Introduction
Learning English is considered a lifeline for marginalized immigrant communities in the United States and yet, mainstream approaches to ESOL (English for Speakers of Other Languages) are not always empowering or liberating. In response to larger societal patterns of xenophobia, ESOL practices have often focused on assimilating immigrants rather than disrupting hierarchies of power (Phillipson Reference Phillipson1992; Pavlenko Reference Pavlenko2002). The structural conditions of education funding in the United States, which increasingly focus on standardized testing and ‘employability’, encourage assimilationist approaches as well (Valdés Reference Valdés2015; Larrotta Reference Larrotta2017). These approaches, in turn, rely upon and re-inscribe hegemonic language ideologies which invalidate the practices of minoritized people and uphold those of the elite as an impossible and idealized standard (Cook Reference Cook, Cook and Wei2016).
Previous research in sociolinguistics and educational linguistics has critically examined the influence of standard language ideologies on ESOL education, rightly noting the damaging consequences of insisting on ‘Standard English’ as a target (Wiley & Lukes Reference Wiley and Lukes1996; Milroy & Milroy Reference Milroy and Milroy1998). Too often, however, discussions about standard language ideologies rely on what Rosa & Flores call a metaphysics of raciolinguistic presence; that is, ‘a sense that languages, varieties, and racial groups are empirical “things” in the first place’ (Reference Rosa and Flores2017:11). Even as ‘Standard English’ is considered problematic, it is treated as a stable and recognizable linguistic object. Such a perspective not only risks perpetuating the colonial project of linguistic boundary-drawing, but also ignores contemporary participation in the maintenance of linguistic inequality and linguicism and obscures the practices that continue to construct ‘Standard English’ today.
As Valdés argues, the construction of the linguistic object is an inherent component of all language teaching, a process she names the curricularization of language (Reference Valdés2015). To be teachable, particularly under conditions that require efficiency and replicability, language must be broken down into ‘a curricular subject or skill the elements of which can be ordered and sequenced, practiced and studied, learned and tested in artificial contexts within which learners of the target language outnumber proficient speakers’ (Reference Valdés2015:262). Similarly, language is delimited through the construction of borders ‘which make salient particular features and characteristics’ (Valdés Reference Valdés, García, Flores and Spotti2017:1). While these processes themselves are not necessarily problematic, they can perpetuate racist and classist conceptions of language, which cause material damage for learners (Cavanaugh & Shankar Reference Cavanaugh and Shankar2017). This is a particular concern in an educational context like community-based adult ESOL, where many learners are themselves members of minoritized populations, as working-class people of color.
Informed by this framework of language curricularization and borderization in education, this study looks at how standard language ideologies unfold in real-time in the context of a non-profit, community-based adult ESOL program in New York City. Drawing on ethnographic observations from eighteen months of fieldwork in ESOL classes, along with interviews with teachers and administrators, I find that the structural conditions of the program, particularly its funding requirements, perpetuate a conception of ‘English’ shaped by linguistic racism and classism, despite individual practitioners’ intentions and the progressive ideals of the program as a whole.
I bring together work on historical linguistic differentiation and enregisterment (Irvine & Gal Reference Irvine and Gal2000; Makoni & Pennycook Reference Makoni and Pennycook2006), raciolinguistic ideologies (Rosa & Flores Reference Rosa and Flores2017; Rosa Reference Rosa2019), and language curricularization and borderization (Valdés Reference Valdés2015, Reference Valdés, García, Flores and Spotti2017) to outline the mechanisms behind this construction of ‘English’ and how they unfold in real-time at the classroom level. By examining recurrent themes and discursive patterns in both teacher-student interactions and the curriculum, I show how linguistic authority is centralized in these classrooms through the presentation of a closed linguistic system, as well as a focus on replication of templatic language that explicitly excludes the linguistic practices of the students and their communities. This, in turn, allows for the drawing of linguistic borders through the pathologization of forms traditionally associated with racialized varieties of English, pointing to the continued relevance of race in the designation of a standard, which is reinforced through teachers’ metalinguistic and metapragmatic commentary. I end by looking at how students challenge these dominant ideas, revealing a disconnect between mainstream understandings of language and the way students actually use language, while also pointing to the possibility of more liberatory teaching practices and conceptions of language.
Conceptual framework
This article brings together various strands of scholarship on language ideologies, applying them to the context of community-based adult ESOL education. While these analyses have often been focused on larger discourses around language, I explore these ideas on the classroom level, looking at how ‘English’ is constructed in real time. Valdés's work on language curricularization and the construction of symbolic language borders (or, language borderization), for example, explores the construction of linguistic objects in education, primarily through policy and program design (Reference Valdés2015, Reference Valdés, García, Flores and Spotti2017). In this study, however, I turn my attention to more micro-level phenomena, investigating how language borders are established (and contested) in curriculum and classroom discourse, in addition to programmatic policies.
In this sense, my project also extends work on standard language ideologies, which are rooted in normativity and the belief that there is one correct form of a language. This resistance to variation, however, is not universal, since all speakers (including those that consider themselves to use the ‘standard’) have variation; the question is how and why certain variation comes to be seen as pathological. Milroy & Milroy (Reference Milroy and Milroy1998) argue that while standard language ideologies exist across industrialized nations, they manifest in context-specific ways. In the United States, standardization is particularly tied to race and ethnicity (mediated by class), with the language practices of Latino/a/e and African American communities often receiving special attention in debates about ‘American English’; the linguistic practices of indigenous Hawaiians and other Native American groups have also been scrutinized in this process (Matsuda Reference Matsuda1991). Consequently, ‘other’ Englishes are devalued, despite their communicative power, resulting in the further marginalization of these speakers, especially during times of increased immigration or heightened racial tension.
Often, these standard language ideologies are understood as a problem of unequal treatment among different (established) codes, but work in critical sociolinguistics and critical applied linguistics calls this into question. This research seeks to denaturalize named languages, reminding us that distinct languages are not natural objects and cannot be described by linguistic criteria alone (Makoni & Pennycook Reference Makoni and Pennycook2006; Agha Reference Agha2007). While certain linguistic features might cluster together, communicative systems are gradient and overlapping, and clear boundaries are only possible when we invoke social criteria. Some of critical sociolinguistic scholarship looks at the historical construction of languages as part of the colonial project, drawing on the histories behind the boundaries that were drawn by colonizers and missionaries alike, as well as the continued relevance of language in national identities (Makoni & Pennycook Reference Makoni and Pennycook2006; Canagarajah & Liyanage Reference Canagarajah, Liyanage, Martin-Jones, Blackledge and Creese2012; Khubchandani Reference Khubchandani2012; McElhinny & Heller Reference McElhinny and Heller2017). This scholarship affirms that the contemporary (Western) understanding of language arose as an instrument of colonialism, a way to divide and socially mark people.
Related work looks specifically at the ideological tools used in this linguistic differentiation (Irvine & Gal Reference Irvine and Gal2000; Agha Reference Agha2005). This work has identified some general mechanisms that seem to recur across contexts in the construction of discrete languages. It also explores how specific linguistic features become linked with particular people or groups of people, connections that are fundamental to the drawing of linguistic borders. In this project, I look for evidence of similar mechanisms occurring in real-time, bringing a feature-level focus to contemporary discussions of linguistic regimentation (Rosa Reference Rosa2019). With this perspective, I contribute to a better understanding of which features and practices constitute the (imagined) linguistic object itself and, consequently, which people and communities are framed as ideal within this educational context.
Other recent work has adopted a raciolinguistic lens to focus on the interconnectedness of race and language as ideological constructs rooted in the colonial project (Flores & Rosa Reference Flores and Rosa2015; Rosa & Flores Reference Rosa and Flores2017). Such work builds on the idea that notions of race intersect with and inform standard language ideologies, while also pointing to a connection between the racialization of minoritized communities (particularly in the United States) and the treatment of their linguistic practices.
There is good reason to pay particular attention to race in a study focused on a community-based adult ESOL program in the United States. Language classrooms are inherently characterized by a certain degree of linguistic normativity, and as previously mentioned, it has long been noted that conceptions of race shape standard language ideologies (Milroy & Milroy Reference Milroy and Milroy1998). Further, race continues to inform both official policy and public opinion towards immigrants in the United States; differential treatment patterns along racial divides, and such inequalities are inevitably replicated within education. With this in mind, a raciolinguistic perspective helps make sense of the language ideologies at play in this context and offers insight into the construction of ‘English’ in the sites I am studying and also in ESOL classrooms more broadly.
My methodology is informed by the long tradition of classroom-based ethnography, along with more recent work in classroom discourse analysis (Luttrell Reference Luttrell2010; Adger & Wright Reference Adger, Wright, Tannen, Hamilton and Schiffrin2015). Sustained observation provides the context-specific knowledge necessary to understand small but meaningful moments and their connection to recurrent processes or themes, and is particularly helpful for the analysis of language and its intersection with other axes of social differentiation (Rampton Reference Rampton2006; Reyes & Wortham Reference Reyes and Wortham2016). In this spirit, I implement an ethnographic approach strongly informed by work on linguistic differentiation in education (Reyes Reference Reyes2011; Rosa Reference Rosa2019), using data drawn from weekly classroom observations conducted over the course of an academic semester to inform my analysis.
Methods
This study is focused on the adult ESOL program run by a multi-service non-profit organization in New York City, which I refer to as NYC-Org. This program is funded through a combination of federal and state grants and private donations, with funding often changing year to year. The classes themselves were free and open to anyone in the surrounding neighborhoods; students did pay a nominal fee for materials, a policy geared towards encouraging sustained attendance rather than fundraising. For some class times, there was a waitlist (due to size restrictions imposed by the classrooms themselves) but otherwise, enrollment was open and rolling. All instructors worked part-time for the organization, teaching one or two classes a few days a week.
During my time at NYC-Org, I regularly observed the classes of two teachers, both self-identified white, native speakers of English who had extensive experience teaching adult ESOL. Their students were all first-generation immigrants but were otherwise a diverse group in terms of linguistic background, national origin, formal educational experience, literacy level, gender, immigration status, and length of time in the United States. These observations occurred weekly over the course of three semesters, though this article focuses on just the first semester of data.
I started volunteering in the adult education program at NYC-Org in January 2018, in the hopes of developing a dissertation research project. I wanted to explore how linguistic variation was addressed (or not) in these classes, something I had grappled with in my own teaching practice. Over time, I began to connect this question more directly with ideology and became interested in how various linguistic phenomena were talked about and reacted to by the different actors, and what that revealed about the conception of ‘English’ in these classes. My guiding questions became: how were the linguistic borders of ‘English’ constructed and contested, and what implications did that have for the students and their learning?
During each session, I took field notes (focused on the linguistic practices of the teachers and students, as well as moments when particular attention was given to linguistic form) and also audio-recorded each class and collected copies of the materials used. Throughout the eighteen-month data collection process, I maintained a reflexive practice of regularly reviewing my field notes to look for patterns and new observations that might require adjustment of my research questions and design. As needed, I referred to the audio recordings to clarify unclear notes or fill in relevant details. As a result, data collection and data analysis were closely intertwined. I followed an informal axial coding approach, initially coding my notes with simple content descriptions, and then sorting them into higher level categories over time, which allowed me to identify emergent patterns, discussed more fully below.
Findings
The teachers and administrators at NYC-Org were deeply committed to their students, and often framed their own work as in opposition to xenophobic and conservative beliefs. In this vein, several teachers explicitly told me that they were not in this work for the money, but because it aligned with their values. The adult education program was intentionally pro-immigrant, with many staff vocally denouncing the inhumanity of the current immigration system and supporting local activism around immigrant issues. The administrators also advocated for the inclusion of undocumented students, despite pushback from funders. The staff in general worked hard to create a pluralistic environment that was explicitly welcoming to those who might have experienced ethnic or religious persecution in the past, as well as LGBTQ students.
Despite these progressive values, the classes themselves adhered to a quite traditional and normative vision of language and language education, one that I recognized from my own experience in adult ESOL, as well as previous research on English teaching (Wiley & Lukes Reference Wiley and Lukes1996). Language was treated as a fixed object, with ‘English’ being something that the teachers had and the students needed. In addition, teachers talked often of ‘Standard English’ (or ‘good English’, or ‘correct English’), pointing to the influence of a standard language ideology which saw certain linguistic practices as more legitimate than others.
This framing, however, had less to with the staff's personally held beliefs about language, and more with the influence of requirements imposed by funders and other structural factors. In fact, interviews with the program director and one of the teachers revealed a surprisingly dynamic view of the students’ linguistic practices and needs. Both acknowledged the validity of students’ established linguistic practices, with the teacher framing ESOL class as a place to develop fluid communicative strategies rather than knowledge of specific ‘English’ norms.
They [students] will tell you that they use it at the grocery store, at the doctor's office… they will use it at school… but I just think that, in the day-in and day-out interactions, you can live—if you pick the right neighborhood where everyone speaks your language—you can live here for a very long time and not need to learn English. And I think that goes back generations. (Program Director)
You know, they've made their own connections. I see myself like, rather than helping them go to the doctor—I mean, sometimes it's just instilling confidence so they feel they can navigate whatever they have to do… it changes over the years. (Teacher)
But this expansive stance was quickly tempered when the issue of funding came up. Throughout our interview, the program director mentioned the impact of funding availability on the classes they were able to offer and the way they ran those classes. In fact, the program started by offering college-credit courses; ESOL classes were only added (eventually replacing the original college program) because of a shift in what sorts of classes the city was willing to fund. New initiatives were often started not to address an identified need, but because a new funding source (with new guidelines) became available; the program director described past changes to programming as “somebody had this goofy idea, and we did it because they funded us”.
Over time, funding has become increasingly tied to employment outcomes, with program success measured by employment rates among students and a new focus on ‘vocational English’. The documentation required by funders has also increased. Pre- and post-test scores, along with a portfolio of classwork, need to be submitted for each student each semester, with one teacher distilling this as “we need to show progress by virtue of papers in their folders”; some grants also mandate tracking of students for five years after they have left the program. This focus on documentable progress led to a reliance on standardized material; the teacher quoted above actually selected the textbook she used in class because of the worksheets that accompanied it, which she thought would work well for the student folders. As described by Valdés (Reference Valdés2015), this is a common mechanism of a curricularization of language shaped by standard language ideologies. As a result, the chosen textbook hugely informed the shape of class, as acknowledged by one of the teachers: “It's hard because it's somebody else's—when you are using a textbook, you're using somebody else's idea of how English should be communicated or taught”. In this vein, the structural conditions of this program, which are typical of community-based programs in the United States, had a decisive influence on the construction of ‘English’ in this context.
As I explored how this construction unfolded in the classroom and the mechanisms supporting it, certain patterns recurred in my notes, ultimately converging on three main themes: the assertion and maintenance of linguistic authority, the designation of target forms, and moments of destabilization. In the sections that follow, I discuss each theme in detail, drawing on specific moments from class as exemplars of how these processes played out. Together, they give a sense of how the construction of this linguistic object actually unfolded and what conditions made it possible.
Through my analysis, I do not mean to critique any individual teacher or particular program but instead examine the mechanisms behind language borderization in dominant approaches to adult ESOL and language teaching as a whole. I was interested in the program at NYC-Org precisely because of how typical it seemed; what I saw there was very similar to what I have seen in other programs, as well as in my own teaching. This program, like many others, is working within very real limitations imposed by structural linguistic inequality, scarce funding, and bureaucratic requirements; I offer this analysis as a starting point for imagining how to contest such limitations and their effect on conceptions of language.
Linguistic authority
While much of contemporary critical scholarship treats language as a social practice, negotiated by and dependent on multiple actors, the understanding of ‘English’ that I observed in these classrooms was much more traditional, standardized, and normative—a dynamic that has been well-documented in other educational contexts (Valdés Reference Valdés2015). In both classes, ‘English’ was treated as a fixed linguistic system, connected to an authority that was centrally located in the curricular texts and in the teachers’ expertise. The teachers (and the textbook writers) knew English and in the spirit of a banking model of education (Freire Reference Freire1970), their job was to pass that knowledge onto the students; the students’ own experience with or knowledge of English was rarely included or referenced in classroom activities, and sometimes explicitly excluded.
As a result, there were a limited number of actors invited into this process, which created the appearance of consensus as to what counted as ‘English’. This illusion is crucial to the construction of a distinct linguistic object. Linguistic practices are messy and overlapping; for them to cohere into an identifiable ‘language’ requires a singular perspective, a centralized linguistic authority which definitively draws borders and simplifies the complexity of actual usage. Locating this linguistic authority and identifying its methods are key to understanding and interrogating the relationship between designation of target languages and larger social processes.
As mentioned above, the textbook laid the foundation of how ‘English’ was conceived of in these classrooms, and in this sense, it was also particularly influential in the centralization of linguistic authority. While both classes made use of a diverse set of materials, their overall organization was set by a pre-selected textbook (and the corresponding workbook), with other curriculum or texts pulled in as supplements. Typically, both classes worked through the textbooks, from beginning to end, though some lessons or chapters were skipped. Supplementary material often corresponded to the stated grammar or vocabulary topics of focus in the current textbook chapter. For example, during a chapter that highlighted the use of auxiliary verbs in present-tense questions, the teacher pulled in from a non-textbook source a fill-in-the-blank worksheet in which students completed questions with the ‘correct’ auxiliary verb. Other supplementary materials corresponded to temporally relevant holidays or provided opportunities for longer-form reading.
The textbooks, and many of the supplementary materials, presented a top-down understanding of language and language learning. For example, the main textbook used in each class (Side by side and Future, respectively) maintained its own sort of self-contained linguistic universe; all the vocabulary and forms to be used were presented within the text itself. Each unit had a vocabulary list and a few target grammatical structures that were recycled over and over again in the activities. Within the text, there was no appeal to outside sources, including the students’ own existing knowledge, reference materials like dictionaries, or online resources (except those provided by the publishing company). Nor was there an acknowledgement of home language practices, except as a data point to be used in personal introductions (e.g. ‘Hi, my name is ___ and I speak ___’).
The presentation of ‘English’ in these texts was also form-focused and templatic. There was an explicit emphasis on repetition—of words, of conversations, of particular syntactic structures—with little attention to mastery of meaning or development of communicative strategies. Most activities were designed with the goal of recognizing and copying certain structures, with initial activities in a unit focused on explicit repetition of forms (listen and repeat) and later activities designed to elicit near-replicas of those forms (complete the sentences using the verbs in the parentheses). Freer-form production activities (write a paragraph) generally came at the end of the unit, after target forms had been practiced multiple times, with the implicit expectation that the students would use those forms.
The ‘English’ in these textbooks was also strikingly invariant. Students were presented with one way to say something, with virtually no alternatives, with a few exceptions like contractions (is not versus isn't) and ‘long’ versus ‘short’ answers (yes, I do like coffee versus yes, I do). Regional or social variation was not mentioned at all, even when it might be expected. For instance, the common example of [ɪŋ] versus [ɪn] for -ing, which the students themselves referenced, was never addressed within either textbook. The lack of authentic text or audio in the textbooks (and accompanying materials) helped maintain this consistency; the language that students saw and heard seemed to be designed to represent prescriptive norms (sometimes to the point of hyper-formality/correction) rather than believable examples of real-life practice.
Every so often, the teachers would introduce a variant, but generally only if it were a variant they themselves used regularly (and was not represented in the textbook materials). For example, one teacher explained that our could be pronounced as ‘are’ or ‘ow-wer’ (her spelling), and that she liked to use the “short one [‘are’]… if it's a long sentence”. Other times, the teachers would appeal to an outside authority, allowing variation if a ‘respectable’ source acknowledged it. During an activity focusing on syllable counting, for example, the teacher was unsure about the ‘correct’ answer to the question (regarding the number of syllables in different) and asked me to check in the dictionary. The dictionary ‘confirmed’ for the teacher that there is only one pronunciation of different, and that it has two syllables. The confusion arose because of the spelling of the word (particularly the first e), itself a reflection of another standard language ideology at play which privileges written forms over spoken.
When variation came up during the use of non-textbook materials (in song lyrics, for instance), it was often explicitly excluded from the target variety, with the teachers encouraging students to ignore the forms in question.
(1) (During an activity focused on ‘What have you done for me lately’ by Janet Jackson)
Teacher: Ain't is not English, ain't is street language for isn't… what the children say.
(2) (During an activity focused on a song containing the lyric ‘You drive too slow’)
Teacher: Should be slowly… it's a song, so they do what they want.
Students’ questions about variant forms also tended to be dismissed, with teachers warning them not to trust the language they heard outside of the classroom.
(3) (During an oral review of the answers to a fill-in-the-blank activity)
Teacher: There are people who were born here [the United States] and speak English but don't pronounce it very well.
More often than not, variation was treated not as a natural feature of language use, but a sign of a lack of proficiency. If students’ responses did not conform to the target template, the teacher often ‘corrected’ them, adjusting word order or inflection when repeating a student's contribution or when writing it on the board. The use of new vocabulary (not included in the target lists) was also censored, except in cases when the student had already demonstrated mastery of the lesson-specific targets.
Overall, the structure of the textbook and the resulting pedagogical structure of the classes discouraged students from sharing their own knowledge and experience, or from working creatively with the language they were learning. This reinforced the notion of an idealized standard. It also upheld the instructors (and eventually me) as figures of linguistic authority, which in turn, lifted up our white, ‘native-speaker’ Englishes as ideal, a process that was further developed through the designation of target forms, as described in the next section (Creese, Blackledge, & Takhi Reference Creese, Blackledge and Kaur Takhi2014). Language learning was rarely framed as a collaborative effort, and while the instructors did privately acknowledge to me that the students had taught them something new about ‘English’ usage, this was never mentioned during class.
Activities tended to be largely teacher-directed, with the teacher (or sometimes me) initiating an activity, giving instructions, and giving feedback on students’ responses. For example, the activity mentioned above in which the students counted the number of syllables in different started with a review of a story the students had read the previous class. The teacher asked me to read the story out loud while the students listened. Then, the teacher handed out the syllable-counting worksheet, which contained a list of vocabulary pulled from the story, with instructions to count the number of syllables in each word and to write the number next to the word. The teacher asked them to work independently and the class remained mostly silent as the students completed the worksheet. Shortly before the time when the class normally had a fifteen-minute break, the teacher stopped the students and initiated a whole-group review of the answers.
This exchange shows a typical routine in these classes, where the teacher organizes the structure of the activity and facilitates the conversation; I almost never observed group discussions that included the teacher and involved students responding directly to each other. The teacher also directed the feedback routines and generally followed the classic IRF/IRE (Initiate-Response-Follow-up or Initiation-Response-Evaluation) pattern, most often providing negative feedback on the students’ production by recasting what the student had said (Sinclair & Coulthard Reference Sinclair and Coulthard1975).
The earlier portion of the activity, in which I read the story aloud and the students listened, also supports the de-centering of students. Virtually every reading activity in both classes began with either the teacher or I (if the teacher asked me to) reading the text and the students listening; the students were generally asked to read only after they had heard a ‘native-speaker’ do so first. This pedagogical pattern is yet another way in which linguistic authority was centralized and made distinct from the students’ knowledge and experiences.
Along these lines, the exclusion of learner practices was also dependent on the erasure of ‘non-standard’ English practices, particularly those associated with immigrant communities. Though the students undoubtedly were part of larger cultural/ethnic communities that included individuals who considered themselves proficient, if not fluent and/or native users of English, the knowledge that students might have gained through experiences in those communities was ignored and sometimes explicitly censored. This included the denigration of what might be seen as hybrid varieties, which students embraced at various moments throughout the class. For example, during a fill-in-the-blank lyrics activity, students in one class positively labeled the song as ‘Spanglish’, with one student exclaiming “Oh good! It's Spanglish!”. Similarly, multiple students (proudly) identified their own linguistic practices as hybrid, while the instructor voiced dissatisfaction of this ‘mixing’.
Hybridity or mixing seemed to be a problem even at the psycholinguistic level, with one teacher reminding students that “we don't want to go English-Spanish-brain, we want to go English-brain”, a comment which naturalizes the distinction between named languages and depicts language ‘mixing’ as the result of some sort of confused or disorganized mental state. Though such practices appeared to come naturally to students, they were deemed undesirable and outside the realm of the ‘English’ to be learned in class, an idea that is further explored in the following section. Ultimately, this serves as one example of a larger pattern that excluded the linguistic knowledge and practices of students and their communities.
Designation of target forms
As mentioned above, the textbook and teachers were positioned as figures of linguistic authority and were therefore the primary actors in the drawing of linguistic boundaries, but how were such boundaries drawn? A significant mechanism was the designation of particular target forms and, accordingly, the censoring of other forms through negative feedback. As described below, the distribution of these forms was largely informed by raciolinguistic ideologies, based on both historical associations attached to certain forms, as well as commentary offered by the teachers.
In these classrooms, the majority of lessons explicitly focused on grammatical structure, treating a number of topics, including word order, question formation, verb tenses, the use of modals, and more, while others emphasized vocabulary or general comprehension or some cultural subject. Regardless of the focus of the lesson, however, certain forms or structural elements seemed more likely to be corrected or commented on, surfacing again and again as objects of attention for both the textbooks and the instructors: final sounds, inflectional morphology, and interdental fricatives.
With each of these categories, the teachers employed various feedback techniques, including explicit correction, exaggerated modelling, and metalinguistic commentary. By explicit correction, I mean any feedback routine in which the teacher explicitly marked students’ language as ‘incorrect’ (through the use of negative evaluative terms like ‘no’ or negative gestures such as the shaking of the head), followed by a recast of the word or phrase the student produced. Exaggerated modelling includes the exaggerated production (that is, louder, longer, and/or involving more extreme gestures than is normal in natural speech) of certain sounds to demonstrate ‘correct’ pronunciation. Metalinguistic commentary consists of explicit (evaluative) statements about language. These pedagogical strategies (and corresponding discursive structures) ultimately served as standardizing processes which worked together to establish linguistic borders and so construct the target language—‘English’.
The categories of forms mentioned above stood out to me because both the curricular materials and the teachers focused on them over and over again, regardless of the stated target of the activity. Every single class session contained moments of attention to these forms, and the teachers modeled interdental fricatives more than any other sounds. At first, I was not at all surprised to see these features throughout my notes, because I recognized these patterns from my own teaching, but I began to wonder why these features? What makes them so worthy of attention?
For the first two categories, there is considerable overlap. While some examples seem to focus on final sounds that do not involve an inflectional morpheme—“in English, we say the final consonant, Tomás (emphasizing s)” or “it's very important to say the final sound”—there were many instances which could fall into both categories. For example, a number of impromptu mini-lessons, inspired by a student-produced ‘mistake’ which the teacher hoped to address, highlighted verbal inflection, focusing on features like the s on third-person present simple verbs or do versus does in question formation. Similarly, the teachers often paused an activity to not only correct students’ pronunciation, emphasizing the final sound, but also comment on the importance of the sound, as shown in the examples below.
Other inflection-related targets of correction were related to a perceived lack of inflection, or ‘incorrect’ inflection (that is, conjugation which did not correspond to the paradigm presented in class).
In terms of specific phonemes, interdental fricatives (especially /θ/) were certainly the most corrected sound, with teachers often producing an exaggerated pronunciation with their tongue obviously between their teeth.
It is important to note that the focus on these forms did not seem to be about intelligibility; in fact, in virtually every instance of a teacher taking whole-group class time to address a ‘mistake’, the teacher seemed to fully understand the content of the student's message. Instead, these corrections seemed to be about normativity, or adhering to a standard. Further, these are not the phenomena that students tended to focus on when giving feedback to each other. When working in pairs or in groups, students generally only corrected each other when they were following a specific template and it sounded like an entire word was omitted.
This reminds us that while the feedback patterns mentioned above may appear typical of an ESOL class, they are not the only possible option. Instead, they are the consequence of a (widespread and deeply entrenched) standard language ideology which is perpetuated through certain pedagogical actions and discourse structures.
The textbooks followed similar patterns, particularly in terms of inflectional morphology. Lessons on the plural suffix or verb conjugation figured prominently in virtually every unit. In contrast, other grammatical topics—like word order, or the usage of function words, for instance—received significantly less attention, only highlighted a handful of times in the texts. This is particularly notable given that variation in inflectional morphology rarely causes intelligibility issues in everyday conversation—the meaning these forms convey is often expressed (more overtly) by content words, or is easily understood from the context. Regardless of their communicative value (or lack thereof), the textbooks framed these forms as central to ‘English’ and ‘English’ learning.
Adopting a raciolinguistic lens offers some insight into why these features received so much attention (Rosa & Flores Reference Rosa and Flores2017). Of course, these corrected forms correspond with features associated with ‘non-standard’ varieties. Final consonant deletion, invariant verb forms in present tense, and TH-stopping have all been noted in linguistic descriptions of ‘non-standard’ varieties, including AAVE/African American Language and Chicano/a/e and Latino/a/e English (Green Reference Green2002; Fought Reference Fought2003). To some degree, it makes sense that teachers would focus on these sorts of ‘non-standard’ features. Many of the students in these ESOL classes (and in others at community-based programs) are working-class people of color who likely live and work in communities where such features are commonly deployed. If teachers want to manage the resulting influence (from these ‘non-standard’ practices), then explicit attention to these forms might be a valid strategy. This also means, however, that they are actively correcting away from community practices, pathologizing the language associated with students’ peers and once again limiting opportunities for the incorporation of real-life language use. Language-learning is directed at an idealized standard which ignores variation and excludes forms associated with certain populations.
Historically, the formation of this standard has been tied to racialization, starting from the co-naturalization of race and language in the colonial project, and that continues to be true in these ESOL classes. Not only are the corrected forms traditionally associated with racialized varieties of English, but the metalinguistic/metapragmatic commentary made by teachers reinforces this association. Sometimes, the teachers acknowledged that certain forms were ‘English’, but not the right ‘English’, such as examples (3), (11), and (12) above.
Other times, the difference between supposed languages or varieties was naturalized, indirectly invoking a natural (and therefore racial) distinction: while it was particularly important to pronounce final sounds in ‘English’, one teacher described ‘Spanish’ as “too relaxed” because of it supposedly high rate of deletion. The use of certain coded terms also invoked racial (and class) difference and more specifically, tropes of linguistic anti-Blackness: “Ain't is not English, ain't is street language for isn't” or “Maybe the guys on the job say that… You don't want to speak street English; you want to speak the good English”. While a larger discussion of anti-Blackness in adult ESOL is beyond the scope of this article and will be reserved for future work, it is important to note how salient it is in these examples, both in the usage of street as a negative descriptor, and in the preoccupation with the word ain't, which is hardly a typical feature of ‘learner language’ but is strongly associated with African American and other Black communities in New York City. Here, the ‘standard’ is defined in explicit contrast to Blackness.
Additionally, other ‘white’ varieties of English were treated differently. Though, as previously mentioned, discussion of variation was limited, a few variants were introduced by the instructors and treated as valid alternatives: y'all (instead of you), coffee [kɔfi] (instead of [kɑfi]), and the family are (instead of the family is). These forms, however, were explicitly linked to regional (not racialized) varieties of English: ‘Southern English’, ‘Brooklyn English’, and ‘British English’, respectively. Ultimately, ‘Standard English’, the apparent target in these classrooms, was discursively reinforced as a racially based construct.
‘Standard English’ as a target is nothing new, and its connections to colonialism and race have been well-established (Milroy & Milroy Reference Milroy and Milroy1998; Lukes Reference Lukes2009; McElhinny & Heller Reference McElhinny and Heller2017). The colonial project sought to justify social hierarchies and exploitation through the invention of racial difference, with linguistic distinction invoked as evidence of this racial distinction, and vice versa. As a result, language and race are coarticulated categories which continue to inform each other in educational contexts (Rosa Reference Rosa2019). In the classrooms at NYC-Org (and likely many other adult ESOL programs), the connection between whiteness and standard language was re-inscribed through the designation of certain forms (and people) as ideal, while others (which map to historically racialized communities) were framed as pathological. The pedagogy and materials in these classes thus build on existing linguistic and ethnoracial enregisterment which has linked particular features to particular people and organizes them according to a colonial hierarchy (Silverstein Reference Silverstein2003; Agha Reference Agha2005, Reference Agha2007; Rosa Reference Rosa2019). In this way, the construction of ‘English’ in this context replicates larger systems of inequality, a phenomenon that is worth interrogating, especially if ESOL education is meant to empower adult immigrants.
Destabilization
Though linguistic authority was centralized in these classrooms and the target language was cast as immutable, there were nevertheless moments when these norms were challenged and destabilized. Both during class and in peripheral interactions, students found ways to continually embody flexibility and criticality in the face of a dominant standard language ideology. While such moments did not fully disrupt currents of linguistic normativity, they do point to the disconnect between traditional classroom language ideologies and real-world practices, and also remind us of the possibility of other ways of understanding and doing language.
Although teachers periodically deviated from standard language norms, in general, they fell short of genuinely contesting traditional conceptions of language. Such deviations often occurred when communication (or, the one-way transfer of information) was deemed vital; you can use Spanish was a familiar refrain during these moments. This was especially common when passing along non-class related information, including information about immigration services or other programs. Sometimes, the teachers themselves employed linguistic practices they seemed to classify as ‘non-English’. When greeting students one morning, for example, one teacher said “welcome, bienvenidos, marhaba”, though the plurilingualism potentially represented by this gesture diminished when she continued to say that “we're not really speaking, just trying to be friendly”. They would also use words from ‘other’ languages in explaining instructions, though these were largely limited to cognates (“this is the grammar, gramática”) or certain idiosyncratic expressions (“poquito pequeño”) and as in the greeting, seemed to be gestures to build rapport rather than sincere attempts at clarification.
As mentioned before, there were also rare instances when a teacher proactively acknowledged variation, but these variants were generally implied to be external to the target variety (i.e. not what the students were learning) and so such discussion ultimately served to reinforce language borders rather than promote internal diversity or variation. Socially or racially marked variants were not discussed, except to mark them as inappropriate (as in the comment made about the word ain't; see example (1) above). There were also a handful of moments in which teachers mentioned another word as an equivalent alternative to a target vocabulary term (“fall, or the fancy word is autumn”) but these seemed to be restricted to non-marked variants which the teachers themselves used. Otherwise, teachers tended to maintain a focus on ‘Standard English’.
The students, however, continually destabilized the norms explicitly established by the teachers and the course materials, both intentionally and incidentally. For example, though students tended to stick to templates when participating in whole-group activities (when the teacher was listening to their production), they almost immediately switched into more creative, more hybrid practices when unmonitored, adopting a translanguaging stance that defied named language boundaries (Otheguy, García, & Reid Reference Otheguy, García and Reid2015). This was especially true in one of the classes, where the teacher often had them work in pairs or in groups and almost never offered explicit feedback during this time. During such activities, students tended to be much more talkative and used a range of ‘out-of-class’ codes (including ones that could be labeled separate languages) to negotiate a task, returning to the templatic language when it came time to a record an answer (usually in writing). Sometimes, the teachers would try to discourage this by pairing students from different cultural backgrounds (e.g. Latino/a/e and North African), though much of the same linguistic dynamism persisted. Even when the teacher interpreted the students as speaking two mutually unintelligible languages to/at each other (“you're speaking to her in Spanish! She doesn't understand Spanish!”), this never appeared to be much of an issue for the students themselves.
In fact, students seemed to be not only comfortable with the linguistic diversity within their class, but actively interested in it, often asking each other about vocabulary and teaching each other phrases. There was one student in particular that I repeatedly observed asking others about their home language practices, but generally only when she was seated far from the teacher. On one occasion, when she got ‘caught’ deviating from the designated activity (and the corresponding linguistic forms) she defended herself by explaining that they were “speaking English but writing Arabic, it's good!”, though the teacher was not in agreement.
As mentioned before, the students also positively identified with language-mixing and linguistic hybridity, describing themselves as combining languages or using terms like ‘Spanglish’ (see Linguistic authority). In this way, students destabilized the ‘standard’ language norms which keep languages separate and instead viewed linguistic borders as permeable. This is especially notable given the teachers’ own negative comments about Spanglish and other hybrid varieties; the students were defying a particular linguistic hierarchy that was explicitly established within their classroom. The students’ comfort with ‘mixing’ (as well as their knowledge of their teachers’ disapproval) was underscored by the way they spoke to me when the teacher was out of earshot. Before or after class, or in whispered conversations during class, students often made use of a much wider array of linguistic practices with me than they did during whole-group activities, sometimes using target forms, but also adding in forms that the teacher had corrected in the past or words that I associated with ‘Spanish’ or ‘Arabic’.
Similarly, students were much more likely to engage in linguistic play when they felt they were not being monitored. This included impersonations of each other or the teacher; they were particularly fond of mimicking the teachers’ feedback styles. For example, in one class, when the teacher was out of the room, a student pretended to look over his classmate's work, raising the pitch of his voice and saying “very good, no, different please”; he stopped this routine as soon as the teacher re-entered the classroom. Other times, the students would play off each other and use similarities in sound to make jokes.
Such practices display dexterity with language and an appreciation for creativity and humor but were restricted to conversations that did not include the teacher. When the teacher was involved, the students overwhelmingly stuck to target forms and rarely made jokes like this, showing that while they seemed to value linguistic creativity, they assumed it was not part of the target language or of the sanctioned acquisition process.
At times, the students would also disrupt the focus on ‘standard’ language by bringing up certain sociolinguistic topics, though these discussions similarly tended to take place outside of official class time—before or after class, during the break, or after they had finished an activity. For example, during the break one day, one student came over to me and asked if I had heard about the lawyer (Aaron Schlossberg, in May 2018) who had recently gone on an anti-Spanish rant in a Manhattan deli. This turned into a larger group discussion about the importance of being able to speak without censorship and the connection between anti-immigrant sentiment and an insistence on ‘English-only’ policies. This was especially striking given that the teacher often reminded students to speak ‘English only’ in the class, but of course, the teacher was not in the room during the conversation. Such discussions not only challenged the standard language ideology of the classroom which upheld a ‘pure English’ and expected speakers to follow certain linguistic norms, but also indicated that students were aware of the sociopolitical implications of different beliefs about language.
Not all moments of destabilization were restricted to non-class-time, however. Periodically, students would question the language presented by the textbook or the teacher. One student, who was arguably the most politically active in the class, was particularly vocal about disagreeing with the text. For example, in one activity, in which the students were asked to assign a job title to a picture, she disputed the supposedly correct answer ‘doctor’, insisting that ‘nurse’ made more sense for the picture.
During another activity, the same student questioned the appropriateness of the word server, which the teacher had offered as a more ‘modern’ alternative to waiter, saying that “server for Hispanic person is not good” because of its similarity to serviente. The student not only asserted the legitimacy of her own knowledge, but also pointed to the interdependent relationship between codes that were often considered distinctly separate objects in the class. Here, an ‘English’ word was connected to and influenced by a word (and its demeaning social connotations) that was not considered part of the target language and that was unfamiliar to the teacher, blurring the linguistic borders that standard language ideologies impose and destabilizing larger structures of normativity. Despite the inherent inequality of the classroom setting (and of the immigrant experience), students did retain agency and found ways to limit the power of centralized linguistic authority in the class and make room for alternate notions of ‘English’ and language in general.
Discussion
So, how was the target language—‘English’—ideologically constructed in the adult ESOL program at NYC-Org? The centralization of linguistic authority was key, which minimized student input and created the coherent perspective necessary in standardization. This was achieved through the presentation of a closed linguistic universe which only included those forms introduced explicitly by the textbook and other curricular materials and by the teachers, as well as a focus on replication of templatic language. Teacher-centered pedagogy also reinforced this centralization.
These figures of linguistic authority, in turn, constructed a target language that was stable, unvarying, and delimited by clear borders. These borders were maintained by an outsized focus on certain categories of forms—final sounds, inflectional morphology, and interdental fricatives—which cast any ‘English’ which does not contain those forms as deviant. Unsurprisingly, the absence of these features is traditionally characteristic of a number of racialized varieties of ‘English’, pointing to the continued relevance of race in the designation of a standard. This relationship between racial categories and linguistic objects was discursively reinforced through teachers’ metalinguistic/metapragmatic commentary.
This notion of ‘English’, however fixed and monolithic, was nevertheless contested and destabilized throughout the class, especially by the students. Diverse linguistic practices were employed and embraced, including those that the students positively identified as non-‘English’ and/or hybrid. Creativity persisted in the face of normativity, with students periodically engaging in wordplay and humor, and in a handful of moments, some sociolinguistic discussion crept in. At times, students even asserted themselves as their own figures of linguistic authority, questioning or ignoring the norms established by the teachers and the curriculum. Together, these translanguaging moments, in which students defied ‘standard’ language norms, point to alternative conceptions and reveal a disconnect between mainstream understandings of language and the way students actually use language.
Of course, this disconnect is not exactly surprising. Language teaching in general has long been associated with the imposition of imperial power—and consequently, the pathologization of popular practice (Phillipson Reference Phillipson1992; McElhinny & Heller Reference McElhinny and Heller2017). English teaching in the United States, in particular, was wielded as a tool in the forced assimilation of immigrants (Pavlenko Reference Pavlenko2002). Contemporary ESOL education is undeniably linked to this tradition, and so today's policy makers and funders, and consequently practitioners too, are influenced by this legacy.
This relationship between ‘standard’ language and social legitimacy continues to pervade wider society as well. The Official English movement, which consists of various efforts to make English the official language of the United States and so restrict resources for ‘other’ languages, remains strong (Hill Reference Hill2008). Similarly, much of mainstream political discourse regards so-called ‘non-native’ or ‘non-standard’ language practices as suspicious—the discussion about former US President Barack Obama's speech styles (and name) is just one example (Alim & Smitherman Reference Alim and Smitherman2012). Immigrants are harassed in public for not conforming to ‘standard’ language practices (e.g. the language policing incident in the Manhattan deli, mentioned above) and linguistic discrimination persists in housing, the criminal justice system, and other arenas (Baugh Reference Baugh, Makoni, Smitherman, Ball and Spears2003; Rickford & King Reference Rickford and King2016).
These are the ideologies which shape our societal understanding of language; naturally, they make their way into the ESOL classroom. Not only are such discourses the most familiar and most available to teachers, they also dominate the world that teachers are trying to prepare their students for. The staff at NYC-Org were deeply aware of the xenophobia and discrimination their students faced and saw ESOL class as a place to develop tools to help navigate such challenges. If their students were able to learn ‘English’, they would be better able to advocate for themselves and if they learned ‘Standard English’, maybe they would be able to move up the linguistic hierarchy and access more ‘opportunities’.
But the ‘master's tools will never dismantle the master's house’ (Lorde Reference Lorde1984:112). In order to equip our students to negotiate structures of inequality, ESOL teachers (particularly white teachers, myself included) often (unwittingly) replicate the same structures in our own classrooms. The larger systems are only reinforced, and our students—the working class, racialized immigrants who such systems are designed to exploit—are still left fighting their way up from the bottom. Their ability to succeed within such a structure is limited by design; the continued co-articulation of race and language, as well as the persistence of linguistic hierarchy, means that their linguistic practices might always be interpreted as deviant, regardless of what they are actually producing (Flores & Rosa Reference Flores and Rosa2015). Learners are disempowered and linguistic (and social) inequality is re-inscribed (Rosa & Flores Reference Rosa and Flores2017; Baker-Bell Reference Baker-Bell2020).
And yet, as dominant as these ideologies may be, there is room to contest and challenge them, as students’ own practices within the classrooms at NYC-Org show. On some level, language learning is inherently normative, but it can also be creative and hybrid and critical and connected to popular practice. I offer these descriptions of the ideological processes at play in these classrooms to remind us of such potential, and to illustrate that dominant ways of constructing and understanding language are not natural laws, but are themselves patterns which can be disrupted. This is the value of a critical approach—not the critique itself, but the unpacking of something opaque and seemingly natural, which can allow us to imagine other possibilities.
A range of alternative pedagogical techniques have been developed in other contexts and can serve as inspiration for new approaches to community-based adult ESOL in the United States (Otheguy, García, & Reid Reference Otheguy, García and Reid2015; Cooke, Bryers, & Winstanley Reference Cooke, Bryers and Winstanley2018; Macedo Reference Macedo2019). At the same time, it is important to recognize the structural pressures and limitations which encouraged this program's focus on ‘Standard English’ in the first place, and which would constrain the efficacy of such approaches; we must also push for material conditions that could actually support such endeavors, as well as envision entirely new contexts for language learning. And these language borders are of course not limited to the ESOL classroom; they persist within public and institutional policy, as well as contemporary scholarship, and language scholars must also interrogate the borders they are upholding in their work and in their larger activism, or lack thereof (Pennycook Reference Pennycook2019; Tupas Reference Tupas2020).
In the same vein, the implementation of more radical pedagogies is not possible without a critical examination of current practices. Translanguaging, for instance, has been influential in K–12 education in the United States and holds much promise for adult ESOL; in fact, it has been a guiding force in my own development as an educator. But translanguaging is not a set of teaching tips as much as a transgressive linguistic and political orientation (Otheguy, García, & Reid Reference Otheguy, García and Reid2015). It calls us to make way for decolonial approaches by identifying and disrupting the colonial legacies we are still enacting, and to work with a conception of language that aligns with the perspective of racialized students and immigrant communities. The analysis I present here represents my effort to heed that call, to lay bare what has been constructed to make space for something new, in adult ESOL education and beyond, and to inform future work that imagines more concrete pedagogical transformations.
