CONICET | Buscador de Institutos y Recursos Humanos

1. Contrary to assumed wisdom (e.g. Kayne1994), the Double Object Construction [DOC] could not only be possible in Romance, but actually accommodate interesting facts; 2. SE-CL, known for its many functions could be serve enabling DOC variants in Spanish. 3. Configurational asymmetries associated with Benefactive (BEN) Alternation in English (Pesetsky1995, Larson1990) may: (i)capture systematic availability of diverse readings for SE in Romance, also as an aspectually-relevant constituent, along with availability of Recipient/Possessor interpretation; and (ii)bring out finer structural contrasts among verbs otherwise lumped together, by exposing deviant behavior. Notably, while some change-of-possession [COP] verbs in Spanish like vender ?sell?, perder ?lose?, pagar ?pay? in (1)a yield a passive-impersonal (α) reading generally seen across COP verbs; others, like comprar ?buy?, asegurar ?secure?, ganar ?win?, robar ?steal? (1)b systematically yield not-often discussed benefactive reading (β) as well. We note that: (A) (β) is unavailable in (1)a, and different to the reciprocal reading (γ) seen in (1)c. (B) Even if not all verbs giving (β) systematically take benefactive PPs, this is no impediment for SE to license (β)-reading with a consequent telicity computation (as in (2)), nor for the verb to accommodate an additional PP expressing COP beneficiary (different from unselected affected dative in anticausative SE; Mendikoetxea & F. Soriano 2010) ((8) below) if cliticized. Further: (C) the relation between SE and aspect is key to (β), insofar as the benefactive quirk concurs with a type of telicity independent of (theme) measuring-out, unseen in (α)|(γ); and (D) even if (β) may seem a further instance of aspectual SE seen with universal quantifiers (todo ?all?) in cases like (3) (cf. Campanini & Schäfer 2011, Mc Donald 2016), the contrast in (4) is key to the fact that event endpoint in (β) is defined by BENEFACTIVE (POSSESSIVEGOAL, Levin 2008).(1)a. Se vendieron (todos los) coches importados. (Lit. SE sell.3P.PST all the cars imported) (α)?All the imported cars have been sold [out]? b. Se compraron (todos los) coches importados.(Lit. SE buy.3P.PST all the cars imported) (α)?All the imported cars have been bought *?[in/out]? (β)?They bought all the imported cars for themselves? c. Se asignaron (todos los) coches importados. (Lit. SE lend.3P.PST all the cars imported) (α)?All the imported cars have been assigned? (β)?They assigned all the imported cars to themselves?(γ)?They assigned (all) the imported cars to each other?(2)a. (Se) {vendieron/perdieron} *({todos los/?algunos}) coches importados en una hora. (α)-reading ?(All the/Some) imported cars were sold/lost within an hour?b. *(Se) {compraron/ganaron} ({todos los/algunos}) coches importados en una hora. (β)-reading ?They got (lit. bought/won) (all the/some) imported cars [for themselves] within an hour?.c. (Se) distribuyeron/asignaron *({todos los/algunos}) coches importados en una hora. (γ)-reading?They distributed/assigned (all the/some) imported cars [among themselves] within an hour?(3)Se comió todo.(Lit. She ate all) ?She ate it all [up]? (argument-to-event homomorphism, Krifka 1998)(4)a. Se comió *(el) postre (*para ella) (Lit. She ate the desert *for herself) ?She ate up the desert?b. Se compró (el) postre (para ella) ?She bought (the) desert (for herself)?The analysis combines Hale&Keyser?s [HK](2005) P-conflation and Larson?s (1990) Alternative Projection, but also semantic works contributing new conclusions on DOC (cf. Levin 2008 i.a.). We suggest that verbs like comprar involve a nondynamic birrelational head (po) similar to the one found in possessive verbs (Benveniste 1966, Hale 1986, Freeze 1992, Kayne 1993; Guéron 1995), also identified with the lexical P with (HK 2002, McIntyre 2002, Rigau 2005, Rapoport 2012). It would be this head that yields the result state (POSS Small Clause) accounting for the contrasts above. Yet, the light verb combined with pHAVE would not semantically trivial: its content can be noted in the fact that COP like comprar involve, unlike have, a caused event producing spatial contiguity (cf. central coincidence, HK) (possession) as result. Verbs like comprar would thus decompose into two heads: (i) a monadic (causative) vo typically selecting for SCs (vCAUSE, Harley 2004, Folli & Harley 2007); (ii) a pº heading the stative (SC) predicate instantiating the POSS final state disallowed by vender. Such asymmetry captures the contrast between CAUSEDPOSSESSION|CAUSEDMOTION, along with CAUSEDPOSS meaning (10)a-b across DOC/?dative to? variants (cf.(10)c-d, both expected patterns in transfer verbs crosslinguistically attested in DOC (Levin 2008 i.a.)), SE imposing the expected animacy restrictions (≈possible possessor), just as instances motivating vCAUSE+PHAVE account in other languages (Harley & Miyagawa 2016).Advantages: ?This configuration (also used to explain possessive meaning in DOC, Harley & Jung 2015 i.a.) accommodates event complexity (see Krifka 2004), along with the fact that the semantic relation between the two arguments is actually not one of change (motion) but of stasis (the inner POSSstate, empirically brought out by for-x-time adverbials (6)) articulated by pHAVE. ?The additional benefactive PP in (8) is not unexpected, but readily explained as further instantiation of the incorporated P, as its hyponymic (=possessor) interpretation suggests (expected in P-incorporation, Haugen 2009, Mateu 2012). ?THEME being trivial to telicity (2) also follows from the proposed configuration, as this argument is not sitting in the relevant (measure-out) position. Instead, pHAVE is, defining telicity accordingly (i.e., based on the POSSstate yielded by the SC it heads, Harley 2005) (5). ?Presence of pHAVE explains the distribution in (1) (SE being regarded as silent ?indefinite? (possessor) subject of the predicative relation (subject binder, Torrego 2013) acting as adequate controller (8)). ?Configurations incompatible with pºHAVE thus do not allow possessors to c-command possessees, and disallow (β)-reading, where GOALS c-command themes (basic requisite in HK for P-incorporation).In (1)a SE can only be handled under passivization; a benefactive PP does not occur; and the configuration is different: THEME in int argument position and GOAL (headed by the lexical P a ?to?) as its complement (vs. comprar, allowing a structure with GOAL as inner subject of the SC (Kayne 1993) and THEME as its complement (7)). ?This structural opposition lines up with a major aspectual opposition between delimitness (GOAL-driven) and measure-out (THEME-dictated telicity) (Tenny 1994), and hence with a structural contrast behind potential roles of SE in aspectual determination (3)-(4). DOC availability, with the consequent opposition between CAUSED POSSESION|CAUSED MOTION, could explain why lexically-coded direction of motion is trivial (traer|llevar ?bring from|to? both give (β)), as long as pHAVE is possible (9)a, or why dejar (9)b (and verbs of dipossession, e.g. robar ?steal?, Haspelmath 2014) apparently similar to vender, allow (β)/DOC.(5)a. Poco a poco se vendieron todos los coches. ?Little by little, all the cars were sold?b. Poco a poco se compraron todos los coches.?Little by little, all the/every car(s)ended up being theirs?(6)(se) compró un coche por una hora. ?She bought a car for an hour? (≈the car was hers/with her for 1h)(7)[vP DP1i [vCAUSE, √pP [GOALsei, p? [pºHAVE THEMEDP2]]]] (see also Harley 2002, Folli & Harley 2007)(8)(Los inspectoresi) *(sei) {quedaron/acapararon/incautaron/alquilaron} coches (para ellosi).The inspectors kept/monopolized/confiscated/rented cars for themselves (≈they got them by renting) (9)a. Se trajeron/llevaron los coches ?They brought/carried the cars to themselves? (≈they got the cars).b. Se dejaron/cortaron/sirvieron/separaron postre ?Theyi left/cut/served/cut off dessert (for themi)?(10)a. DOC: se compró el coche (CAUSEDPOSSESSION)b. DC: compró el coche a Juan (CAUSEDPOSSESSION)c.DOC: se llevó el coche(CAUSEDPOSSESSION)d. DC: llevó el coche a Juan (CAUSEDMOTION)(11)El presidente fue robado (lit. The president was stolen) ?The president was burglarized?In this way, contrasts between otherwise similar ditransitive verbs, occurrence of benefactive PPs, and argument interpretation get a unified explanation. The proposal: (i) allows to consider SE as aspectually-relevant element in a different way (cf. more traditional approaches, Zagona1996, deMiguel&Lagunilla 2000), while it preserves key oppositions in telicity calculation (measure-out|delimitness) along with distinctions relevant to (4) (e.g. Kemmer?s 1993 inherent refl/disjoint verbs, also capturing distribution/ DOC availability with speech verbs, Krifka 2004); (ii) may capture otherwise unexpected data like (11); (iii); while it remains amenable to other major analyses (Pylkännen 2000, see Miyagawa&Tsujioka 2004 on Japanese); (iv) reflects crosslanguage regularities linking verb meaning to argument realization, but also contrasts necessarily following from distinct morphosyntactic resources available in each language for argument expression (e.g. cl-SE in Romance). Future work will confirm if observations generalize across Romance, as preliminary Italian/Portuguese/Catalan data suggest: e.g., although some Italian natives allow (β) in all COPs, this could explain distinct clitic order correlated with an (α)-(β) contrast (cf. la si compra (ACC.f|SE|buy) ?one buys it? vs. se la compra (SE|ACC.f|buy)?he buys it for himself?), along with nontrivial contrasts in participle agreement (PPT-ObjAgr. (ce la si è comprata) in (α); PPT-SubjAgr. (ce la si è comprate) only in (β). Then, a lot more crosslanguage variation could be accounted for.

enviar mensaje