Null Directional Prepositions in Spanish & Romanian: Non-reflexive & Reflexive Clitics, Lecture notes of Spanish

Download Null Directional Prepositions in Spanish & Romanian: Non-reflexive & Reflexive Clitics and more Lecture notes Spanish in PDF only on Docsity! NULL DIRECTIONAL PREPOSTIONS IN ROMANIAN AND SPANISH JONATHAN E. MACDONALD Stony Brook University 0. Introduction The main goal of this paper is to provide a formal account of non- reflexive non-argumental clitic pronouns of Spanish (1) and reflexive non- argumental clitic pronouns of Romanian (2).1 The clitics are in bold. (1) (Yo) le lavé el coche (I) to-him washed the car “I washed his car’ or ‘I washed the car on/for him.” (2) Eu mi-am spălat maşina2 I myself-have washed car-the “I washed my/the car.” I will focus on five properties associated with these constructions: 1. the inability to produce a telic interpretation of a predicate; 2. the lack of on/with entailment; 3. the ability to express temporary relations; 4. the ability to prevent idiomatic interpretations; and 5. the resistance to adjectival secondary predication. The Spanish reflexive shows all five properties; the Romanian reflexive shows the first four. To account for these properties, I propose that both the Spanish non- reflexive and the Romanian reflexive are introduced as the complement of a null directional-like preposition that merges as a complement of the verb (3). 1 I refer to non-reflexive non-argumental clitic pronouns as non-reflexives, and to reflexive non-argumental clitic pronouns as reflexives. 2 Romanian also has clitics in accusative case. The accusative clitics show fundamentally different properties from the dative clitics. Throughout the paper, all of the Romanian clitics are dative, unless otherwise noted. JONATHAN E. MACDONALD 170 The directional-like nature of the null preposition is motivated by a set of properties shared with overt directional prepositions. (3) …V’ ri V PP wo P DP ØDIR 4 Span. non-refl./Rom. refl. To motivate the structure in (3), it is instructive to first consider the properties of Spanish reflexive non-argumental pronouns. Therefore, in section 1, I outline a discussion of Spanish reflexives from MacDonald (2004); he shows that reflexives pattern with goal prepositions. In section 2, I compare Spanish reflexives with Spanish non-reflexives; I show that non-reflexives pattern with directional prepositions. In section 3, I show that Romanian reflexives pattern with Spanish non-reflexives. In section 4, I conclude by briefly considering the question of why Spanish reflexives pattern differently from Romanian reflexives and Spanish non-reflexives. 1. The properties of Spanish reflexives In this section, I discuss properties of Spanish reflexives. The discussion is taken from MacDonald (2004), in which he shows that Spanish reflexives share three properties with goal prepositions: 1. they produce a telic interpretation of the predicate; 2. they express an on/with entailment; and 3. they can prevent idiomatic interpretation. This section provides a starting point for illuminating the properties associated with Spanish non-reflexive constructions. 1.1. Telic interpretation of the predicate The presence of a Spanish reflexive results in a telic interpretation of a predicate (MacDonald 2004, Nishida 1994, Zagona 1996). As evidence, observe the (in-)compatibility of the durative phrase (e.g. durante una hora ‘for an hour’) in (4). (4) a. Lavé el coche durante una hora. washed the car for an hour “I washed the car for an hour.” NULL DIRECTIONAL PREPOSITIONS IN ROMANIAN AND SPANISH 173 complement of the goal preposition. Goal prepositions and reflexives express an on/with entailment. 1.3. Preventing idiomatic interpretation Finally, MacDonald (2004) observes that the presence of a reflexive can prevent idiomatic interpretation.5 Consider the Spanish idioms in (11). (11) a. Jose cortó los lazos con Ana Jose cut the ties with Ana “Jose broke up with Ana.” b. Juan come la sopa boba Juan eats the soup stupid “Juan is a sponger.” Observe that when a reflexive pronoun is added to the utterances in (11), their idiomatic interpretation is lost. This is illustrated in (12). (12) a. # Jose se cortó los lazos con Ana. b. # Juan se come la sopa boba. Observe that goal prepositions can prevent idiomatic interpretation as well (13). (13) a. # John kicked the bucket to the barn. b. # John spilled the beans into the sink. Goal prepositions and reflexives can prevent idiomatic interpretation.6 1.4. The structure of Spanish reflexive pronouns Given the properties in common with goal prepositions, MacDonald (2004) proposes that the Spanish reflexive is introduced as a complement of a null goal-like preposition that merges as the complement of the verb. This is illustrated in (14). 5 De Miguel and Fernández Lagunilla (2000) make a similar observation. 6 Goal prepositions and reflexives do not always prevent idiomatic interpretation: John spilled the beans to the police; sincomer(se)lo ni beber(se)lo, “without deserving it”. This ability depends in part on the semantic contribution of the additional element (see Nunberg et al. 1994). MacDonald (2006a) discusses the effect of Spanish reflexives (and non-reflexives) on idiomatic interpretation. See footnote 11. JONATHAN E. MACDONALD 174 (14) …V’ ri V PP ru P DP ØGOAL 4 Span. refl MacDonald (2004) accounts for the range of properties associated with the reflexive in the following way. The property of an overt goal preposition that allows it to produce a telic interpretation of a predicate is also present on the null goal-like preposition.7 The on/with entailment results from a property of the null preposition that contributes to a compositional theta-role assigned to the direct object by the verb and prepositional phrase8, such that the object is interpreted as on/with the denotation of the complement of the goal-like preposition. That is, the object is interpreted as on/with the denotation of the reflexive and given that the reflexive is coindexed with the subject, the object is interpreted as on/with the denotation of the subject. Finally, MacDonald (2004) argues that the ability to prevent idiomatic interpretation results from the low merger of the null preposition. Typically, only elements merged within the verb phrase can affect idiomatic interpretation (Marantz 1984 and Svenonius 2004). Let us now consider the properties of Spanish non- reflexives. 2. The properties of Spanish non-reflexives In this section, I discuss Spanish non-reflexives. Non-reflexives pattern with directional prepositions in four ways: 1. they do not produce a telic interpretation of the predicate; 2. they do not express an on/with entailment; 3. they can prevent idiomatic interpretation and; 4. they resist adjectival secondary predication. As such, I propose that the non-reflexive is introduced as a complement of a null directional-like preposition that merges as a complement of the verb. I make these properties of the non-reflexive salient by contrasting them with the properties of the reflexive. I argue that the differences between the reflexive and the non-reflexive result from a distinct set of properties associated with the null prepositions introducing them. I supply more evidence 7 Directional prepositions lack this property. See (16) below. 8 Larson (1988) argues for compositional theta-roles assigned by a verb plus prepositional phrase. NULL DIRECTIONAL PREPOSITIONS IN ROMANIAN AND SPANISH 175 for the distinct set of properties by drawing attention to yet another contrast between these constructions in section 2.2. 2.1. A comparison of Spanish non-reflexives and Spanish reflexives As we saw in section 1, Spanish reflexives pattern with goal prepositions in the following ways: 1. they produce a telic interpretation; 2. they express an on/with entailment; and 3. they can prevent idiomatic interpretation. We will see that Spanish non-reflexives pattern with directional prepositions in the following ways: 1. they do not produce a telic interpretation of the predicate; 2. they do not express an on/with entailment; and 3. they can prevent idiomatic interpretation. Consider the utterance in (15). (15) (Le) lavé el coche durante una hora (to-him) washed the car for an hour “I washed his car for an hour.” or “I washed the car for him for an hour.” The non-reflexive does not affect the compatibility of the durative. The non-reflexive does not produce a telic interpretation. Observe that directional prepositions pattern in the same way (16). (16) a. John drove the car (toward the garage) for an hour. b. John carried the bag (toward to beach) for an hour. They do not affect the compatibility of the durative. They do not produce a telic interpretation. Consider the utterances in (17a-b). (17) a. Le abroché la camisa to-him buttoned the shirt “I buttoned his shirt.” or “I buttoned the shirt on/for him.” b. Le lavé el coche to-him washed the car “I washed his car.” or “I washed the car on/for him.” (17a) can be said in a context in which the shirt is on a hanger during the buttoning. In (17b), the individual denoted by the non-reflexive does not have to be present while the car washing takes place. The non-reflexive does not express an on/with entailment. Observe that directional prepositions pattern in the same way (18). (18) a. John drove the car toward the garage. JONATHAN E. MACDONALD 178 verb, it merges lower than the secondary predicate and, as such, the non- reflexive cannot control PRO. Thus, no predication relation can be established between a non-reflexive and a secondary predicate.10 Recall that MacDonald (2004) analyzes Spanish reflexives as the complement of a null goal-like preposition that merges as the complement of the verb (14). Given that proposal, and given the structure proposed here for non-reflexives (24), it is natural to conclude that any diverging patterns between the two (i.e. the (in-)ability to produce a telic interpretation and express an on/with entailment) result from distinct properties of the null prepositions themselves. In fact, I claim that the source of variation between reflexives and non-reflexives is the distinct properties associated with the null prepositions that introduce them. Observe another property that is a point of divergence between these null prepositions.11 Consider a context in which I work at a car wash and I am assigned cars to wash. My boss, who is concerned with whether or not I washed a particular car assigned to me, asks whether I did so. To tell him that I washed my car, I cannot use the reflexive pronoun. Thus, the utterance in (25) is not a felicitous response to his question. (25) # Me lavé el coche myself washed the car. “I washed my/the car.” Consider a similar situation in which my coworker is assigned a particular car that my boss asks about. Let us say that I washed his car. I can utter the sentence in (26) with the non-reflexive pronoun to respond to my boss.12 10 Demonte (1988) has an account of secondary predication that depends on mutual c- command between the secondary predicate and the DP with which it establishes a predication relation. Given that the non-reflexive is housed in a prepositional phrase on the present account, it cannot c-command out to establish the mutual c-command relation. Thus, under Demonte’s proposal, the resistance to adjectival secondary predication provides evidence for the existence of the null prepositional phrase itself. However, it is not clear that complements of prepositions generally cannot c-command out of a prepositional phrase. Observe that the complement of the preposition can c-command out and control the PRO of an embedded clause: John gave money to Billj PROj to buy milk. 11 MacDonald (2006a) discusses how the distinct properties associated with these null prepositions can account for the variable behavior of reflexives and non-reflexives in the idiom sin comerlo ni beberlo. Reflexives do not prevent idiomatic interpretation: sin comerselo ni beberselo, while non-reflexives do #sin comemelo ni bebermelo. 12 Example (26) is typical of a benefactive or malfactive in Spanish. Consider another: Juan me compró un libro ‘Juan bought me a book.’ Observe that the benefactive/malfactive NULL DIRECTIONAL PREPOSITIONS IN ROMANIAN AND SPANISH 179 (26) Le lavé el coche to-him washed the car “I washed his car.” or “I washed the car for him.” In the carwash scenario, the reflexive pronoun cannot express a temporary relation established between an object, in this case a car, and the denotation of the reflexive.13 The non-reflexive pronoun can express this temporary relation. This temporary relation is one example of a general difference between these reflexive and non-reflexive constructions with respect to the interpretation of the direct object. In general, the non-reflexive is not as restricted as the reflexive is. The non-reflexive tolerates temporary relations and is not restricted to objects interpreted as on/with. The reflexive does not tolerate temporary relations and is restricted to objects interpreted as on/with. The direct objects in non-reflexive constructions are less restricted in their interpretation than the direct objects in reflexive constructions. More formally, I claim that the range of possible relations between the direct object and the denotation of the (non-)reflexive is regulated via the compositional theta role assigned by the verb plus prepositional phrase. The restrictions on the interpretation of the direct object vary according to the properties of the preposition. 2.3. Recapping the findings The properties of Spanish non-reflexives that we have uncovered are listed in (27). (27) a. Do not produce telic interpretation → like directional PP b. Do not express on/with entailment → like directional PP c. Can express a temporary relation d. Can prevent idiomatic interpretation → like directional PP e. Resist adjectival secondary predication → like directional PP The common properties between the non-reflexive and the directional preposition motivate the structure in (24) in which the non-reflexive is introduced as a complement of a null directional-like preposition that merges as a complement of the verb. The properties in (27a-c) are argued to result interpretation is independent of the reflexive/non-reflexive contrast: Juan se compró el libro. ‘Juan bought himself a book.’ The benefactive/malfactive interpretation is independent of the present discussion. 13 MacDonald (2004) refers to this as temporary possession. I use relation to remain neutral. JONATHAN E. MACDONALD 180 from the nature of the null preposition itself, while the properties in (27d-e) are argued to result from the low merger of the null preposition. Consider now the properties of Romanian reflexives. 3. The properties of Romanian reflexives In this section I show that Romanian reflexives share the first four properties in (27) with Spanish non-reflexives.14 Based on these common properties, I propose that Romanian reflexives, like Spanish non-reflexives, are introduced as the complement of a null durational-like preposition that merges as the complement of the verb. Consider the utterance in (28). Observe that the reflexive does not affect the compatibility of the durative. The reflexive does not produce a telic interpretation.15 (28) Eu mi-am spălat maşina timp de zece minute I myself-have washed car-the time of 10 minutes “I washed my/the car for 10 minutes.” Consider the utterances in (29). (29) a. Eu mi-am pătat cămaşa I myself-have stained shirt-the “I stained my/the shirt.” b. Eu mi-am spălat maşina I myself-have washed car-the “I washed my/the car.” (29a) can be said in a context in which the shirt is not on the subject16, and (29b) in a context in which the subject dropped the car off at the carwash 14 We cannot test whether or not the reflexives resist adjectival secondary predication, for we cannot determine whether predication is established with the subject or with the reflexive itself. 15 Observe that in the following sentence, the durative phrase is incompatible and there is a reflexive pronoun present: Eu mi-am mâncat mărul *timp de zece minute “I ate my/the apple for an hour.”; observe nevertheless that the durative phrase remains incompatible when the reflexive is not present: Eu am mâncat mărul *timp de zece minute. This suggests that the reflexive does not produce the telic interpretation here; the telic interpretation most likely results from the properties of the verb itself. 16 Manoliu-Manea (1996) observes that when the reflexive is accusative, an overt preposition (la) is required, and the shirt is necessarily interpreted as on/with the denotation of the clitic: Eu m-am pătat la cămaşă “I stained my shirt”(The example is my own). Given the case difference and the presence of the overt preposition, I take this to be a distinct construction NULL DIRECTIONAL PREPOSITIONS IN ROMANIAN AND SPANISH 183 null telic morpheme while non-reflexives and Romanian reflexives are not? There may be telling correlations among the properties in Table 1 that can provide an answer to this specific question. The patterns in Table 1 give the impression that there is a correlation between an on/with entailment and the ability to produce a telic interpretation. It is not clear, however, that this correlation exists. MacDonald (2006b) observes in French that the presence of a reflexive pronoun results in an on/with entailment (32a) that otherwise does not exists without the reflexive (32b). (32) a. Jean s’est boutonné la chemise Jean himself-is buttoned the shirt “Jean buttoned his shirt.” b. Jean a boutonné la chemise Jean has buttoned the shirt “Jean buttoned his shirt.” Moreover, he notes that the presence of the reflexive pronoun does not affect the compatibility of the durative phrase (33).18 (33) a. Jean s’est/a boutonné la chemise pendant une heure Jean himself-is/has buttoned the shirt for an hour “Jean buttoned his shirt for an hour.” These facts suggest that the property responsible for producing a telic interpretation is independent of the property responsible for expressing an on/with entailment. If these properties are indeed independent, then, there is no way to determine if a preposition can produce a telic interpretation of a predicate if it expresses an on/with entailment, or vice-versa. Nevertheless, as an independent property, we can assume that some heads have the property and some do not. This does not tell us why some heads have the property nor why some heads do not, but it provides us with a way to understand the scope of possible (cross-linguistic) variation based on the presence or absence of the properties in Table 1. That is, given the range of properties, we can determine whether or not any is contingent on the presence or absence of another property, or whether each is a property truly independent of all others. In the former case, we could begin to find insight into and perhaps understand more clearly the patterns of these Romance (non-)reflexive constructions. In the 18 There does seem to be some variation with respect to the compatibility of the durative phrase here. JONATHAN E. MACDONALD 184 latter case, we expect to find a (null) preposition for each of the possible combinations of the presence and/or absence of each property in Table 1. In this latter case, it would not be clear what the insight into the divergent patterns of these distinct Romance (non-)reflexives would be. References Beck, Sigrid and William Snyder. 2001. “Complex Predicates and Goal PP’s: Evidence for a Semantic Parameter”. Proceedings of the 25th Boston University Conference on Language Development, ed. by Anna H.-J. Do, Laura Dominguez & Aimee Johansen, 114-122. Somerville: Cascadilla Press. Bowers, John. 2000. “Predication”. The Handbook of Contemporary Syntactic Theory, ed. by M. Baltin & C. Collins, 299-333. Malden, Mass.:Blackwell. De Miguel, Elena & Marina Fernández Lagunilla. 2000. “El operador aspectual se”. Revista Española de Lingüística 30:1.13-43 Demonte, Violeta. 1988. “Remarks on Secondary Predicates: C-command, Extraction, and Reanalysis”. The Linguistic Review, 6:1-39. Dowty, David. 1979. Word Meaning and Montague Grammar. Dordrecht: D. Reidel. Larson, Richard K. 1988. “On the Double Object Construction”. Linguistic Inquiry 19:3.335-391. MacDonald, Jonathan E. 2004. “Spanish Reflexive Pronouns: a Null Preposition Hypothesis”. Proceedings of WCCFL 23.528-540. ed. by Vineeta Chand, Ann Kelleler, angelo J. Rodríguez & Benjamin Schmeiser. Somerville: Cascadilla Press. -----------. 2006a. “Non-reflexive, Non-argumental Clitic Pronouns of Spanish”. Proceedings of WECOL 2004, ed. by Asier Alcazar, Roberto Mayoral Hernández & Michal Temkin Martinez. Fresno: California State University. ----------. 2006b. The Syntax of Inner Aspect. Doctoral Dissertation, Stony Brook University, Stony Brook, NY. Manoliu-Manea, Maria. 1996. “Inalienability and topicality in Romanian: Pragma-semantics of syntax”. The Grammar of Inalienability: A typological Perspective on Body Part Terms and the Part-Whole Relation, ed. by H. Chappell & W. McGregor, 711-743. Berlin: Mouton de Gruyter. Marantz, Alec. 1984. On the Nature of Grammatical Relations. Cambridge, Mass.: MIT Press. Nishida, Chiyo. 1994. “The Spanish reflexive clitic se as an aspectual class marker”. Linguistics 32.425-458. NULL DIRECTIONAL PREPOSITIONS IN ROMANIAN AND SPANISH 185 Nunberg, Geoffrey, Ivan A. Sag and Thomas Wasow. 1994. “Idioms”. Language 70.3.491-538 Pustejovsky, James. 1991. “The Syntax of Event Structure.” Cognition 41.47- 81. Snyder, William. 1995. “A Neo-Davidsonian Approach to Resultatives, Particles, and Datives.” Proceedings of NELS 25.1, ed. by Jill Beckman, 457-471. Amherst: GLSA. Svenonius, Peter. 2004a. “Slavic Prefixes inside and outside the VP”. Nordlyd 32.2, ed. by P. Svenonius, 205-253. Tromsø, Norway: Faculty of Humanities, University of Tromsø. Online at www.ub.uit.no/munin/nordlyd/ Zagona, Karen. 1996. “Compositionality of Aspect: Evidence from Spanish Aspectual Se”. Aspects of Romance Linguistics. ed. by Claudia Parodi, Carlos Quicoli, Mar89 Saltarelli, and María Luisa Zubizarreta, 475-488. Washington DC: Georgetown University Press.