In the wider applied linguistics literature, the phrase heritage language (HL) learner is used in various ways that refer to a highly heterogeneous population with diverse historical, linguistic, and cultural backgrounds (for discussions on this subject, see Hornberger & Wang, Reference Hornberger, Wang, Brinton, Kagan and Bauckus2008; Kondo-Brown, Reference Kondo-Brown2003). However, in the literature that specifically deals with HL acquisition and pedagogy issues, the term HL learners usually refer to those who have acquired some competence in a nondominant language as their first language (L1) mainly through socialization at home, but did not achieve full-control over it due to a switch to the dominant language (Kim, Reference Kim, Kondo-Brown and Brown2008; Polinsky, Reference Polinsky, Brinton, Kagan and Bauckus2008; Valdés, Reference Valdés1995).Footnote ¹ The reported or demonstrated proficiency levels of HL learners differ widely because of at least three main factors: their diverse L1 backgrounds, degree of HL use and contact, and related sociopsychological factors (such as identity, attitudes, and motivation; Kondo-Brown, Reference Kondo-Brown and Kondo-Brown2006a).

Heritage language instruction in the United States is extensively available within and outside the formal education system.Footnote ² As the nation's most widely spoken immigrant language, Spanish seems to lead in pedagogy and curriculum studies on HL instruction, which is commonly known as Spanish for native speakers. At the same time, the work on other HLs taught in American schools and universities (many of which are what we call less commonly taught languages) is growing rapidly (e.g., Kondo-Brown, Reference Kondo-Brown, Kondo-Brown and Brown2008). Behind the expansion of HL pedagogy and curriculum studies is a national interest in promoting advanced-level competence using approaches that go beyond the traditional foreign language (FL) instruction (Malone, Rifkin, Christian, & Johnson, Reference Malone, Rifkin, Christian and Johnson2005; Robinson, Rivers, & Brecht, Reference Robinson, Rivers and Brecht2006).Footnote ³ One of the challenges for FL teachers in promoting advanced-level competence through traditional FL instruction is to provide the students with sufficient opportunities for exposure to and use of the target language in naturalistic, contextualized settings inside and outside classrooms (Omaggio Hadley, Reference Omaggio Hardley2001). In contrast, HL teachers are primarily concerned with how best to provide adequate instruction that builds on the students’ HL backgrounds or their rich linguistic and cultural repertories.

The goal of this article therefore is to make curricular and pedagogical recommendations as well as to suggest a future research agenda that will better serve HL learners in advancing their HL competence. To this end, this article first discusses selected second language (L2) curriculum development frameworks with specific reference to HL instruction. Second, a number of curricular, pedagogical, and assessment issues are examined concerning the advancement of HL competence during precollegiate period. Third, the same issues focusing on HL instruction offered at postsecondary institutions are explored.

CURRICULUM DEVELOPMENT FOR HERITAGE LANGUAGE STUDENTS

In the last 30 years, approaches to L2 curriculum development have been recommended from various perspectives. While a comprehensive discussion of these approaches is beyond the scope of the present article, it will review some of the key approaches with specific references to HL curriculum development, namely learner-centered, content-based, and outcomes-based.Footnote ⁴

With learner-centered curriculum, “it is the learners themselves who should be taken as the central reference point for decision-making regarding both the content and the form of language teaching, and . . . this goal should be realized interactively by a process of consultation and negotiation between the participants in the learning situation” (Tudor, Reference Tudor1996, p. 23). Because of the unique backgrounds and features of HL students, the notion of learner-centeredness is important and useful in teaching these students (Douglas, Reference Douglas2005; Webb & Miller, Reference Webb and Miller2000; Wu, Reference Wu, Kondo-Brown and Brown2008). From this perspective, the teacher first needs to determine what the students already know or what they want or need to know, and if there is a discrepancy between teachers’ and students’ expectations, it should be resolved in negotiations between the teacher and students (Nunan, Reference Nunan1988).

Recent survey research that examined HL students’ motivations for and preferences in learning an HL suggested that negotiations between the teacher and students are necessary and helpful in curriculum development (e.g., Ducar, Reference Ducar2008; Husseinali, Reference Husseinali2006; Jensen, Reference Jensen2007; Lee & Kim, Reference Lee, Kim, Kondo-Brown and Brown2008). For example, Ducar's (Reference Ducar2008) survey of Spanish HL students (N = 152) suggested that, while HL researchers and educators generally emphasize the importance of advancing HL students’ knowledge and skills in the formal, academic variety (see the last section on postsecondary HL instruction), the acquisition of academic Spanish may not be very interesting to the students. In this study, the majority of the students were not Spanish majors and had little interest in pursuing a career in academics. Instead, they were more interested in working on nonstandard varieties of Spanish (such as Mexican or Mexican American varieties) because they were more relevant to their lives and career goals. Based on this finding, Ducar recommended the inclusion of sociolinguistic projects where the students critically examine nonstandard varieties of Spanish, for example, linguistic autobiographies, language use surveys in the community, and so forth.

Content-based curriculum aims at advancing the students’ language competence by integrating particular content with the teaching of language skills. The focus for the students is on acquiring information in the L2, and in the process, they develop academic language skills in the L2 (Brinton, Snow, & Wesche, Reference Brinton, Snow and Wesche2003). In the United States, content-based instruction in an HL is available to precollegiate HL students via two-way immersion or in some FLES (Foreign Language in the Elementary Schools) programs, where positive effects on HL development have been reported (see, e.g., Potowski, Berne, Clark, & Hammerand, Reference Potowski, Berne, Clark and Hammerand2008; also see the next section on precollegiate HL instruction). Recently, the potential benefits of content-based curriculum for postsecondary HL students have also been proposed (e.g. Curriculum Guidelines, 2003; Kagan & Dillon, Reference Kagan and Dillon2001, Reference Kagan, Dillon, Byrones and Maxim2003; Lee & Kim, Reference Lee, Kim, Kondo-Brown and Brown2008; Morioka, Takakura, & Ushida, Reference Morioka, Takakura and Ushida2008). For example, the survey results reported in Lee and Kim (Reference Lee, Kim, Kondo-Brown and Brown2008) indicated that Korean HL students have a strong desire to connect with their Korean heritage and identity. Accordingly, they recommended content-based instruction that uses Korean culture as an organizing principle for the course content for such HL learners.

While interest in content-based curriculum for postsecondary HL students is growing, there seems to be a general lack of research on the effects of content-based instruction on postsecondary HL students. Future research should perhaps investigate these issues. For example, the potential for incorporating project-based activities, especially in mixed-level abilities HL programs might be examined (e.g., see Ilieva, Reference Ilieva2008). Project-based activities can be highly adaptable in terms of degrees of student ownership, purposes, data collection methods, products, classroom interactions, assessment procedures, and so forth (Beckett & Miller, Reference Beckett and Miller2006). In Ilieva's (Reference Ilieva2008) study, project work not only provided HL students an opportunity to integrate their cumulative language abilities and cultural knowledge through individualized, project-based learning but also allowed the students to engage in community-based as well as literature-based projects that dealt with specific cultural, linguistic, and sociolinguistic issues of interest to them.

Recently, outcomes-based curriculum has received considerable attention in the FL literature (e.g., Glisan & Foltz, Reference Glisan and Foltz1998; Mathews & Hansen, Reference Mathews and Hansen2004; Norris & Pfeiffer, Reference Norris and Pfeiffer2003; Rifkin, Reference Rifkin2003). This is partly due to an increasing public awareness of accountability and assessment issues for all educational programs including FL programs. In the midst of the current outcomes-based curriculum movement, the integration of good assessment practices is increasingly demanded for any language program offered within a formal educational system (Norris, Reference Norris2006). For example, assessment is a key component of curriculum development frameworks recommended for school-based two-way immersion programs, where learners of both languages are in the program, usually English L1 speakers learning the HL and HL speakers learning English (e.g., Howard, Sugarman, Christian, Lindholm-Leary, & Rogers, Reference Howard, Sugarman, Christian, Lindholm-Leary and Rogers2007).

As the national interest in outcomes-based curriculum and assessment grows, the use of standardized assessment instruments such as the American Council on the Teaching of Foreign Languages (ACTFL) oral proficiency test (OPI) may become more prevalent for HL instruction (see Ricardo-Osorio, Reference Ricardo-Osorio2008). While validity and reliability issues concerning the ACTFL OPI seem to remain controversial,Footnote ⁵ this instrument is likely to continue influencing FL and HL education in the United States. For example, the use of ACTFL OPI has recently been recommended by the National Council for the Accreditation of Teacher Education (NCATE) for teacher training accreditation purposes (Pearson, Fonseca-Greber, & Foell, Reference Pearson, Fonseca-Greber and Foell2006).Footnote ⁶ In the past, the appropriateness of using the ACTFL guidelines with HL students was examined, but it was mainly in the context of making placement decisions (see the last section on postsecondary HL instruction), and little is known about the degree to which the ACTFL guidelines will work for program assessment or evaluation purposes. For example, we do not know if the ACTFL guidelines can effectively measure the progress of postsecondary HL students, some of whom are believed to have advanced-level oral proficiency when they start the program (Campbell & Rosenthal, Reference Campbell, Rosenthal and Rosenthal2000). Future research on this issue is recommended. Also, future research could explore the possibility of using portfolios as an assessment option for HL students. Previous HL research has suggested that portfolio assessment has potential for use with HL students (e.g., Schwarzer & Petrón, Reference Schwarzer and Petrón2005), but the benefits of portfolios have not been adequately investigated.Footnote ⁷

PRECOLLEGIATE HERITAGE LANGUAGE INSTRUCTION

For precollegiate students, HL instruction is usually offered either as two-way immersion programs within the formal education system (largely for Spanish speakers) or as HL programs at community-based schools. This section examines issues concerning precollegiate HL instruction in these two settings, starting with the review of research on two-way immersion programs and followed by a discussion of community-based schools.

Two-way immersion programs, where the students are given content instruction in two languages, are rapidly expanding in the United States. According to the Center for Applied Linguistics online Directory of Two-Way Bilingual Immersion Programs in the U.S. (http://www.cal.org/twi/directory/), as of June 2009, 346 two-way immersion programs existed in the nation. Most of them are offered at elementary schools (85%), and the vast majority are operated in Spanish and English (92%). In some programs, the proportion of time taught in an HL gradually decreases from about 90% to 50%, but in others, an approximately equal amount of time is spent in the two languages from the beginning (Christian, Reference Christian, Brinton, Kagan and Bauckus2008). There is a vast amount of literature on two-way immersion programs, but the primary interest of the investigation seems to be its impact on the students’ overall academic achievement in the mainstream school, not HL maintenance or development per se. Unfortunately, in the United States bilingual education politics seems to have contributed biases to the procedures and interpretations of research investigating this issue (Crawford, Reference Crawford1992).

Research on two-way immersion programs that has a focus on HL maintenance and development is, however, expanding (e.g., Christian, Reference Christian, Brinton, Kagan and Bauckus2008; Hayashi, Reference Hayashi and Kondo-Brown2006; Potowski, Reference Potowski2007; Sohn & Merrill, Reference Sohn, Merrill, Brinton, Kagan and Bauckus2008). In these studies, the students attending a two-way immersion program in the United States are mostly English-dominant students, and their HL proficiency levels appear to vary considerably depending on their language use, attitudes, and identity constructions (e.g., Hayashi, Reference Hayashi and Kondo-Brown2006; Potowski, Reference Potowski2007). Despite individual differences, recent research generally suggests that attendance in a two-way immersion program is an effective approach to HL maintenance and development (Montrul & Potowski, Reference Montrul and Potowski2007; Sohn & Merrill, Reference Sohn, Merrill, Brinton, Kagan and Bauckus2008).Footnote ⁸ Future studies could continue to analyze the impact of a two-way immersion program on HL development, and in the process, one area that could be improved is the development of assessment instruments and procedures (Payton, Reference Payton, Brinton, Kagan and Bauckus2008; Tucker, Reference Tucker2005). For example, the Center for Applied Linguistics Web site mentioned earlier provides a comprehensive list of Spanish-language assessments for dual language programs. Assessment instruments and procedures for students of different HL backgrounds have also been developed, but the reliability and validity of these instruments do not seem to be adequately investigated (see Hasegawa, Reference Hasegawa, Kondo-Brown and Brown2008).

Community-based after-school or weekend HL programs are widely available for different language groups (see Kondo-Brown, Reference Kondo-Brown and Kondo-Brown2006a; Lee & Shin, Reference Lee and Shin2008; McGinnis, Reference McGinnis, Brinton, Kagan and Bauckus2008; Payton, Reference Payton, Brinton, Kagan and Bauckus2008). The students attending these schools come from diverse socioeconomic backgrounds, and the schools are operated and supported largely by the community leaders and volunteer parents. Community-based HL programs in the United States seem to differ widely in terms of curriculum (e.g., some use the target L2 as the main tool of communication, and others use it sparsely) as well as in terms of the resources available. Therefore, any generalization about these programs should be interpreted cautiously. However, one consistent finding across different studies is that parents usually have positive perceptions of community-based HL schools: They generally believe that attendance to HL schools helps their children maintain their ethnic identity and culture (Kondo-Brown, Reference Kondo-Brown and Kondo-Brown2006a).

Despite the benefits perceived by the parents, many students who are attending HL schools experience a sense of burden as the demands for academic work and extracurricular activities in the mainstream school increase (Kondo, Reference Kondo1998). Also, previous quantitative studies that investigated the effect of HL instruction at community-based schools were not able to find a positive correlation between proficiency levels and length of instruction at community-based HL schools (see Kondo-Brown, Reference Kondo-Brown and Kondo-Brown2006a). A number of pedagogical and situational problems associated with HL schools, whose operation largely relies on the parents, seem to explain the lack of relationship between the length of attendance in these schools and the demonstrated proficiency levels (e.g., Kataoka, Furuyama, & Koshiyama, Reference Kataoka, Furuyama and Koshiyama2000; Lee & Shin, Reference Lee and Shin2008; Li, Reference Li2005; Wang, Reference Wang2003).Footnote ⁹ Clearly, many community-based HL schools need substantial support for professional development, especially assistance in curriculum and materials development.

Among the variety of HL programs offered outside the formal education system in the United States, one of the most thoroughly researched types of HL programs is perhaps the set of immersion programs offered at nihongo hoshuu jugyookoo (Japanese language supplementary instruction schools), which are commonly referred to in the literature by the abbreviated form hoshuukoo. The hoshuukoo offer after-school or weekend immersion programs (usually from kindergarten to ninth grade) for Japanese L1 and HL children, which are supported from afar by the Japanese government as well as the local Japanese community.Footnote ¹⁰ Research suggested that the hoshuukoo plays a critical role not only for HL maintenance and development but also ethnic identity development (e.g., Chinen & Tucker, Reference Chinen, Tucker and Kondo-Brown2006). Research also suggested that attendance at hoshuukoo can help HL students advance their proficiency in Japanese (e.g., Kanno, Hasegawa, Ikeda, Ito, & Long, Reference Kanno, Hasegawa, Ikeda, Ito, Long, Brinton, Kagan and Bauckus2008; Nagasawa, Reference Nagasawa1995). For example, in Kanno et al.'s (Reference Kanno, Hasegawa, Ikeda, Ito, Long, Brinton, Kagan and Bauckus2008) study, among HL and non-HL students who demonstrated Japanese proficiency at the advanced or superior levels on the ACTFL OPI, HL students who attended a hoshuukoo also showed superior performances on other kinds of spoken and written Japanese proficiency tests.

As in the case of two-way immersion programs, factors such as age of arrival, amount and quality of study in Japanese, and exposure to Japanese at home seem to directly affect the Japanese proficiency levels of hoshuukoo students (Kataoka, Koshiyama, & Shibata, Reference Kataoka, Koshiyama, Shibata, Kondo-Brown and Brown2008). Kataoka et al.'s (Reference Kataoka, Koshiyama, Shibata, Kondo-Brown and Brown2008) large-scale study (N = 1,591) in fact revealed sizable individual differences in Japanese proficiency development among hoshuukoo students, and based on this finding, the authors suggested that the existing hoshuukoo curriculum, which requires the use of Japanese textbooks designed for Japanese L1 speakers, may need to be modified.

POSTSECONDARY HERITAGE LANGUAGE INSTRUCTION

This section examines issues raised in recent research on the teaching of HL students in postsecondary institutions: issues concerning the placement of these students; work on a subgroup of HL students whose proficiency levels in spoken and written language skills are critically unbalanced; issues of individual differences in HL proficiency levels among postsecondary HL students; and finally, work that deals with discrepancy between HL learners’ advanced levels of informal, nonprestigious language varieties and their low levels of formal, prestigious language varieties.

Beginning with placement issues, recent national survey study suggested that most university FL programs in the United States use locally or internally developed tests and self-ratings, but certain commercial or externally developed tests are also widely used (e.g., AP subject tests, SAT II subject tests, Spanish-Computer-Adaptive Placement Exam [S-CAPE], OPI, etc.; Brown, Hudson, & Clark, Reference Brown, Hudson and Clark2004). The receptive skills (e.g., reading comprehension and grammatical knowledge) are much more commonly assessed than production skills. To date, no national survey has specifically examined the use of placement tests for HL learners. Most of the published studies concerning the placement of HL students are case studies conducted at the authors’ home institutions. These studies suggested that the most useful placement tools for HL students are (a) autobiographic questionnaires that include self-reported language use and ratings (Domingo, Reference Domingo, Hudson and Clark2008; Kagan, Reference Kagan2005; Ranjan, Reference Ranjan, Hudson and Clark2008) and (b) faculty-made performance tests such as oral tests, which were mostly based on the ACTFL guidelines (e.g., Dolgova, Reference Dolgova, Hudson and Clark2008; Takahashi, Reference Takahashi, Hudson and Clark2008; Zamar & Robotham, Reference Zamar, Robotham, Hudson and Clark2008).

The HL placement case studies are largely descriptive, but empirical studies that provided HL student placement data are also emerging (e.g., Fairclough, Reference Fairclough2006; Kagan & Friedman, Reference Kagan and Friedman2003; Kondo-Brown, Reference Kondo-Brown2004, Reference Kondo-Brown2005). For example, as discussed earlier, some placement test procedures for HL students were developed based on the ACTFL guidelines. Kagan and Friedman (Reference Kagan and Friedman2003) examined the degree to which this procedure is appropriate based on their data collected from university HL students of Russian.Footnote ¹¹ Their study suggested that the ACTFL proficiency guidelines work as well for Russian HL learners as they do for their FL equivalents, while noting that, if they are at the same level on the ACTFL rating scale, an HL learner in Russian will have “better pronunciation, higher fluency, and a wider vocabulary range” than an FL learner (p. 543). Based on this study, they recommended the use of an OPI based on the ACTFL guidelines as part of the placement procedures for HL students.

Kondo-Brown (Reference Kondo-Brown2004, Reference Kondo-Brown2005), on the other hand, suggested that the multiple-choice placement tests intended to measure the students’ receptive skills may effectively place non-HL students as well as less proficient HL students into various courses, but the same placement procedures may not be effective for more proficient HL students.Footnote ¹² With more proficient HL students, Kondo-Brown's (Reference Kondo-Brown2004) study additionally suggested that a simple essay test can serve as an effective placement procedure for discriminating proficiency levels of all incoming HL and non-HL students, including those at the higher levels.Footnote ¹³ Likewise, the Sohn and Shin (Reference Sohn and Shin2007) study of university Korean students indicated that their multiple-choice tests (listening, reading, and usage) have less discrimination power than a composition test. They also observed that the correlation between the multiple-choice tests scores was high, but the correlation between the multiple-choice and a composition test was relatively low. Based on this finding, they suggested the inappropriateness of making placement decisions for HL students based on a single test. They recommended the use of (a) a writing task that requires the use of formal academic language and (b) a diagnostic oral interview test on the first day of class for HL students, especially those whose performances on the multiple-choice test and composition tests are unbalanced in order to avoid possible misplacement of these students.

A second major issue for postsecondary HL instruction (especially in lower-division, i.e., first- and second-year courses) is how to deal with HL students who have acquired native-like fluency in their HLs at home but had received little or no formal instruction in literacy skills (Kondo-Brown, Reference Kondo-Brown2003). The unbalanced development of spoken and written language is a concern for teaching any HL, but this issue seems more central in teaching HLs with other than Roman-alphabet orthographies. For example, research suggested that there are notable differences in literacy skills among university HL students of East Asian and Slavic languages depending on the quality and length of previous formal instruction or schooling in the target HL (e.g., Friedman & Kagan, Reference Friedman, Kagan, Brinton, Kagan and Bauckus2008; Kondo-Brown, Reference Kondo-Brown2004). In the case of Chinese HL instruction, the situation is a bit more complicated than other languages because one important pedagogical consideration for Chinese classrooms is whether the learners prefer traditional or simplified characters (Wiley, Reference Wiley, Brinton, Kagan and Bauckus2008).Footnote ¹⁴

Many postsecondary institutions seem to offer special courses for HL students with highly unbalanced spoken and written language abilities (Kondo-Brown, Reference Kondo-Brown2003). Although little is known about how such classes are designed, implemented, and evaluated, one study suggested that a homogeneous accelerated class for HL students of Chinese offers more effective literacy instruction than a regular Chinese class where both HL and non-HL students study together (Shen, Reference Shen2003). Some recent case studies also reported on the development of an online program for HL students who need special help with literacy skills (e.g., Wu, Reference Wu, Kondo-Brown and Brown2008; Zhang & Davis, Reference Zhang, Davis, Kondo-Brown and Brown2008). For example, Wu (Reference Wu, Kondo-Brown and Brown2008) reported on the development of an online literacy program specifically designed for a subgroup of Mandarin-speaking HL students who had not received formal instruction in Chinese. The program was developed based on a needs analysis as well as the National Standards for Foreign Language Learning in the 21st Century (ACTFL, 1999), and it required students to work on assignments and projects where they regularly interacted with their family and community. This line of research should continue in the future.

Individual learner differences in HL proficiency levels are another major concern. In some university programs, FL and HL students may study together in the same classroom from the very beginning (e.g., Gambhir, Reference Gambhir2008; Ilieva, Reference Ilieva2008; Yu, Reference Yu, Kondo-Brown and Brown2008), but in others, the FL and HL students study in separate courses initially but later merge into the same classes (e.g., Angelelli & Kagan, Reference Angelelli, Kagan, Leaver and Shekhtman2002; Sohn & Shin, Reference Sohn and Shin2007). In either context, the teacher more or less needs to deal with the issue of individual differences. In teaching less commonly taught languages such as Hindi, dealing with proficiency differences seems to be a serious matter (Gambhir, Reference Gambhir2008).Footnote ¹⁵ As discussed earlier, one recommended model for individualized instruction is project-based learning (Ilieva, Reference Ilieva2008). However, the dynamics and outcomes of mixed-abilities classrooms are largely unknown, and future research could continue to investigate the use of project work as a strategy to deal with large individual differences. Some researchers have proposed a differentiated instruction model for HL students because it allows students to work toward their learning goals at their own pace (e.g., Carreira, Reference Carreira, Potowski and Cameron2007). Another recommended model is computer-assisted instruction (e.g., Douglas, Reference Douglas, Brinton, Kagan and Bauckus2008; Meskill & Anthony, Reference Meskill and Anthony2008). For example, Douglas (Reference Douglas, Brinton, Kagan and Bauckus2008) examined the effect of using computer-assisted tools for a group of advanced-level Japanese HL students whose demonstrated Japanese proficiency levels were considerably different, especially literacy skills, including the ability to read and write kanji. The program provided individualized instruction on kanji and vocabulary learning as well as kanji learning strategy activities. The results of the pretest-posttest tests on kanji knowledge and kanji learning strategies indicated significant gains.

Some recent studies have revealed sizable intragroup differences in demonstrated or perceived knowledge and skills within each of the advanced-level HL and non-HL student groups (Kanno et al., Reference Kanno, Hasegawa, Ikeda, Ito, Long, Brinton, Kagan and Bauckus2008; Kim, Reference Kim, Kondo-Brown and Brown2008; Kondo-Brown, Reference Kondo-Brown2009; Kondo-Brown & Fukuda, Reference Kondo-Brown, Fukuda, Kondo-Brown and Brown2008; Rubio, Reference Rubio2003).Footnote ¹⁶ Clearly, some degree of individualized instruction is necessary in teaching HL students, especially when they study together with non-HL students in the same class. Future curriculum development studies that specifically address the issue of individual differences would be helpful in this regard. In doing so, researchers should especially consider the students’ views on the presence of both HL and non-HL students in the same classroom. Research has suggested that some non-HL students may view the presence of HL students positively as language models, while other non-HL students may find the presence of HL students intimidating and also have concern about grading issues (e.g., Lee & Kim, Reference Lee, Kim, Kondo-Brown and Brown2008).

Finally, learning academic language is an important matter for HL learners. Postsecondary HL students who have acquired informal, vernacular varieties at home experience challenges in learning the standard variety, which is valued for academic learning (e.g., Jo, Reference Jo2001; Parodi, Reference Parodi, Brinton, Kagan and Bauckus2008). Some researchers have suggested that HL students should improve their knowledge and skills in formal academic language by (a) having more opportunities to interact with speakers of the formal variety, (b) receiving explicit instruction on the notion of registers, and (c) engaging in classroom activities focusing on the high-level registers (Ducar, Reference Ducar2008; Valdés & Geoffrion-Vinci, Reference Valdés and Geoffrion-Vinci1998). Additionally, this process should be viewed as complementary to, not as a replacement for, the students’ abilities to speak and write nonstandard varieties in informal, colloquial contexts (Ducar, Reference Ducar2008; Lacorte, Reference Lacorte2005; Parodi, Reference Parodi, Brinton, Kagan and Bauckus2008).

A second issue is how best to teach academic vocabulary that goes beyond the domain of immediate personal use (e.g., Angelelli & Kagan, Reference Angelelli, Kagan, Leaver and Shekhtman2002; Campbell & Rosenthal, Reference Campbell, Rosenthal and Rosenthal2000; Kanno et al., Reference Kanno, Hasegawa, Ikeda, Ito, Long, Brinton, Kagan and Bauckus2008; Parodi, Reference Parodi, Brinton, Kagan and Bauckus2008). Unfortunately, to date, little research has examined the effectiveness of academic vocabulary instruction for HL students. One notable exception is McQuillan's (Reference McQuillan1996) quasi-experimental study that examined the effect of extensive free reading on HL students’ academic vocabulary learning.Footnote ¹⁷ Without a control group, this study may not accurately measure the actual effect of the experiment, but it is notable that the HL students made significant gains in vocabulary knowledge. Given that reading authentic texts for authentic purposes is often recommended for HL learners (Schwarzer & Petrón, Reference Schwarzer and Petrón2005), the option of teaching academic vocabulary to HL learners via extensive reading academic texts should be further examined in future HL study.Footnote ¹⁸

A third issue that is often raised specifically for advanced-level HL students is to how to help them acquire the meta-language necessary for success in advanced-level academic courses (e.g., Angelelli & Kagan, Reference Angelelli, Kagan, Leaver and Shekhtman2002; Valdés & Geoffrion-Vinci, Reference Valdés and Geoffrion-Vinci1998). Earlier studies suggested that HL students placed in advanced-level courses may produce grammatically correct utterances, but have difficulty in describing and analyzing them due to their lack of metalinguistic knowledge (Kondo, Reference Kondo1999; Valdés, Reference Valdés1995). Angelelli and Kagan (Reference Angelelli, Kagan, Leaver and Shekhtman2002) reported on the development of a university Spanish course that specifically aimed at helping advanced- and superior-level HL students acquire the meta-language for academic Spanish. The course, which “reflects the latest research available in the field of HL teaching” (Angelelli & Kagan, Reference Angelelli, Kagan, Leaver and Shekhtman2002, p. 207), included reading, speaking, and writing assignments that required students to analyze certain aspects of Spanish grammar (e.g., the use of past tenses in literary narratives) and report their analyses verbally and in writing. The end-of-course evaluations indicated that the students expressed satisfaction with their improved understanding of the Spanish linguistic system and their ability to discuss it using terminology appropriate for academic contexts.