THE GROWTH OF HERITAGE LANGUAGE EDUCATION IN THE UNITED STATES

Although English has been the dominant language in U.S. public education, there is a long history of private and parochial schools using other languages, such as German, French, and Spanish, as the medium of instruction (Schmid, Reference Schmid2001). Community-based schools providing Chinese language instruction to Chinese immigrants and their children date to the 1800s (McGinnis, Reference McGinnis, Brinton, Kagan and Bauckus2008; Wang, Reference Wang and Hornberger2008), and other non-English languages have been used in schools run by religious institutions and local organizations for centuries (Fishman, Reference Fishman, Peyton, Ranard and McGinnis2001). In some parts of the country, Spanish-language courses specifically for home speakers of Spanish have been offered at the secondary and postsecondary level at least since the 1970s (Valdés, Reference Valdés, Valdés, Lozano and García-Moya1981). It is only since the 1990s, however, that the heritage language education has come into its own as a distinct subfield of applied linguistics and language pedagogy (Valdés, Reference Valdés2005).

In the United States, the recent growth of, and augmented interest in, heritage language education can be attributed to several factors. For one, increased immigration has led to a rise in the percentage of residents who speak languages other than English at home (Ryan, Reference Ryan2013). While there is still significant societal pressure on minority language speakers to assimilate to English monolingualism, there are also increased recognition and codification of linguistic rights (Lee & Wright, Reference Lee and Wright2014), an increased interest among second-generation Americans in maintaining and developing their heritage languages (Van Deusen-Scholl, Reference Van Deusen-Scholl2003), and a simultaneous trend toward valuing bilingualism in English and another language (King, Reference King, Lacorte and Leeman2009). Second, since the first Gulf War in 1991, new federal initiatives have given more prominence to developing advanced language proficiency in “critical languages” (Bale, Reference Bale2014). Together, these trends have contributed to a broadening of postsecondary language offerings to include less commonly taught languages such as Arabic, Chinese, Japanese, and Korean, which have seen large enrollment gains over the past few decades (Modern Language Association, 2013). From the top down, the sense that developing the proficiency of heritage speakers is the most efficient means to meet the nation's perceived language needs (Peyton et al., Reference Peyton, Ranard, McGinnis, Peyton, Ranard and McGinnis2001) has led to federal support for heritage language education and research initiatives, while from the bottom up, the large percentages of heritage language speakers in many less commonly taught language courses has led language educators to seek pedagogical materials and practices better suited for such learners. Another factor worth noting is the so-called multilingual turn (May, Reference May2013) in international applied linguistics, with its accompanying rejection of the assumed monolingual language learner.

IDENTITY AND LANGUAGE LEARNING

In addition to the growing interest in heritage language education, since the mid-1990s there has also been a blossoming of the study of identity in language learning, thanks in part to a broader social turn in applied linguistics theory and research, and the concomitant increased interest in the sociocultural context in which language learning and use take place (Block, Reference Block2009; Canagarajah, Reference Canagarajah, Norton and Toohey2004; Kubota & Lin, Reference Kubota and Lin2009; Norton, Reference Norton Peirce1995, Reference Norton2000, Reference Norton and Chapelle2012; Menard-Warwick, Reference Menard-Warwick2005). In contrast with earlier essentialist views that conceived of identity as a static entity that people have, contemporary social constructivist accounts emphasize that the people's sense of themselves and of their relationship to the world is shifting and multiple. Identities are not fixed within the individual but instead are shaped and constrained by the macro- and micro-level sociohistorical contexts, including societal ideologies, power relations, and institutional policies. Language learning is a particularly fruitful arena in which to study identity because identities are instantiated in discourse, and learning a new language involves taking on new ways of being (Canagarajah, Reference Canagarajah, Norton and Toohey2004; Kanno & Norton, Reference Kanno and Norton2003; Norton, Reference Norton Peirce1995, Reference Norton2000). However, until recently the study of language learning and identity focused on second language learning, and it is only in the past decade that scholars have begun to expand this line of inquiry to heritage language education (e.g., Abdi, Reference Abdi2011; Blackledge et al., Reference Blackledge, Creese, Baraç, Bhatt, Hamid, Wei and Yağcioğlu2008; Creese, Bhatt, Bhojani, & Martin, Reference Creese, Bhatt, Bhojani and Martin2006; He, Reference He2006; Hornberger & Wang, Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008).

While the explicit focus on identity in heritage language education research is relatively recent, heritage language educational policy and pedagogy have always been intricately tied up with cultural identity and identity politics (Leeman, Rabin, & Román-Mendoza, Reference Leeman, Rabin and Román-Mendoza2011a). In the United States, secondary and postsecondary heritage language courses first appeared in Spanish, the non-English language most commonly spoken in the United States as well as the most commonly studied foreign language, in the context of the Civil Rights movements of the 1960s and 1970s. The struggles for Chicana/o and Puerto Rican rights were accompanied by demands for schools and universities to offer courses and curricula that reflected the experiences and specific educational needs of Latina/o students. Because Spanish was seen as central to Latina/o experiences and political activism (Limón, Reference Limón, Amastae and Elias-Olivares1982), Spanish-language courses specifically for home speakers of Spanish were framed as one element of a more inclusive and culturally responsive education, and the development of Spanish proficiency was viewed as a way to strengthen heritage speakers’ connections to their own communities (Leeman & Martínez, Reference Leeman and Martínez2007). Similarly, community-based schools have typically sought to promote heritage language culture and traditions—such as calligraphy, traditional dancing, or martial arts in the case of Chinese—as well as language (Lee & Wright, Reference Lee and Wright2014). Nonetheless, although identity has long been central in heritage language educational discourse, it is only recently that researchers have begun to investigate heritage learners’ sense of themselves, as well as how identities are constructed, performed, and represented in heritage language educational contexts.

CLARIFYING THE TERMS

Whereas earlier pedagogical programs for similar populations had made reference to language education for “bilinguals” or “native speakers” (or had not labeled students at all, as in the case of many complementary schools), the establishment of heritage language education as a field has gone hand in hand with the emergence of the new label and category “heritage language learner” (as well as heritage language learner of a particular language such as “heritage language learner of Korean,” and “heritage language learner of Spanish”) (Doerr & Lee, Reference Doerr and Lee2013; Valdés, Reference Valdés2005). This new construct has been supported by the tremendous growth of heritage language education, the adoption of the term in academic and policy discourse, and the intensive marketing of distinct pedagogical materials to address heritage language learners’ unique needs, particularly in the case of Spanish. However, as has been widely noted, there is no universal understanding of just what the terms “heritage language” and “heritage language learner” mean (e.g., Bale, Reference Bale2010; Carreira, Reference Carreira2004; Hornberger, Reference Hornberger2005; Hornberger & Wang, Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008; Polinsky & Kagan, Reference Polinsky and Kagan2007; Van Deusen-Scholl, Reference Van Deusen-Scholl2003).

Different definitions are based in part on whether the primary focus is on the languages, their societal status, or individuals’ linguistic proficiency, although of course these are not completely independent, as societal status has an impact on language maintenance or shift. Researchers adapting the original Canadian usage by defining heritage languages as “languages other than the national language(s)” (Duff & Li, Reference Duff and Li2009) are generally more attuned to the sociopolitical status of a given language or to the collective rights and needs of the speakers of that language as a group. It is in this research and policy context where there are also ongoing debates about whether terms such as “ancestral language,” “minority language” or “community language” are equivalent or preferable to “heritage language” (e.g., Brutt-Griffler & Makoni, Reference Brutt-Griffler and Makoni2005; De Bot & Gorter, Reference De Bot and Gorter2005; Wiley, Reference Wiley2005).

On the other hand, researchers focusing more specifically on educational policy and curricular design tend to give greater weight to linguistic proficiency and cultural connections in their discussions of who should be classified as a heritage language speaker: either all individuals with an ancestral or family tie to the language—even if they have extremely limited or no proficiency in the language—or just those who have some productive and/or receptive ability. Implicit in the construct of heritage language speaker is the notion that the individual's heritage language proficiency developed prior to her/his exposure to the national language, although this is not always technically the case. Thus, while the emphasis is on the individual speaker, language status generally is also implicated in proficiency-based definitions.

Since the 1990s an additional meaning of “heritage language education” has become widespread particularly (but not exclusively) in the United States, where it is used to refer to language instruction designed specifically for students who have prior home or community-based exposure to this language (Valdés, Reference Valdés2005). In this usage, which is the focus of this article, heritage language education is a specific type of minority language instruction that happens at the margins of regular primary or secondary schooling in the national language(s), rather than all-day educational programs that use a minority language as the medium of instruction for a broad range of subjects (Leeman & King, Reference Leeman, King, Bigelow and Ennser-Kananen2015). This definition of heritage language education includes special sections of foreign language classes meeting for a few hours during the week in secondary schools or universities, as well as instructional programs at complementary schools, which generally hold class on weekends or after regular day school, and are typically run by community groups, cultural associations, churches, or other nonprofits. Generally excluded in this usage are bilingual education, dual immersion, and international schools, which typically offer full-time schooling to the children of expatriates and other elites, although the vast diversity of educational programs and contexts makes classifications extremely complex. The fuzziness of these boundaries is evident, for example, in the case of hoshuko, complementary schools that teach the language arts component of the Japanese government–sanctioned national curriculum to Japanese residents living abroad as well as to U.S.-born Japanese Americans (Doerr & Lee, Reference Doerr and Lee2012).

“HERITAGE LANGUAGE LEARNER” AS AN IDENTITY

The discussions about what counts as a heritage language or as heritage language education, and about who is—and who isn't—a heritage language learner are more than just terminological debates. First, “heritage language learner” is not simply an educational classification but also an identity, one constructed largely by researchers, educators, and administrators and assigned to a group of students, rather than by heritage language learners themselves (Doerr & Lee, Reference Doerr and Lee2013; Hornberger & Wang, Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008; Wiley Reference Wiley2005). As Hornberger and Wang (Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008) noted, the debates about whether the definition of heritage language learner should be based on linguistic proficiency, familial ties, or ancestry fail to consider how learners see themselves or their relationship with the language in question. Further, these debates do not take the educational or policy context into account, nor do they acknowledge individuals’ agentive role in constructing their own identities, which is central to current understandings of identity. This is especially important because many putative heritage language speakers resist being categorized as such (Dressler, Reference Dressler2010; Cho Reference Cho2014). Hornberger and Wang (Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008) addressed these issues in their ecological definition of heritage language learners as “individuals with familial or ancestral ties to a language other than English who exert their agency in determining if they are heritage language learners of that language” (p. 6).

The competing understandings of whom to include under the heritage language learner rubric are a stark reminder that identities are not fixed categories with objective boundaries but socially constructed groupings that are contextually determined, as are the meanings of different labels. They also illustrate the contested nature of membership claims, including conflicts between ascribed and claimed identities, and struggles for legitimacy and authentication (see Bailey, Reference Bailey2000; Bucholtz & Hall, Reference Bucholtz, Hall and Duranti2004; Shenk, Reference Shenk2007). Labels have an impact on how referents are understood and experienced, and they are one way in which identities are discursively produced (Eckert & McConnell-Ginet, Reference Eckert, McConnell-Ginet, Hall and Bucholtz1995; Modan, Reference Modan2001). Because the choice of terms has implications both for how heritage language speakers are perceived by others and for how they perceive themselves, the association of the term heritage with the past has led some researchers to raise concerns that it positions non-English languages as historical relics (e.g., García, Reference García2005). García's (Reference García2005) critique of the shift in U.S. educational policy from bilingual education to heritage language programs stresses the backward-looking connotations of the terminological shift, lamenting that “our multiple identities have been silenced, with one identity reduced to heritage” (p. 605).

HERITAGE LANGUAGE LEARNERS’ PERSPECTIVES: SURVEY-BASED RESEARCH

The notion that ethnic or national identity is embodied in the heritage language is taken for granted in many models of heritage language education (Francis, Archer, & Mau, Reference Francis, Archer and Mau2009), as well as in assumptions that learners enroll in heritage language classes in order to (re)claim their ethnic identity or because they are motivated by “an identification with the intrinsic cultural, affective, and aesthetic values of the language” (He, Reference He2006, p. 2). More recently, a parallel discourse regarding the instrumental value of heritage language development has also taken hold in the United States as language knowledge is increasingly perceived as a valuable job skill, and a primary goal of education has come to be seen as preparing students for the job market (Leeman, Reference Leeman2007; Leeman & Martínez, Reference Leeman and Martínez2007). Heritage languages are also portrayed as valuable for U.S. national interests understood as security, defense, and international competitiveness (Ricento, Reference Ricento2005).

Given the prevalence of these competing discourses, as well as the common essentialist understanding of the relationship of language to culture among heritage language researchers and educators, it is not surprising that studies of heritage language education have considered the impact of identity on student motivation, even when identity was not the primary research focus. One such study is Carreira and Kagan's (Reference Carreira and Kagan2011) report on the National Heritage Language Survey (NHLS) conducted by the federally funded National Heritage Language Resource Center. The online survey gathered data on the language history, use, proficiency, and attitudes from 1,732 students enrolled in university-level heritage language courses in 22 different languages. In response to a multiple-choice question about their reasons for enrolling in heritage language courses, the most common responses were “to learn about their cultural and linguistic roots” (p. 48), although speakers of Cantonese, Japanese, Mandarin, and Spanish gave priority to professional goals. The responses to open-ended questions included the ability to be a go-between for limited English-proficient speakers of Cantonese and the ability to participate more fully in a local Ethiopian church for a student of Amheric, highlighting that the diverse ways that respondents understand the value of their heritage language proficiency go beyond the narrow purposes that flavor so much heritage language education discourse.

In another survey-based study investigating student perspectives, Beaudrie, Ducar, and Relaño-Pastor (Reference Beaudrie, Ducar and Relaño-Pastor2009) surveyed 152 students enrolled in Spanish heritage language classes at a Southwestern university. A combination of closed and open-ended questions were used in order to assess how students perceived their own identity, whether they felt that their Spanish heritage language courses helped them learn about “their own culture” and the culture of other Latina/o and non-Latina/o groups in the United States, as well as the extent to which they felt the program offered opportunities for them to connect with local Spanish-speaking communities. Although many students seemed to presuppose an inherent connection between heritage languages and ethnic identity, as well as to perceive Latina/o cultures as delimitable along Latin American national boundaries, at least one student comment pointed in the opposite direction. In contrast with the essentialist discourse that frames heritage language proficiency as a prerequisite of heritage language identity, one student wrote, “This class has made me more aware of how many Mexican-Americans experience the Spanish language and made me feel as though I have rights to my Mexican-American identity despite my trouble with the language” (p. 167).

Large-scale surveys such as these are an efficient means for providing a sense of student demographics, experiences, and opinions, all of which can prove extremely useful for program design and administration. However, designing survey questions that reflect heritage language learners’ sense of their own identity is extremely challenging, if not impossible. The NHLS self-identification question was open-ended, with “American, Vietnamese-American, Vietnamese, Asian, Asian-American, etc.” provided as sample responses, while Beaudrie et al. (Reference Beaudrie, Ducar and Relaño-Pastor2009) asked respondents to choose among Hispano, Latino, American, and Other. The results are difficult to interpret, due to the multiple and contested valences of these category labels (Alcoff, Reference Alcoff2005), as well as the fact that the meaning of a response option is impacted by the implied contrast with other choices. In addition, recent research on ethnoracial self-identification in the U.S. Census suggests that, like interactional stance-taking and positioning, choosing among various identity labels can be strategic (Dowling, Reference Dowling2014); as one respondent in the NHLS put it, “[my identity] depends on who I'm talking to . . . and what my agenda with that person is” (National Heritage Language Resource Center, 2013).

HERITAGE LANGUAGE LEARNERS’ INVESTMENT AND IDENTITY: ETHNOGRAPHIC INTERVIEWS

Social constructivist approaches to identity generally go hand in hand with qualitative and ethnographic research methods, as they allow for a more nuanced and less essentialized understanding of heritage learners’ identities (He, Reference He2010). Several recent studies have conducted ethnographic interviews with heritage learners and analyzed them utilizing Norton's (Reference Norton Peirce1995, Reference Norton and Chapelle2012) framework of investment and imagined community (see also Kanno & Norton, Reference Kanno and Norton2003). Investment refers to the extent to which learners see the language of study as able to provide material and symbolic resources or cultural capital, which Norton stressed is shaped by power relations and unequal status in different learning contexts. The notion of imagined community recognizes the impact on language learning of learners’ desire to gain or strengthen their membership in social groups associated with the language, whether these be ethnic, national, or classroom communities. Thus, investment and imagined community recognize the identities that learners aspire to, as well as their current sense of themselves.

Two recent studies utilized Norton's framework to report on interviews with heritage language learners of Chinese: Wong and Xiao (Reference Wong and Xiao2010) interviewed 64 university-level students, while Wu, Lee, and Leung (Reference Wu, Lee and Leung2014) interviewed 14 students at a charter middle school for Asian American students as part of a larger ethnographic project. In the United States, school- and university-based Chinese programs generally teach Mandarin, the official language of China, whereas many Chinese and Chinese American communities speak Cantonese or another of the mutually unintelligible languages labeled with the umbrella term Chinese (He, Reference He2006; Lawton & Logio, Reference Lawton and Logio2009; Wu et al., Reference Wu, Lee and Leung2014). The mismatch between students’ actual home language and their supposed heritage language had an impact on their academic success, with heritage speakers of non-Mandarin languages struggling to keep up with their peers from Mandarin-speaking homes. So too, speakers of Cantonese and other Chinese languages showed less investment in Mandarin, and thus less interest in continuing to study it, particularly in the case of the middle school learners. In contrast with their lack of investment in Mandarin, some heritage speakers of Cantonese expressed a desire to expand their proficiency in Cantonese, in order to interact with local family or other members of their social networks, underscoring both that identity concerns were highly relevant for these heritage language students and that the officially constructed Chinese did not meet their needs. Similarly, while the university-level heritage language speakers of Mandarin saw the study of Mandarin as a way to strengthen their integration in an imagined Chinese community, speakers of other Chinese languages generally did not report an affective or identity-based connection to Mandarin. Still, this does not mean that identity was irrelevant: Whereas some focused on the commodified value of Mandarin in the global economy, others saw the acquisition of Mandarin as a way to position themselves as more international or cosmopolitan.

This research clearly shows that the construct Chinese is inadequate for understanding the experiences of heritage language speakers and their relationship to the language of study. Similarly, constructs such as Spanish or Korean are also insufficient for describing the linguistic repertoires and language practices of heritage language speakers (and others). On one hand, such labels erase the geographic, social, and stylistic variation that they encompass. On the other, languages are not objectively bounded entities but ideological artifacts that are tied up with projects of nationalism and colonialism (Makoni & Pennycook, Reference Makoni and Pennycook2005). Further, heritage language speakers’ draw from language varieties, styles, and registers distributed across separate languages, a phenomenon known as “translanguaging” (García & Wei, Reference García and Wei2013). Of course, varieties, styles, and registers are also constructs, ones that gain social meaning when they are noticed and enregistered by speakers (Agha, Reference Agha2005; Johnstone, Reference Johnstone2011). Monolingual and multilingual speakers alike utilize diverse ways of speaking (including specific languages, words, pronunciations, and styles) to signal particular stances or values that in turn index particular social identities (Kiesling, Reference Kiesling2001; Mendoza-Denton, Reference Mendoza-Denton2008; Modan Reference Modan2008).

Recent research on heritage language learner investment and identity illustrates the need to consider not only the imagined languages but also particular varieties, styles, and translingual practices and the ways that these are linked to the imagined communities to which students aspire. This can be seen in Park's (Reference Park and Davis2011) study, in which she interviewed three Korean American students enrolled in a university-based 4th-year Korean heritage language class. Park used narrative analysis to shed light on how these women experienced their identities in relation to their use of Korean outside of the classroom in local Korean immigrant churches and with peers. On one hand, they sometimes reflected ideologies that linked Korean identity to linguistic proficiency, as can be seen in one participant's expressed sense of loss regarding her “Koreanness” when her move to another city resulted in fewer opportunities to speak Korean. On the other hand, the narratives also showed that the construct of “Korean” was overly broad for understanding heritage language learners’ investment; another participant recounted her desire to speak to, and to speak like, a group of bilingual native Korean-speaking youth as well as her reluctance to speak to older Korean immigrants with whom she did not identify.

In Park's (Reference Park and Davis2011) study, the subjectivities expressed by participants varied not just across participants but also within individuals, illustrating the shifting and contextual nature of identities. Thus, these different alignments were not simply a case of some participants identifying more strongly as Koreans and others emphasizing a Korean American or second-generation identity. Instead, participants aligned themselves with different identities in different moments of their narratives. For example, one woman, Taehi, at one point claimed a Korean identity but then distanced herself from what she saw as overly hierarchical Korean linguistic and cultural practices, such as the use of honorifics, and aligned herself with the children of Korean immigrants who understood the contrast between U.S. and Korean parenting styles.

Although speaker agency plays a key role in the construction and performance of identity, identity claims and performances are also constrained by the identities ascribed by others (Bailey, Reference Bailey2000; Brodkin, Reference Brodkin1998; Rampton Reference Rampton1995). The tension between claimed and ascribed identities can take many forms, but one way that people question or deauthenticate identity claims is via reference to language, and a speaker's failure to use the “right” kind of language for the identity they are claiming (Pavlenko & Blackledge, Reference Pavlenko and Blackledge2004; Shenk Reference Shenk2007). The lack of legitimacy afforded to heritage speakers is reflected in the narratives of participants in Park's (Reference Park and Davis2011) study, who reported feeling marginalized by first-generation Korean immigrants such as the minister at a Korean church and the principal of the Korean heritage school where they taught. In these narratives, ethnic identity was inseparable from the demonstration of native-like linguistic behavior, with the heritage language speakers often judged as not fully competent. One participant commented on her frustration at being labeled “second-generation,” as she saw this label as challenging both her linguistic ability and the legitimacy of her Korean identity. The fact that participants also used this term to self-identify is a reminder that the meaning of identity labels is also fluid and contextual.

The legitimacy of different types of heritage language knowledge is a central theme in Doerr and Lee's (Reference Doerr and Lee2009, Reference Doerr and Lee2010, Reference Doerr and Lee2012, Reference Doerr and Lee2013) ethnographic research in a Japanese-language weekend school. The school is one of 78 community-based complementary schools in the United States that offer the same Japanese government–sanctioned language arts curriculum as schools in Japan. Their curriculum was originally targeted to Japanese nationals temporarily in the United States, with the goal of facilitating reintegration upon return, and depending on their size, such schools receive support from the Japanese government. Since the early 2000s, many of these Japanese-language complementary schools, including the one where Doerr and Lee conducted their research, have added a heritage learner track. In this context, heritage language learners are implicitly contrasted with the Japanese students in the original track, which highlights how institutional practices and discourses come together to construct “heritage language learner” as an identity as well as an educational category. Still, while many of the American-born heritage learner students at the school distinguished themselves from the Japanese students in the other track, they simultaneously claimed Japanese identity, underscoring the problematicity of binary identity categories.

The second track was designed for longer-term residents, immigrants to the United States and U.S.-born Japanese Americans, and Doerr and Lee's (Reference Doerr and Lee2009, Reference Doerr and Lee2012, Reference Doerr and Lee2013) interviews with administrators, teachers, parents, and students revealed that differences between the tracks were discursively linked to imagined constellations of national origin, parental ethnicity, language proficiency, and learning style. However, students’ actual biographies and subjectivities were often more complex and less easily categorized. Some parents considered the native speaker track as the only legitimate way to receive a Japanese education and to maintain a Japanese identity, regardless of place of birth or ethnic identity. These questions of legitimacy were echoed in interviews with students about their understanding of their identities and of the differences between the two tracks, which did not always correspond to the factors typically referenced in definitions of heritage speakers (i.e., place of birth, parental ethnicity, language proficiency, or time in the United States). Various students in both tracks framed the heritage track as being easy, for dropouts, or for students with disabilities. For this reason, some U.S.-born students who began in the native speaker track initially resisted moving to the heritage track despite experiencing difficulty keeping up. Doerr and Lee (Reference Doerr and Lee2012) described one student who came to modify her opinion of the heritage language track and another who eventually switched back to the native speaker track. Students’ different understandings of the two tracks, as well as their movement back and forth between them, demonstrate the two-way relationship between students’ subjectivities and their participation in heritage language education, as well as the instability of student identities over time.

The shifting points of contrast and differentiation in constructing identities are also salient in Kang's (Reference Kang2013) study of eight second-generation Korean Americans in a beginning level Korean as a heritage language class. Participants in that study perceived themselves as different from both Koreans and “mainstream Americans,” positioning themselves as somewhere in between. Like participants in Park's (Reference Park and Davis2011) research several of these learners defined Korean American identity in opposition to Korean identity as largely linked to a lack of proficiency in Korean, while also distancing themselves from what they saw as conservative Korean culture. Although Kang emphasized the hybridity of the participants’ identities and the third space they defined for themselves, at least some of the interview data suggest that participants framed “Korean” and “American” as rigid, well-delineated quantifiable categories even when discussed their identity as a combination of the two. For example, one participant described herself as “70% . . . American and 30% . . . Korean” (p. 257), reflecting an ideology that “in-between” identities can be neatly arranged on a continuum between two mutually exclusive poles, with different elements corresponding to one category or the other.

CONSTRUCTING AND INDEXING IDENTITIES IN HERITAGE LANGUAGE CLASSROOMS

This section focuses on the analysis of the linguistic forms and practices that heritage language learners use to perform, index, and negotiate identities in heritage language classrooms. Many of the ethnographic interviews discussed in the previous section also employed observation in classroom settings, although not all of them report on these data or include linguistic analyses. For example, in addition to conducting and analyzing interviews with heritage language learners of Korean as described above, Kang (Reference Kang2013) collected classroom data and analyzed students’ use of Korean and English, as well as how teachers and students utilized pronouns to position students in particular ways. One challenge of such analyses of code-switching is that they involve projecting intentions onto speakers’ behavior—such as assuming that shifts to English reflect gaps in Korean proficiency. Similarly, whereas some students' use of “we” and “they” clearly shows that these students position themselves in opposition to others it is impossible to know just who “they” refers to, or how that group is defined and constructed when there isn't an explicit reference to the group.

Schiefflin and Och's language socialization framework (Schieffelin & Ochs, Reference Schieffelin and Ochs1986; Ochs, Reference Ochs1993), which examines how individuals are socialized into a community through language use and how they are socialized to use language in ways that are consistent with community norms, has been productively applied to the analysis of language and identity in second language learning (Duff, Reference Duff2007), but has been relatively rare in heritage language education (but see He, Reference He2004, for an exception). Lo (Reference Lo2009) also adopted a language socialization approach in her recent ethnographic study of a Korean heritage language complementary school, which examined how the elementary-school-aged students learned Korean while simultaneously being socialized into “Korean ways,” including demonstrating respect for elders. In contrast with the common assumption that heritage language identity and cultural values are embodied in the language, and the related notion that the relationship between specific linguistic forms and particular stances is inherent or fixed, Lo found a reworking of the social meaning of linguistic forms. Specifically, teachers were willing to accept heritage learners’ failure to use honorifics, which would have been seen as disrespectful among Korean children. Instead of interpreting their students’ nonhonorific language as rude or as a sign of a less respectful American cultural identity, teachers saw acceptance of non-honorific language simply as a result of the circumstances of their acquisition of Korean. In contrast, teachers were far less indulgent regarding students’ nonlinguistic signs of disrespect, such as body language or failure to make eye contact, demonstrating that it was not simply a case of tolerating students’ bad behavior. This reshaping of teachers’ ideologies illustrates that socialization is not a one-way process in which children take up the linguistic and cultural norms of more competent community members.

Unlike the ethnographic research discussed thus far, which was conducted in complementary schools, Showstack's (Reference Showstack2012) study examined classroom interactions in Spanish as heritage language classes at a public university in Texas. The case of Spanish, and particularly in the Southwest, is unique among heritage languages in the United States in that it has a long-standing historical presence (prior to U.S. independence), and it is by far the most commonly spoken non-English language, making it the unofficial second language of the country (Alonso, Reference Alonso2006). The widespread recognition of Spanish's presence is not matched by acceptance, however. Instead, Spanish is often portrayed negatively and is discursively linked to unauthorized immigration, and it serves as a key ideological symbol of a racialized minority (Hill, Reference Hill1999, Reference Hill2008; Leeman, Reference Leeman and Del Valle2013). Despite the large number of Spanish speakers in the United States, English-only educational policy means that Spanish-medium schooling is rare, and Spanish is instead offered as a foreign language. Thus, Spanish as a heritage language classes are generally offered as alternatives to regular Spanish as a second language classes. As a result, heritage learners of Spanish are implicitly contrasted with the presumed monolingual Anglophone students to whom Spanish-language education has been targeted. Rather than valuing the knowledge that heritage speakers of Spanish bring to the classroom, Spanish instructors and materials have often portrayed the contact varieties and practices typical of many U.S.-born Spanish speakers as deficient and substandard (Hornberger & Wang, Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008; Leeman, Reference Leeman2005; Valdés, Reference Valdés, Valdés, Lozano and García-Moya1981; Villa, Reference Villa2002). Ideologies regarding the superiority of so-called standard and pure varieties in comparison to local varieties of Spanish circulate outside of the academic setting as well (Velázquez, Reference Velázquez2009).

The ideological elevation of the monolingual standard variety is evident in the classroom interactions among Mexican heritage students in Showstack's (Reference Showstack2012) research. The data were drawn from discussions on assigned topics such as bilingual and bicultural identities, and the pros and cons of Spanglish. Showstack's analysis showed that students equated language with ethnic identity and culture in their discursive construction of social categories. In particular, only “authentic bilinguals” were categorized as “authentic cultural hybrids” (p. 16); and while participants did not eschew cultural mixing and hybridity, they framed linguistic mixing more negatively, with several participants disparaging the “Pocho” Spanish spoken by locals with limited Spanish-language education as well as their own “deficient” Spanish.

Helmer's (Reference Helmer, Scherff and Spector2011, Reference Helmer2013) ethnography of a Spanish as heritage language class at a charter high school in the Southwest also documented classroom ideologies that framed Spanish proficiency, as well as native-like accent, as necessary for constructing an authentic “Mexican” persona. However, Helmer's analysis revealed that while those who spoke with an English-inflected accent were ridiculed for “speaking White” (Helmer, Reference Helmer, Scherff and Spector2011, p 144), performance or ascription of a Mexican identity did not always correlate with linguistic ability. Nor did it depend on ethnic or national origin, as Helmer showed in her discussion of a self-identified Mexican boy who spoke fluent Spanish but used the label “Mexicans” to refer to three Native American girls who bullied their classmates, presumably tapping into local and national discourses that link Mexicans to criminality and danger and confirming that ethnic identity labels are not as straightforward as they might appear. Also coexisting with ideologies about the necessity of Spanish for Mexican identity was resistance to Spanish, and particularly “proper Spanish” and Spanish literacy, which students saw as irrelevant in their daily lives and professional futures. In addition, like research on the production of classroom identities such as “good student” in other educational settings (e.g., Pomerantz, Reference Pomerantz2008; Wortham, Reference Wortham2006), Helmer identified a variety of self-positionings not directly related to Mexicanness, ethnicity, or Spanish. For example, one group of students referred to themselves as “the mean girls” and a young woman framed herself as a “bully.” This serves as an important reminder that there are always a multiplicity of intersecting and overlapping discourses and subjectivities circulating in heritage language classrooms.

HERITAGE LANGUAGE PEDAGOGY AND IDENTITY

As noted above, much heritage language education in the United States is at least implicitly linked to ethnocultural identity and/or questions of educational equity. Building on this history while also drawing from critical pedagogy, some heritage language educators have sought to challenge the privileging of an imagined monolingual standard variety and the stigmatization of heritage speakers for their nonstandard or deficient varieties and practices by engaging students in analyses of the relationship of language to identity in discussions of language variation and multilingual practices (see Leeman, Reference Leeman and Lacorte2014). Within heritage language education, such proposals have been most common in Spanish, probably due to the salience of Spanish in the United States, and of U.S.-based varieties and practices, as well as the growth of the fields of Spanish in the United States and Latina/o studies. Nonetheless, there have also been recent calls to incorporate critical perspectives in the teaching of other heritage languages (e.g., Wu et al., Reference Wu, Lee and Leung2014).

A few recent studies have examined the implementation and impact of critical heritage language pedagogies with an eye to questions of learner identity. For example, in Helmer's (Reference Helmer, Scherff and Spector2011, Reference Helmer2013) study, discussed in the previous section, one of the teachers included several elements of critical pedagogical approaches to Spanish as a heritage language. Specifically, she incorporated a mixed language text by Chicana author Gloria Anzaldúa that describes the author's use of “Pachuco Spanish,” which she characterized as a language of rebellion against both standard English and standard Spanish. The (White Anglo) teacher went on to say that some of the students’ own slang had its roots in Pachuco subculture. While the teacher presumably thought she was validating students’ identities by incorporating their language varieties and practices into the classroom, this was not students’ interpretation. Instead, they saw the references to Pachucos as a slur, one that linked them to “gangster” or “ghetto” identities, rather than the imagined Mexican community with which some of them identified (Helmer, Reference Helmer2013). According to Helmer, student resistance and their negative attitude toward the teacher had its roots in her failure to recognize what they felt were their true identities.

Another recent study examining critical pedagogy in Spanish as a heritage language is Leeman et al.'s (Reference Leeman, Rabin and Román-Mendoza2011a, Reference Leeman, Rabin and Román-Mendoza2011b) investigation of a critical service-learning internship program in which college students of Spanish (primarily heritage learners) ran an afterschool Spanish heritage language program at a local elementary school. One of the motivations for the program was that while the college students gained confidence regarding their Spanish within their heritage language class, they reported feeling linguistic insecurity in other Spanish classes and in the world outside the university. Faculty mentors sought to create opportunities for them to gain a sense of themselves as legitimate speakers of Spanish and to develop identities as language experts. Written reflections and interviews conducted at the end of the program showed that the student teachers did gain confidence as well as a greater sense of themselves as part of an imagined community of Spanish speakers. But while the program was successful in this regard, and despite numerous readings and online discussions critically examining monoglossic language ideologies, some participants’ newfound confidence seemed linked to repositioning themselves as native speakers rather than validating local varieties and practices.

The bilingual Korean, Chinese, and Samoan immigrants in Cho's (Reference Cho2014) study were undergraduates teaching in complementary heritage language programs while enrolled in a teacher preparation course in which critical language awareness was a major component. Cho, who was the teacher of the course, engaged students in discussions of the constructs of native and nonnative speaker, power differentials between languages, and the relationship of heritage languages to identity and culture. Her analysis of classroom interactions and student narratives revealed that some program participants subscribed to the essentialism equating language and ethnic or national identity, at least at first. For example, one Korean woman expressed surprise at how “American” her heritage language students were, having assumed they would exhibit the classroom cultural norms of overseas Koreans, while another insisted on the importance of teaching Korean culture and national symbols such as the national anthem and the flag, in order to help them accept their Korean identities (despite students’ negation of this identity). Like Leeman et al.'s (Reference Leeman, Rabin and Román-Mendoza2011a, Reference Leeman, Rabin and Román-Mendoza2011b) research, Cho's study highlights the challenges of destabilizing teachers’ ideologies regarding heritage language learner identities and language.