CORRELATIONAL STUDIES OF SOCIOCULTURAL DIMENSIONS OF HERITAGE LANGUAGES

Rooted in essentialist paradigms and commonly adopted in developmental psycholinguistics and sociolinguistics, a correlational approach assumes sociocultural parameters as a priori givens and takes for granted these parameters as independent variables. It considers sociocultural parameters as having a sociohistorical reality that is independent of language behavior. It assumes that sociocultural traits are persistent and consistent across times and situations. It asks how an HL learner with specific, static traits behaves in the process of HL development. For example, do members of specific socioeconomic, ethnic, gender, generational or linguistic groups maintain their HLs, and if so, how? What is the relationship between phonological development and the learner's place of birth? What is the relationship between motivation and success in HL learning? In this framework, the researcher treats sociocultural variables as explanations for HL variation. Among the most common means of data collection are various measures of language proficiency (listening, speaking, reading, and writing), tests of academic performance, quantitative observations of language use, questionnaires, and attitude or motivation scales.

In one of the first few full-length books dedicated to one single HL population in a given setting, Gibbons and Ramirez (Reference Gibbons and Ramirez2004) examined the simultaneous development of English and Spanish as an HL in a given speech community in Sydney, Australia. The study investigated a set of variables that contribute to varying degrees of proficiency—societal factors, interpersonal contact, education, media and literacy practice, and attitudes and beliefs. It considered the impact of macro-level societal infrastructures and institutions on the development and maintenance of bilingualism and biliteracy. It listed societal variables including political history, geography, demography, status, institutions and media, and the many subcategories within each of these, and then examined each of them one by one for both English and Spanish. The authors correlated the sociodemographic information of the Hispanic teenager population under study with their levels of language proficiency, concluding that there was little correlation between social class and language achievement.

Gibbons and Ramirez (Reference Gibbons and Ramirez2004) also evaluated the impact of social practices and resources on spoken language development by correlating various categories of social network ties, the strength of the contact, and the mode of the contact (whether face to face or distant) with learners’ oral proficiency, grammar, vocabulary, and overall confidence in using the language. By the same token, this study examined attitudes and beliefs that support or undermine the maintenance of minority languages. These attitudes and beliefs included aesthetic and social judgments of a particular language or variety, views concerning bilingualism and biculturalism, projections of the vitality of the minority language, and positioning of the learner vis-à-vis different cultures and languages, as well as beliefs about language proficiency. The researchers used a combination of open-ended and close-ended questions to provide a profile of the attitudes of their subjects and to identify attitude clusters and performed statistical correlations with the proficiency measures.

Similarly, Jia (Reference Jia, He and Xiao2008) reported findings from a study on first-generation Chinese immigrants in New York City in terms of their speaking, reading, and writing skills for each 2-year interval of their residence in the United States. A grammaticality judgment task in Mandarin was also used to measure participants’ sensitivity to Mandarin grammar. The findings showed that, in certain contexts, at the same time as the exposure to English and the English skills grew steadily, HL skills continuously declined over the years. Jia concluded that age of immigration, social economic status, and self-reported Chinese cultural identity are major variables that correlate with current proficiency in Chinese as a heritage language.

Treating learners’ language status (HL learner versus foreign language learner) as an independent variable and adopting both proficiency tests and self-assessment measures, Kondo-Brown (Reference Kondo-Brown2005) investigated (a) whether Japanese as a heritage language learners would demonstrate language behaviors distinctively different from those of traditional Japanese as a foreign language learners, and (b) which domains of language use and skills would specifically exhibit such differentiation. Kondo-Brown's findings suggested that there were striking similarities between the learner group and students with at least one Japanese-speaking grandparent, but without a Japanese-speaking parent, and students of Japanese descent without either a Japanese-speaking parent or grandparent. In contrast, students with at least one Japanese-speaking parent proved to be substantially different from other groups in (a) grammatical knowledge, (b) listening and reading skills, (c) self-assessed use or choice of Japanese, and (d) self-ratings of a number of can-do tasks that represented a wide range of abilities. Similarly, Kaufman (Reference Kaufman, Ravid and Bat-Zeev Shyldkrot2005) took HL and native language as independent variables, investigated narratives produced by speakers of Hebrew as an HL and compared them to monolingual norms. The heritage narrative data showed considerable fragmentation in all aspects of the language. In comparison with monolingual native speakers, the heritage learners were lacking in communicative fluency, grammatical accuracy, and lexical specificity as evidenced in their use of developmental forms characterized by present-tense temporal anchoring, frequent pauses, false starts, repairs, lexical substitution, simplification, redundancy, and circumlocution.

Using place of birth as well as age as independent variables, Jia and Bayley (Reference Jia, Bayley, He and Xiao2008) investigated the (re)acquisition of the Mandarin Chinese perfective aspectual marker -le by 36 children and adolescents who either initially acquired Mandarin as a first language or were acquiring it as an HL. The results of several different measures indicate that, as expected, participants who were born in China outperformed their U.S. born counterparts, as did participants who reported using primarily Mandarin at home. Results for age show a more complicated picture, with younger speakers outperforming older speakers on a narrative retelling task, but older speakers outperforming younger speakers on cloze and sentence completion tasks.

Rothman (Reference Rothman2007) compared Brazilian Portuguese heritage speakers’ knowledge of inflected infinitives to advanced adult L2 learners and educated native controls. Unlike the latter groups, heritage speakers, who did not have formal education in the standard dialect, were shown to not have target knowledge of inflected infinitives. Rothman concluded that, whether it is the case of attrition or that of incomplete acquisition of the HL, literacy plays a significant role in the acquisition of this grammatical property in Brazilian Portuguese. Also conceptualizing literacy activities and print materials as crucial variables for HL growth, Koda, Zhang, and Yang (Reference Koda, Zhang, Yang, He and Xiao2008) addressed literacy development in Chinese as a heritage language among school-age students. These children typically use Chinese at home, receive primary literacy instruction in English at school, and pursue ancillary literacy in Chinese in a weekend school. As such, their primary literacy tends to build on underdeveloped oral proficiency, and secondary literacy reflects heavily restricted print input and experience. Hence, their literacy learning in both languages lacks sufficient linguistic resources. Despite these inadequacies, however, many children succeed in their primary literacy, and some even in HL literacy. Based on theories of cross-language transfer, reading universals, and metalinguistic awareness, their study explored what additional resources—metalinguistic and cognitive—are available to these children, and how such resources correlate with children's level of literacy.

Many researchers have correlated HL achievement with learner identity formation or transformation. Li (Reference Li1994) posited that proficiency correlates positively with a well-developed sense of ethnic identity and network with their ethnic group, such that group members have a greater understanding and knowledge of their groups’ cultural values, ethics, and manners. The same is echoed in Abdalla (Reference Abdalla2009), Bhatt (Reference Bhatt2009), Carreira (Reference Carreira2009), Chinen and Tucker (Reference Chinen and Tucker2005), Cho (Reference Cho2000), Kaufman (Reference Kaufman, Ravid and Bat-Zeev Shyldkrot2005), Kondo-Brown (Reference Kondo-Brown2005), and Lee (Reference Lee2002), all of whom suggested that in addition to internal factors such as attitudes, motivation, and social identity, ethnic identity is also a key factor in HL development. Tse (Reference Tse2000) attempted to explain the relationships among ethnic identity, attitudes and motivation, and HL development. She examined published narratives of Americans of Asian descent to discover whether feelings of ethnic ambivalence or evasion extend to the HL, and if so, how they affect language beliefs and behaviors. The results suggested that for many, the HL is closely associated with the ethnic group so that attitudes toward the ethnic group and its language speakers also extend to the narrators’ own language ability and their interest (or lack of interest) in maintaining and developing their HL. Tse concluded that language acquisition is facilitated when an individual has positive attitudes toward the language and feels positively about her ethnic group.

Bermel and Kagan (Reference Bermel, Kagan, Kagan and Rifkin2000) examined correlations between variables such as the number of years the learners have spent in the United States, the number of years of schooling in Russia or the former Soviet Union, and learners’ geographical background (whether originally from Armenia, Azerbaijan, Belarus, Georgia, Ukraine, Russia, etc.) with their self-reported Russian abilities in speaking, listening, reading, and writing as well as their received grades as a measure of command of written language. Although they were not able to identify a significant correlation, they speculated that émigré Russians’ self-reported social status (whether their families belonged to the class of so-called intelligentsia and whether their parents held respectable professions in the former Soviet Union) may correlate with retention of greater speaking capacities in Russian.

Lu and Li (Reference Lu, Li, He and Xiao2008) correlated different motivational factors (integrative, instrumental, and situational) and heritage and nonheritage college students’ Chinese learning in mixed classrooms. Their results indicate that both integrative and instrumental motivations are important to students’ self-confidence in their language proficiency, but integrative motivation is more important to students’ overall tests scores. Further, students are more influenced by instrumental motivation than nonheritage students but less influenced by situational factors (such as teacher effect and effect of mixed-classes).

Correlational research has long been one of the major approaches to second language research. It is no surprise that in researching the sociocultural factors in HL learning, this is a commonly adopted approach as well. Whether it is multivariate studies that examine relationships among a number of variables or bivariate studies that examine associations between two variables, collectively, these studies alert us to the paramount associations among the sociocultural variables that are important to HL learning and suggest general tendencies at certain given point in time. However, research from this perspective tends to rely on average frequencies or probabilities of usage and has yet to explain why a particular variable is associated with another. It tends to evaluate complex and evolving constructs such as motivation, attitude, ethnic identity, proficiency, and literacy in terms of numerical values and leads one to think that these sociocultural traits are essential, built-in, and unchanging qualities.

SOCIAL CONSTRUCTIVIST STUDIES OF SOCIOCULTURAL DIMENSIONS OF HERITAGE LANGUAGES

The constructivist approach to the relationship between (heritage) language learning and its sociocultural dimensions has its origin in the sociology of knowledge (Berger & Luckmann, Reference Berger and Luckmann1966), phenomenology (Schutz, Reference Schutz1967), ethnomethodology (Garfinkel, Reference Garfinkel1967), practice theory (Bourdieu, Reference Bourdieu1977), and sociohistorical psychology (Vygotsky, Reference Vygotsky1978), among others. A social constructivist view sees sociocultural dimensions such as identities, attitudes, and motivation as accomplishments (outcomes) of linguistically encoded acts and stances. With respect to the notion of identity, for example, the cultural theorist Stuart Hall (Reference Hall and Woodward1990) wrote:

Postcolonial theorist Homi Bhabha (Reference Bhabha1994) similarly asserted that individuals have not one, but multiple, identities and that the identification of those identities is “never the affirmation of a pre-given identity” (p. 64). In other words, identities are not determined by essence or nature, but are derived from and maintained through social interaction. One of the implications of this standpoint is an emphasis on the possibility of choice and transformation of identity.

In this view, then, sociocultural concepts and labels are constantly changing, constructed through human interactions and the conditions of our lives. The qualities and attributes that we attach to any specific type of human activities are products of social conventions that are open to revision and renewal. In applied linguistics, this view is most saliently expressed in the concern with the dynamic nature of language learning (de Bot, Reference de Bot2008; Larsen-Freeman & Cameron, Reference Larsen-Freeman and Cameron2008; Markee, Reference Markee2008); the ecology of language learning (Kramsch, Reference Kramsch2002; van Lier, Reference van Lier2004); the sociocultural foundation of language learning (Atkinson, Reference Atkinson2002; Lantolf, Reference Lantolf2000, Reference Lantolf2006; Lantolf & Thorne, Reference Lantolf and Thorne2007); and the coconstructed nature of activities, affect, abilities, and identities (Ochs & Jacoby, Reference Ochs and Jacoby1995; Young & He, Reference Young and He1998).

Rather than treating HL as an independent variable, a constructivist approach considers HL as a dependent variable. It seeks to unravel the kinds of social cultural dimension a learner is attempting to construct by engaging in a specific kind of social interaction or expressing a particular kind of affect (Ochs, Reference Ochs1993). This approach emphasizes that in any given actual situation, at any given interactional moment, participants are actively (re)constructing themselves as members of a particular ethnicity, nationality, speech community, social rank, and profession and as learners of HLs at various proficiency levels (Ochs & Jacoby, Reference Ochs and Jacoby1995). Research from a social constructivist approach focuses on analyzing the organization of communicative practices through which HL learners and users acquire or maintain sociocultural knowledge and interactional competence and on the open-ended, negotiated, contested character of the interactional routine as a resource for language growth, maintenance, and change. In this line of work, the forms of language and the sociocultural contexts of language use become symbiotic with each other.

A social constructivist approach reconceptualizes (heritage) language development in the following ways. First, what does it mean to know an HL? A social constructivist approach considers language acquisition and socialization as an integrated process (Garrett & Baquedano-López, Reference Garrett and Baquedano-López2002; Kramsch, Reference Kramsch2002; Ochs, Reference Ochs, Stigler, Shweder and Herdt1990, Reference Ochs, Gumperz and Levinson1996; Ochs & Schieffelin, Reference Ochs, Schieffelin, Fletcher and MacWhinney1995; Schieffelin & Ochs, Reference Schieffelin and Ochs1986, Reference Schieffelin, Ochs, Slobin, Gerhardt, Kyratzis and Guo1996; Watson-Gegeo, Reference Watson-Gegeo2004). Linguistic meanings and meaning makings are therefore necessarily embedded in cultural systems of understanding. An account of linguistic behavior must then draw on accounts of culture. The heritage culture is by definition a complex, developing, transnational, intercultural, cross-linguistic, and hybrid one. Accordingly, to know an HL means not merely to command the phonetic, lexical, and syntactic forms in both speech and writing, but also to understand or embrace a set of continually evolving norms, preferences and expectations relating linguistic structures to multifaceted, dynamic contexts (Polinsky & Kagan, Reference Polinsky and Kagan2007).

Second, how does heritage culture relate to HL? From a social constructivist perspective (Ochs & Schieffelin, Reference Ochs, Schieffelin, Fletcher and MacWhinney1995), HL learners’ acquisition of linguistic forms requires a developmental process of delineating and organizing contextual dimensions in culturally sensible ways. Learners are tuned into certain indexical meanings of grammatical forms that link those forms to, for example, the social identities of interlocutors and the types of social events. This model relates learners’ use and understanding of grammatical forms to complex yet orderly and recurrent, dispositions, preferences, beliefs, and bodies of knowledge that organize how information is linguistically packaged and how speech acts are performed within and across socially recognized situations.

Just as foreign or second language learners may have varying degrees of investment across space and time (Norton, Reference Norton2000), HL learning is often motivated by neither strictly instrumental nor integrative goals; learner motivations are derived not merely from pragmatic or utilitarian concerns but also from the intrinsic cultural, affective, and aesthetic values of the language (Andrews, Reference Andrews1999; Bhatt, Reference Bhatt2009; Campbell & Christian, Reference Campbell and Christian2003; Carreira, Reference Carreira2009; Cummins, Reference Cummins2005; He Reference He2006). Unlike mother tongue acquisition in a monolingual environment, an HL is in constant competition with the dominant language in the local community. How do HL learners position themselves vis-à-vis mainstream culture or language? A social constructivist approach enables us to examine the construction of multiple yet compatible or congruent identities, as well as blended or blurred identities in multilingual, multicultural, immigrant contexts.

Third, what constitutes evidence of learning? Social constructivist research has looked for culturally meaningful practices across settings and situations (Agha & Wortham, Reference Agha and Wortham2005; Bartlett, Reference Bartlett2007; He, Reference He2006, Reference He2009b; Jeon, Reference Jeon2008; Markee, Reference Markee2008). Heritage language learners can be viewed as acquiring repertoires of language forms and functions associated with complex and changing contextual dimensions (e.g., evolving and shifting role relationships, identities, acts, events) over developmental time and across space (He, Reference He2006). It is not context-free frequency of occurrence of some linguistic forms but rather the understanding of the situational and interactional contingencies of the use of those forms that indexes the learner's competence.

The last but not the least important question to consider is the route by which HL is acquired and socialized. The transmission of HL takes place in not merely formal settings (e.g., classrooms) but also, and perhaps more importantly, informally (e.g., across generations at homes and in the communities). Both the propositional contents of message conveyed in the HL and the ways in which HL is used (e.g., how instructors or parents communicate with the learner) have a direct impact on how learners perceive the language and its associated culture (He, Reference He2000, Reference He, Bayley and Schecter2003).

There is a long tradition of conceptualizing language as an integral part of the development of the self, the mind, and the society that complements language learning. When language is seen not as a self-contained coherent system, but a context-specific tool for achieving our purposes, learner identity is then structured in the everyday flow of language and stabilized in the pragmatic narratives of our day-to-day, fluid social life. For HL learners, HL learning is thus constitutive of identity, which is accomplished in the everyday social conversations.

Rampton (Reference Rampton1995), though not focusing on HL matters exclusively, provided a compelling example of such a social constructivist approach. Rampton examined the hybrid, emergent identities created as students navigate social relations through the use of home languages. He described language crossing in urban, multiethnic groups of adolescents in the United Kingdom, as White, south Asian, and Caribbean youth mixed features of Panjabi, Caribbean Creole, and stylized Asian English. Crossing involves inserting linguistic features from other languages into speech that takes place in a predominant language. It is a discursive strategy with which diverse youth contest and create relations around race, ethnicity, and youth culture. Instead of correlating the social status of minority home languages with the use of home languages by youth to resist stigmatization and discrimination, Rampton showed how these and other sociocultural effects are realized in situated practice. The uses of minority HLs involve contestation, teasing, resistance, irony, and other stances with respect to social issues surrounding minority identities in Britain. Home language, in this case, along with the dominant language, is shown to be used by minority youth as a resource to confront the specific challenges presented to them in a multiethnic society, rather than an abstract variable to be correlated with social injustice.

Researchers working from a socioconstructivist perspective attend to creativity and indeterminacy in HL use. In addition to learner identities, cultural values and speech roles have also been subjects of inquiry. Lo (Reference Lo2004) demonstrated how expressions of epistemic stance relate to moral evaluations by looking at cases in which teachers at a Korean HL school claim to read their students’ mind with a high degree of certainty. Lo argued that Korean HL learners are socialized to portray their access to the thoughts and sensations of other individuals differently depending upon who these individuals are. If the individuals are perceived as morally worthy, then the access is portrayed as distant; if they are perceived as morally suspect, then the access is presented as self-evident.

He (Reference He2001, Reference He, Bayley and Schecter2003, Reference He2004, Reference He, Holten and Frodesen2005, Reference He, Reyes and Lo2009a) examined the development of from a linguistic-anthropological and conversation-analytic perspective that aims to synthesize cognitive and sociocultural approaches by considering language development as largely originating in social interaction and shaped by cultural and social processes. She investigated, for example, speech roles and speech exchange systems in the classroom (He, Reference He, Bayley and Schecter2003, Reference He, Holten and Frodesen2005, Reference He, Reyes and Lo2009a), discourse markers (Chen & He, Reference Chen and He2001), modal verbs and pronominal references (He, Reference He2000, Reference He2004, Reference He2009b), and sequential and social bases of semantic ambiguity in learning (He, Reference He2001) as rich resources for learner identity (trans)formation. He (Reference He2000), for example, detailed the discourse processes by which learners are socialized to values of respect for authority and group conformity through teachers’ directives in weekend Chinese language schools, where teachers do not merely impart knowledge and facts but also function as moral guides to the students. She not only documented how teachers often use three-part moral directives to discipline students’ disruptive behavior but also analyzed how teachers and students transform these directives as they construct particular stances in context.

Park (Reference Park2008) investigated how parents and grandparents in three-generational Korean American households socialize young children through their use of a particular linguistic feature in Korean, the verb suffix –ta. Drawing on naturally occurring interaction in the household, Park showed that utterances ending in –ta are used mostly by Korean adults to socialize children into the distinction between culturally desirable and culturally dispreferred behavior, thereby reconstructing the hierarchical relationships among different generations. Park highlighted the importance of the three-generational Korean households which make it possible for children to observe and imitate the culturally appropriate verbal behavior of their parents as the latter are engaged in interactions with the children's grandparents. In Park's study, the language form of verb suffix –ta was not the end of socialization (i.e., children are not expected to use this form in their own speech), but rather the means to socialize children into cultural norms and values.

Drawing on Bourdieu's (Reference Bourdieu1991) notion of habitus and symbolic capital, Dai and Zhang (Reference Dai, Zhang, He and Xiao2008) considered HL acquisition and maintenance as identity processes whereby the learner acquires and adjusts additional and alternative voices as resources for the positioning of self and of others. The authors situated the Chinese HL learners’ habitus in the fields of language, culture, and social interaction and utilized linguistic habitus, cultural habitus, and social habitus to differentiate the ambivalent threshold of what the learners are inheriting. Linguistic habitus refers to a certain articulatory style (as reflected in the variations of pronunciations in various dialects, for instance) that learners have acquired at home. Sociocultural habitus refers to a set of durable dispositions regarding core values, customs, lifestyles, and demeanors that heritage learners have acquired through socialization by family, school, friendship groups, institutions, and the mass media. The sociocultural habitus disposes the learner to cope with different social agents in certain ways.

Jeon (Reference Jeon2008) explored the role of language ideology in the maintenance of Korean. In a 3-year, multisite ethnographic study, she examined the range of language ideologies espoused by individuals in different phases of life. Drawing upon data from three separate venues—a university Korean language class of mostly heritage Korean speakers, a community-based English as second language program for Korean American senior citizens, and the home of a recent Korean immigrant family with teenage children—Jeon examined and compared the language attitudes of the mostly Korean-speaking elders in the community, those of the largely English-speaking second generation, and those of a recently emigrated father who insists that his children speak only English. Jeon concluded that language ideologies are continuously shaped by changing life circumstances and that promoting bilingualism at the societal level is a critical requirement in any language maintenance effort.

As we can see, the social constructivist approach is antithetical to correlational explanations of language development. It refrains from universalizing and allows social and language behavior to speak for itself, without imposing predetermined definitions of essential characteristics of HL learners. We fully expect future research along the social constructivist line to explore the challenges HL learners face across a spectrum of linguistic components and language skills and to reveal and specify the culturally situated ways in which these and other linguistic forms are learned and taught over interactional time, over historical time, and over developmental time (Lemke, Reference Lemke and Kramsch2002). We also expect to see research to be more sensitive and responsive to multiple sociocultural dimensions as heritage learners evolve in the course of social interaction, transforming in response to the acts and stances of other participants in order to provide a “rain forest” account of language learning with a “lush ecology” (Atkinson, Reference Atkinson2002, p. 526).

In sum, from both correlational and social constructivist approaches, we have seen an increasingly large body of empirical studies documenting the various formal and functional aspects of HL learning that are concerned with different subgroups of HL learners ranging from developmental traits in learners who have minimal proficiency in the HL to maintenance issues in the case of highly proficient HL learners. These studies largely focus on one language proficiency level of the subjects at one life stage in one specific life circumstance. As Campbell and Christian (Reference Campbell and Christian2003) pointed out, successful HL education is inseparable from the role of school systems, social institutions, and historical experiences of particular language communities, as well as language ideologies, suitable proficiency assessment instruments, and adequate literacy development. We have learned from previous research that a range of variables may influence language maintenance, including social prestige of the language, number of speakers, affinity to native country, vitality of HL schools, learners’ social and ethnic positionings, degree of family bond, and discourse and interactional practices (Baker, Reference Baker2006; Creeze & Martin, Reference Creeze and Martin2006; Feuerverger, Reference Feuerverger1991; Gibbons & Ramirez, Reference Gibbons and Ramirez2004; He, Reference He2006; He & Xiao, Reference He and Xiao2008; Jo, Reference Jo2001; King, Reference King2000; Shin, Reference Shin2005; Stalikas & Gavaki, Reference Stalikas and Gavaki1995; Tannenbaum & Howie, Reference Tannenbaum and Howie2002; Zentella Reference Zentella1997). Given our definition of HL, neither the classroom nor the family is the only domain relevant to HL development or maintenance. Efforts to understand HL will be most fruitful if we take into account not only formal, institutional settings such as schools (e.g., Byon, Reference Byon2003; He, Reference He2000; Lo Reference Lo2004) but also patterns of HL use in informal settings such as home and communities (Bayley & Schecter, Reference Bayley and Schecter2003, Brinton & Kagan, Reference Brinton and Kagan2008; Kondo-Brown, Reference Kondo-Brown2006; Park Reference Park2008; Xiao Reference Xiao, He and Xiao2008), not only the impact of face-to-face interaction but also the role of technology and popular culture (Lam, Reference Lam, Duff and Hornberger2008; Lee, Reference Lee2006). Temporally, like learning in other settings (e.g., Markee, Reference Markee2008; Wortham, Reference Wortham2005), HL development is not limited to any specific given period of time; HL competencies, choices, and ideologies change over the learner's life span, reflecting changing motivations, social networks, opportunities, and other variables. In other words, we need to attend to the sociocultural complexity of HL development, a point to which I turn next.

THE SOCIOCULTURAL COMPLEXITY OF HERITAGE LANGUAGE DEVELOPMENT AND ITS IMPLICATIONS

In her analysis of efforts to protect cultural heritage, Kirshenblatt-Gimblett (Reference Kirshenblatt-Gimblett, Buntinx, Rassool, Kratz, Szwaja, Ybarra-Frausto, Kirshenblatt-Gimblett and Karp2006) pointed out that people are not only passive cultural transmitters but also conscious, reflexive agents in the heritage enterprise itself. HL is not static but dynamic; it is constantly undergoing transformation by its learners and users, so that at the same time it serves as a resource for the transformation of learner identities, it is also transformed itself as a result of learners’ and users’ language ideologies and practices. It is important to realize that expert guidance in HL learning may be multiple, conflicting, and contested, as a result of different language ideologies held by different participants in the process. What constitutes an HL and what amounts to appropriate and adequate language learning can become debatable due to varying participant positions in terms of class, gender, generation, and so on. Beliefs about language articulated by learners, parents, and teachers to rationalize or justify the pedagogical decisions often play an important role in the outcome of school or community-based HL education. Ideological inconsistencies and tensions among those involved in HL education often affect the individuals’ attitudes about their language maintenance and language learning.

Furthermore, the HL learner is engaged in multiple speech events in multiple settings for multiple purposes. The learning of HL, for example, takes place through the learner's interactions with multiple participants including language instructors, parents, grandparents, siblings, and peers, each of whom positions the learner in unique speech and social roles, and each of whose reactions and responses to the learner helps to shape the path of his or her language development. Research on sociocultural dimensions of development thus entails examination of the different stages as well as different domains of development and of the coconstructed, interactive nature of learning activities. He (Reference He2006), for example, delineated the complexity of HL socialization along temporal and spatial dimensions, with a focus on Chinese as an HL. Along the temporal dimension, she underscored the nonlinear, iterative, dynamic nature of development. Along the spatial dimension, she highlighted the multiagency and multidirectionality of development. In her study, learner identity is the centerpiece rather than the background of HL development. Identity is to be understood in association with its verb form, to identify, and thus as identification. In other words, identity is treated not as a collection of static attributes or as some mental construct existing prior to and independent of human actions, but rather as a process of continual emerging and becoming, a process that identifies what a person becomes and achieves through ongoing interactions with other persons (Bucholtz & Hall, Reference Bucholtz, Hall and Duranti2004; Laroche, Kim, Hui, & Tomiuk, Reference Laroche, Kim, Hui and Tomiuk1998; Ochs Reference Ochs1993). He (Reference He2006) asked whether development will place learners in interactional conditions of cultural and linguistic ambiguity that they are prepared to handle and whether the growing cultural complexity of communication as a result of development will lead to the withering away or the emergence of certain types of identity constructs.

In short, HL development is grounded in the learner's participation in social practice and continuous adaptation to the unfolding, multiple activities and identities that constitute the social and communicative worlds that he or she inhabits. Such a position compels us to take a more dialectical, dialogical, and ecological perspective on socialization, in the sense that the process will be viewed as reciprocal. It is important to keep in mind that learners are not merely passive, uniform recipients of socialization. As learners’ allegiances and competencies evolve, the language choices and competencies of their parents, siblings, neighbors, and friends will also change, consequently and/or concurrently. In other words, learners contribute to the socialization process of the very people who socialize them to use the HL. Heritage language learning has the potential to transform all parties involved in the socialization process. It can also be expected that HL research will contribute to the very disciplines that have served as its theoretical or methodological guidance in terms of fundamental theoretical constructs, research methods, and units of analysis, among others. For example, HL learning provides fertile ground for us to reconsider dichotomous concepts such as native language versus target language, native speech community versus target speech community, instrumental versus integrative motivations, and basic interpersonal communication skills versus cognitive academic language proficiency. Last but not least, research will challenge us to reevaluate our unit of analysis from single snapshots of one-on-one, unidirectional interactional processes to trajectories of growth and change over space and time for all participants.