Becoming a Teacher of Writing in an AI-Assisted World: Considering Personal Epistemologies, Writing Theories, and the Use of AI Platform Technologies

 

This article appears as part of a special issue series of CITE English Language Arts Education focused on critical perspectives on digital platforms in ELA teacher education (Volumes 24:4 to 25:3).

What does it mean to see oneself as a teacher of writing in this technological moment, particularly as a teacher of children who are learning to write? Teacher educators are struggling to reconcile, understand, and conceptualize how teacher educators support English language arts preservice teachers as they develop conceptions of teaching writing in a landscape saturated with digital platforms, increasingly embedded with artificial intelligence (AI) technologies.

We perceive ourselves, our preservice candidates, and our P-12 practicing colleagues to be engaged in a dilemmatic space (Fransson & Grannäs, 2013) of forced change created by an avalanche of AI-assisted technologies sold and encouraged for educational use. What is this space of challenge and dilemma? In this conceptual article, we articulate the often unstated, nested, and multidimensional features of teacher identity development with regard to writing content and pedagogical content knowledge, children’s writing, and the landscape of AI-assisted technologies.

The field needs more elaborated theorizing that highlights ethical issues around equity, humanization, and AI writing interactions, which must be addressed by teacher educators so they may thoughtfully prepare preservice English language arts (ELA) and literacy teachers to teach writing in culturally sustaining (Paris & Alim, 2017), affirming (Allen et al., 2013), and ambitious ways (Anagnostopoulos et al., 2020). The focus of this critique is confined to preservice teachers of ELA and the preparation needed to teach with “technoethics” (Krutka et al., 2019) and to think “technoskeptically” when engaging with generative AI technologies in schools (Krutka et al., 2020; Krutka et al., 2023; Pleasants et al., 2023).

Digital platforms, particularly those with AI-assistant chatbots and chatbot features, increasingly mediate teachers’ work in secondary ELA classrooms in which all students have individual devices. Before the global pandemic that began in 2020, the Education Week Research Center reported that 65% of middle and high school students learned in 1:1 classroom environments (i.e., 1 student:1 device; Bushweller, 2022); however, by 2021 that number had increased to 90%. Given this reality, teacher educators are mindful of preparing teachers to develop the skills necessary to teach within and through device-filled environments and, importantly, to develop the teacher identity dispositions necessary to engage critically in their use of device-mediated learning interactions, learning management systems (LMS) like Canvas, Moodle, and Google Classroom, and now, Generative AI-assisted technologies either embedded within the LMS, or adjacently, such as large language models (LLMs) like ChatGPT and Claude, and applications like Grammarly, Quillbot, and Adobe’s Firefly, which include AI GPT-technologies.

These technologies have designed interoperability (Kumar et al., 2019), which allows for user ease of movement between tools and platforms, streamlining data collection and aggregation, and efficiency. These features have accelerated teachers’ use in classrooms, such that use is now ubiquitous, normalized, and a centralizing force. In other words, many teachers and learners cannot go through a school day without direct interaction with screen-based discourse.

Both teachers and learners expect to mediate many of their interpersonal and intrapersonal interactions through screens. Arguably, especially at the high school level, screen-based, digital interactions have become the centering hub of learning by which all other learning and teaching interactions occur as spokes off of and connected to the digital center (Andrejevic & Selwyn, 2023; Garcia & Nichols, 2021; Kumar et al., 2019). Rather than argue the quality of this condition, this critique takes up this reality to consider specifically what it means to be a teacher of writing of secondary students when the task of writing is mediated through layers of devices, digital platforms, and now, artificial intelligences.

In 2008, Pamela Grossman called for teacher educators to engage in their own professional learning. Grossman’s plea called for soul searching that would support the defense of teacher educators as professionals ready to meet the challenges presented by forces advocating for accountability and privatization. Today’s moment is similarly precarious. Practicing and preservice teachers are the constant recipients of messaging to utilize learning management systems and AI assisted technologies to support data tracking, grading, learning goals and metrics, the development of instructional materials, and even feedback to young writers.

Most of these messages redefine teaching and learning as measurable performance outcomes that favor efficiency and consistency over interactions informed by issues of equity, accessibility, depth, and curiosity. This article calls to action teacher educators who work with preservice teacher candidates to support the development of technocritical, technoskeptical dispositions as teachers of writing such that we are prepared to meet the challenges of our time.

In what follows, we situate our theorizing using activity theory (Grossman et al., 1999) and a critical theory lens (Giroux, 2011; hooks, 2014) to consider how teachers’ identities and personal epistemologies inform learning and teaching interactions of writing and composition within dilemmatic spaces (Fransson & Grannäs, 2013; Honig, 1994), in which educators must struggle through challenge. These fields of research are relevant because they provide a foundation for theorizing the practices of preservice and early career writing teachers at the horizon line of AI-assisted technologies in the English language arts classroom.

Conceptual Framework

We posit activity theory (Grossman et al., 1999) as a useful conceptual framework for considering how preservice teachers engage in the becoming of teachers of writing when interfacing with AI platform technologies capable of producing written documents simulating human compositions. Activity theory informs our theorizing because it is “predicated on the assumption that a person’s frameworks for thinking are developed through problem-solving action” (p. 4). In this way, we consider the development of preservice teachers’ personal epistemologies around teaching and the teaching of writing as they influence and are influenced by learning interactions with students.

In their framing of activity theory, Grossman et al. (1999) referred to personal epistemologies as conceptual tools and the curriculum resources and instructional methods and modalities used by teachers as practical tools. In what follows we refer to teachers’ identities and epistemic beliefs about writing and writing instruction as the conceptual tools employed when teaching writing in platformitized spaces. This choice reflects our inclination to forward literacy research into epistemic beliefs and epistemic processing (e.g., Graham, 2018; Woodard & Cho, 2020), as well as the need to highlight the AI platform technologies as the practical tools to which we refer.

Activity theory remains an overarching influence, however, as it spotlights the interplay between the construction of classroom environments as sociocultural places in which humans act out their tacit and espoused beliefs mediated through the use of tools and artifacts. Activity theory focuses the researcher to examine the use of tools (Wertsch, 1991) as a way to consider the epistemological underpinnings educators carry with them.

Connected conceptually to an activity approach and situated in the growing body of research focused on examining the motivations and effects of platformized technologies on children’s learning, we use a critical lens (Giroux, 2011; Macgilchrist, 2021; hooks, 2014) to contend with multiple theoretical constructs related to the preparation of preservice teachers as teachers of writing. This interrogation finds its footing in the research of personal epistemologies (Hofer & Pintrich, 2017), writing teacher identity development (Graham, 2018; Kohnen, 2019), and writing instruction for P-12 learners (Graham, 2019; Smith, 2019).

With this foundation we theorize how monetized, for-profit, and private AI platform technologies, such as Google Docs, ChatGPT, Claude, and GPT-powered tools (e.g., Grammarly, Quillbot, and MagicSchool) mediate preservice teachers’ developing conceptions of themselves as teachers of writing. Our goal is to provide a technoethical (Krutka et al., 2019) thought heuristic to spur conversation in our field, as learners and teachers are increasingly incentivized to use AI-assisted platform technologies in today’s ELA classrooms.

Following a discussion aimed at articulating the boundaries of our interrogation, we theorize harms that arise from engagement in the dilemmatic space of AI and writing instruction. Taken together, the sections of this paper walk the reader through a proposed understanding of (a) writing teacher identities, (b) the ways identity positioning occurs in the dilemmatic spaces of teaching writing with AI assisted technologies, and (c) the potential harms present in those dilemmatic spaces.

As the potential harms suggest, there is cause for concern regarding how teachers’ identities respond, adapt, and reconfigure themselves as they normalize the use of AI-assisted technologies into their writing instruction. Our focus attends to the extent to which teachers feel empowered (or disempowered) to direct AI use strategically and in ways that correspond to their intrinsic identity constructs of ethics, justice, and the development of young, human writers.

In our conclusion, we contend that critical consideration of potential harms, especially as they relate to teachers’ abilities to reconcile uncomfortable dilemmas that they believe impact students’ present and future learning opportunities (or the preclusion of such), is vital in a time when the voices espousing technooptimistic and technodeterministic views seem louder, and importantly, have been monetized and purchased for use in ELA classrooms. Finally, we point out the implications this conceptualization has for conducting classroom research within teacher education on writing teacher identity, AI assisted technologies, and teacher development.

Recently, A.O. Scott (2023), writing in the New York Times Book Review, highlighted the role of thinking and ideation in writing for communication and publication. Scott invited readers to consider where ideas originate when writing and the extent to which it matters if AI-assisted composing is the genesis. Postulating that perhaps “intellectual labor is on the brink of a transformation as sweeping as the Industrial Revolution,” Scott described a series of “co-authored” human-AI chatbot essays, stories, and novels published in the recent months (p. 12). He archly compared these publications with authorial inspiration from other muses—lovers, ghosts, Ouija boards, and seances. In the end, Scott resisted a conclusive answer to his own question, leaving the reader with the queries, “Is this a matter of metaphysics, or of technique? Are we interested in the messengers — the chatbots and the Ouija-board revenants — or in the messages they deliver?” (p. 13).

As teachers of children and their writing, Scott’s casual attitude left us concerned. It is not a thought experiment for us to consider how the creative and critical modes of thinking built through writing could be short-circuited and rerouted by overreliance on writing with AI platforms like ChatGPT and Grammarly. Less muse than muscle, these platforms provide opportunities to engage with humans and produce (or coproduce) writing products throughout process writing, thus upending conceptions of ideation and imagination, organization and outlining, and a writer’s purposeful resistance to staid writing formats like ACE paragraphs and five-paragraph essays. If AI-assisted platforms can be a writer’s muse, are there developmental and ethical concerns teachers should consider when introducing this muse to young writers?

On Teacher Identities

Teacher identity research asserts the continuousness of teachers’ identity construction (Akkerman & Meijer, 2011; Mockler, 2011), such that “teachers’ identities are (re)constructed over the course of their careers” and teacher identity is “dynamic, flexible… hybrid and… influenced by a range of individual factors” (Rushton et al., 2023, p. 3). Synthesizing identity research, Brown (2022) argued for an identity work perspective that recognizes the similarities and connections amongst identity scholarship across fields (author’s use of italics, p. 1223). He identified a core set of assumptions that identities can most often be described as reflexive, evolving, fluid, and affected through both individual agency and organizational and social contexts, especially within relations of power.

Negotiation is also a defining concept of teacher identity (Holland & Lave, 2001) as teachers must negotiate their internal epistemologies within nested contexts of classrooms, schools, and local communities. Negotiation also occurs with external policy and accountability contexts (e.g., at state and national levels) that exert pressure about what constitutes successful performance and whose metrics and definitions may be in conflict with teachers’ or local constituents’ expectations and belief systems.

Within the burgeoning context of AI-assistance in the teaching and learning of writing, ELA teachers develop triadic identities of themselves as teachers of children, teachers of content, and teachers of pedagogical processes, each of which is nested in multiple contexts and shares the core assumptions of fluidity and change. This triad of identities includes a plethora of complex and overlapping dispositions and methods of teaching that teachers evolve and refine through classroom practice.

Developing an Identity as Teacher of Children

Research into the construction of teacher identity as it relates to being a teacher of children and adolescents brings forward particular identity aspects. Being a teacher of children and youth is emotional work often defined by shifting between conflicting expectations for emotional comportment (Hargreaves, 1998), such that teachers are expected to be caring and personable, while at the same time objective and neutral. Teachers, particularly exemplary African American teachers who employ culturally responsive pedagogies based in asset-based perspectives of their learners, have been identified as “warm demanders” (Ware, 2006), who balance caregiving, discipline, and high academic expectations for their students.

Lasky’s (2005) research identified high school teachers who shared “the belief that their core purpose as a teacher was to teach academic content while also attending to social and emotional elements of their students’ development” (p. 909). These teachers valued “human connections as key to their students’ increased involvement in their learning,” who described being “willingly vulnerable” as a means of relationship building that corresponded to students’ academic engagement (p. 908).

Similarly, Sapir and Mizrahi-Shtelman (2024) found that, amongst homeroom teachers in Israeli schools, meaningfulness was a core component of teachers’ professional identities. Teachers asserted that being identified as a “meaningful figure for students” and “doing meaningful education work with societal impact” were key aspects to their identities as successful and effective teachers of youth (p. 2).

Taken together, it is clear that teachers who teach young writers must configure identities that incorporate perspectives on caring and relationships to teach children effectively within a particular age group. With this in mind, we ask, “As learning increasingly occurs in net-based societies and cultures, to what extent will interpersonal relationships amongst teachers and students continue to be a pivotal element?”

Developing Identities as Teachers of Content Knowledge and Pedagogical Processes

A second identity that ELA teachers develop is as teachers of content. ELA content knowledge consists of foundational knowledges in various literatures, genre studies, and literary theory and criticism, as well as literacy development and language acquisition theories. As English language arts teachers build their literary content knowledge, they are also creating a schema of grammar and composition knowledges, which then contribute to their epistemologies of how reading and writing should be taught, separately and together. Grammar and composition content can be understood as anchoring knowledge focused on how English(es) work, their grammatical constructions, aberrations, histories, and evolutions, and how language is used in its many forms to communicate, identify and solve problems, imagine, and invent. Writing content knowledge also consists of conceptual beliefs regarding writing processes, approaches, audiences, and ways a writer develops and effectively employs a writing voice.

Pedagogical processes knowledge is the knowledge of how to teach (i.e., pedagogical content knowledge, Grossman, 1990; Shulman, 1986). This third corpus of knowledge informs the identities teachers develop as they make instructional decisions about how time is used, social interactions (e.g., individual work and group work), instructional scaffolds, direct instruction and teacher modeling, and student-led exploration and inquiry.

All of these “how” decisions about the processes of teaching and learning affect what is taught. Highly complex, pedagogical processes’ knowledge interacts with knowledges of content and human development to create actionable knowledge specific to classroom teaching (Grossman, 1990). With regard to writing instruction, pedagogical content knowledge is widely believed to be developed through teacher engagement in acts of writing (Romano, 2007), evidenced by the National Writing Project’s dominant practice of engaging teachers as writers. Thus, identities as teachers-of-writing overlap with identity constructs in which teachers understand themselves to also be writers.

Both of these identities reveal spaces of confliction for teachers of writing between preparation-to-teach and enactment as classroom teachers (Kohnen, 2019; Smagorinsky, 2010; Zuidema & Fredricksen, 2016). Whether these conflictions arise from juxtapositions between past experiences or the cultures of the schools into which teachers are acculturated varies, but it is clear that the espoused values of teachers of writing and their actions may not align (Elmore, 1995).

For example, Kohnen’s (2019) study identified preservice teachers who stated beliefs in the importance of creative and personal writing, while at the same time indicated that specific forms of writing were more “academic” and, as such, focused instruction on standardized test-related writing strategies. Participants also shared a belief in the importance of time for revision in process writing, while simultaneously expressing “the calendar and lack of time mean that students rarely can be asked to revise” (p. 360).

These conflicting views are particularly salient when considering the deluge of AI technologies focused on efficiency in writing production. For instance, if process-based writing that engages students to participate in peer editing is believed to be an effective instructional strategy, what happens when peer-to-peer relationships are replaced by chatbot editing applications that can more efficiently identify and correct student writing? Similarly, if AI-assisted platforms like ChatGPT replace classroom brainstorming discussions in the ideation stages of writing, what opportunities for the community of learners are lost and which opportunities may be gained? Indeed, Kohnen’s (2019) participants emphasized the importance of grading student writing in ways that reinforced notions that writing already exists primarily as a transaction in which “writing is exchanged for a grade” (p. 360). This research highlights the need for teacher educators to consider or reconsider writing theory such that we examine how AI technologies not only shape identity beliefs but may also accelerate and reinforce practices that ultimately harm the establishment of classroom communities in which learners share and support one another through written expression.

In summary, ELA teachers are multidimensional professionals whose identities evolve, especially around the identity aspects of adult-child relationships, content knowledge of both literature and writing, and pedagogical processes. These kaleidoscopic identities inform the design of learning opportunities that may support or hinder learner engagement, belonging, and academic achievement. We highlight these three aspects because the infusion of AI-assisted technologies into writing instruction may disequilibrate current identities. Over time, these shifts in practice will reshape the personal epistemologies of teachers that will then effect changes in practice.

Personal Epistemology and Theories About Writing and Composing

Personal epistemologies are the theories individuals construct and reconstruct about knowledge and knowing over a lifetime (Hofer & Pintrich, 1997). Personal epistemologies include beliefs about what kinds of information individual believe constitutes knowledge, in their lives more generally, as well as the professional and disciplinary knowledges that make up their fields. These personal beliefs also shape how individuals think about what they know, how knowledge can be used, and how they know what they know (Sandoval et al., 2016). These ways of knowing that individuals come to theorize in their day-to-day lives reveal their beliefs about how other people come to know things — the experiences, interactions, and learning scenarios that create or result in knowing — as well as how they know how to use their knowing.

Epistemic beliefs are an important starting place when considering the development of teachers’ identities as teachers of writing. Beginning with beliefs about knowledge and knowing is particularly salient because there exists so little understanding of the preparation preservice candidates receive to teach writing (Zuidema & Fredricksen, 2016). Yet, at the same time, all teachers have developed extensive epistemic beliefs about writing, as students are socialized through years of writing instruction in schools (Lortie, 1975; Street, 2003).

It is well articulated that many students do not receive adequate writing instruction (Graham, 2019), and teacher candidates’ “apprenticeship of observation” (Lortie, 1975) may “provide student teachers with a deep, though not necessarily accurate, sense of what it means to be a writing teacher” (Street, 2003, p. 35). Thus, failure to create accessible, relevant, and high-quality opportunities to compose and learn how to write as an expression of creative and critical thinking is due, at least in part, to teachers’ (lack of) preparation to teach writing and their epistemic beliefs about writing (Graham & Harris, 2018; Kohnen, 2019; Smagorinsky, 2010; Street, 2003). Preservice teacher preparation in writing instruction has also been limited by the nation’s myopic focus on the “science of reading” and the financial dismantling of the National Writing Project in 2011, creating a situation ripe for exploitation by edtech innovators and their solutionist technologies.

Four Dimensions of Writing Theory

As described previously, preservice and early career teachers’ identities as teachers of writing are influenced by the dispositions and practices of teachers they have observed as students and colleagues, via Lortie’s (1975) “apprenticeship of observation.” Smith’s (2019) work  summarized four interrelated dimensions of writing instruction: a products dimension, a cognitive processes dimension, a social and contextual practices dimension, and a pathways dimension. Teachers often favor one dimension over others, and trends in instruction “ebb and flow” regarding “when and how they [trends] have been taken up — at all — in pedagogical practice” (p. 67). These dimensions lend further understanding to the corpus of writing knowledge and lived knowing that preservice teachers bring into the dilemmatic space that asks them to consider if, how, and when they should integrate AI platforms into their teaching of writing. 

These dimensions of writing theory shape how writing is conceived and taught. The products dimension aligns most easily with the insertion of AI chatbots into writing instruction. If the creation of grammatically correct products in American Standard English is the high-value goal, AI powered editing platforms like Grammarly and Quillbot seem like a logical fit (Toncic, 2022). Yet, if writing is reduced to datasets for AI pattern-matching, students may internalize a view of writing as simply regurgitating expected linguistic sequences rather than an artistic craft of understanding contexts and making rhetorical choices. What is lost when these platforms edit students’ writing? At what developmental point should these platforms be introduced in the classroom? Should third graders run their essays through Grammarly? Should sixth graders? What are the social costs when peer editing and revision conferences are replaced by the singularity and efficiency of Quillbot’s nonhuman to human “explanations”? Questions like these start to draw the boundaries of the dilemmatic space in which writing teachers now exist.

Presentism: Contextual Challenges in Classroom Teaching

To describe ways in which teachers orient themselves within the workplace, Lortie (1975) argued that the teaching environment within schools could be characterized by three impeding “orientations” that overlap and reinforce one another: presentism, conservatism, and individualism. He defined conservatism as teachers’ mistrust in reform efforts and their need to continue to adhere and be loyal to established practices, even if they were not fundamentally supported by research. Individualism defined teachers’ preference to work alone and autonomously. Presentism is the predisposition to privilege short-term goals, which may hamper one’s abilities to plan for the long term (Hargreaves & Shirley, 2009).

As the controlling temporality of life in schools, presentism positions teachers to value efficiency and consistency, which further contextualizes the dilemmatic space teachers of writing exist within as they engage instructional encounters with AI technologies. Do teachers, especially secondary ELA teachers, feel compelled to use AI to help alleviate some of the stress of grading student writing and providing feedback? Presentism would indicate yes, they do. In addition, the breathtaking and profligate pace of marketed AI technologies spawns a predilection to short-term planning, because it seems another just-in-time extension, app, or platform is always ready to solve the immediate problem. This presentist view compounds resistance to a long-term plan for teaching multidraft, process writing with students.

Assessing multiple drafts feels daunting and impossible. Teacher educators need to assist their preservice candidates to consider the extent to which AI-assisted opportunities perpetuate a profession-wide proclivity toward  presentism. Arguably, when a writing teacher values efficiency and consistency over creative and critical exploration — the messy and time-consuming work of thinking through writing — those valuations of short-term strategy and immediate results are translated to students. Teachers ought to consider that they may become what we teach them to become.

In a Dilemmatic Space With AI-assisted Technologies

The Dilemmatic Space: AI Platforms Mediating Writing Teacher Work

Political theorist Bonnie Honig suggested that dilemmas be reimagined as more than discreet events, instead as “dilemmatic space or spaces which both constitute us and form the terrain of our existence” (1994, p. 568). These spaces of difference exist as “ever present in people’s living spaces” (Fransson, 2016, p. 187) and allow humans to engage the “political and moral projects of ordering subjects, institutions, and values” required as living in a complex, global world (p. 567).

Dilemmatic spaces in education can also be understood as spaces of social construction, within which people are positioned by themselves and others while also navigating the tools of their work and the challenges that arise through day-to-day interactions in classrooms and schools. Importantly, the concept of dilemmatic space provides analytic means for considering how changes in policy, politics, and society intrude into the lives of teachers and forces them to define (or redefine) the decisional spaces and actions of their work (Fransson & Grannäs, 2013).

For example, the neoliberal market values of accountability and efficiency forwarded by the No Child Left Behind Act policies created opportunities for privatization and high-stakes testing enterprises that changed the everyday lives of teachers and students and, for many, required them to engage in practices they believed were not in the best interest of children or themselves (Valli & Buese, 2007). When education settings are engaged in dramatic change from outside forces, the dilemmas move from solitary events to endemic in the context. Thus, the space itself is transformed, “positioning teachers into situations in which there is no right way of acting, but only acting for the best — where their professional judgement tells them one thing and the policy directives something else” (Fransson & Grannäs, 2013, p. 6). The digital tools of AI and platform technologies represent another seismic shift in education settings, particularly for teachers of writing.                                               

Dilemmatic spaces contain a spatial element, within which negotiations occur as a way of grappling with the complexities created when one is constructing, deconstructing, reconstructing the self and others, while at the same time having one’s identity constructed by others sharing the same space (Fransson & Grannäs, 2013; Honig, 1994). While Honig’s work is situated in political theory, we see utility in the concept of dilemmatic space as a useful heuristic for examining the turning point for ELA teachers.

When applied to the preparation of teachers of writing, Honig’s (1994) propositions push us to consider what is happening now that our epistemic beliefs about the knowing, knowledge, and nature of writing in classrooms characterized by presentism come up against predictive analytics. To understand and locate the terrain of dilemmatic space created by the infusion of AI-assisted technologies into writing classrooms is part of our quest in this article.

Since the public beta testing of ChatGPT in November 2022, teachers have participated, both willingly and unwillingly, in the normalization of pedagogical and learner interactions with AI-assisted chatbot technologies and writing instruction. However, prior to the public launch of ChatGPT, practical, commercial, and research communities recognized the profound influence of digitalization on learning and teaching interactions (Fransson, 2016). Fransson summarized this dilemmatic space as follows:

Teaching in the 21^st century is about maneuvering in an intertwined physical and digital context, where teachers are expected to implement digital technologies, deal with their own expectations, concerns and skills, work with multiple loyalties and conflicting tasks that arise and work out how digital technologies (could) reshape the processes of teaching and learning. (p. 196). 

More recently, scholars across disciplines have wrestled with the effects of platformitization (Garcia & Nichols, 2021; Nichols & Garcia, 2022; Pangrazio et al., 2022) and artificial intelligence in education (AIED; Selwyn, 2023). Nichols and Garcia conceptualized platforms as “dynamic environments that materialize a range of competing interests and imperatives in the educational settings where they are introduced” (p. 211). Their work invited educators to consider relational aspects of platform ecologies — the social, technical, and political-economic relations (p. 214) — and how these relations interact to enhance and constrain classroom learning. The call for additional research into platformitization underlines the need for cautious innovation and the support for teachers to develop critical attitudes toward and deeper understandings of these technologies.   

As Fransson (2016) described the current dilemmatic teaching space as intertwining the physical and digital context, this is especially true for teachers of writing for whom platforms like Google Docs frame digital spaces, thus shaping and constraining student writing, thought, and social interaction (Garcia & Nichols, 2021). Teachers also position themselves and their colleagues in terms of technology use, which may come with social currency. Some teachers may identify themselves as “early adopters,” who take up and try out new platforms and AI technologies quickly, playing with, integrating, and then discarding a variety of tools into their classrooms. Seeing oneself as an early adopter signals to others to consider the extent to which they are or are not early adopters.

Other teachers may define themselves as “late adopters.” These binary positions may then open up additional positions within the sociopolitical space of teaching, as teachers ask themselves and one another, when, how, and to what extent they, too, should be incorporating new technologies into their classrooms.

Thus, when educators position ourselves in social spaces, we signal to others that there are other positions available. Positioning distinctions within the dilemmatic space of writing instruction, platformitization, and AI are salient because the technologies are monetized, advertised, lobbied, and sold as “solutions” to the “problems” in education, and school and district leaders have heavily invested in their use.

The construction of the problems faced by teachers of writing by for-profit companies like Google, Open AI, Anthropic, and Microsoft leads to the products they sell and profit from as the material solutions. The immense profit and prestige available to companies motivates them to favor some positions over others, which might negatively affect the critical dispositions teachers need to employ while teaching and to develop in preservice preparation.

When school districts have spent millions of dollars to purchase hardware/software, digital infrastructure, and platform licenses, there is a strong endorsement for teachers to use the tools that have been given to them. Teachers may not feel they have the space, time, or opportunity to interrogate the ethical uses of tools, the motivations of those directing their use, and the effects of use on their teaching practices, or the social, emotional, and academic learning of children.

Taking up Nichols and Garcia’s (2022) construction of platform ecologies, teachers may not be engaged in collaborative conversations that move beyond the typical conception of tool use to consider the ecosystem technologies exist within, create, and effect. The concept of dilemmatic space provides a framework for educators to peer into the current context and reconsider alliances made, the tools used, and the positions taken as they relate to learning and teaching in digitally mediated environments. 

Such positionings within the dilemmatic space are made more complex by the constraints of the contexts described earlier. Presentism remains a powerful force in teaching contexts. The need to work more efficiently and with greater consistency is a felt need for many teachers of writing. It is not abstract to be faced with a Canvas page or Google folder full of student papers to read, assess, and evaluate.

It is important to begin analytic conversations with preservice teachers in which educators analyze the differences between human-generated (e.g., the teacher) and machine generated feedback. Some teachers have told us that the “feedback” provided by ChatGPT and Claude is “better” — more precise, more technically correct, and just more, than anything they typically provide. High school ELA teachers, of course, exist in schools wherein they may be teaching as many as 150 different students a day. This systemic constraint has held true for generations. Today’s diverse student body, however, brings new challenges, such as limited, sheltered support for multilingual students to develop English proficiency with confidence and skill, high stakes exams whose scores have potential for punitive use in grading districts, schools, and individual teachers, and cell phones that dominate attentional space resulting in students who may be off task and disengaged.

These conditions exacerbate teachers’ attention to the present moment and encourage get-arounds for improving their own efficiency and consistency when dealing with student writing. While grading via AI saves time, another appeal is that it allows the user to delegate hard decisions and final judgments to a machine (Andrejevic & Selwyn, 2023). Ideally, as teachers score work, they consider who the student is, what their background is, where they are in their progress as a writer. Ultimately, they read a paper and see the human writer attached to it. And while rubrics were designed to help with these hard decisions of attaching a score to something that is innately human — a student’s writing — the decision still lies with the teacher. Offloading the judgement to a machine may ease the burden of some moral and ethical stressors, while creating new dilemmas.

Let us be clear. Probability engines cannot give thought-filled, human feedback to writers. Anthropomorphizing platforms like ChatGPT and Claude as having “intelligence” was the first slip felt by teachers of writing on the slippery slope of AI-assistants as the solution to teachers’ need to respond to student writing. Lauren Goodlad of CriticalAI grapples with the misnomers “artificial intelligence” and “stochastic parrot” by referring to these technologies as “probablistic mimics” (Pearson & Burns, 2023, 4:24; see also Bender et al., 2021).

Goodlad’s moniker makes brighter the fact that chatbot platforms are neither intelligent nor thinking. Built on and with access to almost unfathomable and increasing amounts of data, these large language models (LLMs) mimic human thought, discernment, and dialogic interactions. This mimicry occurs through sophisticated predictive analytics, the microinteraction of which can be understood as the best guess of the next word in a sequence of words. They also are not like parrots, because as Goodlad said, “Parrots are actually really smart animals” (6:06).

Harms in the Dilemmatic Space: Writing and Composing as Human Thinking

The next section focuses on harms arising in the dilemmatic space of assessing and grading student writing with AI chatbots. We articulate the salient harm to both teacher/users and student writers as a loss of opportunity to develop the cognitive architecture of writing as thinking, which connects to losses of agency and community. In addition, we highlight current scholarship grappling with the ways AI and platform technologies interact with and affect preservice teachers’ developing understandings of writing instruction, with an eye toward the development of preservice teachers as critical consumers, especially as it relates to shepherding student work through assessment and evaluative processes.

Seeking a balance between innovation and caution, educators should balance enthusiasm for the inclusion of AI tools with a critical eye for finding the boundaries of their utility and potential harms. This requires the preparation of teachers who are equipped to use AI ethically and responsibly, not seeing it as a neutral tool, but instead adopting the ecological perspective toward educational technologies (Garcia & Nichols, 2021). This preparation and support should include understandings of the political, social, and ethical implications of AI (Bender et al., 2021; Selwyn, 2022). Within the context of education, the social dimension of platformization is often the focus. However, writing teachers would benefit by a greater understanding of the relationship between the three dimensions forwarded by Nichols and Garcia (2022; see also van Dijck, 2013). Once a platform is viewed via a relational lens, multiple “frictions” arise (Nichols & Garcia, 2022).

In the social dimension, educators focus on the uses of the products and the outcomes. However, when the technical and political-economic dimensions are included, writing teachers can contend with the ways Google products, for example, lure school systems in with inexpensive products but fail to clearly disclose the company’s collection of user data, which in turn, furthers their corporate economic dominance. In other words, as students utilize Google Docs as a social writing platform, they are also freely sharing their writing — their original and creative thinking — with a for-profit company that benefits from the monetization of their ideas. Similarly, AI platforms and assisted applications offer not only inexpensive products (often a limited, free version), but they also promise to alleviate the stressors of presentism with efficiency in scoring student work, in return for data collection as surveillance capitalism (Zuboff, 2019) and the continuance of data interpolation that forwards and reinforces stereotypes, bias, and hegemonic viewpoints (Bender et al., 2021; Buolamwini, 2023; Selwyn, 2022; Tassoni, 2024). Without a relational lens, preservice teachers risk remaining neutral because they are unable to see or understand the friction in and between systems (Nichols & Garcia, 2022).           

Many secondary ELA teachers already struggle with conceptualizing writing instruction beyond linear process writing and a focus on product over process. As well as considering that most educators are planted within the social dimension, only seeing the products for their uses and locked into presentism (seeking efficient means of scoring work), a dilemmatic space exists. The content of the outdated “writing process” often swings between structured expository writing (e.g., five paragraph essays and claims, evidence, and commentary paragraphs) and personal nonfiction essays. The current adherence to linearity suggests many teachers, especially preservice teachers and early career teachers, will engage with AI tools in ways that reify dominant practices, highlighting how preservice teachers engage in uncritical integration of AI technologies and increased reliance on the built tools within digital platforms, hindering the quality of students’ opportunities to learn to write as a human meaning-making endeavor in ways that are both dialogic and recursive.

Possibly leading pedagogy to over-emphasize formulaic writing skills over experimentation, creativity, and students development of authentic voices as writers, the operationalism and preemption in AI systems helps highlight a need to implement informed writing pedagogy to help counteract these tendencies (Andrejevic & Selwyn, 2023). Yet, Jamieson (2022) proposed that AI — like the introduction of computers and the internet — by itself is not the problem. Instead, the problem resides in ignoring the rhetoric of the “AI crisis,” as well as the implications its use raises in the discipline of writing (p. 153). Within writing instruction pedagogy, preservice teachers must learn to engage with AI technologies critically instead of blindly accepting their use as an efficient and effective tool for grading and assessing student writing.

Pangrazio (2022) suggested that educators need to consider two mechanisms as they engage with AI and platforms: first, to shed light upon critical understandings of the technology, then to move beyond a critique to reimagining. One way of doing this is through constructive engagement (Selwyn, 2023) situating AI as opportunity (Graham, 2023; Jamieson, 2022; Morrison, 2023; Tassoni, 2024). The next section briefly explores recent writings that assert a way forward with AI-assisted writing in composition studies. While this paper focuses on P-12 teachers and classrooms, learning from colleagues in university composition studies can provide an important counterpoint.

At the collegiate level, within composition studies, scholars have advocated for the introduction and incorporation of artificial intelligence technologies as opportunities for reimagining a less linear writing process (Graham, 2023; Jamieson, 2022) and a “post-product” space (Tassoni, 2024). Challenging the common concerns that AI-assisted writing will damage student learning and writing skills by shortcutting the writing process, Graham proposed an alternative framework that sees AI as an opportunity to add new dimensions of recursion and dialogue to the writing process, rather than as a replacement for it. Relegating writing judgments to AI could deprive students of the social processes involved when teachers give evaluative feedback, shaping students’ understanding of writing as a dialogue between human interlocutors (Andrejevic & Selwyn, 2023). In this postprocess view, using AI-LLMs like ChatGPT becomes a new form of prompt engineering and output curation that can augment, rather than replace, the iterative, meaning-making aspects of writing.

In this same vein, Jamieson (2022) called for a reframing of the AI crisis as an opportunity for the field of composition studies to reengage with pedagogy. This reengagement would focus on teaching students explicitly how to use AI tools effectively and incorporating them into the writing process. Morrison (2023) also argued that AI should be seen as a new tool for writing and that the field of composition studies has the pedagogical expertise to meet the challenge. While Morrison’s article is more conceptual, focusing on the philosophical and affective dimensions of the relationship between AI and writing, both Jamieson and Morrison argued for a pedagogical focus and the value of human writing. Both authors emphasized the importance of teaching students how to use and engage with AI technology effectively in their writing, rather than merely trying to police against its misuse. Both articles also express a deep appreciation for the unique qualities of human writing, such as its capacity for meaning-making, creativity, and personal voice. They argued that these human elements should be the focus of writing instruction, not just technical correctness.

Tassoni (2024) explored how the emergence of generative AI can affirm dominant discourses, values, and practices, potentially marginalizing the diverse literacies of underserved students in first-year composition courses. In response to this harm, Tassoni proposed a “postproduct” approach to writing instruction that shifts the focus from finished products to the negotiations and reflections students undertake when collaborating with AI. This approach aims to make the hidden curriculum of higher education more visible and open for critical exploration by students, helping students resist uncritically adopting AI-generated text and developing a more nuanced understanding of how AI can both challenge and reinforce traditional academic writing norms.

Tassoni (2024) asserted that this postproduct pedagogy, combined with labor-based grading, can provide underserved students more opportunities to explore their own voices and linguistic practices, rather than feeling pressure to conform to dominant standards. Teacher educators can support preservice candidates and in-service colleagues in the development of lessons, rubrics, and discussions aimed to identity when, where, and how AI should intersect with particular kinds of writing and the extent to which human writers must acknowledge the influence and use of AI assistance.

Interrogating ways in which AI may be working to marginalize the diverse literacies of underserved students, postproduct pedagogy (Tassoni, 2024) situates itself within the negotiations and reflections that students undertake when they are collaborating with AI. Each of these critical ways of viewing the implementation of AI use in writing classrooms provides a reimagining to help preservice teachers become critical examiners of the implications of technology to develop discerning eyes and ethical stances that may inform how LLMs and digital platform technologies subtly guide, change, and even dictate instruction methods and learner outcomes (Bender, 2021; Selwyn, 2022).  

In summary, the conversations at the intersections of AI technologies and platformization in education indicate a need for teacher educators to more directly and intentionally acknowledge and engage with the dilemmatic space of teaching in this frontier space. Teacher educators need to provide opportunities for preservice teachers to view and interact with AI technologies and platforms through a relational lens, so they can see, interrogate, and seek to understand the friction created within the various components. This action is needed within the field, in general, and, most especially, within the space of supporting teachers of writing, where writing teachers’ orientations confront a unique set of internal and external pressures.

A Closer Look at AI Assessment and Evaluation of Student Writing

A first point of contention illuminated when noticing how teachers rely on AI LLMs (e.g., Claude and ChatGPT) to evaluate and assess student writing is the reduction of writing and composing as something other than acts of thinking. Regardless of the dimension of writing theory forwarded by teachers, or which may be privileged by the curriculum materials teachers use, acts of writing and composing represent thinking. These forms of thinking may be critical or creative and likely are both. At times, acts of thinking may demonstrate development such that learners develop more sophisticated ways of thinking about, with, and within the world.

Thinking as and through writing also represents intention, emotion, memory, and embodiment. Writing, whether by hand or typing, literally engages the body in the creation of acts of the mind. As people write, their shoulders may tense with the felt emotion of the words the use to tell the story. Their faces may grimace or smile. And even the youngest writers write for an audience, proudly sharing their writing with others and telling what it says and who or what it represents. Taken together then, writing and composing are acts of thinking and, as such, reflect, embody, and bring into the social world epistemic beliefs.

First, building upon the premise that writing is thinking, then the student’s drafted essay may represent their best, most current thinking on the topic under study. To disregard the human in that creation of thinking, the essay, by asking a nonhuman to read, respond, and grade it raises ethical questions about role definition, intentions, the purposes of the writing assignment to begin with, and how one values (or not) the student-teacher dialectic in learning interactions. As such, the first dilemma is fundamentally human. What is lost when the audience for my thinking is nonhuman, or worse if I believe I am writing for my teacher and peers, when my thinking will be read by neither?

This situation also highlights assumptions held by the teacher and, over time, held by the student, that AI-driven writing assessment is universally applicable (Heuer  et al., 2021; Macgilchrist, 2021) and that the feedback provided by the LLMs will be culturally congruent to and understandable by the student. This “eradication of the subject”  (Andrejevic & Selwyn, 2023) through AI claiming to know the student writer better than they know themselves could undermine the role of writing in self-expression, identity formation, and subjectification.

Second, to the possible deceptions and costs of having a nonhuman reader, in both scenarios the teacher is losing an opportunity to hone their instruction by studying their own practice through the lens of student performance. We define reviewing student work as more than a task of assessing and grading student learning; it is also an opportunity to reflect on the quality and effectiveness of our instruction. Did I coherently and clearly convey the assignment? Were the materials universally designed, accessible, interesting, and relevant for the students in front of me? Reading student writing can and should be as much about developing teaching practice as it can be about honing my abilities to respond to writing with consistency and greater efficiency. This disposition toward reflective practice is particularly important for preservice teachers to develop. Being a part of professional learning communities in which talking about how effective teachers make meaning of student writing, assess it, grade it, and learn from it establishes an inclination to continue those behaviors into classroom teaching.

Third, research into the ways writing is taught in contemporary schools (Graham, 2019) indicates that some teachers do provide instruction using evidence-based methods. In Graham’s review, exemplary instruction included ample time allotted for writing instruction and student writing (Cutler & Graham, 2008; Dockrell, et al., 2016; Hsiang & Graham, 2016) using a variety of instructional strategies to engage students in process writing and writing-to-learn activities (Gillespie et al., 2014; Ray et al., 2016). Mention of these examples matter for two reasons.

First, high-quality, evidence-based writing instruction and opportunities to learn to write do more than simply teach writing as correct use of grammatical constructions or adherence to formula. High-quality, generative process writing opportunities, over time, build students’ cognitive schema of writing and composing, its uses, connections, and utility, as well as students’ epistemic beliefs about writing. In other words, over years, students’ capacities as writers, their identity beliefs about themselves as writers (or not), and their epistemic beliefs of writing cohere together primarily through their experiences writing in school. Writing also serves reciprocal relationships with reading comprehension, disciplinary content knowledge learning, and multilingualism (Graham, 2019).

Thus, we can extrapolate that as learners engage in writing as composing activities, such as those focused on planning and revising their ideas as expressed through writing, youth writers are building a schema of writing as recursive, intentional, and communicative, as well as attentive to the constraints posed by audience, genre, purpose, grammars, and organization (Graham, 2018). Likewise, if learners spend the majority of their writing interactions engaged in closed-ended activities or those constricted by artificial confines, such as the “ACES” writing strategy (i.e., the  Answer, Cite, Explain, and Summarize paragraph structure) and five paragraph essay (e.g., focused around the three-prong thesis statement, with the corresponding body paragraphs detailing the three aspects of the thesis), their cognitive architecture of writing (i.e., learners’ schema) and their epistemic beliefs about writing may focus on writing as lower level thinking which is primarily product oriented. In epistemic beliefs about composing as a tool for knowledge generation; writing’s utility for thinking and problem-solving; how complex products are created; the social, recursive, and multimodal acts writing encompasses; and how it serves to foment additional learning—many of these aspects are stunted at worst and ill-defined at best.

If writing is framed as a tool for building a flexible and networked human cognitive architecture for ideation, curiosity, evaluation, and communication, then teachers’ reading of and response to student writing becomes a space of peering into the learner’s developing conceptions of thinking, from both content and process perspectives. The loss of this opportunity might be highly detrimental. In other words, assessing and grading student writing is about much more than the acts in real time and their direct correlates (i.e., a grade in the grade book). This instructional interaction represents the development of linguistic and communicative thinking structures, influences socioemotional interpolation of interactions with and productions of text, and provides teachers insights into the effects of their instructional choices. Arguably, many more facets of purpose could be articulated here. Offloading reading, assessing, and grading to a nonhuman technology run on predictive analytics calls into question these purposes.

There are additional considerations related to introducing AI systems like LLMs into P-12 classrooms to serve as personal writing assistants. These technologies could harm students’ developing cognitive schema around writing and what it means to develop an identity as a writer. Just as writing offers a window into the minds of our students that teachers miss if they offload the assessment of student work to AI, these offloading acts also reduce the social and communal aspects of writing as communication. Andrejevic and Selwyn (2023) cautioned that AI’s biases toward individualization, operationalism, and depersonalization could work against the social, interpersonal, subjectivity-developing aspects involved when students grow into their identities as writers. Careful implementation informed by writing pedagogy would be needed to counteract these tendencies.

In addition to concerns about the costs of AI proxy grading as dehumanizing thinking and communicative human acts, we are also concerned about the losses of human agency and data privacy that occur in this scenario. Data privacy is a complex topic worth its own address, thus we highlight its importance but do not address this harm in detail here.   However, the losses of human agency and privacy are notable when writing that is personally created, considered, sweated over, and owned is shared with a profit-oriented platform which will use the writing as training data. As teachers of writing we know that writing is often intimate work that reveals who the writer is inside. Sharing student writing without the permission of learners seems to be a betrayal of a foundational professional responsibility of teaching.

Conclusions

The Benefits of This Moment’s Reflection

Teaching writing was not working for everyone before AI technologies entered the education setting. The contextual pressures of presentism in United States’ schools have been a documented constraint for 50 years (Lortie, 1975). Schools have accelerated their maladaptation to neoliberal, presentist pressures after policy responses to the A Nation at Risk report (Strauss, 2018), the No Child Left Behind Act (Hursh, 2007; Valli & Buese, 2007), and privatization and marketization initiatives resulting from Race To The Top legislation (Sleeter, 2012). Neoliberal influences in policy changed writing instruction dramatically, moving it away from the tendencies toward creative and process foci to sharpen attention to short-form response writing and highlighting writing assessment aimed to measure argumentation skills, the analysis of which is more easily quantified. The Covid-19 global pandemic further exacerbated education’s already marketized, solutionist relationship with edtech companies and their platforms. In addition, it has been suggested that the emphasis on science of reading instruction further decreases educator attention on the importance of writing instruction (Sawchuck, 2023).

Taken together, these variables in the dilemmatic space of teaching writing shape teachers’ epistemic beliefs about effective classroom practices. With this in mind, we recognize the opportunity AI brings as a catalyst for honest appraisal of teacher education systems, and how they do or do not prepare teachers to teach writing in ways that holistically develop dispositions in learners that writing is a valuable thinking and creative process, with multiple paths, outcomes, and endless possibilities. And yet, we are worried.

There was a brief moment in summer 2020 when conversations felt hopeful that the pandemic offered a reset opportunity for American education. Voices previously unheard were speaking out demanding racial justice and calling out unjust school discipline, inequitable learning opportunities, and the pervasive culture of low standards in many schools. Now, almost 4 years later, we have watched in dismay how many education settings have doubled down on the practices that should have changed. It is this doubling down effect that we fear with regard to the infusion of AI in the writing classroom.        

AI-assisted platform technologies are changing what it means to compose and write. These technologies will continue to proliferate within and across sites of learning. As such,  teacher educators, and specifically those who prepare preservice teachers of writing and understand themselves to be teachers of writing, must respond to this swiftly shifting landscape with articulated theory and ethics. Such response must consider how preservice teachers’ interior framing, their personal epistemologies about writing and teaching, mediate their interactions with AI-assisted technologies, especially considering the “coded gaze” of racist and gendered biases built into AI technologies (Buolamwini, 2023) and the corporate gain opportunities from forwarding technodeterminist and technoptimist narratives that AI tools will solve the problems of education (Foroohar, 2019; Krutka et al.,  2020; Zuboff, 2019).

Our Call to Action

We take heart, though, in the work of scholars grappling with and forwarding technoethical (Krutka et al., 2019) and technoskeptical positions (Krutka et al., 2020; Krutka et al., 2023; Pleasants et al., 2023) toward AI platform technologies. This developing line of scholarship must inform our conversations and practices with preservice teachers of writing. Indeed, the “technology education iceberg” (Pleasants et al., 2023, pp. 503-504) model offers teacher educators a framework for exploring the kinds of challenges that arise in the dilemmatic space of teaching writing with AI. This model frames three layers of perspective that can serve as a path for analyzing layers of interfacing with tools, values, and systems. Models like the technology education iceberg hearten us as we encourage teacher educators to engage in the challenging work of self-examination and critical digital literacy use of AI platform technologies such that they are better prepared to support preservice teachers of writing and their students to “confront technological dilemmas and make informed decisions for their futures” (p. 507).

Historically, research indicates that teacher educators have not participated in robust, reflective conversations about their roles as scholar-practitioners (Goodwin et al., 2014), which in turn, may further distance them from rapidly changing practices and policies in real-time P-12 classroom settings. Our research into theorizing writing instruction in a platformized landscape takes up the charge from Goodwin et al. (2014) that teacher educators “reflect upon… [and] learn more about negotiating among competing political demands” of teacher education work to more carefully articulate the work of educating preservice teachers to become compassionate, responsive, and effective teachers of children and their writing (p. 299).

This research work also highlights the need for teacher educators to engage in our own ongoing professional development, such that we are able to “aggressively investigate the practice of teacher education and offer professional education that reflects the needs of our students and the needs of our schools” (Grossman, 2008, p. 22). Such theorizing of practice is of vital importance when educational technology companies seek to profit off the integration and ubiquity of their products and the learning and teaching interactions in preservice teacher education and our children’s classrooms. 

In this conceptual article, we have argued that the fields of English language arts and composition studies require a more robustly elaborated theoretical perspective on the identity development of preservice teachers as teachers of writing due to the rapid proliferation of AI-assisted technologies in learning spaces. Reframing education settings as dilemmatic spaces (Fransson & Grannäs, 2013; Honig, 1994) within which teachers contend with constructions of self/other as teachers of writing provides the opportunity to consider the dynamic spatial, interpersonal, and intrapersonal conditions that shape the positions we take toward ourselves, others, and the tools we choose in day-to-day learning and teaching interactions.

Ultimately, teacher educators must forward technocritical inquiry opportunities and conversations with preservice teachers, such that all of us develop technoskeptical dispositions toward AI platform technologies and writing instruction. Doing so is more likely to ensure that decisions to integrate AI platforms into writing instruction are thoughtful, intentional, well-reasoned, and student-centered rather than driven by presentist concerns to maximize productivity.

References

Akkerman, S. F., & Meijer, P. C. (2011). A dialogical approach to conceptualizing teacher identity. Teaching and Teacher Education, 27(2), 308–319.

Allen, A., Scott, L. M., & Lewis, C. W. (2013). Racial microaggressions and African American and Hispanic students in urban schools: A call for culturally affirming education. Interdisciplinary Journal of Teaching and Learning, 3(2), 117-129.

Anagnostopoulos, D., Cavanna, J., & Charles-Harris, S. (2020). Managing to teach ambitiously in the first year. The Elementary School Journal, 120(4), 667-691.

Andrejevic, M., & Selwyn, N. (2023). Education in an era of pervasive automation. Postdigital Science and Education, 5(1), 220–231. https://doi.org/10.1007/s42438-022-00336-x

Bender, E. M., Gebru T., McMillan-Major, A., &  Shmitchell, S.. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–623. Virtual Event Canada: ACM. https://doi.org/10.1145/3442188.3445922 

Brown, A. D. (2022). Identities in and around organizations: Towards an identity work perspective. Human Relations, 75(7), 1205–1237.

Buolamwini, J. (2023). Unmasking AI: My mission to protect what is human in a world of machines. Random House.

Bushweller, K. (2022, May 17). What the massive shift to 1-to-1 computing means for schools, in charts. Education Week. https://www.edweek.org/technology/what-the-massive-shift-to-1-to-1-computing-means-for-schools-in-charts/2022/05

Cutler, L., & Graham, S. (2008). Primary grade writing instruction: A national survey. Journal of Educational Psychology, 100, 907-919.

Dockrell, J. E., Marshall, C. R., & Wyse, D. (2016). Teachers’ reported practices for teaching writing in England. Reading and Writing: An Interdisciplinary Journal, 29(3), 409–434. https://doi-org.ezproxy.lib.ou.edu/10.1007/s11145-015-9605-9

Elmore, R. F. (1995). Structural reform and educational practice. Educational Researcher, 24(9), 23-26.

Foroohar, R. (2019). Don’t be evil: How big tech betrayed its founding principles – and all of us. Currency.

Fransson, G. (2016). Manoeuvring in a digital dilemmatic space: Making sense of a digitised society. Nordic Journal of Digital Literacy. 11(3), 185-201. doi:18510.18261/issn.1891- 943x-2016-03-04

Fransson, G., & Grannäs, J. (2013). Dilemmatic spaces in educational contexts—Towards a conceptual framework for dilemmas in teachers’ work. Teachers and Teaching: Theory and Practice, 19(1), 4-17. http://dx.doi.org/10.1080/13540602.2013.744195

Garcia, A., & Nichols, T. P. (2021). Digital platforms aren’t mere tools —they’re complex environments. Phi Delta Kappan, 102(6), 14-19. https://doi.org/10.1177/0031721721998148

Gillespie, A., Graham, S., Kiuhara, S., & Hebert, M. (2014). High school teachers use of writing to support students’ learning: a national survey. Read and Writing, 27, 1043–1072 . https://doi.org/10.1007/s11145-013-9494-8

Giroux, H. A. (2011). On critical pedagogy. Bloomsbury.

Goodwin, A. L., Smith, L., Souto-Manning, M., Cheruvu, R., Tan, M. Y., Reed, R., & Taveras, L. (2014). What should teacher educators know and be able to do? Perspectives from practicing teacher educators. Journal of Teacher Education, 65(4), 284-302. https://doi.org/10.1177/0022487114535266

Graham, S. (2019). Changing how writing is taught. Review of Research in Education, 43(1), 277-303. https://doi-org.ezproxy.lib.ou.edu/10.3102/0091732X18821125

Graham, S. (2018). Introduction to conceptualizing writing. Educational Psychologist, 53(4), 217-219.

Graham, S. S. (2023). Post-process but not post-writing: Large language models and a future for composition pedagogy. Composition Studies, 51(1), 162-168.

Graham, S., & Harris, K. R. (2018). An examination of the design principles underlying a self-regulated strategy development study. Journal of Writing Research, 10(2), 139-187.

Grossman, P. L. (1990). The making of a teacher: Teacher knowledge and teacher education. Teachers College Press.

Grossman, P. L. (2008). Responding to our critics: From crisis to opportunity in research on teacher education. Journal of Teacher Education, 59(1), 10-23. https://doi.org/10.1177/0022487107310748

Grossman, P. L., Smagorinsky, P., & Valencia, S. (1999). Appropriating tools for teaching English: A theoretical framework for research on learning to teach. American Journal of Education, 108(1), 1-29.

Hargreaves, A. (1998). The emotional practice of teaching. Teaching and Teacher Education, 14(8), 835–854.  https://doi.org/10.1016/S0742-051X(98)00025-0

Hargreaves, A. (2000). Mixed emotions: Teachers’ perceptions of their interactions with students. Teaching and Teacher Education, 16, 811–826.

Hargreaves, A., & Shirley, D. (2009). The persistence of presentism. Teachers College Record, 111(11), 2505-2534.

Heuer, H., Jarke, J., & Breiter, A. (2021). Machine learning in tutorials – Universal applicability, underinformed application, and other misconceptions. Big Data & Society, 8(1). https://doi.org/10.1177/20539517211017593

Hofer, B. K., & Pintrich, P. R. (1997). The development of epistemological theories: Beliefs about knowledge and knowing and their relation to learning. Review of Educational Research, 67(1), 88-140. https://doi.org/10.3102/00346543067001088

Holland D., & Lave J. (2001). History in person: Enduring struggles, contentious practices, intimate identities. School of American Research Press.

Honig, B. (1994). Difference, dilemmas, and the politics of home. Social Research, 61(3). 563-597. https://www.jstor.org/stable/40971048.

hooks, b. (2014). Teaching to transgress. Routledge.

Hsiang, T. P., & Graham, S. (2016). Teaching writing in grades 4–6 in urban schools in the Greater China Region. Reading and Writing, 29(5), 869–902. https://link.springer.com/article/10.1007/s11145-015-9597-5

Hursh, D. (2007). Assessing “No Child Left Behind” and the rise of neoliberal education policies. American Educational Research Journal, 44(3), 493–518. http://www.jstor.org/stable/30069426

Jamieson, S. (2022). The AI “crisis” and a (re)turn to pedagogy. Composition Studies, 50(3), 153–157.

Kohnen, A. M. (2019). Becoming a teacher of writing: An analysis of the identity resources offered to preservice teachers across contexts. English Education, 51(4), 348–375. https://www.jstor.org/stable/26797046

Krutka, D. G., Heath, M. K., & Mason, L. E. (2020). Editorial: Technology won’t save us – A call for technoskepticism in social studies. Contemporary Issues in Technology and Teacher Education, 20(1), 108-120. https://citejournal.org/volume-20/issue-1-20/social-studies/editorial-technology-wont-save-us-a-call-for-technoskepticism-in-social-studies

Krutka, D.G., Heath, M.K., & Willet, K. B. S. (2019). Foregrounding technoethics: Toward critical perspectives in technology and teacher education. Journal of Technology and Teacher Education. 27(4), 555-574.

Krutka, D. G., Pleasants, J., & Nichols, T. P. (2023). Talking the technology talk. Phi Delta Kappan, 104(7), 42-46.

Kumar, P.,Vitak, J., Chetty, M., & Clegg, T. (2019). The platformization of the classroom: Teachers as surveillant consumers. Surveillance & Society, 17, 145-152. doi: 10.24908/ss.v17i1/2.12926

Lasky, S. (2005). A sociocultural approach to understanding teacher identity, agency and professional vulnerability in a context of secondary school reform. Teaching and Teacher Education, 21(8), 899-916.

Lortie, D. C. (1975). Schoolteacher: a sociological study. University of Chicago Press.

Macgilchrist, F. (2021). What is ‘critical’ in critical studies of edtech? Three responses. Learning, Media and Technology, 46(3), p. 243-249. doi: 10.1080/17439884.2021.1958843

Mockler, N. (2011). Beyond ‘what works’: Understanding teacher identity as a practical and political tool. Teachers and Teaching, 17(5), 517–528.

Morrison, A. (2023). Meta-writing: AI and writing. Composition Studies, 51(1), 155-219.

Nichols, T., & Garcia, A. (2022). Platform studies in education. Harvard Educational Review 92, (2), 209–30. https://doi.org/10.17763/1943-5045-92.2.209.

Pangrazio, L., Stornaiuolo, A., Nichols, T.P.,  Garcia, A., & Philip, T.M. (2022). Datafication meets platformization: Materializing data processes in teaching and learning. Harvard Educational Review, 92(2), 257–83. https://doi.org/10.17763/1943-5045-92.2.257.

Paris, D., & Alim, H. S. (2017). Culturally sustaining pedagogies: Teaching and learning for justice in a changing world. Teacher College Press.

Pearson, M., & Burns, P. (Hosts). (2023, Sept. 25). CriticalAI and the probably machines: What we don’t know just might hurt us. (S2 E6) [Audio podcast episode]. In The Bot Brothers: AI for Educators. RSS.com  https://rss.com/podcasts/thebotbrothers/1139306/

Pleasants, J., & Krutka, D. & Nichols, T. (2023). What relationships do we want with technology? Toward
technoskepticism in schools. Harvard Educational Review, 93(4), 486-515. doi: 10.17763/1943-5045-93.4.486

Ray, A. B., Graham, S., Houston, J. D., & Harris, K. (2016). Teachers use of writing to support students’ learning in middle school: A national survey in the United States. Reading and Writing. 29(5), 1039 – 1068. https://doi.org/10.1007/s11145-015-9602-z

Romano, T. (2007). Teaching writing from the inside. In K. Beers, R. Probst, & L. Rief (Eds.), Adolescent literacy: Turning promise into practice (pp. 167-178). Heinemann.

Rushton, E. A., Rawlings Smith, E., Steadman, S., & Towers, E. (2023). Understanding teacher identity in teachers’ professional lives: A systematic review of the literature. Review of Education, 11(2), e3417. doi: 10.1002/rev3.3417 

Sapir, A., & Mizrahi‐Shtelman, R. (2024). ‘We are not babysitters’: Meaningfulness and meaninglessness in homeroom teachers’ identity work. European Journal of Education, e12714. https://doi.org/10.1111/ejed.12714

Sawchuck, S. (2023, January 17). How does writing fit into the ‘science of reading’? Education Week. https://www.edweek.org/teaching-learning/how-does-writing-fit-into-the-science-of-reading/2023/01

Sandoval, W. A., Greene, J. A., & Bråten, I. (2016). Understanding and promoting thinking about knowledge: Origins, issues, and future directions of research on epistemic cognition. Review of Research in Education, 40(1), 457-496. https://doi.org/10.3102/0091732X16669319

Scott, A. (2023, June 21). Reading in the time of books bans and A.I. The New York Times. https://www.nytimes.com/2023/06/21/books/review/book-bans-humanities-ai.html

Selwyn, N. (2023). Constructive criticism? Working with (rather than against) the AIED back-lash. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-023-00344-3

Selwyn, N. (2022). The future of AI and education: Some cautionary notes. European Journal of Education, 57(4), 620–31. https://doi.org/10.1111/ejed.12532

Shulman, L.S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher, 15(2), 4-14.

Singer, N. (2017, May 13). How Google took over the classroom. The New York Times. https://www.nytimes.com/2017/05/13/technology/google-education-chromebooks-schools.html

Sleeter, C. E. (2012). Confronting the marginalization of culturally responsive pedagogy. Urban Education, 47(3), 562-584. https://doi.org/10.1177/0042085911431472

Smagorinsky, P. (2010). The culture of learning to teach: The self-perpetuating cycle of conservative schooling. Teacher Education Quarterly, 37(2), 19–31.

Smith, A. (2019). Waves of theory building in writing and its development, and their implications for instruction, assessment, and curriculum. In D. E. Alvermann, N. J. Unrau, M. Sailors, & R. Ruddell (Eds.), Theoretical models and processes of literacy (7th ed.; pp. 65-83). Routledge.

Strauss, V. (2018, April 26). ‘A Nation at Risk’ demanded education reform 35 years ago. Here’s how it’s been bungled ever since. The Washington Post. https://www.washingtonpost.com/news/answer-sheet/wp/2018/04/26/the-landmark-a-nation-at-risk-called-for-education-reform-35-years-ago-heres-how-it-was-bungled/

Street, C. (2003). Pre-service teachers’ attitudes about writing and learning to teach writing: Implications for teacher educators. Teacher Education Quarterly, 30(3), 33–50. http://www.jstor.org/stable/23478439

Tassoni, J. P. (2024). Fit to resist in post-product space: Underserved student populations and generative AI’s writing norms. In B. Buyserie & T. Thurston (Eds.), Teaching and generative AI (pp. 201-216). Utah State University.

Toncic, J. C. (2022). Advancing a critical artificial intelligence theory for schooling. Teknokultura. Revista de Cultura Digital y Movimientos Sociales, 19(1), 13–24. https://doi.org/10.5209/tekn.71136.

Valli, L., & Buese, D. (2007). The changing roles of teachers in an era of high-stakes accountability. American Educational Research Journal, 44(3), 519-558. https://doi-org.ezproxy.lib.ou.edu/10.3102/0002831207306859

van Dijck, J. (2013).  The culture of connectivity: A critical history of social media. Oxford Academic.  https://doi.org/10.1093/acprof:oso/9780199970773.001.0001.

Ware, F. (2006). Warm demander pedagogy: Culturally responsive teaching that supports a culture of achievement for African American students. Urban Education, 41(4), 427-456.

Wertsch, J. V. (1991). Voices of the mind: A sociocultural approach to mediated action. Harvard University Press.

Woodward, L., & Cho, B. Y. (2020). How students’ beliefs about knowledge matter in multiple‐source reading online: Implications for classroom instruction. Journal of Adolescent & Adult Literacy, 64(2), 135-144.

Zuboff, S. (2019). The age of surveillance capitalism: The fight for a human future at the new frontier of power. Public Affairs.

Zuidema, L. A., & Fredricksen, J. E. (2016). Resources preservice teachers use to think about student writing. Research in the Teaching of English, 51(1), 12–36. http://www.jstor.org/stable/24889932