Subjunctive use variation among monolingual native ...

John Benjamins Publishing Company

This is a contribution from Spanish in Context 11:3 © 2014. John Benjamins Publishing Company This electronic file may not be altered in any way. The author(s) of this article is/are permitted to use this PDF file to generate printed copies to be used by way of offprints, for their personal use only. Permission is granted by the publishers to post this file on a closed server which is accessible only to members (students and faculty) of the author’s/s’ institute. It is not permitted to post this PDF on the internet, or to share it on sites such as Mendeley, ResearchGate, Academia.edu. Please see our rights policy on https://benjamins.com/#authors/rightspolicy For any other use of this material prior written permission should be obtained from the publishers or through the Copyright Clearance Center (for USA: www.copyright.com). Please contact [email protected] or consult our website: www.benjamins.com

Subjunctive use variation among monolingual native speakers of Spanish A cross-dialect analysis Muriel Gallego and Emilia Alonso-Marks

Research conducted with monolingual native Spanish speakers has yielded discrepant results concerning the loss or maintenance of the subjunctive, whereas studies conducted with heritage speakers point to a gradual subjunctive loss (Mikulski 2010; Montrul 2009; Silva-Corvalán 1994a, 1994b, 2001). Crossdialectal studies (Blake 1981; De Sterck 2000; Fernández Ulloa & Portillo Mayorga 2000; Menegotto 2003, 2004, 2008) have mainly focused on past tenses, and scarce are cross-dialectal studies investigating monolingual speakers’ subjunctive use in the present tense. This study examines mood use by means of an acceptability test and a forced choice cloze test conducted by native speakers of two varieties of Spanish. One hundred and four adult monolingual native speakers of Spanish participated in the study — N = 56 from Rosario, Argentina and N = 56 from Toledo, Spain. Results indicate an overall vitality of the subjunctive and also reveal slight differences regarding acceptability based on gender and region. Keywords: Subjunctive mood in Spanish, vitality of mood contrast, dialectal variation

1. Introduction Spanish features the concept of grammatical mood, which according to the majority of approaches describes the relationship of a verb to reality and intent. Syntactically, the subjunctive mood is found primarily in dependent clauses and in a variety of constructions typically categorized as: complement clauses, relative clauses, and adjunct clauses. One key element that has been the center of many semantically based approaches to the study of the Spanish subjunctive has been the concept of assertion. Terrell and Hooper (1974) suggested three possible contexts for mood

Spanish in Context 11:3 (2014), 357–380. doi 10.1075/sic.11.3.03gal issn 1571–0718 / e-issn 1571–0726 © John Benjamins Publishing Company

358 Muriel Gallego and Emilia Alonso-Marks

alternation: asserted (not presupposed), presupposed (not asserted), and neither asserted nor presupposed. Assertion would trigger the indicative in the complement clause whereas presuppositions and neither (nor assertion or presupposition) would trigger the subjunctive. Therefore, it has been traditionally established that the most significant difference between the indicative and subjunctive moods in Spanish is the difference between certainty/objectivity (indicative) and possibility/subjectivity (subjunctive). However, mood choices obey syntactic, semantic, and pragmatic factors, which has generated much discussion and no consensus among researchers (Borrego et al. 1986; Collentine 1995; Farley 2004; Haverkate 2002; Lipski 1978; Lozano 1972, 1975; Mejías-Bikandi 1994; Sastre Ruano 1997). Native speakers (NSs) seem to achieve mood selection abilities fairly early. Blake (1983) carried out a study in which he compared dialectal differences (native speakers from Mexico and Spain). Results indicated that mood choices appear to become more evident amongst children of around 5 years of age, especially in indirect commands or adverbial clauses triggered by para que. Similarly, Montrul (2004) pointed out that children use the subjunctive in direct informal commands by the age of 2, however, as she remarks, despite the fact that subjunctive morphology emerges very early, lexical, semantic and pragmatic competencies develop later. Pérez Leroux (1998) found that the first structure acquired by children is commands (direct and indirect) while the last one is complement clauses in which the subjunctive is not only determined by syntactic factors but is also context/ meaning related, thus implying additional complexity. Investigations grounded in a variationist approach to the study of subjunctive use have focused on analyzing both linguistic and non-linguistic factors influencing mood selection and the behavior of mood alternation. Several studies have provided descriptive evidence of native speakers’ L1 use and have also established subjunctive frequency in contrast with non-subjunctive frequency (Gudmestad 2010; Lastra & Butragueño 2012). It has been observed that subjunctive use presents a great deal of variation among native speakers as shown when investigated among heritage, bilingual, and monolingual speakers (Blake 1981; De Streck 2000; Fernández Ulloa & Portillo Mayorga 2000; Kowal 2007; Lastra & Butragueño 2012; Lynch 2000; Menegotto 2003, 2004, 2008; Mikulski 2010; Montrul 2009; Silva-Corvalán 1994a, 1994b, 2001; inter alia). Previous research has contributed to the understanding of subjunctive use, including signs of loss, reduction or vitality in several groups of speakers, as well as documenting frequency regarding mood alternation, different types of clauses, and verb tenses. Studies conducted with NSs reveal certain disparities, mainly concerning the maintenance or debilitation of the subjunctive, while studies carried out with heritage speakers in the U.S. point to a more consistent loss or reduction

© 2014. John Benjamins Publishing Company All rights reserved

Subjunctive use variation among monolingual native speakers of Spanish 359

of the subjunctive mood that has been attributed to attrition, simplification or incomplete acquisition (Mikulski 2010; Montrul 2009; Silva-Corvalán 1994a, 1994b). Cross-dialectal studies have mainly investigated issues such as variability of -ra and -se in past tenses, and alternation of the subjunctive, the conditional and the indicative (Blake 1981; De Streck 2000; Fernández-Ulloa & Portillo Mayorga 2000; Menegotto 2003, 2004, 2008). To the best of our knowledge no studies have compared mood use in the context of volitional constructions in the present tense by monolingual native speakers of two non-contact varieties across continents. In order to contribute to the existing knowledge of subjunctive mood use among monolingual native speakers, the present paper seeks to expand the current understanding of mood use and vitality. 2. Subjunctive use across varieties Dialectal differences concerning subjunctive use have been explored in adult NSs of Spanish of different regions, considering the variables of age, gender, and socioeconomic status. One of the landmark works was that of Blake (1982), who compared subjunctive use in complement clauses in participants from Mexico and Spain. He found that a significant percentage of participants (more so in Mexico than in Spain) preferred the indicative in subjunctive-triggering contexts, such as that of doubt (dudar, no creer, no ser/estar seguro). Kowal’s (2007) findings were in line with Blake’s (1981) and showed that NSs in Costa Rica also selected the indicative in cases of doubt or denial, and volition. The overall analysis of a corpus composed of 50,000 words recorded by 14 participants indicated that only 21 of the 747 complement clauses registered could be considered subjunctive-triggering contexts. It was observed that the subjunctive was only used categorically in constructions with decir, esperar, hacer falta, preferir and ser muy raro, and that either the subjunctive or the indicative was used with verbs such as dejar, importar, encantar, and querer. Additionally, it was noted that participants selected only the indicative in lieu of the subjunctive, but not the infinitive. Another study investigating subjunctive vitality was conducted by Lastra and Butragueño (2012) in which they analyzed over 1,000 samples produced by 18 speakers as registered in the “Sociolinguistic Corpus of Mexico City.” Taking into account a number of linguistic and sociolinguistic variables — i.e. form of the utterance, syntactic context, semantic and pragmatic value, educational level, gender and age, results showed that the subjunctive was widely used, especially among educated young women, which was interpreted as a clear sign of vitality. The most frequent constructions were complement clauses, among which volitional



constructions ranked second. As far as tenses used, the present was identified as most frequent while the imperfect was placed second. When investigating imperfect subjunctive and pluperfect subjunctive use among college freshmen in Bilbao and Santander, Fernández-Ulloa and PortilloMayorga (2000) found that neither of the three variables under consideration — gender, place and language — had a significant correlation with the use of the subjunctive. The study confirmed the vitality of the subjunctive among first-year college students. The -ra form of the imperfect subjunctive was often used alternatively with the form -se. Nevertheless, the authors determined that -ra was widely employed in the spoken domain, whereas the -se form was commonly found to the written domain. Those findings are in line with most of the studies reported by De Streck (2000), who determined that the forms -se, -ra, and -ría could alternate in their subjunctive operationalization. However, when in subjunctive contexts, the frequency varies: -ra > -se > -ría, in that order. As for the comparison of occurrences of -ra, -se across continents, -ra appears in spoken language more than in written language, and in Latin America, -ra is overall overwhelmingly more frequent. One study that analyzed subjunctive tense alternation in Argentinian and Castilian NSs is that of Menegotto (2003, 2004, 2008), who investigated possible reasons why Argentinian NSs would not accept an utterance such as Ojalá que ganaran anoche and prefer Ojalá que hayan ganado anoche, while NSs in the Castilian region would indicate the opposite preference pattern. Menegotto (2003, 2004, 2008) indicated that Castilian speakers accept the imperfect subjunctive as categorically [+past], while Argentinian speakers do not. Therefore, the variation seen in these two regional groups responds to certain behavioral patterns in the indicative, especially when indicators of [+past] (e.g. anoche, ayer, la semana pasada) are included. Castilian speakers would not consider the indicative form he visto as [+past], but as [+simultaneous] or even [+present], therefore, they tend to consider he visto+ayer ungrammatical. Menegotto’s participants in both regions categorically accepted the use of the present subjunctive in the complement clause: Espero que los chicos ganen el partido. Volitional constructions of the type: Nosotros[SUJ1] queremos [modality] que él [SUJ2] escriba [mood] un libro are typically considered acceptable by monolingual and bilingual speakers (Gudmestad 2006, 2010, 2012) and have displayed consistency and relatively high frequency across varieties (Gudmestad 2006, 2010, 2012). The high acceptability rates could be dependent on the operationalization of the Subjunctive Disjoint Reference (SDR) effect (Kempchinsky 1986), which determines that when both subjects coincide, the construction is considered non target-like: *Yo [SUJ1] quiero [modality] que yo [SUJ1] escriba [mood] un libro. Utterances such as Ana lamenta que tenga tanto trabajo in which the subject in the complement could be considered ambiguous tend to be accepted (Kempchinsky



1986). In a later study, Kempchinsky (2009) indicated that this type of construction can be considered an embedded imperative and remarked: “co-reference is possible in those cases where a direct imperative would be infelicitous, i.e., where the subjunctive subject is not the agent in control (p. 1796).” Therefore, while the subjunctive typically implies an agent other than the subject in the matrix, the imperative implies any agent other than the speaker. 3. Objectives and research questions In an attempt to advance knowledge concerning degrees of variability in subjunctive use, the goal of the present study was to investigate mood use by monolingual NSs of two varieties of Spanish: Toledo, Spain, and Rosario, Argentina, and to identify whether the use of the present subjunctive in complement clauses is currently displaying vitality. These two regions were chosen because they have been historically monolingual with little or no contact with other languages. Grounding our analysis in a variationist approach allowed us to examine language behavior, taking into account not only geopolitical influences but also sociolinguistic variables such as age group and gender. We focused on complement clauses in the present tense because they have been traditionally established as more consistent, stable, and frequent, to determine if said stability is such in the investigated regions and sociocultural groups. Additionally, our study intended to follow up on some of the aspects investigated in Mikulski (2010) by elucidating whether the SDR effect operates in monolingual native speakers of these two varieties. The following research questions guided this investigation: Do monolingual native speakers of Spanish of the two varieties analyzed accept subjunctive forms over non-subjunctive forms in complement clauses in contexts in which the subjunctive is expected? Do monolingual native speakers of Spanish of the two varieties analyzed indicate different degrees of acceptability towards the use of the present subjunctive in complement clauses based on linguistic and sociolinguistic factors? 3. The study 3.1 Participants Participants were recruited in Toledo, Spain (N = 56) and Rosario, Argentina (N = 56). Both men and women ranging from 18 to 73 years of age took part in the



study. For the analysis of the data, participants were categorized in three age groups (18–30, 31–50, 51–73) in order to elicit generational differences. Participants in both regions were monolingual speakers of their respective varieties of Spanish and had a high school diploma (the equivalent of a total of 12 years, K-12). Table 1 summarizes information about the social distribution of participants. Table 1. Social distribution of participants. Age Group 1 18–30 yrs old

Age Group 2 31–50 yrs old

Age Group 3 51–73 yrs old

Male

Female

Male

Female

Male

Female

Toledo

10

9

10

9

8

10

Rosario

8

9

9

11

10

9

3.2 Instruments Three written instruments were designed for this study: 1) a brief background questionnaire; 2) a cloze forced-selection test (CT); and 3) an acceptability judgment (AJ) test. Given that Rosario, Argentina is a voseante region and Toledo, Spain is a tuteante region, the instruments were adapted to reflect this difference. Both the CT and the AJ test contained items that elicited mood selection and mood use of the present subjunctive in complement clauses. The AJ test was constructed following a 5 point Likert scale, where 5 indicated “I definitely say this-people in my city definitely say this-I normally hear that”, and 1 indicated “I never say this-people in my city never say this-I have never heard something like that.” Items were artificially created and adapted from Mikulski (2010), designed to account for generally accepted target-like and non target-like constructions: Volitional constructions: (1) Ella quiere que su madre cocine todas las noches. She wants that her mother cook [SUB 3rd present] every night. (2) * Yo espero que mi amigo vende su casa. I hope that my friend sells [IND 3rd present] his house.

Volitional constructions with subject co-reference: (3) * Nosotros queremos que nosotros vayamos juntos. We want that we all go [SUB 3rd present] together.



Although volitional constructions were the main focus of the investigation, constructions with verbs of emotion, appreciation, doubt, and influence were also included: (4) Yo recomiendo que Susana estudie para el examen. I recommend that Susana study [SUB 3rd present] for the exam. (5) Es importante que Susana estudie para el examen. It is important that Susana study [SUB 3rd present] for the exam.

Distractors using the indicative in different tenses were included as well: (6) Las casas son muy caras. Houses are too expensive. (7) Anoche vi una película muy buena. I saw a very good movie last night.

As indicated by Mikulski (2010) and Silva-Corvalán (2001) inter alia, the implementation of grammaticality judgment tests or acceptability rate tests might not be entirely accurate, in that the researcher is not always able to tease out which specific element in the sentence was the one influencing the judgment. This is particularly problematic in the case of the present subjunctive-indicative alternation, because of the similar morphology. Therefore, as in Mikulski (2010), we complemented the AJ test with a cloze forced selection test, in order to increase the reliability of the data set, and to eliminate confounding factors such as participants basing their judgment on issues other than mood selection per se. Therefore, a forced choice cloze test was created, comprised also of volitional constructions and other semantic categories of complement clauses: (8) Esperamos que tú llames/llamas a tu mamá durante tu viaje. We hope that you call [IND 3rdpresent]/[SUB 3rdpresent] your mom during your trip. (9) ¿Quieres que vamos/vayamos al cine? . Do you want (us) to go [IND 3rdpresent]/[SUB 3rdpresent] to the movies? (10) Recomiendo que Mariana trabaje/trabaja menos horas. I recommend that Mariana work [SUB 3rdpresent]/[IND 3rdpresent] less hours.

Since, as indicated by Lastra & Butragueño (2012) the subjunctive appears less in interrogative constructions, our instruments included more declarative than interrogative constructions.



3.3 Data collection and analysis The instruments were administered during private face-to-face interviews with the participants at each of the research sites. Participants were contacted at different work sites, social events, and sometimes through contacts of the researchers. They were informed of the nature of the research and were given instructions for conducting the tasks, however, no specific information was provided concerning the use of subjunctive to avoid biased responses. Then, participants accessed the printed material that included brief instructions and the tasks. No specific time was allocated for the completion of the tasks. Participants were required to complete the brief background questionnaire followed by the AJ task and finally the CT. Data were coded and all items were analyzed independently and grouped in categories according to the corresponding test. For the AJ test, simple frequency calculations and a Multivariate ANOVA was performed given that every item was considered an independent variable. Gender, age group, and region were the dependent variables. For the CT, since the responses were considered categorical data, a Chi-square test was performed. Similarly, all items were analyzed separately and in each analysis there were two variables: (1) independent variable: age, gender, or region; (2) dependent variable: subjunctive, non-subjunctive, or both. 4. Results 4.1 Acceptability judgment (AJ) test The Multivariate ANOVA revealed that for the majority of the items in the AJ test, no interaction or main effect was significant. Participants across regions, age groups and gender displayed similar patterns of acceptability. The majority of the target-like constructions as well as the non-target-like constructions were considered more accepted and less accepted respectively. However, some items were less categorical and showed interesting patterns, signaling significant differences across variables. Results are presented independently. Table 2 summarizes frequency calculations for target-like tokens (subjunctive-triggering contexts) that presented a slight nominal (but not statistically significant) variation across regions.



Table 2. Acceptability percentages for target-like constructions with differences across variables. 1 2 Don’t say/ hear this %* %

3

4

%

%

5 Say/hear this %

T**

R

T

R

T

R

T

R

T

R

Pablo prefiere que el novio llegue temprano.

9

7

9

9

17

18

17

18

49

49

Becky insiste en que su lugar de trabajo sea más organizado.

15

0

15

12

13

19

10

21

48

47

Raúl quiere que su amigo esca- 2 le montañas con él.

0

13

0

17

16

17

30

52

54

*. percentages are displayed according to degree of acceptability and region. **. T = Toledo; R = Rosario.

In addition, slight regional differences were found in some non target-like tokens, presented in Table 3. For some of these tokens, the differences are nominal, and some are also statistically significant, with an effect based on gender and age as well. Table 3. Acceptability percentages for non target-like constructions with differences across variables. 1 2 Don’t say/ hear this %* %

3

4

%

%

5 Say/hear this %

T**

R

T

R

T

R

T

R

T

R

El médico recomienda que Luciana no fuma.

86

100

4

0

2

0

2

0

6

0

Es importante que lees el libro de historia.

85

100

2

0

0

0

6

0

6

0

Ale te sugiere que visitas a tu amigo.

83

71

10

13

4

4

0

4

2

9


Lastly, nominal regional differences were identified in volitional constructions with subject co-reference. These are presented in Table 4.



Table 4. Acceptability percentages for subject co-reference constructions. 1 2 Don’t say/ hear this %* %

3

4

%

%

5 Say/hear this %

T**

R

T

R

T

R

T

R

T

R

Chazz y yo deseamos que dejemos de comer carne.

75

84

8

16

8

0

4

0

4

0

Yo prefiero que yo deje de fumar.

85

84

0

0

2

2

6

5

6

9

Nosotros esperamos que vaya- 83 mos juntos.

61

10

21

4

18

0

0

2

0


While generally accepted, results of the multivariate ANOVA showed a significant main effect of gender for the construction Pablo prefiere que el novio llegue temprano, Fgender(1,92) = 4.387, p = .039, partial eta2 = .046, and the overall mean for males = 3.53 (SD = 1.40), whereas for females the overall mean was = 4.11 (SD = 1.19). The mean for men in the three age groups respectively was 4.33, SD = 1.155; 3.55, SD = 1.572, and 2.50, SD = .707 in Toledo, whereas for females, the mean value was 4.08, SD = .862, 4.00, SD = 1.700, 4.00, SD = 1.323. Similarly, the mean value for male participants in Rosario was 3.14, SD = 1.345, 3.58, SD = 1.621, and 3.75, SD = 1.165 for the three age groups, while for females the mean value was 3.91, SD = 1.375, 4.17, SD = 1.329, and 4.50, SD = .674. These results indicate that this construction was accepted indistinctly across regions, but more widely accepted by females than males. Contrarily, the construction *Ale te sugiere que visitas a tu amigo was generally not accepted. However, it showed an interesting pattern. Results indicated that all the interactions and main effects for all three independent variables (region, age, and gender) were significant: Fage*gender*region(2,92) = 3.493. The effect of each independent variable was dependent on the other independent variables. Older male Toledo participants, those in the age group 51 to 73 years old showed higher acceptability rates (M = 3.00, SD = 2.828), even though the use of the indicative (visitas) would be considered generally as less normative. Females in both regions did not consider this item as acceptable. The means and standard deviations for the female Toledo participants in the three age groups are as follows: 1.23, SD = .599, 1.10, SD = .316, and 1.11, SD = 333 respectively, and those for the female participants in Rosario are: 1.09, SD = 302, 1.00, SD = .000, and 1.00, SD = .000. Given the many interactions, and due to space constraints, figures are not presented for this particular construction.



Both *El médico recomienda que Luciana no fuma and *Es importante que lees el libro de historia are overall rejected and the main effect for gender and region is significant. In *El médico recomienda que Luciana no fuma there is a significant interaction between gender and region, F(1,92) = 6.264, p = .014, partial eta2 = .064. The main effects of gender and region are also significant, Fgender(1,92) = 6.264, p = .014, partial eta2 = .064; Fregion(1,92) = 13.830, p fuyo, sedeo > seyo, video > veyo), and whose present subjunctive had a -ya ending (oya, foya, seya, veya) (472–73). [translation ours]. García Macho and Penny (2013) point out that forms such as vamos and vais were still given subjunctive value during the Golden Age and they still maintain optative value. To further consider the use of vamos with subjunctive value, we conducted a preliminary search in both CREA and CORDE. Vamos was both searched as: (1) vamos, (2) ¿quieres que vamos?, (3) ¿querés que vamos?, and (4) ¿quereis que vamos?. Options (3) and (4) rendered no results. Plausibly, option (1) rendered numerous results, however, we sorted those occurrences in which vamos was used in a subjunctive context, for example, from CORDE: (11) salgamos del baile con el lector y vamos un momento a recoger los pormenores de otra escena bien diferente en otra parte, Amalia, José Mármol (1851–1855)

Results from CORDE, for option (2) ¿quieres que vamos?, (12) Ypólito, ¿quieres que vamos a la posada y tomemos toda aquella (13) Si quieres que vamos, ¿para qué me lo dices con invencio (14) dios que por tu respeto lo he bien trabajado: si quieres que vamos mostrar te donde mora tu nueuo mari (15) E dixeron le los esclauos: quieres que vamos e la cojamos. E respondio (16) ¿Y nos preguntaréis todavía porqué combatimos? Querréis que vamos a nuestra patria a gritar ¡viva la Federación! Porque la escena no ha cambiado.

Similarly, from CREA (17) En la familia. Bueno, ¿quieres que vamos a jugar?

We also contemplate the possibility of a pragmatic alternation considering that ¿Quieres que vayamos al cine? could indicate futurity and hypotheticality, whereas *¿Quieres que vamos al cine? denotes a more immediate and more likely to be accomplished proposal. Concerning these differences, Lastra & Butragueño (2012) stipulate that: La relativa probabilidad del cumplimiento de la acción está también relacionada con la intercambiabilidad por el indicativo, especialmente en las prótasis y



apódosis de condicionales, como en si vinieras, te cansarías, frente a si venías, te cansabas, aunque el cambio de modo provoca algunas diferencias de significado…(119) The relative probability of performing the action is also related to the interchangeability with the indicative, especially in if-clauses, as in if you could come, you would get tired versus if you came, you would get tired, although the change in mood results in changes in meaning … (119) [translation ours]

Moreover, Lastra & Butragueño (2012) show that the syntactic context in which the indicative and the subjunctive appear to be more interchangeable is that of adjective clauses, yet it is less likely to happen in complement clauses. In the case of our example, *¿Quieres que vamos al cine?, the indicative triggers a semantic and pragmatic change. As in Lastra & Butragueño’s (2012) findings, the independent variable of gender plays an interesting role in our study. In the intricacies of the construction of gender one must look for explanations for correlation between gender and linguistic variables. According to Labov’s Gender Paradox (Labov 1990, 2001), females could be both conservative and innovative with respect to language variation and change. Labov’s (1990) study and his review of research in 2001 showed that, in general terms, women led linguistic changes, i.e. changes in formal styles towards prestigious norms, such as standard varieties, or changes in conversational styles towards innovative norms, but used more conservative forms of stable linguistic variables. Our findings reveal that women indicated less acceptability with regard to the present indicative in constructions where the present subjunctive would be considered normative. According to the Gender Paradox, women tend to gravitate toward the use of stable linguistic variables that may be feared to be in danger of disappearing and adhere to heritage patterns to construct a credible story of traditions. 6. Limitations and future research This study is not without limitations. Our findings add to existing evidence regarding subjunctive vitality in complement clauses and introduces some particularities concerning the SDR effect and mood alternation. Nevertheless, it does not constitute generalizable evidence due to the rather small sample size. One initial step that future research could take would be to further investigate subjunctive use in non-contact varieties, examining the same aspects that were explored here, or others, collecting a vaster sample size. Another potential limitation of this study pertains to the decontextualized nature of the instruments, which could have posed ambiguity issues for participants.



For example, an utterance such as Pablo prefiere que el novio llegue temprano, does not have a clear reference for the subjunctive subject. This means that el novio (the boyfriend) can either be Pablo’s boyfriend or somebody else’s boyfriend. This difference could have had an impact on acceptability rates. Participants in both regions accepted that particular utterance (more so females than males), however, it is not clear whether the element determining the acceptability was mood alternation or not. Similarly, further evidence is needed to determine why some of the participants indicate a slight degree of acceptability for utterances such as *El médico recomienda que Luciana no fuma and *Yo prefiero que yo deje de fumar. Given the scarcity of interrogative items of the kind ¿Quieres que vamos/vayamos al cine? in our instrument, the generalization of results is not fully possible. Further research could validate this point, in addition to determining whether the historic subjunctive value of vamos subsisted in other varieties and in other constructions. Furthermore, it would be interesting to determine whether speakers in Rosario recognize a difference, for example of pragmatic origin, between ¿Querés que vayamos al cine? and ¿Querés que vamos al cine? and if so, which would be the unmarked form. Given the nature of acceptability tests and considering that participants couId have selected normative options even when not using or hearing them frequentIy, further data elicited during sociolinguistic interviews is needed. Future research could also validate these results and tendencies with data collected in spontaneous speech. Additionally, data that explores subjunctive use in other tenses and/ or constructions, such as relative clauses and adjunct clauses would complement existing evidence. 7. Conclusion The data presented here contributes to variationist studies in Spanish and expands on previous research concerning the vitality of the subjunctive and subjunctive use across dialects. Our study, conducted in two research sites with monolingual native speaker participants from two non-contact regions, provides insight regarding subjunctive use, taking into account sociolinguistic variables such as age group, gender, and region. Congruent with previous research conducted among monolingual speakers, our findings indicated that subjunctive use, particularly in the present tense in complement clauses, is still displaying vitality in the two regions investigated, independent of age and gender. A variationist framework allowed for a description of the phenomena free from a prescriptivist approach and with the sole objective of describing acceptability rates in the two communities investigated.



Note We are deeply grateful for the valuable comments provided by the anonymous reviewers. We thank our participants, who kindly volunteered their time. Also, thanks go to Jiuqing Cheng, César Gutierrez, Rebecca Pozzi and Andrea Menegotto who contributed with suggestions, comments and material. All errors remain our own.

References Blake, Robert. 1981. “Some Empirically Based Observations on Adult Usage of the Subjunctive Mood in Mexico City.” In Current Research in Romance Languages, ed. by J. Lantolf, and G. B. Stone, 13–22. Bloomington, IN: Indiana University Linguistics Club. Blake, Robert. 1983. “Mood Selection among Spanish-speaking Children: Ages 4 to 12.” The Bilingual Review 10: 21–32. Borrego, J., J. G. Asencio, and E. Prieto. 1986. El subjuntivo: valores y usos. Madrid : Sociedad General Española de Librería, S.A. Collentine, Joseph. 1995. “The Development of Complex Syntax and Mood-Selection Abilities by Intermediate-Level Learners of Spanish.” Hispania 78: 122–35. DOI: 10.2307/345232 De Streck, Goedele. 2000. Registros y áreas geográficas en lingüística.Valores y usos de las formas verbales en -RA, -SE, -RÍA y –RE. Salamanca: Ediciones Universidad de Salamanca. Farley, Andrew. 2004. “Semantic and Syntactic Perspectives on Mood Selection in Spanish: An Analysis of Presupposition, Government, and Binding.” Estudios de Lingüística Aplicada 22: 13–35. Fernández Ulloa, Teresa, and Rosario Portillo Mayorga. 2000. “Uso del imperfecto y pluscuamperfecto de subjuntivo entre estudiantes universitarios de Bilbao y Santander.” Onomázein 5: 167–187. García, Mary Ellen, and Tracy Terrel. 1977. “Is the Use of Mood in Spanish Subject to Variable Constraints?” In Studies in Romance Linguistics: Proceedings of the Fifth Linguistic Symposium on Romance Linguistics. ed. by M. P. Hagiwara, 214–226. Rowley, MA: Newbury House Publishers. Guitart, Jorge. 1982. “On the Use of the Spanish Subjunctive among Spanish-English Bilinguals.” Word 33 (1): 59–67. Gudmestad, Aarnes. 2006. “L2 Variation of Spanish Subjunctive: Linguistic Features Predicting Use.” In Selected proceedings of the 7th Conference on the Acquisition of Spanish and Portuguese as First and Second Languages, ed. by C. Klee, and T. Face, 170–184. Somerville, MA: Cascadilla Press. Gudmestad, Aarnes. 2010. “Moving Beyond a Sentence-level Analysis in the Study of Variable Mood Use in Spanish.” Southwest Journal of Linguistics 29 (1): 25–51. Gudmestad, A. 2012. “Acquiring a variable stmcture: An interlanguage analysis of second language mood use in Spanish.” Language Learning 62(2), 373–402. Haverkate, Henk. 2002. The Syntax, Semantics, and Pragmatics of Spanish Mood. Amsterdam: John Benjamins Publishers Co. DOI: 10.1075/pbns.96 Kempchinsky, Paula. 1986. “Romance Subjunctive Clauses and Logical Form.” PhD. diss., University of California-Los Angeles.



Kempchinsky, Paula. 2009. “What Can the Subjunctive Disjoint Reference Effect Tell Us About the Subjunctive?” Lingua 119: 1788–1810. DOI: 10.1016/j.lingua.2008.11.009 Kowal, Jerzt. 2007. “La elección del modo subjuntivo en las subordinadas nominales.” Linguística Española Actual 29 (1): 45–72. Labov, William. 1990. “The Intersection of Sex and Social Class in the Course of Linguistic Change.” Language Variation and Change 2: 205–54. DOI: 10.1017/S0954394500000338 Labov, William. 2001. Principles of Linguistic Change, Vol. 2: Social Factors. Oxford: Blackwell. Lastra, Yolanda, and Pedro Butragueño. 2012. “Aproximación al uso del modo subjuntivo en el Corpus sociolingüístico de la ciudad de México.” Boletín de Filología 47 (2): 101–131. DOI: 10.4067/S0718-93032012000200004 Lipski, John. 1978. “Subjunctive as Fact?” Hispania 61 (4): 931–934. DOI: 10.2307/340945 Lloyd, Paul.1993. Del latín al español. Madrid: Gredos. Lozano, Anthony. 1972. “Subjunctives, Transformations and Features in Spanish.” Hispania 55 (1): 76–90. DOI: 10.2307/338249 Lozano, Anthony. 1975. “In Defense of two Subjunctives.” Hispania 58 (2): 277–283. DOI: 10.2307/338950 Lubbers Quesada, Margaret. 1998. “L2 Acquisition of the Spanish Subjunctive Mood and Prototype Schema Development.” Spanish Applied Linguistics 2: 1–22. Lynch, Andrew. 2000. “The subjunctive in Miami Cuban Spanish: Bilingualism, contact, and language variability.” PhD. diss., University of Minnesota. Dissertation Abstracts International 60, 8. Mejías-Bikandi, Errapel. 1994. “Assertion and Speaker’s Intention: A Pragmatically Based Account of Mood in Spanish.” Hispania, 77 (4): 892–902. DOI: 10.2307/345752 Menegotto, Andrea. 2003. “Tiempo y aspecto en dos variedades de español.” Actas del IX Congreso de la Sociedad Argentina de Lingüística. Edición en CD. Córdoba: Comunicarte Editorial. Menegotto, Andrea. 2004. “Variación dialectal en los pretéritos del español. Un acercamiento minimalista.” In Debates actuales. Las teorías críticas de la literatura y la lingüística, Facultad de Filosofía y Letras, UBA. 18–21 Oct. 2004. Edición en internet. Menegotto, Andrea. 2008. “Variación dialectal en los pretéritos simple y compuesto del español” In Lengua viva. Estudios ofrecidos a César Hernández Alonso, ed. by Álvarez Tejedor et al, 175–201.Valladolid: Universidad de Valladolid. Mikulski, Ariana. 2010. “Receptive Volitional Subjunctive Abilities in Heritage and Traditional Foreign Language Learners of Spanish.” The Modern Language Journal 94 (2): 217–233. DOI: 10.1111/j.1540-4781.2010.01018.x Montrul, Silvina. 2004. The L2 Acquisition of Spanish. Morphosyntactic Development in Monolingual and Bilingual L1 and Adult L2 Acquisition. Amsterdam: John Benjamins. DOI: 10.1075/lald.37 Montrul, Silvina. 2009. “Knowledge of Tense-aspect and Mood in Spanish Heritage Speakers.” International Journal of Bilingualism 13 (2): 239–269. DOI: 10.1177/1367006909339816 Perez-Leroux, Ana. 1998. “The Acquisition of Mood Selection in Spanish Relative Clauses.” Journal of Child Language 25: 585–604. DOI: 10.1017/S0305000998003614 Real Academia Española: Banco de datos (CREA). Corpus de referencia del español actual. http://www.rae.es. Real Academia Española: Banco de datos (CORDE). Corpus de referencia del español actual. http://www.rae.es


380 Muriel Gallego and Emilia Alonso-Marks Sastre Ruano, María Ángeles. 1997. El Subjuntivo en español. Salamanca: Ediciones Colegio de España. Silva-Corvalán, Carmen. 1994a. “The Gradual Loss of Mood Distinctions in Los Angeles Spanish.” Language Variation and Change 6 (3): 255–272. DOI: 10.1017/S095439450000168X Silva-Corvalán, Carmen. 1994b. Language Contact and Change: Spanish in Los Angeles. Oxford: Clarendon Press. Silva-Corvalán, Carmen. 2001. Sociolingüística y pragmática del español. Washington, D.C.: Georgetown University Press. Terrell, Tracy, and Joan Hooper. 1974. “A Semantically Based Analysis of Mood in Spanish.” Hispania 57: 484–494. DOI: 10.2307/339187

Authors’ address Muriel Gallego and Emilia Alonso-Marks Ohio University Gordy Hall 283, Athens, OH 45701 [email protected] [email protected]
