Keywords

Introduction

During the past decade, a major gap has emerged between typical instructional practice in second language (L2) and bilingual teaching and the perspectives of researchers regarding optimal instructional practice. Language teaching is still largely based on monolingual instructional assumptions. In the case of teaching L2 as a subject, curriculum guidelines typically emphasize the desirability of maximizing instructional use of L2 and minimizing instructional use of students’ home language (L1). In bilingual and L2 immersion programs, the “monolingual principle” (Howatt 1984) dictated that the bilingual student’s two languages should be kept rigidly separate. Lambert (1984) expressed this assumption clearly in the context of Canadian French immersion programs:

No bilingual skills are required of the teacher, who plays the role of a monolingual in the target language … and who never switches languages, reviews materials in the other language, or otherwise uses the child’s native language in teacher-pupil interactions. In immersion programs, therefore, bilingualism is developed through two separate monolingual instructional routes. (p. 13)

This monolingual principle or “two solitudes” assumption (Cummins 2007) has increasingly been called into question in recent years on the basis of both theoretical and empirical considerations. In this chapter, I discuss the theoretical issues and empirical research relevant to instructional language use in bilingual and L2 teaching programs. This literature is definitive in refuting the monolingual principle and the legitimacy of instructional approaches based on that principle. When we free ourselves from monolingual instructional assumptions, a wide variety of opportunities emerge for developing students’ L1 and L2 proficiencies by means of bilingual/multilingual instructional strategies that acknowledge the reality of, and strongly promote, cross-language transfer.

Fading Credibility of Monolingual Instructional Approaches

Changing perspectives regarding the nature of bi-/multilingualism and L1/L2 relationships in the instructional process are illustrated in the 2014 publication of two edited volumes focusing on “the multilingual turn” in language education (Conteh and Meier 2014; May 2014). Prior to these publications, TESOL Quarterly, the major international journal focused on teaching English as an additional language, published a series of papers in two symposia focused on “Imagining Multilingual TESOL” (2009) and “Plurilingualism in TESOL” (2014) (Taylor 2009; Taylor and Snoddon 2013).

An immediate catalyst for this increased focus on “teaching through a multilingual lens” (Cummins 2014) was García’s (2009) book Bilingual Education in the 21st Century: A Global Perspective in which she elaborated on the construct of translanguaging . This construct was originally proposed in the Welsh context by Cen Williams (1996) to refer to the alternation of input and output mode in bilingual instruction. Thus, students may receive information through the medium of one language (e.g., Welsh) and then talk or write about this information through the medium of the other language (e.g., English) (Lewis et al. 2012).

García (2009) extended the notion of translanguaging to refer to the “multiple discursive practices in which bilinguals engage in order to make sense of their bilingual worlds” (p. 45) (emphasis original). This conception highlighted the fact that there are no clear-cut boundaries between the languages of multilingual individuals. The instructional implications include a shift away from a focus on teaching two separate autonomous linguistic systems to a more flexible set of arrangements that might include strategies such as students writing initially in their stronger language and using this as a stepping stone to writing in their weaker language (e.g., Fu 2009; Luk and Lin, 2014) or strategic use of code-switching by teachers and students, as well as a variety of other instructional strategies that focus on the affordances provided by students’ bilingualism (see, for example, Celic and Seltzer [2011] for a comprehensive compilation of translanguaging instructional strategies).

Although these recent developments have accelerated the fading credibility of monolingual instructional practices, the theoretical and empirical roots of this evolution go back much further.

Evolution of the Multilingual Turn

Theoretical Contributions

Two sets of theoretical contributions that predate recent discussions of translanguaging are discussed below. The first of these was the “linguistic interdependence” hypothesis that posited a common underlying proficiency that made possible cross-linguistic transfer. The second involved the elaboration of a dynamic systems view of multilingualism (e.g., Herdina and Jessner 2002), which drew on Cook’s (1995) articulation of the notion of “multicompetence” (Cook 1995) and Grosjean’s (1989) discussion of the very different mental structures that distinguish bilinguals from monolinguals.

Linguistic interdependence. One of the earliest observations of productive cross-lingual transfer in bilingual programs was made by Lambert and Tucker (1972) who noted that students in the Montreal-area French immersion program they evaluated engaged in a form of contrastive linguistics where they compared grammatical and lexical aspects of French and English. This spontaneous focus by students on similarities and differences in their two languages occurred despite the fact that, as noted above, teachers kept the two languages rigidly separate.

A theoretical rationale for teaching for cross-linguistic transfer was articulated in the late 1970s by Cummins (1979, 1981) who noted consistently significant relationships among academic aspects of L1 and L2 (e.g., reading comprehension). On the basis of these cross-lingual relationships and a variety of other data (e.g., bilingual program outcomes, age effects in L2 learning), he formulated the “interdependence hypothesis,” which posited that at a cognitive level, languages are not separate but connect with each other by means of a common underlying proficiency. This hypothesis was formally expressed as follows (Cummins 1981):

To the extent that instruction in Lx is effective in promoting proficiency in Lx, transfer of this proficiency to Ly will occur provided there is adequate exposure to Ly (either in school or environment) and adequate motivation to learn Ly. (p. 29)

In concrete terms, what this hypothesis implies is that in, for example, a dual language Spanish-English bilingual program in the USA, Spanish instruction that develops Spanish reading and writing skills is not just developing Spanish skills, it is also developing a deeper conceptual and linguistic proficiency that is strongly related to the development of literacy in the majority language (English). In other words, although the surface aspects (e.g., pronunciation, fluency, etc.) of different languages can be distinguished, there is an underlying cognitive/academic proficiency that is common across languages. This common underlying proficiency makes possible the transfer of cognitive/academic or literacy-related proficiency from one language to another.

There is extensive empirical research that supports the interdependence of literacy-related skills and knowledge across languages (see reviews by Baker 2001; Dressler and Kamil 2006; Cummins 2001; Genesee et al. 2006). Thomas and Collier (2002), for example, found that immigrant students’ L1 proficiency at the time of their arrival in the USA was a strong predictor of English academic development. The research trends can also be illustrated by the research of Verhoeven (1991) in the context of two experimental transitional bilingual programs involving Turkish-background students in the Netherlands. These programs promoted L1 literacy over several elementary school grades. Verhoeven reported that bilingual instruction resulted in better literacy results in L1at no cost to L2. In fact, in comparison to students receiving Dutch-only instruction, those in the bilingual transitional classes showed somewhat better performance in Dutch and a more positive orientation toward literacy in both L1 and L2. The study also supported the interdependence hypothesis by showing that “literacy skills being developed in one language strongly predict corresponding skills in another language acquired later in time” (p. 72).

The evidence supporting cross-lingual interdependence is clearly summarized by Dressler and Kamil as part of the Report of the National Literacy Panel on Language-Minority Children and Youth (August and Shanahan 2006). They conclude:

In summary, all these studies provide evidence for the cross-language transfer of reading comprehension ability in bilinguals. This relationship holds (a) across typologically different languages …; (b) for children in elementary, middle, and high school; (c) for learners of English as a foreign language and English as a second language; (d) over time; (e) from both first to second language and second to first language; (p. 222)

A recent study carried out in Taiwan provides strong support for cross-linguistic interdependence. In a sample of 30,000 grade 9 students, Chuang et al. (2012) reported correlations of 0.79 between Mandarin and English reading ability. The fact that more than 60 % of the variance in English reading could be accounted for by Chinese reading suggests that cross-lingual interdependence operates even when there are few linguistic commonalities between the languages.

Research examining cross-linguistic relationships between natural sign languages (e.g., American Sign Language [ASL] and spoken languages reinforces this conclusion (see Hoffmeister and Caldwell-Harris (2014) for a review). For example, Strong and Prinz (1997) investigated relationships between English literacy and ASL in a sample of 155 students between ages 8 and 15 attending a residential school for the deaf in California. Forty of the students had deaf mothers and 115 had hearing mothers. They reported that ASL skill was significantly correlated with English literacy and children with deaf mothers outperformed children with hearing mothers in both ASL and English reading and writing. They also reported evidence that the differences in English literacy between children of deaf mothers and children of hearing mothers could be accounted for by the differences in ASL proficiency between these two groups.

The research evidence suggests six major types of cross-lingual transfer that will operate in varying ways depending on the sociolinguistic and educational situation:

  • Transfer of conceptual elements (e.g., understanding the concept of photosynthesis)

  • Transfer of specific linguistic elements (e.g., knowledge of the meaning of photo in photosynthesis)

  • Transfer of more general morphological awareness (e.g., awareness of the function of –tion in acceleration [English] and acceleration [French])

  • Transfer of phonological awareness – the knowledge that words are composed of distinct sounds

  • Transfer of metacognitive and metalinguistic learning strategies (e.g., strategies of visualizing, use of graphic organizers, mnemonic devices, vocabulary acquisition strategies, etc.)

  • Transfer of pragmatic aspects of language use (e.g., willingness to take risks in communication through L2, ability to use paralinguistic features such as gestures to aid communication, etc.)

The question sometimes arises as to whether we are talking about transfer or the existence of underlying attributes based on cognitive and personality attributes of the individual. For example, can the relationship between L1 and L2 reading comprehension be explained by the fact that both are related to underlying cognitive attributes of the individual? In reality, transfer and attributes are two sides of the same coin. The presence of the underlying attribute makes possible transfer across languages . Attributes (e.g., verbal cognitive abilities) develop through experience; in other words, they are learned. Once they exist within the individual’s cognitive apparatus or operating system (Baker 2001), they are potentially available for two-way transfer across languages. In other words, transfer will occur from Lx to Ly or from Ly to Lx if the sociolinguistic and educational context is conducive to, or supports, such transfer.

Dynamic systems theory. Grosjean (1989) originally emphasized that “the bilingual is not two monolinguals in one person” and this insight was subsequently elaborated in Cook’s (1995, 2007) concept of multicompetence, which highlighted the fact that multicompetence is not comparable to monolingual competence in each language. Herdina and Jessner (2002) and de Bot et al. (2007) elaborated this perspective by proposing a dynamic systems theory which argued that the presence of one or more language systems influences the development not only of the second language but also the development of the overall multilingual system, including the first language. Dynamic systems theory goes beyond the notion of interdependence across languages by highlighting the fact that the entire psycholinguistic system of the bi- and multilingual is transformed in comparison to the relatively less complex psycholinguistic system of the monolingual. As expressed by Jessner (2006), there is “a complete metamorphosis of the system involved and not merely an overlap between two subsystems” (p. 35).

Dynamic systems theory and the concept of multicompetence are not in any way inconsistent with the notion of a common underlying proficiency. The interdependence hypothesis and common underlying proficiency construct were addressed to the explanation of a different set of issues and clearly do not aim to provide an elaborated cognitive model of bi/multilingualism . What all these constructs share is a recognition that the languages of bi- and multilinguals interact in complex ways that can enhance aspects of overall language and literacy development. They also call into question the pedagogical basis of monolingual instructional approaches that appear dedicated to minimizing and inhibiting the possibility of two-way transfer across languages .

Empirical Contributions

Lin’s (1996) study of classroom Cantonese-English code-switching in Hong Kong schools was one of the first to cast doubt on the legitimacy of linguistic segregation in bilingual education and L2 immersion contexts. The classrooms she observed were ostensibly “English-medium” but operated in a Cantonese-English oral mode and English written mode. Rather than characterizing this instructional code-switching as inherently problematic and a failure to faithfully implement an English immersion model, Lin highlighted the sociocultural, linguistic, and educational functions it served. According to Lin, these practices represented teachers’ and students’ pragmatic and expedient response to cope with the symbolic domination of English in the Hong Kong context. However, because official policies discouraged bilingual classroom practices, teachers were largely unwilling to acknowledge code-switching and bilingual language use in the classroom. As a consequence, there was no discussion at either policy or school level of the most appropriate approaches to bilingual language use for instructional purposes.

Lin (1996, 1997) pointed to the negative consequences of the lack of inquiry into alternative approaches to developing bilingualism among Hong Kong students. The bilingual practices observed did not affirm the value or support the development of Chinese academic literacy, thereby perpetuating the ideological domination of English academic monolingualism. Lin (1997) called for the development of “viable bi/tri/multilingual education approaches that will enable the majority of students to bridge the multiple linguistic gaps between their home world and their school world: the gaps between their mother tongue (Cantonese) and Chinese literacy, between Cantonese and spoken English, and between Chinese literacy and English literacy” (p. 288).

Over the past 15 years, educators and researchers working collaboratively have begun to move in the direction advocated by Lin by exploring alternative approaches to bringing home and school languages into productive contact with the goal of affirming and developing both. These bilingual instructional strategies (Cummins 2007) or translanguaging strategies (Canagarajah 2011; Creese and Blackledge 2010; García 2009; García and Li Wei 2014) have evolved within “monolingual” L2-medium classes for emergent bilingual students (e.g., Auger 2008; Cummins and Early 2011; Cummins and Persad 2014; Hélot et al. 2014), bilingual programs for emergent bilingual students (e.g., Celic and Seltzer 2011; García and Li Wei 2014) and L2 immersion programs for dominant language speakers (e.g., Lyster et al. 2009, 2013).

Two examples of students’ reflections (from Cummins and Early 2011) will illustrate both interdependence across languages and the effects of encouraging productive L1-L2 contact in the learning process. These late elementary grade students were in their first year of learning English in Lisa Leoni’s English-as-a-second language class in the Greater Toronto Area. Lisa had encouraged students to carry out creative writing and assignments in their L1 and generally use their L1 as a stepping stone to English.

When I am allowed to use my first language in class it helps me with my writing and reading of english because if I translation in english to urdu then urdu give me help for english language. I also think better and write more in english when I use urdu because I can see in urdu what I want to say in english. (Aminah; original spelling retained).

When I am allowed to use Urdu in class it helps me because when I write in Urdu and then I look at Urdu words and English comes in my mind. So, its help me a lot. When I write in English, Urdu comes in my mind. When I read in English I say it in Urdu in my mind. When I read in Urdu I feel very comfortable because I can understand it. (Hira; original spelling retained)

Emerging Issues

Two related issues relevant to the conceptualization of cross-linguistic transfer have emerged in recent years. The first concerns the terminological question of whether it is more appropriate to refer to individuals’ knowledge of multiple languages as “plurilingualism” or “multilingualism”. The second concerns the question of whether it is legitimate to refer to “languages” as constructs or entities at all as opposed to using the verb form “languaging” to express the integrated or fused nature of how people draw on their linguistic repertories. Clearly, if the concept of “languages” is not legitimate, then it is problematic to talk about cross-linguistic transfer or bringing students’ languages into productive contact.

Plurilingualism or Multilingualism?

The Council of Europe (2001) elaborated the construct of plurilingualism to refer to the dynamically integrated and intersecting nature of bilingual and plurilingual individuals’ linguistic repertoires, which include unevenly developed competencies in a variety of languages, dialects, and registers (Cenoz and Gorter 2013; Coste et al. 2009; Piccardo 2013). Researchers who have adopted this terminology make a clear distinction between plurilingualism and multilingualism: the former is seen as expressing the mutual influence and dynamic relations among languages and dialects while the latter is characterized as implying a static and autonomous conception of languages. It is clear that this conception of plurilingualism is highly congruent with dynamic systems theories of multilingualism. The distinction is elaborated by Piccardo (2013):

Multilingualism keeps languages distinct both at the societal level and at the individual level. It also tends to stress the separate, advanced mastery of each language a person speaks. Plurilingualism, on the contrary, is focused on the fact that languages interrelate and interconnect particularly, but not exclusively, at the level of the individual. It stresses the dynamic process of language acquisition and use, in contrast with coexistence and balanced mastery of languages. (Council of Europe 2001, p. 601)

This distinction, however, is not universally accepted. Conteh and Meier (2014), for example, use both terms and point out that the choice of term depends largely on the researcher’s intellectual tradition, with plurilingualism being used more commonly in the francophone scientific community and multilingualism in the anglophone scientific community. Gajo (2014) points out that within the francophone tradition, multilingualism refers to the societal level and plurilingualism to the individual level.

The relevance of this debate in the current context is that none of the researchers who continue to use the term “multilingualism” to refer to both societal and individual realities would associate the term with the negative characterization outlined above. None of the researchers who discuss the “multilingual turn” in language education (e.g., Conteh and Meier 2014; May 2014) have characterized the languages of bi/multilinguals as autonomous systems separated by rigid boundaries. Their use of the term is entirely consistent with its use within a dynamic systems theory and largely indistinguishable from the conception advocated by those who prefer the term “plurilingualism.” Similarly, the pedagogical approaches implied by notions such as “teaching for transfer” or “teaching through a multilingual lens” (e.g., Cummins 2014) in no way imply a static notion of multilingualism. My preference is to use both terms with “plurilingualism” preferred when the communicative goal is to emphasize specifically the dynamic and integrated relationships among language varieties within the individual.

Do Languages Exist?

The notion of cross-linguistic transfer has recently been questioned by García and Li Wei (2014) on the grounds that the construct of “a language” itself is illegitimate. They argue that there is only one linguistic system with features that are totally integrated rather than being associated with any one language. The terms languaging and translanguaging are preferred in order to position “language” as a social practice in which learners engage rather than a set of structures and functions that they learn. As noted above, the relevance of this position for the present chapter is that if languages do not exist, then it is meaningless to talk about transfer from one language to another.

Based on this conceptualization of translanguaging, García and Li Wei (2014) highlight some problematic issues in relation to scholars “who still speak about L1, L2 and code-switching ” (p. 62). They also argue that we can now “shed the concept of transfer… [in favor of] a conceptualization of integration of language practices in the person of the learner” (emphasis original) (p. 80). They question the notion of a common underlying proficiency because it still delineates separate L1 and L2 and separate linguistic features (p. 14): “Instead, translanguaging validates the fact that bilingual students’ language practices are not separated into an L1 and an L2, or into home language and school language, instead transcending both” (p. 69). They do admit that the linguistic features of the single integrated system are “often used in ways that conform to societal constructions of ‘a language’, and at other times used differently” (p. 15).

In light of the issues raised by García and Li Wei (2014), it is important to clarify the status of terms such as “home language” and “school language” (and L1/L2). Carried to its logical conclusion, the critique of the construct of “language” would mean that it would be illegitimate for a child to express an utterance such as “My home language is English but my school language is French.” It would also be illegitimate for web sites such as Ethnologue (www.ethnologue.com) to refer to and provide information about the 7,106 languages and dialects that humanity has generated. One could also not talk about Spanish-English (and other) bilingual programs since these languages do not exist. To claim that languages exist as social constructions but have no legitimacy “in reality” raises the issue of what is “reality” and what is a “social construction.”

García and Li Wei’s (2014) critique focuses on the linguistic reality of the construct of “language” rather than its social reality. Languages are clearly social constructions with arbitrary boundaries (e.g., between a “language” and a “dialect”) but these social constructions generate an immense material and symbolic reality (e.g., dictionaries, school curricula, wars, profits for corporations that teach and test languages, etc.). It is entirely possible to reconcile the construct of translanguaging , which highlights the integrated conceptual/linguistic system through which plurilingual individuals process and use language, with the social reality of different languages, understood as historical, cultural, and ideological constructs that have material consequences and determine social action (e.g., language planning, bilingual programs, etc.).

An analogy can be made with the construct of “colors.” We commonly talk about distinct colors such as red, yellow, and blue as though colors had an autonomous and objective existence. Yet we know that these “colors” represent arbitrary cut-off points on the visible spectrum. Although each color corresponds to a particular wavelength range, the spectrum represents a continuum with no objective divisions. In western society, we typically distinguish about seven major colors even though the human eye can distinguish about ten million color variations. In short, the major colors we distinguish are social constructions that we use to make sense of and act on our world (e.g., paint our house). Despite their lack of “objective” reality, few people would argue that we should abandon any reference to distinct colors. In the same way, it can be argued that the boundaries between different languages represent social constructions, but it is nevertheless legitimate to distinguish languages in certain contexts and for certain purposes in order to make sense of and act on our worlds. Thus, it is no more problematic for a 10-year old to talk about her “home language” and “school language” than it is for the same child to distinguish her red toy from her blue toy.

The essence of the conceptualization of translanguaging proposed by García and Li Wei (2014) can be maintained by acknowledging that: (a) the boundaries between languages/dialects are fluid and socially constructed; (b) as emergent bilinguals gain access to their two languages, these languages become fused into a single system (common underlying proficiency); (c) languages and languaging are socially contested sites and encounters where the legitimacy of cultures and identities are negotiated; and (d) school programs serving plurilingual/multilingual students should connect with students’ background linguistic and conceptual knowledge and teach for transfer and greater integration across languages.

Conclusion

There is overwhelming research evidence that literacy-related skills transfer across languages as bilingual development progresses through the school years. Educators and researchers working collaboratively have begun to identify multiple ways in which teachers can use bilingual instructional strategies to support this transfer process both in order to increase students’ overall metalinguistic awareness and promote academic development in both languages. As our understanding of bilingual and multilingual development has advanced, researchers have elaborated constructs such as translanguaging and plurilingualism to express the dynamic nature of bilingual and multilingual cognitive processing. These constructs expose the intellectual fragility of the notion that the two (or more) languages of the bilingual or L2 learner should be kept rigidly separate in bilingual instruction. However, the integrated nature of bilingual language processing does not require us to relinquish the construct of specific “languages” nor to banish from the lexicon constructs such as “home language,” “school language,” L1/L2, etc. Similarly, it is legitimate to talk about and promote instruction that teaches for transfer across languages.

Cross-References