Verb structure and SE syntax: verb-conflated components and aspectual closure in Romance se constructions

MANGIALAVORI RASIA, MARÍA EUGENIA; MÚGICA, NORA

Caen

Conferencia; Chronos 12. 12th International Conference on Actionality, Tense, Aspect, Modality/Evidentiality; 2016

CRISCO research centre - Normandie Université, UCBN

Verb structure and SE syntax: verb-conflated components and aspectual closure in Romance se constructions.Introduction. The clitic SE/SI may take pride in being one of the most studied topics in Romance grammar. The reasons for its importance essentially lie the considerable range of constructions in which the clitic may appear (including reflexives, reciprocals, impersonals, passives, middles, anti-causatives) and the variety of functions it has been pointed out to serve, which add up to its variable status, both regarding its (still not so clear) classification as a lexical/functional element and its (not always) mandatory status in certain configurations, its relevance vs. its semantic triviality in others, etc. In sum, the syntax of this clitic comprises a vast range of theoretical issues it bears on such as movement, agreement, Case assignment, binding theory. Yet, a most promising topic lies, in our view, in syntax-semantics and morphology-syntax interface. In particular, a closer look at the availability of different SE constructions for specific verbs suggests relevant connections between internal VP structure, aspectual closure and the role played by SE in these respects.Goal. We will address the potential relations between alternative SE constructions and syntactically-relevant aspects of verb meaning with a view to identifying?or at least, gaining a better understanding of?finer divergences regarding the internal structure of the VP and morphosyntactically-non overt components involved in such differences. These elements would eventually be not only able to accommodate the distribution of transitive verbs in different SE constructions (i.e., the admission and exclusion of what seem to be semantically equivalent verbs); they would also provide locus for the encoding of relevant aspects of verb meaning in such a way that semantics would not be acting freely, but it would be regulated by (and present a significant homomorphism with) syntax. If correct, this would allow us to consider the potential role of SE as an aspectually-relevant element in a way rather different to the one commonly entertained (Aspectual SE; e.g. Zagona 1996, de Miguel & F. Lagunilla 2000). The work will be conducted under the by now apparent premise that both the morpho-syntactic status of SE and its syntactic function vary depending on the specific construction in which it appears?further, that these variations can shed light into semantic consequences of relations established in the (lexical)syntactic level.Problem & Data. We will primarily focus on the fact that whereas certain (transitive) verbs seem to only allow a passive/impersonal reading of SE ((1)a); other (apparently analogous) transitive verbs allow both a passive reading and non-passive one comprising a benefactive quirk that does not arise in vender ((1)b). To complicate things further, other verbs like distribuir [distribute] allow a third reading, corresponding to the well-known reciprocal SE ((1)c).(1) a. Se vendieron (todos los) coches importados.SE sell.3PL.PST all the cars imported.?All the imported cars have been sold [out]?b. Se compraron (todos los) coches importados.SE buy.3PL.PST all the cars imported.?All the imported cars have been bought *[out]??They bought all the imported cars for themselves?c. Se distribuyeron (todos los) coches importados.SE distribute.3PL.PST all the cars imported.?All the imported cars have been distributed??They distributed all the imported cars to themselves??They distributed all the imported cars among themselves?Besides, if we look at the effects of SE in the aspectual specification of the event, curious results obtain. For one thing, the benefactive quirk concurs with an aspectual closure not observed in the passive SE, and which happens to be crucial in the determination of telicity, according to classical tests (2). Yet, and unlike other verbs that seem to yield the same scenario?buscar [search], guardar [keep], preparar [prepare], escribir [write], alquilar [rent], reservar [preserve] etc.?not all verbs entering the pattern of comprar allow a dative argument outside the SE construction (as is the case for ganar [win], acabar [finish], quedar [keep], acaparar [hoard]). Somehow, this is no impediment for SE to trigger a benefactivereading with a direct consequence on aspectual closure; nor for the predicate to accommodate an additional PP explicitating this benefactive semantic quirk (3)?which should not be confused with the affectedness reading in anticausautive SE with unselected dative (Mendikoetxea & F. Soriano 2010). On the other hand, the definiteness of the NP (a crucial variable for the aspectual closure of most transitive verbs as noted since Tenny 1985 i.a. and also relevant for the passive SE alternative; cf. (2)a) is not relevant in these cases either. Even when the case at stake may seem to be just another example of the aspectual SE commonly appearing with universal quantifiers such as todo [all] especially with consumption verbs (cf. (4))?in which case the passive reading is not available?, in the case of verbs like comprar the determiner is not a requirement for the occurrence of SE and its telicity-inducing effect (5). In other words, telicity is triggered independently of the definiteness contributed by the internal argument?as noted with certain (e.g. motion) verbs since Talmy 1985. In this sense, comprar/buy seem to have richer descriptive content.(2) a. (Se) vendieron *(todos los) coches importados en una hora.?All the imported cars were sold in an hour?b. *(Se) compraron (todos los) coches importados en una hora.?They bought (all the) imported cars [for themselves] in an hour?.c. *(Se) distribuyeron (todos los) coches importados en una hora.?They distributed (all the) imported cars [to themselves] in an hour?(3) Se {compraron/quedaron/acapararon/ganaron/*perdieron/*vendieron} coches [para ellos].?They bought/kept/hoarded/won/*lost/*sold} cars [for themselves]?(4) a. Se comió todo.?[He/She] ate it all [up]? (cf. *Everything has been eaten)b. Se compró todo.?[He/She] bought it all [for herlsef]? (cf. Everything has been bought)(5) a. Se comió *(el) postre.?[He/She] ate up *(the) desert?b. Se compró (el) postre.?[He/She] bough (the) desert [for herself]?Analysis. Even if the problem can be addressed from multiple perspectives, we believe that more secure grounds for unveiling the relevant differences between the verbs under consideration, the grammatical status of SE and its participation in aspectual closure in these cases are provided by (l)syntactic analysis of verb structure. Specifically, we will appeal to both Hale & Keyser?s (1993, 2002, 2005) theory of argument structure and P-conflation processes involved in the delivery of lexical verbs; but also to Hale?s (1996) and Kayne?s (1993, 1994) account of possessive semantics ensuing from V-incorporation of abstract prepositions; in our case building on Central Coincidence prepositions [PC] (an element central to Hale & Keyser?s analysis). Specifically, we will argue that verbs like comprar have its source in a stative locative relational element (PC) like the one found in verbs such as have (cf. Hale 1986, Freeze 1992, Kayne 1993, 1994); this element being basically defined as a non-eventive predicate that expresses the relation of spatial contiguity whereby some entity is associated or in contact with another entity, roughly comparable in meaning with the lexical P 'with' (Cf. Rapoport 2012, Rigau 2005). However, as opposed to have, in our case the verb hosting the stative predicate (PC) is not semantically empty. The burden of its semantic properties can be noted in the fact that comprar [buy] involves, unlike have, a process giving the spatial contiguity (i.e., the central coincidence) as result. If the proposed analogy holds, then verbs following this pattern would also be monadic structures (V0) hosting a semantically contentful non-eventive predicate in its complement (i.e., the covert PC) (6). Following Hale & Keyser?s schema of verbal derivation, V0 would conflate with the pohonologically null PC, giving verbs like vender as a result. This (birrelational) head (i.e.,PC) would instantiate the additional (possession) state lacked by vender?ultimately, it is true that the relation between the two arguments involved is not one of change but of stasis?, besides locally selecting the two arguments involved: the (unspecified) subject of the possessive relation?SE having been regarded as silent ?indefinite? subject (cf. Cinque 1988, Raposo and Uriagereka 1996, Rivero & Sheppard 2002 i.a.) or as a binder of a null subject (e.g. Mendikoetxea 1999), or, alternatively, as an argument-like element merged in the vPinternal subject position (cf. Torrego 2013)?and the object ultimately possessed by the former as a result of the event taking place. Hence, the alternative reading in (1) would fall out from the circumstance that SE would be needed to bind the VP external subject (not always phonologically realized, Spanish being a NS language) and the internal one (i.e., would identify the subject of the process event and of the result [possessive] stative relation) (7). In its turn, the impossibility for the object to measure out the event (cf. (2)b-c) would be easily accommodated by the fact that this argument is not setting in the correct position (i.e., sister-to-V position; cf. Harley 2005, following Hale & Keyser 1993). Finally, the additional benefactive PP (para ellos) would be accommodated as a double of the V-conflated P (cf. Gallego 2012)?note, in fact, that there is an anaphoric relation linking the PP and SE. In the case of vender no PC would be conflated into the verb. From here, two consequences follow: the subject (if realized) would originate in the position of external arguments?i.e., outside the VP projected by the monadic V0 which in turn selects the internal argument; therefore, vender can only handle SE under passivization. In its turn, the object would be sitting in sister-to-V position and, hence, object-event homomorphism (cf. (2)a)?i.e., the correlation between bounded object and bounded/telic event?shall be expected. Finally, the benefactive PP does not occur, for there is no P for doubling to apply.(6) VPV PCPDP1 PCPC DP2(7) Cristina y Néstor se {compraron/distribuyeron/*vendieron} los coches importados.Cristina and Néstor SE bought distributed sold the cars importedCristina and Néstor {bought/distributed/*sold} the imported cars [to/for themselves]?Conclusion. Whereas the ambiguity pointed out in data such as (1) can be analysed from very different angles (and brought under different frameworks), we believe that this evidence operates the other way round. That is, the availability (or not) of benefactive readings for SE constructions?and the relevance of object boundedness?should rather been seen as an excellent probe into relevant contrasts determined by verbal meaning and syntax which would go otherwise unnoticed. At any rate, the fact that a SE structure allows a specific reading with specific implications for aspectual closure only with a certain class of transitive verbs in Spanish is an important clue that an approached privileging components encoded in the V may well be on the right track.Selected references.De Miguel, E. & M. F. Lagunilla. 2000: ?El operador aspectual se?. REL, 30, 1, pp. 13-43.Hale, K. & Keyser, S.J., 2002, Prolegomenon to a Theory of Argument Structure, MIT Press.Kayne, R. S. 1993. "Toward a Modular Theory of Auxiliary Selection." Studia Linguistics 47,3-3 1.Mendikoetxea, A., 1999, ?Construcciones con se: medias, pasivas e impersonales?, in I. Bosque & V. Demonte (eds), Gramática descriptiva de la lengua española. 2 1631-1711. Espasa Calpe, Madrid.Rapoport, Tova. 2012. ?Central coincidence: The preposition with?. In Jean-Marie Merle & Agnès Steuckardt. Prépositions & aspectualité. Bibliothèque de Faits de Langues, Paris, Ophrys.Rigau, G. 2005. ?Number agreement variation in Catalan dialects?. In Cinque, G. & R. Kayne (ed) The Oxford handbook of comparative syntax; p. 775-805.Torrego, Esther 2013. ?Revisiting Romance SE?. In Gramatika Jaietan: Patxi Goenagaren Omenez, X. Artiagoitia and J. Lakarra (eds.), Universidad del País Vasco, 785-791.Zagona, Karen. 1996. Compositionality of aspect: Evidence from Spanish aspectual se, in C. Parodi et al. (eds.), Aspects of Romance Linguistics. Washington, DC: GUP, 475-488.