Rejecting abyssal thinking in the language and education of racialized bilinguals: A manifesto

ABSTRACT

Following Boaventura de Sousa Santos, the authors of this article reject the type of “abyssal thinking” that erases the existence of counter-hegemonic knowledges and lifeways, adopting instead the “from the inside out” perspective that is required for thinking constructively about the language and education of racialized bilinguals. On the basis of deep personal experience and extensive field-work research, we challenge prevailing assumptions about language, bilingualism, and education that are based on raciolinguistic ideologies with roots in colonialism. Adopting a translanguaging perspective that rejects rigid colonial boundaries of named languages, we argue that racialized bilingual learners, like all students, draw from linguistic-semiotic, cultural, and historical repertoires. The decolonial approach that guides our work reveals these students making a world by means of cultural and linguistic practices derived from their own knowledge systems. We propose that in order to attain justice and success, a decolonial education must center non-hegemonic modes of “otherwise thinking” by attending to racialized bilinguals’ knowledges and abilities that have always existed yet have continually been distorted and erased through abyssal thinking.

We are a group of scholars who have worked in language education for years. We are situated within two of the most powerful and interconnected English language empires – the United States and Great Britain. The lenses we have used for our work have been different, but the objective of our work has been the same: to center the experiences and knowledges of racialized bilinguals, their language, and their education. By racialized bilinguals we mean people who, as a result of long processes of domination and colonization, have been positioned as inferior in racial and linguistic terms. We hold that much of the scholarship on language education has been tainted by what the Portuguese decolonial philosopher Boaventura de Sousa Santos (2007) has called “abyssal thinking.” This hegemonic thinking creates a line establishing that which is considered “civil society,” and declares as nonexistent those colonized knowledges and lifeways positioned on the other side of the line, thus relegating them to an existential abyss. Our critique of abyssal thinking aims to unsettle European colonialism’s division of populations into superior “civilized” races and inferior “uncivilized” ones¹; and it aims to challenge too the insidious legacies of these colonial logics in the contemporary world (Quijano, 1991, 1993, 2000). We point to how the colonial logics stemming from abyssal thinking have been so well established that they are not readily apparent. We start with two vignettes drawn from some of our own data that exemplify how these logics operate in the lives and education of racialized bilinguals:

Vignette #1

Two 17-year-old British-born Chinese students, Tian and Ming, are speaking with Li Wei in a weekend complementary school where they are studying Chinese. They are discussing the subjects they are pursuing in their day schools, and Tian has just revealed that he is studying Latin, besides math and physics. When Ming is asked about his studies, he says:

Ming:和他一样, but no Latin. [Same as him, but no Latin.]

Li:数学和物理. [Math and physics].

Ming: Mm. Typical Chinese, isn’t it?

Li: Why do you say that?

Ming: Well, even when I was a kid, the teachers at school say: “Oh

Chinese kids are good at maths and science,” so always encouraged me

to do math and science subjects.

Tian: I think I’m pretty good at languages. But they never

encouraged me.

Ming: It’s true though, teachers think we are good at science and not

good at English stuff.

Li:So, what do you think of the term, EAL?

Tian:Don’t know. Makes you feel secondary I suppose

Ming: I hate it. English is my main language. I’m actually struggling with

Chinese. That’s why I’m here, doing weekend Chinese class. I’ve no problem with English.

(Li Wei, 2011, London)

Vignette #2

After a long discussion about language, in an 11^th grade English Language Arts (ELA) classroom in New York City, the teacher Ms. Winter, asks: “Do you have to use ‘proper’ English to sound smart?” Yessica, a bilingual Latina who identifies as Dominican, says matter-of-factly, “Miss, people are gonna judge you either way. It’s not even about your language; it could be about how you look.” The students in the group nod and voice their agreement, but one, Faith, is quiet.

Despite sitting in an 11^th grade ELA classroom, Faith is in 12^th grade, a senior in high school preparing to graduate. Repeating this class is meant to help her pass the examination required for graduation. Despite the myriad labels that have been placed on her over the years – English Language Learner, repeater, struggling reader – Faith’s proud self-descriptions of her identity as a poet, as well as her peers’ high praise for her poetry and comments about her being “smart” and “deep” tell a different story.

Faith speaks up. She begins, “No shade to what anyone said, but I think there’s some people in the world that are very ignorant towards those people who have high vocabulary standards, in which we are able to articulate ourselves. For example,” and here Faith smiles, “like I’m doing right now.” She continues, “They feel like, oh, she’s using these words and she has no idea what they mean and that’s a wrong judgment.” Ms. Winter asks, “Why do you think people would assume you don’t understand the words you’re using?” Faith replies proudly, “Cause I use a lot of them. And if you look at me as a young Latina, Brown, from the South Bronx … ”

(Seltzer, March 14, 2016, NYC)

These two vignettes demonstrate how racialized bilingual students are continuously positioned by society and categorized in schools as deficient in language, despite the students’ own understandings about their linguistic abilities. The potential of these Chinese and Latinx students is made invisible by abyssal thinking that assigns legitimacy only to the knowledge systems and practices stereotypically associated with dominant white monolingual people. The perceived deficiencies of these racialized bilingual students are produced by raciolinguistic ideologies that perpetually stigmatize their language practices (Flores & Rosa, 2015; Rosa, 2019; Rosa & Flores, 2017).

The task, then, is to challenge what Quijano (1991, 2000) has called ongoing coloniality, the imagined line in which some language practices and ways of life are understood as more academic, standard, or legitimate. By rejecting abyssal thinking and focusing on the vast linguistic complexity and heterogeneity of people and language, we challenge the line itself, rather than simply try to help people live with or overcome it. Influenced by this decolonial perspective that challenges the universal logic and matrix of power produced by colonialism,² we approach language, in Santos’ terms “from the inside out” (Santos, 2007, p. 54), putting at the center of our work the racialized bilingual students themselves as well as their languaging, that is, their everyday language interactions through which they make sense of their world (Maturana & Varela, 1984) – rather than their “language” as defined, taught, and assessed in schools.

In this article we bring together our perspectives to articulate how approaching language education from the inside out challenges traditional understanding of language, bilingualism, and education. We begin by situating our positions in our experiences. We then share our understandings of two key terms that define our fields – the term language and the term education. With respect to the first term, we clarify our understandings of language and our rejection of bilingualism as simply additive, as well as our rejection of descriptions of the language use of bilinguals in terms such as cross-linguistic transfer and code-switching. With respect to the second term, we explain how some of the policies and practices that are common in the education of racialized bilinguals are inadequate, and then provide some of the principles that guide our understandings of other practices that would be of much greater benefit to them. We end this article by reflecting on the enduring mismatch between, on the one hand, the theory and practice of much of traditional language education, and on the other hand the actual experiences of racialized bilinguals. We show that the persistent refusal of many to perceive this mismatch stems from abyssal thinking and raciolinguistic ideologies.

Situating our work

Our understanding of language and education has been constructed from our collective experience over time and in collaboration with many scholars who have called attention to the lack of justice in the education of racialized students (see, among many others, Bartolomé, 1998; López, 2017; Paris & Alim, 2014; Valdés,1996). Our work has also been inspired by many language scholars who have defended the rights of minoritized communities as a matter of justice (see, for example, Corson, 1993; Fishman, 1977; Skutnabb-Kangas & Phillipson, 1995). However, rather than perceive minoritized languages as autonomous entities that are entitled to rights, our work focuses on the rights of racialized people to be educated on their own terms and on the basis of their own language practices. For these rights to be enacted, it is almost certain that political changes outside the educational arena will be necessary. But regardless of the broader structural changes that the future may or may not bring, the decolonial perspective that we adopt makes us take note that a different world is already here – a world made by racialized bilinguals themselves as they engage with their own knowledge systems and cultural and linguistic practices (Martínez & Mejía, 2019). This allows us to reject abyssal thinking and point to already existing possibilities with what decolonial scholar Walter Mignolo (2000) calls “thinking otherwise”.

Most of the writers of this manifesto are members of racialized bilingual groups. Living our lives as racialized bilinguals has enabled us to witness and partake of knowledge systems that are not always accepted as modern, valid, or scientific, and yet hold much importance and value for our communities. As students, some of us have experienced what our bilingual practices got us – a remedial education that focused on perfecting our English rather than pushing us to think critically. We have felt what it means to be stuck and misunderstood in classrooms that did not challenge us, that did not engage our entire beings, our imaginations, philosophies, and aesthetics. We have experienced how it feels to be derided as (im)migrants or delinquent citizens, to be told that English is not our language, that we speak incorrectly, or even – literally in our experiences – that we are stupid because we do not speak English. Some of us have been told many times to “go back to your country” and to speak English, when in fact we were born in the U.S. or arrived in the U.S. or the U.K. as children or young adults. Educationally, we have attempted to respond to our purported linguistic deficiencies – including pronunciation patterns, grammatical constructions, and orthographic conventions. Yet when we mastered the skills identified as the causes of our attributed language problems, the target somehow always seemed to shift, raising in our minds questions about the fundamental nature of linguistic mastery, skills, and targets.

As teachers we have witnessed the absurdity of trying to teach only in English according to a curriculum formulated for the most part in narrow, white-Eurocentric terms, when in fact our bilingual students were much more developed linguistically, but also historically, philosophically, geographically, politically, and scientifically. We have often witnessed students excited to share a poem or piece of literature from outside the dominant Anglo-American tradition, only to be told to stick to what was in the book and the curriculum. As teacher educators we have also been challenged with the lack of attention to racialized bilingual students, whom teachers evaluate only through what they can do in English. And we have witnessed the stigmatizing effects of language policies in schools that work against the students’ bilingualism, policies that are found even in bilingual and heritage language education programs. Our experience in the U.S and the U.K., which informs much of the present manifesto, has been to engage educators and their students in critical decolonial thinking about language and education, enlisting them in the effort to combat raciolinguistic ideologies.

Our position on language education for racialized bilinguals has thus emerged from years of hands-on labor alongside teachers, children, and youth in classrooms. We have witnessed the perverse consequences that many traditional theories and practices in education have had for racialized bilingual children and youth, as we tried to briefly capture in the vignettes above through the voices of Ming, Tian, Yessica and Faith. We speak about theory/practice because we know that they are mutually constitutive and that neither one has priority. We root our work in rich empirical and experiential sources that focus on the consequences of the different kinds of theory/practice that can prevail in the education of racialized bilinguals.

Our understandings of language

Our work is centered on language education. We have affirmed unequivocally that languages do exist, and that they are socially constructed realities (Li, 2018; Otheguy et al., 2015, 2019). The socially constructed nature of named languages can be illustrated by the fact that, to take just two simple points, linguists cannot, through sole reliance on lexical and structural tools, tell you how many languages there are in the world nor determine what counts as two languages as opposed to two varieties of the same language. Linguists, for example, cannot resolve, based solely on lexical and structural criteria, whether Catalan and Valencian or Hokkien and Teochew are the same or different languages. The distinction between them can only be drawn, if it is to be drawn, by taking into account cultural, historical, and political considerations (Otheguy et al., 2015). The process of socially engineering named languages is well known (Fishman, 2000; Kloss, 1967). The unavoidably situated character of named languages is readily acknowledged by a website such as Ethnologue, a reference for languages of the world driven by Christian missionary colonizing work. Ethnologue gives a count of languages based on changing sociopolitical considerations, because the number cannot be based on purely lexical and structural characteristics.

Psycholinguistic research describes persistent simultaneous activation of what are regarded a priori as the two separate languages of bilinguals (Costa, 2005; Kroll & Bialystok, 2013). We question the framing of these findings in terms of socially constructed notions of separate languages and see instead a lack of discrete correspondence in the mental representation of bilinguals between their two presumably separate named languages (Otheguy et al., 2015). We use this perspective as a point of entry for stressing that the socio-politically imposed concept of a named language has little to do with how racialized bilinguals language or, for that matter, how any bilingual languages (García, 2009; García & Li, 2014). This ignorance of the languaging of bilinguals often leads to the marginalization of the linguistic practices of racialized bilinguals, as evident in the institutionalized perceptions on display in the first vignette above. That is, we maintain that bilingual people language with a unitary, not dual, repertoire from which they draw features that are useful for the communicative act in which they are engaged (Otheguy et al., 2015, 2019). We refer to this conceptualization of language and bilingualism as translanguaging (García, 2009; García & Li, 2014; Li, 2011, 2018). Translanguaging rejects abyssal thinking; it is a way to understand the vast complexity and heterogeneity of language practices, avoiding their conception as problems and their evaluation in the negative terms of the colonial imaginary line that values only those socially situated as being above and making invisible those assigned to being below. Translanguaging also leads us to include in the study of language the role of meaning-making resources long considered outside of language – as simply para-linguistic or pragmatic. How bilinguals deploy the sights, the sounds, the objects, and instruments at their disposal is important in our conception of language (Li, 2018; Li & Lin, 2019; Zhu et al., 2019).

Named languages do not simply exist as neutral objects but rather are brought into existence through sociopolitical forces that are part of the broader (re)configuration of the world that serves dominant interests (Makoni & Pennycook, 2007). This difference between existing and being brought into existence is not simply a terminological one. On the contrary, as in the vast number of experiences of which the vignettes above are a tiny sample, the abyssal thinking that produces strict boundaries around named languages co-articulates with raciolinguistic ideologies that perpetually stigmatize the language practices of racialized bilingual students (Flores & Rosa, 2015; Rosa, 2019; Rosa & Flores, 2017). Examples like those of Ming and Faith in the vignettes above, which show how racialized bilingual students tend to be consistently framed as linguistically deficient, are linked to broader ideologies of languagelessness that can position racialized subjects as illegitimate language users altogether (Flores et al., 2015, 2020; Rosa, 2016).

We argue that raciolinguistic ideologies undergird the notion that racialized bilinguals lack a construct known in schools as “academic language.” Efforts to purportedly teach racialized students to use academic language are fundamentally flawed. These efforts emerge from abyssal thinking claiming that there is an inductively established set of features that defines academic language that distinguishes it from non-academic language. But all we have, in fact, is the a priori category of academic language – assumed, not discovered – deductively supported by a meager number of defining shibboleths.

Because of the impossibility of clearly dividing language into academic and non-academic, attempts to identify detectable linguistic characteristics of academic language tend to stem from idealized representations of texts produced mostly by white monolingual English-users occupying a socially dominant position. This does not take into account myriad language traditions, or even ostensibly academic traditions that encompass the humanities and some social sciences – poetry, spoken word, narratives, novels, essays (García & Solorza, 2020). In addition, these efforts erase the inherent heterogeneity and defiance of boundaries found in all language practices, including as a matter of fact those deemed academic and standardized (Flores, 2020; Guerra, 2016; Martínez & Mejía, 2019). As a result, even when racialized and/or bilingual writers have attained recognition by having their work included in the school curriculum, the language practices of the authors are tagged as exceptional and unique. This means that their works are never placed at the core of the curriculum, but are presented as written in, for example, African American “dialect,” or as exceptionally including English and another language. In other words, these works, even when room is made for them in academic spaces, are defined as in opposition to those works whose language is regarded as normative and standard.

This relegates racialized bilinguals’ languaging to a place outside the school norm, resulting in their subjection to remedial educational approaches. These approaches are guided by the notion that the so-called “achievement gap” is the result of racialized students’ purported failure to master so-called academic language (Flores, 2020; García & Otheguy, 2018). For example, in the U.S. the assessment of five-year old children entering kindergarten consists, for the most part, of having them orally describe pictures. One of us once observed the case of Margarita, a Mexican American girl born in the U.S. who was asked to describe a picture of a mother making cookies from dough. In this assessment, the word dough was assigned more points than the word cookie because it was considered academic language. Margarita was very familiar with her mother making tortillas with masa, but she had never made cookies with her mother or used the word dough. So even though she could describe the picture of the mother making cookies, she could not come up with the word dough. As a result of numerous examples of this kind, she was put in an English-as-a-second language program and kept out of the school’s dual language bilingual program, which was reserved for those students who tested as “gifted and talented,” and thus inappropriate for a child considered to have limited vocabulary. In the ESL program, Margarita was not challenged because the focus was simply on having her add more presumably academic vocabulary to her lexicon. Our question has always been: What would have happened to Margarita if she had been challenged by teachers and classmates who believed in what she could do with language? What if she had been engaged in funny, imaginative, and challenging work that built on her existing linguistic and cultural knowledge?

This marginalization of racialized language practices connects to broader colonial histories that have questioned the linguistic competence of racialized communities as part of their dehumanization (Rosa, 2016). Indeed, in the direct genealogy of the concept of academic language is the concept of semilingualism, which suggested that racialized bilingual students failed to develop native-like proficiency in any language (Cummins, 1979; Skutnabb-Kangas & Toukomaa, 1976). Combining our linguistic analysis of the elusiveness of an inductively identifiable academic language with our analysis of this racialized history, we have connected the ideological construction of semilingualism and the subsequent emergence of discussions of academic language to what Flores and Rosa (2015) have called the “white listening subject.” Of course, a white listening subject is not always just listening, nor is it only white. The term refers to those who inhabit positions of institutionalized power that are produced and maintained, on the one hand, through structures of white supremacy, and on the other hand, through modes of perceiving and apprehending language, including but not limited to listening. Through the conceptualization of the white listening subject, Flores and Rosa make explicit the effect that the construction of a subjectivity based on claimed, ascribed, and socialized racial superiority has had in deeming the language practices of racialized bilinguals as inferior and non-academic. As in the case of the student named Faith in the vignettes, Flores and Rosa have shown that this assessment of inferiority persists in many instances independently of the actual structural features underlying linguistic practices.

In short, our contention is that academic language is not a set of empirically derived linguistic features, but rather a category that emerges as part of broader raciolinguistic ideologies that overdetermine racialized communities as linguistically deficient and unacademic, even as the concept of academic language itself remains impossible to define objectively. That is, racialized populations are often perceived by the white listening subject as using non-academic language that needs to be corrected even when engaging in ostensibly the same linguistic practices that are unmarked for white subjects. This was the case of Ming in the vignette who in spite of considering English his “main language” and “having no problem with English,” was categorized as “EAL” and “not good in English” based on raciolinguistic ideologies associated with Chineseness (on the complexity of Chineseness, see Li, 2021). And it was the case of Faith who acknowledges that despite her use of a lot of “high vocabulary,” she was not judged to be a competent language user because she is a “young Latina, Brown, from the South Bronx.” In a similar vein, Flores, Phuong & Venegas (2019) documented how “yeah” was heavily policed for being “non-academic” in a school serving predominantly low income Latinx students, despite being pervasive at the university, used frequently by doctoral students and professors. One would imagine that if the response to university colleagues when they used the term “yeah” was to correct them and insist that this was not appropriate for an academic setting, they would be offended. And yet, this is precisely the type of raciolinguistic policing that low-income students of color experience on a daily basis under the guise of providing them access to an imagined academic language (see also Martínez & Mejía, 2019).

We need to critically examine how narrow sociopolitical definitions of language imposed by nation-states and schools have little to do with the languaging of racialized bilinguals. We resist evaluating the language practices of racialized bilingual students based on norms that overdetermine them as linguistically lacking. And we seek to bring attention to the fact that these communities are already engaging in the types of meaning-making processes that schools demand (Martínez, 2018; Martínez & Mejía, 2019). Our position regarding the spurious nature of named and academic language does not constitute a barrier to the creation of high standards. On the contrary, we make these claims to open up the possibilities for pedagogical approaches that reject the abyssal thinking that has produced dichotomous framings of language. We favor a focus on language architecture (Flores, 2020) that supports racialized bilingual students by recognizing that they already have the linguistic knowledge that is required for school-related tasks. Our position is that their existing cultural and linguistic knowledge is neither a barrier nor a bridge to academic language, but rather legitimate on its own terms, and a necessary component in ensuring these students’ success on school related tasks (Flores, 2020). We maintain that our efforts have to be directed at challenging the colonial line that has been produced through abyssal thinking, and not directed to helping students accept the line and its pernicious strictures.

Rather than approaching education for racialized bilingual students in relation to a perpetually deferred future in which they will eventually acquire linguistic legitimacy, it is crucial to center the linguistic knowledge and broader skills these students possess in the present, which have been distorted and erased through abyssal thinking. The constructs of named languages and academic language have for too long been barriers to a meaningful education in the present for too many racialized bilinguals. We are not suggesting that the solution is to reframe racialized bilingual students’ existing language architecture as academic. Instead, we are indicating that the attempt to sort language practices into those deemed academic and those deemed non-academic is fundamentally flawed, both as a project of linguistics, and as a project of promoting racial equity, with the two always going hand-in-hand in our activism and scholarship.

Our understandings of bilingualism

Scholarship on the bilingualism of groups and on bilingual education grew in the mid-20^th century, spurred by research in Québec that then impacted the United States and Europe. At that time, Canada and the United States were dealing with different sources of political unrest. In Canada the struggle in the 1960s was between two white settler linguistic communities – – Anglophones and Francophones – ignoring indigenous and other racialized communities (Haque, 2012). In contrast, the struggle in the U.S. was in terms of the civil rights of racialized bilingual communities, especially Mexican Americans, Native Americans, and Puerto Ricans. In the Canadian case, language was foregrounded in the struggle for political power between the two white communities, culminating in Francophones gaining political power in Québec, a region in which they were the numerical majority. Immersion bilingual programs designed by scholars at McGill University (see, for example, Lambert & Tucker, 1972) responded to the needs of socially and economically powerful Anglophone communities that wanted their children to become bilingual. That is, the focus was on developing what was labeled additive bilingualism, with French added to English. In the U.S. case, however, language was one of many factors in the struggle over civil rights for racialized people facing marginalization and exclusion. For Mexican Americans, Puerto Ricans, and Native Americans who were involved in the early Civil Rights movement, education for their children was not solely about language, but about their rights to fair housing, jobs, income, as well as their right to educate their own children (Flores, 2016, 2017; Flores & García, 2017). For Mexican Americans and Puerto Ricans, the intent was never simply to add English, but to ensure that Latinx children, as colonized people, were able to use their bilingualism to exercise their rightful participation in society and in education (Otheguy, 1982). It is true that the bilingual education that was institutionalized in the U.S. as the Bilingual Education Act (1968) was meant to advance the shift to English of Latinx people (García, 2009), promoting what was termed by Canadian scholars as subtractive bilingualism (Lambert, 1974). On the ground, however, bilingual education efforts of Latinx communities were not focused on bilingualism in isolation, whether additive or subtractive, but rather were meant to advance the community’s overall well-being (Flores & García, 2017; García & Sung, 2018).

The notion of additive bilingualism took root in bilingual education programs all over the world, bolstering the colonial lines that had been established between dominant and non-dominant people and their languages and histories, as well as between native and non-native students. To combat the form of abyssal thinking that continually stigmatizes colonized populations’ language practices as deficient based on a static notion of linguistic legitimacy, we conceptualize bilingualism as “dynamic” (García, 2009). Dynamic bilingualism does not start from monolingual end points from which languages are added or subtracted. It starts from racialized bilingual students’ own languaging and broader knowledge systems that exist in what Anzaldúa (1987) has called the entre mundos/borderlands. This space between worlds is where our racialized bilingual students live fully, and where our educational practice is centered. The continued focus on additive bilingualism, as opposed to additive schooling (Bartlett & García, 2011), fails to account for how, without broader structural transformations, the bilingualism of racialized bilingual students will be perceived as deficient by those positioned as white listening subjects.

Furthermore, when bilingualism is described as simply additive, bilingual speech is often framed in relation to two discrete language systems. Jim Cummins, one of the most longstanding and prominent bilingual education scholars, first introduced the notion of cross-linguistic transfer (Cummins, 1979; Cummins & California State Department of Education, 1981), that is, the idea that linguistic skills in one language transfer to another language. Cummins’ theory rests on the concept that there is an interdependence, or a Common Underlying Proficiency (CUP), between named languages. The two named languages are entities with linguistic features that are viewed as separate, even though language proficiency is common to both languages. But we believe that the notion of cross-linguistic transfer, when both languages are conceived as separate and autonomous entities, has proven harmful to the education of racialized bilinguals.

Our position has emerged from work in classrooms. We have often heard teachers ask: When and how does transfer occur? How can I accelerate transfer? Teachers who think this way take up what is said to be the bilingual child’s first language (L1) simply as a scaffold to develop what they conceive of as their second language (L2). The focus of the teachers thus remains on the teaching of one or two named languages, and not on the process of teaching racialized bilingual children; in other words, the teachers are concerned more with language than with children. The result of this pattern of teacher interest continues to be that racialized bilingual students are often rendered as inadequate in one language or another, or even in both, with some teachers insisting that students have not reached the appropriate threshold in their L1 to be able to transfer knowledge to their L2. In our different proposal, the language acts of racialized bilinguals always leverage their translanguaging because students are acting not with one language system or another, but with a unitary network of meanings. Nothing is being transferred; everything is being accessed.

In line with the code-centered view of cross-linguistic transfer that we reject is a code-centered view of what is seen as the simultaneous use of multiple named languages that has typically been referred to in the literature as codeswitching. Gumperz (1982) defined codeswitching as “the juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or subsystems” (p. 59). In justifying this language use, many scholars of bilingualism have long insisted – benignly in their eyes – that this behavior is ruled-governed (MacSwan, 2017; Poplack, 1980). However, very early on, racialized bilingual scholars argued that the proposed orderliness and constraints on codeswitching, well-meaning as they were, did not correspond to their observations of practices in the community. For example, the Language Policy Task Force of the Centro de Estudios Puertorriqueños under the leadership of Pedro Pedraza et al. (1980) argued that in the East Harlem Puerto Rican community they studied, they did not find such constraints. We ourselves have examined, for example, the code-switching constraints offered by MacSwan (2017) and found that in our own homes and lives these constraints did not always hold (Otheguy et al., 2019). Consequently, we have argued that MacSwan’s claim that the repertoire of a bilingual person must also contain what he calls language specific internal differentiation, that is, that there must be two lexico-grammatical systems, does not hold up when confronted with the bilingual practices that we experience in many of our communities and homes. Our concern is that the insistence on the difference between “grammatical” and “ungrammatical” codeswitching is yet another mechanism for marginalizing the language practices of racialized bilingual students, many of whose daily language practices would be considered, from such a perspective, ungrammatical.

Our account of the languaging of bilinguals, what we have called their translanguaging (García & Li, 2014; Li, 2018), thus goes beyond the notion of cross-linguistic transfer (see, for example, Figure 1.1 in García & Li, 2014), as well as beyond the notion of codeswitching (Otheguy et al., 2015, 2019). Both notions clash with our proposal that bilinguals do language with a unitary linguistic system. Our proposal advocates effacing the line of cognitive demarcation purportedly separating the languages of the bilingual, a line that, born of abyssal thinking, is sustained by hegemonic sociocultural structures and ideologies but not by psycholinguistic reality. Because of the unitary nature of bilingual repertoires, assessment in one named language or another, or even in both separately, can never tell us the full picture of what bilingual children know and are able to do.

Since the emergence of translanguaging scholarship, some have taken up the term in ways that we regard as unfortunate, since they hold on to the classificatory distinction between named languages that reproduces abyssal thinking. This is the case, for example, of MacSwan’s (2017) call for multilingual translanguaging, discussed above. This is also the case of Cummins’ cross-linguistic translanguaging theory (CTT) (Cummins, 2021a, 2021b) which distinguishes his position from the way he refers to ours as unitary translanguaging theory (UTT). CTT follows Cummins’ (2017) proposal of the term “active bilingualism”, which emphasizes the agency of bilinguals while holding on to the concept of additive bilingualism and cross-linguistic transfer. These approaches reify the presumption of discrete languages that arose from colonialism and nation-building efforts, as well as give credence to the imaginary line imposed by colonial logics, enabling the continued identification of racialized bilinguals’ language practices as fundamentally deficient when compared to those of dominant monolingual language users.

For us, the unitary repertoire of bilinguals, that is, their translanguaging, serves as a point of entry for identifying the inherent heterogeneity in all language practices (Guerra, 2016). That is, by beginning from the perspective that bi/multilingualism is the norm, the translingual orientation is able to show that all language users leverage their repertoire in ways that are not compartmentalized into different grammars and modes. We frame this approach as a strategy for challenging abyssal thinking and raciolinguistic ideologies, enabling us to place our views in a broader social justice frame that not only gives racialized bilinguals the same opportunities to communicate and learn as their white monolingual peers, but also centers racialized bilingual students’ repertoires and lifeways rather than attempting to remediate them.

Our understandings of education for racialized bilinguals

We have had occasion to mention above that our understandings of how to educate racialized bilingual students have not emerged de la nada, out of context. They come from experiences of our own that have helped us to understand racialized bilinguals’ worldviews. And they are guided, not only by Santos’ rejection of abyssal thinking, but also by his position with regard to education that “preference must be given to the form of knowledge that guarantees the greatest level of participation to the social groups involved in its design, execution, and control, and in the benefits of the intervention” (Santos, 2007, p. 73). But we have seen that, ignoring Santos, many traditional theories and practices in language education are deeply connected to the knowledge of only one side of the line, rendering the other side non-existent, popular, intuitive, subjective, incomprehensible, magical. And we have seen that it is in this abyssal light that normative approaches to language and education continue to generate absences and failures.

For years, we have witnessed how the much-discussed principle of one-language-at-a-time in language classrooms and bilingual education is hardly ever applied. This has been documented not only in the U.S. (Martínez, Hikida & Durán, 2015; Tian & Link, 2019) and the UK (Creese & Blackledge, 2010; Li, 2014a), but also in Hong Kong (Lin & He, 2017), Puerto Rico (Mazak & Herbas-Donoso, 2014), South Africa (Makalela, 2017; McKinney & Tyler, 2019; Prinsloo & Krause, 2019), Malaysia (Rajendram, 2021), Nepal (Phyak, 2018) and many other places around the world. The question for us has been why, despite all the evidence of translanguaging as a productive frame for the actual behavioral norm in schools and communities throughout the world, does applied linguistic scholarship continue to insist on language separation as the most important characteristic of a language classroom? In an attempt to challenge this tendency in the field, our thinking around pedagogical practices has attempted to move beyond what in school is called one language or another as if these were bounded entities (García et al., 2017; García & Li, 2014; Li, 2011, 2018). To help teachers think beyond languages, we have found it useful to provide space for them to critically reflect on the pedagogical implications of shifting their understandings from a focus on cross-linguistic transfer to leveraging the emergent network of meanings of racialized bilingual children. Of course, racialized bilingual children may, on certain occasions, have to produce texts in one named language or the other, depending on the teaching context. But the teachers with whom we have worked understand that when allowed to act on texts as thinkers, listeners, speakers, readers, and writers, racialized bilinguals bring their whole emergent network of meaning into the texts (García, 2020; García & Kleifgen, 2019; Seltzer, 2019a).

Pedagogical practices informed by this shift in perspective build on a long tradition of work that has critiqued the strict separation of languages, especially in the teaching of colonized populations (for sources and examples, see García & Li, 2014, pp. 56–60). Indeed, it is important to point out that the original use of the term translanguaging, trawsieithu in Welsh, referred to a “bilingual instructional approach” for Welsh/English bilingual students where they would use one language for input and another for output (Baker, 2001; Lewis et al., 2012a, 2012b; Williams, 1996). In 2003, Danling Fu (2003) proposed what she called “a bilingual process approach” to teach writing to Chinese emergent bilingual students. And Cummins (2007) recommended “bilingual instructional strategies,” proposing, for example, the creation of “dual language books.”

Our shift from a focus on separate bilingual practices to unitary emergent networks of meaning moves pedagogical practices beyond simply the “dual.” As stated above, we recognize that named languages are sociopolitical categorizations that shape the very fabric of modern society; but we also understand that these named languages do not correspond to discrete dual linguistic systems. With this in mind, a translanguaging pedagogical design does not require bilingual students to hold their named languages as separate cognitive linguistic entities or to use one of them for the purpose of learning the other. Educators, then, are free to encourage bilingual students to leverage their entire semiotic repertoire and to select from it the features and modes that are most appropriate to building their worlds and understandings.

Li (2011) has called for the opening up of translanguaging spaces to confront the mono and dual logic operating in classrooms. Many other scholars around the world have extended the theory/practice of translanguaging (Blackledge & Creese, 2014; Busch, 2014; Canagarajah, 2011; Cenoz & Gorter, 2015; Fu et al., 2019; Makalela, 2017; Paulsrud et al., 2017; Pennycook, 2017; Scibetta & Carbonara, 2020). We reject language exclusion and separation in the education of racialized bilinguals in order to bring down barriers that prevent these students from making legitimate use of their full meaning-making resources. We take critical note of the fact that students from dominant white groups are permitted to use their full linguistic repertoire to participate in foreign- or second-language activities, in CLIL/Content and Language Integrated Learning, and in bilingual education programs, whereas, in contrast, racialized bilinguals are discouraged or even prohibited from making use of their full repertoire.

Our specific focus on the experiences of racialized bilinguals also affords us a point of reference for the concept of translanguaging that differs from the concept of plurilingualism which has been popularized in the European context. Like translanguaging, plurilingualism has challenged idealized notions of bounded languages and of their strict separation, focusing instead on the learner’s ability to use a repertoire of several named languages to varying degrees as part of what is understood as intercultural action (see, for example, Coste, 2000). Yet, because plurilingual policies in education evolved from a need by the European Union to have a common European citizen who could communicate across languages to trade, sell, and enlarge markets (Hélot & Cavalli, 2017), these European educational policies are rooted in sociopolitical dynamics that differ greatly from those that inform our decolonial perspective (García & Otheguy, 2019). Plurilingualism and translanguaging have emerged from what Mignolo (2000) has called different loci of enunciation, with plurilingualism responding to global economies, and translanguaging offering a way to delink from the logics derived from colonialism and global capitalism.

We are aware that any depoliticized approach to multilingualism lacking an explicit social justice perspective (no matter whether named plurilingualism or even if called translanguaging) will contribute to the production of neoliberal subjects in ways that exacerbate existing global racial and class inequities (Flores, 2013). In contrast, translanguaging for us, issued from our locus of enunciation and informed by decolonial thinking, places questions of equity for racialized bilinguals and broader societal inequities at the center of the analysis. This allows us to bring attention to how, for example, in the United States, bilingual education has lost its commitment to racialized bilingual communities, and has instead been used to attract monolingual students, often from dominant white groups, to stratify public schools and gentrify neighborhoods (Cervantes-Soon et al., 2017; Flores & Chaparro, 2018; Flores & García, 2017; Palmer et al., 2014; Poza, 2016; Valdez et al., 2016). Our equity inspired, translanguaging approach also allows us to bring attention to the impact of racial and socioeconomic segregation on the quality of bilingual education programs (Flores & McAuliffe, 2020).

Our understandings of pedagogical practices

We conceptualize pedagogy as more than a series of “strategies,” seeing it instead as a way to create in-school spaces that leverage the language and knowledge systems of racialized bilingual students. To repeat, we recognize that named languages are sociopolitical categorizations that have shaped societies, and, as such, must be acknowledged and made part of teaching in schools. However, as we have also mentioned, we recognize – and work with teachers and schools to recognize – that these named languages do not neatly correspond to the mental representations nor the language practices of racialized bilingual speakers (or of any speakers). Relatedly, we have seen in our work with teachers and students that the ideologies often accompanying named languages – and that inform such seemingly common-sense notions as native speaker, standard language, first/second language, and academic language – can be disrupted through pedagogical approaches that recognize and support students’ dynamic languaging and ways of knowing. In this section, we outline a few examples of pedagogical practices that, in dialogic relationship between theory and practice, have co-constructed and extended the theoretical understandings we have laid out here.

We start by restating our belief that pedagogical practices to teach, assess, learn about, engage, and challenge students must emerge from teachers’ stances and students’ meaning-making practices. This interplay makes up what García et al. (2017) call a translanguaging pedagogy, a framework for educators who wish to cultivate culturally and linguistically sustaining learning spaces (Paris & Alim, 2017) and who wish to respond to, or shift with, the corriente of the students’ languaging. A particularly valuable instance of widely and deeply applied translanguaging pedagogy is the collaborative initiative between The City University of New York and the New York State Education Department known as CUNY-NYSIEB.³ To take up a translanguaging pedagogy in this initiative, teachers were challenged to perceive their students anew, to challenge their own raciolinguistic ideologies and abyssal thinking. The teachers then designed opportunities for students to do the important work of school (i.e., engage meaningfully with content and integrate into their repertoires new linguistic features and ways of languaging). But the instructional and assessment designs of these teachers communicated that they see their students as capable, gifted, and already engaged in the kind of thinking and languaging expected of them in school. Thus, integral to this pedagogy, and to perceiving racialized bilingual students anew, is an interrogation of the ideologies that obscure those gifts and capabilities, and that negatively shape teachers’ (and others’) perceptions of them as students.

As CUNY-NYSIEB teachers and research teams worked together to design and implement lessons, analyze student work, and adapt unit plans and assessments, a picture emerged of what it looks like to take up translanguaging in different classrooms. For example, researcher Ann Ebe and English Language Arts teacher Charene Chapman-Santiago (2016) worked together to document students’ engagement with Inside Out and Back Again, a novel written in verse that contained passages in both English and Vietnamese. By analyzing the author’s translanguaging as a literary device, students had deep conversations about the text and were led to create their own translingual poetry, which enabled them to tell stories about their own families and cultures. This experience enabled Chapman-Santiago to tap into her own experiences as an immigrant from Jamaica and to face the “daunting” task of “incorporate[ing] all these children’s cultural identities and backgrounds” (CUNY-NYSIEB, 2020, p. 79). Instead of throwing her hands up and teaching only in English – a normative approach in ELA classrooms – Chapman-Santiago made space for her students’ home language practices, while also providing scaffolded and supportive instruction that helped them integrate English into their repertoires (For other examples, see CUNY-NYSIEB, 2020; see also García & Kleyn, 2016).

Teachers’ engagement with translanguaging theory emboldens them to design instruction by building centrally on the linguistic gifts they know their students possess. These gifts are often stifled by ideologies that see languages as homogeneous entities; that conceive of bilingualism as two sharply separate named languages; and that formulate language policies that are reproductive of these instances of monoglossic ideologies. Flores (2020) drew from his ethnographic research in a majority-Latinx elementary school to describe the framework he calls language architecture. Instead of organizing curriculum and instruction around remediating perceived lack of academic language in racialized bilingual students – a perception rooted in raciolinguistic ideologies that Flores takes pains to critique – teachers develop, with the support of researchers and teacher-educators, “new listening/reading subject positions that recognize the complex linguistic knowledge that their students have developed as part of their lived experience and make this central to the work that they are doing in classrooms” (Flores, 2020, p. 24). In taking up these new positions, teachers can perceive their students as language architects who already “navigate socially constructed linguistic boundaries on a daily basis” and, thus, “have unique affordances” (p. 25) for understanding many of the language and literacy practices expected of them in schools and on high-stakes exams.

For example, Flores describes a unit of instruction in Ms. Lopez’s second grade Spanish-English bilingual classroom centered around the book Abuela by Arthur Dorros. Flores documents the sophisticated attention to language that students brought to their close readings of the text, highlighting how this kind of reading – a skill students are expected to demonstrate on standards-based assessments – was part of “the language architecture that Latinx children from bilingual communities engage in on a daily basis [that] is legitimate on its own terms and is already aligned to [the standards]” (p. 28).

In a similar vein, Seltzer’s (2019a, 2019b, 2020) classroom-based work with 11^th grade English Language Arts teacher Ms. Winter (the teacher featured in one of the vignettes above) shaped what she has called a critical translingual approach, one that invites language and literacy teachers into engagement with theory and critical self-reflection and asks that they translate new theoretical understandings and reflections on their own positionalities into curriculum and instruction. With Ms. Winter, Seltzer co-designed a year of instruction that aimed to bring forth students’ translanguaging and translingual sensibilities (Seltzer, 2020). This work also sought to make space for students’ understandings of the links between language, race, power, and identity through engagement with translingual texts and their own translingual writing and text production. As Ms. Winter participated in this work, her students took up the invitation to connect the texts they read (and listened to and viewed) to their own experiences as racialized, language-minoritized people. Listening to her students and reading their translingual writing had the effect of transforming Ms. Winter’s raciolinguistic literacies (Seltzer & de los Ríos, 2018) and reshaped her ongoing efforts to forge a translingual English classroom that focuses on equity and anti-racism. As Seltzer, Winter, and her students explored language ideologies at work in their lives, they also honed the language and literacy practices expected of the students on the high-stakes standardized English exam they took at the end of the course.

What these examples from our own school-based research show is that the teachers with whom we collaborate are eager to take up theories that name and give voice to the work that many of them already do in the classroom. They are also eager to engage with these theories in ways that extend their practice, help them ask critical questions, and reflect on their positionalities and pedagogies anew. In our collaborations with teachers, we have found that as they explore new theories and take them up in their pedagogical practices, they also make them their own, shaping them in ways that align with their teaching. It is this reshaping of theory through practice – and through continuing collaborations with researchers and teacher-educators – that pushes the theory forward and creates more dynamic, sustaining, and equitable learning environments for racialized bilingual students.

Conclusion

This manifesto has framed our understandings of language and education from the inside out, and from our place on the side of the line obscured by hegemonies and ideologies that render racialized bilingual communities as deviant, deficient, and in need of remediation. We argue that any meaningful shift in the education of language-minoritized students must start from valuing the languaging and ways of knowing that prevail in families, communities, and yes, in many classrooms. By taking up the lens of translanguaging, the ways of languaging of racialized bilingual students can be seen, not as deviations from a monoglossic norm, but as those of full human beings who – like all human beings – make meaning by drawing from complex, interrelated linguistic-semiotic and multimodal repertoires grounded in deeply valued cultural-historical roots. This important shift in the perception of racialized bilinguals can disrupt oppressive raciolinguistic ideologies that thrive on the dominant side of the abyssal line. Guided by the writings of Boaventura de Sousa Santos, we invite everyone to engage in thinking – and thus in research and teaching – that effaces the lines of colonial logic and to join us in the creation of more equitable educational policies and practices.

We have purposely come up with new terms. We speak and write about racialized bilinguals, raciolinguistic ideologies, translanguaging, and a critical translingual approach. As Mignolo (2000) has taught us: “An other tongue is the necessary condition for ‘an other thinking’ and for the possibility of moving beyond the defense of national languages and national ideologies – – both of which have been operating in complicity with imperial powers and imperial conflicts” (p. 249). To be sure, perceptions of the language of racialized bilinguals have been shaped by ideologies that are not always reducible to nation-state or economic dynamics. But these perceptions are nevertheless part of broader racial-colonial distinctions that separate out legitimate from illegitimate ways of being in the world. In this imperial context, these distinctions permit the exclusion of minoritized populations from privileges granted to dominant populations, by ideologically constructing the minoritized population as inferior and undeserving of rights and resources. Racial-colonial distinctions also inform pernicious characterizations of the work of scholars targeted by racism and white supremacy as incomprehensible, political, ideological, superficial, trendy, or otherwise lacking in scientific validity or objective truth.

Our work emerges from heridas which we have both observed and experienced as learners and teachers. The force of these injuries as evidence of the need for fundamentally different educational approaches is not diminished because some on the receiving end, as in our case, have now achieved positions of relative comfort and local power and influence. We have witnessed the dynamic interplay among knowledge systems, words, and languages. And through this experience we have sought to open up a space of possibility, as we attempt to unsettle the abyssal line that we have inherited from colonial logics about race, language, and broader knowledges and lifeways.

Our path has not been linear. We have had to find the spaces, the cracks that the late Lillian Weber so eloquently spoke about at the City College of New York in the 1980s, so as to connect our experiencias personales with scholarly theory and educational practice. We know that our work on its own will not lead to the kinds of social transformations that may be needed for the creation of educational practices that we have advocated for and illustrated here. Perhaps our work merely creates ripples that can contribute to broader salutary effects on racialized bilinguals and their teachers. As Mignolo and Walsh (2018, p. 8) remind us: “Decoloniality … does not imply the absence of coloniality, but rather the serpentine movement toward possibilities of other modes of being, thinking, knowing, sensing and living, that is, an otherwise in plural”.

Disclosure of potential conflicts of interest

No potential conflict of interest was reported by the author(s).

Notes

1. We focus in this paper on the historical processes of white European colonization and their continued effect on those who now live in the U.S. and the U.K. We recognize, of course, that the processes of colonization and dominance over others have not been solely carried out by white Europeans.

2. The decolonial theory and approaches that we take up in this article have been advanced by scholars such as Boaventura de Sousa Santos Walter Mignolo, and Aníbal Quijano, whose work we cite here. They have also been developed by other Latin American scholars, such as Silvia Rivera Cusicanqui, Enrique Dussel, Arturo Escobar, Ramón Grosfoguel, María Lugones, and Nelson Maldonado-Torres, among others. In the Asian context, Kuan-Hsing Chen (2010) has been advocating a similar approach that he calls ‘deimperialization’.

3. Four of the authors participated in this initiative, known by its acronym, CUNY-NYSIEB. Otheguy served as principal investigator, García as co-principal investigator, Flores as founding director, and Seltzer as the third director.