Fitting in with Porteños: Case Studies of Dialectal Feature Production, Investment, and Identity During Study Abroad

Abstract

In recent years, several studies across a variety of target languages (e.g., Chinese, French, and Spanish) have demonstrated that students who study abroad acquire target-like patterns of variation. In Spanish-speaking contexts, recent research has moved beyond investigating the acquisition of features specific to Spain to examine that of features used in immersion contexts such as Mexico, the Dominican Republic, Ecuador, Peru, and Argentina. Nevertheless, many of these studies either rely on quantitative variationist analysis or implement qualitative analysis of one or two target dialectal features. In addition, learner omission and expression of pronominal subjects in these contexts have been largely underexplored. Using a mixed-methods approach, this study not only quantitatively examines learners’ production of several features of Buenos Aires Spanish, including sheísmo/zheísmo, /s/-weakening, voseo, and subject pronoun expression, but it also qualitatively relates the production of these features to learners’ experiences during a five-month semester in Argentina. It aims to answer the following research questions: When and to what degree do three English-speaking students studying abroad for five months in Buenos Aires, Argentina acquire target-like production of [ʃ] and/or [ʒ], s-weakening, vos, and subject pronoun expression? How do participants’ experiences, communities of practice, investments, identities, and imagined communities relate to this production? Speech data were gathered prior to, at the midpoint, and at the end of the semester by means of sociolinguistic interviews and elicitation tasks. To further understand the connection between these learners’ use of the target features and their overseas experiences, we explored the case studies of three learners of Spanish of differing proficiency levels (beginning, intermediate, and advanced) using qualitative data collected during semi-structured interviews at each interview time. The results suggest that all three learners increased their production of the prestigious, salient dialectal features of sheísmo/zheísmo and vos during the sojourn and that the amount of increase was greater at each proficiency level. While the beginning and intermediate learners did not move toward target-like norms in their use of the often-stigmatized, less salient, variable features of /s/-weakening and subject pronoun expression, the advanced learner did. As such, stigma, salience, and variability, as well as proficiency level, may play a role in the acquisition of variable features. Learners’ investment in the target language and participation in local communities of practice increased at each proficiency level as well, and learners’ imagined communities beyond their study abroad experiences were related to their identity construction and linguistic choices abroad.

1. Introduction

Study abroad (SA) programs have long been regarded as the ideal vehicle through which a language learner can develop advanced linguistic and sociocultural skills in the target language (TL). In comparison to the traditional university language classroom, the SA setting can provide learners with more robust and contextualized exposure to the language, with continual opportunities for interaction with native speakers (NSs) of the host community and authentic cultural experiences. These opportunities often result in SA participants returning home with increased accuracy in linguistic structures, a richer and more diverse vocabulary, increased fluency and oral proficiency, and more advanced pragmatic, intercultural, or sociolinguistic competencies (Isabelli-García & Isabelli, 2020).

This notion of sociolinguistic competence refers to the ability to understand and appropriately use language in different social contexts. It involves not only linguistic knowledge but also an understanding of the social and cultural factors that influence language use, allowing individuals to navigate various social situations and adjust their language to fit the norms and expectations of different social contexts. One way to demonstrate sociolinguistic competence in a language is to acquire target-like patterns of variation. In Spanish-speaking contexts, the first studies on this topic explored features specific to Peninsular varieties of Spanish (Geeslin & Gudmestad, 2008; George, 2013, 2014; Knouse, 2013; Ringer-Hilfinger, 2013) but later expanded to include a range of countries, including Mexico (Geeslin et al., 2013; Kanwit et al., 2015; Kanwit & Solon, 2013), the Dominican Republic (Linford et al., 2021), Ecuador (Escalante, 2018a, 2018b), Peru (Grammon, 2018), and Argentina (Pozzi & Bayley, 2020; Pozzi, 2021, 2022). Nevertheless, many of these studies either rely solely on quantitative variationist analysis or implement qualitative analysis of one or two target dialectal features. In addition, features such as subject pronoun expression (SPE) have been largely underexplored in these contexts, including Argentina, the focal context in this study.

Using a mixed-methods approach, this study examines the development of sociolinguistic competence of three learners of Spanish of differing proficiency levels studying for a five-month semester in Buenos Aires, Argentina. The quantitative analysis investigates learner production of four features of Buenos Aires Spanish (BAS): sheísmo/zheísmo, /s/-weakening, voseo, and SPE. The qualitative analysis explores how participants’ experiences, communities of practice (CoPs), investments, identities, and imagined communities relate to their production of these features throughout the sojourn.

Research on the development of sociolinguistic competence has clear implications for language teaching in and outside of SA. The view of language as socially situated helps us to understand the variable nature of the input that learners receive and how they might adopt that variation into their own speech (Geeslin, 2022). It also emphasizes the need for language teachers to provide students with opportunities for practice outside the classroom (Gurzynski-Weiss et al., 2018) and encourages practitioners to consider the norms and targets they establish for learners (Beaulieu, 2016; Drewelow & Theobald, 2007).

2. Literature Review

2.1. Acquisition of Sociolinguistic Competence

Due to the importance of sociolinguistic competence in communicative competence overall, a growing number of studies across languages has explored the importance of SA in developing this knowledge (see, for example, Arabic: Trentman, 2017; French: Kennedy Terry, 2017; Regan et al., 2009; German: Barron, 2006; Mandarin Chinese: Li, 2010, 2014; Japanese: Iwasaki, 2010; Spanish: Geeslin et al., 2013; George, 2013, 2014; Knouse, 2013; Ringer-Hilfinger, 2013). In general, this research has found that learners who go abroad use dialectal variants at a higher rate than those who do not, but typically use them to a lesser degree than NSs in the host community. The degree to which a speaker adopts local features is highly variable and mediated by a wide range of linguistic, social, and programmatic factors, such as the linguistic variables themselves, language proficiency, and the degree to which they interact with locals, which can depend on one’s social identities. However, this research has largely been conducted using quantitative methods and has not sufficiently examined the reasons for individual differences.

2.2. Argentine Spanish

Over the past 20 years, Buenos Aires, Argentina has been one of the top Spanish-speaking destinations for US students studying abroad, typically attracting between 1000 and 5000 students a year (Institute of International Education, 2023). In this section, we detail four features of BAS and the patterns of variation found among NSs in the city, who are often referred to as porteños, as well as previous research on learner acquisition of these features.

2.2.1. Sheísmo/Zheísmo

One hallmark feature of BAS is sheísmo/zheísmo, or the use of the prepalatal fricatives [ʃ] and/or [ʒ] for the sounds that correspond to the graphemes “ll” and “y”. For example, yo (“I”) and allá (“there”) are pronounced as [ʃo] or [ʒo] and [a.ˈʃa] or [a.ˈʒa]. This is different from most varieties of Spanish, which use a voiced palatal fricative ([ʝo] and [a.ˈʝa]) or affricate ([ɟʝo] and [a.ˈɟʝa]). Studies on variation in BAS indicate that the prepalatal sounds [ʃ] and [ʒ] are considered prestige norms. However, there is still some variation between the voiced and voiceless variants, with younger speakers of BAS preferring the devoiced variant [ʃ] and some older speakers using the voiced and devoiced variants (Rohena-Madrazo, 2015). Although the devoicing of these phonemes has neared completion (Chang, 2008; Rohena-Madrazo, 2015), due to the variation still present in BAS, in this study, we examine the use of the voiced or devoiced variants.

A few studies have explored the acquisition of BAS phonological features in SA sojourns (Hoffman-Gonzalez, 2015; Pozzi & Bayley, 2020; Pozzi, 2022). These studies found that, by the end of a semester in Buenos Aires, participants made gains in nativelike production of the target phonological features, using [ʃ] and/or [ʒ] 84–90% of the time. These findings demonstrate that participants increased their use of these features throughout SA, with the most dramatic increase occurring during the first 2.5 months. In fact, learner production of these features in Buenos Aires was greater than that of previously studied phonological features of Peninsular Spanish in Spain, perhaps due to the ease of implementation, prestige, salience, and stability of the BAS phonemes.

2.2.2. Syllable-Final /s/-Weakening

Syllable-final /s/-weakening is a widespread phonological process in which nearly half of the world’s Spanish speakers participate (Lipski, 2011), but it tends to be used more often among speakers of lower socioeconomic status in vernacular speech (File-Muriel, 2007). In many Spanish dialects, /s/-weakening is viewed as a gradient progression from more to less robust realizations of /s/: a maintained voiceless sibilant [s], a partially weakened aspirated variant [h], and the fully reduced, deleted, or elided variant (Lipski, 2011).

BAS is known to be an /s/-weakening variety of Spanish (Rasmussen & Zampini, 2010) in which /s/ tends to be maintained before a vowel or a pause and aspirated or deleted before a consonant within a word or in word-final position (Hualde, 2005).

Table 1. /s/-weakening in BAS.

Buenos Aires
	[s]	[h]	0	Tokens
_C	12%	80%	8%	4150
_##C	11%	69%	20%	5475
_##V	88%	7%	5%	2649
_//	78%	11%	11%	2407

Adapted from Bybee (2000).

shows /s/-weakening rates in BAS as presented by Bybee (2000), based on Terrell’s (1977, 1978, 1979) studies. These numbers suggest that in BAS, aspirating /s/ is more common than completely deleting it, though deletion, which is often considered stigmatized (Colantoni & Kochetov, 2016), does occur.

Studies on learner acquisition of /s/-weakening (Escalante, 2018a, 2018b; Geeslin & Gudmestad, 2008; Pozzi, 2022; Sayahi, 2005; Schmidt, 2011) have examined the perception and production of the feature. First, there is evidence that learners can acquire the ability to perceive /s/-weakening. Escalante (2018b) found that nearly all of her eleven participants who spent a year in coastal Ecuador made significant gains in the perception of the variant, and learners with higher proficiency levels were more likely to perceive it. The same pattern was evident in a study conducted by Schmidt (2011), who found that the perception of /s/-aspiration in the United States emerged at the high-intermediate level and continued to increase at advanced levels.

However, there is little evidence of learner adoption of the feature, despite its prevalence in many dialects of Spanish. For example, Escalante (2018a) found that after a year of immersion in coastal Ecuador, participants weakened /s/ only 4.6% of the time, and one of the participants was responsible for most of this use. Similarly, Sayahi (2005) found that learners maintained /s/ 94% of the time in Morocco, and Geeslin and Gudmestad (2008) found that only 5 out of 130 learners exhibited /s/-weakening in the United States. Exploring the acquisition of /s/-weakening among learners in Buenos Aires, Pozzi (2022) found that participants exhibited /s/-weakening just 0.56% of the time mid-program and only 1% of the time after 5 months in Buenos Aires, with two students accounting for most of this use.

2.2.3. Voseo

The authentic voseo, which is used in Argentina, refers to the use of the pronoun vos and the verb forms associated with it. It is not only highly frequent, salient, and recognizable (Lipski, 1994; Schreffler, 1994) but it is also used uniformly across social levels and contexts in Argentina in place of tú (Lipski, 1994), the informal address form used in many other varieties of Spanish.

Few studies (Hoffman-Gonzalez, 2015; Pozzi, 2021, 2022) have explored learner acquisition of vos in Argentina, finding that participants used vos 59.6% (Hoffman-Gonzalez, 2015) to 70.4% of the time (Pozzi, 2021, 2022) by the end or after SA in Buenos Aires. Pozzi (2021, 2022) found that the stronger the learners’ social networks and the higher their proficiency level, the more they used vos verb forms. In fact, learners in Buenos Aires produced vos at a much higher rate in oral tasks (59.6–70.4%) than learners in Spain produced vosotros in oral tasks (20.96% in George, 2013). This could be because although neither vos nor vosotros forms tend to be explicitly taught in US Spanish classes (Cameron, 2012; LeLoup & Schmidt-Rinehart, 2017), students might struggle to not only acquire the vosotros form and use it to address a group informally in Spain but also to remap the ustedes form to only address a group formally there. In contrast, it may be relatively simple for learners to acquire and use vos verb forms instead of tú verb forms to address an informal interlocutor in Argentina, as no remapping is necessary. The higher rate of vos acquisition in Argentina may also occur because students are increasingly exposed to vos in Argentina, as it is used in a growing number of situations, including service encounters, in which usted was previously used (Kapovic, 2007).

2.2.4. Omission or Expression of Pronominal Subjects

The vast literature on SPE (i.e., voy “(I) go” vs. yo voy “I go”) among NSs in Spanish-speaking contexts suggests that rates vary widely across dialects. Varieties with lower overt expression rates include Andean Spanish (16%, Cerrón-Palomino, 2018) and Mexican Spanish (19%), while varieties with the highest rates have been found in the Caribbean, specifically in the Dominican Republic (41%, see Otheguy et al., 2007). To our knowledge, only three studies have examined this topic in BAS, but there is some discrepancy between the rates found. Barrenechea and Alonso (1977) and Soares da Silva (2006) found overt subject pronouns present in oral interviews at rates of 21 and 32%, respectively, but Pešková (2013), who explored overt SPE in semi-spontaneous speech, found a higher rate at 48%. This difference could be attributed to methodology; Pešková asked participants to respond to tasks, such as “Preguntale a tu padre qué Ø opina de Buenos Aires” (“Ask your father what (he) thinks about Buenos Aires”). This was intended to lead to a target question from the participant, such as ¿Qué opinás (vos) de Buenos Aires? (“What do [you] think of Buenos Aires?”). Since interviews typically allow participants to speak longer about the same person, which would lead to fewer changes of referent—the strongest predictor of overt SPE across many languages, including Spanish (Guy et al., forthcoming)—it may be that the semi-spontaneous prompts used in Pešková (2013) led to a higher overt SPE rate because speakers were not encouraged to continue the discourse further. Taking this into consideration, we expect that learners in Argentina would be exposed to overt SPE by NSs somewhere between 21 and 32% of the time.

A growing number of studies using a variationist methodology has examined the acquisition of SPE among English-speaking L2 learners, often comparing cross-sectional L2 SPE rates to NS rates (see Geeslin et al., 2015; Geeslin & Gudmestad, 2008, 2011, 2016; Gudmestad & Geeslin, 2010; Linford, 2009). Overall, these studies show that, as learner proficiency increases, the use of null subject pronouns increases. Linford (2009) found that beginning learners produced null subjects 60% of the time, and advanced learners produced them 87% of the time, with the advanced rate overshooting that of NSs, which is consistent with Geeslin et al. (2015).

2.3. Case Studies in Study Abroad Research

Since language learning in an SA context “is as much a process of socialization as it is of acquisition” (Kinginger, 2009, p. 156), students’ (non) participation in the host community, determined by the learner and the community, relates to the acquisition of sociolinguistic competence. One way that learners might gain participation in the target community is through a community of practice (CoP, or group of individuals who interact together and share common goals, Wenger, 1998), for example, a host family. Nevertheless, host families do not always effectively integrate host students (Wilkinson, 1998). For example, in Kinginger’s (2008) study, Ailis stayed with an older single woman with whom she did not share much conversation. In contrast, Bill reported having 4 hour dinners with his host family three times a week, during which they discussed his interests. Thus, both the quality and the quantity of NS interactions are important in SA (Magnan & Back, 2007; Segalowitz & Freed, 2004).

Not only are host families potential settings for quality interactions in the TL in SA, but social networks can be instrumental in L2 learning as well (Isabelli-García, 2006). In Kinginger’s (2008) study, Deidre lived by herself, only spoke French in service encounters and class, and did not make French friends. With time, she decided that French was too difficult and her cultural norms were better than those of the French. Ultimately, she opted to interact on the internet, barely improving her French in SA.

In addition to social networks, language development is mediated by identity and investment in the TL (Menard-Warwick & Palmer, 2012; Norton, 2001; Watson-Gegeo, 2004). This notion of investment refers to one’s desire to learn the TL due to the benefits such learning may bring as it relates to one’s identity (e.g., increased value in the social world, access to privileges afforded to TL speakers). As learners interact in the TL community, they continually construct their identities and adjust how they relate to others based on how they perceive themselves and how they believe they are perceived by TL users (Norton Pierce, 1995). This complexity of co-constructed learning is exemplified in the case of Beatrice (Kinginger, 2008), who spent extensive time with her host family and wanted to become a French teacher upon her arrival in France, yet due to conflicts with her hosts about the war in Iraq, her national identity conflicted with her developing identity as a French speaker. She retreated from interactions with NSs, adopted a negative attitude toward French culture, and made minimal linguistic gains. In this way, students’ investment in their TL, identities, attitudes toward the target culture, and experiences in SA affect their interactions and language learning.

Previous work has also examined the often negative ways in which students of color have been positioned in SA contexts. In Talburt and Stewart’s (1999) study, Michiela, an African American female studying in Spain, reported facing racial discrimination and hypersexualization, which affected her desire to engage with locals. Similarly, in Quan’s (2018) study, an Iranian American student of color who was studying in Granada, Spain, Vera, felt she was often stared at on the street and believed that “her inability to access communities of TL speakers in Spain [was] a product of her race and ethnicity” (p. 37). In fact, Vera felt “positioned as an incompetent Other because of the way she looked” (Quan, 2018, p. 37).

Heritage speakers (HSs), or individuals who have personal or familial connections to a language other than English in the United States (Beaudrie & Fairclough, 2012), have also been positioned in diverse ways when studying the language of their heritage in an immersion context. In fact, some HSs have faced racial, class-based, and linguistic discrimination (Shively, 2016). For example, Riegelhaupt and Carrasco (2000) reported that a Mexican American aspiring bilingual teacher, Lidia, was judged as low-class by her host family in Guanajuato, Mexico, due to speaking a contact variety of Spanish used in her hometown in Yuma, Arizona, whereas her Euro-American peer was accepted and welcomed by the same host family, despite making more errors. Similarly, Quan (2018) examined the ways in which a half-African, half-Mexican American student, Caroline, distanced herself from her host family in Spain because she felt the family had positioned her as an incompetent Spanish speaker.

HS perceptions of the local dialect, as well as their connections with locals, have also been found to be related to their acquisition (or lack thereof) of dialect-specific features in Spain (see George & Hoffman-González, 2019; George & Salgado-Robles, 2021; Peace, 2021; Salgado-Robles & George, 2019). For instance, in Peace’s (2021) study, four HSs chose to use Peninsular variants while studying in Spain, in some cases because they viewed Peninsular Spanish as “correct”. On the other hand, one student, Julia, maintained her home dialect due to her desire to speak Spanish in a way her Mexican family would approve of and due to her greater connection to her home community than to the SA community. Similarly, Kentengian (2020) found that most of her HS participants chose not to adopt salient features of Peninsular Spanish while studying in Spain, with the exception of those who had more social relationships with Spaniards and those who positively evaluated the culture. Finally, in George and Hoffman-González’s (2019) study, although Jessica, a Mexican American who studied in Argentina, was ridiculed by her family for her use of Argentine Spanish (specifically vos) upon her return to the United States, she felt proud of her ability to use different varieties of Spanish, as she believed it marked her as “worldly”.

A speaker’s imagined communities (Norton, 2001) can also impact language learning during SA. Imagined communities are the communities constructed through learners’ imaginations—the communities that learners participate in or wish to join or gain membership to in the future. Kanno and Norton (2003) explain that since our actions will be driven by our aspirations, “our identities then must be understood not only in terms of our investment in the ‘real’ world but also in terms of our investment in possible worlds” (p. 284). Few studies (Kentengian & Peace, 2019; Pozzi et al., 2023; Pozzi & Reznicek-Parrado, 2021) shed light on the complex relationship among identity, imagined communities, and the acquisition of dialectal features. Kentengian and Peace (2019) found that their HS SA participants accommodated minimally to Peninsular Spanish, instead choosing linguistic practices that aligned with their imagined community of educated, global speakers of transnational varieties of Spanish. In Pozzi et al.’s study, Juan was an HS of Mexican descent who initially wanted to use vos but confirmed his Mexican American identity while in Argentina, ultimately figuring out how to “be himself” in Spanish by using tú, the address form used in his imagined community at home, where he planned to return and speak Spanish after SA. In Pozzi and Reznicek-Parrado (2021), three HSs of Mexican descent studying in Argentina affirmed their heritage identities, gained awareness of linguistic variation, and confirmed their imagined communities in the United States. In these studies, the HSs’ imagined communities were not rooted in the host communities abroad, but rather in their desire to become multidialectal speakers of Spanish or in their commitment to returning home. Thus, their evolving identities and imagined communities were closely tied to their adoption or rejection of local dialectal features.

As seen here, learners are not fixed people with predictable abilities or experiences, nor can we presume that they have the goal of speaking a particular monolingual variety of Spanish. Still, the field tends to classify students as either L2 learners, assuming they are monolingual in English without previous experience with Spanish and whose goal is to speak like an NS, or HSs, who are believed to use Spanish in their families and communities and to study abroad in order to connect with a single identity related to the country of their heritage (Leeman, 2015). These fixed categories and assumptions, however, are limiting in that they do not reflect the diverse, complex realities or aims of our students. For example, how do we classify students who may have grown up in a home in which the dominant language is English but live in a community in which the dominant language is Spanish, and who are committed to using their community’s variety of Spanish in their daily lives and in immersion contexts? As seen here, these L2 and HS labels do not necessarily accurately represent learners’ rich, diverse stories. Nevertheless, case studies that examine qualitative information in greater detail can provide additional insight regarding learners’ trajectories and goals, moving beyond the limitations of any imposed scholarly constructed terminology.

In sum, previous studies on the acquisition of dialectal features are generally quantitative, while few have adopted a qualitative case study approach, and these studies tend to focus on one or two features (see Hoffman-Gonzalez, 2015; Pozzi et al., 2023). This study explores the cases of three focal participants of differing proficiency levels using mixed methods to relate each learner’s quantitative production of linguistic features used in the host community to qualitative data collected during a five-month semester in Buenos Aires, Argentina. It aims to answer the following research questions:

• When and to what degree do three English-speaking students studying abroad for five months in Buenos Aires, Argentina acquire target-like production of [ʃ] and/or [ʒ], /s/-weakening, vos, and subject pronoun expression?
• How do participants’ experiences, communities of practice, investments, identities, and imagined communities relate to this production?

3. Methodology

3.1. Context and Participants

This study’s participants—Kim, Eddie, and Brittany (pseudonyms)—were English-speaking undergraduate students whose home institutions were in different parts of the United States. They attended distinct host universities in Buenos Aires affiliated with different semester-long SA programs that lasted 5 months (20 weeks). All three lived with host families, although Brittany later moved to an apartment alone, and Eddie switched host families toward the beginning of the sojourn. None of them had previously visited Argentina or had contact with people from the country before participating in SA in Buenos Aires. Each student was assessed by their respective SA program to determine their placement in a particular level of Spanish classes, which we refer to in this study as their proficiency level. Focal students took courses with international students or with Argentines. Our only participant who completed an internship was Eddie (intermediate proficiency), who chose to work in the area of human rights, which was also his concentration while abroad.

In order to recruit participants, the first author, who was an L2 speaker of BAS who had previously lived in Argentina, sent an email about the study to a company that organized SA programs in Buenos Aires. This company then distributed the study information to SA students, and 23 of these students chose to participate in the study. Three of those twenty-three were chosen as focal participants for the case studies: Kim, Eddie, and Brittany. These focal students, outlined in

Table 2. Case study focal students.

Name	Age	Gender	Level	Racial/ Ethnic Background	Living Situation	Internship	Program Type
Kim	20	Female	Beginning	African American	Host family	No	International students
Eddie	20	Male	Intermediate	Mexican American	Host families	Yes	Argentines
Brittany	22	Female	Advanced	Non-Hispanic White	Host family, Apartment alone	No	Argentines

below, were chosen because their production of the features under investigation varied, they had different pre-program proficiency levels and racial/ethnic backgrounds, and their cases were quite distinct from one another.

3.2. Procedures

These students completed a background questionnaire prior to SA. In addition, the first author conducted three recorded interviews that lasted approximately one hour each and included a semi-structured interview in English about their SA experiences, followed by a variety of instruments at each interview time: prior to or at the beginning, in the middle (after 2.5 months), and at the end or immediately following their return from Buenos Aires (after 5 months). The instruments included a sociolinguistic interview in Spanish about their daily routines and their experiences abroad lasting 20 minutes or more, a reading passage from Todo Mafalda (Quino, 2007) selected from that used in Chang (2008), a word list based on Davies’ (2006) frequency dictionary (which incorporated 24 words from the comic strip and 28 distracters), an oral discourse completion task (DCT) based on George (2013), and two role plays adapted from Kinginger (2008) and Villareal (2014). The oral DCT consisted of 28 situations written in English that asked for oral responses in Spanish, 16 of which elicited vos in the imperative or present indicative and 12 of which were distractors. The role plays aimed for the students to use address forms to obtain information from someone else (Villareal, 2014) while participating in spontaneous speech and imitating real-life turn-taking (Bardovi-Harlig, 2013).

3.3. Linguistic Features

With respect to sheísmo/zheísmo, every time a student uttered an instance of “y” or “ll” during the sociolinguistic interviews, the reading passage, and the word list, the realization of those graphemes was coded as [ʃ], [ʒ], [ʝ], and [ɟʝ]. The first listen of each interview involved an impressionistic analysis performed by the first author, followed by another impressionistic analysis performed by an applied linguist who was a NS of BAS. A total of 95% of the tokens were coded consistently after the applied linguist’s initial training. Afterward, any discrepancies in our coding of the 3000 tokens were examined using acoustic analysis through Praat (Boersma & Weenink, 2017). For /s/-weakening, auditory analysis (File-Muriel & Brown, 2011) was used to code each instance of coda /s/ in the sociolinguistic interviews as maintained (as in mosca “fly” realized as [moska]) or weakened (including aspiration as in [mohka] and deletion as in [moØka]). A linguistics student who was a NS of BAS was trained in this impressionistic analysis as well, and the interrater reliability was verified at 98% based on a random sample of 10% of tokens. As for vos, the first author examined students’ responses on the oral DCT and the role plays to determine whether they used the tú or vos verb form, not the subject pronoun. The purpose of this was to determine whether students acquired the morphosyntactic form, as opposed to the ability to replace one pronoun (tú) with another (vos). The use of subject pronouns was examined in the sociolinguistic interviews to determine the extent to which students used overt subject pronouns as opposed to null subject pronouns in natural speech. In terms of the envelope of variation, the SPE data analysis consisted only of clauses with animate subject pronouns that have a conjugated verb for which there is the possibility of using a null or overt pronoun (see Orozco & Hurtado, 2021).

To examine the connection between learners’ use of the target features and their overseas experiences, each case study draws on mixed methods, exploring the results of a quantitative and qualitative analysis. The quantitative analysis was conducted to determine the percentage of use of the aforementioned linguistic features by each focal student at each interview time. The qualitative analysis examined the ways in which learners’ experiences, CoPs, investments, identities, and imagined communities relate to their production of the linguistic features under investigation.

4. Case Studies

4.1. Kim

4.1.1. Background Information

Kim was a 20-year-old sophomore Spanish major at a large, private university in the Northeast who identified as African American (pre-SA interview, mid-SA interview). She had taken intermediate Spanish classes at her home university; however, through her placement test results for her SA program and her pre-SA interview, it became clear that she had a novice proficiency in Spanish. In her background questionnaire and her initial interview, Kim reported that she was more of an introvert than an extrovert and that she was nervous about her trip to Argentina. Although Kim might traditionally be considered an L2 learner since she did not grow up in a home in which Spanish was used, she reported having personal ties to the language in her home community. Not only was Kim’s best friend from the Dominican Republic, but Kim described her hometown as having a large community of Spanish speakers from the Caribbean, so she had likely been exposed to Caribbean Spanish before studying abroad; however, no data were collected regarding the extent of this exposure.

4.1.2. Quantitative Analysis

Overall, Kim’s use of the features under investigation moved toward target-like norms 2.5 months into the sojourn and away from those norms immediately following SA (after 5 months, see

Table 3. Kim’s tokens and percentages of BAS feature use.

Feature	Interview #1			Interview #2			Interview #3
Sheísmo/ zheísmo	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS
	0	71	0%	48	45	51%	52	50	49%
S-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening
	0	23	0%	0	78	0%		115	0%
Vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos
	0	14	0%	5	9	36%	4	8	33%
Subject pronoun expression	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt
	7	10	41%	39	12	31%	47	22	68%

and Figure 1). At the beginning of SA, Kim did not use sheísmo/zheísmo, /s/-weakening, or vos. She used overt pronouns 41% of the time, which, curiously, is the SPE percentage that studies have found in Dominican Spanish (41%, Otheguy et al., 2007), the dialect Kim reported she had been exposed to prior to arriving in Argentina. By the second interview, although Kim did not exhibit /s/-weakening, her use of sheísmo/zheísmo and vos increased significantly to 51% and 36%, respectively, moving toward BAS norms. Her use of overt pronouns also moved toward BAS norms, reaching 31% overt pronoun use, which is in the range of the target NS SPE rate of 21–32% found in BAS (Barrenechea & Alonso, 1977; Soares da Silva, 2006). In the final interview, she used sheísmo/zheísmo and vos slightly less than in interview 2 (49% and 33%, respectively), exhibited no /s/-weakening, and moved away from NS overt pronoun use across dialects of Spanish, with her highest SPE rate of 68%. Kim’s move away from BAS norms in interview 3 could have been affected by the final interview being conducted immediately after her return home, where she had returned to hearing English and had reunited with Dominican friends.

4.1.3. Qualitative Analysis

Experiences, Identity, and Communities of Practice

Kim’s SA program in Buenos Aires focused on the Spanish language. Her classes were taught in Spanish with exchange students from different countries. This set-up was difficult for Kim, especially at first when she “didn’t know castellano”, she “was confused all day long”, and she “didn’t understand a thing” (mid-SA interview). However, over time, she began to understand and participate more in Spanish (post-SA interview).

Kim’s classes were spread out in different universities across the city. During her commute, she was unpleasantly surprised by stares, catcalls, and theft. She explained it this way:

People stare a lot, so that’s different. Staring in the U.S. is very, like, rude, but here it’s totally different, that’s just normal, people are very curious, which also, especially for me, there aren’t a lot of African Americans here, so I don’t know if I get stared at more than a normal person, but I feel like …maybe that’s it, I don’t know….

(mid-SA interview)

In addition to the staring, Kim was uncomfortable with the way men would try to call her attention. She said, “Catcalling is huge here and like any sign of the catcalling is just like interest and that guy like thinks he can get with you…it makes me very uncomfortable…sometimes I have no idea what they’re saying and I don’t think I wanna know” (mid-SA interview). Another aspect of the culture that made her uneasy was the need to be vigilant. She mentioned that “Theft is a huge thing, like pickpocketing…so I’m constantly aware and watching my surroundings, it just, like, takes a lot of energy to do that” (mid-SA interview).

Kim lived with a host mother who was in her sixties in an apartment in a wealthy neighborhood of Buenos Aires. Kim described having a positive, “well-matched” relationship with her, reporting that she spoke English and that they often spoke “Spanglish”. However, Kim struggled to participate in conversations in Spanish with her host mom and her 25-year-old host brother, who visited occasionally, especially when the conversation did not slow down and focus on her. She explained, “I’m… spending a lot of time confused about what’s going on. With my host mom we’d talk for a little bit about what we did and then they would talk really fast and I wouldn’t understand what they were saying. It’s really hard” (mid-SA interview). Thus, Kim’s language abilities seemed to prohibit her from full participation in her host family CoP.

Aside from her host family, Kim did not have many opportunities to interact with Argentines. In her mid-program interview, Kim explained that she had fallen into a routine of attending her classes, sharing meals with her host mom, and watching Netflix in her room. In her post-SA interview, Kim admitted that, despite her desire to interact with Argentines in the second half of her program, that did not happen. Instead, she barely managed to get to and from her classes, share meals with her host mom, and meet American classmates for dinner on the weekends. She explained that she did not feel confident enough in her language abilities to have real conversations with Argentines, and she did not feel that she had much of an opportunity to do so. She described it this way:

I haven’t really made a lot of connections or relationships with Argentines. It’s really difficult I think at this age… I’ve met a lot of people from other countries in South America or just around the world but I think it’s harder with other Argentines...just because they have their own lives and they have their own friends and their own routines so I think it’s harder to facilitate friendships beyond like, “Hi, bye, oh you’re here” type of situation.

(post-SA interview)

Although Kim lived with a host family, she struggled to have meaningful interactions with Argentines in and outside of the host family. In addition to not feeling like a part of Buenos Aires culture, certain aspects of the culture, including the way people stared at her, the catcalls, and the possibility of theft, made her feel uncomfortable. These experiences may have contributed to her lack of investment in the target variety and her confirmation that her imagined community was elsewhere.

Investment and Imagined Communities

Kim began studying Spanish because of her love for Hispanic cultures, and she decided to study abroad due to her desire to be engulfed in the Spanish-speaking world, to “push [her] to [her] limits” (pre-SA interview), and to grow in her knowledge of the Spanish language and as a person. She chose Argentina because her university offered programs in Spain and Latin America and she preferred to study in Latin America, where she had always wanted to travel. In addition, since her long-term career goal was to become an immigration lawyer, she preferred to study in a Latin American country since her future clients were more likely to be from Latin America than from Spain. As such, from the beginning of her sojourn in Buenos Aires, her imagined community involved working as an immigration lawyer in the United States with people from different Latin American countries. Furthermore, she wanted to study in a country that was beyond her comfort zone that she would likely not have the chance to visit again.

Prior to the semester abroad, Kim had never traveled outside the United States. Although she did not have any family that spoke Spanish, she reported having several native Spanish-speaking friends from a variety of Spanish-speaking countries (primarily from the Caribbean, specifically the Dominican Republic) in her hometown. During her time in Argentina, she hoped to become fluent in Spanish; however, she did not wish to adopt a specific variety of Spanish from a particular region.

During Kim’s mid-SA interview, she provided a somewhat negative evaluation of her SA experience. She related this to the feeling of being Othered, even about the way she ate. She mentioned, “I eat things with my hands and that is seen as taboo and barbaric”. When reflecting on this, she alluded to trying to get through her time in Argentina until she could return home. This negative evaluation was also related to the Spanish that was spoken in Buenos Aires. Kim said she was shocked to “first of all...[hear] Spanish all the time to begin with but then also [hear] this specific accent all the time” (mid-SA interview). Moreover, she had difficulty navigating the Spanish used in different contexts. For example, she became accustomed to the way her host mom spoke Spanish, but struggled to understand the way her professors spoke in her classes, the way people spoke in the streets, and particularly the way young people spoke. Regarding the castellano (Spanish) that is spoken by porteños (people from Buenos Aires), Kim did not hesitate to explain that she preferred Caribbean Spanish as opposed to BAS because that is what she had been exposed to from her Spanish-speaking friends in the United States. She added that she thought Caribbean Spanish “sounds better”, is “smoother”, and is more “beautiful” (post-SA interview) than Argentine Spanish. Despite this, she made an effort to speak like an Argentine, particularly toward the middle of the program, although she felt she was not able to do so successfully. By the end of the program, she confirmed that Caribbean Spanish better aligned with her identity as a Black speaker of Spanish and her home community of Dominican Spanish speakers, saying, “I will always just be partial to it [Caribbean Spanish]”. In addition, throughout the sojourn, she confirmed her commitment to her imagined community as an immigration lawyer in the United States, where she hoped to use Spanish to interact with Spanish-speaking clients from Latin America, but did not anticipate needing to use the castellano that she had been exposed to in Argentina.

This shift in Kim’s investment, as well as the negotiation of her identity as a Black Spanish speaker and the confirmation of her imagined community in the United States, seemed to align with her linguistic choices during SA. At the beginning of the program, some characteristics of Kim’s Spanish resembled the variety she had been exposed to, Dominican Spanish, in her use of tú (as opposed to vos) and her production of overt subject pronouns. This was not the case, however, with respect to Kim’s lack of use of /s/-weakening (a prominent feature of Caribbean Spanish), which might be attributed to a possible lack of perception of /s/-weakening in her Dominican community as a beginning Spanish speaker. In fact, she did not display /s/-weakening at all throughout the program. Toward the middle of SA, she tried to speak like an Argentine, perhaps related to her early attempts to interact with Argentines and improve her Spanish. By her mid-SA interview, she had increased her use of sheísmo/zheísmo and vos, and her use of subject pronouns moved toward the SPE rate of BAS Spanish. However, by the end of SA, her linguistic choices had moved away from BAS norms, decreasing her use of sheísmo/zheísmo and vos and increasing her use of overt pronouns as she invested in her identity as an African American speaker of Spanish in a Dominican community in the United States, where she hoped to one day use her Spanish with Latin Americans as an immigration lawyer.

4.2. Eddie

4.2.1. Background Information

Eddie was a 20-year-old who was a double major in political science and environmental studies with a minor in Spanish at a large, public university located in the Southeast. He described himself as learning Spanish as a foreign language and reported learning Spanish in school, not at home. He said that he had heard Spanish growing up around some of his Mexican friends in Los Angeles. He explained that his parents are Mexican, and they had learned Spanish in school, so they could speak Spanish for their jobs, but they did not speak it at home. That said, Eddie had been exposed to Spanish in his community, but no empirical data were collected on the amount of his exposure to the language prior to SA. He had never studied abroad or left the United States before his trip to Argentina, where he took several classes with locals as part of his SA program in Buenos Aires and was placed into intermediate classes. He lived with two different host families and participated in a human rights internship, for which he worked at a high school for students who identified as transgender. He described himself as Mexican (pre-SA interview, mid-SA interview), and, through studying abroad, he hoped to not only learn to speak Spanish fluently but also to gain a more global perspective regarding race, ethnicity, sexuality, and gender. He had been exposed to Mexican Spanish in school and in his Spanish-speaking community in Los Angeles, but in Buenos Aires, he hoped to speak like an Argentine, “fit in as much as possible, and adapt to the local community” (pre-SA interview). Thus, due to his personal and familial connection to Spanish in the United States (see Beaudrie & Fairclough, 2012), Eddie might be considered an HS of Spanish. However, since this term has been constructed by scholars and there is a lack of understanding of how this label may shape learners’ trajectories (Leeman, 2015), in this case study, we seek to examine Eddie’s own perceptions, interpretations, and ways of representing himself in various contexts (see Hornberger & Wang, 2008, p. 6).

4.2.2. Quantitative Analysis

Overall, Eddie’s use of sociolinguistic features seemed to move toward target-like norms with respect to the characteristics traditionally associated with BAS, sheísmo/zheísmo and vos; however, this was not observed for the other characteristics under investigation, /s/-weakening and SPE (see

Table 4. Eddie’s tokens and percentages of BAS feature use.

Feature	Interview #1			Interview #2			Interview #3
Sheísmo/ zheísmo	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS
	10	57	15%	55	11	83%	72	6	92%
S- weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening
	0	68	0%	0	282	0%	0	426	0%
Vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos
	0	18	0%	2	15	12%	13	7	65%
Subject pronoun expression	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt
	1	22	4%	3	44	6%	3	56	5%

and Figure 2). Eddie’s first sociolinguistic interview was conducted upon his arrival in Buenos Aires, but he had already begun to exhibit the use of BAS phonemes 14.9% of the time in his pronunciation of “y” and “ll”, and he approximated NS norms significantly more as time went on (83.3% of the time 2.5 months in and 92.3% after 5 months). Similarly, although he did not begin to use vos immediately, he moved toward NS norms, exhibiting the use of vos 12% of the time after 2.5 months and 65% of the time at the end of 5 months abroad. In contrast, Eddie did not move toward target-like norms in his use of /s/-weakening and SPE; he did not exhibit /s/-weakening at all during SA, and he produced few overt subject pronouns throughout the sojourn (4%, 6%, and 5% of the time, respectively, in interviews 1, 2, and 3). This could be due to Eddie’s contact with Mexican Spanish prior to SA, which is known for /s/-maintenance and which tends to have lower SPE rates (19%) than other dialects of Spanish (see Otheguy et al., 2007). In addition, Eddie’s lack of /s/-weakening could have been related to the tendency of lower-proficiency learners not to perceive the feature.

4.2.3. Qualitative Analysis

Experiences, Identity, and Communities of Practice

Upon his arrival in Argentina, although Eddie was used to hearing Mexican Spanish with friends in the United States, he expressed interest in learning Argentine Spanish in Argentina. He explained, “I’m in this culture and that’s the way they speak, I’m going to try to fit in as much as possible and adapt to the local thing” (pre-SA interview). Despite his positive attitude regarding SA, he faced several challenges with his host family and with attempted robberies.

At the beginning of SA, Eddie was placed in an upper-class neighborhood with a host family that initially made an effort to include him but turned out not to be a “good fit” (mid-SA interview). First, they complained about his cooking. Eddie explained, “I was making chilaquiles which is like a Mexican breakfast food that my family makes, I’m Mexican, and it’s like my favorite food, I was like, really excited to make it once I saw there was tortillas in the supermarket that we live next door to and she asked me not to make ‘cause it smells really bad” (pre-SA interview). He felt he was stereotyped as an exchange student who makes smelly food, and this rejection felt particularly hurtful because it was related to his identity as a Mexican.

In addition, the family showed little empathy after an attempted robbery near their home. He explained it this way:

I was walking home from class the other day and somebody tried to rob me, I was surrounded by like three people and they like grabbed me and I ended up just like pushing one of the guys really hard and ran, um so nothing was stolen, and it was completely fine, but, I came home and I was like…talking to my host mom and I was like yeah, someone tried to rob me, and her reaction was just like ‘oh, where were you walking?’ and, I was like, ‘Córdoba’ and she was like, ‘oh, you cannot walk at Córdoba at night’, and that was just like the end of the conversation and we ate dinner and it was just like really cold, but I think that’s pretty emblematic of a lot of the interactions I’ve had with her so far, so I’m pretty excited to be moving to a new family.

(mid-SA interview)

Unfortunately, this was not the only time someone attempted to rob Eddie near his first host family’s home, but luckily, he was not hurt either time.

Despite these initial challenges while living with his first host family, Eddie seemed to maintain an overall positive attitude toward his experience abroad. He moved to another more eclectic, diverse neighborhood in the city, which is known for its art scene. In this neighborhood, he moved in with a new host family, which consisted of an older woman from Lima, Peru, named Teresa, whom he described as “the nicest woman in the entire world”. She had lived alone in Buenos Aires for 40 years and spent several hours conversing with Eddie in Spanish during the three meals she shared with him each day. In addition, Eddie started an internship with an organization that supported transgender students in obtaining funding for their schooling. Although he had six friends from home who were studying in Buenos Aires with whom he spoke English, Eddie made friends from Bolivia with whom he spoke Spanish and spent a great deal of time. While he managed to make some Argentine friends, he did not spend much time with them or feel like part of their community. By the end of SA, Eddie settled into several CoPs (Peruvian host family, Argentine internship, Bolivian friends) that allowed him to improve his Spanish while accepting and respecting his identity as a Mexican, not asking him to change who he was to be accepted, as he had felt the pressure to do with his first host family.

Investment and Imagined Communities

Although Eddie seemed to be invested in his adaptation to the language and culture of Buenos Aires upon his arrival, the negative experiences with his first host family and the attempted robberies seemed to shake that investment in the local community and, ultimately, his imagined community. This became apparent as he recognized upsetting aspects of Argentine culture, particularly those he associated with the neighborhood and ideals of his first host family. He said,

At the moment I’m not very ecstatic about Buenos Aires…it’s very overwhelming. I’m having…not too positive experiences recently, just like a lot of racism is really overwhelming and people are like very open about it and it’s like very difficult to navigate society cause it’s just like so much racism and it’s just like really terrible…and I don’t know if I’m like very into the big city feel.

(mid-SA interview)

Upon asking him to explain more about the xenophobia he described, he explained it as follows:

A lot of people, more than 60 or 70% of people that I met in the city say things that are hard to hear…for example people from Peru, from Chile, from Bolivia, also Colombia…[they are] compared to thieves, prostitutes… bad people.

(mid-SA interview)

The comments that Eddie described hearing in the upper-class neighborhood he had just moved from were quite ironic since he was robbed in that neighborhood and came to feel more comfortable and safe once he moved to the neighborhood that was characterized as “dangerous” due to the presence of the aforementioned immigrant populations. In fact, the person who showed him the most acceptance, care, and respect during his time in BA was Teresa, his new host mom, who was from one of those countries (Peru).

Although Eddie maintained a relatively positive attitude toward his SA experience and BAS, he seemed to become less invested in Argentine society as time went on, and instead invested in his relationships with people from other countries (e.g., friends from Bolivia and his host mom from Peru) and in his efforts related to social justice (e.g., his internship that supported transgender students). While he invested his time and energy in these relationships and activities, he also began to miss home, where he felt he had more of a place in the community. In this way, although Eddie began the sojourn viewing Buenos Aires as his imagined community, where he wanted to adapt to the TL and culture, over time, he embraced his acquisition of the BAS features of sheísmo/zheísmo and vos but not those of /s/-weakening or SPE. This split in Eddie’s production of BAS features—using some but not others—was perhaps a way to embrace the target variety while holding onto his Mexican identity in connection with his imagined community at home, where he felt more accepted and aligned with societal views.

4.3. Brittany

4.3.1. Background Information

Brittany was a 22-year-old pre-law major and Spanish minor at a university in the North Central United States who identified as “White” (pre-SA interview). She was an advanced Spanish speaker who had never been to Argentina, but she had studied abroad in Chile. She explained that she did not have many opportunities to use Spanish back home prior to studying abroad and that she had not made many Chilean friends when she studied in Chile. As such, Brittany might be considered a traditional L2 learner from the United States, who had not been exposed to Spanish outside of school in her home community. She began SA in Argentina living with a host family but moved to an apartment after a few weeks, as she was used to living alone in college. She took classes in history and pre-law with locals at the university, was involved in extracurricular activities, and utilized dating websites, which helped her make connections with Argentines. During SA, she hoped to fit into the local community and speak more like an Argentine.

4.3.2. Quantitative Analysis

At the beginning of SA, Brittany described her Spanish as slow and non-native. She did not exhibit linguistic characteristics of BAS, and she used overt pronouns 22% of the time (which is less than the 35% SPE rate found by Cifuentes, 1980, in Santiago, Chile, the dialect she had previously been exposed to). Nevertheless, she moved toward target-like norms in her use of the features under investigation by the second interview 2.5 months into SA, using sheísmo/zheísmo categorically and vos nearly categorically (95% of the time). She also exhibited /s/-weakening 23% of the time and overt pronouns 22.5% of the time. By this time, she would correct herself when she used tú and opt for vos, making an effort to “fit in” by using Argentine Spanish (mid-SA interview). At the end of the sojourn, she used sheizmo/zheísmo and vos categorically, exhibited /s/-weakening 35% of the time, and produced overt pronouns 28.5% of the time (see

Table 5. Brittany’s tokens and percentages of BAS feature use.

Feature	Interview #1			Interview #2			Interview #3
Sheísmo/ zheísmo	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS	Tokens (BAS)	Tokens (No BAS)	% BAS
	0	82	0%	80	0	100%	102	0	100%
S-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening	Tokens (s-weakening)	Tokens (No s-weakening)	% s-weakening
	0	74	0	23	76	23%	117	215	35%
Vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos	Tokens (vos)	Tokens (tú)	% vos
	0	19	0	21	1	95%	19	0	100%
Subject pronoun expression	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt	Tokens (overt)	Tokens (null)	% overt
	29	32	22%	9	31	22.5%	22	55	28.5%

and Figure 3). This SPE rate remained within the target BAS SPE rate of 21–32% found in Barrenechea and Alonso’s (1977) and Soares da Silva’s (2006) studies.

4.3.3. Qualitative Analysis

Experiences, Identity, and Communities of Practice

Unlike the other two participants, who faced a variety of challenges in the host community, Brittany’s experiences seemed to be quite positive overall. In fact, since the beginning of SA, her biggest concern was the way she spoke Spanish. At first, she was so conscious of sounding non-native that she did not want to speak, even in service encounters, because she felt offended when people would ask where she was from. However, over time, she realized that someone who is not Argentine but identifies as American (pre-SA interview) can still participate in the Argentine community. She explained, “I don’t even try to pretend that I am going to sound convincing, I don’t even let it offend me” (mid-SA interview). She decided that she was going to use Spanish, even if it meant she was not perceived as Argentine: “If I don’t speak Spanish I’m not trying to insert myself in the community, so I have to speak Spanish” (mid-SA interview).

She made friends right away thanks to her classes with locals, her ability and willingness to speak Spanish, and her use of social media. She used a dating application called Tinder initially to find a date, but the application ended up facilitating the establishment of her social network. She said, “I actually went… [out] with some Tinder guy…and through him I met like my core group of friends, so yeah, honestly, I have Tinder to thank for that, for my kind of principal friend group here…” (mid-SA interview). Through her Tinder date’s roommate, she was introduced to another group, and, by the end of SA, she had made friends with several different groups of Argentines.

In addition to making friends through Tinder, although she said she was not a dancer, she decided to take dance classes with locals, including tango, salsa, and folklore. She reported that these classes allowed her to feel like part of the community and even helped her learn things about Argentine culture that some of her Argentine friends did not know. Participating in these activities led her to feel and speak more like an Argentine, too. She explained,

Yeah, it’s definitely that feeling, if I say a phrase and it’s exactly how they would have said it and no one comments or no one kind of smirks a little bit and there’s a whole conversation like that is such a rewarding feeling, kind of this feeling like I am, you know, “part of the gang”. We’re just having a conversation where my strange speech like isn’t really a factor anymore. And yeah when you can have just conversations that make you feel like a normal player, that is a really cool feeling”.

(mid-SA interview)

To Brittany, moving from feeling clearly marked as non-native and robotic when she arrived to feeling like “part of the gang” confirmed that she was accepted as part of the host community.

Investments and Imagined Communities

Brittany was clearly invested in the target community at the beginning of the sojourn and sought to learn and use Argentine Spanish during SA. Once she stopped worrying about how she spoke and how that represented her as an outsider, she was able to invest her time and energy in adapting culturally and socially, which led to becoming an insider in a variety of CoPs. During this participation in CoPs, she increasingly approximated target-like norms of Argentine speech.

Despite her integration into the local community, Brittany always thought she would return home after the sojourn. However, while in Argentina, she did not often think about that return. Instead, she put her time and energy into building her life overseas. In fact, she said that she was pleasantly surprised by her flourishing social life since she thought that making a social network abroad would be harder and that she was going to be alone often. On the contrary, in Argentina, she spoke Spanish most of the time, did not have American friends, attended classes, went to friends’ houses to drink mate, attended asados (barbeques), and learned Argentine dances. She mentioned that when she had arrived in Buenos Aires, she had used terms from Chilean Spanish, but her Spanish had changed based on the context she was in. Although her Spanish may change in the future if she is exposed to another variety of Spanish for an extended period of time, she thought it would be hard to revert to using tú, as she had been conditioned to use vos. In sum, since Brittany did not have Spanish-speaking friends or family in her home community, when she arrived in Argentina, she did not feel tied to any particular dialect of Spanish, nor did she have to worry about whether her home community would approve of her speaking like an Argentine. As such, her investment and her imagined community were rooted in the host community in Argentina throughout the sojourn, which seemed to play an important role in her approximation of BAS norms of all four features under investigation.

5. Discussion

The first research question examined when and to what degree the three focal students studying abroad in Buenos Aires acquired [ʃ] and/or [ʒ], /s/-weakening, vos, and SPE. All three participants (Kim, Eddie, Brittany) made rapid gains in target-like production of sheísmo/zheísmo and vos, with the largest increase generally occurring during the first 2.5 months, reaching 52% (Kim—beginner), 83% (Eddie—intermediate), and 100% (Brittany—advanced) by mid-SA for sheísmo/zheísmo and reaching 36% (Kim—beginner), 12% (Eddie—intermediate), and 95% (Brittany—advanced) by mid-SA for voseo. In addition, post-SA production rates for sheísmo/zheísmo reached 49% (Kim—beginner), 92% (Eddie—intermediate), and 100% (Brittany—advanced) and post-SA use of vos verb forms reached 33% (Kim—beginner), 65% (Eddie—intermediate), and 95% (Brittany—advanced). These gains increased as learners’ proficiency levels increased, which is consistent with previous literature on the acquisition of these BAS features (Pozzi & Bayley, 2020; Pozzi, 2021).

On the contrary, not all participants approximated target-like norms with respect to the other two features under investigation: /s/-weakening and overt SPE. With respect to /s/-weakening, Kim (beginner) and Eddie (intermediate) did not exhibit the feature at all during SA. Only Brittany (advanced) produced it, reaching 22% mid-SA and 35% post-SA. These findings are in line with previous work, which has found that learners rarely produce weakened versions of /s/, even after extensive exposure to the variant (Escalante, 2018a), although they do acquire the ability to perceive it over time as proficiency increases (Escalante, 2018b; Schmidt, 2011). The lack of production among the beginning and intermediate learners could be related to the stigmatization of /s/-weakening in Spanish, generally speaking, which might lead students to avoid adopting it in their speech, or to the possible lack of perception of the feature among these learners, as /s/-weakening has been found to be perceived starting at an intermediate-high proficiency level and increasing in the advanced levels (Schmidt, 2011).

The findings regarding overt pronoun use were varied. Kim started SA producing overt pronouns 41% of the time, which is the SPE rate in Dominican Spanish; she used them 31% of the time by mid-SA, approximating the BAS SPE rate of 21–32% (Barrenechea & Alonso, 1977; Soares da Silva, 2006); and she used them 68% of the time post-SA, moving back toward but greatly overshooting the SPE rate in Dominican Spanish (41%, Otheguy et al., 2007). This was perhaps due to having returned to the United States prior to the last interview, where she had been exposed to English and had likely heard Dominican Spanish again among her imagined community at home. Eddie’s SPE rates remained relatively stable throughout SA, starting with an overt rate of 4% pre-SA, reaching 6% mid-SA, and ending with an overt pronoun rate of 5%, which is closer to the SPE rate of Andean Spanish (16%, Cerrón-Palomino, 2018) and Mexican Spanish (19%, Otheguy et al., 2007) than that of BAS (21–32%, Barrenechea & Alonso, 1977; Soares da Silva, 2006). Since Eddie’s SPE rates did not change much mid- and post-SA after meeting his host mom and Bolivian friends (Andean speakers), his SPE rates were not likely influenced by theirs. Instead, Eddie’s SPE rates may be related to his confirmation of his imagined community at home, where he would likely be in contact with Mexican Spanish in his social circles. In contrast, Brittany continually moved within the BAS SPE range of 21–32% (Barrenechea & Alonso, 1977; Soares da Silva, 2006) toward the higher limit of 32% found in Soares da Silva’s (2006) study. In particular, she moved from 22% pre-SA to 23% mid-SA and reached 29% post-SA, as she confirmed her imagined community among BAS speakers in Argentina.

The difference between the participants’ rapid adoption of sheísmo/zheísmo and vos, on the one hand, and their non-adoption of /s/-weakening and mixed results with SPE, on the other, may be related to the fact that the first two features are fairly obligatory in BAS, and the latter two are more variable and context-dependent. According to usage-based linguistic theory, our experience with language creates and impacts our cognitive representations of that language (Langacker, 1987, 2000). As input is received, language users process and encode utterances and categorize them based on phonetic form, meaning, and context, building up cognitive representations in their memory (Bybee, 2013). These representations are then strengthened or weakened by incoming utterances, which are sorted and matched by similarity to existing representations. Applying these concepts to the acquisition of variable forms among language learners, it can be argued that learners would be more efficient in processing and organizing forms that are less variable in nature, as a less variable input would provide a more robust representation of that variable in a learner’s cognitive system. As such, the degree of variability of a particular feature may play a role in its acquisition.

Additionally, /s/-weakening has been found to index lower socioeconomic status, informal registers, and rapid speech (see Ryant & Liberman, 2016 and sources within) in many parts of the Spanish-speaking world, suggesting that it generally has significant social stigma in those places. In BAS in particular, coda /s/-aspiration is expected before a consonant but is less common and stigmatized before a word that starts with a vowel or in sentence-final position, and /s/-deletion is uncommon and stigmatized in any phonological context (Colantoni & Kochetov, 2016). In the classroom, learners are mainly exposed to and produce prestige forms, not stigmatized forms. In fact, they may be hesitant to produce stigmatized markers, including those that contain an element of covert prestige, due to a lack of in-group status. That said, our results suggest that stigma and salience may play a role in L2 learners’ acquisition of variable features, particularly for beginning and intermediate-level learners, who may not perceive the less salient features or who may not perceive the nuances of the prestige or stigma related to the use of those features in distinct linguistic contexts. While this study did not examine learners’ perception of the features under investigation, future research should continue to explore the role of variability, stigma, and salience, as well as learners’ awareness and perception of local features.

The second research question examined how participants’ experiences, CoPs, investments, identities, and imagined communities relate to their production of the aforementioned BAS features during the sojourn. The findings indicate that learners’ commitment to different varieties of Spanish in their home communities before SA may affect not only their investment in the TL and culture during SA, but also their investment in imagined communities post-SA. In fact, since learner participation in CoPs seemed to increase at each proficiency level, the ability to access and participate in CoPs and thus obtain the quantity and quality input necessary to acquire local features may be related to learner proficiency. Finally, learners’ racial and ethnic identities influence not only their inclusion (or exclusion) from CoPs but also the ways they are positioned and position themselves in the host community. In particular, when the African American and Mexican American participants felt Othered, they retreated from interactions with Argentines and, in some cases, developed negative evaluations of Argentine culture.

Kim, a beginning Spanish speaker who identified as African American, had limited CoPs and reported experiencing confusion regarding castellano. Similar to Deidre in Kinginger’s (2008) study, who did not make French friends, opted to interact on the internet, and barely improved her French during SA, Kim went to class and tried to speak Spanish with her host mom but often resorted to watching Netflix in her room, initially approximating BAS norms but moving away from them by the end of SA. She adjusted her investment in the TL community based on how she believed she was perceived by TL users (see Norton Pierce, 1995), moving away from her investment in the host community as she experienced stares and catcalls on the street and felt judged as “barbaric” due to the way she ate. Consequently, Kim tried to get through her stay in Buenos Aires until she could return home to a Spanish-speaking community from the Caribbean that better aligned with her identity as a Black speaker of Spanish. As Wheeler (in press) notes, for African Americans like Kim, learning Spanish entails more than just linguistic acquisition; it also involves learning how to be a Black speaker of Spanish across geographical contexts vis-a-vis sociolinguistic features that align with imagined racialized identities. That is, varieties of Caribbean Spanish tend to be associated with speakers of African ancestry (Potowski & Shin, 2019), which may resonate more with African American language learners, while varieties like BAS may be seen as more “White” Spanish. Moreover, as noted elsewhere, students of color who feel Othered may retract from local interactions (Goldoni, 2017; Quan, 20189; Talburt & Stewart, 1999). Ultimately, Kim moved away from interactions with locals as she embraced her African American and US American identities while confirming her imagined community among Latin American speakers of Spanish back home in the United States, where she hoped to become an immigration lawyer and interact with speakers of the Caribbean dialect of Spanish she preferred.

Eddie, an intermediate learner who identified as Mexican and had not been exposed to Spanish in the home growing up, began SA with several CoPs in place, including a host family, classes with locals, and an internship. However, as has been found with L2 learners (Kinginger, 2008) and HSs in SA (see Quan, 2018; Riegelhaupt & Carrasco, 2000), Eddie faced challenges with his host family that affected his SA experience. Similar to the case of Beatrice in Kinginger’s (2008) study, who started SA with a great attitude but retreated from interactions with NSs due to conflicts with her host family, ultimately making minimal linguistic gains, Eddie initially wanted to speak Spanish like an Argentine, but due to conflicts with his hosts over his “smelly [Mexican] food”, he moved away from relationships with Argentines and, in some cases, away from target-like norms. Although he moved toward BAS norms throughout SA with respect to two of the target linguistic features, sheismo/zheismo and vos, he did not approximate BAS norms with respect to the variable features of /s/-weakening or SPE post-SA. Meanwhile, he shifted his investment in the host community to make connections with speakers from Bolivia and Peru, who respected his Mexican identity. Over time, similar to Juan’s case in Pozzi’s (2021) study, Eddie embraced his Mexican identity and confirmed his imagined community among Spanish speakers in the United States. Moreover, like Julia’s case in Peace’s (2021) study Eddie ultimately felt more connected to his home community than to the SA community.

Brittany, on the other hand, was an advanced speaker who might be considered a traditional L2 learner, as she had not been exposed to Spanish in her home community in the United States. Even though she had previously studied abroad in Chile, she did not feel tied to any particular variety of Spanish prior to SA in Argentina and, instead, sought to become part of the local community through her linguistic choices and her social interactions. In fact, her co-constructed participation with NSs in a variety of CoPs, including her Argentine friend groups that she made through Tinder and her dance classes, aligned with her deep investment in the TL and culture. Similar to Bill in Kinginger’s (2008) study, who reported having four-hour dinners with his host family three times a week about his interests and making great linguistic gains abroad, Brittany reported having a high number of quality interactions with Argentines in diverse CoPs throughout SA, and she approximated BAS norms with respect to all four sociolinguistic features under investigation. She embraced her US American identity, realizing that she could fully participate in Argentine society even though she was not Argentine, and her imagined community was rooted in the host community from the beginning to the end of SA. Unlike Kim and Eddie, Brittany reported no negative incidents due to her racial, ethnic, national, or gender identities. On the contrary, her racialization as a non-Hispanic White Spanish-language learner in BA may have worked to her advantage, considering the many CoPs to which she was able to gain access. As Riegelhaupt and Carrasco (2000) noted, while Lidia, a Mexican American HS of Spanish, was perceived negatively by her host family, her classmate of European descent was viewed positively. Overall, Brittany seemed free to adopt Argentine Spanish, as she had not been committed to an alternative variety of Spanish in her home community prior to the sojourn, nor did her imagined future CoPs conflict with an Argentine identity.

6. Conclusions

This study is one of the first to examine learners’ case studies regarding the acquisition of sociolinguistic competence. It is also the first to complement the quantitative analysis of multiple features of BAS, including the underexplored feature of SPE, with qualitative data regarding learners’ experiences, CoPs, investments, identities, and imagined communities in SA. The quantitative findings show that all three participants increased their production of the prestigious, salient features sheísmo/zheísmo and vos, but not the less salient, often-stigmatized feature of /s/-weakening, and the results were mixed with respect to SPE. This may have been the case since the first two features are used categorically among NSs and are characteristic of BAS, whereas the latter two are not only variable but are also not necessarily associated with BAS in particular. Thus, stigma, salience, and variability may play a role in the acquisition of sociolinguistic variation abroad, particularly for learners at beginning and intermediate levels. Only the advanced speaker, Brittany, approximated target-like norms with respect to all four features under investigation. As such, Brittany may have been more likely to perceive the variable features than the lower-proficiency learners. Thus, future studies should examine not only learners’ production of sociolinguistic features, including obligatory and variable forms, but also their perception of them.

With respect to the qualitative results, participants’ investments in the target variety and their participation in local CoPs increased at each proficiency level. In this way, higher-proficiency learners may be more capable of accessing and participating in social networks in the host community, which, in turn, might afford them additional opportunities for increasing the quantity and quality of input and interaction and, thus, further approximation of target-like norms of variation. In addition, students’ imagined communities differed based on previous exposure and commitment to alternative varieties of Spanish. In particular, the beginning (Kim) and intermediate (Eddie) learners who had previous exposure to alternative dialects of Spanish in their home communities confirmed their imagined communities back home during SA, whereas the advanced learner (Brittany), who did not have such ties prior to the sojourn, was free to invest in her imagined community in Argentina and adopt the target dialect. As such, previous personal ties to different varieties of Spanish may play a role in dialectal acquisition.

The qualitative findings also show how students were positioned and positioned themselves in diverse ways in the host community in relation to their racial, ethnic, cultural, and linguistic identities. For example, learners who felt Othered in the host community when they were stared at and experienced catcalls on the streets (Kim) or robbed and criticized for preparing “smelly” Mexican food (Eddie) ultimately retreated from interactions with Argentines and developed negative evaluations of Argentine culture. With this in mind, SA program stakeholders should provide pre-SA training for students to anticipate the ways they may be positioned abroad, including possible instances of discrimination. For example, programs might provide information regarding challenges that traditionally underrepresented students have faced overseas in the past, as well as opportunities while students are abroad to unpack their SA experiences as they unfold. Moreover, program stakeholders (e.g., host families, local instructors, and program staff) need to be trained in issues of diversity, equity, and inclusion to best support students who might encounter such challenges and avoid decisions that may exacerbate them. As Quan (2018) states, “if the intention is to diversify and be inclusive of all students in the abroad experience, then researchers, educators, and administrators should listen to the experiences of students of color and other underrepresented students (e.g., students with disabilities, LGBTQ students) abroad” (p. 43).

Finally, these results highlight the need to problematize the ways in which the field classifies learners in terms of instruction research. In this study, the focal students’ diverse histories with Spanish demonstrate the insufficiency of traditional labels used to categorize their experiences. What the field has often conceptualized as an L2 learner implies that their exposure to Spanish is limited to the classroom. However, in the United States, Spanish is no longer conceived of as a foreign language that is only spoken in faraway lands. Instead, it is increasingly recognized as a local language that is used throughout the country (see Quan et al., forthcoming). Kim’s case in particular demonstrates the ways in which students’ experiences are becoming more diverse, as they may encounter Spanish in their communities, even if they do not hear it at home. In addition, while HSs are often assumed to be exposed to the language in the home, Eddie’s case shows that while he did not hear or use Spanish at home, he self-identifies as Mexican and as a foreign language learner of Spanish. In fact, only one of our three focal participants seems to reflect what has commonly been conceived of as an L2 learner—a White English-speaking student who studies Spanish as a foreign language and has no prior exposure to the language outside of an academic or immersion context. As such, these labels imposed on language students do not necessarily reflect their lived realities, and the field needs to move beyond imposing this dichotomy on our increasingly diverse students (see Quan et al., forthcoming).