini-Bonelli, Text and technology: In honour of John Sinclair(pp. 157–176). Amsterdam: John Benjamins.

McEnery, T., & Hardie, A. (2012). Corpus linguistics: Method, theory and practice. Cambridge: Cambridge University Press.

Paumier, S. (2008). Unitex manual. Paris: Universite Paris Est Marne LaValee & LADL.

26


Sinclair, J. (2003). Reading concordance. London: Pearson & Longman.Sneddon, J.-N. (2000). Understanding Indonesian grammar. Sydney: Allen and

Unwin.Whitsitt, S. (2005). A critique of the concept of semantic prosody.

International Journal of Corpus Linguistics 10(3) , pp. 283–305.

27

Documents

SEMANTIC PROSODY OF WORDS OF EFFECTS IN INDONESIAN