The acquisition of dative "se" in L1 Spanish and its implications for the Weak Continuity Hypothesis of Acquisition of Functional Categories.
Spanish makes an extensive use of the clitic "se" in a wide variety of syntactic environments (2). In other words, there are several kinds of "se" in Spanish, and therefore, many analyses that have been developed to account for its distribution and its semantic contribution in different types of predicates (De Miguel and Fernandez Lagunilla 2000; Gonzalez Vergara 2006; Martin Zorraquino 1979; Mendikoetxea 1992, 1997, 1999ab, 2008, 2012; Molina Redondo 1997; Otero 1986, 1999; Raposo and Uriagereka 1996; Rigau 1994; Sanz 1995; Teomiro 2010, 2011; among others).
The semantic contribution of "se" is varied and far from uncontroversial (inchoative marker, reflexive marker, passive marker, telic marker, etc.), and so are the different kinds thereof that have been claimed to exist (see references above). Nevertheless, two kinds of "se" (alternating vs. non-alternating "se") can be distinguished on the ground of purely distributional properties (i.e. whether its presence is mandatory for the sake of grammaticality or not):
A) Non-alternating "se", whose presence is mandatory. It can occur in two kinds of syntactic environments:
A.1) In intransitive configurations, it correlates with a change in the structure of participants of the event denoted by the verb either by increasing or reducing the valence of the verb. Thus, its presence is mandatory when this change is conveyed, otherwise, the sentence is ungrammatical (i.e. it is non-alternating). This kind of "se" occurs with anticausative-inchoative verbs like (1), and reflexive verbs like (2).
A.2) In transitive or accusative contexts, non-alternating "se" correlates with an implication of inalienable possession between the subject and the object: in (3a) the head is Juan's, and so are the knee in (3b) and the hands in (3c) (they are parts of Juan's body, which is encoded by an inalienable possession relationship). If the clitic is missing (as can be seen in the examples in (6), produced by children), the sentence is ungrammatical (i.e. it is non-alternating).
B) Alternating "se", whose presence is not mandatory. This does not amount to saying that its presence is optional, though. Like non-alternating "se", it can occur in two kinds of syntactic environments:
B.1) In intransitive configurations, like non-anticausative intransitive verbs in (5).
B.2) In transitive configurations, like consumption verbs in (4).
In both (B1) and (B2), its presence does not seem to alter the structure of participants of the event (i.e. the valence of the verb remains unchanged) but seems rather related to aspectual factors (cf. De Miguel and Fernandez Lagunilla 2001). Its presence is not mandatory since the grammaticality of the predicates is independent of
whether "se" occurs or no. Nevertheless, this kind of "se" is not optional in the sense that when the clitic is present, there are changes in the aspectual properties of the predicate.
(1) La ventana *(se) [anticausative-inchoative cerro / rompio. verb] The window CL closed / broke. "The window broke / closed." (2) Juan *(se) peino / lavo. [reflexive verb] Juan CL combed / washed. "Juan combed washed (himself)." (3) a. Juan *(se) rompio la crisma. [transitive verb] Juan CL broke the head. "Juan broke his head." b. Juan *(se) ha manchado la rodilla. [transitive verb] Juan CL has stained the knee. "Juan has stained his knee." c. Juan *(se) ha lavado las manos. [transitive verb] Juan CL has washed the hands. "Juan has washed his hands." (4) a. Juan (se) comio el bocadillo. [consumption verb] Juan CL ate the sandwich. "Juan ate the sandwich. b. Juan (se) leyo el libro. [consumption verb] Juan CL read the book. "Juan read the book." (5) a. Juan (se) cayo. [non-anticausative Juan CL fell off. intransitive verb] "Juan fell off." b. Juan (se) murio. [non-anticausative Juan CL died. intransitive verb] "Juan died." c. Juan (se) marcho a Jaen. [non-anticausative Juan CL left to Jaen. intransitive verb] "Juan left for Jaen."
Thus, the distributional properties of "se" (i.e. whether its presence is mandatory or not) define two major classes (alternating vs. non-alternating), which in turn have several subclasses depending on the valence of the predicate where the clitic occurs (transitive or intransitive).
Having defined these kinds of "se", the question that arises is whether they are natural classes or not. If so, one might expect non-alternating "se" (in (1), (2), and (3)) to be earlier acquired than alternating "se" (in (4) and (5)) because the acquisition of the latter should require exposure to more linguistic data due to its non-mandatory nature. However, this is not borne out by the actual data scrutinized in this paper.
Previous work has shown that children do not usually have problems with the acquisition of non-alternating "se" in predicates like (1) and (2), as expected if nonalternating "se" were a natural class. In fact, Spanish children at the age of 3 have full knowledge of argumental clitics (Baauw 2000; Dominguez 2003; Lopez Ornat 1990; Montrul 2004). Besides, Hogdson showed in her experimental works that children do have problems with the so-called aspectual "se" (which we have named alternating "se") (Hogdson 2002, 2005, 2006), as expected if alternating "se" were a natural class.
However, Escobar and Torrens (2010) have observed, considering data from the Child Language Data Exchange System (CHILDES; MacWhinney 2000) (3), that early
children tend to omit "se" when the clitic occurs with accusative contexts as can be seen in (6), where "se" is missing despite being obligatory in adult grammar as shown by the corresponding examples in (3) above (non-alternating "se" in transitive or accusative contexts). This is unexpected if non-alternating "se" were a natural class.
(6) a. cayo y rompio la quisma. (Irene, 1;11) [fell.sub.3SG] and [broke.sub.3SG] the head. "He fell and broke his head." b. ha manchado la rodilla. (Magin, 1;11) has stained the knee. "He has stained the knee." (cf. CHILDES, Spanish).
Moreover, data from CHILDES corpora provide extra evidence of the problems that children have with alternating "se", as can be seen with non-anticausative intransitive verbs in (7), where the clitic is wrongly omitted. One may argue that the absence of the clitic is not completely ungrammatical in certain readings (cf. Montrul 2004). However, it is on the reading of non-anticausative intransitive predicates that makes clitic "se" obligatory that we will concentrate here.
(7) a. cayo la abuela (Magin 1;10) fell the granny "Grandma fell." b. ha caido mami (Magin 1;11) has fallen mom "Mom has fallen." c. Cayo # ota ve cayo # Irene, 1;8) [fell.sub.3SG] again [fell.sub.3SG] "He fell again." d. lo cojo lo tito y Tina cayo (Irene 1;10) [CL.sub.ACC] [take.sub.1SG] [CL.sub.ACC] [throw.sub.1SG] and Tina fell "I take it, I throw it and Tina fell." (cf. CHILDES, Spanish).
The later acquisition of configurations with non-alternating "se" in accusative contexts (6), as shown in (5), is unexpected if the difficulty of the acquisition is to be explained in terms of the possibility of alternation. This seriously compromises an explanation based on the alternating nature of the clitic because it wrongly predicts that the acquisition of the predicates in (6) (non-alternating "se" in accusative contexts) should be as easy as the acquisition of the predicates in (1) (non-alternating "se" with anticausative-inchoative verbs) or predicates in (2) (non-alternating "se" with reflexive verbs). Therefore, it cannot be maintained that alternating and non-alternating "se" are natural classes, despite their distributional commonalities.
This paper sets out to address why non-alternating "se" in transitive contexts is later acquired than the other kinds of non-alternating "se", and why the latter are earlier acquired than alternating "se". We will proceed as follows: in section 2 we put forward an analysis of the different kinds of alternating and non-alternating "se" in Spanish. We argue that the non-alternating clitic in anticausative-inchoative (1) and reflexive (2) verbs is an expletive anaphor that does not require any functional head to be licensed. Contrarily, the alternating clitic in non-anticausative intransitive verbs, as in examples in (5), as well as the non-alternating clitic in accusative contexts, as in examples in (3), are dative arguments within low applicative phrases. This requires a more complex structure (i.e. the presence of an extra functional head, namely a low applicative, along with the structure it projects). In section 3, we argue in favour of a particular view of the Weak Continuity Hypothesis, according to which the full inventory of functional categories in adult grammar is absent in early child grammar (Clahsen 1990; Clahsen, Penke and Parodi 1993; Meisel and Muller 1992). In section 4 we briefly describe the methodology followed to obtain the data from CHILDES database that are presented in section 5, and that support the hypothesis put forward in section 3. Section 6 is intended as the final conclusion of the paper.
2. THEORETICAL BACKGROUND
In this section, we argue that, regardless the kinds of "se" defined by their distributional properties in the previous section, there are two kinds of "se" on the ground of how they are licensed (i.e. the syntactic environment in which they are allowed to appear):
A) An expletive "se" that is inserted in the Specifier of the verbal phrase due to formal reasons, and does not require any functional head that licenses it nor does it project any further.
B) An argument "se" that is licensed by a functional head, more concretely a low applicative head, and augments the predicate either by adding a result state (in the case of non-anticausative intransitives) or a possessor argument (in the case of transitives).
We will provide the theoretical background that allows us to define such classes and we will put forward two hypotheses as to their acquisition that will be checked against data from the CHILDES database.
2.1. Expletive "se"
In this subsection, we provide an analysis of two instances of the clitic "se". More concretely, we pursue Teomiro's (2010, 2011) idea of "se" as an expletive in anticausative-inchoative and reflexive predicates inserted in [Spec,VP] due to formal reasons.
2.1.1. Non-alternating "se" with anticausative-inchoative verbs
Non-alternating "se" with anticausative-inchoative verbs is an expletive anaphor inserted in [Spec,VP] due to formal reasons, as represented in (8) below. More concretely, it is inserted in order to delete certain uninterpretable features that are present on the predicative and temporal heads. These features are normally deleted by the external argument in the transitive alternate of these verbs. However, when the external argument is deleted and the inchoative alternate is generated, the clitic "se" is the element that deletes these features due to its nominal nature (see Teomiro 2010, for a more detailed analysis).
(8) Se rompio el cristal. CL broke the glass. "The glass broke."
2.1.2. Non-alternating "se" with reflexive verbs
Non-alternating "se" with reflexive verbs is an expletive anaphor inserted in [Spec,VP] due to formal reasons too, as represented in (9) below. More concretely, it is inserted in order to delete certain uninterpretable features that are present on the predicative and temporal heads, as it occurs with anticausative-inchoative verbs. In the reflexive alternate, there is only one syntactic argument and some interpretable features cannot be deleted. Therefore, "se" is inserted in order to delete these features due to its nominal nature (see Teomiro 2011, for a more detailed analysis).
(9) Se afeito Juan. CL shaved Juan. "Juan shaved (himself)."
2.2. Argumental (dative) "se"
In this subsection, we provide an analysis of some instances of the clitic "se" based on low applicatives (Pylkkanen 2008)4. Alternating "se" with non-anticausative intransitive verbs, as in (5), is dealt with in section 2.2.1. Non-alternating "se" with verbs that have an accusative object, as in (6), is addressed in section 2.2.2. Consumption verbs that optionally select for the clitic "se", as in (4), are left aside due to lack of space (see note 4 and Romero and Teomiro 2012; Teomiro and Romero 2012).
2.2.1. Alternating "se" in non-accusative contexts (non-anticausative intransitive verbs)
Teomiro (2013) argues that the clitic "se" that appears with verbs like "caer" (fall), "morir" (die) or "marcharse a un sitio" (leave for somewhere) is within a low applicative phrase. This analysis rests upon three statements:
a) These verbs are conflated verbs (based on Hale and Keyser 2002), as represented in (10). They are formed by a verbal head that denotes undergoing a process plus a nominal element that denotes that process.
(10) a. caer - [V, [begin strikethrough]ACC[end strikethrough] [V UNDERGO] + [NP 'caida' (fall)]ACC]
b. morir - [V, [begin strikethrough]ACC[end strikethrough] [V UNDERGO] + [NP 'muerte' (death)]ACC]
c. marchar - [V, [begin strikethrough]ACC[end strikethrough] [V UNDERGO] + [NP 'marcha' (leave) ]ACC]
b) The conflated noun can denote both a process and a state, as can be seen in (11), where the noun "caida" (fall) (which is non-deverbal) is selected by the copula estar, typical of stage-level predicates (5), rather than by the copula ser, typical of individual-level predicates (6).
(11) a. El arbol *es/esta caido. The tree is fallen. 'The tree is fallen.' b. Juan *es/esta muerto. Juan is dead. 'Juan is dead.'
The clitic is assumed to be inserted in the Specifier of a low applicative phrase. This phrase is headed by an applicative head that has the semantics specified in (12). What this head does is to incorporate the subject of the sentence (that binds the clitic "se", which is an anaphor) into a possession relation with the conflated noun that can denote a state, as represented in (13). As a result, the subject enters in a possession relation of the state denoted by the conflated noun, and hence, the predicate denotes a process plus a state, as represented in (14).
(12) Semantics of low applicatives: (Pylkkanen 2008) [lambda]X[lambda]y[lambda]f<e<st>> [lambda](e,x) & theme (e,x) & TO-THE-POSSESSION (x,y)
(13) Syntax of low applicatives: (based on Pylkkanen 2008)
(14) a. Juan cayo. Juan fell. [there exists]e(undergo) & undergoer(e, Juan) & f(e,x) & theme(e, [caida.sub.EVENT])
b. Juan se cayo. Juan se fell. [there exists]e(undergo) & undergoer(e, Juan) & f(e,x) & theme(e, [caida.sub.EVENT]) & TO THE-POSSESSION ([caida.sub.STATE], se) & se=Juan
2.2.2. Non-alternating "se" in accusative contexts
We argue that in sentences where non-alternating "se" occurs with transitive verbs, the clitic "se" is in the Specifier of a low applicative head too. This applicative head has the same semantics as the one introduced in section 2.2.1: it establishes an inalienable possession between the subject, which binds the clitic, and the accusative object. The derivational tree is represented in (15), and the semantic interpretation of such predicates is given in (16).
(15) Syntax of transitive verbs with non-alternating clitic "se": Juan se lava las manos. Juan CL washes the hands "Juan washes his hands."
(16) Juan se lava las manos Juan CL washes the hands "Juan washes his hands." [there exists]e(wash) & agent(e, Juan) & f(e,x) & theme(e, manos) & IN-THE-POSSESSION (manos, se) & se=Juan
Data from Czech show that this kind of "se" is structurally different from the non-alternating "se" of anticausative inchoatives as in (1) and reflexives as in (2): Czech marks its equivalent as dative ("si" rather than "se") in similar structures, as can be seen in (17), whereas anticausative inchoatives, as in (19), and reflexives, as in (18), are unmarked.
(17) Jan *se/si myje zuby kaPdy den. Jan *[CL.sub.ACC]/[CL.sub.DAT] washes [teeth.sub.ACC] every day. "Jan washes his teeth every day."
(18) Jan se/*si myje kaPdy den. Jan CLacc/*CLdat washes every day "Jan washes himself everyday."
(19) Dvefe se/ *si najednou zavfely Door [CL.sub.ACC]/*[CL.sub.DAT] suddenly closed "The door suddenly closed."
3. TOWARDS A REVISED VERSION OF THE WEAK CONTINUITY HYPOTHESIS OF ACQUISITION OF FUNCTIONAL CATEGORIES
We pursue a revised view of the Weak Continuity Hypothesis of Acquisition of Functional Categories (Clahsen 1990; Clahsen et al. 1993; Meisel and Muller 1992). The idea that underlies this approach to L1 acquisition claims that, given the set of functional heads available in UG, not all of them are active by default in adult grammar. In other words, functional heads like [T.sup.0] or [C.sup.0] are active by default, unlike other functional heads such as low and high applicative heads, which in fact may be activated or not in different languages. In Spanish, non-anticausative intransitive verbs with alternating "se" (5) and transitive verbs with non-alternating "se" (3) have in common the presence of a low applicative phrase. Therefore we may conclude that the latter functional heads are the ones that require extra time to be projected in Child Spanish.
According to this revised view of the Weak Continuity Hypothesis that considers applicative heads as non-active by default heads, both kinds of verbs (transitive with non-optional "se" and non-anticausative intransitives with optional "se") fall within a natural class, and we put forward the following hypothesis:
(20) [H.sub.0]: Both transitive verbs with non-alternating "se" (3) and non-anticausative intransitive verbs (5) involve the presence of a functional head that is not active (though available) by default in UG: a low applicative head ([LAppl.sup.0]).
Provided that [LAppl.sub.0] is a functional head that is not active by default in UG, and according to the above mentioned revised view of the Weak Continuity Hypothesis of Acquisition of Functional Heads:
a) [LAppl.sup.0] requires linguistic evidence in order to become active.
b) [LAppl.sup.0] is later acquired than functional heads that are active by default (e.g. [T.sup.0], [C.sup.0]).
Neither anticausative-inchoative verbs with non-alternating "se" in (1) nor reflexive verbs with non-alternating "se" in (2) require [LAppl.sup.0] or any other functional head (see section 2.1), so they are earlier acquired.
To conclude, our hypothesis (20) predicts that both transitive predicates with non-alternating "se" and non-anticausative intransitive predicates with alternating "se" will appear in child grammar:
a) At the same age.
b) And later than non-alternating "se" in anticausative inchoative (1) and reflexive (2) verbs, because in these cases the clitic is an expletive nominal item that is inserted to satisfy certain formal features. As an expletive, it does not require the presence of any extra functional head nor does it project any further, so no more complex structure is required (Teomiro 2010, 2011).
In section 5, we provide data from CHILDES corpora that support the hypothesis (20) by confirming the derived predictions.
A series of corpora of a child (Juan), taken from CHILDES database (MacWhinney 2000) and ranging from 1 year and 9 months to 4 years and 7 months, have been analysed along with corpora from other children to provide extra evidence. The corpora from a child called Juan are transcribed conversations with his father and/or his mother. In some conversations, his younger brother (Jaime) is also present and interacts with him (cf. Linaza, Sebastian and Del Barrio 1981).
We have proceeded to search for different instances of "se": instances in which the child correctly says "se" (i.e. he has an adult-like behaviour, which is indicated by the sign [+]), and others in which the child should say "se" but he incorrectly omits it (i.e. he has a non-adulXt-like behaviour, which is indicated by the sign [-]). Furthermore, we have distinguished various kinds of "se":
a) [ANTC] Anticausative "se": verbs like "romperse" (break), "cerrarse" (close), "abrirse" (open), "derretirse" (melt).
b) [REFL] Reflexive "se": like "lavarse" (wash), "peinarse" (comb one's hair), "levantarse" (get up).
c) [ANC] Alternating Non Causative "se": non-anticausative intransitive verbs like "caerse" (fall), "morirse" (die), and "marcharse a" (leave for). In the readings we are testing, when the clitic is omitted, the result is ungrammatical, i.e. it is not adult-like. These cases are marked with the symbol [-]. In contrast, the cases where the clitic is omitted like adult-like are marked with the symbol [+].
d) [ACC] "Se" in accusative contexts like examples in (6). This kind of "se" has not been attested in the corpora analysed here. Therefore, data from other corpora are provided.
A summary of the data found in the CHILDES database of Juan is shown in Table 1 below.
The data in Table 1 indicate that this child produces anticausative and reflexive "se" in an appropriate way since the age of 2 years, whereas he omits "se" in non-anticausative intransitive verbs until the age of 2 years and 5 months, when he starts producing it.
Not until the age of 3 years and 6 months does the child produce it more often. It is from the age of 3 years and 5 months onwards that Juan already makes use of the dative clitic "me" ([me.sub.dat]), so he is already expected to cope with low applicative phrases, as can be seen in the examples in (21)-(23), taken from CHILDES database. This prediction is borne out by the data. At the age of 4 years and 7 months the dative clitics seem to be consolidated given the examples in (24), and all kinds of "se" are expected to be produced without any difficulty by the child.
(21) me la ha dejado a mi Maria y Paula. (Juan, 3:05a)
CL1sg.dat CL3SG.ACC has lent to me Maria and Paula.
(22) Dame dame una pega. (Juan 3:06a)
give-CL1sG.DAT give-CL1sG.DAT one sticker.
(23) te voy te te voy a ganar. (Juan, 3:06d)
CL2SG.DAT go CL1SG.DAT CL1SG.DAT go to win.
(24) a. me lo dejas. (Juan, 4:07)
CL1sg.dat CL3SG.ACC lend.
b. me lo das. (Juan, 4:07)
CL1sg.dat CL3sg.acc give.
Instances of non-alternating "se" with transitive verbs have not been found in Juan's corpus, which does not necessarily mean that it is not available yet. However, data from other children's corpora (Escobar Alvarez and Torrens 2010) show that non-alternating "se" in transitive contexts is later produced than "se" with anticausative-inchoative and reflexive verbs, as can be seen in examples in (5) (repeated below) and (22):
(5) a. cayo y rompio la quisma. (Irene, 1;11)
[fell.sub.3SG] and [broke.sub.3SG] the head.
"He fell and broke his head."
b. ha manchado la rodilla. (Magin, 1;11)
has stained the knee.
"He has stained the knee."
(cf. CHILDES, Spanish from Escobar & Vicens 2010)
(25) (es)taba Juanito y Aba y encontraron una mariposa. (Irene, 2;0)
was Juanito and Aba and found a butterfly.
"Juanito and Aba were there and found a butterfly."
(cf. CHILDES, Spanish from Escobar & Vicens 2010)
In this paper we have argued in favour of two instances of clitic "se" available at different acquisition stages. In particular, the clitic "se" that occurs with anticausative inchoative and reflexive verbs is assumed to be earlier acquired than the clitic "se" that occurs with non-anticausative intransitive verbs along with other transitive verbs. We have claimed that the latter clitic "se" is an instance of dative "se" located in the Specifier of a low applicative phrase. Under a revised view of the Weak Hypothesis of Acquisition of Functional Categories, we have argued that children require more time to acquire this applicative head including such an extra functional projection, which is, in contrast, absent in the structure deriving anticausative and reflexive predicates. In these other cases, clitic "se" is an instance of an expletive nominal element inserted during the derivation. It does not require the presence of any extra functional head nor does it project any further. Therefore, the predicates that require this expletive "se", namely anticausative-inchoative and reflexive verbs, are earlier acquired than the predicates that require the argumental "se" that is licensed by the low applicative head, namely transitive verbs with non-alternating "se" and non-anticausative intransitive verbs.
Finally, further experimental research is recommended in order to see if these findings are replicated with larger corpora on the one hand, and in experiments with elicitation tasks on the other hand. It would be also recommendable to delve deeper into the differences of the kinds of "se" explored in this work and other types of predicates like consumption verbs. Eventually, L2 acquisition of Spanish different types of "se" should also be studied since different languages may or may not have the same applicative head active in their grammar. The study of the L2 acquisition of dative "se" from one language with low applicatives in comparison with another without them should shed more light on the nature and acquisition of the clitic "se" in Spanish.
Baauw, S. 2000. Grammatical Features and the Acquisition of Reference. A Comparative Study of Dutch and Spanish. Utrecht: LOT.
Carlson, G. 1977. "A unified analysis of the English bare plural". Linguistics and Philosophy 1 (3): 413-458.
Clahsen, H. 1990. "Constrain on parameter setting: A grammatical analysis of some acquisition stages in German child language". Language Acquisition 1: 361-391.
Clahsen, H., M. Penke, M and T. Parodi. 1993. "Functional categories in early child German". Language Acquisition 3: 395-429.
De Miguel, E. and M. Fernandez Lagunilla. 2000. "El operador aspectual 'se'". Revista Espanola de Linguistica 30 (1): 13-44.
Dominguez, L. 2003. "Interpreting reference in the early acquisition of Spanish clitics". Linguistic Theory and Language Development in Hispanic Languages. Eds. S. Montrul and F. Ordonez. Cascadilla Press: Somerville.
Escobar Alvarez, L. and V. Torrens. 2010. "The effect of reflexive clitics on the aspectual properties of child language". New Directions in Language Acquisition: Romance Languages in the Generative Perspective. Eds. P. Guijarro Fuentes and L. Dominguez. Newcastle upon Tyne: Cambridge Scholars Publishing. 25-49.
Gonzalez Vergara, C. 2006. Las Construcciones No Reflexivas con "Se". Una Propuesta desde la Gramatica del Papel y la Referencia. Doctoral dissertation. Universidad Complutense de Madrid.
Hale, K. and J. Keyser. 2002. Prolegomenon to a Theory of Argument structure. Cambridge, MA: MIT Press.
Hogdson, M. 2002. "The acquisition of Spanish reflexive "se" in transitive verb argument structures and singularity". Proceedings from 26 BUCLD. Somerville, MA.
Hogdson, M. 2005. "Children's production and comprehension of Spanish grammatical aspect". Theoretical and Experimental Approaches to Romance Linguistics. Eds. R. S. Gess and E. J. Rubin. Amsterdam: John Benjamins. 125-143.
Hogdson, M. 2006. Telicity and the Syntax-Semantics of the Object and Subject. Doctoral dissertation. University of Massachussets, Amherst.
Linaza, J., M. E. Sebastian and C. Del Barrio. 1981. "Lenguaje, comunicacion y comprension. La adquisicion del lenguaje". Monografia de Infancia y Aprendizaje. 195-198.
Lopez Ornat, S. 1990. "La formacion de la oracion simple: Las omisiones de categorias sintacticas (S-V-O) en la adquisicion del espanol". Revista de Estudios de Psicologia 41: 41-92.
MacWhinney, B. 2000. The CHILDES Project: Tools for Analyzing Talk. Mahwah, NJ: Lawrence Erlbaum.
Martin Zorraquino, M. A. 1979. Las Construcciones Pronominales en Espanol. Madrid: Gredos.
Meisel, J. and N. Muller. 1992. "Verb placement and the position of finiteness in early child grammar". The Acquisition of Verb Placement: Functional Categories and V2 Phenomena in Language Acquisition. Ed. J. Meisel. Dordrecht: Kluwer. 109-138.
Mendikoetxea, A. 1992. On the Nature of Agreement: the Syntax of ARB SE in Spanish. Doctoral dissertation. York University.
Mendikoetxea, A. 1997. "Clitics as AGR, and PRO in finite clauses". Proceedings of the IV Colloquium of Generative Grammar.
Mendikoetxea, A. 1999a. "Construcciones con se. Medias, pasivas e impersonales". Gramatica de la Lengua Espanola. Eds. I. Bosque and V. Demonte. Madrid: Espasa-Calpe.
Mendikoetxea, A. 1999b. "Construcciones inacusativas y pasivas". Gramatica de la Lengua Espanola. Eds. I. Bosque and V. Demonte. Madrid: Espasa-Calpe.
Mendikoetxea, A. 2008. "Clitic impersonal constructions in Romance: Syntactic features and semantic interpretation". Transactions of the Philological Society 106(2): 290-336).
Mendikoetxea, A. 2012. "Passives and se constructions". The Handbook of Hispanic Linguistics. Eds. J. I. Hualde, A. Olarrea and E. O'Rourke. Chichester, UK: Blackwell.
Molina Redondo, J. A. 1997 Usos de Se: Cuestiones Sintacticas y Lexicas. Madrid: Sociedad General Espanola de Libreria.
Montrul, S. 2004. The Acquisition of Spanish: Morphosyntactic Development in Monolingual and Bilingual L1 Acquisition and Adult L2 Acquisition. Amsterdam: John Benjamins.
Otero, C. P. 1986. "Arbitrary subjects in finite clauses". Generative Studies in Spanish Syntax. Eds. I. Bordelois, H. Contreras and K. Zagona. Dordrecht: Foris. 81-109.
Otero, C. P. 1999. "Pronombres reflexivos y reciprocos". Gramatica de la Lengua Espanola. Eds. I. Bosque and V. Demonte. Madrid: Espasa-Calpe. 1428-1517.
Pylkkanen, L. 2008. Introducing Arguments. Cambridge: MIT Press.
Raposo, E. and J. Uriagereka. 1996. "Indefinite SE". Natural Language and Linguistic Theory 14: 749-810.
Rigau, G. 1994. "Les propietats dels verbs pronominals". Els Marges 50: 29-39.
Romero, C. and I. Teomiro. 2012. "La relacion entre estructura eventiva y papeles tematicos: el se aspectual del espanol". Revista de Filologia Romanica 29 (2): 233-243.
Sanz, M. 1995. Telic Clitics in Spanish. Unpublished manuscript, University of Rochester.
Teomiro, I. 2010. Anaphors at the Interfaces. A Comparative Study of the Variation of the Anaphoric Systems of English, Dutch and Spanish. Doctoral dissertation, Universidad Autonoma de Madrid.
Teomiro, I. 2011. "Reflexivity and adjustment strategies at the interfaces". Tromso Nordlyd Working Papers in Linguistics 37: 119-149.
Teomiro, I. 2013. "Low applicatives and optional "se" in Spanish non-anticausative intransitive verbs". Revista de Linguistica y Lenguas Aplicadas 8: 248-270.
Teomiro, I. and C. Romero. 2012. "Theta meets aspect: The Spanish aspectual "se" with consumption verbs". Topics in Linguistics 10: 20-27.
ISMAEL IVAN TEOMIRO GARCIA *
MARIA ANGELES ESCOBAR ALVAREZ
* Correspondence to: Ismael Ivan Teomiro Garcia. Universidad Nacional de Educacion a Distancia. Departamento de Filologias Extranjeras y sus Linguisticas. Paseo Senda del Rey, 7. 28040 Madrid. Spain. E-mail: email@example.com
(1.) This work has been partially funded by research projects FFI2011-23829/FILO and FFI2011-29798-C02-01 (Spanish Ministry of Economy and Competitiveness), and 2013-014-UNED-PROY (Spanish National University of Distance Education). We want to thank Maria Beatriz Perez Cabello de Alba for her useful comments and suggestions.
(2.) As well as its variations for the different grammatical persons available in Spanish: "me" for 1SG, "te" for 2SG, "nos" for 1PL, and "os" for 2PL. Hereinafter, we will use "se" to refer to all the morphological variations of this clitic for expository reasons.
(3.) The CHILDES project (MacWhinney 2000) presents a set of computational tools designed to help record, analyse, and share naturalistic samples of speech. Currently, it provides the largest collection of naturalistic corpora, covering a broad range of phonological, morphological, lexical, syntactic, semantic and pragmatic phenomena. It is freely available at http://childes.psy.cmu.edu/.
(4.) Consumption verbs are left aside for lack of space. The prediction is that they will be also acquired later than anticausative inchoatives and reflexives because they involve a low applicative phrase like non-anticausative intransitives (see section 2.2.1). However, as stated by Teomiro and Romero (2012), these verbs denote a complex predicate wherein a causal relation holds between an eventuality and a state. Therefore, an extra layer like a voiceP or a vP is needed to account for causativity.
(5.) A stage-level predicate is true of a temporal stage of its subject (Carlson 1977). For example, "Paul is thirsty" is true of a certain period of time until he drinks something.
(6.) An individual-level predicate is true through the existence of the individual that is its subject (Carlson 1977). For example, "Mary is kind" refers to a property that the subject has, regardless of which particular point in time we consider.
Table 1. Juan Linanza (data from CHILDES). AGE (year:month) [ANTC] [REFL] [ANC] 2:00 [empty set] [+] 1 [-] 3 2:01 [empty set] [??] [-] 1 2:03 [+] 1 [empty set] [-] 2 2:04 [empty set] [empty set] [-] 12 2:05a [empty set] [empty set] [+] 1 2:08a [empty set] [+] 1 [??] 2:08b [+] 3 0 [empty set] 3:05a [empty set] [+] 1 [??] 3:06c [empty set] [+] 2 [??] 3:06d [empty set] [empty set] [+] 3 3:09 [+] 2 [+] 2 [+] 6 4:24 [empty set] [+] 2 [??]
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||articulo en ingles|
|Author:||Teomiro Garcia, Ismael Ivan; Angeles Escobar Alvarez, Maria|
|Publication:||RaeL Revista Electronica de Linguistica Aplicada|
|Date:||Jan 1, 2013|
|Previous Article:||Analisis de la relacion entre el modelo formativo de aprendizaje del ingles--AICLE y Ensenanza Formal--y la ansiedad.|
|Next Article:||An empirical investigation of lexical and cognitive networks in the brain.|