3.1 Language demands in the era of high-stakes testing

The pressures placed upon emergent bilinguals in secondary schools around the world are only intensified by the fact that many students are expected to participate in high-stakes tests and to perform on par with their peers who are proficient in the language in which such tests are administered. These exams not only impact instruction but also greatly impact the lives of test takers, as has been well documented internationally (see, for instance, Shohamy Reference Shohamy2001; McNamara & Roever Reference McNamara and Roever2006).

In the US, for example, federal education policy entitled No Child Left Behind has since 2002 required that emergent bilinguals take and pass high-stakes exams in English, including tests of mathematics and English language arts. The scores from these tests are used by the federal government as a means of holding each school – and thereby each state – accountable for the federal funding they receive (Menken Reference Menken, Groenke and Hatch2009). The law requires that emergent bilinguals show continual progress on academic content assessments in English as well as on English language proficiency assessments, with failure resulting in high-stakes consequences for individual schools (such as school closure and loss of federal funding) as well as for students (such as grade promotion and graduation) (Menken Reference Menken2008; Solórzano Reference Solórzano2008). Because it is impossible to truly divorce language from content on a test administered in English to emergent bilinguals, researchers argue that the test scores attained are not valid for these students and should not be used for high-stakes decision-making such as school evaluation or to determine high school graduation, grade promotion, and program placement (Gándara & Baca Reference Gándara and Baca2008; Solórzano Reference Solórzano2008). Yet the practice continues, resulting in the disproportionate failure and penalization of emergent bilinguals on such tests, as exemplified in lower secondary school completion rates in US states where emergent bilinguals must pass high school exit exams to graduate.

These findings are supported in Canada by Odo (Reference Odo2012), where emergent bilinguals in British Columbia are required to participate in provincial high school exit exams in English, math, science, and social studies. He reports that emergent bilinguals are two to four times more likely than monolingual English speakers to fail these assessments, and are barred from high school graduation as a result. Odo's (Reference Odo2012) analysis of a sample exam draws into question its validity because, as in the US case, the assessment of language and content on the exam are confounded when administered to students learning English.

The linguistic and literacy demands of these exams are great for emergent bilinguals in all subject areas, not only language tests. Emergent bilinguals in the US score an average of 20–50 percentage points below their English proficient peers on state assessments, not only of English language arts but also of other content-area subjects such as mathematics (Abedi & Dietal Reference Abedi and Dietal2004; Government Accountability Office 2006). A word frequency analysis by Menken (Reference Menken2010) of high school exit exams used in New York for mathematics and English language arts shows that the exams in both subjects are linguistically complex, involving uncommon words in English that would be incomprehensible to emergent bilinguals, in accordance with Nation's (Reference Nation2006) argument that 98% of the words in a given text must be the most frequent English words for them to be comprehensible to an emergent bilingual. Not surprisingly, disparities between emergent bilinguals and other students are wide, not only on language exams but also on mathematics exams, as exemplified in the Israeli research by Levin & Shohamy (Reference Levin and Shohamy2008) cited above.

In Australia, all students since 2008 have been required to participate in the National Assessment Program: Literacy and Numeracy (NAPLAN). While scores are not used to close low-performing schools as in the US case, the Australian exams are still high-stakes in that scores are available to parents and publicized in the media (Lobascher Reference Lobascher2011; Polesel, Dulfer & Turnbull Reference Polesel, Dulfer and Turnbull2012). As elsewhere, these high-stakes exams have been criticized for their inappropriate use with emergent bilinguals (McTaggart & Curró Reference McTaggart and Curró2009; Wigglesworth, Simpson & Loakes Reference Wigglesworth, Simpson and Loakes2011). In their examination of past NAPLAN numeracy tests at the secondary school level, Quinnell & Carter (Reference Quinnell and Carter2011) clarify some of the challenges in subjects such as mathematics when they write:

They note how some words that are used in what they term everyday ‘natural’ English have different meanings when used in ‘mathematics’ English, and that NAPLAN mathematics test items are lexically dense and lack contextual cues (Quinnell & Carter Reference Quinnell and Carter2011).

Negotiating the language demands posed by content tests is challenging for all students, and particularly for emergent bilinguals. Moreover, Luke (Reference Luke2011) argues that this ‘culture of accountability, performance, and measurability’ (p. 370) is indeed a global phenomenon, greatly impacting schools in many different countries, and thereby intensifying the challenges that emergent bilinguals must face in secondary school.

3.2 Defining academic language and literacy for emergent bilinguals in secondary schools

This section clarifies the complex language demands of secondary schooling, particularly for emergent bilinguals. Skutnabb-Kangas & Toukomaa (Reference Skutnabb-Kangas and Toukomaa1976) first noted how emergent bilinguals can often appear to educators to be fluently bilingual on the surface, for example when using language for social purposes, while still performing below grade level on academic skills and tasks. Cummins (Reference Cummins1981, Reference Cummins, Street and Hornberger2008) drew the influential distinction between basic interpersonal communication skills (BICS) and cognitive academic language proficiency skills (CALP). BICS involve contextualized language that is supported by paralinguistic cues such as gestures, facial expression, and tone of voice as well as other interpersonal and situational cues to create meaning. By contrast, CALP involves more abstract language with fewer such cues, and is required of students in order to complete school tasks and assessments like those described above. Cummins (Reference Cummins1981, Reference Cummins, Street and Hornberger2008) found that students typically acquire BICS more rapidly than they do CALP. Amidst many critiques of the BICS/CALP distinction over the years, worth noting is Bailey (Reference Bailey and Bailey2007: 9) who cautions against thinking of social language as less cognitively demanding, and who instead argues that differences between BICS and CALP are ‘in the relative frequency of complex grammatical structures, specialized vocabulary, and uncommon language functions’. MacSwan & Rolstad (Reference MacSwan, Rolstad, Paulston and Tucker2003) take this further, arguing that CALP does not involve more complex language and that the BICS/CALP distinction conflates language ability and academic achievement, a consequence of which is ‘the ascription of special status to the language of the educated classes’ (p. 329). This perspective, in turn, has been found to foster a pathologizing of students who do not use language in ways that schools require – a point I return to later in this manuscript.

In any case, as content grows increasingly complex in secondary schools and literacy practices become more and more specialized within the subject areas, so too do the demands for the language needed to acquire this knowledge (Carrasquillo, Kucer & Abrams Reference Carrasquillo, Kucer and Abrams2004). Like other scholars who study the academic language demands of secondary schooling for emergent bilinguals (e.g., Zwiers Reference Zwiers2007; Gibbons Reference Gibbons2009), Schleppegrell (Reference Schleppegrell2004) employs functional linguistics (following Martin Reference Martin1992 and Halliday Reference Halliday1994) to examine the grammatical features of academic language used in school and to explain why particular aspects of the school curriculum are linguistically challenging for emergent bilinguals as well as speakers of language varieties other than the standard (see also Fang & Schleppegrell Reference Fang and Schleppegrell2008). Colombi & Schleppegrell (Reference Colombi, Schleppegrell, Schleppegrell and Colombi2002) note how secondary students must acquire what they term ‘advanced literacy,’ which they define as:

Implicit within their analysis of the academic language that secondary school entails is a belief that failure to ensure that secondary students acquire advanced literacy is tantamount to denying them opportunities for full participation in school and later in life.

In recent years, research has identified academic language and literacy as a primary reason for differences in performance among emergent bilinguals. Rather than being academically homogenous, emergent bilinguals in secondary schools arrive with disparate levels of academic language and literacy skills, content knowledge, and prior schooling experiences (Freeman, Freeman & Mercuri Reference Freeman, Freeman and Mercuri2002; Abedi Reference Abedi2004; Ruiz de Velasco Reference Ruiz de Velasco and Valenzuela2005). Academic language and literacy skills are crucial for achievement, particularly at the secondary level (Colombi & Schleppegrell Reference Colombi, Schleppegrell, Schleppegrell and Colombi2002; Zwiers Reference Zwiers2007; Menken Reference Menken2008; Freeman & Freeman Reference Freeman and Freeman2009; Gibbons Reference Gibbons2009). Short & Fitzsimmons (Reference Short and Fitzsimmons2007) highlight what they term an ‘academic literacy crisis’ amongst emergent bilinguals at the secondary level, which they argue should be of serious concern.

In the wake of these findings, new studies have emerged that argue the importance of academic literacy for secondary emergent bilinguals (Ruiz de Velasco & Fix Reference Ruiz de Velasco and Fix2000; Colombi & Schleppegrell Reference Colombi, Schleppegrell, Schleppegrell and Colombi2002; Snow & Biancarosa Reference Snow and Biancarosa2003; García & Godina Reference García, Godina, Jetton and Dole2004; Rubinstein-Avila Reference Rubinstein-Avila2004; August & Shanahan Reference August and Shanahan2006; Meltzer & Hamann Reference Meltzer and Hamann2005; Short & Fitzsimmons Reference Short and Fitzsimmons2007; Yi Reference Yi2007). While explicit literacy instruction is usually considered a task for elementary teachers, too remedial for instruction at the secondary level, there is growing recognition that the teaching of academic literacy across content areas also needs to become a regular part of secondary school curricula and instruction, particularly in the education of emergent bilinguals. As Koelsch (Reference Koelsch2006) writes of emergent bilinguals in the US, referred to as ‘English language learners’:

While working to ensure students attain the knowledge and skills that secondary schooling demands, it is equally important to examine those demands critically. Departing from earlier research that was inattentive to issues of language and power, the body of research in New Literacy Studies (NLS) posits that academic language is not neutral, but rather should be thought of as involving a series of social practices embedded in uneven power dynamics (Gee Reference Gee, Barton, Hamilton and Ivanič2000; Street Reference Street2003). Likewise, acquiring language for academic purposes involves written and spoken language across multiple modalities such as internet and digital literacies. As Street (Reference Street2003) writes:

There is a very real need for schools to support emergent bilinguals in their acquisition of academic language and literacy, but a criticism is how the very definition of academic language and literacy privileges some while marginalizing others.

4.1 Bilinguals’ translanguaging practices

Conceptualizations of dynamic languaging extend further, drawing into question traditional notions of bilingualism and the language practices of bilinguals. Brutt-Griffler & Varghese (Reference Brutt-Griffler and Varghese2004) argue for new understandings of how bilinguals both think and use their languages:

Pushing back against traditional views of bilinguals normed on the language practices of monolinguals, which portrays them as possessing two entirely separated language codes as if they were two monolinguals in one, these authors study the interconnectedness of bilinguals’ languages. Brutt-Griffler & Varghese (Reference Brutt-Griffler and Varghese2004) describe the language practices of bilinguals along a continuum of language rather than as a binary between first and second language.

Along this vein, numerous researchers are also now studying and writing about translanguaging to describe the fluid languaging practices of bilinguals, particularly emergent bilinguals in school contexts. The term was coined by Williams (Reference Williams1994) to describe the alternation between languages in instruction as a pedagogical strategy, in the context of Welsh-English bilingual classrooms, and was later extended by García (Reference García2009) to describe the language practices of bilinguals, who use their linguistic resources flexibly to create meaning. García (Reference García2009) rejects views of bilingualism as ‘monolingualism times two’ (p. 70), which she depicts as a bicycle with two wheels, and notes that such views are rooted in monolingual and what she terms ‘monoglossic’ ideologies based on the ideal of the fully balanced bilingual. Instead, she argues that a more accurate depiction of the language practices of bi/multilinguals in the twenty-first century would be an all-terrain vehicle, with wheels that ‘extend and contract, flex and stretch, making possible, over highly uneven ground, movement forward that is bumpy and irregular but also sustained and effective’ (García Reference García2009: 45).

As she writes in defining translanguaging:

García (Reference García2009) offers as an example a bilingual family's mealtime, in which some family members are bilingual and others monolingual, and shows how speakers must translanguage in order to communicate with one another in ways that are inclusive of everyone at the table.

As noted in Canagarajah's (Reference Canagarajah2011a) review of literature to date, translanguaging has been written about in the fields of sociolinguistics, applied linguistics, and literacy studies, as well as language and literacy education. The following terms have been used to refer to translanguaging and the fluid language practices of bilinguals: flexible bilingualism (Blackledge & Creese Reference Blackledge and Creese2010), plurilingualism (Council of Europe 2000), third spaces (Gutiérrez Reference Gutiérrez2008), metrolingualism (Otsuji & Pennycook Reference Otsuji and Pennycook2010; Pennycook Reference Pennycook2010), fluid lects (Auer Reference Auer1999), heterography (Blommaert Reference Blommaert2008), hybrid language practices (Gutiérrez, Baquedano-López & Tejeda Reference Gutiérrez, Baquedano-López and Tejeda1999), and poly-lingual languaging (Jørgensen Reference Jørgensen2008). When describing translanguaging practices in bilinguals’ writing, the following terms have been used: codemeshing (Young Reference Young2004; Canagarajah Reference Canagarajah2011b), transcultural literacy (Lu Reference Lu2009); translingual writing (Horner et al. Reference Horner, Lu, Royster and Trimbur2011), multiliteracies (Cope & Kalantzis Reference Cope and Kalantzis2000), continua of biliteracy and transnational literacies (Hornberger Reference Hornberger2003; Hornberger & Link Reference Hornberger and Link2012), hybrid language practices and textual third spaces (Gutiérrez Reference Gutiérrez2008), pluriliteracies (García, Bartlett & Kleifgen Reference García, Bartlett, Kleifgen, Auer and Wei2007), and heterography (Blommaert Reference Blommaert2008). Moreover, translanguaging research depicts the language and literacy practices of emergent bilinguals in secondary schools as dynamic and highly complex, thereby providing more nuanced understandings of these students, particularly as compared to the literature that simply describes the students’ academic language and literacy skills as seen through a monoglossic lens.

Translanguaging also occurs in classroom contexts, and has been referred to as: hybrid classroom discourse practices (Gutiérrez, Baquedano-López & Tejeda Reference Gutiérrez, Baquedano-López and Tejeda1999), multilingual classroom ecologies (Creese & Martin Reference Creese and Martin2003), a four-quadrant pedagogic framework for developing academic excellence in a bilingual program (Joseph & Ramani Reference Joseph and Ramani2012; Hornberger Reference Hornberger and Hawkins2013), supportive bilingual scaffolding (Saxena Reference Saxena2010), flexible bilingual pedagogy (Blackledge & Creese Reference Blackledge and Creese2010), and a multilingual holistic approach or focus on multilingualism (Cenoz & Gorter Reference Cenoz and Gorter2011a). All of these practices point to permeable boundaries between languages.

While the practice of translanguaging has been observed occurring in classrooms, translanguaging as pedagogy addressing how to leverage the dynamic languaging of emergent bilinguals strategically in order to enhance their learning remains less well understood (Creese & Blackledge Reference Creese and Blackledge2010). As Canagarajah (Reference Canagarajah2011a) writes:

In this review, I apply a translanguaging framework first to critique research that frames emergent bilinguals at the secondary level as ‘deficient’ when they do not possess the academic language and literacy skills that secondary school demands, or that suggests solely monolingual pedagogies to meet their needs. I then overview research that considers the pedagogical implications and potential of translanguaging and other bilingual pedagogies to meet the needs of emergent bilinguals at different places along the literacy spectrum.

4.2 Persistent rigidity in language education and schooling

Schooling practices often stand in stark opposition to the more fluid languaging practices of bilinguals described above.

Rigid language separation in schools would include the association of one teacher with one language, the use of a specific classroom for a specific language, and syllabi and curricula for the different languages (for example, the regular classroom teacher teaching through the dominant language and the foreign language teacher teaching an additional language). This approach prevails in schools serving emergent bilinguals, particularly at the secondary level, and is what Heller (Reference Heller1999: 271) terms ‘parallel monolingualism’ and Cummins (Reference Cummins2005: 588) calls the ‘two solitudes’.

In spite of what is now a robust and well-known body of research in support of bilingual education, schooling in many immigrant-receiving countries remains monolingual for emergent bilinguals. Research provides evidence that emergent bilinguals who are able to develop and maintain their home languages in school through bilingual education typically outperform their peers in monolingual programs and experience greater academic success (Thomas & Collier Reference Thomas and Collier1997, Reference Thomas and Collier2002; Krashen & McField Reference Krashen and McField2005; Baker Reference Baker2011). Notwithstanding the critiques of his work noted above, Cummins (Reference Cummins2000) has shown in research conducted in Canada how the academic language and literacy skills students acquire in their home languages are found to transfer to similar skills in their additional language. His interdependence hypothesis, or theory of linguistic transfer, holds that students who have developed their home language literacy skills ‘will tend to make stronger progress in acquiring literacy in their L2 [second language]’ (Cummins Reference Cummins2000: 173). Likewise, research shows conclusively that teaching students to read in their home language promotes higher levels of reading achievement in their additional language (August & Shanahan Reference August and Shanahan2006; Goldenberg Reference Goldenberg2008). Moreover, maintenance and development of both languages over time is associated with a range of academic, linguistic, and cognitive advantages (Cummins Reference Cummins2000; Bialystok Reference Bialystok2007). When the home languages of emergent bilinguals are not developed in school, as is typically the case, researchers argue that they cannot benefit from their bilingualism in the same ways (Baker Reference Baker2011).

It is well documented how school practices in many countries rebuff the research base in support of home language instruction, particularly at the secondary school level. For instance, Little's report (Reference Little2010) for the Language Policy Division of the Council of Europe seeks to promote home language instruction in the education of migrant children as a ‘basic human right’ (p. 13), in response to the reality that this is rarely the case within the educational systems of its member states. The report reacts to this reality by offering a framework to encourage plurilingual and intercultural education through language policy development.

Likewise, in her research about teachers of English as an additional language (EAL) in the UK who are themselves bilinguals (in Turkish, in this case) and working in mainstream English classrooms, Creese (Reference Creese2004: 189) notes how ‘the bulk of the language support in English schools is non-bilingual and is delivered in English’. She goes on to write:

A report for the Mayor of London's Office conducted by Issa, Allen & Ross (Reference Issa, Allen and Ross2008) in fact attributes poor educational attainment of students from Turkish, Turkish Kurdish, and Turkish Cypriot communities to the lack of bilingual education in English schools. These authors recommend that schools in London develop policies that ‘value and recognise the importance of the first language and distinctiveness of culture within all ethnic groups’, and ‘earmark funds for initiatives that seek to raise the attainment of under-achieving pupils, such as long-term investment in bilingual learning initiatives across the curriculum’ (Issa, Allen & Ross Reference Issa, Allen and Ross2008: 36).

In their report on the education of minority language speakers in Europe, Dooly & Vallejo (Reference Dooly, Vallejo and Unamuno2009) write the following, referring here to several other country reports by the Educational Policies that Address Social Inequality (EPASI) program:

These authors argue that the failure to provide home language instruction results in poor performance of emergent bilinguals in school, limiting the students’ literacy development and opportunities for success.

The attack on bilingual education has been well documented in the US, where home language instruction has become highly politicized and tied to heated national immigration debates. While in the past only a minority of all emergent bilinguals in US schools ever received bilingual education, there are even fewer opportunities for such students to do so now (see, for instance, Wiley & Wright Reference Wiley and Wright2004; Hornberger Reference Hornberger2005; Crawford Reference Crawford2007; Menken Reference Menken2008; Gándara et al. Reference Gándara, Losen, August, Uriarte, Gómez and Hopkins2010; Wright Reference Wright2010; Arias & Faltis Reference Arias and Faltis2012). For example, the states of California, Arizona, and Massachusetts, which serve large numbers of emergent bilinguals, have in recent years adopted ballot measures that seek to eliminate bilingual education altogether, resulting in a dramatic reduction of bilingual programs in those states (as documented in Crawford Reference Crawford2007; Gándara & Hopkins Reference Gándara and Hopkins2010; Arias & Faltis Reference Arias and Faltis2012). The term ‘bilingual education’ has also been altogether removed from federal education policy in the US, so that, for instance, what was the Bilingual Education Act has been replaced by Language Instruction for Limited English Proficient and Immigrant Students and the ‘Office of Bilingual Educational and Minority Languages Affairs’ is now called the ‘Office of English Language Acquisition’ (Menken Reference Menken2008; García Reference García2009). US federal education policy now requires high-stakes testing in English, which has resulted in a further reduction of bilingual education programs in the country (Wiley & Wright Reference Wiley and Wright2004; Hornberger Reference Hornberger2005; Crawford Reference Crawford2007; Gándara & Baca Reference Gándara and Baca2008; Menken Reference Menken2008; Menken & Solorza forthcoming). Moreover, while bilingual education is implemented in many countries around the world, it is not typically the favored approach for emergent bilinguals, who often come from minoritized backgrounds, to acquire the dominant language.

5.1 Students with interrupted formal education/refugee students

As noted above, secondary students who are new arrivals but who are at the earliest stages of academic language and literacy development are termed students with interrupted formal education (or SIFE). These students are also referred to in research as students with ‘Limited Formal Schooling’, ‘Limited Prior Schooling’, or ‘Limited Formal Education’ and as ‘Newcomers’ (e.g., Constantino & Lavadenz Reference Constantino and Lavadenz1993; Short Reference Short2002; Short & Boyson Reference Short and Boyson2004). As a number of these students hold official refugee status and, conversely, many refugees are labeled SIFE, the literature about adolescent refugees is also presented below. Much of the research reported here highlights what researchers perceive as limitations in the students’ language and literacy for academic purposes; while the academic needs of SIFE are important to note, later in this review I critique the tendency among researchers to focus solely on the students’ deficits, and apply a dynamic languaging lens in making my critique. First I outline the research as it is presented, and return to the critique in the sections that follow.

While there remains limited research about SIFE overall (Freeman et al. Reference Freeman, Freeman and Mercuri2002; Ontario Ministry of Education 2008), here I overview what has been published to date. In the US, where much of the research about these students has been conducted, Ruiz de Velasco & Fix (Reference Ruiz de Velasco and Fix2000) reported that 20% of all emergent bilinguals in high school and 12% of emergent bilinguals in middle school have missed two or more years of schooling. Australia, another top refugee resettlement country along with the US and Canada (according to McBrien Reference McBrien2011), generates a good deal of research about refugee students, and the state of Victoria reports that about 8% of the student population is comprised of new arrivals with little, no, or severely interrupted schooling (Brown, Miller & Mitchell Reference Brown, Miller and Mitchell2006). Literature about SIFE across these countries highlights concern for the poor performance of these students and their disproportionate course failure and dropout rates ([Canada] Kanu Reference Kanu2008; Toohey & Derwing Reference Toohey and Derwing2008; [US] Freeman et al. Reference Freeman, Freeman and Mercuri2002; Boyson & Short Reference Boyson and Short2003; [Australia] Brown, Miller & Mitchell Reference Brown, Miller and Mitchell2006; Matthews Reference Matthews2008).

In describing the characteristics of SIFE in the US, Freeman et al. (Reference Freeman, Freeman and Mercuri2002) write:

As with much of the literature about SIFE, these authors point to the need for the students to acquire the academic language and literacy skills that secondary schooling demands both in the state language as well as in their home language, all within a limited period of time. In this passage, Freeman et al. (Reference Freeman, Freeman and Mercuri2002) describe the challenges SIFE face in acquiring language and literacy simultaneously, on top of the need to acquire academic content, particularly within the high-stakes testing context that characterizes US secondary education at present. However, rather than noting the students’ dynamic languaging practices, the students are described here through a monolingual lens with a focus on English acquisition.

Boyson & Short (Reference Boyson and Short2003) confirm the findings of Freeman et al. (Reference Freeman, Freeman and Mercuri2002) with regard to the difficulty of developing language and literacy skills while simultaneously acquiring a new language within such a short time frame. They note how emergent bilinguals who arrive in US secondary schools with strong prior academic preparation have the greatest likelihood of educational success given appropriate instruction, and compare SIFE to these students in the following passage:

As these authors observe, prior schooling and corresponding home language and literacy skills appears to be a great predictor of academic success in an additional language in secondary school (a finding supported by August & Shanahan Reference August and Shanahan2006). Again, the exacting pressure of high-stakes testing in the US and high school graduation requirements demand that SIFE acquire both language and content very quickly, an unrealistic task for most.

As Boyson & Short (Reference Boyson and Short2003) indicate in the quotation above, the schooling of SIFE may have been interrupted for the range of reasons they mention, or it might have been limited for other reasons. What is more, the challenges these students face in school might not have been caused by actual interruptions to their schooling, but rather by differences in the schooling they received. While many school systems define SIFE as having received at least two years less schooling than their peers and functioning at least two grade levels below in reading and mathematics, Klein & Martohardjono (Reference Klein and Martohardjono2006) found in their research study that some SIFE students enrolled in US schools had never missed any schooling at all. Instead, these students attended school consistently in their country of origin without significant interruption, and simply received what these authors term ‘nonparallel schooling’.

Carrasquillo, Kucer & Abrams (Reference Carrasquillo, Kucer and Abrams2004) focus on the role of cognitive development in the challenges that SIFE face. As they write:

In this passage, the authors reference Cummins’ (Reference Cummins2000) linguistic interdependence hypothesis and the idea that academic skills such as literacy that are developed in a students’ home language will transfer to the second language. Or, taking a more dynamic view of bilingualism with fluidity between ‘first’ and ‘second’ language (as per García Reference García2009), ‘skills are skills’; accordingly, academic skills a student develops in either language are skills that the student possesses. What Carrasquillo et al. (Reference Carrasquillo, Kucer and Abrams2004) propose, however, is that emergent bilinguals may need to have a stronger cognitive foundation for them to perform well in school. This portrayal of SIFE students as somehow mentally limited simply because they do not have high levels of language or literacy for academic purposes, is a problematic extrapolation. For instance, the cognitive foundation of English monolinguals is not drawn into question when their literacy skills are limited, a line of critique I return to later in this review.

Whatever the reason for emergent bilinguals to be labeled ‘SIFE’, and in the absence of any description of their linguistic resources, much of the research about this student population generated outside the US also focuses on their need to acquire academic language and literacy. For instance, Brown, Miller & Mitchell (Reference Brown, Miller and Mitchell2006) conducted qualitative research about the experiences of Sudanese refugees attending high school in Melbourne, Australia and, from their interviews with students and teachers, identified academic language and literacy as presenting great challenges to the students in school. These challenges include subject-specific knowledge, involving the grammar, spelling, and technical or specialized vocabulary in subjects such as social studies and science. Further challenges cited by students included cultural literacies (for example, that would allow students to understand a culture-specific concept such as ‘Gold Coast Tourism’) and pedagogy (such as small group work, commonplace in many Australian schools but to which many emergent bilinguals are unaccustomed). The need to focus attention on SIFE academic language and literacy development is addressed elsewhere in Australian scholarship about this student population (Cassity & Gow Reference Cassity and Gow2005; Matthews Reference Matthews2008; Naidoo Reference Naidoo2009; Cranitch Reference Cranitch2010; Dooley & Thangaperumal Reference Dooley and Thangaperumal2011).

Johansson et al. (Reference Johansson, Angst, Beer, Martin, Rebeck and Sibilleau2001) likewise divide Canadian SIFE, whom they term ‘English as a second language (ESL) literacy students’, into the following three categories according to their home language literacy levels: 1) pre-literate and non-literate, 2) semi-literate, and 3) functionally literate. They set guidelines and expectations for the students’ literacy acquisition according to level. Kanu's (Reference Kanu2008) research in Canada notes a number of challenges that SIFE face, including difficulty with academic skills such as a fast-paced curriculum, limited differentiation of instruction for SIFE students in mainstream classrooms, and difficulty with note-taking, studying, academic writing, critical thinking, literacy and numeracy, and organizational skills. Kanu (Reference Kanu2008) notes further challenges specific to the subset of SIFE who are also refugee students that may impede their learning, such as acculturation stress, separation from families, and numerous psychosocial challenges of having experienced trauma.

While the body of research reported in this section provides important understandings of SIFE students and their needs in school, much of it, and particularly that which describes the students and their characteristics, focuses on their lack of academic language and literacy skills and even cognitive deficiencies. As a result, these studies can promulgate the deficit view of the students that is already prevalent in many schools. By contrast, other paradigms, such as translanguaging research promoted by García (Reference García2009) and others cited above, highlight the students’ complex home language and literacy practices. I propose that this more recent body of research about dynamic languaging offers a different way of seeing SIFE students, by sharpening our focus on the linguistic resources the students bring with them to school, rather than seeing them mainly for what they fail to bring.

5.2 Institutional barriers faced by SIFE and refugees in secondary schools

In many contexts, the challenges that SIFE face in acquiring the academic language and content that secondary school demands are compounded by the structure of traditional secondary schools. In the US, for example, traditional high schools are organized by subject area, with strict departmental divisions according to discipline. Such a structure provides little room to make language learning a high priority or to focus on the needs specific to SIFE.

Some SIFE students attending US schools receive specialized instruction focused specifically on their learning needs, typically through attending a one-year or summer program for SIFE, or in SIFE classes set apart from those of other emergent bilinguals. In spite of research highlighting the importance of home language instruction for SIFE (Klein & Martohardjono Reference Klein and Martohardjono2006; DeCapua, Smathers & Tang Reference DeCapua, Smathers and Tang2007; DeCapua & Marshall Reference DeCapua and Marshall2010; Lukes Reference Lukes2011), Short & Boyson (Reference Short and Boyson2011) find in their evaluation of newcomer programs nationally that the majority of SIFE programs offer only English as a second language (ESL) rather than bilingual education. Thus, instruction within these programs serving SIFE is typically in English only. What is more, SIFE in most school systems attend English monolingual classes in which they are mixed with other emergent bilinguals and often with native English speakers as well, in spite of the wide differences between the learning needs of each of these student groups. What is more, many SIFE in Canada and the US are also routinely placed into ninth grade upon arrival, regardless of their age, to offer them more time to meet graduation requirements. These placement practices mean that many SIFE are older than their grade-level peers, which offers a further layer of psychosocial complexity in the schooling experiences of these students (Johansson et al. Reference Johansson, Angst, Beer, Martin, Rebeck and Sibilleau2001; Freeman et al. Reference Freeman, Freeman and Mercuri2002; Dávila Reference Dávila2012).

In his qualitative study of SIFE from the Vietnamese Central Highlands attending school in the southeast US, Dávila (Reference Dávila2012) finds that emergent bilinguals in the high school he examined are enrolled in one 90-minute ESL class daily, with the remainder of their school day spent in ‘mainstream’ or non-ESL classes with native English speakers. He documents how instruction in the mainstream classes is rarely differentiated for emergent bilinguals, let alone for those who are SIFE. Dávila (Reference Dávila2012) terms this a ‘sink-or-swim’ situation in that the needs specific to SIFE students are left unaddressed, and points out how ‘certain normative school structures and interactions prove detrimental to their language learning, their self-esteem and their motivation to succeed’ (Dávila Reference Dávila2012: 139). Specifically, he argues that placing SIFE in mainstream classrooms which demand that they perform to the level of their English monolingual peers, in the absence of modified curricula or assessments, is a form of oppression that will ultimately perpetuate uneven power and class dynamics, limiting chances for upward social or economic mobility.

Research also points to structural issues in Australian secondary schools that limit their ability to meet the needs of students with interrupted formal education, as indicated in the following quote:

Many SIFE in Australia who are of secondary school age attend specialized six- or twelve-month intensive English programs called Intensive English Centres that vary in design from state to state, and are then mainstreamed (Miller, Mitchell & Brown Reference Miller, Mitchell and Brown2005; Ferfolja & Vickers Reference Ferfolja and Vickers2010).

There is a good deal of Australian literature that emphasizes the difficulty of the transition out of the Intensive English Centres for SIFE students, highlighting the lack of school and teacher preparedness in mainstream schools to meet the students’ needs, as well as the demand for longer programming options specialized for this population (Nsubuga-Kyobe & Dimock Reference Nsubuga-Kyobe and Dimock2002; Miller, Mitchell & Brown Reference Miller, Mitchell and Brown2005; Miller Reference Miller2009; Naidoo Reference Naidoo2009; Cranitch Reference Cranitch2010; Ferfolja & Vickers Reference Ferfolja and Vickers2010).

Specific pedagogical challenges that arise for mainstream teachers include appropriately differentiating instruction for students with such a wide range of abilities and prior knowledge, developing curricula and pedagogy appropriate to the literacy needs of SIFE, availability of appropriate materials, and addressing the emotional needs of refugee students who have experienced trauma (Miller, Mitchell & Brown Reference Miller, Mitchell and Brown2005).

To further support secondary school SIFE who are refugees in making the transition from Intensive English Centres into mainstream classrooms, Naidoo (Reference Naidoo2009, Reference Naidoo2010) and Ferfolja & Vickers (Reference Ferfolja and Vickers2010) describe the Refugee Action Support program in Greater Western Sydney. The program provides an approach to literacy development that employs small-group tutoring provided after school by university pre-service teachers in secondary teacher education working as tutors for teenage refugee SIFE. Specific benefits of the program for the students that are cited in these qualitative research studies include increased academic skills particularly in language and literacy, multiple identity construction, improved student performance, student empowerment through involvement and engagement in education agendas, and pedagogy and curricula that are student-centered. The main purpose of the program is to fill the gaps left by the school system's Intensive English Centres, which these authors argue do not offer sufficient support to meet the needs of SIFE.

As is the case in the US, instruction of SIFE in Australia is also typically in English only. For instance, Hatoss & Sheely (Reference Hatoss and Sheely2009) share the results of a sociolinguistic survey-based study of the Sudanese community in a regional settlement in South-East Queensland. Their findings demonstrate that even though the Sudanese refugee community in Australia is ‘strongly attached’ to their home language, Dinka, English was typically the sole language of instruction in schools.

Thus, monolingual instruction persists in schools serving SIFE internationally in spite of research arguing the benefits of home-language instruction for them (Carrasquillo, Kucer & Abrams Reference Carrasquillo, Kucer and Abrams2004; August & Shanahan Reference August and Shanahan2006; Klein & Martohardjono Reference Klein and Martohardjono2006; DeCapua, Smathers & Tang Reference DeCapua, Smathers and Tang2007; DeCapua & Marshall Reference DeCapua and Marshall2010; Lukes Reference Lukes2011). Similarly, culturally relevant instruction for these students to foster their acquisition of language for academic purposes is rarely provided (Nykiel-Herbert Reference Nykiel-Herbert2010), reflecting a continuation of deficit views of SIFE students and their families as well as wide cultural and linguistic dissonance between home and school language practices (García Reference García2009; Roy & Roxas Reference Roy and Roxas2011).

5.3 Long-term English language learners and generation 1.5 students

As noted above, students who have attended school in their receiving country for seven years or more and are still eligible for language support services such as ESL and/or bilingual education programming are labeled Long-Term English Language Learners (LTELLs) in the US (Freeman et al. Reference Freeman, Freeman and Mercuri2002; Menken & Kleyn Reference Menken and Kleyn2010; Olsen Reference Olsen2010). While not conflating the terms, there is, as I describe later in this section, overlap in the literature describing LTELLs and that of students labeled younger-arriving esl learners or generation 1.5; so later in this section I include research about these student groups.

While there has been little research to date about LTELLs or educational programming specifically tailored for them, there is great demand from educators in the field for further information about how to serve this population of emergent bilinguals, whose needs differ from those of new arrivals, and who account for a significant proportion of the emergent bilingual population at the secondary level. In the US, for instance, LTELLs now comprise approximately one-third of all secondary emergent bilinguals in New York City and in Chicago (New York City Department of Education 2008; Chicago Public Schools administrator, personal communication, 12 November 2010). At the state level, LTELLs comprise 23% of all emergent bilinguals in Colorado, and 50% in California (Olsen Reference Olsen2010). As is the case for SIFE, the research about these students typically focuses on their perceived academic deficits, at the expense of exploring their complex and dynamic languaging practices and presenting more holistic ways of seeing this student population. Below I apply this critique to my own earlier research about this student population.

Freeman et al. (Reference Freeman, Freeman and Mercuri2002) conducted one of the earliest formal studies of LTELLs, involving a small sample of students at the secondary school level in the US. Like others since then (e.g. Bernstein Reference Bernstein2004; Menken & Kleyn Reference Menken and Kleyn2010; Olsen Reference Olsen2010; Calderón & Minaya-Rowe Reference Calderón and Minaya-Rowe2011; Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012), they found that these students experienced inconsistencies in their schooling, ‘in and out of various English as a Second Language [ESL] or bilingual programs without ever having benefited from any kind of consistent program support’ (Freeman et al. Reference Freeman, Freeman and Mercuri2002: 5). They note how the academic performance of LTELLs lags far behind that of their grade level peers, a point confirmed in other research. A study of emergent bilinguals in Texas conducted by Flores, Batalova & Fix (Reference Flores, Batalova and Fix2012) shows that LTELLs scored below their peers not only in English but also in mathematics. As they write: ‘Whereas 86 percent of students who exited an ELL program in three years “met the standard” in math in the 11th grade, only 59 percent of long-term ELLs “met the standard”’ (Flores, Batalova & Fix Reference Flores, Batalova and Fix2012: 12). Another study of LTELLs in the Dallas Public Schools (Texas) shows that the overall academic performance of these students does not consistently improve over time, as there appears to be a ceiling to their level of academic language and literacy attainment in English (Yang, Urrabazo & Murray Reference Yang, Urrabazo and Murray2001). Accordingly, LTELLs are disproportionately likely to fail courses, be retained in grade, or drop out (Menken & Kleyn Reference Menken and Kleyn2010).

Menken, Kleyn & Chae (Reference Menken, Kleyn and Chae2012) conducted a descriptive, qualitative study in New York to provide information about the academic and social characteristics of LTELLs. The students are found to regularly translanguage in their daily lives, in that they fluidly move between the languages they speak. Their research highlights inconsistencies in the education of LTELLs within and between programs, schools, and countries. Drawing on interview data, they identified three main categories of LTELLs (noting that students may fall into more than one category):

1. 1) Students with inconsistent US schooling, whereby the system has shifted them between bilingual education, ESL programs, and mainstream classrooms with no language support programming;

2. 2) Transnational students, who have moved back and forth between the US and their family's country of origin; and,

3. 3) Students with consistently subtractive schooling, who were enrolled in subtractive models of bilingual education and/or ESL in the US, which failed to provide a strong foundation in English or their home languages. (Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012: 127–128)

Based on a mixed-methods study of LTELLs in New York City schools, Menken & Kleyn (Reference Menken and Kleyn2010) elaborate how the experiences of these primarily US-educated students can be described as ‘subtractive’, regardless of the above categories into which they fall, in that their home languages have not been developed in school and instead largely replaced with English. They find that this has been a significant contributing factor to the students being labeled ‘long-term ELLs’ over time. The authors offer recommendations to curtail the number of students who eventually become LTELLs, by ensuring that emergent bilinguals ‘be offered the opportunity to develop their native languages in US schools in programs with clear and consistent language policies, which seek to develop bilingualism and biliteracy’ (Menken & Kleyn Reference Menken and Kleyn2010: 413). Together, these studies (Menken & Kleyn Reference Menken and Kleyn2010; Menken, Funk & Kleyn 2011; Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012) promote educational programming for LTELLs at the secondary level that is differentiated from that of other emergent bilinguals, and which teaches academic language and literacy skills explicitly in all subjects through the medium of English and the student's home language.

Olsen (Reference Olsen2010) produced a report seeking to spotlight the large numbers of LTELLs in California, drawing upon a survey of 40 districts in the state. Offering further support for the findings presented above, in spite of differences in methodology, this report identifies challenges LTELLs face in school, notes that their prior schooling has been characterized by inconsistent and inappropriate language programming, and highlights how ‘few districts have designated programs or formal approaches for them’ (Olsen Reference Olsen2010: 2). Olsen (Reference Olsen2010) problematizes that there are any LTELLs at all, a population who she asserts would not exist with improvements to policy and language programming. As she writes, her report, entitled Reparable Harm, is ‘a wake up call to California educators and policymakers to recognize the large number of English Learner students amassing in California secondary schools. . . [who] are still not English proficient and have incurred major academic deficits’ (Olsen Reference Olsen2010: 1). While the author's main point is to improve California secondary schools, and this point is well taken, it is worth noting that the depiction of the students’ language practices is negative and framed in terms of what they lack rather than what they bring.

Rubinstein-Avila (Reference Rubinstein-Avila2004) provides a case study of the complex literacy practices of an 8th grade LTELL named Miguel. The case study found that despite the fact that Miguel used a great deal of literacy in his home, as the translator of important documents for his mother, he continued to struggle in school because of his academic language skills in both English and Spanish. The author documented some strategies that Miguel's 8th grade teachers used to improve his literacy including sustained silent reading and the explicit teaching of strategies such as the use of subheadings and context clues. The author argues that all secondary teachers must be trained in how to effectively teach LTELLs.

Based on their research about these students in Canada, Roessingh & Kover (Reference Roessingh and Kover2002) use the term younger-arriving esl learners to refer to LTELLs, whom they define as immigrant students who were born in Canada or entered Canadian schools at a young age. They note how many were exited from ESL programming, yet experienced academic difficulties at the high school where the study was conducted. Accordingly, the staff created a program to assist this population of students, involving a sheltered English class in the 9th and 10th grades in conjunction with an ESL adjunct class, in which the sheltered English class primarily focused on concepts and the ESL adjunct class focused on explicitly teaching the learning strategies and academic language needed for the students to understand those concepts.

There is some overlap between LTELLs and students termed generation 1.5 in US, Canadian, and Australian literature, a population of emergent bilinguals primarily educated in their receiving country. As Harklau (Reference Harklau2003) defines the term, Generation 1.5 students immigrated to the US when of school age or were born in the US, but speak a language other than English at home and have different learning needs from other emergent bilinguals because they are familiar with US culture and schools, and yet ‘they are usually less skilled in the academic language associated with school achievement, especially in the area of writing’ (Harklau Reference Harklau2003: 1). Roberge (Reference Roberge2002) includes transnational children who migrate with their families back and forth between the receiving country and their country of origin in his description of Generation 1.5 students. Generation 1.5 researchers have primarily focused on the challenges these students face in college, particularly in their writing, rather than the needs of these students in high school (Harklau, Losey & Siegal Reference Harklau, Losey and Siegal1999; Roberge Reference Roberge2002; Harklau Reference Harklau2003; Miele Reference Miele2003; Thonus Reference Thonus2003; Scwartz Reference Scwartz2004; Singhal Reference Singhal2004; for exceptions, see Forrest Reference Forrest2006 and Yi Reference Yi2007).

Singhal (Reference Singhal2004) identifies the following six characteristics of Generation 1.5 students based on research conducted in the US:

1. 1) nontraditional ESL learners (familiar with American culture)

2. 2) ear learners (learned English primarily through hearing it and not through reading or writing)

3. 3) limited knowledge of home language (whom the author terms ‘academically illiterate’)

4. 4) growing knowledge of English

5. 5) good aural/oral skills, and

6. 6) inexperienced readers and writers.

Based on research on Generation 1.5 students placed into mainstream English composition classes rather than ESL classes upon entrance into college, Scwartz (Reference Scwartz2004) notes that the defining characteristics of these students include a lack of academic language in English or their home language, a familiarity with American culture, and a resistance to being labeled ‘ESL’. Similarly, Gawienowski & Holper (Reference Gawienowski, Holper and Blumenthal2006) depict Generation 1.5 students as a disengaged group who no longer consider themselves ESL students and resent being placed in ESL classes. Again, the focus on student deficiencies is worthy of critique.

In writing about Generation 1.5 students in Canada, Vasquez (Reference Vasquez2007) analyzes a student who was able to use her advanced oral/aural proficiency in English to develop positive relationships with her Intensive English Program (IEP) professors and the international students in her class, which positioned her as highly motivated and more knowledgeable then her peers. However, this masked the difficulties she continued to experience with academic language. Thus, despite achieving high marks in her IEP classes, her teachers assessed her writing skills as average or poor. Williamson (Reference Williamson2012) shows similar findings in Australia, and writes:

In this quotation, the author describes the invisibility that this ability to pass as an English monolingual creates for Generation 1.5 students in Australia and elsewhere. The students’ strong oral language, particularly when language is used for social purposes, makes it difficult for educators to recognize their need to develop academic language and literacy skills (Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012). In the next section, I describe the mismatch between existing school structures and the students’ needs, and then move on to further critique of the ways that the students are often negatively framed in the literature and in schools.

5.4 Structural and ideological barriers to meeting the needs of LTELLs in secondary schools

As is the case for SIFE, the challenges that LTELLs and Generation 1.5 students face are compounded by the rigid and monoglossic structures (García Reference García2009) of traditional secondary schools. Research notes how LTELLs are often dismissed by educators as failures, how many teachers are reluctant to teach this sub-group of emergent bilinguals, and how language programming for them is typically mismatched to their actual learning needs (Menken, Funk & Kleyn 2011; Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012). For instance, Menken, Kleyn & Chae (Reference Menken, Kleyn and Chae2012) describe how secondary schools in the US are mainly set up for new arrivals with strong academic language and literacy skills in their home language, and do not differentiate instruction or distinguish programming for LTELLs from that of new arrivals. Accordingly, LTELLs usually attend ESL classes with new arrivals, where the curriculum and instruction they receive fails to account for their strong oral language skills when language is used for social purposes. LTELLs are also found to be easily confused with English monolinguals by their teachers. The reality is that these students require instruction that builds upon their language practices and that expands their academic language and literacy skills both orally and in writing. As these authors conclude: ‘high schools can no longer assume prior literacy ability among their ELL students, but instead must be prepared to teach literacy in explicit ways’ (Menken, Kleyn & Chae Reference Menken, Kleyn and Chae2012: 136).

Menken, Funk & Kleyn (2011) observe how this requires a shift in orientation for teachers, in that content-area teachers must also begin to think of themselves as language teachers, and language teachers (for example, of ESL or ‘foreign’ language) must begin to think of themselves as literacy teachers – a task which many resist, as documented in this study. The authors note that although ESL and bilingual teachers at the elementary level are typically prepared to teach literacy to their emergent bilingual students, secondary school teachers are ill-prepared in their teacher education programs or professional development workshops to meet the needs of students learning literacy as well as language and academic skills. What is more, the rigid structures of discipline-based secondary schools run counter to the needs of SIFE and LTELLs who are acquiring academic language and literacy, making it harder for instruction to be differentiated appropriately for these students.

Many LTELLs and Generation 1.5 identify as native speakers of the language of school, and thus resist their placement in language support programming. Talmy (Reference Talmy2004) observes how LTELLs and Generation 1.5 high school students in Hawai'i often resent being placed in ESL classes and ultimately reproduce the same linguicism they have experienced in school by distancing themselves from new arrivals, whom they term ‘fresh off the boat’ (FOB), and positioning themselves as superior. To illustrate how the students are treated by their teachers, Talmy (Reference Talmy2004) presents a classroom scenario in which the teacher asks the students to pick a holiday from their ‘own country or culture’ for a research project. She then rejects the suggestions of her LTELL students when they propose Christmas or New Year's as their holidays of choice for the assignment (instead, she accepted Chinese New Year's or other ‘foreign’ holidays). The author posits that the teacher's rejection of her students’ holiday choices implies that Christmas and New Year's belong only to her, as an American, and not to the students, whom she positions as ‘Others’ – a positioning that the students in turn resist.

Similarly, Kleyn, Flores & Menken (forthcoming) find that students labeled LTELL often not only contest their placement into ESL classes and their official designation as ‘English language learners’ in the US, but also resist language arts classes taught in their home language, Spanish. The problem is that the home language practices of the students are frequently corrected and thereby marginalized by their Spanish teachers, who were prepared to teach Spanish as a foreign language to English monolinguals. This article rejects portrayals of LTELLs as somehow limited or deficient, and argues that the students instead be seen as complete and dynamic, along the lines of the arguments made in the next section. Moreover, it is evident that schooling needs to change to attend to the diversity of emergent bilinguals who are at different locations along the academic language and literacy spectrum.