Indonesia and the Malay World ISSN: 1363-9811 (Print) 1469-8382 (Online) Journal homepage: http://www.tandfonline.com/loi/cimw20 On the morphosyntax and pragmatics of -in in Colloquial Jakartan Indonesian I Wayan Arka & Nurenzia Yannuar To cite this article: I Wayan Arka & Nurenzia Yannuar (2016): On the morphosyntax and pragmatics of -in in Colloquial Jakartan Indonesian, Indonesia and the Malay World, DOI: 10.1080/13639811.2016.1215129 To link to this article: http://dx.doi.org/10.1080/13639811.2016.1215129 Published online: 13 Sep 2016. Submit your article to this journal View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=cimw20 Download by: [Cornell University Library] Date: 13 September 2016, At: 18:40 INDONESIA AND THE MALAY WORLD, 2016 http://dx.doi.org/10.1080/13639811.2016.1215129 On the morphosyntax and pragmatics of -in in Colloquial Jakartan Indonesian I Wayan Arka and Nurenzia Yannuar ABSTRACT This paper presents a novel description and analysis of the suffix -in in colloquial Jakartan Indonesian (CJI), providing the first explicit account of the syntax-semantics-pragmatics interface of this suffix in Indonesian. It demonstrates that -in has its own place in Indonesian grammar, in which certain morphosyntactic functions overlap with its -i/-kan counterparts in Standard Indonesian. Like -i/-kan, -in is a polysemous suffix; its applicative and causative functions are the product of a complex process involving predicate composition of the root and suffix during affixation. This analysis addresses not only clear cases of functional correspondence between the suffixes, but also the deeper morphosyntactic, semantic and pragmatic constraints responsible for certain intriguing differences between -in and -i/-kan. ARTICLE HISTORY Received 4 February 2015 Accepted 30 June 2016 KEYWORDS Applicative-causative; argument structure; colloquial Jakartan Indonesian; morphosyntax; semantics Introduction Indonesia has a very complex polyglossic situation in which people speak Standard Indonesian (SI) as the high variety (H) of the language and Colloquial Indonesian (CI) as the low variety (L). SI is the most formal and prestigious form of the language and can only be learned through education (Arka 2013). Figure 1 below places CI below SI, marking the ordinary status of CI. It is less prestigious than SI, and minorities with local vernaculars have relatively easy access to it through national and local media. CI develops differently in different regions, and it is often greatly influenced by local languages. One of the most widely used and popular colloquial dialects is one that originated in the capital of the country: Colloquial Jakartan Indonesian (CJI). Jakarta is the centre of the Indonesian economy, politics, media and business, and as a result, the CJI dialect has a powerful influence on other variants across the country. Its speakers include the educated urban population of Jakarta, its neighbouring cities, and other populous cities in Java and Indonesia. It is widespread and is on its way to becoming the everyday language and the homogenous and standard form of informal Indonesian (Sneddon 2006). Proper use of CI or CJI often requires speakers’ socio-pragmatic competence. For example, one should avoid using CI to someone who is of higher status and in formal contexts such as a classroom presentation, a formal speech or a staff meeting. CI has traditionally been identified with informal contexts or conversations, but Indonesian socio-political change has resulted in a looser control over the national language and this has supported the promotion of CI into a domain prescriptively reserved for SI; for example, in teenlit CONTACT I Wayan Arka wayan.arka@anu.edu.au © 2016 Editors, Indonesia and the Malay World 2 I. W. ARKA AND N. YANNUAR Figure 1. The language ecology of Indonesia (Arka 2013: 79). (teen literature), a popular adolescent genre of written literature (Djenar 2012: 37). In such works, CI is used not only in the dialogues of the characters, but also in the narrative. However, unlike in SI, there is no standardised form of writing for both CI and CJI, as words are often written the way they are heard (Sneddon 2006: 14). As a result, a word might be spelled differently: ‘no’ in CJI can be written as nggak, enggak, engga’, ngga’, ga’ or gak (Sneddon 2006: 15). Sneddon (2006: 15) also presents CJI’s simplification of SI spelling and pronunciation. SI’s word-final diphthongs ai and au are shortened into e and o. Thus, SI’s pakai (‘use’) and kalau (‘if’) become pake and kalo in CJI, respectively. Another simplification is apparent in the deletion of consonant initial s, such as in sudah (‘already’) (SI), which becomes udah (CJI) and saja (‘just’) (SI), which becomes aja (CJI) (Sneddon 2006: 18). Wouk (1999) mentions morphophonemic alternations in CJI during verb affixation. Table 1 shows that CJI has richer verbal affixes and that CJI has adopted all SI affixes while also developing its own. CJI has an additional prefix N- in the list of intransitive active, and -ke in the list of non-volitional. It also has the suffix -in as a transitive suffix, while SI only has -i and -kan in its list. The verb morphology of CJI has been discussed extensively in Wouk (1999), who further argues that CJI has established rules and properties of its own and therefore should be treated as a distinct form of Indonesian. Sneddon (2006) provides a comprehensive description of CJI’s grammar, emphasising that any formalised SI grammar will never be able to incorporate CJI grammar. This article focuses on that particular feature in CJI, the applicative1 suffix -in, believed to have originated from Balinese (Sneddon 2006: 30). This suffix is only used in CJI, and does not occur in SI (Wouk 1989; Ewing 2005; Sneddon 2006). Earlier work mostly focused on outlining the basic properties of -in, especially in contrast to its -i/-kan counterparts in SI. Wouk (1999) proposes that -in in Betawi2 is a transitive suffix that is able to neutralise the 1 An applicative affix is broadly, a transitiviser forming a transitive verb from an intransitive verb. However, it can also operate on a transitive to sustain its transitivity (see Aikhenvald 2015: 145). INDONESIA AND THE MALAY WORLD 3 Table 1. Affixes in SI and CJI (adapted from Wouk 1999: 74). Intransitives:a Stative Active Standard Indonesian (SI) Colloquial Jakartan Indonesian (CJI) berbernyanyi berenang meNmenyapa menyerah menyuruh ber-, Øbernyanyi, nyanyi berenang, renang meN-, N-, Ømenyapa, nyapa, sapa menyerah, nyerah, serah menyuruh, nyuruh, suruh Transitivesb Prefixes Gloss ‘sing’ ‘swim’ ‘greet’ ‘surrender’ ‘ask’ meN-, dimeN-, N-, di-, Ømenyapu menyapu, nyapu ‘sweep’ memakan memakan, makan ‘eat’ diborgol diborgol ‘handcuff’ Suffixes -i, -kan -i, -kan, -in memarahi memarahi, marahin ‘angry’ menerjemahkan menerjemahkan, nerjemahin ‘translate’ ter-, ke-an ter-, ke-an, keNon-volitionalc terbuang terbuang, kebuang ‘throw away’ ketinggalan ketinggalan, ketinggal ‘left behind’ a An intransitive verb expresses a one-participant event (i.e. having no object). Wouk (1999) labels meng- as an active verb, and ber- as a stative verb for the sake of practicality. b A transitive verb expresses an event with subject and an object. In SI, prefix meN- is used to denote an active voice, while di- is used to denote a passive voice. c Wouk (1999: 72) uses non-volitional as a term to describe ‘verb forms which prototypically indicate a lack of volition or lack of deliberateness on the part of the actor’. semantic distinctions maintained by -i and -kan in SI. While this observation is generally correct, we will show here that the neutralisation is in certain cases surprisingly prohibited. We investigated this and propose an explicit analysis within LFG (Lexical Functional Grammar) to account for both the expected and unexpected behavioural properties of -in and -i/-kan correspondences. Our study highlights the significance of lexical classes in valence-changing3 morpholexical operation and contributes to existing typological and theoretical studies in causativisation and applicativisation (Kroeger 1988; Guzman 1992; Wechsler and Arka 1998; Donohue 2001; Austin 2005; Good 2005; Heusinger et al. 2011). The article is organised as follows. We begin by carefully discussing the properties of -in, focusing on numerous cases of expected and unexpected correspondences of -in and -i/-kan. The detailed description will then lead to the analysis and discussion sections towards the end of the article, in which we present our LFG-based analysis. Background on Indonesian and the issues Indonesian is an SVO (subject-verb-object) language, with words in SI typically appearing in morphologically longer forms than their CJI counterparts. For example, the actor voice 2 Grijns (1998) refers to -in as similar to -kan in SI. Wouk (1999) argues that -in in CJI originates from Betawi although it was actually borrowed from Balinese (Sneddon, 2006: 30), and that CJI is formed through a koineisation process; that is, a slow and steady process resulting in a merger of more than one linguistic (sub)system because of continuous contact and therefore increased interaction or integration among speakers of different languages (cf. Siegel 1985: 376). In the case of CJI, the languages involved in the contact include Indonesian/Malay, Javanese, and Balinese (and possibly others, subject to further research). 3 The number of core arguments required by a verb describes its valency. The valency-changing process includes the increase of valency (causative, applicative) and the decrease of valency (passive, anti-passive), as defined in Aikhenvald (2015: 331). 4 I. W. ARKA AND N. YANNUAR (AV) meN- in SI corresponds to the shorter form N- or is simply dropped in CJI, e.g. menanam (’>’ (A) U:LOC V-kan | ‘PRED1< ARG1, ARG2, ‘PRED2 < __ , … >’>’ (A) U:NonLOC V-in | ‘PRED1< ARG1, ARG2, ‘PRED2 < __ , … >’>’ (A) U:NonLOC > LOC[+def] speaker(i), addressee(j) background {informal(i,j)} The following points should be noted. Firstly, all of these suffixes are three-place (head) predicates. This means that A (actor, ARG1) acts upon ARG2, which brings about the embedded state of affairs (PRED2). The advantage of using this structure is that we can account for the fact that the same marker can be associated with the causative and applicative, and that the syntactic effect of the applicativisation/causativisation is transitivisation. We have seen that -i/-kan/-in can all function as applicative and causative. Precisely how this is achieved by means of argument fusion is further discussed below. The suffixes -i/-kan/-in differ in the nature of the role linking of ARG2; see (29). This is to capture empirical patterns discussed previously. That is, -i and -kan are in complementary distribution with each other: -i is locative-related, whereas -kan is not. The notation of U:Loc associated with -i (29a), for example, means that there is a linking constraint that requires the ARG2 of -i to be semantically locative-related (locative, goal or source). Importantly, -in has its ARG2 linked to either a locative or a non-locative, capturing the fact that it can correspond to either -i or -kan, as discussed in earlier sections. In addition, -in has a crucial preference constraint that the non-loc is preferred over the locative under certain conditions, which accounts for the contrast shown in (27). We informally represent the definiteness constraint by specifying [+def] to the locative argument to capture the idea that a definite locative would be acceptable [cf. example (27b)]. Another key difference between -in and -i/-kan is the socio-pragmatic structure. The constraint in (28c) says that -in is used when the background context of the speakerhearer interaction is informal. Note that there is no such constraint associated with -i/kan, correctly capturing the fact that -i/-kan is unmarked as far as formality is concerned. That is, -i or -kan is usable for both formal and informal situations, whereas -in cannot be used in a formal context. 16 I. W. ARKA AND N. YANNUAR Having outlined the key points of the argument structure and the semantic role-linking constraints of -i/-kan/-in, we are now ready to demonstrate how the analysis works. In the interest of space, we focus on argument fusion and the related semantic constraint for certain interesting but intriguing patterns, which appear to be unexpected. Before doing so, however, principles of argument fusion will be explained. Argument fusion At the heart of the a-str (argument structure)-based analysis adopted in this article is that argument fusion during predicate composition is due to affixation. The principles discussed in Arka et al. (2009) include underspecification and thematic harmony in fusion. 30 a. b. Fusion is underspecified: it is single or double, giving rise to applicative or causative readings. Thematic role harmony in fusion: arguments of similar types tend to fuse. As an illustration, consider the derivation of tembak-in (‘shoot’), exemplified in (23b) and (24b). The -in verb is ambiguous between two sub-categorisation frames, each of which corresponds to the sub-categorisation of its -kan and -i counterpart, (tembakkan/ tembaki). The correspondence with its -in ambiguity is well captured by the proposed analysis. Consider the argument fusion involved in the -in/-kan/-i derivation with the stem tembak (‘shoot’): 31 a. b. -in/-kan < ARG1_i, ARG2, ‘shoot< agent_i, goal/pt> U:instr -in/-i < ARG1_i, ARG2, ‘shoot< agent_i, goal/pt_j> U:goal/pt_j [cf. (23)] [cf. (24)] As seen in (31), with the same underspecified argument structure, the ambiguity in tembakin arises due to two possibilities of argument fusion with -in. In both cases, agent-like arguments fuse (i.e. ARG1 with the stem’s agent). The difference is associated with the fusion of ARG2. In the first option (31a), ARG2 is the instr, not part of the a-str of the stem tembak; i.e., it is newly introduced by -in. This is also the a-str of -kan. In the second option, ARG2 fuses with the goal/pt of the stem. This is also the a-str of -i. In short, our analysis can capture the correspondences of -in with -kan and -i and the associated patterns of (un)ambiguity. Intransitive stems and function blocking There is an intriguing pattern whereby one type of applicative is blocked. For example, the -in affixation of kencing (‘urinate’) to derive kencingin can only be interpreted as locative applicative only (i.e. -in corresponds to -i), not as a theme applicative -in (corresponding to -kan). In other words, the non-locative (displaced-) theme applicative is blocked, even though with other verbs, -in can correspond to -kan for non-locative applicativisation (e.g. bicara becomes bicarain/bicarakan), or for causative -kan when masuk becomes masukin/ masukkan. However, with certain other intransitive stems such as bicara (‘talk’), the goal-locative applicative function is blocked: bicarain means (‘talk about X [X= a theme]’), not (‘talk to INDONESIA AND THE MALAY WORLD 17 X [X= a goal]’). The fact that different stems give rise to different kinds of blocking of the same suffix strongly suggests that the blocking has something to do with the properties of the stem. We now flesh out our analysis, making explicit the properties of the stem responsible for the function blocking. There are two related properties of the stem responsible for this blocking: the transitivity of the stem and the nature of the theme role in relation to the meaning of the verb. The blocking is observed mainly with intransitive bases and in cases involving theme and goal-locative roles. The constraints make sense in the predicate composition analysis as ARG2 is to be fused with the argument of the stem; otherwise, if there is no argument from the stem’s structure to fuse with, the role of ARG is newly introduced (i.e. not part of the stem’s role). An important consideration is whether or not the specific theme is a semantic constant of the stem. A semantic constant is a conceptual component of the verb, in contrast to a variable (Jackendoff 1990; Carter et al. 1988); e.g. in the causative meaning of ‘kill’, informally decomposed into [CAUSE x, BECOME(y, DEAD(y))], the conceptual elements of CAUSE, BECOME and DEAD are constants, whereas x and y are variables. In the following list of intransitive verbs, the theme has a different semantic status in the meaning of the verbs. 32 Class 1: Class 2: (a) kencing ‘urinate’: invariable constant theme (b) kentut ‘fart’: invariable constant theme (c) berak ‘defecate’: invariable constant theme (d) omong ‘talk/speak’: variable theme (e) bicara ‘talk/speak’: variable theme (f) rembung ‘discuss’: variable theme That is, for verbs in (a)-(c), the theme is unexpressed and is understood without variability. In contrast, the theme of the verbs in (d) and (f) can be unexpressed but allow for variability, e.g. the thing spoken or asked can be of different kinds. The verbs of the first type are expected to behave differently from the second type with respect to the applicativisation, because applicativisation affects the theme or the goal/locative, making it a variable to be overtly linked to a syntactic function. This is indeed the case. In terms of our proposed a-str-based analysis, the different behaviours, including the blocking, can be explained as follows. We will discuss the first type, which is exemplified by kencing, then the second type, which is exemplified by bicara. Consider the following examples with the root kencing (‘urinate’), which exemplify the blocking of the theme applicativisation. 33 34 Die ngencing-in darah itu. 3s AV.urinate-in blood that (i) ‘S/he urinated onto the (pool of the) blood.’ (ii) NOT FOR: ‘S/he urinated the blood.’ Dia mengencing-i/*kan darah itu. 3s AV.urinate-I/*KAN blood that (i) ‘S/he urinated onto the (pool of the) blood.’ (ii) NOT FOR: ‘S/he urinated the blood.’ (CJI) (locative: √) (theme: ×) (SI) (locative: √) (theme: ×) As seen in (33), -in can only function as a locative applicative reading (-i). The theme applicativisation is blocked. That is, ‘blood’ is not consistent with the nature of the theme specified by the verb ‘urinate’; hence the applied object ‘blood’ is forced to be interpreted as a locative, as seen in reading (i). The -in applicative corresponds to -i, not -kan in standard Indonesian, (as 18 I. W. ARKA AND N. YANNUAR Figure 2. Predicate composition of kencing-in shown in (34) and in which -kan is not acceptable). Thus, there is a deep semantic constraint that applies across the informal and formal varieties that blocks the theme applicativisation. In terms of the proposed a-str-based analysis, the semantic constraint can be captured in the derivation shown in Figure 2. The process can be explained as follows. The root kencing is intransitive; it only has a single core argument, actor-like (A), which fuses with the first argument of -in. (Fusion is indicated by a connecting line.) The understood theme (i.e. the excretion) is a semantic constant not shown in the a-str representation as it is not a variable (or semantic argument) of the stem. The theme is therefore not available for fusion with the second argument of -in. In effect, the second argument is left to be identified with a locative-related role. Surely, other roles such as source or instrument are excluded from the event of urinating, perhaps also for semantic reasons (i.e. the nature of the event of urinating, with urine deposited/directed to a location of some kind and naturally not involving any instrument). In addition to the semantic/syntactic information, Figure 2 also captures the sociopragmatic background information carried by -in. The notation background {informal (spk, addrs)} means that the -in structure is used in an informal speaker-addressee interaction. We now move on to the reverse situation associated with class 2 verbs, exemplified by the verb bicara (‘talk’), in which the goal applicative function is blocked. Consider the non-applicative sentences in (35) where the goal argument is marked by the preposition dengan (SI) and ama (CJI), and the corresponding applicative sentences in (36). As seen in the free translation, the expected applied goal argument reading i.e. reading (ii) is not acceptable. The understood applied object must be semantically understood as a theme [reading (i)]. In other words, the theme applicativisation blocks the goal applicativisation. 35 a. b. c. d. Saya tidak ber-bicara 1s NEG MID-talk ‘I did not talk to other people.’ Gue nggak bicara 1s NEG talk ‘I did not talk to other people.’ Saya tidak ber-bicara 1s NEG talk ‘I did not talk about other people.’ Gue nggak bicara 1s NEG talk ‘I did not talk about other people.’ dengan with orang lain person other (SI) (goal Oblique) ama with orang lain person other (CJI) (goal Oblique) tentang about orang lain person other (SI) (theme Oblique) tentang about orang lain person other (CJI) (theme Oblique) INDONESIA AND THE MALAY WORLD 19 Figure 3. Predicate composition of bicara-in 36 a. b. Saya tidak membicara-kan/-i orang lain 1s NEG AV.talk-kan/-i person other A P:theme/*goal i) ‘I did not talk about other people.’ ii) NOT FOR ‘I did not talk to other people.’ Gue nggak bicara-in orang lain 1s NEG talk-in person other A P:theme/*goal i) ‘I did not talk about other people.’ ii) NOT FOR ‘I did not talk to other people.’ (SI) (theme: √) (goal/locative: ×) (CJI) (theme: √) (goal/locative: ×) We can capture this blocking property as a theme over goal preference, part of the general preference constraint of NonLoc>Loc of the applicative suffix. It can be represented in the lexical entry of the suffix. Its application is triggered by the semantic class of the verbal base. The derivation of bicarain from bicara is shown in Figure 3. The representation in Figure 3 shows that, unlike the verb kencing, the verb bicara has understood variable arguments (theme and goal), although not always expressed syntactically. The theme is therefore available for the fusion or identification with ARG2 of -in, which is then realised as the applied object in the AV verb bicarain (36b). Furthermore, the verb bicara triggers the activation of the preference constraint of NonLoc over Goal/Loc. This preference constraint can be thought of as being motivated by the general canonical transitivity constraint wherein a second argument/object should be more patient/theme-like than goal-locative-like. In other words, the theme/patient applicative blocks the goal/locative applicative.7 Conclusion We have presented an integrated analysis of the suffix -in, providing the first explicit account of the syntax-semantics-pragmatics interface of this suffix and the ways in which it is similar and different from its -i/-kan counterparts in Indonesian. The suffix -in is only used in colloquial Indonesian. Like its counterparts -i and -kan, it functions as both an applicative and causative suffix. We claim that -in is, like -i/kan (Arka et al. 2009), a polysemous morpheme that allows alternative argument fusions, which gives rise to its varying functions (causative/applicative) and within 7 This might be a ‘soft’ constraint (cf. Bresnan et al. 2001) in that, given the right context, one might still be able to have the locative applicative reading for (36b). 20 I. W. ARKA AND N. YANNUAR its applicative function, it allows a range of applicative types (instrumental, beneficiary, goal, locative and theme). Our analysis is couched within a parallel-based LFG model (Bresnan 2001; Dalrymple 2001, among others). It makes use of recent advances in the theory of argument structure (and function mapping) as applied to Indonesian-type languages (Musgrave 2001; Arka and Manning, 2008). It is augmented by drawing on the insight of the HPSG (Head-Driven Phrase Structure Grammar) framework (Pollard and Sag 1994; Sag et al. 2003) to represent the locus for sociolinguistic/pragmatic information. We have demonstrated how our analysis can address not only the clear cases of functional correspondence between the suffixes, but also the deeper morphosyntactic/semantic constraints responsible for certain intriguing differences and similarities between -in and -i/-kan. In this final section, we highlight the significance of our study, firstly in Indonesian linguistics, then in the study of applicatives/causatives in wider typological and theoretical contexts. The novel contribution of our study of -in in Indonesian linguistics is both empirical and analytical. Empirically, we provide fresh data showing the most complete patterns of the correspondences and non-correspondences of -in and -i/kan, especially the intriguingly unexpected ones. Analytically, our article provides the first integrated argument-structure-based analysis that explicitly captures the morphosyntactic-semantic properties of -in, e.g. different applicative types, ambiguity and its semantic blocking effect on certain applicative types. Overall, our description and analysis of -in deepens the understanding of the complex properties of -in in Indonesian grammar. Given the properties of -in as discussed here, we argue that -in has its own place in Indonesian grammar on a par with -i/-kan. Its function overlaps with -i/-kan syntactically but, as we have seen, it is not the case that -in always corresponds to -i/-kan morphosyntactically. Crucially, it is a dedicated register marker, signifying an informal communicative context. While originating from Jakartan Indonesian, -in is now used widely across Indonesia in informal contexts, especially by young Indonesian speakers. The grammatical and sociolinguistic constraints of this suffix as discussed in this article are arguably now part of speakers’ Indonesian grammar. Causative/applicative phenomena have long been of typological and theoretical interest (Kroeger 1988; Comrie and Polinsky 1993; Alsina 1992; Cole and Son 2004; Austin 2005; Good 2005; Butt and King 2006; Heusinger et al. 2011, among others). The -in applicative is perhaps unusual in its applicative/causative marking; there has been no report in the literature that we are aware of an applicative/causative marker dedicated to a particular register. Its behavioural properties, such as its polysemy, are not unique, and as such causative/applicative polysemy is also the property of its -i/-kan counterparts and also a property encountered in many other languages (Austin 2005). We have also observed that -in allows a large range of applicative types. Typologically, languages commonly have different markers for different types of applicatives. However, we have pointed out that -in can have a blocking effect and that this is lexically determined (i.e. associated with certain stems). This provides the justification for our proposed argument-structure analysis, which highlights the significance of predicate composition with principled argument fusion. Theoretically, there have been competing frameworks by which causative/applicative phenomena can be analysed. The LFG-based parallel model of grammar, as adopted in INDONESIA AND THE MALAY WORLD 21 here, allows the flexibility to capture not only the intricacies of the grammatical properties of -in, but also its socio-pragmatic information in an integrated way. We have treated this social information as part of a layer that is distinct from, but related to, the argument structure. LFG is a stable and mathematically well understood constraint-based theory of linguistic structure (Kaplan 1982; Dalrymple 2001; Bresnan 2001). The framework demands a highly explicit and precise linguistic analysis that is testable and computationally implementable. The LFG-based computational grammar of Indonesian has been developed with the -i/-kan analysis already incorporated into its current grammar (Arka et al. 2009). Extending the current coverage of the grammar so that it can handle a range of registers, including colloquial Indonesian, is a huge task (given the complexity of such a grammar). However, our findings and analysis represent a significant step in that direction. Abbreviations 1, 2, 3 Adj A APPL ARG AV BEN/ben CAUS CI CJI DEF G/go INCL instr ITR LFG loc N NP OBJ OBL P/pt PASS PL POSS PP prag PRED PROG R/rec REDUP s SI src str SUBJ SVO T/th TR V First, second, third person Adjective Agent Applicative Argument Actor voice Benefactive Causative Colloquial Indonesian Colloquial Jakartan Indonesian Definite Goal Inclusive instrument Intransitive Lexical functional grammar Locative Noun Noun Phrase Object Oblique Patient Passive Plural Possessive Prepositional phrase Pragmatic Predicate Progressive Recipient Reduplication Singular Standard Indonesian Source Structure Subject Subject verb object Theme Transitive Verb 22 I. W. ARKA AND N. YANNUAR Acknowledgements Arka gratefully acknowledges the support of the ARC Discovery Grant (DP0877595) and the Humboldt Georg Forster Fellowship. Nurenzia gratefully acknowledges the scholarship support from the Wurm Funds. For helpful discussions, we thank Simon Musgrave and the audience members at the ALS2014, Melbourne. We also thank the two anonymous reviewers of Indonesia and the Malay World for their useful feedback. For data checking, we thank the following CJI speakers: Nazarudin, Derry Patradewa, Dirgantara Reksa, Melita Tarisa, and Tika Ramadhini. Author biography I Wayan Arka is a senior fellow and associate professor in linguistics at the College of Asia and the Pacific, the Australian National University, and a lecturer at Udayana University, Bali. His interests are in descriptive, theoretical and typological aspects of Austronesian and Papuan languages of Indonesia. Email: wayan.arka@anu.edu.au Nurenzia Yannuar is a faculty member of the Department of English, Universitas Negeri Malang, Indonesia. She is currently a PhD candidate at Leiden University Centre for Linguistics, the Netherlands. Her interests include sociolinguistics and youth languages. Email: nurenzia.yannuar.fs@um. ac.id References Aikhenvald, A. Y. 2015. The art of grammar: a practical guide. Oxford: Oxford University Press. Alsina, A. 1992. On the argument structure of causatives. Linguistic Inquiry 23 (4): 517–55. Alsina, A. 1996. The role of argument structure in grammar. Stanford CA: CSLI. Arka, I. W. 2003. Balinese morphosyntax: a lexical-functional approach. Canberra: Pacific Linguistics. Arka, I.W. 2005. Speech levels, social predicates and pragmatic structure in Balinese: a lexical approach. Pragmatics 15 (2/3): 169–203. Arka, I.W. 2013. Language management and minority language maintenance in (eastern) Indonesia: strategic issues. Language Documentation & Conservation (7): 74–105. Arka, I. W. and Manning, C. 2008. Voice and grammatical relations in Indonesian: a new perspective. In P.K. Austin and S. Musgrave (eds), Voice and grammatical relations in Austronesian languages. Stanford CA: CSLI, pp. 45–69. Arka, I.W., et al. 2009. A linguistic and computational morphosyntactic analysis for the applicative -i in Indonesian. In M. Butt and T.H. King (eds), The proceedings of the LFG 09 conference, . Artawa, I K. 1994. Ergativity and Balinese syntax. PhD thesis, La Trobe University Austin, P. 2005 [1997]. Causatives and applicatives in Australian Aboriginal languages. In K. Matsumura and T. Hayasi (eds), Dative and related phenomena. Tokyo: Hitsuji Shobo, pp.165–225. Bresnan, J. 2001. Lexical-functional syntax. Malden MA: Blackwell. Bresnan, J., Dingare, S., and Manning, C. 2001. Soft constraints mirror hard constraints: voice and person in English and Lummi. In M. Butt and T. H. King (eds), Proceedings of the LFG 01 conference, University of Hong Kong, Stanford CA: CSLI Publications, . Butt, M. 1995. The structure of complex predicates in Urdu. Stanford CA: CSLI Publications. Butt, M. and King, T.H. 2006. Restriction for morphological valency alternations: the Urdu causative. In R.M. Kaplan (ed.), Intelligent linguistic architectures: variations on themes. Stanford CA: CSLI Publications. Carter, R.J., Levin, B.C., and Tenny, C.L. 1988. On linking papers by Richard Carter. Cambridge MA: Center for Cognitive Science Lexicon Project MIT. INDONESIA AND THE MALAY WORLD 23 Cole, P. and Son, M. 2004. The argument structure of verbs with the suffix -kan in Indonesian. Oceanic Linguistics 43 (2): 339–64. Comrie, B. and Polinsky, M. (eds). 1993. Causatives and transitivity. Amsterdam: John Benjamins Publishing Company. Clynes, A. 1995. Topics in the phonology and morphosyntax of Balinese. PhD dissertation, Australian National University. Dalrymple, M. 2001. Lexical functional grammar. Syntax and Semantics (34). New York: Academic Press. Djenar, D.N. 2012. Almost unbridled: Indonesian youth language and its critics. South East Asia Research 20 (1): 35–51. Donohue, M. 2001. Coding choices in argument structure: Austronesian applicatives in texts. Studies in Language 25 (2): 217–54. Englebretson, R. 2003. Searching for structure. Amsterdam: John Benjamins Publishing Company. Ewing, M.C. 2005. Colloquial Indonesian. In A. Adelaar and N. P. Himmelmann (eds), The Austronesian languages of Asia and Madagascar. New York: Routledge. Good, J. 2005. Reconstructing morpheme order in Bantu: the case of causativization and applicativization. Diachronica 22 (1): 3–57. Grijns, C.D. 1998. Between brackets: on ‘vocabulary building’ in Batavia circa 1930. In M. Janse (ed.), Productivity and creativity: studies in general and descriptive linguistics in honour of E.M. Uhlenbeck. Berlin: Mouton de Gruyter. Guzman, V.P. de. 1992. Morphological evidence for primacy of patient as subject in Tagalog. Series A. Occasional Papers 82. Canberra: Pacific Linguistics, pp. 87–96. Heusinger, K. von, Klein, U., and Guntsetseg, D. 2011. The case of accusative embedded subjects in Mongolian. Lingua 121: 48–59. Jackendoff, R. 1990. Semantic structures. Cambridge MA: MIT Press. Kaplan, R. M. 1982. Determinism and nondeterminism in modelling psycholinguistic processes. Paper presented at the Conference on Linguistic Theory and Psychological Reality Revisited, Princeton University. Kaswanti Purwo, B. 1997. The direct object in bi-transitive clauses in Indonesian. In M. Shibatani (ed.), Grammatical relations: a functional perspective. Amsterdam: John Benjamins Publishing Company. Kroeger, P.R. 1988. Case marking in Kimaragang causative constructions. Western Austronesian Linguistics 3: 241–76. Musgrave, S. 2001. Non-subject arguments in Indonesian. PhD thesis, Melbourne University. Pollard, C. and Sag, I.A. 1994. Head-driven phrase structure grammar. Chicago: University of Chicago Press. Sag, I.A., Wasow, T.A., and Bender, E. M. 2003. Syntactic theory: a formal introduction. Stanford CA: CSLI Publications. Siegel, J. 1985. Koines and koineization. Language in Society 14 (3): 357–78. Sneddon, J.N. 2006. Colloquial Jakartan Indonesian. Canberra: Pacific Linguistics. Sneddon, J., Adelaar, A., Djenar, D., and Ewing, M. 2010. Indonesian reference grammar. Crows Nest: Allen and Unwin. Wechsler, S. and Arka, I.W. 1998. Syntactic ergativity in Balinese: an argument structure based theory. Natural Language & Linguistic Theory 16 (2): 387–441. Wouk, F. 1999. Dialect contact and koineization in Jakarta, Indonesia. Language Sciences 21: 61–86. Wouk, F. 1989. The use of verb morphology in spoken Jakarta Indonesia. PhD dissertation. University of California Los Angeles.