Syntactic Reflexes of Emerging Optionality in Spanish as Heritage Language: The Case of Dative Experiencer Verbs

6\QWDFWLF5HIOH[HVRI(PHUJLQJ2SWLRQDOLW\LQ6SDQLVK DVD+HULWDJH/DQJXDJH7KH&DVHRI'DWLYHH[SHULHQFHU 9HUEV 'LHJR3DVFXDO&DER Hispania, Volume 99, Number 1, March 2016, pp. 34-50 (Article) 3XEOLVKHGE\-RKQV+RSNLQV8QLYHUVLW\3UHVV DOI: 10.1353/hpn.2016.0017 For additional information about this article http://muse.jhu.edu/journals/hpn/summary/v099/99.1.pascual.html Access provided by The American Association of Teachers of Spanish and Portuguese (25 Mar 2016 15:34 GMT) Syntactic Relexes of Emerging Optionality in Spanish as a Heritage Language: he Case of Dative-experiencer Verbs Diego Pascual y Cabo Texas Tech University Abstract: his study contributes to current trends of heritage speaker bilingualism research by examining the syntax of so-called Spanish dative-experiencer predicates (gustar-like verbs). Building on previous indings (e.g., Silva-Corvalán 1994; Toribio and Nye 2006), it is hypothesized that Spanish heritage speakers can project an optional agentive syntax (a use deemed ungrammatical by monolingual speakers) while still having access to the obligatory dative-experiencer syntax. It is argued that the availability of this emergent optionality, coupled with inluence from English, the dominant language, conspire to promote the non target-like forms documented in previous research (e.g., Toribio and Nye 2006; de Prada Pérez and Pascual y Cabo 2011). he ensuing predictions are tested with a grammaticality judgment task that examines the informants’ knowledge and use of gustar-like verbs in passive constructions precisely because passivization of stative predicates should be precluded from a grammar that does not allow for an agentive alternation. Keywords: acquisition/adquisición, bilingualism/bilingüismo, psychological verbs/verbos psicológicos, Spanish as a heritage language/español como lengua de herencia, Spanish in the United States/español en los Estados Unidos 1. Introduction B ilingual contexts oten create conditions that give rise to linguistic outcomes that difer from those expected of monolingual speakers of the same language(s) (usually the minority language, but perhaps in both). hese distinctive outcomes, which are generally argued to obtain as a result of some sort of incomplete acquisition (e.g., Montrul 2008; Silva-Corvalán 1994, 2014), are characteristic of the grammars of heritage speakers (HS). Briely deined, a HS is a bilingual individual who was exposed naturalistically to a language at home that is diferent from the greater societal majority language (e.g., Beaudrie and Fairclough 2012; Bennamoun, Montrul, and Polinsky 2013; Montrul 2008, 2009, 2011, 2012; Rothman 2007, 2009; Valdés 2000). he heritage language (HL) is most oten the sole irst language (L1) (or a simultaneous L1 along with the societal language), but with time the societal majority language usually becomes the HS’s dominant linguistic system, relegating the HL to more restricted domains of use. Shits in dominance and in language use preferences typically take place in early childhood, roughly around the age the child is introduced to schooling, and before ultimate attainment in all domains of the HL grammar occurs. As noted by the growing body of HL acquisition studies, HS bilingual competence generally difers from that of their monolingual counterparts’ across properties and across languages in ways not always predictable and/or well understood (e.g., Benmamoun, Montrul, and Polinsky 2013; Laleko 2010; Mikhaylova 2012; Montrul 2008; Pascual y Cabo and Rothman 2012; Pires and Rothman 2009; Polinsky 2006, 2008, 2011; Rothman 2007, 2009; Sok-Ju Kim 2012). his observation also applies to Spanish as a heritage language, as reported in studies examining a variety of properties and domains such as diferential object marking (e.g., Bowles and Montrul, 2009), AATSP Copyright © 2016 Hispania 99.1 (2016): 34–50 Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 35 copula choice (e.g., Silva-Corvalán 1994; Valenzuela, Barski, Diez, Faure, Ramírez, and Pangtay 2012; Valezuela, Iverson, Rothman, Borg, Pascual y Cabo, and Pinto 2015), tense and aspect (e.g., Miller and Cuza 2013; Montrul, 2002, 2009), subject and object expression (e.g., Montrul 2004; Silva-Corvalán 1994), gender agreement (e.g., Montrul, Foote, and Perpiñan 2008), mood (e.g., Montrul 2009; Lynch 1999; Pascual y Cabo, Lingwall, and Rothman 2012), pragmatics/discourse (e.g., Pinto 2012; Pinto and Raschio 2007) and its interface with syntax (e.g., Cuza and Frank, 2011; De Prada Pérez, and Pascual y Cabo 2012; Pascual y Cabo, Lingwall, and Rothman 2012;), or phonetics and phonology (e.g., Henriksen 2015; Rao 2014; Ronquest 2012). he present study contributes to this line of research by examining the syntax of Spanish dative experiencer verbs, also known as reverse psychological predicates (i.e., gustar-like), a property that has been documented to be particularly vulnerable in HS grammars given its structural and semantic opacity (e.g., De Prada Pérez and Pascual y Cabo 2011; Silva-Corvalán 1994; Toribio and Nye 2006). Informed by syntactic theory (e.g., Belletti and Rizzi 1988; Landau 2010; Pesetsky 1995), the hypothesis of the present study is that in heritage-speaker Spanish, this group of verbs has undergone a reanalysis of its argument structure. As a direct consequence of this adjustment, and alongside its obligatory dative-experiencer syntax, Spanish HSs can use and interpret gustar-like verbs with an (optional) agentive syntax, a use deemed ungrammatical by monolingual speakers due to its unaccusative nature. he predictions that fall out from this hypothesis are tested with a grammaticality judgment task that examines gustar-like verbs in passive constructions precisely because passivization of unaccusative predicates should be precluded from a grammar that does not allow for an agentive alternation. A total of 49 HSs and 16 Spanish monolingual speakers completed this task. he analysis of the data obtained reveal that, although HSs show full-ledged knowledge of the semantic and syntactic restrictions of the passive voice, they tend to favor the proposed innovation, a use that is categorically rejected by the control group. Given this background, in the next section, I present the syntax of reverse psychological predicates. 2. he Syntax of Psychological Verbs Psych-predicates are those whose lexical-semantics denote a mental or emotional state. hese verbs subcategorize for two internal roles: an experiencer (i.e., the entity that receives or contains the mental/emotional state) and a theme (i.e., the entity that triggers the mental/ emotional state). Belletti and Rizzi (1988) proposed a three-class distinction of psych-predicates for Italian, also pertinent to Spanish (see Parodi-Lewin 1991), described as follows. Class I psych-verbs are generally treated as transitive verbs with a subject experiencer in nominative case and a theme in accusative case (as in 1). Some Spanish verbs that belong to this class are temer ‘to fear’ or odiar ‘to hate.’ (1) Teo odia la lechuga Teo.NOM hate.3.SG.PRES the lettuce.ACC ‘Teo hates lettuce’ Class II psych-verbs generally contain a theme with nominative case that controls verbal agreement and a postverbal accusative experiencer. Some Spanish verbs that belong to this class are molestar ‘to bother’ or asustar ‘to scare.’ Consider example (2): (2) Pau molesta a Jake Pau.NOM bother.3.SG.PRES to Jake ‘Pau bothers Jake’ 36 Hispania 99 March 2016 It has been noted that class II psych-predicates are hybrid in nature since they can be interpreted as either stative or eventive verbs (e.g., Arad 1998; Landau 2010; Pesetsky 1995). In other words, class II psych-verbs such as asustar ‘to frighten’ or molestar ‘to bother’ can be used to describe the outcome of an action in which the structural subject can have either an agentive, as in (3a), or a non-agentive role, as in (3b). (3) a. Diana asustó a Whitman (intentionally/unintentionally) Diana.NOM scare. 3SG. PRES. to. Whitman.ACC. ‘Diana scared Whitman’ b. A Whitman le asustan las tormentas To Whitman.DAT him (DAT.CLI) scare. 3PL. PRES. storms ‘Whitman is scared of storms’ Note that although (3a) can have two interpretations, they are undistinguishable morphologically. In its agentive reading, (3a) corresponds to ‘Diana intentionally caused Whitman (to) fear’ and, as can be seen in (4) below, can be passivized, showing that indeed it is clearly not stative with this speciic meaning. (4) Whitman es asustado por Diana ‘Whitman is scared by Diana’ On the other hand, if a class II psych-predicate is taken to have a stative interpretation, then it is treated as an unaccusative verb1 and, obviously, it could not be used agentively. When that is the case, stative-unaccusative class II psych-predicates align syntactically with class III psych-verbs. Class III psych-predicates subcategorize for a postverbal theme that controls verb agreement and a preverbal dative experiencer that is 1) obligatorily doubled by a clitic and 2) preceded by the dative marker a (as in examples 5a and 5b) when the experiencer is spelled out. his reversed, yet preferred word order (e.g., Gutiérrez-Bravo 2007), however, can vary for discourse-pragmatic reasons (Franco and Huidobro 2003; Gómez Soler 2012) (as in examples 5b and 5d). (5) a. A Pau le gusta el pan. b. El pan le gusta a Pau. To Pau him (DAT.CLI) like the bread. he bread him (DAT.CLI) like to Pau. ‘Pau likes bread’ ‘Pau likes bread’ c. Le gusta el pan. Him (DAT.CLI) likes the bread. ‘He likes bread’ d. El pan le gusta. he bread him (DAT.CLI) likes. ‘He likes bread’ Unlike class II psych-verbs, which can alternate between having both agentive and stative syntactic representations, class III psych-verbs only have an unaccusative syntax available. In other words, they cannot be used agentively, do not project vP, and are incompatible with passive constructions (6a and 6b). his diference will prove crucial to the syntactic proposal I will make for the representation of gustar-like verbs in heritage speaker Spanish. (6) a. *La película fue gustada por Marta ‘he movie was liked by Marta’ b. Laurie fue asustada por Doug ‘Laurie was frightened by Doug’ Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 37 Unlike Spanish, English does not have dative experiencers that could be mistaken for subjects (White, Brown, Bruhn-Garavito, Chen, Hirakawa, and Montrul 1999: 173) and its verbal argument structure cannot be reversed. hat is, the theta-roles are always mapped onto canonical Subject-Verb-Object word order and verbal agreement is necessarily controlled by the preverbal argument. As can be seen in (7), any deviation to the SVO order results in an ungrammatical sentence. (7) a. Nico likes dogs c. hey like chocolate b. *Nico like dogs d. *hey likes chocolate Additionally, as can be seen in (8a–b), English does allow passive constructions with the verb ‘like,’ the equivalent version of Spanish ‘gustar.’ (8) a. Everyone likes cupcakes b. Cupcakes are liked by everyone Given this theoretical background, in the next section, I present the most important indings related to the acquisition of class III psych-verbs among Spanish heritage speakers. 3. Acquisitional Studies Previous work examining knowledge and use of gustar-like predicates among Spanish HSs has convincingly shown that this is a particularly challenging property to acquire. he general consensus is that, although Spanish HSs can and do have access to target-like forms (Dvorak and Kirschner 1982; Pascual y Cabo 2013; Toribio and Nye 2006), they have a strong tendency to produce (Dvorak and Kirschner 1982; Silva-Corvalán 1994) and accept (de Prada Perez and Pascual y Cabo 2011; Toribio and Nye 2006) what a priori can be considered target-divergent forms.2 For example, Dvorak and Kirschner (1982) tested Puerto Rican HSs in New York City via an English-to-Spanish translation task. heir data revealed patterns of omission of the dative marker a in obligatory environments (though the clitic le remained unaltered) as well as a strong preference towards preverbal experiencer control verbal agreement. Toribio and Nye (2006) examined production and comprehension of psych-verb constructions via an elicited written production task and a scalar grammaticality judgment task. In addition to inding strong evidence in support of Dvorak and Kirschner’s (1982) so-called invariable le, their results revealed 1) indeterminacy with respect to the mapping of the arguments to syntactic positions;3 2) omission of the dative a-marker; and 3) that animate DPs are possibly taken to be the structural subject irrespective of actual grammatical function. In a related study, Montrul and Bowles (2009) tested knowledge and use of the Diferential Object Marker a (DOM) among US Spanish HSs and reported a generalized lack of instantiation of such inherent case in the HSs’ grammars (trends also observed more recently by Montrul and Sánchez Walker 2013). his inding led them to survey other instances of inherent dative case, which indicated that problems with DOM seem to extend to dative experiencers in class III psych-verbs. More recently, De Prada Pérez and Pascual y Cabo (2011) used a scalar grammaticality judgment task to examine basic subject-verb agreement and clitic agreement with two high frequency class III psych-predicates (gustar ‘to like’ and encantar ‘to love’). heir indings revealed four important tendencies. First, HS participants demonstrated robust knowledge of clitic agreement (contrary to Dvorak and Kirshchner 1982; Toribio and Nye 2006). Second, HSs revealed a strong preference for the use of the third person singular verb form regardless of actual subject-verb agreement. hird, no diferences were found between the two class III 38 Hispania 99 March 2016 psych-verbs tested. Fourth, and perhaps most interestingly, these preferences were comparable to the group of Spanish native speakers tested. In general, the previous studies reviewed here indicate that HSs tend to use strategies towards a more transparent mapping of class III psych-verbs. hese results are, however, of no surprise since class III psych-verbs constitute an important learnability problem for the learner since it is necessary to acquire the associated non-canonical mapping of thematic roles to syntactic positions. To provide an adequate explanatory analysis of the nature of the HS diferences observed in this particular domain, the present study further examines the argument structure and semantic mappings of class III psych-predicates among Spanish heritage speakers in the United States. Since such diferences are well documented already, I seek to go beyond further documentation of the diferences alone, by ofering from the outset a hypothesis based on the syntax of class II/III psych-verbs (e.g., Belletti and Rizzi, 1988; Landau, 2010; Pesetsky, 1995). I predict that in HS Spanish, class III type psych-verbs (e.g., gustar ‘to like’), those that only have a stative reading available in monolingual grammars, are being reanalyzed as class II type psych-verbs (e.g., asustar ‘to frighten’), which have both a stative and an agentive syntax available. his syntactic readjustment predicts that Spanish HSs should (variably) accept class III type verbs (e.g., gustar ‘to like’) used in all contexts and forms that are available for class II psych-verbs (e.g., asustar ‘to frighten’), even if these uses are ungrammatical in monolingual grammars. For example, such modiication would allow Spanish HSs to produce gustar taking a nominative experiencer as the structural subject (e.g., *ella gusta el queso ‘she likes cheese’), a inding that has consistently been reported in previous studies (De Prada Pérez and Pascual y Cabo 2011; Silva Corvalán 1994; Toribio and Nye 2006). Furthermore, if I am on the right track, one would expect to also ind HSs’ acceptance of gustar-like verbs used in the passive voice since this should be a valid option in their grammars as a byproduct of the new/optional agentive interpretation. Recall from section 2 that this sort of construction is not allowed in monolingual grammars because class III psych-verbs (e.g., gustar ‘to like’) do not have an agentive syntactic structure available and therefore cannot be passivized. Class II type verbs, however, do have an agentive syntactic structure available (even in monolingual environments) which allows them to be passivized. Before I provide details of the experimental design in section 5, I next present the research question that guides the present study. To conclude, section 6 includes a detailed discussion of the data obtained. 4. Research Questions and Hypotheses he research question at the core of the present study is the following: are class III psychverbs (e.g., gustar ‘to like’) undergoing a reanalysis of their argument structure whereby they are adopting the hybrid nature of class II psych-verbs (e.g., asustar ‘to frighten’)? Building on previous indings (Dvorak and Kirshchner 1982; Toribio and Nye 2006), it is hypothesized that Spanish HSs will show diferences from monolingual norms for class III psych-verbs and the properties associated with them. To test this hypothesis, I designed a grammaticality judgment task that examined the abovementioned argument structure innovation. he speciics of the experimental methodology employed herein as well as the informants that participated in this study are detailed in the following sections. 5. Methodology 5.1 Participants Participants’ information was collected via a linguistic background questionnaire that, among other things, included questions regarding language use, exposure to the heritage Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 39 language, language instruction, and perceived language proiciency. A total of 65 informants, 16 Spanish monolingual speakers and 49 adult HSs, participated in this study. Ranging in age from 18 to 24 years old (average 20;1), all adult HSs were either US-born or had arrived in the United States before the age of two. heir average age of irst exposure to English was 0;2 for the HS intermediate group and 0;7 for the HS advanced group. To be included in this study, all HS informants were required to be of Cuban origin and to speak Spanish on a daily basis (although it need not be their dominant language). his group was divided into advanced (n 5 21), intermediate (n 5 24), and low (n 5 4) proiciency subgroups according to their responses to an adapted version of the DELE, a commonly used Spanish proiciency test (e.g., Cuza and Frank 2011; Montrul 2000, 2002, 2009; White, Valenzuela, KozlowskaMacGregor, and Leung 2004). his test consists of two main sections. he irst one includes a 30-item multiple-choice section that targets lexical proiciency and grammatical competency. In the second section, informants select the most appropriate answers to a contextualized cloze test (n 5 20). Given the small number of low proiciency informants based on the DELE (n 5 4), the results from this HS group are not included herein. All monolingual speakers reported having been born and raised in Cuba by Cuban parents and having very limited knowledge of a foreign language.4 In terms of education, 9 reported having completed secondary education, and 7 reported having completed (or were in the process of completing) post-secondary education. A summary of informants’ background data is provided in Table 1 below. Table 1. Background information by group Group N (Male/ Female) Age (years) AoA English (years) Self-rating English (Max. = 10) Self-rating Spanish (Max. = 10) DELE score (Max. = 50) Advanced HSs 21 (6/15) 20;1 (1;6) 0;7 (0;69) 9.2 (.78) 6.1 (1.38) 41.5 (1.64) Intermediate HSs 24 (8/16) 20;3 (1;6) 0;2 (0;55) 9.6 (.56) 4.9 (1.66) 32.6 (1.66) Monolingual group 16 (4/12) 37;1 (19;2) — 1.87 (1.78) 9.81 (.54) 46.9 (2.04) 5.2 Scalar Grammaticality Judgment Task All participants were trained on and then asked to use a 1 (completely unnatural) to 4 (completely natural) Likert scale to rate a total of 40 sentences (available as an appendix at the end of this paper). Additionally, they also had the option to choose 0 if they were not completely sure, did not know the answer, or did not hear the sentence properly.5 To ensure consistency across testing sessions (e.g., controlling prosody), as well as to avoid any potential literacy efects (diferentially applicable for HSs), all subjects were asked to provide their judgments as they heard the tokens from a recorded video of one male speaker born and raised in Cuba who, at the time of the recording, was 32 years old. he video lasted 8 minutes 16 seconds; the irst minute provided instructions in Spanish and 3 practice items. he remaining 7 minutes included a total of 40 sentences. Between one sentence and the next, there was an 8 second pause. Prior to the beginning of the test, all informants received suicient training and feedback. Of the 40 sentences included in this experiment, half were baseline items and the other half served as critical items. All items were randomized and counterbalanced. he 20 baseline items were divided into three diferent groups: 10 items tested acceptance of grammatical sentences 40 Hispania 99 March 2016 (as in 9); 5 items tested ungrammatical sentences due to noun-adjective agreement (as in 10); and 5 items tested ungrammatical sentences due to subject-verb agreement (as in 11). (9) Nosotros vivimos en un apartamento ‘We live in an apartment’ (10) *El carro blanca fue caro ‘he white.FEM. car. MASC. was expensive’ (11) *Ella hago la tarea todos los días ‘She does.1st SG. Homework every day’ In light of the hypothesis put forth in section 4, the critical items tested the informants’ acceptance/rejection with respect to four types of passive structures: passive sentences with transitive verbs (as in 12); passive sentences with unaccusative verbs (as in 13); passive sentences with class III psych-verbs (as in 14); and passive sentences with class III psych-verbs (as in 15). (12) Transitive verb sample token Esa carta fue escrita por Carol ‘hat letter was written by Carol’ (13) Unaccusative verb sample token *Hugo fue llegado a la casa ‘Hugo was arrived home’ (14) Class II psych-verb simple token Pau fue asustado por Jorge ‘Pau was scared by Jorge’ (15) Class III psych-verb sample token (*)La película fue gustada por Fernando ‘he movie was liked by Fernando’ In order to conirm the hypothesis, HSs’ judgments should converge with those of the control groups in the case of (12) and (14), since agentive verbs can appear in both active and passive voice constructions, and (13), since both Spanish and English do not allow passivization with unaccusitives. Conversely, I expected HSs to variably accept gustar-like verbs in passive constructions (15) as a relex of the new available agentive interpretation. 5.3 Results Regarding the baseline items, which I present irst to show that all participants were able to handle the task itself, a three (type: grammatical, *verb agreement, *adjective agreement) by three (group: advanced HSs, intermediate HSs, control group) repeated measures ANOVA revealed a main efect for type (F (2, 96) 5 716.47, p , .001), group (F (3, 46) 5 4.277, p , .001) and a type-by-group interaction (F (6, 96) 5 9.521, p , .001). he signiicant efect for type was expected because grammatical tokens should yield signiicantly higher ratings than the ungrammatical ones. Figure 1 displays the average ratings for each of the baseline conditions for each group which from let to right are: 1) *noun-adjective agreement; (2) *subject-verb agreement; and 3) grammatical. Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 41 Figure 1. Group means of baseline conditions by group As can be seen, the signiicant diference discussed above stems from the contrast that exists between an almost categorical acceptance of grammatical sentences and the tendency to reject ungrammatical ones. Results from Bonferroni pairwise comparisons revealed that all three baseline conditions (including the two ungrammatical ones) difer at a level of statistical signiicance: *adj-agree vs. *verb-agree, p , .001; *adj-agree vs. grammatical, p , .001; *verbagree vs. grammatical, p , .001. Although such a general trend is also visible among intermediate HSs, this group stands out from the other two in that the ungrammatical sentences receive higher ratings (2.19 in the case of *adj-agree and 1.4 in the case of *verb-agree), while the grammatical sentences are judged lower (3.64). he results from Bonferroni post hoc tests indicate that within this group, these diferences are indeed statistically signiicant (i.e., *adj-agree vs. *verb-agree, p , .001; *adj-agree vs. grammatical, p , .001; *verb-agrere vs. grammatical, p , .001). hese results, I argue, should not be taken as a sign for target-divergent performance for this group, since they do make a signiicant distinction between grammatical and ungrammatical tokens. Instead, they reveal some indeterminacy, which may possibly result from the diferential salient nature of the ungrammatical tokens themselves. hat is, while problems with noun-adjective agreement in Spanish usually correspond to the absence or presence of a low perceptual salient form (e.g., a mere vowel alternation in the case of gender [i.e., most typically ‘o’ for masculine and ‘a’ for feminine]), problems with subject-verb agreement in Spanish tend to be more acoustically salient. Similar morpho-phonological issues pertaining to variable HS performance have been discussed in previous studies examining modality alternations (e.g., Montrul 2009; Pascual y Cabo, Rothman, and Lingwall 2012), gender agreement (e.g., Montrul, Foote, and Perpiñan 2008) and diferential object marking (e.g., Montrul and Bowles 2010) among comparable Spanish HS groups in the United States. Moreover, the testing modality employed in this study would certainly lend support to this explanation since less salient forms tend to be easily concealed in oral speech. 42 Hispania 99 March 2016 Up to this point, I have shown that the informants’ responses to the baseline items were generally on target, though somewhat indeterminate in the case of the intermediate HS group. hat said, all groups were successful at making the necessary distinctions between grammatical and ungrammatical utterances. his distinction is taken as evidence that all groups were able to handle the task itself and that both the experiment and the procedure were valid for the purposes of this study. Next, I proceed to discuss the results for the critical conditions. Figure 2 shows the overall group means and standard deviations for the four critical conditions tested in this experiment, which from let to right are 1) passives with transitive verbs; 2) passives with unaccusative verbs; 3) passives with class III psych-verbs; and 4) passives with class II psych-verbs.6 Recall from section 2 that all of these conditions, except for 2) and 3), are grammatical in the passive voice. Figure 2. Group means of critical conditions by group As expected, all groups clearly accept passive constructions with transitive verbs and class II psych-verbs. his is not surprising given that Spanish and English work the same in this regard. he general tendency is also to reject passives with unaccusative verbs (though not so categorically for the HS intermediate). Interestingly, we observe important diferences across the groups with respect to the acceptance of class III psych-verbs. To test for statistically signiicant diferences, a repeated-measures ANOVA was run with the variables of group (e.g., control group, advanced HSs, intermediate HSs) and verb type (transitive, unaccusative, class II psych-verb, class III psych-verb). he results of this ANOVA showed a main efect for verb type (F (2.572, 149.191) 5 740.303, p , .001), group (F (2, 58) 5 9.015, p , .001), as well as a high order interaction between group and verb type (F (5.145, 149.191) 5 43.193, p , .001). Next, I further examine these results, condition by condition. Individual responses to the condition that tested passives with transitive verbs (see Figure 3) suggest a high degree of homogeneity across all informants and across all four groups; though more categorical judgments were provided by the control participants. As can be seen, all tokens Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 43 included in this condition received a rating of 3 “natural” or 4 “completely natural” by informants from the control group. HS informants follow the same trend. hat said, two advanced HSs (#15 and #20) and six intermediate HSs (#7, #10, #11, #12, #16, #21) provided a rating of 2 “unnatural” to one token (or two in the case of #10). Figure 3. Individual responses for passives with transitives To further establish whether informants were sensitive to the semantic and syntactic restrictions of the passive voice one needs to examine their judgments of passive sentences with unaccusative verbs. As can be seen in Figure 2 above, all groups largely reject this sort of construction. his, again, was the expected outcome given that unaccusative verbs are incompatible with the passive voice. Statistically signiicant diferences were found between the advanced and Intermediate HSs (p 5 .0002). Moreover, the adult monolingual control group difers statistically from the HS-intermediate (p 5 .002) but not from the HS-advanced (p 5 1.0). Crucially, further intragroup comparisons reveal that, in spite of this proiciency efect in adult HSs, all groups without exception make a statistically signiicant diference between passives with unaccusative predicates and passives with transitive predicates (Bonferroni post hoc test: control, p , .001; HS-advanced, p , .001; HS-intermediate, p , .001). Such a distinction is critical for the purposes of the present study as it suggests that all groups are sensitive to the syntactic and semantic restrictions of passive sentences. In an examination of the individual responses to the tokens included in this condition (see Figure 4), it can be observed that all control monolinguals consistently provide judgments that relect their rejection of passive constructions with unaccusative predicates.7 Compared to the control data, HSs reveal a higher degree of uncertainty and variability across and within individuals. hat said, the trend observed is still in line with the ratings provided by the control group as most responses to the grammaticality judgment task fall in the 1–2 range (completely unnatural to unnatural). Figure 4. Individual responses for passives with unaccusatives 44 Hispania 99 March 2016 As was also the case with the condition that examined the informants’ judgments towards passives with transitive verbs, all groups tested herein clearly accept passive constructions with class II psych-predicates. Recall that this is a perfectly grammatical construction for verbs such as molestar ‘to bother’ or asustar ‘to frighten’ given their hybrid nature. Pairwise intergroup comparisons reveal that none of the groups statistically difer from each other for this speciic condition. Individual responses across all four groups (see Figure 5) support this claim since, for the most part, all informants tend to accept the tokens included in this condition. Figure 5. Individual responses for passives with class II psych-verbs Recall from section 2 that gustar is an unaccusative verb and, as such, should be incompatible with the passive voice. hat said, not all groups reveal the same judgments: whereas monolingual adults (M 5 1.10) clearly reject this use, HSs’ intuitions reveal greater variance (HS-advanced M 5 1.86; HS-intermediate M 5 2.22). Upon further investigation, the results from pairwise intergroup comparisons reveal that results are statistically signiicant; this allows for insights into the nature of these HS diferences. For example, these comparisons indicate that whereas the responses from the two HS groups do not difer statistically from each other (p 5 1.0), the responses from the monolingual adults do difer at a level of statistical signiicance from those of both HS groups (HS-intermediate, p , .001; HS-advanced p 5 .037). Congrary to what was observed in the previous conditions, analysis of individual responses for passives with class III psych-verbs (see Figure 6 below) indicates noteworthy diferences between the experimental and the control group. On the one hand, the monolingual controls continued to provide homogeneous and target-like judgments. On the other hand, the two HS groups showed greater data dispersion, with informants responding along the full spectrum of answers. Out of the 45 HSs tested, only four advanced and one intermediate (#17) provided answers that fell within the range obtained from the monolingual controls. In addition to this, it should be noted that the tokens included in this condition elicited a response of 0 “I don’t know/I am unsure” a total of 9 times among the HSs (and none among the controls), something that did not happen in the previous conditions. Figure 6. Individual responses for passives with class III psych-verbs Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 45 To summarize the results, it can be argued that the evidence provided thus far is consistent with the reanalysis of class III psych-verbs in HS grammars whereby gustar-like verbs can adopt an emergent optionality that is absent in monolingual Spanish grammars. 6. Discussion and Conclusion he goal of this study was to examine the current status of gustar-like verbs in HS Spanish and to provide an explanatory analysis of the data obtained. Building on previous indings, it was reasoned that in HS Spanish, this group of verbs may have undergone a reanalysis of their argument structure and may have consequently adopted the hybrid nature (agentivity/stativity) of class II psych-verbs (e.g., asustar ‘to frighten’). To support this hypothesis, one would expect to ind verbs like gustar or encantar used in the same environments and in the same ways that verbs such as asustar ‘to frighten’ or molestar ‘to bother’ are used in monolingual Spanish. his reanalysis, for example, would allow Spanish HSs to produce instances of gustar taking a nominative <experiencer> as the structural subject (e.g., *yo gusto eso “I like that” or *(nosotros) no nos gustábamos trabajar “we did not like to work”); indings that have already been documented on previous studies (Silva-Corvalán 1994: 180–81; Toribio and Nye 2006: 268). Additionally, this innovation can also explain the intraspeaker variation observed in this domain because it involves having access to two diferent syntactic options. hat is, having these two options available in their grammars would help explain why target- and non target-like forms are observed within the same heritage speaker. Given the syntactic overlap between the two groups of verbs, direct evidence in favor of this reanalysis could only be provided via an experimental demonstration of class III psych-verbs taking an optional agentive syntax, a relex of the proposed reanalysis. To be clear, I hypothesized that HSs’ grammars would accept, or at least have a reduced threshold for rejection of this emergent optionality but that Spanish monolinguals’ grammars would categorically reject it. he data reported in the previous section seem to suggest that this is indeed the case since 1) the HS and the control group were able to make precise distinctions between all relevant grammatical and ungrammatical conditions (i.e., transitive verbs, class II psych-verbs, and unaccusative verbs (not including class III psych-verbs)) in similar ways, but crucially; and 2) only the HS groups showed a tendency to accept passives with gustar-like verbs. Admittedly, the diferences observed may not be as categorical as one would want them to be to indisputably conirm the abovementioned hypothesis. However, I argue that this is suicient to show that these results are consistent with the change I proposed in the direction that I hypothesized. hat is, given the premise that the syntax of these HSs is dichotomous in nature, it logically follows that this sort of construction should either be allowed or rejected. In other words, if the HSs’ syntax did not allow for this option, then they should reject it more categorically, very much in the same manner that they reject passives with unaccusatives. So, why is it that HSs’ grammars would allow for this construction, yet their acceptance rates are not as categorical as other grammatical constructions? Here, I would like to argue that the proposed change does not imply that for the HSs, the agentive reading and/or the innovative structure for that matter is now obligatory. Rather, the hypothesis provides HSs’ grammars with an additional alternative, an optionality that is precluded in monolingual Spanish. As a result of this, it was predicted that one could ind variability in the data, whereby HSs sometimes would favor gustar used as true stative class III psych-verbs with the prescribed syntactic structure and some other times with the proposed innovation. he analysis of the individual responses presented herein (see Figures 3–6) adds further support to this hypothesis because of the following three trends. First, it was observed that the individual judgments to the counterbalance and baseline conditions were not only fairly consistent within each informant, but they were also comparable to those of other informants within 46 Hispania 99 March 2016 the same group. Second, the individual judgments to all three counterbalance conditions were generally on target across all individuals and across all three groups indicating that they were able to discern between grammatical and ungrammatical sentences in the passive voice. hird, the larger intraspeaker variable acceptance observed among HSs for the tokens that tested gustar-like verbs in passive constructions (but not to the same extent in the other conditions) speaks to the legitimacy of the emerging syntactic option proposed herein since the HS grammars can opt for, but are not restricted to, an agentive reading of gustar. To conclude, the goal of this study was to describe and explain the current status of gustarlike verbs in HS Spanish. It was hypothesized that this group of verbs had undergone a reanalysis of their argument structure whereby, in conjunction with its obligatory dative-experiencer syntax, Spanish HSs could also have access to an optional agentive syntax. hough not categorical, the main results showed that only the HS groups were able to accept passives with gustar-like verbs. Given these indings, and to the extent that the syntactic and semantic restrictions of the passive voice are maintained across the groups examined, I argue that the data presented are consistent with the proposed hypothesis. Despite inding support for the reanalysis, these data also leave us with a series of unanswered questions: what do these HS outcomes tell us about the nature of HS linguistic knowledge? Also, what is the source of these diferences? Unfortunately, with the data available, it is not possible to determine whether incomplete acquisition, L1 attrition, input diferences, or a theory of multiple grammars are likely the primary contributors to the emergence of such optionality. I hope that the discussion of the data I have presented here will encourage future research exploring these and other possibilities. NOTES 1 An unaccusative verb is a kind of intransitive verb whose only argument is internal. Crucially, structural subjects of unaccusative verbs do not take on agentive roles, but are rather underlying objects. A few examples of unaccusative verbs in Spanish include aparecer ‘to appear.’ caer ‘to fall,’ lorecer ‘to blossom.’ 2 his is not completely surprising given that Spanish monolingual children do not use this group of verbs in an adult-like fashion until approximately the age of six years old (e.g., Gómez Soler 2011, 2012; Torrens, Escobar, and Wexler 2006), an age at which most HSs have already shited to English dominance. 3 Although, again, a preference towards having the experiencer control verb agreement was clear. 4 Of the 16 informants, one reported having lived outside the country (Norway), but only for a period of 2 months. Additionally, besides three adult participants that reported having very basic knowledge of Russian, no other participants reported having knowledge of any other language. 5 Informants were asked to mark their judgments on a piece of paper. hey were not asked to provide a correction of the ungrammatical sentences. 6 To measure the internal consistency of the instrument, Cronbach’s Alpha was calculated for each condition using SPSS 21. he values obtained were 0.584 (passives with transitive verbs), 0.697 (passives with unaccusative verbs), 0.788 (passives with class III psych-verbs), and 0.778 (passives with class II psych-verbs). hese scores reveal that the internal consistency of the scale is highly reliable. 7 It should be noted, however, that although participant #8 from the monolingual control group provided a rating of 4 “completely natural” to one of the tokens included in this condition, I believe this to be an isolated slip-up given the fact that the rest of the ratings from this individual for this condition were consistently 1 “completely unnatural.” WORKS CITED Arad, Maya. (1998). “VP-Structure and the Syntax-lexicon Interface.” PhD diss. University College London. Print. Beaudrie, Sara, and Marta Fairclough. (2012). “Spanish as a Heritage Language in the United States: he State of the Field.” Georgetown: Georgetown UP. Print. Belletti, Adriana, and Luiggi Rizzi. (1987). “Psych-Verbs and θ-heory.” Natural Language and Linguistic heory 6.3: 291–352. Print. Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 47 Benmamoun, Abbas, Silvina Montrul, and Maria Polinsky. (2013). “Heritage Languages and their Speakers: Opportunities and Challenges for Linguistics.” heoretical Linguistics 39 (3–4): 129–81. Web. 8 Dec. 2015 Bowles, Melissa, and Silvina Montrul. (2009). “Instructed L2 Acquisition of Diferential Object Marking in Spanish.” Little Words. heir History, Phonology, Syntax, Semantics, Pragmatics and Acquisition. Ed. Ronald Leow, Héctor Campos, Donna Lardiere. Washington DC: Georgetown UP. 199–210. Print. Cuza, Alejandro, and Joshua Frank. (2011). “Transfer Efects at the Syntax-semantics Interface: he Case of Double-que Questions in Heritage Spanish.” Heritage Language Journal 8(1): 66–89. Database. 8 Dec. 2015. De Prada Pérez, Ana, and Diego Pascual y Cabo. (2011). “Invariable gusta in the Spanish of Heritage Speakers in the US.” Proceedings of the 11th Generative Approaches to Second Language Acquisition (GASLA 2011). Ed. Julia Herschensohn and Darren Tanner. Somerville: Cascadilla Proceedings. 110–20. Web. 17 May 2015. ———. (2012). “Interface Heritage Speech across Proiciencies: Unaccusativity, Focus, and Subject Position in Spanish.” Selected Proceedings of the Fourteenth Hispanic Linguistics Symposium. Ed. Kimberly Geeslin and Manuel Díaz-Campos. Somerville: Cascadilla Proceedings. 308–18. Web. 21 Jul. 2014. Dvorak, Trisha, and Carl Kirschner. (1982). “Mary Likes Fishes: Reverse Psychological Phenomena in New York Puerto Rican Spanish.” Bilingual Review 9.1: 59–65. Print. Figueira, Rosa. (1984). “On the Development of the Expression of Causativity: A Syntactic Hypothesis.” Journal of Child Language 11.1: 109–27. Database. 8 Dec. 2015. Franco, Jon and Susana Huidobro. (2003). “Psych Verbs in Spanish Leísta Dialects.” Linguistic heory and Language Development in Hispanic Languages. Ed. Silvina Montrul and Francisco Ordóñez. Somerville: Cascadilla. 138–57. Print. Gómez Soler, Inmaculada. (2012). “Acquiring Spanish Psych-Verbs: Maturation or Continuity?” Online Proceedings Supplement of the 36th Boston University Conference on Language Development. Ed. Alia Biller, Esther Chung, and Amelia Kimball. Somerville: Cascadilla Proceedings. Database. 8 Dec. 2015. Gutiérrez-Bravo, Rodrigo. (2007). “Prominence Scales and Unmarked Word Order in Spanish.” Natural Language and Linguistic heory. 25.2: 235–71. Database. Henriksen, Nicholas. (2015). “Acoustic Analysis of the Rhotic Contrast in Chicagoland Spanish: An Intergenerational Study.” Linguistic Approaches to Bilingualism 5(3): 285–321. Web. 8 Dec. 2015. Laleko, Oksana. (2010). “he Syntax-pragmatics Interface in Language Loss: Covert Restructuring of Aspect in Heritage Russian.” PhD diss. U of Minnesota. Web. Landau, Idan. (2010). he Locative Syntax of Experiencers. Cambridge: MIT P. Print. Lynch, Andrew. (1999). “he Subjunctive in Miami Cuban Spanish: Bilingualism, Contact, and Language Variability.” PhD diss. U of Minnesota. Database. 8 Dec. 2015. Mikhaylova, Anna. (2012). “Aspectual Knowledge of High Proiciency L2 and Heritage Speakers of Russian.” Heritage Language Journal 9.1. Database. 8 Dec. 2015. Miller, Lauren. and Alejandro Cuza. (2013). “On the Status of Tense and Aspect Morphology in Child Heritage Spanish: An Analysis of Accuracy Levels.” Proceedings of the 12th Generative Approaches to Second Language Acquisition Conference. Ed. Jennifer Cabrelli Amaro, Tifany Judy, Diego Pascual y Cabo. Somerville: Cascadilla Proceedings. 117–29. Print. Montrul, Silvina. (2000). “Causative Psych Verbs in Spanish L2 Acquisition.” Spanish Applied Linguistics at the Turn of the Millennium. Ed. Ronald P. Leow and Cristina Sanz. Somerville: Cascadilla Press. 97–118. Print. ———. (2002). “Incomplete Acquisition and Attrition of Spanish Tense/aspect Distinctions in Adult Bilinguals.” Bilingualism: Language and Cognition. 5: 39–68. Web. 8 Dec. 2015. ———. (2004). “Subject and Object Expression in Spanish Heritage Speakers: A Case of Morpho-syntactic Convergence.” Bilingualism: Language and Cognition 7: 1–18. Database. 8 Dec. 2015. ———. (2006). “On the Bilingual Competence of Spanish Heritage Speakers: Syntax, Lexical-Semantics and Processing.” International Journal of Bilingualism 10: 37–69. Database. 8 Dec. 2015. ———. (2008). Incomplete Acquisition in Bilingualism. Re-examining the Age Factor. Amsterdam: Benjamins. Print. ———. (2009). “Knowledge of Tense-aspect and Mood in Spanish Heritage Speakers.” International Journal of Bilingualism. 13.2: 239–69. Database. 8 Dec. 2015. ———. (2011). “The Linguistic Competence of Heritage Speakers.” Studies in Second Language Acquisition. 33.2: 155–61. Database. 8 Dec. 2015. 48 Hispania 99 March 2016 ———. (2012). “he Grammatical Competence of Spanish Heritage Speakers.” Spanish as a Heritage Language in the United States: he State of the Field. Ed. Sara Beaudrie and Marta Fairclough. Washington, DC: Georgetown UP. Print. Montrul, Silvina, and Melissa Bowles. (2009). “Back to Basics: Incomplete Knowledge of Diferential Object Marking in Spanish Heritage Speakers.” Bilingualism: Language and Cognition 12.3: 363–83. Database. 8 Dec. 2015. Montrul, Silvina, and Noelia Sánchez-Walker. (2013). “Diferential Object Marking in Child and Adult Spanish Heritage Speakers.” Language Acquisition 20: 1–24. Web. 8 Dec. 2015. Parodi-Lewin, Claudia. (1991). “Aspects in the Syntax of Spanish Psych-verbs.” PhD diss. University of California, Los Angeles. Print. Pascual y Cabo, Diego. (2013). “Knowledge of Gustar-like Verbs in Spanish Heritage Speakers.” Proceedings of the 12th Generative Approaches to Second Language Acquisition Conference (GASLA). Ed. Jennifer Cabrelli Amaro, Tifany Judy, Diego Pascual y Cabo. Somerville: Cascadilla Proceedings. 162–69. Print. Pascual y Cabo, Diego, Anne Lingwall, and Jason Rothman. (2012). “Applying the Interface Hypothesis to Heritage Speaker (HS) Acquisition: Evidence from Spanish Mood.” BUCLD 36: Proceedings of the 36th annual Boston University Conference on Language Development. Somerville: Cascadilla Press. 564–76. Database. 8 Dec. 2015. Pascual y Cabo, Diego, and Jason Rothman. (2012). “he (Il)logical Problem of Heritage Speaker Bilingualism and Incomplete Acquisition.” Applied Linguistics 33.4: 1–7. Database. 8 Dec. 2015. Pesetsky, David. (1995). Zero Syntax: Experiencers and Cascades. MIT: Cambridge. Print. Pinto, Darren. (2012). “Pragmatics and Discourse: Doing hings with Words in Spanish as a Heritage Language.” Spanish as a Heritage Language in the United States: he State of the ield. Ed. Sara Beaudrie and Marta Fairclough. Georgetown UP. Print. Pinto, Derrin, and Richard Raschio. (2007). “A Comparative Study of Requests in Heritage Speaker Spanish, L1 Spanish, and L1 English.” International Journal of Bilingualism 11.2: 135–55. Database. 8 Dec. 2015. Pires, Acrisio, and Jason Rothman. (2009). “Disentangling Sources of Incomplete Acquisition: An Explanation for Competence Divergence Across Heritage Grammar.” International Journal of Bilingualism 13.2: 1–28. Print. Pierce, Amy. (1992). “he Acquisition of Passive in Spanish and the Question of A-chain Maturation.” Language Acquisition 2: 55–81. Print. Polinsky, Maria. (2006). “Incomplete Acquisition: American Russian.” Journal of Slavic Linguistics 14.2: 191–262. Web. 8 Dec. 2015. ———. (2008). “Gender Under Incomplete Acquisition: Heritage Speakers’ Knowledge of Noun Catergorization.” Heritage Language Journal 6 (1): 40–71. Database. ———. (2011). “Reanalysis in Adult Heritage Language: A Case for Attrition.” Studies in Second Language Acquisition 33 (2): 305–28. Web. 8 Dec. 2015. Rao, Rajiv. (2014). “On the Status of the Phoneme /b/ in Heritage Speakers of Spanish.” Sintagma (26): 37–54. Web. 8 Dec. 2015. Ronquest, Rebecca. (2012). “An Acoustic Analysis of Heritage Spanish Vowels.” PhD diss. Indiana. Database. 8 Dec. 2015. Rothman, Jason. (2007). “Heritage Speaker Competence Diferences, Language Change and Input Type: Inlected Ininitives in Heritage Brazilian Portuguese.” International Journal of Bilingualism 11.4: 359–89. Web. 8 Dec. 2015. ———. (2009). “Understanding the Nature and Outcomes of Early Bilingualism: Romance Languages as Heritage Languages.” International Journal of Bilingualism 13.2: 155–65. Web. 8 Dec. 2015. Silva-Corvalán, Carmen. (1994). Language Contact and Change: Spanish in Los Angeles. Oxford: Oxford UP. Print. ———. (2014). Bilingual Language Acquisition. Spanish and English in the First Six Years. Cambridge UP. Print. Sok-Ju, Kim. (2012). “Acquisition of Scrambling of Negative Polarity Items (NPIs) by Heritage Korean Speakers.” Proceedings of the 12th Generative Approaches to Second Language Acquisition Conference (GASLA). Ed. Jennifer Cabrelli Amaro, Tifany. Judy, and Diego Pascual y Cabo. Somerville: Cascadilla Proceedings Project. 60–75. Web. 8 Dec. 2015. Toribio, Almeida Jacqueline, and Carlos Nye. (2006). “Restructuring of Reverse Psychological Predicates.” New Perspectives on Romance Linguistics. Benjamins: Amsterdam. 263–77. Print. Valdés, Guadalupe. (2000). “Introduction.” Spanish for Native Speakers: AATSP Professional Development Series Handbook for Teachers K–16. Vol. 1. New York: Harcourt. Web. 8 Dec. 2015. Pascual y Cabo / Emerging Optionality in Spanish as a Heritage Language 49 Valenzuela, Elena, Ewelina Barski, Adriana Diez, Ana Faure, Alma Ramirez, and Yolanda Pangtay. (2012). “Gender in the Code-mixed DPs of Heritage Spanish Bilinguals.” Hispania 95 (3) 481–94. Print. Valenzuela, Elena, Mike Iverson, Jason Rothman, Kristina Borg, Diego Pascual y Cabo, and Manuela Pinto. (2015). “Eventive and stative passives and copula selection in Canadian and American heritage speaker Spanish.” New Perspectives on the Study of Ser and Estar. Ed. Isabel Pérez-Jiménez, Manuel Leonetti, Silvia Gumiel-Molina. Amsterdam: John Benjamins. 267–92. Print. White, Lydia, Cynthia Brown, Joyce Bruhn-Garavito, Dongdong Chen, Makiko Hirakawa, and Silvina Montrul. (1999). “Psych Verbs in Second Language Acquisition.” he Development of Second Language Grammars: A Generative Approach. Ed. Elaine Klein and Gita Martohardjono. Amsterdam: John Benjamins. 173–99. Print. APPENDIX Grammatically Judgment Task 1) Transitive Passives La casa fue diseñada por los arquitectos. El libro fue escrito por unos escritores. El libro fue leído por los estudiantes. La tarea fue completada por los alumnos. El paquete fue enviado por mis amigas. 2) *Unaccusative Passives La mesa fue llegada por los estudiantes. El teléfono fue entrado por las chicas. El chico fue aparecido por mis primos. El vestido fue desaparecido por las chicas. El paquete fue salido por los estudiantes. 3) *RPPS Passives (Class III) La pizza fue gustada por mis amigos. El pastel fue gustado por los niños. La película fue gustada por mis amigos. El chocolate fue gustado por mis amigos. El helado fue gustado por los niños. 4) PP Passives (Class II) La niña fue asustada por los perros. El profesor fue asustado por los estudiantes. El chico fue asustado por sus amigos. Mis padres fueron asustados por los ladrones. Mi hermano fue asustado por los profesores. Distracters 1) *Ungrammatical (adjective agreement) La casa blanco es muy grande. El carro blanca fue caro. El café americana estuvo caliente. Las niñas altos eran de Costa Rica. Los niños guapas vivían en Colombia. 50 Hispania 99 March 2016 2) Grammatical (adjective agreement) Nosotros vivimos en un apartamento. Mis amigos son altos. La comida está caliente. Los libros de aventuras son interesantes. Las películas muy largas son aburridas. Por las mañanas siempre bebo un café con leche. Ellos tienen mucho dinero. El perro bonito es pequeño. Las películas cómicas son divertidas. El libro rojo es aburrido. 3) *Ungrammatical (verb agreement) Nosotros voy al centro comercial los ines de semana. Ellas tengo dos hermanos muy altos. Ella hago la tarea todos los días. Nosotros hago la comida algunos días. Yo compramos la comida en el supermercado.