Dismantling Persistent Deficit Narratives About the Language and Literacy of Culturally and Linguistically Minoritized Children and Youth: Counter-Possibilities

Despite decades of efforts, deficit narratives regarding language development and use by children and students from historically marginalized backgrounds remain persistent in the United States. Examining selective literature, we discuss the ideologies that undergird two deficit narratives: the notion that some children have a “word gap” when compared to their White middle-class peers, and students must develop “academic language” to engage in rigorous content learning. The “word gap” concept came from a study wherein a group of young children in low-income families heard fewer words than those in middle-class families. It assumes that language can only be acquired in one way—vocabulary exchange from one parent to one child—and ignores decades of research on diverse pathways for language development. We highlight an alternative perspective that language development builds on children’s experience with cultural practices and the harm on minoritized children by privileging a specific form of vocabulary acquisition. The second deficit narrative concerns “academic language,” a concept championed by scholars aiming to address educational inequity. The construct runs the risk of undervaluing the potential of students from historically marginalized backgrounds to engage in learning using language that is “informal,” nonconventional, or “non-native like.” It also is sometimes used as a rationale to relegate students to special programs isolated from more rigorous academic discourse, thus ironically removing them from opportunities to develop the academic registers they are deemed to be missing. We explore alternative frameworks that shift the focus from linguistic features of academic talk and texts as prerequisites for academic work to the broad range of linguistic resources that students employ for academic purposes in the classroom. Finally, we turn to a positive approach to youths’ language development and use: translanguaging by multilingual learners and their teachers. Translanguaging demonstrates the power of a resource-oriented perspective that values students’ rich communicative repertoires and actively seeks to disrupt language hierarchies. We argue that this approach, however, must be considered in relation to the broader social context to meet its transformative aims. Together, our analysis suggests counter-possibilities to dismantle deficit-oriented narratives and points to promising directions for research and practices to reduce inequity in education.

Despite decades of efforts to dismantle deficit perspectives on culturally and linguistically minoritized students in United States education, and a wealth of literature that has provided evidence to counter these perspectives, deficit narratives regarding students' language development and language use remain persistent. Our goal in this paper is to focus on examples where deficit perspectives have taken hold, and to consider alternative approaches in those domains. We begin by contrasting deficitbased and strengths-based perspectives on language development and educational outcomes. We then consider the broader, and orthogonal, contrast between individual cognitive approaches to the study of language development versus more social-pragmatic contextual approaches. We explore three topics of studies where these different perspectives are particularly relevant.
Deficit views involve a narrow focus on what students do not have or cannot do, derived from a long-lived perspective that attributes the failure of individuals to internal or presumed deficiencies of their families and communities. Deficit lenses are typical of research that examines "gaps" of different kinds, positioning individuals, rather than structural inequities, as the subjects of scrutiny. Valencia (2010) describes this process as identifying differences between groups, labeling them as causes, and then creating interventions as remedies. A hallmark of deficit approaches is their tendency to propose unidimensional, commonly-understood constructs or metaphors to explain educational failure: "[b]ased on the 'law of parsimony,' deficit thinking is a type of cognition that is a relatively simple and efficient form of attributing the 'cause' of human behavior." (Valencia, 2010, p. 22, emphasis in original). By focusing narrowly on individuals, deficit thinking obscures structural factors, like school segregation, disinvestment, or tracking. This approach also ignores the potential harm done to minoritized populations by standards and curricula that ignore non-dominant disciplinary epistemologies, for example, the invention of mathematical "illiteracy" of Black students (Martin, 2019) or the "settled expectations" that restrict the content and form of the science valued and communicated in science education (Harris, 1995;Bang et al., 2012).
When deficit thinking is applied in classroom settings, the results often include segregation of students who are viewed as inferior, and arguments about the educability of certain groups of students that rely on pseudoscientific beliefs in cultural or genetic deficits. Further, deficit views often are embedded in what Bourdieu and Nice (1977) described as defense of the doxa, or the deep unquestioned assumptions of a society (see also Valencia, 2010;Valdés, 2017). A good case in point for unquestioned assumptions (or doxa) that support deficit thinking is the notion of coloniality (Quijano, 2000). Whereas coloniality is more fully described in a later section, it is important to introduce here the still-alive colonial idea that positions non-European people and cultural practices (which include language practices) as primitive and, therefore, deserving of their places within social hierarchies. These types of beliefs, which originated alongside colonial relationships of power, can remain unquestioned when we presume that students need to use privileged forms of English in order to engage with certain forms of intellectual work. As such, it is important to take issues of power and larger historical contexts into account when considering the role of cultural experience in language development as it manifests in educational settings.
In contrast, scholars have explored a range of alternatives to "blaming the victim" (Valencia, 2010, p. 19) for understanding and addressing persistent inequalities in educational outcomes. Some have put the focus on inequalities in educational systems and broader social and economic structures, such as Valencia (2010), who has argued for the importance of understanding how "racialized opportunity structures lead to racialized achievement patterns" (p. 21). Others have taken frameworks commonly used for explaining disadvantages facing racially and linguistically minoritized children and families to shine light on the unrecognized and undervalued resources that they possess. Yosso (2005), for example, reexamines the notion of social, cultural, and linguistic "capital"-often used to highlight what minoritized people are missing-to instead explore the "community cultural wealth" that they bring. Such resources include particular kinds of cultural, linguistic, and social capital not held by dominant populations, but also aspirational, navigational, and resistant capital. Individuals, according to Yosso (2005), have developed this wealth not in spite of but because of their familial, community, and cultural experiences navigating multilingual and intercultural contact as well as systems of oppression.
Another approach to avoiding deficit perspectives is using a cultural-historical framework in developmental and educational research. Gutiérrez and Rogoff (2003) argue that a culturalhistorical approach can help researchers avoid the deficit thinking that treats culture as a static individual trait and cultural groups as homogeneous (see also Medin et al., 2010;Akhtar and Jaswal, 2013;Callanan and Waxman, 2013;Rogoff et al., 2017). This framework brings us to the contrast between individual cognitive approaches and social-pragmatic contextual approaches to the study of language development. Individual, cognitive approaches have dominated the field of language development at least since the rise of Chomsky's theory of syntax (e.g., Chomsky, 1959). Bruner (1983), Tomasello (1996Tomasello ( , 2003, and others taking sociocultural and interactionist perspectives have argued, however, that it is not necessary to presume a genetic endowment to explain language. Instead, social and cognitive capacities such as infants' ability to infer intentions of others (e.g., Baldwin, 1993;Tomasello and Akhtar, 1995;Akhtar et al., 2019) and to recognize communicative routines (e.g., Ninio and Bruner, 1978) can explain language learning without relying solely on innate structures. Social-pragmatic accounts of language learning have gained traction over the past 50 years, and yet individual cognitive accounts are still dominant. In second language and bilingual language development, similar debates have pitted individual structural theories against theories that forefront the social context of language learning (Firth and Wagner, 1997;Firth and Wagner, 2007;Lantolf and Thorne, 2006;Atkinson, 2011;Dixon et al., 2012). Hawkins (2019) outlines the fundamental shift associated with a sociocultural approach to language: From a sociocultural perspective, language does not stand alone; a language cannot be conceived as a codified set of structures, grammars, and lexical items. Rather, languages shift and change across context, users, places, and time . . . language is entangled with other semiotic resources to convey meaning in virtually every communication, and how meanings are made (between people) in large part depends on cultural models of communication and cultural interpretations of semiotic resources (p. 15).
From a sociolinguistic perspective, van Lier and Walqui (2012) described the limitations of viewing language as "form" or "function" alone, arguing instead that language is an "inseparable part of all human action, intimately connected to all other forms of action, physical, social, and symbolic," and "thus an expression of agency, embodied and embedded in the environment" (p. 4). Linguistic form and function, in other words, are "subservient to action," and language itself "ceases to be an autonomous system, but is part of larger systems of meaning making" (van Lier and Walqui, 2012, p. 5). Others in the field of second language acquisition have criticized formalist and cognitivist theories (Ortega, 2009;Valdés et al., 2014), arguing to replace input and output models with a focus on language as "a communicative repertoire that is apprenticed in social practice" (Valdés et al., 2014, p. 21). Similar moves have occurred in literacy studies. Lea and Street (2006), for example, describe alternatives to what they called an individual, cognitive "study skills" approach that, they argue, privileges "surface features of language form" and views literacy as transferring "unproblematically from one context to another" (pp. 368-369). Alternatively, they propose an "academic literacies" approach privileging literacy practices within disciplines and larger academic settings (p. 368).
One advantage of more social approaches to the study of language is that seriously considering the context of language use makes it easier to highlight the strengths that language learners bring to the task. As such, the social-pragmatic accounts afford a research approach to recognizing cultural assets for language development in early childhood and facilitate the thinking of instructional practices as ways to foster these assets in the classroom setting. By examining what young children and school-aged youth are able to do with their linguistic resources, rather than comparing language to a presumed norm, researchers may more easily see the purposes of language use in particular cultural contexts and avoid deficit assumptions.
To dismantle the persistent deficit views of language development, in this article we explore selective literature in three topics-the word gap, academic language, and translanguaging-and offer our perspectives as researchers in education and developmental psychology regarding their role in both the construction of deficit discourses and the potential to dismantle such discourses. To be clear, the purpose of this paper is not to provide comprehensive reviews of each topic, which are available elsewhere (e.g., Cummins, 2000;DiCerbo et al., 2014;Poza, 2017), but to connect these three topics to elucidate why the deficit views have remained so powerful and to offer suggestions on how to move past this limiting framework. The cited work was chosen to highlight some gaps in the deficit views and to illustrate how the three topics may converge to support the alternative assets-based views. As a preview, we first discuss the dominant language ideologies that undergird current "word gap" and "academic language" narratives and propose theoretical and conceptual alternatives. We then turn to translanguaging by multilingual learners in classrooms, which we contend presents an alternative approach with potential to challenge linguistic hierarchies, but that can also unintentionally contribute to reifying such hierarchies. Finally, we close with a discussion on overarching issues, research gaps and potential developments in the field.

WORD GAP
The 30 million word gap concept grew out of Hart and Risley's study (1995), which found that a group of young children in lowincome families heard fewer words than a group of children in middle-class families. More than 25 years old, the study continues to have staying power in education and policy realms and to be cited as a cause for larger "achievement gaps" in low-income children (e.g., Golinkoff et al., 2019;Romeo et al., 2018;Rowe, 2018;Walker et al., 2020). The word gap is mentioned on the websites of the US Department of Health and Human Services and the National Association for the Education of Young Children, and has inspired interventions such as Providence Talks. Yet, in a recent attempt to replicate Hart and Risley's (1995) findings, Sperry et al. (2019) did not find a large "language gap" between low-income and middle-class children. The work by Sperry et al. (2019) has received sharp criticism (e.g., Golinkoff et al., 2019), suggesting that the word gap argument maintains a powerful hold (for an overview of the language gap debate see Kuchirko, 2019). Here we argue that the word gap narrative is emblematic of other deficit-oriented frameworks for understanding the role of language in development and education. The narrative also points to the lingering influence of research orientations and methodologies that have conceptualized language development as a process that occurs within the mind of the individual child, without attention to the sociohistorical and structural contexts in which the child is situated.
Proponents of the word gap concept are committed to young children's learning and believe that ameliorating the perceived gap will lead to educational equity. However, the word gap concept is rooted in a cognitive theory that positions language as an individual ability that develops through child-directed speech, can be measured in a decontextualized way, and can be compared across individuals, like body height and body weight. But even body height cannot be compared completely out of context. Individuals of the same height can be seen as unusual in one cultural community and well within the norm in another community. To continue the metaphor, the word gap concept acts as a "ruler" to define what is normal and implies this standard level of vocabulary exposure is necessary for academic achievement. We do not dispute that child-directed speech is important; however, the focus on vocabulary exposure and dyadic conversation reflects the dominant norm of language practices by White middle-class families and fails to capture the diversity of social practices around language across different communities (Ochs and Schieffelin, 1986;Lieven, 1994). Rowe (2018) argues that studies of the word gap have taken a sociocultural approach, in that they "stress the importance of children's early social and communicative experience for language development" (p. 122). The Emergentist Coalition Model (e.g., Hollich et al., 2000), for example, attempts to integrate the social-pragmatic approach into consideration by factoring in social cues such as eye gaze and pointing in word learning. However, the assumptions made in these accounts seem to contradict sociocultural perspectives, especially the view that the quality and quantity of parents' language "input" is the major determinant of children's vocabulary development, and that unpacking the reasons for this deficient input should be the basis for designing interventions.
A number of scholars have demonstrated, however, that although dyadic conversation is important, it is not equally emphasized in all cultures, and is not the only way for children to develop language (Sperry et al., 2019). Children's everyday experience with language involves a vast array of types of discourse (e.g., Heath, 1983;Miller and Fung, 2012), and children's engagement in types of talk depends on cultural practices. Everyday language experience can include personal storytelling, arguments, jokes and teasing, and other verbal and nonverbal forms of language socialization (e.g., Gaskins and Paradise, 2010;Miller, 2014;Rogoff, 2014), and children also learn language through overhearing speech (e.g., Akhtar, 2005;Shneidman et al., 2009).

Conceptual Critiques
Specifically, the original word gap argument assumed a singular pathway for early language developmentvocabulary exchange from one parent to one child, restricting the observation and analysis of children's language experience to the word level and focusing on parental utterances that are directly addressed to the focal child (Hart and Risley, 1995). As such, subsequent research using this methodology ignores decades of research on how children learn a language in diverse contexts and leaves out discursive practices shown to be important for children's language development, rendering it a partial and biased framework of analysis.
Although dyadic conversation contributes to language development, and is perhaps the most relevant practice for White middle-income children, it is not the only way for children to develop language (Lieven, 1994). Children learn language in cultural settings where parents do not often address them directly (Ochs and Schieffelin, 1986). Even within the Western middle-class setting, observational studies have indicated that toddlers monitor third-party conversations when not directly engaged in the conversation (Dunn and Shatz, 1989). Children as young as 18 months old have been shown to learn words through overhearing when other cognitive demands are not too high (Floor and Akhtar, 2006). In a study on word learning, for example, 2.5-year-old children either were spoken to directly or overheard a conversation that included a label for a novel object or a novel verb. The children were able to learn novel labels and novel verbs by overhearing speech to others at the same rate as when being directly spoken to (Akhtar et al., 2001). Furthermore, when relating children's learning through overhearing to their social experience, Shneidman et al. (2009) found in their sample of 20-month-olds that the extent to which the toddlers picked up novel labels through overhearing was positively correlated with the number of hours that they spent with multiple adults or older children.
It is important to acknowledge that some studies of the word gap have distinguished between child-directed speech and overheard speech. For example, Weisleder and Fernald (2013) found, in a sample of low-SES Spanish-speaking United States families, that the frequency of child-directed speech to 19-monthold children predicted both children's vocabulary and the efficiency of their language processing at 24 months. In contrast, the number of overheard words did not predict children's language measures. This finding supports the claim that the number of words spoken to children may be important for children's word learning. And yet, contrary to the Hart and Risley (1995) argument, Weisleder and Fernald (2013) found a great deal of variability in the number of words spoken to children within their low-income sample, and this variability was not related to parents' education or SES.
In addition to the dominance of the individual cognitive perspective on language development, the persistence of the word gap narrative can be attributed to broader, raciolinguistic ideologies around language and power in the United States (Rosa and Flores, 2017). Focusing analysis on dyadic conversation and word quantity privileges White middle-class norms of speaking with children and disregards the linguistic strengths that children develop through other discursive and communicative practices in their communities. The word gap narrative also complements the values of meritocracy and capitalism by relating social class to individual success (Avineri et al., 2015). By side-lining poverty and educational inequity, and focusing on children's language learning as a deficit related to individuals, families, or communities, the word gap narrative presents itself as an alluring quick fix to educational injustice. The narrative has also informed current educational policies and curricular frameworks; for example, research on the word gap was presented as part of the initial justification for the concept of "Developmentally Appropriate Practice" (or DAP) put forward by the National Association for the Education of Young Children (NAEYC), and researchers have argued that this affirmation of the word gap is likely to have powerful impact on educational practices with young children (Souto-Manning and Rabadi-Raol, 2018). 1 Growing critiques of the word gap narrative represent more than merely a difference of interpretation. Instead, they clarify the harm that can be done to families by accepting an individual cognitive perspective on language development that positions marginalized children and their families as linguistically deficient. Using this approach to rectify inequity has resulted in ineffectual-or even harmful-interventions. For example, Adair et al. (2017) observed that teachers of 1st and 2nd grade Latinx children in Texas cited the word gap as evidence that their students were not able to engage in active learning. Without more vocabulary, the teachers argued, children needed to be quiet listeners, and children in their classrooms seemed to have accepted their teachers' view that learning must happen through keeping quiet and listening to the teacher. Adair et al. (2017) argued that the word gap causes harm by depriving students of dynamic and complex learning opportunities. In another critique, Morelli et al. (2018) argued that the intervention study undertaken by Weber et al. (2017) placed too much emphasis on talk and ignored indigenous ways of thinking.

Recognizing Strengths
The word gap argument narrowly focuses language development on the number of words heard by children in the mother-child dyadic context. However, children's everyday experience with language typically occurs at the discourse level across a wider array of contexts, as demonstrated by the wealth of evidence from research of language socialization (Ochs and Schieffelin, 1986;Gaskins and Paradise, 2010;Miller, 2014;Rogoff, 2014). For example, story-telling is a common oral practice found in diverse cultural communities (Heath, 1983;Miller et al., 2005). A crucially relevant body of work by Miller and others (e.g., Lin et al., 2012;Miller and Fung, 2012) detailed children's active participation in personal storytelling, a joint social practice that involves the child, caregivers, and siblings, and routinely occurs in family conversation across different cultural communities. Children participate in different roles depending on the values emphasized in their cultural community. For example, middleclass families in Taipei, Taiwan tend to engage their children in personal storytelling with the focal child serving as a bystander and a co-narrator (Lin et al., 2012). At the age of 2.5 years, children in Taipei, Taiwan actively listened and readily contributed to the narration of past events that frequently included their misdeeds for didactic purposes. From 2.5 to 4 years of age, children in Taipei gradually increased their contribution as a co-narrator while their role of a bystander remained prominent over time. As a comparison, children in the Longwood (pseudonym) neighborhood of Chicago, who were raised in a primarily European-American middle-class community, more frequently took and maintained the role of co-narrator, rather than bystander, across the age range. Although the Longwood children also sometimes took dual roles of co-narrator and bystander in personal storytelling, active listening was observed less frequently in their bystander role than was seen with the Taipei children. The stark contrast shown in this research (e.g., Lin et al., 2012;Miller and Fung, 2012) is linked to the underlying meaning system of cultural ideologies and principles that motivate, privilege, and maintain the unique language experiences available to children.
European-American children's everyday language experience can differ profoundly depending on their socio-economic class backgrounds, and these differences also transcend the number of words heard (e.g., Miller and Sperry, 1987;Miller, 1994;Wiley et al., 1998). For example, children of working-class families in the Daly Park (pseudonym) neighborhood of Chicago were immersed in everyday discursive practices in a host of ways different from the middle-class version of personal storytelling in the Longwood neighborhood. First, co-narration of personal stories was deeply valued by working-class families, which was observed much more frequently in Daly Park than in Longwood. Second, conflict episodes in which mothers contradicted children's different versions of a story tended to occur in a direct and matter-of-fact manner in the working-class families in Daly Park. The children participated in opposition exchanges that were often lengthy, standing up for themselves and defending the right to express their own views. In contrast, oppositional exchanges were almost non-existent in the personal storytelling in the middle-class families at Longwood; the right to tell the child's different version of story was readily given, and taken for granted (Wiley et al., 1998). These culturally unique ways to engage children in personal storytelling have been observed in European-American working-class families across different communities (Miller and Sperry, 1987;Cho and Miller, 2004) and in African-American working-class families as well (Corsaro et al., 2002).
The working-class version of personal storytelling showcases the linguistic strengths and narrative skills of children who are immersed in artful performances. In a striking set of examples provided by Miller et al. (2005), working-class parents in Daly Park shifted from the past tense to the historical present tense, fluidly built up punchlines, and deployed parallel constructions through artful storytelling in front of their 2-to 4-year-olds. Compared to their middle-class counterparts, these children also heard more verbs of emotion and attributions as a bystander or co-narrator in the telling and reconstruction of everyday negative experiences that the family encountered. The assets for language development by children from working-class families can only be recognized when researchers consider children's discursive experiences as valuable practices of a cultural group in their own right, rather than reducing them to a variable of income level.
These studies highlight how studying language as a social practice can make the linguistic strengths of children in minoritized communities visible. Focusing solely on dyadic conversations or on individual cognitive processes, however, can obscure these strengths and highlight language practices commonly associated with White middle-class parenting.

ACADEMIC LANGUAGE
Another idea that has had considerable purchase, particularly in educational research and practice, is the notion that elementary and secondary students from certain linguistic backgrounds are in need of special instruction in "academic language" because their current language practices deviate from those that some consider essential for academic success. This argument, using different terms (e.g., academic English and the language of schooling) and different constructs, has been advanced by psychologists, linguists, literacy scholars, and educational researchers (see Cummins, 2000;DiCerbo et al., 2014;Jensen and Thompson, 2020). Tellingly, dosages of "academic language" instruction are most often prescribed for (a) speakers of languages other than the main language of instruction in schools (e.g., "English Learners" in United States schools) and (b) speakers of non-dominant varieties of that same language of instruction (e.g., speakers of what linguists and educators have referred to as African American Vernacular English, Black English, or Black Language). Although important differences exist in the linguistic repertoire of students who grew up speaking more than one language (including English) since early childhood, those not exposed to significant amounts of English before entering school, and those who grew up speaking non-dominant varieties of English (Heath, 1983;Zentella, 1997), all these populations are subject to the argument that their language, no matter how nativelike or fully developed, is not the right kind for school. In the United States, for example, it has been argued that English learners may be developing quite a lot of English, but it is often "social", "informal", or "everyday" English, and not what is needed in school. In fact, this distinction between the linguistic features of "academic" language and those of its putatively non-academic counterpart have dominated both empirical research (see review by DiCerbo et al., 2014) and guidance for educational practitioners in the field of Teaching English to Speakers of Other Languages (Short et al., 2018). Similarly, the argument goes, speakers of stigmatized varieties of English such as Black English may be "native" monolingual speakers of English, but their dialect gets in the way of their academic success. Baker-Bell (2020), drawing on Alim and Smitherman (2012), argues that the notion of "academic language" is actually a proxy for White Mainstream English; indeed, decades of empirical research documents both the linguistic sophistication of Black English and its stigmatization in United States schools (see also Baugh, 2004;Green, 2004;Wolfram, 2004). Godley and Minnici (2008), in summing up this perspective, quoted a teacher's comments about one of their Black students as representative of this position: "Rajid can't do challenging work-just listen to the way he talks" (p. 30).
In this section, we briefly discuss the well-intentioned arguments that have been used to highlight the "academic language" vs. "everyday language" contrast and point out problems with this approach from the perspective of equitable learning opportunities for students from nondominant linguistic, cultural, and racial backgrounds. We then offer alternatives that recognize and build on the linguistic and intellectual assets that students from minoritized backgrounds already bring with them to academic settings, without diminishing the importance of students' expanding their linguistic repertoire to include language used by a variety of disciplinary audiences. As discussed earlier, sociocultural perspectives on language development and language use help to illuminate students' strengths, which often emanate directly from students' own background, including their family and community experiences and language practices, as well as the ways in which marginalized students can gain access to opportunities to develop more dominant discourses helpful for success in academic settings.
To be clear, the concept of academic language has been championed by scholars committed to addressing educational inequity. For example, Cummins (1981Cummins ( , 2000, an educational psychologist whose commitment to improving education for linguistically minoritized populations is undisputed, proposed the controversial but still-influential distinction between "basic interpersonal communication skills" (BICS) and "cognitive academic language proficiency" (CALP) out of concern that students from minority language backgrounds were being transitioned out of bilingual support programs too quickly or identified as having learning disabilities. Cummins (2000) argued that students' relatively rapid development of BICS in English masked their lack of competence in the more cognitively demanding and "decontextualized" CALP. English language proficiency tests which, according to Cummins, measured BICS but not CALP, were incorrectly indicating that students had mastered English, leading school officials to seek explanations other than English language proficiency, such as learning disabilities, for some students' lack of academic progress. In justifying his distinction, Cummins (2000) drew on a wide variety of language-related fields and constructs, from both monolingual and bilingual contexts, to argue for the validity of his distinction, including Vygotsky's distinction between spontaneous and scientific concepts (Vygotsky, 1962;Kozulin, 1998), Bruner's communicative and analytic competence (Bruner, 1975), and Snow's distinction between contextualized and decontextualized language (Snow et al., 1991). More recently, in an attempt to envision interventions to facilitate the language development of students who are developing additional languages as well as speakers of stigmatized varieties of English, scholars have attempted to articulate a number of linguistic dimensions that make oral and written language in school settings distinct from "everyday" language (e.g., DiCerbo et al., 2014;Lesaux et al., 2016;Uccelli and Phillips Galloway, 2017).
Depending on the context in which it is used, however, the focus on academic language as a set of linguistic features contrasting with everyday uses of language presents a number of problems, both conceptually and in terms of potentially deleterious impacts on the very students that those who have advanced the construct have been interested in serving. If either teachers or assessment developers look only for students' acquisition or use of features of language that have been predetermined to be "academic" (i.e., different from language likely to be used in "everyday" settings), then students' contributions to the learning tasks at hand may be missed. As we will discuss throughout this section, such perspectives have "material consequences" for students from a range of nondominant linguistic and cultural backgrounds (Martínez and Mejía, 2020, p. 53; see also Baker-Bell, 2020), and alternative approaches are necessary to envision and enact more equitable learning opportunities.

Conceptual Critiques
Even some who themselves have catalogued linguistic features used in academic settings acknowledge constraints associated with this approach. For example, Snow and Uccelli (2009) reviewed the literature to compile an "inventory" of multiple dimensions marking some language as "more colloquial" and others as "more academic," but they also raised concerns about such lists: [D]ozens of traits have been identified that contrast with primary or colloquial language and that might function as markers of academic language, but it is unclear that any of them actually defines the phenomenon. Any of these traits might be present in casual spoken language: Is it their co-occurrence that defines some language as academic? Is it their frequency? . . . How does the list . . . help us with the tasks of assessment or instruction? (p. 121).
Marítnez and Mejía (2020) go further, arguing that academic language, rather than an "empirically observable set of linguistic features," is actually an "idealized notion of the kinds of language valued in schools" (p. 53, emphasis added). If, as Martínez and Mejía (2020) argue, "'[a]cademic language' is an idea," then it is worth interrogating that idea. Educators often confuse the notions of "dialect" and "register" (Ferguson, 1994), attempting to change students' home and community language practices (dialects) instead of working with them to expand their repertoire of language uses for particular audiences and purposes (registers).
Depending on how it is used, the construct of academic language can also conflate language and literacy, in that reading and writing on many English language proficiency tests measures academic achievement more than English language proficiency (Wiley, 1996;Brooks, 2020). Both researchers and educators also often confuse lists of linguistic features present in academic texts that students are expected to read with characteristics we would expect to see in language that students themselves produce to engage in academic work (Bunch and Martin, 2020). However, even instructors and scholars in higher education consistently use features of "informal" or "social" language in their academic work (Bamford and Bondi, 2005;Bunch, 2014); navigating higher education requires a wide range of oral and written language practices, including those closer to the "conversational" end of the spectrum (Biber et al., 2002). Approaches to academic language that focus only on linguistic features of academic texts also often conflate text complexity (features of the text itself) with text difficulty, i.e., the degree of challenge that individual readers may face accessing a particular text based on their own language proficiency, background knowledge, interest level, and support provided see also;Bernhardt, 2011). The conflation ignores the fact that even linguistically and structurally "simple" texts may be difficult without background knowledge, support, and a motivating reason to read, while the most "complex" texts may be accessible with background knowledge, support, and high levels of interest in reading . Finally, with regard to linguistic complexity more generally, scholars have pointed out that (a) all language is complex, just in different ways; (b) all language is contextualized, again in different ways; and (c) all language is capable of doing cognitively challenging work (Bartolomé, 1998;MacSwan and Rolstad, 2003;Schleppegrell, 2004).
As is the case with the word gap discussed earlier, it can be argued that, despite the best intentions, some conceptions of academic language are rooted in deficit orientations that contribute to maintaining the dominance of White middleclass language ideologies (Baker-Bell, 2020). MacSwan (2020), for example, describes the purchase of "standard language ideology": "the view that the language variety of socioeconomic elites is intrinsically more complex than other varieties" (p. 29). Similarly, Flores and Rosa (2015) argue that "discourses of appropriateness" invoke a "conceptualization of standardized linguistic practices as objective sets of linguistic forms" that are actually "raciolinguistic" ideological constructions that "conflate certain racialized bodies with linguistic deficiency unrelated to any objective linguistic practices" (p. 150). As Paris and Alim (2014) have suggested, it is worth asking what alternatives are available if "the goal of teaching and learning with youth of color was not ultimately to see how closely students could perform White middle-class norms but to explore, honor, extend, and problematize their heritage and community practices." We devote the rest of this section to exploring such alternatives.

Recognizing Strengths
As we argued in the introduction, one potential alternative to dominant approaches toward academic language is to shift away from viewing language as a set of structures toward a social practice or action (Valdés et al., 2014;van Lier and Walqui, 2012; see also; Lea and Street, 2006). Viewing language and literacy as social practices allows educators-and researchers-to recognize what students are able to do in academic settings using the wide range of language and other semiotic resources already available to them. Rather than asking "what are the features of academic language?" we might ask "what can students DO with language to engage in academic work?" (Bunch, 2014). Rather than focusing exclusively on the linguistic features of particular academic texts, it can be productive to examine the nature of the language practices that students are called upon to use to participate in the participant structures, transactions, genres, and other social practices common in classrooms (Bunch and Willett, 2013). For Frontiers in Education | www.frontiersin.org July 2021 | Volume 6 | Article 641796 example, Bunch (2006Bunch ( , 2009 explored the "transactions" that 7th grade students, most of whom identified as Latina or Latino, skillfully negotiated in the group work and whole-class structures, including delivering oral presentations to their peers while simultaneously addressing the teacher and responding to questions. Even when focusing on language structures, however, a number of studies have demonstrated that students from linguistically and culturally marginalized backgrounds are able to engage in academic work using language that would not be considered "academic" according to many definitions of the construct. When focusing on the linguistic structure of students' utterances, Bunch (2006Bunch ( , 2014 found that students adapted the language used in conversations surrounding the academic tasks according to audience and purpose. During a discussion while interpreting political cartoons, small groups of 7th-grade students first used informal, interactive utterances to engage in a discussion about the cartoon, using what Bunch called the language of ideas. When preparing to present to their teacher and classmates, they shifted to the language of display, characterized by linguistic features more commonly used for communicating in academic presentations. For example, they clarified antecedents of pronouns and minimized interpersonal, informal discourse markers (e.g., "like" and "you know") for the larger group. Bunch (2006Bunch ( , 2014 argued that these differences could not be explained simply as students' transitioning in their conversation from using "everyday" to "academic" language. That is, students using the language of display did not necessarily advance the academic work of the group, while some of the most important insights in the group's interpretation of the cartoon came from students using the language of ideas. In fact, the more informal language might have productively facilitated, rather than detracted from, the group's intellectual work, as students interacted more naturally as they discussed the key ideas. Thus, both the language of ideas and the language of display were critical for academic work in this setting, albeit in different ways. In short, students' intellectual contributions might have been missed using lenses focusing on the presence or absence of academic language. Rodriguez-Mojica (2018) also challenged traditional notions of academic language, using a different approach to identify how students use the language of ideas. Drawing on the sociolinguistic notion of speech acts (Austin, 1975;Searle, 1976;Bachman, 1990;Flowerdew, 2013), defined as units of meaning-making that transcend grammatical structure or ideological notions of "standard" language, she explored language use among emergent bilinguals in Fourth Grade English Language Arts (ELA) classrooms. The research highlighted seven speech acts that were particularly relevant to academic ideas and tasks related to ELA speaking and listening standards: providing feedback, making requests for clarification, organizing peer talk and activities, indicating one is following along with the discussion, making supportive assertions, describing a partner's ideas, and attempting to save face following a mistake. Notably, Rodriguez-Mojica's analysis showed that even students classified at the lowest levels of English language development and ELA performance on state standardized tests were able to navigate a range of academic speech acts in English: These students "made academic comments, attempted to explain and describe, sought clarification, and posed and responded to questions all in English"; these accomplishments would have been missed by measuring English language proficiency "by how closely speech adheres to traditional notions of academic language" (pp. 57-58).
Beyond identifying ways of valuing the use of marginalized linguistic forms, we call for attention to the disciplinary practices that students are called upon at the heart of subject-area learning in schools and the ways in which students might leverage nondominant linguistic and cultural practices to engage in these practices (Lee, 2001). Following Valdés et al. (2014), we conceive of disciplinary practices as encompassing the conceptual understandings, analytic skills, and language and literacy practices at the heart of the discipline. The focus on students' use of particular linguistic features often comes at the expense of recognizing what students are actually doing and learning, and talking about. Using disciplinary lenses can help illuminate how students from racially, culturally, and linguistically minoritized backgrounds successfully use language and literacy practices, including those they have expertise in from their homes and communities but that are often devalued in educational settings, to engage in key subject-area practices (e.g., Orellana, 2009;Martínez and Mejía, 2020).
Other scholars have explored how speakers of Black English (also referred to as African American Vernacular English, African American English, and Black Language) use linguistic and stylistic features associated with these varieties of English to engage in disciplinary work (Baker-Bell, 2020;Lee, 2004;Lee, 2007). Lee (2001), for example, highlighted how Black students' home and community language practices, including "language play" such as "signifying," are similar to the strategies valued in literary analysis. As Lee (2001) explained, one particular category of signifying is "playing the dozens," which takes the form of ritual insults often beginning with "your mother . . ." (see Percelay et al., 1994, p. 49). There is similarity between these linguistic practices and the inferences that students make when analyzing literary texts: Signifying always involves indirection and double entendre and invites participants to look beyond the surface meaning to subtle interpretations to be inferred. It is vivid in its use of metaphor and often involves satire, irony, and shifts in point of view. African American adolescents who routinely participate in such talk make tacit use of strategies for interpreting metaphors, symbols, irony, and satire (Lee, 2001, p. 100, p. 100).
In a high school classroom taught and researched by Lee (2001), students used a number of aspects of African-American Vernacular English to engage deeply, vigorously, and thoughtfully as they analyzed a novel at the focus of instruction: "the talk among the students is entirely in African-American English Vernacular, not simply in terms of vernacular syntax forms, but more importantly in terms of the Frontiers in Education | www.frontiersin.org July 2021 | Volume 6 | Article 641796 performance of the discourse. Students signify on one another, display body language for emphasis, and reflect a rhythm and prosody in their speech that is dramatic and culturally Black" (p. 108).
As the examples in ELA demonstrate, a focus on disciplinary practices will illuminate what students from non-dominant backgrounds can do to engage in academic work in ways that would be missed by focusing on linguistic features of academic language. Supporting evidence has been obtained in other subject areas as well, including mathematics (Moschkovich, 2007;Moschkovich, 2008;Moschkovich, 2015), science (Rosebery et al., 1992;Lee, Quinn, and Valdés, 2013), and history (De La Paz et al., 2016;Bunch and Martin, 2020). However, it is important to acknowledge that disciplines themselves are social, cultural, and ideological constructions (Medin and Bang, 2014). Lea and Street (2006), for example, argue that academic literacies must identify subject-based discourses and genres, but that it is also essential to critically examine the "institutional nature of what counts as knowledge in any particular academic context" (p. 269) including the role of epistemological and social processes, power relations, and social identities.
We conclude this section by pointing out how educators can bridge students' home language practices and the language more commonly used in academic settings, an idea that has been explored for many years with regard to monolingual English speakers (e.g., Heath, 1983). A disciplinary focus can help illuminate the ways in which educators can simultaneously recognize students' existing linguistic resources and create conditions under which students can expand that repertoire to include language practices commonly used in disciplinary communities. For example, using the Vygotskyian concept of mediation, Gibbons (2003) described how teachers and English language-learning students from low SES backgrounds in an Australian elementary science classroom collaborated to transform students' initial ways of talking about scientific phenomena into the "specialist discourse" of the curriculum. To illustrate, in a lesson on magnetism, students' language changed from experimenting in small groups ("Look, it's making them move. Those didn't stick"), to relating their experiences to their teacher and classmates ("We found out the pins stuck on the magnet") and finally writing reports about the experiment ("Our experiment showed that magnets attract some metals") (p. 252). Drawing on systemic functional views of language (Halliday, 1985;Halliday, 1993;Halliday and Hassan, 1985), Gibbons argued that specialist discourse can be understood as variation in register at one end of a "mode continuum": "The continuum reflects the process of formal education itself, as students are required to make shifts within an increasing number of fields and to move from personal, everyday ways of making meanings toward the socially shared and more written like discourses of specific disciplines" (p. 252).
It is crucial to point out that Gibbons is not arguing that students' everyday language is irrelevant or counterproductive to their learning. Rather, students' scientific understandings begin with their articulation of the phenomena using existing linguistic resources, which serve as a foundation on which students, with the teacher's guidance, learn to communicate with different audiences for different purposes. In this work, Gibbons explores a wide range of ways in which the teacher builds "linguistic bridges" between "learner language" and the target register, all of them beginning with students' own words (see also Gibbons, 2015).
Importantly, recognizing the value of students' existing language practices not only serves as a means for them to express their disciplinary understandings but also for teachers to present content to students. Brown and Ryoo (2008), for example, conducted an experiment with mostly Spanishspeaking 5th-grade students whose English language proficiency levels were not known. The students were provided an introduction to photosynthesis using "everyday" vocabulary (treatment group) or "scientific" vocabulary (control group). Students in the treatment group showed greater learning gains across all content measures on a postinstruction multiple-choice and short-answer test than the control group. Remarkably, the treatment group outperformed the control group on assessment questions that were asked in scientific language. Both groups, however, performed better when answering questions asked in everyday language than in scientific language. This research underscores the importance of recognizing the value of students' existing language and providing opportunities for students to bridge to greater facility in using scientific language.

TRANSLANGUAGING
So far, we have presented alternatives to two persistent deficit perspectives on the language practices of minoritized students, one that sees minoritized children as lacking words and another that views them as lacking the kind of language necessary for academic success. Our third section explores translanguaging, a sociocultural approach to understanding and leveraging multilingual communicative practices used by linguistically minoritized students. By valuing the range of cultural and linguistic resources students bring to the learning environment, designing instructional activities that intentionally build on those resources, and interrogating language hierarchies, we argue that translanguaging presents a potentially transformative approach to teaching in linguistically diverse classrooms. Unlike the word gap debate and many interpretations of academic language, translanguaging begins with what students are able to do-with the goal of both expanding their repertoires and actively challenging the language practices framed as necessary for academic success in educational contexts ranging from those serving young children (e.g. Gort and Sembiante, 2015;Bengochea and Gort, 2020) to higher education (e.g. Mazak and Carroll, 2016). Yet, as Poza (2017) pointed out, as principles of translanguaging are increasingly shaping pedagogy both in the United States and globally, transformation of power relationships through translanguaging is not inevitable. We suggest that an ecological orientation toward translanguaging in classrooms offers a nuanced approach to translanguaging for both Frontiers in Education | www.frontiersin.org July 2021 | Volume 6 | Article 641796 researchers and practitioners by illuminating how various layers of context-from the broadest sociopolitical dynamics to the more immediate relationships within an individual classroom-shape opportunities for language development. Translanguaging, or trawsieithu, is generally attributed to Cen Williams (1994Williams ( , 1996 in reference to a pedagogical practice that involved alternating English and Welsh for "input" (reading or listening) and "output" (speaking or writing) in bilingual classrooms as part of an effort to revitalize the minoritized Welsh language (Lewis et al., 2012). Colin Baker (2001) first provided an English translation of the term as translanguaging. In the United States, Ofelia García (2009) was largely responsible for popularizing the term, especially in contexts where students from immigrant and linguistically minoritized backgrounds use English and their home languages fluidly in the same classroom. For García (2009), translanguaging referred to the complex, dynamic, and discursive communicative practices in which bilinguals engage to make meaning. García and others, along with other language scholars, have since extended the concept of translanguaging as a pedagogical orientation within a range of multilingual instructional settings (García and Sylvan, 2011;García et al., 2012;Hornberger and Link, 2012;Sayer, 2013;García and Li, 2014;. This literature positions translanguaging as a theoretical approach that views language as a unified linguistic repertoire that includes features associated with various "named" languages rather than two separate language systems (Otheguy et al., 2015;Otheguy et al., 2018), and as a pedagogical approach in which educators actively challenge dominant notions of languages as separate from one another and support students in leveraging their full linguistic repertoires (García and Kleyn, 2016;. In practice, translanguaging strategies might encourage students to translate instructions or key concepts to their home or community languages, present ideas multilingually, or work collaboratively with peers leveraging all students' unique linguistic repertoires. For many years, linguists and educational scholars have worked to dismantle deficit views of bilingualism, underscoring that theories of language, second language development, and bilingualism fundamentally shape instructional arrangements and opportunities for learning (Valdés et al., 2014;Kibler and Valdés, 2016). Although there are parallels in the translanguaging movement to the efforts to contest dominant academic language ideologies, translanguaging is somewhat distinctive in its explicit focus on multilingual practices. Research on bilingual language practices in classrooms in the United States expanded, at least in part, in response to growing racist sentiments and deficit orientations that positioned immigrant-origin students as "semi-lingual," or incompetent in both their home languages and English, particularly when they "mixed languages" (Martin-Jones, 1995;Skilton-Sylvester, 2003;Crawford, 2004).
Historically, linguists have used the term code-switching to refer to the use of multiple languages in conversation. Early studies of code-switching challenged conceptualizations that framed such use of languages as a result of limited mastery of one or both languages; using the tools of linguistics, researchers have demonstrated the systematicity and communicative function of code-switching (Blom and Gumperz, 1972;Poplack, 1980). Subsequent research on code-switching in classrooms more explicitly addressed the need to revisit theories of bilingualism and pedagogical practices that reinforce inequitable educational opportunities for linguistically minoritized students and the need to recognize the wealth of linguistic resources that students bring to their classrooms (Arthur and Martin, 2006;Sayer, 2008;Martínez, 2010). This research documented the pedagogical power of codeswitching, demonstrating that when students have access to all of their linguistic resources, opportunities are widened for them to participate and engage in grade-level academic content.
Yet, in an era of increased globalization, scholars began to argue that the concept of code-switching failed to capture the complexity of multilingualism and the "superdiversity" (Vertovec, 2007) present within today's societies. As part of what has been described as the multilingual turn in applied linguistics (May, 2013;Ortega, 2013), numerous scholars have called for a theoretical perspective that individuals draw from a unitary linguistic repertoire that includes features associated with different "named" languages (Makoni and Pennycook, 2007;Makoni and Pennycook, 2012;Pennycook, 2010;García and Li, 2014;Otheguy et al., 2015;Otheguy et al., 2018). They argue that "named" languages such as English, Japanese, or Arabic are defined by social and political, rather than linguistic, boundaries; they underscore that seeing languages as bounded entities fails to capture the perspective of the language user (Otheguy et al., 2015); and they highlight the shortcomings of concepts such as "mother tongue" and "native speaker" (Makoni and Pennycook, 2007;Makoni and Pennycook, 2012).
This theoretical orientation toward language was accompanied by the call for an instructional approach that would actively draw on the entire linguistic repertoires of students to facilitate both expansion of students' repertoires and engagement in content-area learning (García, 2009;Creese and Blackledge, 2010;, and an explicitly critical view of the role of language ideologies in maintaining dominant power relations (García and Leiva, 2014;García and Li, 2014). It was out of these discussions that the concept of translanguaging gained traction (Poza, 2017). Notably, as scholars have pointed out, neither the practice nor the study of translanguaging is new (e.g., Canagarajah, 2011). For example, Leung and Valdés (2019) noted that for several decades, scholars have challenged the view that students' use of home languages in classrooms should be limited; instead, they have highlighted the value of translating texts and of teachers' use of students' home languages. Similarly, Poza (2017) points out that although the notion of translanguaging has become popular only recently, multilingual practices in classrooms have been studied for many years, particularly in parts of the world where multilingualism is widespread and celebrated.
Indeed, over the past decade, scholars have documented a range of positive outcomes associated with translanguaging in a variety of learning environments. For example, numerous studies have found that translanguaging in classrooms contributed to the expansion of students' communicative repertoires (Hornberger and Link, 2012;Martin-Beltrán, 2014;Velasco and García, 2014;Poza, 2018), including academic literacy practices in English and increased metalinguistic awareness (García et al., 2012;Sayer, 2013;Palmer et al., 2014). Martin-Beltrán (2014), for example, explored the role of translanguaging in a high school "Language Ambassadors" program that served newcomer speakers of Spanish enrolled in ESL classes, bilingual heritage Spanish speakers, and students who spoke English at home and were studying Spanish at school. The students engaged in reciprocal ways of teaching and learning English and Spanish by analyzing each other's writing, often asking "what do you want to say?" and inviting their peers to respond in Spanish, English, or a mixture of both languages. Martin-Beltrán (2014) concluded that translanguaging facilitated meaning making by students, contributed to greater metalinguistic awareness, and allowed students to bridge discourses between home or everyday language and new language and literacy practices in classrooms. This orientation stands in contrast to the notion that students lack the "academic language" necessary for participation in complex classroom tasks.
Other studies found that translanguaging expanded students' access to curricular content, engaged students' prior knowledge, and led to greater participation (Creese and Blackledge, 2010;Allard, 2017;Garza, 2018;Garza and Arreguín-Anderson, 2018;Poza, 2018;Duarte, 2019). For example, Flores and García (2014) described translanguaging by teachers at the Pan American International High School in Queens, New York, which served recently arrived immigrant students who were speakers of Spanish. They described the case of how Ms. C, a Chilean-American English teacher who was a bilingual speaker of Spanish and English, incorporated translanguaging pedagogies into instruction by creating "Hip-Hop Monday," which consisted of analysis of two songs, one in Spanish and one in English, that explored issues of social justice related to United States Latinos or Latin America. After listening to both songs, [t]hey analyze the song with a critical socio-political as well as language/literary lens, using language practices that incorporate Spanish features as well as English features in order to provide a deep critical lens. They then translate a section of the song into English in writing using their full linguistic repertoire . . . They do so using dictionaries, asking each other questions, and always collaboratively in heterogeneous groups that consist of students at different levels of English proficiency (p. 248).
As a pedagogical approach, translanguaging underscores the flexibility of multilingualism, considering the use of racialized, dynamic, and fluid language practices as valuable for their own sake and for the purpose of expanding students' semiotic repertoires and facilitating content-area learning. Some studies have highlighted that translanguaging pedagogies leverage students' familiar language practices to access new practices in the language of instruction and those valued within particular disciplines (e.g., Probyn, 2019). For example, Poza (2018) found that 5th-grade students' translanguaging practices facilitated the development of numerous skills central to those valued in science learning, including the use of discipline-specific vocabulary and visual supports to navigate complex texts, as well as skills surrounding the categorization of objects. Others have underscored that beyond bridging students' existing language practices with those valued in schools, in order to be transformative, translanguaging pedagogies must be paired with a translanguaging stance, a philosophical orientation which "informs everything from the way we view students and their dynamic bilingual performances and cultural practices to the way we plan instruction and assessment" (García et al., 2017, p. 50). That is, the goal is to change what is valued by schools.
The potential to contest and transform dominant power relations in classrooms is central to translanguaging pedagogies. In addition to supporting the development of new language practices and facilitating content learning, scholars have found that translanguaging presents opportunities for students to examine their bilingual identities, interrogate linguistic inequality, and resist colonial ideologies (de los Ríos and Seltzer, 2017;Flores and García, 2014;García and Leiva, 2014;García and Li, 2014;García-Mateus and Palmer, 2017;Sayer, 2013). Taking a critical perspective, classroom spaces can be sites of either contesting or reinforcing dominant power structures. García et al. (2012) contend that translanguaging pedagogies allow for actively contesting those structures by opening up opportunities for racially and linguistically minoritized students to use all of their linguistic resources and access identities that have been constrained by linguistic hierarchies rooted in colonialism.

Conceptual Critiques
As research on translanguaging has proliferated, some scholars have expressed concerns about the potential for (mis) interpretations of translanguaging that can undermine the goal of addressing educational inequity at the heart of translanguaging pedagogical orientations. For example, MacSwan (2017) has argued that by rejecting the notion of separate languages, the concept of translanguaging undermines decades of research that documented the bilingual expertise involved in code-switching and explicitly focused on deconstructing deficit myths about bilingualism. It is further contended that rejection of concepts such as code-switching, and more fundamentally, the absence of linguistically distinguishable systems called "languages," is ultimately counterproductive to efforts to recognize and value students' multilingual linguistic expertise.
Other critics have noted that the multitude of interpretations of translanguaging make it difficult to identify a consistent theoretical orientation, language practice, or pedagogical practice associated with the concept. Specifically, research sometimes misses the emphasis on contesting dominant language ideologies with the goal of transforming power relations that is central to the concept. In a comprehensive review of empirical studies, theoretical essays, literature reviews, textbooks, and practitioner guides on translanguaging, Poza (2017) found that although a majority of them emphasized social justice orientations, those primarily concerned with classroom practice tended to define translanguaging without Frontiers in Education | www.frontiersin.org July 2021 | Volume 6 | Article 641796 including contestation of linguistic and racial hierarchies. In other words, materials on translanguaging that are most likely to be accessible to K-12 teachers tended to ignore the connections among students' language practices, coloniality, and systemic inequality. A third concern relates to interpretations of translanguaging as expanding educational opportunities for all students, regardless of the broader sociopolitical context or instructional setting. While acknowledging the valuable contributions of translanguaging research, Leung and Valdés (2019) suggested that sweeping interpretations of translanguaging have the potential to further marginalize linguistically minoritized students if those interpretations limit students' opportunities to develop the dominant societal language. Allard (2017), who examined the role of translanguaging in ESL and sheltered science classes for adolescent recent immigrants, expressed similar concerns. While the teachers' translanguaging supported the students' comprehension and facilitated greater participation, translanguaging among teachers who were not experienced users of their students' home language (Spanish) ended up curtailing students' opportunities for learning. Similarly, in the context of a high school program for recent immigrant students, Lang (2019) found that the English Language Development teacher's narrow conceptualization of translanguaging as a means to avoid discomfort inadvertently undermined the goal of expanding students' bilingualism. Notably, students in both Allard's and Lang's studies expressed concern that teachers' translanguaging resulted in fewer opportunities for them to participate in English language practices and meet their critical long-term goal of learning English.
A number of scholars have pointed out that as a result of dominant language ideologies rooted in monolingualism, most content assessments of students developing bilingualism reflect only a fraction of what those students are capable of doing (Shohamy, 2011). Embedded within concerns about the potential for translanguaging pedagogies to reify marginalization is apprehension about alignment between translanguaging in classrooms and the assessment of linguistically marginalized students-particularly given the monolingual ideologies often entrenched in high stakes assignments and exams (Taylor and Snoddon, 2013). Canagarajah (2011) points out that even teachers who allow or encourage translanguaging as students negotiate meaning verbally often do not allow the practice in writing, "which they consider a more formal activity where students' performance is assessed" (p. 7). This calls for the need to explore the possibilities for translanguaging theories to inform assessment (e.g., Lopez et al., 2016;Schissel et al., 2018). García and Lin (2016) argue for a "strong" version of translanguaging theory that challenges the notion of languages as separate entities, yet they also acknowledge the material impacts associated with access to particular language practices. They underscore that more just and accurate assessments would take into account what students can communicate and achieve utilizing their entire linguistic repertoire rather than with a single language. Ultimately, they contend that a combination of multiple approaches is necessary: "On the one hand, educators must continue to allocate separate spaces for the named languages, although softening the boundaries between them. On the other hand, they must provide instructional space where translanguaging is nurtured and used critically and creatively without speakers having to select and suppress different linguistic features of their own repertoire" (García and Lin, 2016, p. 11). As Li and Lin (2019) pointed out, translanguaging pedagogies will not solve broader issues of racism, classism, and a legacy of colonialism; yet they present a powerful opportunity to challenge linguistic hierarchies by reframing students' and teachers' views of students' rich semiotic repertoires, and to facilitate meaningful engagement in the joint construction of meaning.

DISCUSSION
This review aims to highlight the threads that connect three areas of research that are not typically considered together. Focusing on early language development, the word gap narrative privileges certain ways of talking to young children and certain ways of learning that are commonly observed in Western, educated, industrialized, rich, and democratic (WEIRD) societies (Henrich et al., 2010). By focusing the research of language development on the number of words spoken to children in one-on-one speech without examining recurrent discursive practices, it obscures the rich linguistic and communicative assets being developed in early childhood situated in minoritized communities. Well-intended interventions, aiming at promoting equity but designed based on a deficit view, could be harmful by discounting cultural differences and undermining the development of culturally unique assets for language development. As children enter school, the notion of academic language, conceived as an intention to promote academic success, may further undermine their learning in classroom settings. Academic language refers to a category of language deemed necessary for school success. However, as we pointed out, defining such a category is conceptually problematic. In addition, placing too much focus on the form of students' language, instead of the content of their ideas and their ability to communicate those ideas, hides what students can do and may exclude them from deep learning opportunities. A focus on disciplinary practices, rather than language forms, has been proposed as an alternative. Following this rationale, we described translanguaging as a transformative approach with the potential to subvert language hierarchies and highlight students' linguistic strengths. In the final section below, we discuss issues that underlie the deficit approach to studying language development, suggest promising directions for research, and specify implications for educational practices.

Overarching Issues
The research reviewed in this paper converges to demonstrate how dominant language ideologies influence persistent deficit narratives of language development and language use by minoritized children and students. Moreover, underlying these ideologies are historical roots and structural issues that are crucial to consider in future research and field development. For example, all the well-intended interventions would be in vain without taking into account the historical and structural backdrop of colonialism. Coloniality presents globalization as a process of social stratification based on race and capitalism as they relate to labor. Race-which is "co-naturalized" with language (Rosa and Flores, 2017)-emerged as categories for conquered and conquering peoples to explain external, mental, and cultural differences (Quijano, 2000). For this reason, the dismantling of deficit views is inherently an anti-colonial project. If historical contexts and colonial relations of power are not explicitly and consistently considered, educators run the risk of reproducing the same social hierarchies produced in the past.
The critiques we raised in the three areas of research consistently point to issues of language ideology that have been with us for a long time. For example, the assumption that particular, prescribed forms of English are required for academic and intellectual endeavors positions alternative ways of using English and languages other than English as inadequate to the task. These assumptions are carried over from their historical roots into modern language ideologies. For example, English was promoted as a "language of salvation and progress" (Shahjahan, 2013;as cited in Hsu, 2015) during colonial times. Campaigns to promote the dominance of English have taken a variety of forms, including "Americanization" projects in the American Southwest and Puerto Rico, "Benevolent Assimilation" efforts in the Philippines, and efforts to force Native Americans to learn English under the philosophy of "Kill the Indian, save the man" (Hsu, 2015;Hsu, 2017). By considering the dominance of English and the influence of the colonial past, we can begin to see how they conspire to reproduce the same kinds of colonial relationships along racial and linguistic lines that foster the promotion of eliminating the word gap and teaching academic language for minoritized children. As Hsu (2015) notes, these histories are "explicitly included" in teacher education and TESOL programs because they promote the dominance of English for its utility in seeking certain economic and social goals (p. 141). Unfortunately, well-intentioned programs might promote the utility of English while simultaneously denigrating the language practices of students and their communities. According to Quijano (2000), an element of Western European modern rationality was that non-European people and things belong to the past. In this way, non-European expressions of language and culture are viewed as inferior and positioned as primitive.

Promising Approaches to Research
Our analysis calls for research attention to study language as a sociocultural practice, as opposed to solely an individual cognitive process, to leverage the linguistic strengths of minoritized children, youth, and students. There is a dire need to consider the role of cultural experiences in research that attempts to make sense of individual variations in language development, rather than assuming a universal pathway in place. Taking into account children's everyday lived experiences situated in their cultural communities (Solórzano and Yosso, 2002;Miller, 2014;Rogoff, 2014;Rogoff, Dahl, and Callanan, 2018) can provide important insights into how learning is coconstructed by young learners and their social partners.
The cultural construction of language development and language use has been examined in the decades-long research on language socialization through cultural practices (e.g., Ochs and Schieffelin, 1986;Miller and Goodnow, 1995;Weisner, 2002). This research has specified diverse building blocks that contribute to the development and use of language by children from nondominant backgrounds, including those from low-income families and historically underserved communities. However, this research is rarely leveraged by research that aims to map individual differences to productive ways of supporting language development. This disconnection, as we alluded to in the Introduction, may stem from the predominantly cognitivebased approaches to studying language development. The field will benefit from stronger integration of the social-pragmatic contextual approaches into the research on individual differences, to identify the assets for learning that underlie these differences. These theory-bridging endeavors will also point the field to culturally sensitive recommendations for everyday practices in supporting children's learning and use of language at home and in school.
To dismantle dominant power relations while attending to the complex network of relationships that shape language development at home and in classrooms, researchers will need to broaden the scope of investigation beyond the individual level to consider interpersonal and cultural aspects of language development. On translanguaging, for example, Allard (2017) argued that researchers must situate the classrooms they are studying within broader policies, practices, and ideologies such that the outcomes of a particular pedagogy can be understood in terms of its relationship with other aspects of the environment, including (but not limited to) lived experience of students and teachers, patterns of student interactions in and out of school, language policies, language ideologies, history of migration to the area, and services available to immigrant students. Similarly, Creese and Blackledge (2010) argued that the best ways to build on students' skillful multilingual practices is contingent on the unique sociopolitical and sociohistorical context in which those practices occur. A research orientation that considers engagement in language practices beyond the classroom-the practices that teachers can draw on to engage students in the classroom-will facilitate the design of activities that are most likely to empower students.

Promising Approaches to Curriculum and Pedagogy
The consequences of language manifestations on school outcomes reflect a complex problem that exists in educational systems. Research and practice partnerships have potential for addressing the systemic issues, rather than identifying norms of language practices and applying them across cultural contexts. One way of removing the consequences in school outcomes is to allow for variation in how students communicate and to value multiple styles of communication in classrooms, for example, by way of strengths-based policy making (e.g., proposals set in motion by the Ann Arbor decision of 1979; see Smitherman, 1981).
How can students be encouraged to engage in academic work using their existing linguistic and other semiotic resources, as well as cultural and community knowledge base, while being apprenticed into other academic discourses that may serve them well in the future? Unfortunately, many remedies proposed for students who "lack" academic language may lead to the prescription of instruction that detracts from the language development desired by advocates of academic language. Rather than "curricularizing" language as a subject to be learned piece by piece (Valdés, 2018), the focus should be shifted toward ways to apprentice students into the language, literacy, and conceptual practices at the heart of intellectually engaging work (Walqui and van Lier, 2010;Valdés et al., 2014;Walqui and Bunch, 2019), recognizing the strength of students' language for its own sake and using that language to bridge to other disciplinary discourses (Lee, 2007). In doing so, students' home, community, and cultural language practices-along with the other linguistic resources they develop as they expand their repertoire to communicate with a wide variety of audiences-are recognized as valued resources for engaging in academic work. Such approaches to curriculum and pedagogy would present a powerful opportunity to challenge linguistic hierarchies by reframing students' and teachers' views of students' rich semiotic repertoires, and to facilitate meaningful engagement in the joint construction of meaning.
In a similar vein, we join language-socialization theorists to recommend that language learners should be supported to engage in intellectually rich content-area instruction while building language proficiency (Faltis and Valdés, 2016). The present analysis of the word gap, academic language, and translanguaging research makes it clear that the ill-founded gatekeeping notion-that children and students must learn more vocabulary or grow English proficiency before engaging in rigorous content-area learning-positions multilingual learners and speakers of minoritized varieties of English as lacking the means to engage such learning, resulting in inequitable practices in education. Robust professional development for pre-service and inservice K-12 teachers is therefore one of the key directions to mitigate the problem. Most teacher education programs continue to rely on theories and practices that are detrimental to students' developing bilingualism. Moreover, many teachers hold deficit views of language learners and feel unprepared to teach them (Faltis and Valdés, 2016;National Academies of Sciences, Engineering, and Medicine, 2018). Professional development that addresses these beliefs and supports teachers to recognize and embrace the linguistic strengths students bring to the classroom can potentially remedy the persistent inequitable outcomes experienced by language learners. By raising teachers' awareness of their values and beliefs and encouraging critical examination of teaching practices, teachers can recognize the strengths students bring to the classroom (National Academies of Sciences, Engineering, and Medicine, 2018;e.g., Hudicourt-Barnes, 2003;Jones Brayboy and Maughan, 2009) and support students with culturally sustaining pedagogies such as translanguaging. The violent histories associated with the imposition of English are important for teachers to understand so that they can better recognize their own positionality related to these histories regarding how languages other than English have been devalued (Hsu, 2017). Opportunities should be created and supported for teachers to reflect upon these histories and critically examine their teaching practices (National Academies of Sciences, Engineering, and Medicine, 2018) and to understand the connections between linguistic hierarchies and pedagogical practices. As we have highlighted in this paper, language learners require access to robust learning opportunities, rich with semiotic resources and invitations to use meaningful language (Tharp and Dalton, 2007).

CONCLUDING REMARKS
Although approaches to language learning that highlight deficits may be designed to help improve life for some children, we argue that they may instead cause harm by placing responsibility on the individual, family, and community and by ignoring structural, historical, and cultural reasons for children's varying academic achievement. This is evident in the word gap research which emphasizes the importance of direct communication with children but ignores other ways of learning a language. In order to gain a holistic perspective of children's language learning, it is crucial that researchers investigate beyond dyadic practices and look at diverse social practices in children's homes and communities. In school, we have suggested the need to amplify the curriculum by providing multiple alternatives in classroom activities for students to engage in learning, and leveraging the linguistic practices students bring to the classroom instead of focusing only on developing "academic language." Finally, we have suggested that educators support the strengths of their multilingual students by taking an ecological approach for students to practice translanguaging. Instead of a "one size fits all" approach to working with students with varying language backgrounds, lessons should be jointly constructed with students to amplify their skills while taking the unique social, cultural, and political contexts into consideration.

AUTHOR CONTRIBUTIONS
S-hW led the team in conceptualizing, drafting, revising, and completing the paper. The Introduction section was written primarily by MC, S-hW, SH, and GB; the word gap section by S-hW, SB, MC, and SRM; the academic language section by GB; the translanguaging section by NL; and the discussion section by S-hW, SH, SB, and SRM. All authors contributed to the revisions of the full manuscript, with S-hW bringing it to final completion.

ACKNOWLEDGMENTS
This article germinated, took its shape, and was completed in collaborative activities of the interdisciplinary learning community, New Gen Learning, funded by the Spencer Foundation and by the University of California, Santa Cruz. The authors would also like to thank Jennifer Figueroa for invaluable comments and feedback on earlier drafts of this article.