Accueil

Syntaxe et sémantique

 

Michèle Conjeaud
INALCO
De l'usage de จะ(cà) dans la langue siamoise (ou thaï)
 
cà est défini dans le premier dictionnaire et la première grammaire publiés (Pallegoix, 1850) comme le "marqueur du futur" de la langue thaï. Cette opinion est encore partagée par la plupart des dictionnaires et grammaires actuels. Malheureusement beaucoup des usages de cà observables sont difficilement compréhensibles si on se contente d'une telle définition.
Un certain nombre de linguistes et de grammairiens se sont penchés sur le problème de cà avec comme objectif d'affiner et de compléter cette caractérisation en termes de "marqueur de futur", par exemple, en y adjoignant quelques précisions, telles que "futur absolu et relatif" (Kanchanavan, 1978), "futur à valeur modale désidérative" (Gsell, 1997). D'autres chercheurs, eux, contestent, implicitement ou explicitement, le rattachement de cà aux marqueurs temporels. Ils considèrent, pour certains, que cà appartient aux marqueurs d'aspect en lui attribuant par exemple une valeur de "marque d'une étape préliminaire d'un procès ou d'un état" (Muansuwan 2002) ou de "prospectif" (Hennequin, 1998). Pour les autres, ils s'accordent pour affirmer que cà est un marqueur de modalité et ils lui attribuent différentes valeurs, telles que "putative action" (Noss 1964), "non-assertive modal" (Rangkupan, 2001), "dispositionnal necessity, epistemic necessity, epistemic possibility, deontic necessity and deontic possibility" (Srioutai, 2004), "challengeability marker" (Iwasaki et Ingkaphirom, 2005). Malgré les avancées certaines dans la connaissance de cà qu'apportent ces études, les hypothèses faites et les résultats obtenus sont insuffisants pour expliquer tous les usages de cà que nous connaissions et, a fortiori, pour répondre à la question que nous nous posions : "quand doit-on, quand peut-on, et aussi quand ne doit-on pas mettre cà dans une phrase ?" 
Notreh étude aborde le problème du morphème cà dans une optique différente de celle des études faites jusqu'à ce jour sur ce morphème, tant en ce qui concerne son objectif que la méthodologie utilisée. Le but poursuivi n'est plus de placer cà dans l'une des grandes catégories grammaticales, Temps, Aspect ou Modalité, mais d'essayer de trouver les règles qui gouvernent ses multiples usages. Quant à la méthodologie, celle choisie est issue de la Théorie des Opérations Prédicatives et Enonciatives (TOPE) développée par Antoine Culioli. En outre, une approche paradigmatique est utilisée pour dégager les valeurs sémantiques de chaque occurrence de cà du corpus aussi objectivement que possible. En se fondant sur cette analyse, nous avançons une hypothèse sur l’identité sémantique pour le morphème cà avec trois grands domaines de variation. Sur cette base, nous décrivons les valeurs sémantiques locales de cà comme résultant de l'interaction entre l'identité sémantique que nous proposons, le co-texte et le contexte.
Bibliographie
GSELL, René (1997). On verb serialisation in standard thai. (in) IVth International Conference of Far East, Southeast Asia and West-African : Grammar and Lexicon, Moskow, September 1997. 13p.
HENNEQUIN, Laurent (1998) Les relations syntaxiques dans la langue thai, thèse. Paris : INALCO. 474 p.
IWASAKI, Shoichi & INGKAPHIROM, Preeya (1995) A Reference Grammar of Thai. Cambridge : Cambridge University Press. 392 p. 
KANCHANAWAN, Nitaya (1978) Expression for time in the thai verb and its application to thai-english machine translation. Ph.D. Dissertation : The University of Texas at Austin. 380 p.
MUANSUWAN, Nuttanart (2002) Verb Complexes in Thai. Ph.D. Dissertation : University of Buffalo, The State University of New-York. 260 p.
NOSS, Richard B. (1964) Thai reference grammar. Washington, D.C. : Foreign Service Institute Press. 254 p.
PALLEGOIX, Jean Baptiste, Mgr (1854) Dictionarium linguae thai sive Siamensis interpretatione Latina, Gallica et Anglica. Bangkok : Parisiis Jussu Imperatoris Impressum in Typographeo Imperatorio. 897 p.  
RANGKUPAN, Suda (2001) Characteristics of Psychological Perspective in Thai Narrative Discourse. Ph.D. Dissertation : University of Buffalo, The State University. 232 p.
SRIOUTAI, Jiranthara (2004) The Thai ca2a : A marker of tense or modality, Communication à Chronos VI, International Symposium on Syntax, Semantics and Pragmatics of tense, Mood and Aspect, Genève, September 22-24. 2 p.
 
 
Nick J. Enfield
Max Planck Institute for Psycholinguistics, (Netherlands)
Person Reference in Kri, a Vietic language of upland Laos
In this paper, I draw upon a corpus of video-recorded conversations in Kri, a previously undocumented language of the Vietic sub-branch of Austroasiatic, spoken in the mountains of upland Laos, near the Vietnamese border in Khammouane Province. This paper investigates the means with which Kri speakers refer to persons in first mentions during natural conversational discourse. While much work has been done on person reference in research on topic continuity and reference-tracking, this has concentrated on pronominals and other minimal forms of non-initial reference. There has been less investigation of the range of choices that a speaker has in making a first mention of a person, the usual account being simply that the speaker should use ‘a full noun phrase’ (e.g., Barbara Fox, Discourse Structure and Anaphora, Cambridge University Press). Recent comparative research has begun to look at this problem (Person Reference in Interaction edited by N. J. Enfield and T. Stivers, Cambridge University Press, 2007). The arguments presented in that work suggest that there is a universal set of preferences for constructing initial references to persons, including preferences for minimality, recognizability, and a preference for association. It was also suggested that these preferences may be ordered differently depending on cultural differences. In Person Reference in Interaction, Enfield wrote a chapter on Lao, a major semi-standardized language of Southeast Asia, with a literary tradition. The work to be presented here will be complementary to that study, in that it provides data from another end of the sociolinguistic spectrum, i.e., from an endangered and isolated, unwritten language spoken only by a few hundred people. The Kri materials reveal a similar system of person reference to that found in Lao, one in which social-hierarchical structure is marked through the use of titles together with names, with a special emphasis on kinship as an organizing principle within a hierarchically-grounded society. The Kri system, however, is much more elaborated than was found for Lao, because (1) the Kri kinship system is significantly more complex, (b) the Kri personal pronoun system is more complex (e.g., including duals and inclusive/exclusive distinctions), and (c) the system of personal naming in Kri is more complex. This study is not only a contribution to comparative work on person reference in grammar, it is also a unique empirical contribution to research on mainland Southeast Asian languages, grounded in data that is exclusively drawn from naturally-occurring conversation rather than elicitation or monologic narrative.
 
 
 
David Gil
Max Planck Institute for Evolutionary Anthropology (Leipzig)
The Illusion of Hidden Complexity in Southeast Asian Languages
 
Anthony Burgess wrote that "[w]hat strikes the learner of Malay is the complete lack of those typically Indo-European properties — gender, inflection, conjugation. It is like diving into a bath of pure logic. Everything is pared to a minimum. [...] If one digs deeply enough into Malay, one comes to the conclusion that the Western concept of 'parts of speech' is alien to it." More generally, the languages of Southeast Asia are prime exemplars of the isolating linguistic type, with little or no morphosyntactic encoding of numerous categories which play a central role in the grammars of more familiar languages.
In analyzing such languages, many linguists argue that the overt simplicity of isolating languages is merely apparent, and that beneath such simplicity lurks hidden complexity of various kinds. In accordance with one approach, the impoverished morphological structures of such languages are compensated for by more complex syntactic structures; this approach is characteristic not only of most generative grammarians but also of many functionalists and typologists, who argue that the absence of morphological agreement and case marking correlates with more rigid word order. However, in other papers, I have provided extensive empirical evidence against the existence of such a compensatory mechanism linking morphological and syntactic structures.
A rather different approach to the isolating languages of Southeast Asia accepts that such languages may be simpler in both morphology and syntax, but assumes that formal simplicity is compensated for by more complex rules of pragmatics. The idea is an intuitive one, namely, that if the grammar of a language does not encode a particular category, then the relevant information must be inferred from the context of the utterance. For example, if a language lacks a grammatical category of number, then speakers must work harder, exercising their pragmatic competence, in order to figure out whether words are singular or plural. With pragmatics supposedly doing the job of an impoverished morphosyntax, such languages accordingly end up with a more complex pragmatic component. However, although intuitively appealing, this paper argues that such an approach is misguided on both empirical and conceptual grounds.
Taking Malay/Indonesian as a typical example of an isolating Southeast Asian language, an extensive corpus of naturalistic speech reveals that in real live language use, speakers do not necessarily fill in absent grammatical encodings by means of pragmatic inferences. The corpus is examined with respect to six selected semantic categories lacking obligatory encoding in the grammar of Malay/Indonesian: number, gender, tense, aspect, thematic role and ontological category. For each of these six categories, examples are adduced of utterances in which (a) the category in question is not encoded in the morphosyntax of the utterance; and (b) the context of the utterance is such that the hearer had no reason to fill in any supposedly missing information by means of pragmatic inferences. It is therefore concluded that the overt morphological and syntactic simplicity of Malay/Indonesian is not compensated for by hidden complexity in the pragmatic domain.
This result should not come as a surprise. All languages make use of pragmatic inferences to fill in missing information up to a contextually appropriate level of specificity. However, the suggestion that languages with simpler morphosyntactic structures make greater use of such inferences is based on the fallacy of Eurocentrism, and the assumption that the contextually appropriate level of specificity just happens to be that which is encoded in the grammars of Standard Average European. (By the same token, a speaker of Larike, with grammatical dual and trial number, would be entitled to assume that all English speakers, when hearing a noun in the plural, have to figure out whether a dual, trial, or a four-or-more plural was intended — a clear reductio ad absurdum.) Thus, there is no reason to expect to find any systematic relationship between overt morphosyntactic simplicity and greater recourse to pragmatic inferences. Here too then, the notion of hidden complexity in Southeast Asian languages is nothing but an illusion, whose origin lies in the grammatical categories of the linguists' own native languages, and the assumption that such categories correspond to an ideal "language of thought" that speakers of all the worlds' languages must somehow strive to express.
 
 



Philippe Grangé
Université de la Rochelle
Le génitif dans quelques langues des Petites Iles de la Sonde
L’est de l’Indonésie se caractérise par une très grande diversité linguistique, liée à la fois à la géographie de cette vaste région archipélagique et à des migrations préhistoriques ou historiques complexes. L’exploration linguistique est récente et encore lacunaire pour certaines langues, sans parler du foisonnement linguistique de la Papouasie. On situe traditionnellement au sud-est de l’Indonésie (Petites Iles de la Sonde Orientales, Moluques) et Timor Lorosa’e le groupe des langues Centrales-Malayo-Polynésiennes (CMP) mais sa délimitation en tant qu’aire linguistique et sa définition typologique ne font pas consensus.
Dès la fin du XIXème s., l’ordre syntaxique “possesseur-possédé” (1) y est noté par des linguistes, et deviendra le principal critère définissant les langues CMP, elles-mêmes sous-ensemble des langues Centrales-Est-Malayo-Polynésiennes (CEMP).
(1) 
guru
sa’o
 
 
 
 
 
 
 
 
 
 
enseignant
maison
 
 
 
 
 
 
 
lio (Austronesian, Central-Malayo-Polynesian)
 
maison du professeur
 
 
(2) 
rumah
guru
 
 
 
 
 
 
 
 
 
 
maison
enseignant
 
 
 
 
 
 
 
indonésien (Austronesian, Western-Malayo-Polynesian)
 
maison du professeur
 
Certains critères typologiques et la notion même d’aire linguistique CMP ont été remis en cause, voir Himmelmann (2005 : 112) et Klamer et al. (2007). Klamer et Ewing (à paraître) précisent avec finesse la typologie des langues austronésiennes de l’aire linguistique qu’ils nomment East Nusantara (Est de l’Insulinde). Ils mentionnent, entre autres caractéristiques, la syntaxe « possessor-possessed », une « morphological distinction between alienable/inalienable nouns » et la possibilité de « metathesis » phonologiques.
Nous n’entrerons pas dans ces débats typologiques, mais compte tenu de la pertinence réaffirmée de l’ordre possesseur-possédé, nous souhaitons examiner l’expression du génitif dans quelques langues austronésiennes des Petites Iles de la Sonde Orientales (CMP Bima-Sumba). Nous prendrons des exemples surtout en lamaholot (Est de Florès, Adonara, Solor et Lembata). Il apparaît que l’ordre possesseur-possédé, s’il est attesté dans la quasi-totalité de ces langues, concerne les noms, mais pas les clitiques référant au possesseur : « if a language has a possessor morpheme, it is generally a suffix/enclitic, not a prefix/proclitic » (Klamer 2002 : 372). C’est le cas en lamaholot pour la particule enclitique d’accord (3) (4), comme pour le pronom libre (5).
(3) 
guru
langu
=n
 
 
 
 
 
 
 
 
 
enseignant
maison
3sg.gén
 
 
 
 
 
 
lamaholot, dialecte d’Adonara Est
 
maison du professeur
 
 
(4) 
go
Langu
=k
 
 
 
 
 
 
 
 
 
1sg
maison
1sg.gén
 
 
 
 
 
 
 
 
ma maison
 
 
(5) 
lango
go’
=en
 
 
 
 
 
 
 
 
 
maison
1sg
gén
 
 
 
 
 
 
 
 
ma maison
 
La morphologie du géntitif est assez diverse parmi les langues des Petites Iles de la Sonde Orientales. Même les différents dialectes du lamaholot contrastent dans la morpho-phonologie accompagnant le génitif, qui entraîne éventuellement des alternances vocaliques (voir (3) et (4) ci-dessus : /o/ > /u/ pour lango “maison”), des épenthèses et des métathèses.
La prédication d’existence implique dans certaines langues l’emploi du génitif. En lamaholot, cette contrainte ne s’exerce qu’à la forme négative. Ainsi, on ne dira pas “je n’ai pas d’arbre” ou “mon frère ne travaille pas” mais “mon arbre il n’y a pas” ou “mon frère n’a pas de travail.”
L’opposition morphologique entre la possession aliénable et inaliénable occupe également un périmètre divers selon les langues ; les parties du corps et les membres d’une famille sont en principe inaliénables, mais il y a des exceptions intrigantes.
Klamer et al. (2007) soulignent les influences réciproques des langues austronésiennes et non-austronésiennes (i.e. langues du Phylum Trans-Nouvelle-Guinée, dites aussi langues papoues). Ces langues papoues ont vraisemblablement légué aux langues austronésiennes de l’Est de l’Insulinde trois caractéristiques principales : l’ordre possesseur-possédé (pour les noms), l’opposition aliénable-inaliénable et la négation en fin de phrase. Le lamaholot semble être une des rares langues des Petites Iles de la Sonde, voir Klamer (2002 : 377), peut-être la seule, à avoir adopté ces trois caractéristiques à la fois.
Des influences certainement plus récentes se sont exercées sur les divers dialectes malais de l’est de l’Indonésie, voir Paauw (2008 :176). Ainsi, en malais de Larantuka (à l’extrême est de Florès, au cœur de l’aire linguistique du lamaholot) on constate l’adoption de l’ordre possesseur-possédé, au moyen de puN “posséder”, fonctionnant désormais comme préposition (7).
(6)                                                                                             
a. rumah
saya
      
b. rumah
=ku
       
c. Saya
punya
rumah.
 
 
 
     maison
1sg
 
     maison
1sg.gén
 
     1sg
posséder
maison
malais/indonésien
 
    ma maison                     ma maison                         J’ai une maison.
 
 
(7)                                                                                             
saya
puN
rumah
 
 
1sg
prép
maison
malais de Larantuka
 
ma maison
 
 
 
Bibliographie
HIMMELMANN, Nikolaus P. (2005). – « The Austronesian languages of Asia and Madagascar: typological characteristics ». in The Austronesian languages of Asia and Madagascar, sous la direction de Alexander ADELAAR et Nikolaus P. HIMMELMANN, pp. 110-181. New York: Routledge.
KLAMER, Marian (2002). – « Typical features of Austronesian Languages in Central/Eastern Indonesia ». in Oceanic Linguistics 2, volume 41 pp. 364-383
KLAMER, Marian ; REESINK, Ger et STADEN, Miriam van (2007). – « East Nusantara as a Linguistic Area ». in From linguistic areas to areal linguistics, sous la direction de Pieter MUYSKEN, pp. 53. Amsterdam: Benjamins.
KLAMER, Marian et EWING, Michael C. eds. (à paraître). – The languages of East Nusantara: an introduction: to appear
PAAUW, Scott Horan (2008). – The Malay contact varieties of Eastern Indonesia : a Typological Comparison (thèse), PhD dissertation, State University of New York at Buffalo
 
 
Yury A. Lander
Institute of Oriental studies RAS (Moscow)
 
Remarks on relativization in western Austronesian
The aim of this paper is to account for some features and variation observed in western Austronesian (WAN) relative constructions. WAN relatives show some specifics as compared to many other languages:
(i) For many WAN languages (mainly Philippine and Formosan ones, but also Malagasy and Murut languages of Sabah, etc.) a constraint has been proposed that restricts relativization to subjects. These languages were actually the most reliable examples that supported the subject’s primacy within the NP-accessibility hierarchy (NPAH; Keenan & Comrie 1977): Subject (S) > Direct Object (DO) > Indirect Object (IO) > Oblique O (OO) > Possessor (Poss) > Object of Comparison (OComp), requiring that S should always be relativizable, while O should not.
(ii) Seemingly, it were mainly WAN languages that made Keenan and Comrie give up the idea that if a language relativizes some position in NPAH, then it should be able to relativize all positions that are higher: such WAN languages as Tagalog, Balinese, Sundanese, Gayo, Buginese, etc. allow for relativization of S and Poss only. Importantly, in many of these languages Poss-relativization is restricted to possessors of subject.
(iii) A related fact is that some WAN languages such as Nias and Muna exploit very similar relativization strategies for S and Poss, but a rather different strategy for relativization of various kinds of objects.
I argue that albeit these features apply to different languages, they are related and some of them have more historical than structural nature:
1. My starting observation is that languages that do not show the features (i) and (ii) (e.g., Colloquial Indonesian, Aceh, Toba-Batak, Tukang Besi, etc.), including those displaying the feature (iii), do not belong to the so-called Philippine grammatical type, characterized in particular by rich “voice systems” and weak syntactic differentiation of nouns and verbs. Given the fact that the Philippine type can almost certainly attributed to Proto-Austronesian, it can be assumed that non-Philippine-type languages showing either (i) or (ii) inherited it from their ancestors displaying Philippine-type features.
2. The feature (iii) can, then, be explained by the fact that O-relativization in non-Philippine-type languages is secondary. Notably, languages showing (iii) are close to violate Keenan & Comrie’s constraint that any relativization strategy must apply to a continuous segment of NPAH, so any historical explanation of this fact, which does not presuppose strict grammatical consistency, should indeed be favored here.
3. A few languages (especially in the South Sulawesi area) have forms that are likely to be classified as participles used for S- and O-relativization. Their morphology (at least partly) goes back to the Philippine-type “voice”-morphology, which suggests that the corresponding voice forms were reanalyzed as participles exactly when objects became available to relativization.
Why was the transition from the Philippine type to other types accompanied by the expansion of relativization potential? The answer, I argue, can be found in Starosta, Pawley and Reid’s (1982; n.d.) idea that Philippine-type languages at least originally mostly used equative sentences with predicates expressed by nominalizations equivalent to (or serving as) relative clauses. The restriction of relativization to S’s and their possessors can therefore be seen as a manifestation of Ross’s (1967) constraints prohibiting relativization out of relatives. Typologically, this is supported by the fact that the clause structure of some other languages showing similar constraints (e.g., West Greenlandic) have also been argued to develop from equative constructions.
Languages deviating from the Philippine type apparently reanalyzed the equative clause scheme as a predicative one. This required integration of objects into the clause structure, which partly was reflected by the appearance of applicative morphology but also by the breakdown of constraints on O-relativization.
To conclude, I argue that while some well-known restrictions on relativization in the Philippine-type languages were motivated structurally, in languages deviating from this type these restrictions occur only as a kind of inheritance, which led to variation observed in Indonesian languages. From the typological point of view, this casts doubts on S’s primacy in the NPAH.
 
Selected bibliography
Anceaux, J. C. 1952. The Wolio Language. ‘s-Gravenhage.
Arka, I W. 1998. From morphosyntax to pragmatics in Balinese: A lexical-functional approach. Ph.D. diss., University of Sydney.
Brown, L. 2001. A Grammar of Nias Selatan. Ph.D. diss., University of Sydney.
Chung, S. 2008. Indonesian clause structure from an Austronesian perspective. Lingua 118(10): 1554-1582.
Ceña, R. M. 1979. Tagalog counterexamples to the Accessibility Hierarchy. Studies in Philippine Linguistics 3: 119–124.
Cole, P. and G. Hermon. 2005. Subject and non-subject relativization in Indonesian. Journal of East Asian Linguistics 14:59-88.
Donohue, M. 1996. Relative clauses in Tukang Besi: grammatical functions and thematic roles. Linguistic Analysis 26: 159-173.
Donohue, M. 1999. A Grammar of Tukang Besi. Berlin: Mouton de Gruyter.
Durie, Mark. 1985. A Grammar of Acehnese on the Basis of a Dialect of North Aceh. Dordrecht: Foris.
Eades, D. 2005. A Grammar of Gayo: a Language of Aceh, Sumatra. Canberra: Pacific Linguistics.
Eades, Y. 1998. Relativization. Working Papers in Sasak 1: 119-129.
Elkins, Richard E. 1970. Major Grammatical Patterns of Western Bukidnon Manobo. Norman: Summer Institute of Linguistics of the University of Oklahoma.
Fox, B. A. 1987. The Noun Phrase Accessibility Hierarchy reinterpreted: Subject primacy or the Absolutive Hypothesis? Language 63.856–870.
Goudswaard, Nelleke Elisabeth. 2005. The Begak (Ida’an) Language of Sabah. Utrecht: LOT Publ.
Hardjadibrata, R. R. 1985. Sundanese: A Syntactic Analysis. Canberra: Pacific Linguistics.
Healey, P.M. 1960. An Agta Grammar. Manila: Bureau of Printing.
Kähler, H. 1974. Relative clause formation in some Austronesian languages. Oceanic Linguistics 13: 257-277.
Keenan, E. L. 1974. Relative clause formation in Malagasy (and some related and not so related languages). In P. M. Peranteau et al., The Chicago Which Hunt. CLS.
Keenan, E. L. and B. Comrie. 1977. Noun phrase accessibility and universal grammar. Linguistic Inquiry 8: 63–99.
Liao, H.-C. 2000. The Noun Phrase Accessibility Hierarchy revisited: A view from ergative languages. University of Hawai’i Working Papers in Linguistics 31: 121-142.
Liu, A. K. 2005. The structure of relative clauses in Jianshi Squliq Atayal. Concentric: Studies in Linguistics 31: 89-110.
Musgrave, S. 2001. Non-subject arguments in Indonesian. Ph.D. diss., The University of Melbourne.
Prentice, D. J. 1971. The Murut Languages of Sabah. Canberra: Pacific Linguistics.
Reid, L. A. 1966. An Ivatan Syntax. Honolulu: University of Hawaii.
Ross J. R. 1967. Constraints on Variables in Syntax. Ph.D. dissertation, MIT.
Ross, M. 2002. The history and transitivity of western Austronesian voice systems. In F. Wouk and M. Ross (eds), The History and Typology of Western Austronesian Voice Systems. Canberra: Pacific Linguistics.
Starosta, S., A. K. Pawley and L. A. Reid. 1982. The evolution of focus in Austronesian. In A. Halim et al. (eds), Papers from the 3rd ICAL. Vol. 1. Canberra: Pacific Linguistics, 145-170.
Starosta, S., A. K. Pawley and L. A. Reid. N.d. The Evolution of Focus in Austronesian. Unpunlished monograph.
Svelmoe, G. and T. Svelmoe. 1974. Notes on Mansaka Grammar. Huntington Beach: SIL.
Thurgood, E. A. 1998. A description of nineteenth century Baba Malay: A Malay variety influenced by language shift. Ph.D. diss., University of Hawaii.
van den Berg, R. 1989. A Grammar of the Muna Language. Dordrecht: Foris.
Zeitoun, E. (ed.) Nominalization in Formosan Languages. Language and Linguistics 3.
 
 
 
François Langella
Chulalongkorn University, Thailand
On the relation between causation and purpose: a case study of hâj in Thai
Aside from well-known cases of syncretism with allative (Rice & Kabata 1997, Schmidtke to appear), purpose also frequently overlaps with cause, in languages as genetically diverse as French (pour, pourquoi), English (for), Konda (Dravidian language, Croft 1991) or Qiang (Tibeto-Burman, LaPolla 2003), to name a few.
Thai seems to follow the latter tendency. For one, the non-verbal purposive marker phɯ̂a, also used to mark remote benefaction (Jenny, m.s.), derives its modern meaning from its original use as a causal marker (Langella, 2009). In this paper, I propose to investigate the marking of resultative constructions by hâj ‘to give’, as reported in the literature (Rangkupan 2005, Thepkanjana and Uehara 2008). I will argue that this putatively purposive use of hâj actually relates to its otherwise permissive meaning, permission being understood as a type of “enabling causation” as defined by Talmy (1976).
(1) and (2) illustrate the type of construction at stake here. Whereas (1) is a resultative construction asserting the factuality of the resulting event, the insertion of hâj in post-verbal position in (2) supplies intentionality to the agent and entails the non-factuality of his intended result, thus possibly leading to a purposive reading.
(1)   lék   tɔ̀j        jà:j     lóm
Lek punch Yai    fall.down
‘Lek fell Yai down with a punch’
 
(2)   lék   tɔ̀j        jà:j   hâj   lóm
Lek punch Yai   give fall.down
a. ‘Lek punched Yai in such a way that (/so that) Yai would fall to the ground’
b. *‘Lek punched Yai in such a way that (/so that) Yai fell to the ground’
 
However, a purposive interpretation fails to make sense when hâj is negated. Assuming that (3) is uttered by Lek’s coach in a rigged boxing game, (3) is best interpreted as a directive, meaning that Lek must not allow the result of his own punching (Yai falling to the ground) to happen. Indeed, both permission and purpose are characterized by the non-factuality of the event following the act of allowing and the intentional action, respectively.
 
(3)   lék    tɔ̀j       jà:j    mâj hâj lóm
Lek punch Yai    neg give fall.down
a. ? ‘Lek, punch Yai so that he does not (/lest he) fall(/s) down’
b. ‘Lek, keep on punching Yai but don’t let him fall down (i.e. don’t knock him down)’
 
Eventually, the ungrammaticality of (4) demonstrates that the possibly purposive reading of hâj remains ultimately grounded in the semantics of “interpersonal manipulation” (Givón 1975). Indeed, hâj cannot be used to express a purposive relation between two events which are carried out by the same agent.
 
(4)   kháw nâng rót-faj (*hâj)   / (phɯ̂a thî:      càʔ      ) paj chiang-mài
3sg    sit    train    (*give) / (purp compl irrealis) go Chiang Mai
‘He took the train (in order) to go to Chiang Mai’
 
References
Croft, W. 1991. Syntactic categories and grammatical relations: The cognitive organization of information. Chicago: University of Chicago Press
Givón, T. 1975. “Cause and control: On the semantics of interpersonal manipulation”. In J.P. Kimball (ed.), pp 59-90
Jenny, M. m.s. “Benefactive Constructions in Thai”. To appear in F. Zúñiga & S. Kittlä (eds.).
Kimball, J.P. (ed.). 1975. Syntax & Semantics - Volume 4. New York: Academic Press Inc.
Lakoff, G. & M. Johnson. 1999. Philosophy in the flesh: The embodied mind and its challenge to western thought. New York: Basic Books
Langella, F. 2009. “From cause to purpose and benefactive – A case study of phɯ̂a ‘for’ in Thai”. Poster presented at the 19e meeting of the Southeast Asian Linguistics Society, May 28-29 2009, Ho Chi Minh City, Viet-Nam
LaPolla, R.J. 2003. A grammar of Qiang. Berlin: Mouton de Gruyter
Rangkupan, S. 2005. “The Syntax and Semantics of GIVE-Complex Constructions in Thai”. Paper presented at the 2005 International Course and Conference on Role and Reference Grammar, Institute of Linguistics, Academia Sinica, Taipei, Taiwan. 
Rice, S. & K. Kabata. 2007. “Crosslinguistic grammaticalization patterns of the allative”. Linguistic Typology 11:451-514
Schmidtke, K. “The role of benefactives and related notions in the typology of purpose clauses”. To appear in F. Zúñiga & S. Kittlä (eds.).
Shibatani, M (ed.). 1976. Syntax & Semantics – Volume 6. New York: Academic Press Inc.
Talmy, L. 1976. “Semantic causation types”. In Shibatani, M. (ed.), pp 43-116.
Thepkanjana, K. & S. Uehara. 2008. “The verb of giving in Thai and Mandarin Chinese as a case of polysemy: A comparative study”. in Languages Sciences 30 pp 621-651
Zúñiga F. & S. Kittlä. To appear. Benefactives and Malefactives. Case Studies and Typological Perspectives. Amsterdam/Philadelphia: John Benjamins
 
 
 
 
Hsiu-chuan Liao
National Tsing Hua University, Taiwan
 
The Development of Case-marking Systems in Ilongot and Other Southern Cordilleran Languages
Southern Cordilleran languages, consisting of Ibaloy (or Inibaloi), Kalanguya (or Kallahan), Karao, I-wak, Pangasinan, and Ilongot (or Bugkalot), are a group of Austronesian languages spoken in the southern parts of the Cordillera Central and the surrounding lowlands of Northern Luzon, the Philippines. Southern Cordilleran languages consist of two major subgroups: (a) Ilongot; (b) West Southern Cordilleran languages: Ibaloy, Kalanguya, Karao, I-wak, and Pangasinan. The West Southern Cordilleran group can be further divided into two groups: (a) Pangasinan; (b) Nuclear Southern Cordilleran languages: Ibaloy, Kalanguya, Karao, and I-wak (Reid 1991, 2006; Himes 1998). 
This paper deals with the development of case-marking systems in Ilongot and other Southern Cordilleran languages. On the basis of data from my own fieldnotes and from other available linguistic materials (including Benton 1971; Brainard 1997; Brainard 2003; Ruffolo 2004; Hohulin and Hale 1977; Himes 1998; etc.), the following observations are made:
First, like other Southern Cordilleran languages, Ilongot employs different phrase markers to introduce personal nouns and non-personal nouns. 
Second, as in other Southern Cordilleran languages, Nominative NPs in Ilongot are introduced by phrase markers containing reflexes of *si (for singular personal nouns) or reflexes of *Ɂi (for singular non-personal nouns) (Reid 1979). However, unlike Nuclear Southern Cordilleran languages, the formative Ɂi- occurs optionally in Ilongot (as well as in Pangasinan) Nominative phrase markers (for singular non-personal nouns). More specifically, (singular, non-personal) Nominative phrase markers can either be Ɂito / Ɂita / Ɂima or to / ta / ma in Ilongot. The optionality of the formative Ɂi- in Ilongot (as well as in Pangasinan) can be considered as a result of “unmarking the (grammatical) subject” (see Reid 1978 for details).
Third, unlike phrase markers in West Southern Cordilleran languages, phrase markers for non-personal nouns in Ilongot are deictic in nature. That is, phrase markers in Ilongot contain not only a case-marking formative (Ɂi-(singular, Nominative), ni-(singular, Genitive/Dative), di- (plural), etc.), but also a deictic component (to ‘proximal’ / ta ‘medial’ / ma ‘distal’) to specify the relative distance between the entities referred to and Speech Act Participants (SPAs). The deictic component ma appears to be an innovation shared by all Southern Cordilleran languages (cf. *di ‘distal’ in Ilokano and Proto-Central Cordilleran). The deictic component to ‘proximal’ appears to be a retention in Ilongot, but not in West Southern Cordilleran languages (cf. *ya ‘proximal’ in Proto-West Southern Cordilleran (Himes 1998:143)). 
The development of case-marking systems in Ilongot and other Southern Cordilleran languages appears to support Himes’s subgrouping of Southern Cordilleran languages.
[Total: 414 Words]
Selected References
Benton, Richard A. 1971. Pangasinan reference grammar. PALI Language Texts: Philippines. Honolulu: University of Hawai‘i Press.
Brainard, Sherri. 1997. Ergativity and grammatical relations in Karao. In Grammatical relations: A functional perspective, ed. by Talmy Givón, 85–154. Amsterdam and Philadelphia: John Benjamins Publishing Company.
Brainard, Sherri, comp. 2003. Karao texts. Studies in Philippine Languages and Cultures 13. Manila: Linguistic Society of the Philippines and Summer Institute of Linguistics.
Himes, Ronald S.  1998.   The Southern Cordilleran group of Philippine languages. Oceanic Linguistics 37(1):120-177.
Hohulin, E. Lou, and Austin Hale. 1977. Notes on Keley-i relational grammar—I. Studies in Philippine Linguistics 1(2):231-63.
Reid, Lawrence A. 1978. Problems in the reconstruction of Proto-Philippine construction markers. In Second International Conference on Austronesian Linguistics: Proceedings, ed. by S. A. Wurm and Lois Carrington, 33-66. Canberra: Department of Linguistics, Research School of Pacific Studies, The Australian National University.
Reid, Lawrence A. 1979. Evidence for Proto-Philippine nominative marking. Philippine Journal of Linguistics 10:1-20.
Reid, Lawrence A. 1991. The Alta languages of the Philippines. In VICAL 2: Western Austronesian and contact languages: Papers from the Fifth International Conference on Austronesian Linguistics, ed. by Ray Harlow, 265–297. Auckland: Linguistics Society of New Zealand.
Reid, Lawrence A. 2006. On reconstructing the morphosyntax of Proto-Northern Luzon.  Philippine Journal of Linguistics 37(1):1-64.
Ruffolo, Roberta. 2004. Topics in the morpho-syntax of Ibaloy, Northern Philippines. Ph.D. dissertation, Australian National University.
 
 
Dara Non
LLF UMR 7110
Université Paris Diderot
 
A propos du groupe nominal complexe en khmer contemporain
La combinaison de deux groupes nominaux GN1 et GN2 met en jeu tout un éventail de « relateurs » (nous employons désormais ce terme à la place de « préposition »), parmi lesquels nous pouvons citer : sɑmrap (pour),kʰnɔŋ (dans),nɘɨ (rester),nɘɨkʰnɔŋ (dans), kʰnɔŋcɑmnaɔm (parmi), ciɜ (être), nɘj (de) et rɔbɑh (de)… Cette mise en relation peut également se faire sans l’intervention d’un relateur (ø). La structure d’un syntagme nominal peut être illustrée par l’exemple suivant :
(a)    tɑmlaj           ø/sɑmrap/ciɜ/nɘj/rɔbɑh       mənʊh        kɨː     phiɜpsəmaɔhtrɑŋ
Valeur             ø/sɑmrap/ciɜ/nɘj/rɔbɑh        être humain     être      fidélité
            La valeur de l’homme est la fidélité.
La traduction généralisante donnée ci-dessus permet sans doute d’homogénéiser les valeurs des différents relateurs qui peuvent apparaître dans l’exemple (a) mais elle ne donne pas accès aux affinités syntaxiqueset sémantiques qu’apporte chaque relateur. La prise en compte de ces affinités n’est possible que dans la mesure où la valeur de chaque relateur est mise en évidence.
Dans cette communication, nous proposons l’analyse d’une de ces configurations syntaxiques à savoir les groupes nominaux construits sans l’intervention d’aucun relateur (ø). Nous posons l’hypothèse qu’il s’agit d’une relation sous-spécifiée, inférable sur la base des propriétés lexicales respectives de GN1 et de GN2.
Dans un premier temps, nous essayerons de repérer à partir des données dans lesquelles il n’y a pas de concurrence entre ø et la présence d’un relateur, toutes les relations qu’on peut avoir dans la configuration GN1 ø GN2. Parmi ces relations, nous pouvons trouver l’identification, possession, appartenance, détermination, localisation… :
(b) koat ruɜh nɘɨ    tiːkrɔŋ        ø      pʰnɔmpeɲ
            il/elle   vivre    rester   ville                 ø         Phnom Penh
            Il/elle habite à la ville de Phnom Penh.
(c) daj     ø      koat   thɔm ciɜŋ   cɜːŋ   ø      kʰɲɔm         tɘɨ     tiɜt
            bras     ø         il/elle   grand   comp   jambe ø         je                     part    part
            Son bras est même plus grand que ma jambe.
(d) koat ciɜ     nɛak   ø      pʰleːɲ
Il/elle   être      humain            ø         musique
Il/elle   est musicien/musicienne.
Les interprétations des groupes nominaux figurés dans les exemples ci-dessus, que ce soit identification, possession inaliénable ou détermination, résultent non seulement des propriétés sémantiques respectives de chaque composante mais aussi de la position syntaxique qu’elle occupe dans le groupe : dans (b) et (c) l’inversion de l’ordre combinatoire n’est pas possible tandis que dans (d) on aura des interprétations complètement différentes. Dans tous les cas, GN2 doit fonctionner comme déterminant qui est source de détermination pour NG1.
Lorsqu’on analyse les cas où la construction avec ø est la seule possible, on est confronté au problème relevant de la distinction entre un nom composé et un groupe nominal : les noms composés en khmer sont formés des juxtapositions des mots de natures variées sans aucun connecteur et que la plupart d’entre eux sont des groupes nominaux figés (Antelme, 2004). Cela nous amène à remettre en cause le statut de la séquence composée des mots nɛak et pʰleːɲ dans (d), qui est généralement considéré comme un nom composé de type N générique-N spécifique : on devrait noter par conséquence /nɛakpʰleːɲ/ et non pas /nɛakøpʰleːɲ/.
Si nous considérons cette séquence dans (e), son interprétation ‘chanteur professionnel’ n’est qu’une des interprétations possibles :
(e) koat   nɛak ø      pʰleːɲ                   məsɘlmeɲ   nɘŋ
Il/elle   humain            ø         musique          hier                  déict
Il/elle   est le/la musicien/musicienne (que nous avons vu) hier.
Il/elle est celui/celle qui a joué de la musique hier.
Par ailleurs, la variation interprétative de la séquence nɛakøpʰleːɲ ne se limite pas à ces deux valeurs, selon les contextes on peut avoir :
(f)    La distribution des tâches à faire pour chaque membre d’un groupe d’organisateurs de fête :
baːŋ ʔaɛŋ   nɛak ø      pʰleːɲ                   hɛh
aîné     pro       humain            ø         chanson           part     
Dis donc, (toi, dans la mesure où tu t’occupes des chansons) t’es “monsieur chanson”?
            Nous essayerons à partir de l’anyse des données, de définir la relation mise en place en cas de ø et d’identifier les paramètres à l’œuvre dans la variation interprétative d’un groupe nominal.
BIBLIOGRAPHIE
ANTELME M. 2004, "Khmer", in : Le Nom composé. Données sur seize langues, sous la direction de Pierre J. L. Arnaud (Travaux du C.R.T.T.), Lyon : Presses Universitaires de Lyon : 149-183.
AURNAGUE, M., 2002, "Relation de partie à tout, configurations typiques et dépendances : analyse sémantique de quelques constructions génitives du basque", Revue de Sémantique et Pragmatique n°11, p. 69-85.
BENVENISTE, E. 1966, "Problèmes de linguistique générale", Tome 1, Editions GALLIMARD.
———— , 1975, "Noms d’argent et noms d’action en Indo-Européen", LIBRAIRIE D’AMERIQUE ET D’ORIENT.
CHUON N., 1967, "Dictionnaire Cambodgien", Edition de l’Institut Bouddhique.
DE VOGUE S., 1993. « Des temps et des modes », in Le gré des Langues n°6, pp 65-91
DE VOGUE S., 1997. « Identité lexicale et hétérogénéité de la variation co-textuelle : le cas de suivre.», En collaboration avec PAILLARD D., inC. Guimier (ed .), Co-texte et calcul du sens, Actes de la Table-ronde organisée par l'ELSAP, février 96, Presses Universitaires de Caen.
FRANCKEL J.J., PAILLARD D. 2007, "Grammaire des prépositions", Tome 1 Editions OPHRYS.
HUFFMAN, Franklin E. 1970, "Modern Spoken Cambodian", Cornell University, Southeast Asia Program.
KHIN, S. 2000, "Manuel de khmer", Editions YOU-FENG.
—     ,1999, "La grammaire du khmer moderne", Editions YOU-FENG.
KLEIBER G., SCHNEDECKER C., THEISSEN A., 2006, "La relation partie-tout", Leuven, Belgique, Peeters.
SAK-HUMPHRY, C. 1993, "The syntax of nouns and noun phrases in dated pre-Ankorian inscriptions", in Mon-Khmer Studies no22, pp 1-126.
(ouvrage collectif), 1996, La relation d’appartenance, Faits de langues n°7, Paris OPHRYS.

 

 
Antonia Soriente
Max Planck Institute for Evolutionary Anthropology
(Jakarta Field Station)
Focus and voice in some languages of the Kayan-Kenyah group
 
This paper will focus on the morphosyntactic features of some Kayan-Kenyah languages belonging to the Western Malayo-Polynesian branch of Austronesian, mainly Kenyah, Penan Benalui and Punan Tubu’ spoken in East Kalimantan. I will present a description of the morphosyntax of Kenyah, Penan Benalui and Punan Tubu’ languages from naturalistic and elicited data and also will use some secondary source data from Kayan and Kelabit to shed light on the typological morphosyntactic features of the area for the expression of focus and voice comparing the way grammatical relations are marked and the way actor focus and undergoer focus contrast is expressed.
As pointed out by Claire (1996), the voice system in Bornean languages is much reduced in comparison to the Philippine-type languages. Indeed there is a wide range of voice systems, from very complex ones like in some languages in Sabah where ablaut, affixation, nominal marking and word order play a relevant role, to much simpler systems like Kayan and Kenyah where a very simple morphological process is employed and only personal pronouns and word order play a role. Penan Benalui and Punan Tubu’ employ the -EN- infixation to mark the undergoer voice, and fall in the middle of this range of voice systems whereas Kayan and Kenyah seem to have lost the system or probably never developed it.
In Penan Benalui and Punan Tubu’ the undergoer focus is productively marked by the -EN- infix as exemplified by 1) and 2) although few examples with bare verbs have been recorded. In most Kenyah languages there is no specific passive morphology, but thematic roles are expressed pragmatically or analytically through the word order or the use of words like kè’en ‘by’, like in the example 3) from Òma Lóngh where the verb is in the active form.
 
1)
balak
yaq
pengau
senuaq
pengah
kinan
 
banana
REL
new
-EN-buy
PFCT
-EN-eat
       “The bananas that were just bought were eaten up”                             Penan Benalui
 
2)
putiq
awoq
teniuq
unih
tubit
kinan
 
 
 
banana
REL
-EN-buy
earlier
already.happened.once
-EN-eat
 
 
‘the banana that was bought earlier has been eaten’                                                Punan Tubu’
 
3)
udeq
jé
kè’en
dévó
fadi
jé
metóngh
te
zómó
laminy
 
dog
that
by
two
relative
that
N-hit
at
front
house
        “That dog was his by the two siblings in the front of the house”        Kny Ò. Lóngh
 
As a conclusion I will try to answer questions on how these languages relate to each other and to other languages in the area in the reflex of the Proto-Austronesian infix -IN- marking simultaneously voice and aspect, how we interpret the lack of the undergoer voice markers in Kenyah and Kayan and whether it is possible to use morphosyntactic features to define subgroupings.
 
 
Hein Steinhauer
Leiden University
 
Sex and the pronouns in Indonesian
 
This paper discusses two aspects of modern Indonesian: the rise of new gender specific vocabulary and the expression of gender in relation to the personal pronominal system.
Feminist observers of Indonesian have hailed the language for its failure to differentiate between men and women in its personal pronominal system. The fact that also actor nouns formed with inherited morphology are gender neutral only endorsed the superficial impression that Indonesian in contrast to European languages like Dutch, French, or Russian is a "non-sexist" language.
Whatever such qualifications may say about the language, the culture or the observer, communication without reference to gender would be utopian: also in Indonesian most proper names and a number of kinship terms are gender specific. In fact, the absence in the language of a gender specific personal pronoun for the third person singular turns out to be a handicap in a variety of contexts. In the following scene, for instance,
            Michel men-cium Michèle.     Dia sangat men-cintai-nya
            M. ACT-kiss M.                     3SG very ACT-love-3SG
the second sentence is ambiguous since it can be translated as 'he loved her very much' or 'she loved him very much'.
To make up for this gap in the lexicon Indonesian has to use disambiguating phrases, such as gadis itu 'that girl'. The use of such strategies is demonstrated by a comparison of Tolstoy's novel Krejcerova sonata and its translation into Dutch and Indonesian.
Alongside inherited gender specific vocabulary such as gadis 'girl', modern Indonesian is developing such vocabulary by means of borrowed suffixes. The suffix pair –a vs. –i (for male vs. female) is hardly used with inherited roots, the parallel pair –wan vs. –wati, however, is productive, although it is said that forms with –wati are often absent. The paper discusses to what extent this is true and what would be the exact semantic relation between both forms.
In contrast to Indonesian, some Malayic vernaculars such as Minangkabau have gender specific pronouns for the second person singular. The Indonesian first and second person pronouns also contain information on the social relationship between the speech partners. Frequently employed strategies to avoid a possibly embarrassing choice are the use of a proper name, a kinship term or a combination of both. By such "pro-pronominal" strategies (to use a Eurocentric term) gender distinctions are often overtly expressed, where European languages would use a gender neutral pronoun.
Meanwhile the stable diglossic languagescape in which the standard Indonesian was used on official occasions, in writing, and in the media, whereas a gamut of local varieties of Malay was used in less official oral communication, has changed. Contributing factors among others are the Internet, nationwide Jakarta-based commercial television (soaps), and the popularity of the booming Jakarta-centric teenage literature. The former diglossia is now turning into an acrolect-basolect continuum with common spoken Jakarta Indonesian developing into a basolect koine. The paper discusses some effects of these developments on the Indonesian pronominal system.
 
 
Tue Trinh
Massachusetts Institute of Technology
Hypothesis. In one version of the 'copy theory of movement', a chain is a sequence CH = (α,β), where α and β are copies of the moved constituent at the derived and the base position, respectively. A PF-rule, Delete, applies to CH and eliminates phonetic material from β (Chomsky 1995: 202). In this paper, I propose the following constraint on Delete.
(1)           Constraint on Copy Deletion (CCD)                                                                                                                                                                                                                        Delete applies to (α,β) only if β ends an XP
Predictions for XP-movement. Another way to state the CCD is that Delete eliminates phonetic material from the lower copy β of a chain only if the right edge of β coincides with the right edge of an XP. It follows that if β is an XP, the relevant chain satisfies the CCD. Movement of the subject from [Spec,V] to [Spec,T] illustrates this case (2).
Predictions for V-topicalization across languages. When the lower copy is not an XP, it is deleted only if it is at the right edge of an XP. We then predict a systematic difference between VO and OV languages that permits V-topicalization, exemplified here by German and Vietnamese. Both of these languages allow V to undergo A-bar movement to [Spec,C] (cf. Fanselow 2002 for German). In Vietnamese, the fronted verb is pronounced at both the topic and the base position (3). In German, this is not the case (4). Given that VP is head-final in German and head-initial in Vietnamese, this difference follows from the CCD.
More generally, it is predicted that any SVO language that permits V-topicalization will show 'double pronunciation' of the fronted verb, while no SOV language will. This is supported by data from Hebrew (5), Spanish (6), Dutch (7) and several other languages (see e.g. Aboh and Dyakonova (2009) for Gungbe, Cable (2004) for Yiddish, Cozier (2006) for Trinidad English, Kandybowicz (2008) for Nupe, and Harbour (2008) for Haitian).
It is also predicted that if an SVO language seems to show topicalization of a bare V but fails to show double pronunciation, the topicalization in question must actually be an instance of remnant VP-movement. I argue that Norwegian and Swedish confirm this prediction.: both languages seem to allow V to occupy [Spec,C] alone only to the extent that VP-internal constituents can be scrambled or extraposed. In this connection, a case is made against Holmberg (1999) – who claims that Swedish does have V-topicalization – by developing the VP remnant movement analysis of Fox and Pesetsky (2005).
Predictions for N-topicalization in Vietnamese. The CCD explains a number of observations internal to Vietnamese. This language is a classifier language: NPs cannot combine with numerals without the mediation of a classifier. The head of NP can undergo A-bar movement to [Spec,C], stranding other elements of the nominal complex behind. When N is topicalized, double pronunciation of the fronted N is optional if N is non-relational (8) and obligatory if N is relational (9). I show that the solution to this puzzle follows straightforwardly from the CCD and the proposal on classifier languages in Chierchia (1998).
Implication for head-adjunction. Head-to-head movement (head-adjunction) constitutes a prima facie counterexample to the CCD: the chain created by head-adjunction always undergo Delete, whether the lower copy ends an XP or not. Thus, both V-to-T movement in German (10) and T-to-C movement in English (11) results in the moved element being pronounced only once, even though German VP is head-final English TP is head-initial.
This problem is solved by the assuming that head-adjunction is post-syntactic and does not leave a copy (Chomsky 1995, Grodzinsky and Finkel 1998). This assumption automatically provides an account for another fact. In Hebrew (12) and Spanish (13), the topicalized infinitival verb can be doubled by a tensed form inside TP. In has been proposed that (12) and (13) are derived by parallel syntactic movement of V to [Spec,C] and to T (14) (cf. Landau 2006, Vicente 2006). If that analysis is correct, we would expect – contrary to fact – overt copies of the verb at three positions: at [Spec,C] because that is where the higher copy is, at T because T is an affix and cannot be stranded, and at the base position because there is no XP boundary to its right. However, if V-to-T does not leave a copy, (12) and (13) are exactly what we expect.
Future work.The CCD makes reference to edges of syntactic constituents of designated type in the X-bar hierarchy. This is a distinctive feature of syntax-phonology mapping rules (cf. Selkirk 1984, Chen 1985, Hale and Selkirk 1987, Truckenbrodt 1995). For example, Selkirk (1984) proposes the following principle: The right/left edge of XPs is aligned with the right/left edge of phonological phrases. This suggests a parameterized reformulation of the CCD with reference to prosodic categories.
(15)        Prosodic Constraint on Copy Deletion (parameterized)                                                                                                                                                                                       Delete applies to (a,β) only if β is followed/preceded by a phonological phrase boundary
The PCCD makes the same empirical predictions as the CCD for the languages we have considered. However, a cursory look at Japanese suggests that the new formulation (PCCD) might be the correct option. Japanese is similar to German in that its VP is head-final. However, Japanese does not have A-bar V-topicalization of the relevant type (Yasutada Sudo, Shigeru Miyagawa p.c.). If double pronunciation is for some reason banned in both languages, the difference between Japanese and German may be derived from the fact that the parametric setting for both Selkirk's principle and the PCCD in Japanese is opposite to that in German (cf. Selkirk and Tateishi 1991). This opens a new direction for future research.
 
(2)           the man]XP will the man]XP kick the ball                                                                      Þ lower copy ends an XP: deletin
(3)           doc         thi            no          nen                         *(doc)    sach                                                       Þ lower copy does not end an XP: no deletion
read              TOP he                should  *(read)  book
'He should read books'
(4)           gelesen hat          Hans      das         Buch      (*gelesen)                             Þ lower copy ends an XP: deletion
read                             has          Hans      the          book                      (*gelesen)
'Hans has read the book'
(5)           liknot     Dan        kiva                        *(liknot)                et                            ha-sefer                                                Þ SVO: double pronunciation                     
to-buy          Dan        hoped    *(to-buy)              ACC     the-book
                "Dan hoped to buy the book'                                         (Omer Preminger p.c.)
(6)           jugar                      Juan       suele                      *(jugar) al             futbol     los domingos       Þ SVO: double pronunciation     
to-play         Juan       HAB.3SG           *(to-play)              at            football the Sundays
'Juan used to play soccer on Sunday'   (Vicente 2007: 7)
(7)           gedronken            heeft      hij           snel                        een         biertje     (*gedronken)                      Þ SOV: no double pronunciation     
drunk           has          he           quickly a              beer                        (*drunk)
'He has quickly drunk a beer'                                (Hedde Zeijlstra p.c.)
(8)           sach        thi           no           nen                         doc         quyen    (sach)     ve                           vat-ly                     Þ non-relational N: doubling optional
                book      TOP       he           should   read        CL                         (book)    about                     physics
                'He should read the book about physics'
(9)           vo           thi           toi           da                           gap         nguoi                    *(vo)                      cua         ong         Nam                      Þ relational N: doubling required
                wife       TOP       I              PERF    met         CL                        *(wife) of            Mr.         Nam
                'I have met wife Mr. Nam's wife
(10)        dass        sie           das         Buch      (*lesen) lesen                      Þ German V-to-T movement: deletion
that               they        the          book                      (*read)  read
'that they will read the book'                                                                               
(11)        will the man (*will) kick the ball                                                   Þ English T-to-C movement: deletion
(12)        liknot     hi            kanta                     et                            ha-praxim                                           
buy.INF      she          buy.PAST.3SG ACC     the-flowers
'she bought the flowers'         (Landau 2006)
(13)        conducir               Juan       condujo                                                un           camion                                                                                
drive.INF  Juan       drive.PRES.3SG               a              truck
                'Juan drives a truck'            (Vicente 2007)
(14)        V     … T+V   …              V object
 
 

 

References :
 
ABOH, E. O. & DYAKONOVA, M. (2009).Predicate doubling and parallel chains. Lingua, 119, pp. 1035-1065. CABLE, S. (2004). Predicate clefts and base-generation: Evidence from Yiddish and Brazilian Portuguese. Ms., MIT. CHEN, M. (1985). The syntax of phonology: Xiamen tone sandhi. Ms., University of California, San Diego. CHIERCHIA, G. (1998). Reference to Kinds across Language. Natural Language Semantics, 6, pp. 339-405. CHOMSKY, N. (1995). The Minimalist Program. MIT Press, Cambridge. COZIER, F. (2006). The Co-occurrence of Predicate Clefting and WH-Questions in Trinidad Dialectal English. Natural Language & Linguistic Theory, 24, pp. 655-688. FANSELOW, G. (2002). Against remnant VP-movement. In: ALEXIADOU, A., ANAGNOSTOPOULOU, E., BARBIERS, S. & GAERTNER, H.-M. (eds.). Dimensions of Movement (pp. 91-127). Amsterdam: Benjamins.FOX, D. & PESETSKY, D. (2005). Cyclic linearization of syntactic structure. Theoretical Linguistics,31, pp 1 - 45. GRODZINSKY, Y. & FINKEL, L. (1998). The neurology of empty categories: Aphasics' failure to detect ungrammaticality. Journal of Cognitive Neuroscience,10, pp. 281–292. HALE, K. & SELKIRK, L. (1987). Government and tonal phrasing in Papago. Phonology Yearbook, 4, pp. 151-183. HARBOUR, D. (2008). Klivaj predika, or predicate clefts in Haitian. Lingua, 118: 853-871. KANDYBOWICZ, J. (2008). The Grammar of Repetition. Nupe Grammar at the Syntax-Phonology Interface. Amsterdam: Benjamins. LANDAU, I. (2006). Chain resolution in Hebrew V(P)-fronting. Syntax, 9, pp. 32-66. SELKIRK, L. (1986). On derived domains in sentence phonology. Phonology Yearbook, 3, pp. 371-405. SELKIRK, L. & TATEISHI, K. (1991). Syntax and downstep in Japanese. In: CAROL, G. & ISHIHARA, R. (eds.). Interdisciplinary Approaches to Language: Essays in Honor of S.-Y. Kuroda (pp. 519-543). Dordrecht: Kluwer.TRUCKENBRODT, H. (1995). Phonological Phrases: Their Relation to Syntax, Focus, and Prominence. Dissertation, MIT. VICENTE, L. (2006). The Syntax of Heads and Phrases. A Study of Verb (Phrase) Fronting. Dissertation, Leiden University.
 
 
San San Hnin Tun
Cornell University
 
L'analyse des particules discursives en birman
 
En birman, langue nationale de la Birmanie et langue de la famille tibéto-birmane, il y a un emploi fréquent des éléments dits particules qui s’attachent à un autre mot (au groupe verbal ou nominal) ou même à un énoncé. Elles sont donc syntaxiquement dépendantes et leurs valeurs sémantiques se définissent aussi selon le contexte. Le terme "particule" est ici défini provisoirement par exclusion, et correspond à tout morphème qui n'appartient pas à une classe grammaticale traditionnelle, telles que nom, verbe, préposition, etc. En birman les particules sont traditionnellement décrites selon leurs fonctions syntaxiques, telles que marqueur de sujet ou d’objet, marqueur de fin de phrase [cf. (1-2.a)]
(1)       /Sh-ya-má        go                               mé                   ba/
professeur       ptcl (marqueur d’objet) demander     ptcl (de politesse)
Demandez au prof.
(2)       /wɛ-ɵà                         sà         dɛ /
(du) porc        manger ptcl (fin de phrase, realis)
(Je) mange du porc
 
Or il s’avère que telles descriptions semblent parfois insuffisantes. Si certaines particules sont grammaticalement obligatoires [ainsi dans les exemples (2.a – 2.b) sans la particule / dɛ / la phrase est agrammaticale], d’autres ne le sont pas [cf. / nɔ / dans (3.b) tandis que dans (3.a) / mɛ / est obligatoire]
(3.a)    /wɛ-ɵà                         sà                    ɵín       dɛ /
(du)porc         manger           devrais            ptcl (fin de phrase, realis)
(Je) devrais manger du porc
 
(3.b)    */wɛ-ɵà           sà                    ɵín- /
(du) porc        manger           devrait (sans marqueur de fin de phrase)*
 
(4.a)    /wɛ-ɵà                         sà                    mɛ /
(du) porc        manger                       ptcl (marqueur de fin de phrase, irréalis/futur)
(Je) vais manger (du) porc.
 
(4.b)    /wɛ-ɵà                         sà                    mɛ       nɔ/
(du) porc        manger          ptcl     ptcl (non-obligatoire, solicitant consentement)
(Je) vais manger (du) porc,d’accord?
 
Dans d’autres cas, les particules peuvent être remplacées par une particule différente sans affecter le contenu propositionnel, ou la grammaticalité de l’énoncé [ex. /tɛ/ et /ta/ dans (5-6)]. Cependant, le choix de l’une ou l’autre dépendra de la situation discursive comme en témoignent les traductions des exemples.
(5)       /wɛ-ɵà                         ɵeiˀ                  kyaiˀ    tɛ/
(le) porc          beaucoup        aimer  ptcl (fin de l’énoncé)
(J’) aime beaucoup le porc.
 
 
(6)       /wɛ-ɵà                         ɵeiˀ                  kyaiˀ    ta/
(le) porc          beaucoup        aimer  ptcl (fin de l’énoncé)
(J’) aime beaucoup le porc, (tu sais).
 
De précédentes recherches sur les particules birmanes à partir de corpus [cf. Hnin Tun, 2004] suggèrent que les "particules" birmanes partagent des caractéristiques avec les marqueurs discursifs (en anglais), tels qu’ils sont décrits dans la littérature sur les analyses de discours [cf. Schiffrin, 1987, Fraser 1996, Carter & McCarthy 2006, Schourup 1985, Maynard 1993, inter alias]. Ces particules ont des fonctions discursives, telles que signaler à l'allocutaire des attitudes énonciatives, signaler l'intention de céder ou refuser les tours de parole, exprimer un reproche. Ainsi dans l’exemple (7) /ka lɛh/ exprime un reproche de la part de la fille à sa mère.
 
(7)       /Me Me           ka lɛh             khɛˀ     laiˀ       ta/
Maman           ptcl ptcl           difficile            ptcl      ptcl
Oh maman, comme tu es difficile!
Comme la plupart de ces particules ont un sens qui varie selon le contexte, il est important de les
examiner non pas dans des phrases isolées, mais généralement au-delà de l'énoncé.
La présente étude tentera d’examiner l’emploi de quelques particules (telles /tɛ/ et /ta/ qui occupent la position finale de la phrase affirmative ou /ka lɛh/ après le syntagme nominal), avec l'objectif d'identifier leurs fonctions énonciatives (ou discursives) à l'aide d'un corpus de birman parlé contemporain.
 
Références
Carter, Ronald and Michael McCarthy. 2006. Cambridge Grammar of English. Cambridge: Cambridge University Press.
Fraser, Bruce. 1996. Pragmatic markers. Pragmatics 6 (2): 167-190.
Hnin Tun, San San. 2004. Not a study of English! A corpus analysis of discourse features in spoken Burmese. Teanga: The Irish Yearbook of Applied Linguistics 21: 75-92.
Maynard, Senko. 1993. Discourse Modality: subjectivity, emotion and voice in the Japanese language. Amsterdam: John Benjamins Publishing Co.
Schiffrin, Deborah. 1987. Discourse markers. Cambridge: Cambridge University Press.
Schiffrin, Deborah. 1994. Approaches to discourse. Oxford, UK & Cambridge, USA: Blackwell Publishers
Schourup, Laurence. 1985. Common discourse particles in English conversation. New York & London: Garland Publishing, Inc.

 

»