Early-Career Language Teachers and Critical Incidents: Agency as a Driver of Identity Construction

May 2026 – Volume 30, Number 1

https://doi.org/10.55593/ej.30117a2

Cameron Flinn
Waseda University, Tokyo, Japan
<cameron.flinnfuji.waseda.jp>

Gary G. Fogal
Sophia University, Tokyo, Japan
<ggfogalsophia.ac.jp>

Abstract

Critical incidents (CIs) are unexpected events that shape early-career language teachers’ professional development. Although research has explored how CIs impact teachers’ beliefs and practices, less attention has focused on agency—how teachers narrate, respond to, and position themselves in relation to these events—in shaping how CIs are integrated into their professional identities. This study draws on complex dynamic systems theory and Hiver and Dörnyei’s (2017) language teacher immunity framework to examine how seven early-career English language teachers interpret and respond to CIs, and the role agency plays in shaping their emerging professional identities. Analysing interview data, we trace participants’ responses across the four phases of Hiver and Dörnyei’s framework: triggering, linking, realignment, and stabilisation. Findings show that teachers’ practices either adapted into productive, stable states of growth and resilience by leveraging contextual and social affordances or transformed into maladaptive states of resistance. We interpret outcome differences based on variations in how teachers assert agency in response to initial conditions such as their backgrounds, experiences, and environments. The findings contribute to understanding the interplay between agency and context in identity construction and underscore the importance of communities of practice and supportive environments that help early-career TESOL professionals grow from CIs.

Keywords: complexity theory; critical incidents; L2 teaching; language teacher cognition; professional development

As a profession, language teaching is shaped by a range of pedagogical and structural tensions. These include competing beliefs about how languages are best acquired (consider, for instance, different orientations to explicit and implicit language knowledge—Foryś-Nogala et al., 2022; Fu & Li, 2021) as well as tensions between teachers’ own beliefs and language policies that shape and constrain L2 classroom practices. For example, in Japan, national policies call for greater attention to communication skills and increased use of English in classrooms, potentially giving rise to tensions with teachers’ pedagogical beliefs and other policy pressures such as entrance-exam requirements (Fairbrother, 2022).

For early-career language teachers (ECLTs), these challenges are experienced with even greater intensity. As novices, ECLTs often negotiate tensions between their emerging beliefs about language learning and institutional expectations (e.g., ECLTs reconciling process- versus product-oriented approaches—Yu et al., 2020). Moreover, because competence in the language of instruction cannot be fully developed through formal teacher training alone and requires sustained practice, ECLTs who teach their L2 often report low subject-specific self-efficacy (Borg, 2006; Sulis et al., 2023). Together, these tensions frequently lead to “reality shock,” where new language teachers struggle to reconcile their expectations with the complex demands of the profession (Davis & Borden, 2025; Farrell, 2012).

One way these early-career challenges have been explored is through critical incidents (CIs). Brookfield (1990) defines a CI as “any vividly remembered event which is unplanned and unanticipated” (p. 84). Documented CIs include, for example, a student’s unexpectedly cheerful reaction to punishment (Megawati et al., 2020), a teacher’s fear of losing authority after being corrected (Morales & Mena, 2017), and students being excluded due to gender imbalances in group work (Farrell, 2008).

However, not every unexpected incident qualifies as a CI. As Tripp (1993) notes, an incident becomes “critical” when a teacher interprets it as a significant turning point that triggers reflection and change. Reflection involves making sense of the CI and deliberately selecting the best course of action to address it—often as part of a broader reassessment of one’s beliefs about language teaching and learning. Because CIs present unanticipated challenges, this reassessment can lead teachers to confront previously taken-for-granted assumptions about, for example, their role as teachers, pedagogical decisions, or classroom management. They may then experiment with new strategies to better align their practices with the demands of the moment (Shapira-Lishchinsky, 2011). In this way, CIs become turning points because they reveal tensions or blind spots in teachers’ beliefs about teaching and professional practice, thus spurring personal and professional growth (Farrell, 2013). The reflective nature of CIs also underscores their situatedness, so that turning points will differ among teachers. To discuss such occasions within the context of ECLTs, we define CIs as unexpected events that interrupt the routine course of teachers’ work and compel them to make sense of the disruption.

One important but underexplored aspect of how language teachers adapt after experiencing a CI is agency—teachers’ capacity to make intentional choices and take meaningful action within the opportunities and constraints of their specific contexts (Edwards & Burns, 2016). Agency is both internal and external: it encompasses teachers’ internal interpretations and sense-making of their experiences, as well as the external conditions that enable or constrain their actions (Nazari et al., 2023; Schutz et al., 2018). In the context of CIs, the internal aspect of agency can be seen in how teachers frame incidents—for example, as within their control (e.g., “I can adjust my lesson plan to address this issue”) or outside their control (e.g., “Institutional policies prevent me from changing this aspect of my class”) (Schutz et al., 2018). Thus, exercising agency post-CI involves both internal processes of interpreting and making sense of these unexpected challenges, as well as deciding how to act within the external constraints and possibilities of a given teaching context.

Because agency shapes—and is shaped by—teachers’ professional identities (Barkhuizen, 2016; Norton, 2013; Trent, 2017), responses to CIs often have lasting implications for how teachers see themselves and their work (Babaii et al., 2022; Kılıç & Cinkara, 2020). For example, in Williams and Grierson (2016), the CIs of teacher trainers working in the Cook Islands and Kenya caused them to feel unsettled by the countries’ colonial histories, emerging through narratives in which they characterised themselves as a force imposing Western educational values. Reflecting on CIs they experienced, they came to see their role in terms of a “double image”—that is, not only who they are but who they are from the perspective of others. This interpretation enhanced their cultural sensitivity and was integrated into their identities as teacher educators. Understanding these agentive responses is important because examining how instructors react to CIs can reveal key processes of negotiation that impact pedagogy, personal wellbeing, and identity. These insights can, in turn, contribute to policy, administrative, and teacher-training decisions about how best to support ECLTs dealing with CIs. This is especially critical for ECLTs because their identities are still in flux, and they are more vulnerable to experiencing CIs than their more experienced counterparts (Farrell, 2008).

Despite these insights, very little research has explored how ECLTs exert agency when faced with uncertainty and adversity, such as that frequently experienced in CIs (Davis & Borden, 2025). Although identity development is documented in CI research (e.g., Babaii et al., 2022), far less is known about the agentive processes through which teachers’ interpretations and actions feed into these identity outcomes, particularly in relation to ECLTs. For example, a small number of existing studies have demonstrated how structured, researcher-designed interventions—such as communities of practice (CoPs) formed through instant messaging apps (Nazari & De Costa, 2021) or formal training programmes (Karimi & Nazari, 2021)—can positively influence language teachers’ agentive decision-making. However, much less is known about how ECLTs exercise agency when responding to CIs in more informal and unstructured settings where such support may not be readily available.

To address these gaps, this study investigates how seven early-career EFL teachers navigated CIs in their everyday teaching contexts, how they exercised agency in responding to these challenges, and what consequences these experiences had for their emerging professional identities. To address our focus on agency and identity, we draw on complex dynamic systems theory (CDST) to conceptualise teachers’ evolving selves as emergent and continually reshaped through interaction with the teaching environment (Norton & De Costa, 2018). As a framework, CDST is well suited for dealing with the well-referenced unpredictable, chaotic, and nonlinear nature of teacher cognition (Feryok, 2010; Zheng, 2013), one that conceptualises teacher cognition as adaptive, self-organising, and sensitive to initial conditions (Hiver, 2022). Within this framework, we regard identity as one component of a broader cognitive system (i.e., of teacher cognition) (Borg, 2019), which encompasses knowledge (Golombek, 1998), emotions (Song, 2016; Taşdemir & Seferoğlu, 2024), and agency (Li, 2020), all of which interact in complex ways as teachers make sense of their professional CI-related experiences. Following a review of the literature, this framework is unpacked in detail. We then outline our methodological approach and discuss the results by focusing on four participants who best illustrate divergent identity outcomes. The paper concludes with implications for teaching and teacher training contexts.

Literature Review

Critical Incidents

Two notable foci in the CI literature highlight gaps where a CDST approach may prove fruitful. The first major strand concentrates on categorising incidents according to their cause, such as student behaviour, materials, or assessment (Atai & Nejadghanbar, 2016; Karimi & Nazari, 2021; Megawati et al., 2020). For example, Atai and Nejadghanbar detailed six categories of CIs of EFL teachers in Iran, describing each and how teachers responded to them. Farrell (2008) also identified categories of CIs (n = 9) from teacher trainees in Singapore, but noted the difficulty in generating distinct classifications. These studies tend to oversimplify the nature of CIs by attributing them to specific, isolated factors.

In reality, however, CIs are rarely triggered by a single variable, and related univariate analyses fail to capture related complexities. For example, in Karimi and Nazari’s (2021) study of Iranian EFL teachers’ CIs, the authors discuss the complexity of learners’ fixation on materials—noting that the source of such CIs may be driven by parental expectations and broader societal pressures. With this point in mind, categorisation-based studies often assume that such categories may render a degree of generalisability concerning how CIs form and how teachers respond. However, studies focused on classification overlook that CIs often emerge within a broader, interconnected educational landscape (Farrell, 2008). Moving beyond a classification agenda, accounting for this complexity (through our approach, via CDST) is crucial for understanding how CIs are formed and subsequently for developing effective strategies to address them.

A second major strand of CI research leans on narrative inquiry to provide descriptive insights into teacher cognition, defined as “an enveloping term covering states and processes that may be as different (or similar) as beliefs, attitudes, and knowledge” (Feryok, 2010, p. 274). Although not always explicitly framed through a “teacher cognition” lens, these studies (e.g., Farrell, 2013; Golombek & Johnson, 2004; Lengeling & Pablo, 2016) nevertheless reveal how instructors think, understand, and implement their roles and practices through CIs. For example, Lengeling and Pablo (2016) used a narrative approach to analyse reflections of CIs by eight novice EFL teachers in Mexico, noting that the CIs allowed instructors to gain new insights and adjust their teaching.

Without discounting the empirical value of these discursive and descriptive inquiries, narrative methods often lack a systematic analysis of how various factors informing cognition—such as existing beliefs, context, and other agents—interact dynamically to influence teachers’ development from a CI. The relatively recent attention in our field to interaction-based studies under the CDST umbrella compel us to view teachers’ thinking and behaviour as agentive and context-embedded (Burns et al., 2015), emerging through dynamic interactions with past experiences (Kiss, 2012), sociopolitical discourses (Kubanyiova & Crookes, 2016), and environmental constraints (Kiss, 2012).

In the following section, we discuss the literature on these dynamic interactions between cognition, context, and agency and how they play a central role in shaping teachers’ identities, including whether and how they adapt following the occurrence of a CI.

Critical Incidents: Identity Development and Agency

As ECLTs enter the profession, they must navigate different CIs informed by, for instance, tension between their imagined teaching ideals and the unpredictable realities of the classroom (Taşdemir & Seferoğlu, 2024). For many, a mismatch here leads to anxiety (Farrell, 2009), dissatisfaction (Sulis et al., 2022), or even leaving the profession altogether (Trent, 2017). Yet not all ECLTs respond to CIs in the same way. While some struggle, others adapt and establish themselves as legitimate, confident members of the profession (Gu & Benson, 2015). This variation has turned attention toward how teachers’ sense of identity and their capacity to act—that is, their agency—shape these different outcomes (Cobb, 2022).

Language teacher identity can be understood as instructors’ “dynamic self-conception and imagination of themselves as teachers, which shifts as they participate in varying communities, interact with others, and position themselves (and are positioned by others) in social contexts” (Yazan, 2018, p. 21). Agency, meanwhile, refers to teachers’ capacity to make purposeful choices and act within the opportunities and constraints of their working contexts (Edwards & Burns, 2016). While agency is widely recognised as crucial to how teachers “defend, negotiate or transform their identities” (Cobb, 2022, p. 3), there is still limited understanding of how ECLTs draw on contextual resources and exercise agency to deal with unexpected dilemmas (such as CIs) (Davis & Borden, 2025).

Recent scholarship on language teacher identity highlights that agency is not only shaped by external structures but also enacted through internal processes of self-authorship—that is, the active sense-making and self-positioning through which they construct who they are as professionals (Kayi-Aydar, 2015). In this view, teachers draw on personal factors such as prior experience when determining how to respond in ways that either reinforce aspects of their professional identity or enable it to shift in light of new challenges (Cobb, 2022). In the context of CIs, teachers’ agency shapes not only how they navigate external constraints but also how they make sense of disruptive moments through the lens of their own beliefs, histories, and imagined selves. For example, Hurst and Brantlinger (2022) documented the dynamic ways in which agency influences whether a CI is seen as positive or negative, shaped by factors such as the level of perceived support from the school or whether teachers saw aspects of their former selves in their students. In one of the very few studies that explicitly address the link between agency, identity, and CIs, Sisson (2016) reported on how one teacher’s CI ignited a series of agentive decisions, which helped shift her identity from low-achieving to high-achieving. This identity shift empowered her to become a teacher who was caring and advocated for her students’ needs despite various curricular constraints. Sisson’s focus on agency as an outcome demonstrates how CIs can also become opportunities to exert agency and perform new identities in response. Taken together, these examples illustrate that identity and agency cannot be disentangled: teachers continuously (re)author who they are through the ways they interpret and respond to CIs they may experience.

Further studies suggest that this sense-making process—how teachers interpret and frame CIs—can significantly shape the developmental pathways of their professional identities (Nazari & De Costa, 2021; Schutz et al., 2018). For example, Kılıç and Cinkara (2020) found that pre-service EFL teachers leveraged CIs to adopt more positive attitudes toward language learning and envisioned new ways to apply these insights to their practices as in-service teachers. In contrast, Morales and Mena (2017) showed that negatively framed CIs could destabilise teachers’ sense of self, leading to lost motivation or even leaving the profession. Similarly, McGarr and McCormack (2016) illustrated how teachers drew upon personal experiences and habits to form non-critical interpretations of CIs, resulting in counterfactual thinking and the strengthening of existing beliefs rather than productive and critical reflection. Taken together, these studies show that differences in how teachers agentively author and interpret CIs—framing them as obstacles, opportunities, or something to be resisted—lead to very different identity outcomes (Edwards & Burns, 2016; Kiss, 2012; Rahman & Singh, 2021; Ruohotie-Lyhty, 2018).

Although meaningful, much of the empirical work on CIs remains largely descriptive, often lacking a robust theoretical framework that captures the dynamic role of agency in shaping identity development (e.g., Atai & Nejadghanbar, 2016; Kılıç & Cinkara, 2020; Megawati et al., 2020). In addition, most existing studies on CIs occur “in formal course settings monitored and managed by language teacher educators” (Yuan & Wang, 2025, p. 485), where the mechanisms of identity construction—such as reflection and emotional labour—are guided through in-service training (e.g., Atai & Nejadghanbar, 2016; Badia et al., 2021) or structured interventions like scaffolded CoPs (e.g., Nazari & De Costa, 2021). However, these settings are rarely available to ECLTs whose professional identities are still in flux and who often find themselves in relative isolation (Kidwell, 2025). Given that identity is fluid (Kubota, 2016), emergent (Miller et al., 2017), and socially negotiated (Yazan, 2018), we saw a need to capture the situated and unfiltered ways in which ECLTs encounter and agentively respond to CIs during this critical transition period of their careers.

To address these gaps, this study investigates how seven ECLTs make sense of CIs in their everyday teaching contexts and the role that agency plays in shaping their evolving professional identities. Moreover, in the absence of a dedicated framework for capturing these dynamic processes, we draw on CDST.

A Complex Dynamic Systems Framework for Exploring Critical Incidents

Research on the impact of CIs on ECLTs may benefit from exploring them as a complex system mediated by teachers’ sense-making, agency, and broader socio-institutional conditions. Doing so may help researchers understand how CIs shape teachers and their emergent identities, thus providing more principled means of assisting ECLTs when engaging with CIs.

Interpreted as a complex system, teacher cognition encompasses cognitive processes related to how teachers think, understand, and implement their roles and practices in educational settings (Li, 2020). Teacher cognition is marked by agency and its ability to adapt and self-organise in response to change, thereby creating coherent patterns of thought from chaotic experiences (Hiver & Dörnyei, 2017; Larsen-Freeman & Cameron, 2008). To better appreciate these processes, a key feature of complex dynamic systems—initial conditions—proves helpful. Initial conditions underscore the sensitivity of said systems to earlier states, signifying that small differences in factors such as context or prior experience can lead to variation in outcomes (Kiss, 2012; Larsen-Freeman & Cameron, 2008). This focus on initial conditions facilitates a novel view of teachers’ responses to CIs as shaped by past experiences, belief systems, and broader educational ecologies. These initial conditions impact how teachers perceive and respond agentively to CIs, influencing the adaptations they make in their professional practices and identities (McGarr & McCormack, 2016). By understanding how these conditions inform teacher agentive behaviour through their reactions to CIs, we can better support the paths that teachers take in response to professional challenges.

To assist with our understanding, the concept of language teacher immunity (Hiver & Dörnyei, 2017), grounded in CDST, provides a structured analytical framework by which to examine teacher cognition. Hiver and Dörnyei posit that teachers develop adaptive coping mechanisms during times of hardship, potentially enhancing their resilience and effectiveness in the face of adversity. The process of building this immunity involves four stages (triggering, linking, realignment, and stabilisation), each representing a phase in the teacher’s journey from experiencing a CI to integrating it into their professional identity (Hiver & Dörnyei, 2017). These stages can lead to resilience and growth, or rigidity and hindered development. Moreover, these stages are a part of a broader cognitive system that includes developing a deeper understanding of one’s professional role and agentively recalibrating one’s identity in response to feedback from the environment. Thus, the language teacher immunity framework aligns closely with teachers’ developmental process after a CI, by highlighting how adaptive changes, triggered by CIs, facilitate the emergence of (un)stable professional identities. The stages of Hiver and Dörnyei’s framework include the following.

Triggering stage: contains a “destabilising event” (Hiver & Dörnyei, 2017); CIs can function similarly. Early-career teachers—those with less than seven years of experience (Day, 2017; Sulis et al., 2022)—are more vulnerable to destabilising events due to their limited exposure to teaching (Farrell, 2008). Moreover, initial conditions (e.g., prior experiences) create unpredictability surrounding when and if an event is “triggered” as a CI because these are “beyond the control of the educator” (Kiss, 2012, p. 19). For a CI to have a triggering effect, it first has to be “noticed” (Finch, 2010) by imbuing it with a situated and subjective meaning (Tripp, 1993). Accordingly, identifying an event as “critical” is tied to initial conditions shaped by teachers’ unique backgrounds and experiences, and these conditions also impact how educators interpret and evaluate a CI.

Linking stage: involves teachers developing strategies which are “linked” to the CI, allowing them to cope with change (Feryok, 2010). Initial conditions embedded within the teaching context—such as strong interpersonal relationships (Le Cornu, 2013) and CoPs (Wenger, 1998)—can positively mediate teachers’ ability to overcome dilemmas, particularly in their early years. In contrast, insufficient training and a lack of mentor support can create unfavourable initial conditions for teachers (i.e., they may lack much needed strategies for dealing with adversity). Thus, recognising the impact of initial conditions is crucial for understanding teachers’ co-adaptive responses and long-term career trajectories post-CI.

Realignment stage: represents how teachers’ coping strategies reflect a shift in their cognition. Realignment often involves reconciling conflicting beliefs and practices (Zheng, 2013) or adjusting one’s approach in response to the unique conditions of the classroom (Li, 2013). The contextual nature of cognition thus compels us to view teachers’ thinking as nested within other micro-, meso-, and macro-level systems (Ludlow et al., 2017) such as classroom, administrative, and policy concerns, respectively. Realignment is not simply a matter of adopting new approaches but is also a reflection of teachers’ evolving interpretations and understandings of their experiences and the educational landscape.

Stabilisation stage: recognizes that teacher cognition self-organises by integrating “the emergent outcome as a component of their professional identity” (Hiver & Dörnyei, 2017, p. 413). Stabilisation occurs when teachers construct a narrative around the event which helps to contextualise and create meaning from the experience. The relationship between teachers’ constructed narratives and their career decisions is complex and may be unpredictable. Thus, how teachers’ identities adapt post-CI can result in different career trajectories (Hurst & Brantlinger, 2022) and these differences stem from conditions largely associated with their workplaces, in turn affecting how teachers perceive themselves and construct their professional narratives.

The framework posited by Hiver and Dörnyei (2017) supports a view of teacher cognition as a shifting phenomenon—one that is nested within a complex ecological structure which mediates the formation, noticing, and emergent identity-related outcomes of CIs. In doing so, the framework serves as a valuable tool for addressing the previously mentioned gaps in the literature: the absence of a systematic approach to CIs and the lack of a comprehensive framework for tracing cognitive development through CIs. With this framework in mind, then, the following research questions explore the self-organisation process of teachers’ cognitions and practices after experiencing CIs:

RQ1.   What CIs (if any) do early-career stage language teachers experience, and what leads to their formation?

RQ2.   What role(s) might teachers’ agency play as they negotiate their CIs?

RQ3.   How do the reported CIs shape early-career stage language teachers’ cognition, including their identities?

Methods

The study used qualitative data obtained through recorded interviews with seven practicing language teachers. Working with a small sample size is consistent with recent publications on identity and CIs (e.g., Derakhshan & Nazari, 2023; Nazari & De Costa, 2021) and provides researchers access to a detailed, nuanced exploration of individual cases that may be lost with larger samples. The sample size also reflects the aim of our study, which was not to generalise across all ECLTs, but to offer a contextualised, theory-informed account of how teachers make sense of and develop from CIs, and the role that agency plays in that process.

The participants were recruited from a large urban centre in Japan over a two-month period. Their prior teaching experiences were diverse and spanned different countries, including Vietnam, the United States, Hong Kong, and Japan, as well as a wide array of institutional types, including high schools, cram schools, universities, and private instructional settings.

Participants

Given our focus on teachers’ identity development processes in authentic teaching contexts, the study targeted participants outside of controlled teacher training conditions. Accordingly, convenience sampling was used to access practicing EFL instructors enrolled in a TESOL-related graduate program at a Japanese university. This program was research-based and not designed as a professional development course for language teachers. The first author was an insider within this community, sharing a similar professional background and career stage with the participants. This dual role as peer and researcher informed our positionality and helped to build the rapport and trust needed for participants to share their CIs candidly and in depth. This trust was essential for ensuring the authenticity of the data, as it helped the participants feel that their experiences would be interpreted fairly.

Biodata were collected to obtain a profile of each participant (see Table 1). This information helped identify diverse starting points (e.g., unique first languages and distinct home countries) and commonalities (e.g., all were working as university-level EFL instructors in Japan) of each participant, allowing us to explore how initial conditions might shape the ways they experience CIs and exercise agency in response.

Table 1. Participant Profiles

Name	Context of critical incident	Years teach-ing	Home country & L1	Educational history	Teaching history	Qualification(s) at time of CI
Minh	University instructor in Japan	~3.5	Vietnam	Completed schooling and BA in Vietnam; MA in Japan	EFL at language center in Vietnam and university in Japan	BA and MA in TESOL
Aiko	University instructor in Japan	~4	Japan	Completed elementary school in US and UK; completed all subsequent education in Japan	EFL at high schools and universities in Japan	BA and MA in TESOL (enrolled in PhD in TESOL)
Erica	University instructor in Japan	7	Vietnam	Completed all education in Vietnam; enrolled in Ph.D. studies in Japan	EFL at language center in Vietnam and university in Japan	BA and MA in TESOL (enrol-led in PhD in TESOL)
Maria	University instructor in the U.S.A.	~4	Italy	Completed all education in Italy; study abroad program in the USA for university	Italian as a FL in Italy; EFL at university in Japan	Enrolled in BA in Foreign Languages in Italy
Kenji	University instructor in Japan	~4.5	Japan	Completed all education in Japan	EFL at several universities in Japan	MA in Language Teaching and Literature
Roxanne	Cram school instructor in Japan	~6	Japan	Completed all education in Japan (elementary, secondary school, and university in English-medium institutions)	EFL to young learners at two cram schools in Japan	BA in Liberal Arts; MA in Sociolinguistics
Emily	Private teacher in Hong Kong	3	Hong Kong	Completed all education in Hong Kong	EFL individual tutoring to students in Hong Kong; EFL at several universities in Japan	Enrolled in BA in English Studies

Data Collection

After agreeing to participate in the study, the instructors were contacted via email and requested to reflect upon at least one CI they had experienced as an ECLT. The email also outlined the definition and nature of a CI (following Brookfield’s (1990) definition, noted earlier) to ensure clarity and consistency. Brookfield’s definition is broad, and we used this intentionally as a general starting point for teachers to reflect back upon their experiences. The teachers were also informed that CIs usually promote reflection (Schön, 1987) and often have an impact on their beliefs and practices (Romano, 2006).

Semi-structured interviews were conducted individually with the seven participants (six participants volunteered to discuss two CIs). Due to scheduling concerns, the interviews were conducted via Zoom. We adopted Farrell’s (2013) four-stage framework (Orientation, Complication, Evaluation, and Result) to guide the interviews. Orientation involves answering the “what, where, when, and who” of the incident. Complication addresses in more detail how the incident unfolded and what occurred. Evaluation encourages the participant to articulate the significance of the event—in this case why the CI was meaningful for them as a language teacher. Finally, Result considers the resolution and how the incident was dealt with in practical terms.

Following this interview protocol allowed us to probe each stage of the incident in detail and to elicit data about the teachers’ thought processes and decision-making. Doing so enhanced the trustworthiness of the data collection process by obtaining a thick and rich description (Carlson, 2010), encompassing not just what happened, but why, how, and what the CI meant to the participants themselves.

Further, to enhance the authenticity and confirmability of the data, the first author routinely rephrased participants’ answers throughout the interviews to check their understanding of participants’ accounts and ensure that their intended meanings were accurately interpreted. This approach functioned as a form of in-interview member checking, consistent with Sigstad and Garrels’ (2018) observation that “paraphrasing may fill the role of ‘member-checking’” (p. 702). All of the interviews were recorded with the interviewees’ permission and later transcribed. The interviews were conducted primarily in English, however, some interviewees used Japanese, and those segments were translated into English during the transcription process. The interviews ranged from 17–45 minutes (M = 32 min).

Ethical Procedures

This study was conducted in accordance with the “Responsibilities of Researchers” outlined in Japan’s Ministry of Education, Culture, Sports, Science and Technology’s Guidelines for Responding to Misconduct in Research Activities (MEXT, 2014). Accordingly, we followed strict data handling procedures and ensured that participants were informed of their rights and the voluntary nature of their participation before providing written consent. The teachers were fully informed about the purpose of the study, that their data would be used for this research and may be published, that their participation was voluntary and uncompensated, and that they could withdraw at any time without penalty. Participants were also informed that they might be quoted under a pseudonym and that all identifying details would be kept confidential. To ensure anonymity, all interview transcripts used pseudonyms, and any references to schools or other unique identifiers were removed.

Using Critical Incidents

A major challenge in CI research is operationalising “criticality.” In line with our focus on agency and with other studies that emphasise criticality as an interpretive process (Tripp, 1993), we wanted to allow for some degree of “openness” in how participants selected, framed, and discussed their CIs, recognising that what may seem insignificant to some teachers can hold profound meaning for other instructors (Farrell, 2008). At the same time, we wanted to ensure that the CIs aligned with our developmental framework—that is, that they had some meaningful (i.e., “critical”) impact on shaping the participants’ identities. To achieve this, we adapted Musanti and Pence’s (2010) criteria to discern CIs, as follows:

The CI should have:

• elicited a strong affective response (e.g., surprise) (Shapira-Lishchinsky, 2011);
• involved a degree of conflict which prompted teachers to seek (but not necessarily develop) coping mechanisms; and,
• demonstrated some sort of consequence or development regarding their professional identity (“who they are as a teacher”) (Halquist & Musanti, 2010).

All the CIs reported here met each of these three criteria.

Analysis

Analytical framework. To analyse the data, this study employed a grounded theory (GT) approach, compatible with CDST (Hiver & Al-Hoorie, 2016; Serafini & Roca-Ramirez, 2024). GT emphasises the dynamic nature of social phenomena, enabling an understanding of the interdependent relationship between teachers’ management of CIs, their associated cognitive processes, and the broader teaching context (Birks & Mills, 2015)—all key features of a CDST perspective on teacher cognition. This method followed previous studies analysing CIs (Atai & Nejadghanbar, 2016), including those using CDST (Karimi & Nazari, 2021).

Our analytical approach also extends upon previous work with CIs and teacher development (Atai & Nejadghanbar, 2016; Karimi & Nazari, 2021) by incorporating constructivist grounded theory (CGT) (Charmaz, 2006) and informed grounded theory (IGT) (Thornberg, 2012). Constructivist grounded theory aligns with our focus on agency and the influence of initial conditions on emergent outcomes by emphasising how individuals construct meanings and actions within larger social, institutional, and historical contexts. Similarly, IGT (Thornberg, 2012) rejects pure induction and instead embraces an abductive approach, recognising that data are social constructions shaped by historical, ideological, and socio-cultural contexts. In this view, insights from existing theories do not replace data-driven analysis but instead serve as “data sensitizing principles” (p. 249). This perspective allowed us to iteratively integrate the four stages of language teacher immunity as an organising tool and sensitising framework, while ensuring that our findings remained grounded in the data.

Coding. NVivo (version 14.23.0) was used to code the interview data following the three-stage procedure outlined by Strauss and Corbin (1990). In this approach, the researcher begins with open coding, which involves initially “breaking down, examining, comparing, conceptualising, and categorising data” (p. 61). This process is fully data driven and requires close reading and constant comparison, allowing codes to emerge out of raw data rather than relying on predetermined categories. This step primarily addressed RQ1 by surfacing what CIs teachers experienced (e.g., students’ lack of participation, struggles teaching online) and what conditions or triggers led to their emergence (e.g., assumptions about students’ cultural knowledge, Covid-19 pandemic). Extensive memos were also taken (Creswell, 2013), noting down key participant quotes and details of each code.

The next stage, axial coding, builds on the first stage by establishing relationships between established codes. This step involves examining how the codes relate to each other and identifying concepts that link them. While traditional GT (Glaser & Strauss, 1967) emphasises purely inductive coding, in this stage we drew on IGT (Thornberg, 2012), which rejects the notion of pure induction and argues that GT analysis can be “thoroughly grounded in data by GT methods while being informed by existing research literature and theoretical frameworks” (p. 249). In this way, such frameworks “can also work in tandem with Glaser’s as well as Strauss and Corbin’s versions of GT” (p. 249).

Based on our engagement with the literature, we found that the relationships between codes in our data naturally aligned with the four stages of language teacher immunity (triggering, linking, realignment, and stabilisation) (Hiver & Dörnyei, 2017). Therefore, in axial coding, we used these stages as a sensitising framework (Thornberg, 2012) to structure our findings.

This step primarily addressed RQ2 by identifying how teachers agentively negotiated the incidents in context—that is, how teachers positioned themselves to respond to the CIs by drawing on, or being constrained by, particular resources, relationships, and other initial conditions. We therefore grouped separately codes which pertained to initial conditions within each stage. For example, within the “linking” stage, three codes “contacted a colleague,” “connected with teachers online,” and “sought advice but failed” were all grouped into the parent code “Linking: Agentive mechanisms,” as they captured teachers’ efforts to mobilise support and coordinate action. We then created a separate parent code, “Linking: Coping mechanisms” to capture teachers’ concrete responses for managing the CI. This process ensured that our analysis remained grounded in the data, as codes emerged inductively in the open coding stage before being mapped onto the framework.

In addition to examining each stage, we identified higher-order patterns that connected teachers’ developmental trajectories across the four stages. For example, we wanted to understand if conditions like institutional support in the linking stage had any impact on the way teachers’ cognition changed in the later stages. This directly addressed RQ3 by tracing how teachers’ experiences with CIs shaped their cognition and identity outcomes. Therefore, in the selective coding stage, we refined our categories and traced links between conditions in earlier stages (e.g., triggering, linking) and their later consequences (e.g., realignment, stabilisation). This was an iterative process that allowed us to reconstruct the developmental trajectory of teachers’ identities, capturing how changes in cognition and practice were linked across the stages. Through this process, we identified two emergent CI pathways, described subsequently.

Results and Discussion

The NVivo analysis identified a total of 73 codes, comprising nine distinct subcategories across three major CI types, addressing RQ1 (see Table 2). The interviews resulted in identifying 13 CIs. In what follows, we first provide a brief overview of the most common types of CIs, contextualising them with reference to prior literature. We then present and interpret our findings according to Hiver and Dörnyei’s (2017) developmental framework.

Table 2. Categories of Critical Incidents (N = 13) and Interview Extracts

Major critical incident categories	Critical incident subcategories	Sample interview extract
Learner-related (n = 8)	Behaviour (n = 1)	“Normally what I would do is I would give them time to prepare, like 3 or 4 minutes… but then what I get from that preparation is nothing. They kind of just look at each other… and say nothing.” (Minh)
	Wellbeing (n = 2)	“I had a student who was mute [sic] in my English-speaking class. I didn’t understand much about this kind of student… if teachers force students to speak without readiness, that’s not good. This was one extreme case, but in a broad sense it applies to many students.” (Kenji)
	Misunderstanding tasks (n = 2)	“I provided them with all the instructions about how to work in pairs or groups, and how give feedback to their partners… [But] during the time when I went around the class to check their progress, one of the students asked me what we need to do…I thought, what just happened to my class? Was my instruction clear enough?” (Erica)
	Underperformance (n = 3)	“I kept teaching the same thing and the students still don’t get it. I was like okay, I told you about this last week or even 10 minutes ago, why wouldn’t you understand that?” (Roxanne)
Materials-related (n = 2)	Cultural references (n = 1)	“I had prepared this slide where I had … the [Donald] Duck family … If they had been in Italy, everyone would have known the relationship between those characters. The problem was these kids didn’t know at all. I did that because I assumed that people would know the same cultural reference, which they didn’t.” (Maria)
	Too many materials (n = 1)	“I finished the class in time, but then if you ask me what did I teach that day, then I probably wouldn’t be able to answer you, because there were so many activities… if I couldn’t figure out what I taught that day, how can I expect my students to learn?” (Minh)
Environment-related (n = 3)	Teaching hybrid (n = 1)	“It was extremely awkward … in hybrid teaching there are many things to check in the same time… I didn’t know much, like which screen to look at in the classroom.” (Kenji)
	Teaching online (n = 1)	“It was my first time teaching at university as well so it was kind of overwhelming… I didn’t really know how to teach online in general.” (Aiko)
	Transition back to F2F (n = 1)	“I was getting used to teaching online on Zoom… but from this semester I had to teach in person… and it was just, I don’t know. I was so used to teaching online it felt really different teaching in person with actual students in front of me… so that in itself was very overwhelming.” (Aiko)

Excerpts discussed below and in Table 2 illuminate the nature of ECLTs’ cognition as a complex and adaptive system and describe how said teachers make sense of and learn from their CIs through the process of self-organisation. A contribution of this study, then, is the revealing of these processes—something the CI literature has yet to explore—and the establishment of a novel research framework adept at capturing such interactions.

To help situate our analysis within the broader CI literature, we first organised the CIs into categories. Although we have argued earlier that the categorisation process risks oversimplifying CIs, in this study it functioned simply to identify the central “trigger” of each incident and to establish a starting point for examining the initial conditions from which they emerged, rather than as an analytical endpoint. Building on prior studies, we then move beyond categorisation to analyse the agentive and identity-related processes that unfolded in response, areas that have received comparatively less attention.

The CIs were categorised into three major groups based on their primary locus of influence: learner-related, materials-related, and environment-related incidents. As noted in other studies (Farrell, 2008), these categories are interrelated; for instance, environment changes (e.g., hybrid teaching) can exacerbate learner behaviours or comprehension difficulties. We determined both the major categories and subcategories according to how the focus of the CI was presented during interviews, and in cases of ambiguity we confirmed with participants how they felt the cause of the CI should be framed.

Consistent with prior studies, learner-related CIs were the most common category in our data (e.g., Karimi & Nazari, 2021). In Farrell’s (2008) study, the most frequently reported CIs concerned learners’ language proficiency—a pattern that also emerged in our dataset. Five of our reported CIs (two related to task misunderstanding and three to perceived underperformance) were linked to learners’ limited English proficiency. This is likely because many participants taught introductory-level university students in required English language classes or worked with beginner-level young learners. Moreover, most of our participants taught entirely in the learners’ L2, either because they did not share a common L1 or because of institutional language policies prohibiting L1 use.

Learner-related CIs (especially those involving language) may be particularly salient for ECLTs because they often enter classrooms without realistic expectations of what limited-proficiency learners can accomplish and without the strategies to adapt quickly, such as engaging in teacher talk or accepting repeated clarification requests as routine (Farrell, 2008). This concern was particularly true for our ECLTs, many of whom were studying TESOL in a non-English speaking country (i.e., Japan) while themselves being L2 English speakers from a third country (e.g., Vietnam). As a result, unlike a local teacher, they were teaching without knowledge of students’ cultural background or L1, and were doing so through their L2 or L3. For TESOL programs in non-English-speaking countries (like Japan) that recruit international students from third countries, this reinforces the need to prepare ECLTs for the layered socio-linguistic challenges of teaching in a local context that is both linguistically and culturally unfamiliar, using an additional language.

Triggering Stage

As the interview data underscore, a CI can act as a high-intensity threat, triggering the destabilisation of instructors’ professional and personal beliefs. For example, Erica’s “misunderstanding tasks” CI occurred when her Japanese students remained silent despite not understanding her instructions. This contrasted sharply with her prior EFL teaching experience in Vietnam, where her highly active and vocal students would immediately seek clarification, prompting her to realise that her expectations about student behaviour were shaped by assumptions from her earlier context. This reflective thinking is consistent with studies showing that teachers’ beliefs and practices are mediated by their prior teaching experience, which can cause dissonance when working in different linguistic or cultural contexts (Borg, 2011). This dissonance is important because, as we demonstrate later, if left unresolved it can shape how teachers move into subsequent developmental stages. Moreover, such dissonance has been associated with longer-term risk such as demotivation, isolation, or fossilisation (Derakhshan & Nazari, 2023), and in some cases attrition (Morales & Mena, 2017; Trent, 2017).

Similarly, conditions related to teachers’ prior learning (rather than teaching) experiences, especially their own cultural and educational backgrounds, led to CIs which involved conflict between their (often unconsciously held) beliefs and classroom practices. For example, Maria’s prior learning experience in Italy led to a surprising realisation that students in the United States did not share knowledge of the same cultural artefacts (i.e., the Donald Duck family—see Table 2) as students in her home country.

In addition, during the Covid-19 pandemic, Kenji was asked by his university at short notice to teach classes in hybrid format. He described himself in this situation as being extremely “awkward” (Table 2), uncomfortable deciding which group of students to focus on (those on screen or those in the physical classroom). This incident occurred immediately after he completed his MA in TESOL, which highlighted a contradiction between his own preconceived notions of teaching (e.g., a conventional classroom set up), nurtured by his prior learning in the graduate training course, and the pragmatic constraints encountered in real-world classroom settings. Maria’s and Kenji’s CIs show how teachers’ prior learning experiences shaped which classroom events were experienced as destabilising (Romano, 2006). What counts as “critical” is therefore highly individual and context-dependent (Atai & Nejadghanbar, 2016), reflecting the complex relationship between CIs and teacher cognition.

Initial conditions relating to the teaching environment and contextual factors also contributed to triggering CIs, underscoring the nested structure of cognition as embedded within multiple layers of other complex independent systems that inform the educational space (Fogal, 2023; Ludlow et al., 2017). For example, Roxanne recounted an incident where a student in her cram school class cried after receiving a low score on a vocabulary test, which, following the school rules, put them at risk of being expelled:

She screwed it up. So she got like 60, or 70, I don’t know. And she cried like so hard that she couldn’t even, you know, stay in class… So I can see if she gets a lower score, of course it’s her self-satisfaction, and everything like that. But also it’s about the parents, right? The parents wouldn’t like it if she gets a lower score, she would get scolded… They were stressed out because if they don’t get at least the average score, it was like, at least 70% of the test, then they cannot come to class anymore.

She later added that “most students get 100%, because their parents give them full support,” implying that the student did not receive full parental support. In this sense, the incident could be related to the learner’s psychological sense of self-fulfillment, parental expectations, or the harshness of the school policy. Roxanne’s incident shows that the “triggering” of CIs is not only spurred by initial conditions but occurs within layers of independent yet interrelated and equally complex systems: learners, parents, and institutional policies. This nested dynamic can be particularly challenging for ECLTs like Roxanne, who are still navigating the multilayered landscape and environment of their profession (Nall & Hiratsuka, 2023). This study contributes to documenting these interconnected and non-hierarchical structures.

Revealing these networks shows that the formation of CIs—and instructors’ relationship to them—is far more complex than previous studies tasked with categorisation would suggest (Atai & Nejadghanbar, 2016; Farrell, 2008). While categorisation helps to identify broad types of CIs, this approach does not capture the full complexity of how CIs unfold and how instructors engage with them. Our findings thus suggest that there is no straightforward, top-down progression where information flows from a single, central source. In other words, the triggering process unfolds within a boundless and interconnected system where different educational, psychological, social, and policy-related conditions converge in ways that are largely unpredictable (Kiss, 2012).

Linking Stage

RQ2 was concerned with how teachers agentively negotiated their CIs. The analysis in this section explores the strategies that teachers employed and shows how the process of “linking” these strategies was unpredictable, complex, and sensitive to initial conditions.

The findings highlight that initial conditions tied to teachers’ work environments supported or constrained their agentive ability to handle CIs in unpredictable ways. For example, interpersonal dynamics and a lack of supportive mentorship at some teachers’ institutions were barriers to handling the CIs. Aiko, for example, struggled with teaching online but was apprehensive about approaching her supervisor for help because of his stressful workload: “I just felt kind of awkward asking him [the coordinator] … because I knew that he was really busy, he was teaching a lot of classes.” Similarly, after Erica’s students all completed peer feedback incorrectly, she experienced a lack of collegiality and close interpersonal relationships at her institution, which prevented her from receiving sufficient support:

I tried to seek advice from a senior, but I didn’t build a good rapport with them. So I think it’s important for me to develop a sense of collegiality to get advice from different seniors, especially the experienced ones… if I can have a good relationship with them maybe they can spend some time with me and help me out.

In the absence of institutional support mechanisms, these early-career teachers’ exercised agency by drawing on their social networks and CoPs (Wenger, 1998) which became important conditions mediating the linking process. Aiko, for example, was involved in a web community for online language teaching because of a related research project (her initial condition): “Through trial and error, connecting with other teachers online through online communities…I managed to teach online for two years.” This CoP thus helped her navigate the online teaching environment.

Social exchanges through these pre-established CoPs also created feedback loops (Larsen-Freeman & Cameron, 2008) through which the teachers co-constructed meanings around their CIs, feeding “energy” back into the system and allowing them to cope (Rahman & Singh, 2021). For instance, Minh initially interpreted his “materials-related” CI as a general “lack of learning” by students. He struggled to comprehend their underperformance: “I know that there was something there. There was a problem. There was something I need to figure out, I need to fix… I need to change but the thing is I don’t know what to change.” Discussing his lessons with a community of teacher trainers in Vietnam helped him understand that students’ poor learning progress was partly due to his tendency to plan an unrealistic amount of teaching material. Minh’s membership in this community (his initial condition) helped to transform his interpretation of the CI from a broad, unconstructive framing around a fixed outcome (i.e., students not learning) to a more actionable response (i.e., reducing student workload). The important role of these social exchanges post-CI corroborates Nejadghanbar’s (2021) work, who found that group reflection could help “raise teachers’ awareness and understanding of teaching-learning processes and better prepare them to resolve some of the problems they may encounter” (p. 14). This type of socially-mediated reflection facilitates a “linking” of strategies to the CIs by creating a feedback loop, where new knowledge or insights can circle back to their origins, altering the initial understanding (Kiss, 2012; Nazari & De Costa, 2021).

The common thread in developing coping strategies is ECLTs’ ability to leverage CoPs to constructively reflect on and dialogically reinterpret CIs, leading to an informed classroom response (Cheng et al., 2023). Our findings corroborate other recent studies (e.g., Strong, 2023) which show that CoPs can support well-being, reflection, mentorship, and teachers’ ability to adapt post-CI (Nejadghanbar, 2021). Yet, as illustrated by Aiko and Erica’s difficulty accessing institutional colleagues when seeking support, membership and integration into these social spaces is not assured for all EFL practitioners, especially in their early years. This issue was especially evident for our participants in Japan, who were all working as part-time university instructors on fixed-term contracts. Japanese universities increasingly rely on part-time lecturers to teach EFL courses (Talbot & Mercer, 2018). Although such instructors frequently enjoy a high degree of autonomy, they are institutionally isolated due to a lack of permanent office space, access to opportunities for collaborative research, and inclusion in regular faculty meetings (Strong, 2023). This isolation severely limits their agentive capacities to draw on institutional or collegial support when navigating CIs.

Although teachers may be agentive in principle, how this agency can be exercised in practice is primarily shaped by initial conditions—largely beyond teachers’ direct control—which influence how they adapt following a CI. As our later findings will highlight, this leads to unpredictability regarding if or how they can best exercise their own agency. As working conditions for language teachers deteriorate globally (Kubanyiova & Crookes, 2016), rather than leaving early-career teachers’ adaptive capacities to chance, our data highlights that institutions must be responsible for establishing supportive communities through which ECLTs can reflect on and productively reframe challenging experiences (Gu & Day, 2013). Because CIs ultimately affect teachers’ career decisions (Hurst & Brantlinger, 2022), without such support systems educational institutions risk losing more ECLTs (Derakhshan & Nazari, 2023; Mason, 2017).

Realignment and Stabilisation Stages

RQ3 was concerned with how teachers’ beliefs and practices changed as a result of the CIs, and how these changes intersected with shifts in their teacher identities. These two stages are thus discussed together. As the findings below highlight, such changes were grounded in context (e.g., institutional affordances or constraints) and initial conditions (e.g., beliefs informed by teacher training), driving both teachers’ sense-making and agentive decision-making processes. Variations in these conditions produced different forms of agency and, consequently, distinct identity outcomes. Our findings show that the ECLTs navigated identity shifts that were either productive or restrictive, depending on how they enacted agency: Some reconstructed their professional selves in adaptive ways, while others held onto more rigid, limiting identity positions. Because we cannot detail the developmental trajectories of all participants, we focus on a selection of cases that best illustrate these two outcomes.

In one of Roxanne’s CIs, she described herself as “stressed out” due to her learners’ lack of knowledge retention, which suddenly prompted a moment of introspection:

I kept teaching the same thing and the students, like, still don’t get it. And I was like, okay, I told you about this last week or even 10 minutes ago. Why wouldn’t you understand that? And then one time I realized, okay, so I learned these, because I went to an international school for seven years. So I learned the same grammar seven years, like myself as well. So I just realized that, and from that moment, okay, so until I teach the same thing, like the same grammar seven times, I can’t get angry at them…. I started to think of it as a default. That I have to teach the same thing many times, because I was taught that way too.

Recognising parallels between her students’ struggles and her own learning experiences in an international school (her initial condition), she agentively shifted her perspective from a deficit view of students to an appreciation for iteration as a learning tool (Larsen-Freeman, 2018). Although this incident may seem minor, she described it as a “complete change of mindset,” becoming more patient with her students as she began to “understand them at a different level.” Crucially, this shift in her identity as a more empathetic teacher was not an inevitable outcome but a result of her deliberate assertion of agency in reframing the incident and consciously embracing iteration as part of the learning process. Her past experiences mediated this agency, allowing her to see aspects of her former self in her students.

In the second case, Aiko struggled with the uncertainty of teaching during the Covid-19 pandemic, becoming “overwhelmed with everything” as she navigated both online instruction and teaching in a university setting for the first time. Her stress was further compounded by the curriculum coordinator’s habit of sending class materials at the last minute:

He [the coordinator] always sent us the materials like two days before the actual day. And at first it was really stressful because I wasn’t really sure what we were teaching every week. … it ended up being like, oh, like I don’t know what I’m teaching, and then like two days before the class I get the materials and then I work over the weekend and then I teach on Monday morning.

Aiko reframed online teaching as a protective buffer that allowed her to manage this uncertainty. She recognised that in an online environment, she could “wing it” without the immediate pressures of an in-person classroom which she experienced in her earlier work at a secondary school (her initial condition):

When I make a mistake my hands get really like shaky, but yeah, if it’s online no one can see … If it was in person I would have been more stressed out and I probably would have complained, but because everything was online it was, I was able to like wing it.

At the same time, she agentively adapted her approach to student interactions by shifting communication to the LMS, which enabled her to prepare responses in advance and establish herself as more competent and authoritative:

I had time to like think about the answers or if it was like a technical question about like the course. So, then I would be able to ask another teacher if I couldn’t answer. But I think in person if I don’t know a lot of things about the course, students would not really respect you, you know, like, you assume that the teacher knows how the course is like conducted or like, or like the details of it.

Here, Aiko leveraged the CI in two ways: First, by reframing online teaching as a safeguard against the unpredictability of last-minute preparation, and second, by using asynchronous communication to manage her own vulnerability (Song, 2016) and present herself as a knowledgeable instructor. Rather than viewing the online setting as a hindrance, her agentive choices led her to take advantage of the CI as a bridge between her nervous, inexperienced self and the competent, authoritative teacher she aspired to become.

In contrast to these positive adaptations, Maria and Kenji faced CIs that led to stable yet maladaptive outcomes which emerged through narratives of restriction, shaped by how they exercised agency within the context of initial conditions such as personal beliefs and institutional constraints. For example, Maria’s prior belief in traditional teaching methods (her initial condition) became reinforced after an incident with an allergic student while playing a ball game in class. This caused her to become hyper-vigilant about inclusivity, reaffirming her core belief in safer alternatives such as textbooks. She noted that “you can make a class funner but if you always have to think about every single possibility, as a teacher that makes you less capable of doing a lot of different things.” Maria, then, viewed changes in her practice through a risk-reward lens: Although she acknowledged potential accommodations (e.g., wrapping the ball in cloth), she saw this as “bothersome” and eschewed doing interactive “fun stuff.” This perceived loss of freedom driven by inclusivity manifested into a habit of extreme risk avoidance to prevent other episodes from occurring in the future.

Similarly, Kenji, feeling restricted by his institution’s abrupt shift to hybrid teaching and a lack of support, resorted to “satisficing”—a strategy in which the solution is not ideal, but sufficient to achieve a learning outcome and accelerate the transition into a desired state (Maistre & Pare, 2010). His decision to focus only on the face-to-face students brought stability at the cost of his professional satisfaction as hybrid teaching prevented him from applying theoretical knowledge from university to the classroom (his initial condition). Failing to step into his “imaginary role” led him to feel a sense of disenfranchisement with the profession, describing it as one with “limited resources” and “a kind of job with restriction[s].”

What differentiates Roxanne and Aiko from Maria and Kenji is how their agency is both viewed and exercised. For Roxanne and Aiko, their agentive capacity transforms the CIs into positive learning curves and productive identities. In contrast, Maria and Kenji’s narratives of restriction show that agency manifested as a protective shield or simply doing “good enough” which provided stability but limited their professional growth (Hiver & Whitehead, 2018). Although the outcomes vary, these teachers’ agency is contextually situated and tempered by initial conditions, such as prior beliefs, membership in CoPs, and institutional power dynamics (Tsui, 2007; Wesely et al., 2021), which mediate the adaptation and realignment stages, leading to states of growth, resistance, and even regression. These findings substantiate Hurst and Brantlinger’s (2022) work, who showed that teachers’ reactions and emergent identities post-CI were not only influenced by structural conditions (e.g., the school environment) but also their individual agency (how they perceive and interpret events). Thus, how teachers’ cognitions self-organise and settle into adaptive or maladaptive states hinges significantly on “the meaning that teachers attach to their practices” (Feryok, 2010, p. 277), which, as this study shows, is sensitive to initial conditions.

As this and previous studies (Kubanyiova & Crookes, 2016; Miller et al., 2017) underscore, identity development is not merely about accumulating experience but involves ongoing, critical engagement with one’s professional self. As ECLTs navigate CIs, they need to make agentive, value-driven decisions about the kind of teachers they wish to become. Without institutional structures that encourage teachers to interrogate their evolving identities, CIs may push them into restrictive rather than expansive professional trajectories.

Attending to meaning-making processes should be a priority for researchers and educational leaders alike given that (a) ECLTs are more likely to encounter CIs that have a destabilising effect on their classroom practice and professional outlook—as this study underscores with Maria and Kenji—and (b) teachers’ interpretations of CIs can negatively impact their identity and career trajectory (e.g., Hurst & Brantlinger, 2022).

At a time when destabilising forces are prevalent (e.g., the lingering classroom effects of the Covid-19 pandemic and the introduction of AI and large language models in TESOL), attending to the CI-related concerns of ECLTs is crucial for teacher wellbeing and long-term professional development (Sulis et al., 2022). Our findings, then, provide institutions with principled reasons to carve out spaces where in-service, early-career instructors can engage with and process CIs (e.g., faculty workshops aimed at reflecting on agency and professional identity).

This work also demonstrates the utility of Hiver and Dörnyei’s (2017) stage-based framework for exploring CIs among ECLTs—the first study to explicitly do so. Given that the processes informing how ECLTs engage with CIs remain understudied (Derakhshan & Nazari, 2023), revealing these interactions can provide insights into what obstacles ECLTs face and what kind of support they need to succeed. Many of the initial conditions which constrained our participants’ agency were related to employment conditions in Japan (Strong, 2023). Similarly, this stage-based framework may prove useful for observing how shared initial conditions in other contexts influence changes in teachers’ identities and practices (e.g., a graduate TESOL program that transitions into a practicum model in the same school or school district) (see Ludlow et al., 2017). These stages allow researchers to pinpoint where breakdowns may occur. For example, the triggering stage may reveal where ECLTs are vulnerable; the linking stage may highlight where instructors lack the tools or affordances to deal with CIs; and the realignment and stabilisation stages provide insights into how CIs shape teachers’ thinking, practices, and identities, and whether these transformations foster positive growth or stagnation.

Moving forward, there are several limitations to this study which may inform future research. First, the sample size is relatively small. While the number of participants is consistent with related studies and allowed us to balance breadth and depth, we acknowledge that a larger sample could widen contextual understanding and speak to developmental nuances that this work was unable to detect. Second, this study relied only on qualitative interview data. However, system mapping (Fogal, 2022; Ludlow et al., 2017) could add a quantifiable element to research on CIs and teacher cognition by assigning a value to connections and components within the educational system. For example, the influence of initial conditions on teacher adaptation could be rated based on their perceived impact, thus allowing for comparisons across teachers and teaching contexts. This approach might also help identify leverage points that promote adaptive outcomes following CIs, or conversely, where further mechanisms and interventions might be needed to support co-adaptation and emergence.

Another limitation is the retrospective nature of participants’ accounts. We see this as both a strength and a weakness. We acknowledge that retrospective narrations may evoke some degree of temporal distortion. Still, we suggest that the passage of time provided participants with the benefit of hindsight and the opportunity to articulate how they made sense of their past experiences and integrated the CIs within the scope of their broader developmental trajectories over time, as evidenced in Feryok (2020). Along these lines, future studies could consider comparing how teachers’ post-CI narrated selves change across different points in time and space.

Conclusion and Implications for Teaching

This study examined how CIs influence ECLTs’ cognition and identity construction through the lens of CDST. Drawing on Hiver and Dörnyei’s (2017) language teacher immunity framework, we traced teachers’ responses across four key phases—triggering, linking, realignment, and stabilisation—illustrating how CIs are not merely disruptive events but developmental turning points within a nonlinear, agentive process of professional becoming.

Our findings show that ECLTs in the initial stages of shaping their professional identity must make high-stakes interpretive decisions when faced with CIs. These decisions and their outcomes hinge on agency—the capacity to make sense of, respond to, and act upon moments of disruption. However, agency does not operate in a vacuum. It is also mediated by initial conditions—teachers’ prior experiences, belief systems, imagined selves, as well as the structural affordances and limitations of their immediate teaching contexts.

While some teachers had access to mentors or collegial networks that helped them reframe CIs as opportunities for growth (e.g., Minh), others operated in relative isolation (e.g., Erika), navigating these incidents with limited external input. For many ECLTs, the absence of support required them to rely on their individual agency and the limited resources at their disposal to navigate CIs.

Our findings also reveal that ECLTs author the meanings of their CIs through the exercise of agency, constructing internalised narratives—adaptive or maladaptive—which are deeply shaped by their initial conditions. Teachers like Aiko and Roxanne, for example, constructed narratives that positioned their CIs as a catalyst for learning and productive identity realignment, whereas others like Maria and Kenji, interpreted them as restrictive, retreating into defensive or maladaptive states. Although prior work suggests that identity and agency are mutually constitutive (e.g., Cobb, 2022; Ruohotie-Lyhty, 2018), our analysis foregrounds the ways in which agency leads to identity shifts in the context of our participants’ early-career trajectories. From a CDST perspective, these identity shifts can be understood as emergent outcomes of teachers’ agentive sense-making in response to CIs—patterns that arose through the ongoing interaction of initial conditions such as their beliefs, personal histories, and contextual factors. Thus, this study contributes to a situated understanding of the identity-agency relationship in the early stages of language teachers’ careers.

Although ECLTs cannot change the beliefs and experiences they bring to a CI, we can change the conditions that influence how these are mobilised in response to unexpected events. In our data, teachers who were able to critically reframe CIs tended to have access to supportive networks that facilitated productive strategies for sense-making, a resource that can be intentionally cultivated in teacher-training programs. Given that identity-driven sense-making inquiry is typically lacking in teacher education programs (Taşdemir & Seferoğlu, 2024), ECLTs especially could benefit from structured opportunities to practise reframing real or simulated CIs. As Schutz et al. (2018) argue, intensive internship-like experiences can help teachers see how their methods play out in real classrooms, and “discover how it makes them feel and how it might inform their emergent identity” (p. 58). Training programs can strategically integrate such experiences with targeted reflection activities, such as guided CI debriefs where novice teachers revisit challenging moments from practicum or early teaching, articulate varied interpretations, and explore how alternative framings might lead to different identity-related outcomes. In doing so, we can better support the development of resilient, reflective professionals capable of navigating unexpected challenges with both agency and purpose.

About the Authors

Cameron Flinn is a doctoral researcher and lecturer in Japan, where he teaches English for Academic Purposes. His research interests include early-career language teachers, critical incidents, and English-medium instruction. ORCID ID: 0000-0001-5036-5575

Gary Fogal is a professor of Applied English Language and Linguistics at Sophia University, Japan. His primary research engages a complex dynamic systems theory account of L2 writing development. His recent publications appear in Journal of Second Language Writing, Language Awareness, International Journal of Complexity in Education and Applied Linguistics. ORCID ID: 0000-0001-6012-5188

To Cite this Article

Flinn, C., & Fogal, G. G. (2026). Early-career language teachers and critical incidents: Agency as a driver of identity construction. Teaching English as a Second Language Electronic (TESL-EJ), 30(1). https://doi.org/10.55593/ej.30117a2

References

Atai, M. R., & Nejadghanbar, H. (2016). Unpacking in-service EFL teachers’ critical incidents: The case of Iran. RELC Journal, 47(1), 97–110. https://doi.org/10.1177/0033688216631177

Babaii, E., Molana, K., & Nazari, M. (2022). Contributions of assessment-related critical incidents to language teacher identity development. Innovation in Language Learning and Teaching, 15(5), 442–457. https://doi.org/10.1080/17501229.2020.1824234

Badia, A., Becerril, L., & Gómez, M. (2021). Four types of teachers’ voices on critical incidents in teaching. Teacher Development, 25(2), 120–135. https://doi.org/10.1080/13664530.2021.1882549

Barkhuizen, G. (2016). Narrative approaches to exploring language, identity and power in language teacher education. RELC Journal, 47(1), 25–42. https://doi.org/10.1177/0033688216631222

Birks, M., & Mills, J. (2015). Grounded theory: A practical guide. Sage.

Borg, S. (2006). The distinctive characteristics of foreign language teachers. Language Teaching Research, 10(1), 3–31. https://doi.org/10.1191/1362168806lr182oa

Borg, S. (2011). The impact of in-service teacher education on language teachers’ beliefs. System, 39(3), 370–380. https://doi.org/10.1016/j.system.2011.07.009

Borg, S. (2019). Language teacher cognition: Perspectives and debates. In X. Gao (Ed.), Second handbook of English language teaching (pp. 1149–1170). Springer.

Brookfield, S. D. (1990). The skillful teacher. Jossey-Bass.

Burns, A., Freeman, D., & Edwards, E. (2015). Theorizing and studying the language-teaching mind: Mapping research on language teacher cognition. The Modern Language Journal, 99(3), 585–601. https://doi.org/10.1111/modl.12245

Carlson, J. A. (2010). Avoiding traps in member checking. The Qualitative Report, 15(5), 1102–1113.

Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis. Sage.

Cheng, M. M. H., Tang, S. Y. F., Wong, A. K. Y., & Yeh, F.-Y. (2023). How early career teachers overcome everyday work-related challenges: An analysis from the perspective of teacher buoyancy. Journal of Education for Teaching, 49(5), 753–767. https://doi.org/10.1080/02607476.2022.2150962

Cobb, D. J. (2022). Metaphorically drawing the transition into teaching: What early career teachers reveal about identity, resilience and agency. Teaching and Teacher Education, 110, 103598. https://doi.org/10.1016/j.tate.2021.103598

Creswell, J. W. (2013). Qualitative inquiry and research design: Choosing among five approaches (3rd ed.). Sage.

Davis, W. S., & Borden, R. S. (2025). Examining the resilience of two early career world language teachers: A two-year case study. Language, Culture and Curriculum, 38(1), 1–18. https://doi.org/10.1080/07908318.2024.2369581

Day, C. (2017). Teachers’ worlds and work. Routledge. https://doi.org/10.4324/9781315170091

Derakhshan, A., & Nazari, M. (2023). “I am fed up with the criticisms”: Examining the role of emotional critical incidents in a novice teacher’s identity construction. The Asia-Pacific Education Researcher, 32, 449–458. https://doi.org/10.1007/s40299-022-00666-1

Edwards, E., & Burns, A. (2016). Language teacher-researcher identity negotiation: An ecological perspective. TESOL Quarterly, 50(3), 735–745. https://doi.org/10.1002/tesq.313

Fairbrother, L. (2022). Interests and power in English education policy in Japan: A focus on the high school ‘Teaching/learning English in English’ policy.’ In M. Nekula, T. Sherman, & H. Zawiszová (Eds.), Interests and power in language management (pp. 237–265). Peter Lang.

Farrell, T. S. C. (2008). Critical incidents in ELT initial teacher training. ELT Journal, 62, 3–10. https://doi.org/10.1093/elt/ccm072

Farrell, T. S. C. (2009). The novice teacher experience. In A. Burns & J. C. Richards (Eds.), The Cambridge guide to second language teacher education (pp. 182–189). Cambridge University Press. https://doi.org/10.1017/9781139042710.024

Farrell, T. S. C. (2012). Novice-service language teacher development: Bridging the gap between preservice and in-service education and development. TESOL Quarterly, 46, 435–449. https://doi.org/10.1002/tesq.36

Farrell, T. S. C. (2013). Critical incident analysis through narrative reflective practice: A case study. Iranian Journal of Language Teaching Research, 1(1), 79–89.

Feryok, A. (2010). Language teacher cognitions: Complex dynamic systems? System, 38(2), 272–279. https://doi.org/10.1016/j.system.2010.02.001

Feryok, A. (2020). Perezhivanie and experiencing in language teacher development. European Journal of Applied Linguistics & TEFL, 9(2), 1–25.

Finch, A. (2010). Critical incidents and language learning: Sensitivity to initial conditions. System, 38(3), 422–431. https://doi.org/10.1016/j.system.2010.05.004

Fogal, G. G. (2022). System mapping simplex spaces: Facilitating change in L2 educational contexts from a complexity theory perspective. International Review of Applied Linguistics in Language Teaching, 60(1), 103–121. https://doi.org/10.1515/iral-2021-0027

Fogal, G. G. (2023). Educational landscapes through a complexity theory lens: Using system mapping to investigate L2 Shakespeare studies in secondary schools. International Journal of Complexity in Education, 4(1), 115–138. https://doi.org/10.26262/IJCE.V4I1.9481

Foryś-Nogala, M., Krajewski, G., & Haman, E. (2022). The interface issue in second language acquisition research: An interdisciplinary perspective. Lingua, 271, 103243. https://doi.org/10.1016/j.lingua.2022.103243

Fu, M., & Li, S. (2021). The associations between implicit and explicit language aptitude and the effects of the timing of corrective feedback. Studies in Second Language Acquisition, 43(3), 498–522. https://doi.org/10.1017/S0272263121000012

Glaser, B., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. Sociology Press. https://doi.org/10.1097/00006199-196807000-00014

Golombek, P. R. (1998). A study of language teachers’ personal practical knowledge. TESOL Quarterly, 32, 447–464. https://doi.org/10.2307/3588117

Golombek, P. R., & Johnson, K. E. (2004). Narrative inquiry as a mediational space: Examining emotional and cognitive dissonance in second-language teachers’ development. Teachers and Teaching: Theory and Practice, 10(3), 307–327. https://doi.org/10.1080/1354060042000204388

Gu, M., & Benson, P. (2015). The formation of English teacher identities: A cross-cultural investigation. Language Teaching Research, 19(2), 187–206. https://doi.org/10.1177/1362168814541725

Gu, Q., & Day, C. (2013). Challenges to teacher resilience: Conditions count. British Educational Research Journal, 39(1), 22–44. https://doi.org/10.1080/01411926.2011.623152

Halquist, D., & Musanti, S. I. (2010). Critical incidents and reflection: Turning points that challenge the researcher and create opportunities for knowing. International Journal of Qualitative Studies in Education, 23(4), 449–461. https://doi.org/10.1080/09518398.2010.492811

Hiver, P. (2022). Complexity theory as a conceptual framework for language teacher research. In Y. Kimura, L. Yang, T.-Y. Kim, & Y. Nakata (Eds.), Language teacher motivation, autonomy and development in East Asia (pp. 15–33). Springer. https://doi.org/10.1007/978-3-030-93467-5_2

Hiver, P., & Al-Hoorie, A. H. (2016). A dynamic ensemble for second language research: Putting complexity theory into practice. The Modern Language Journal, 100(4), 741–756. https://doi.org/10.1111/modl.12347

Hiver, P., & Dörnyei, Z. (2017). Language teacher immunity: A double-edged sword. Applied Linguistics, 38(3), 405–423.

Hiver, P., & Whitehead, G. E. K. (2018). Sites of struggle: Classroom practice and the complex dynamic entanglement of language teacher agency and identity. System, 79, 70–80. https://doi.org/10.1016/j.system.2018.04.015

Hurst, C., & Brantlinger, A. (2022). Patterns in critical incidents: Understanding teacher retention through career decision making. Teaching and Teacher Education, 109, 103557. https://doi.org/10.1016/j.tate.2021.103557

Karimi, M. N., & Nazari, M. (2021). Examining L2 teachers’ critical incidents: A complexity theoretic perspective. Innovation in Language Learning and Teaching, 15(1), 81–98. https://doi.org/10.1080/17501229.2019.1676755

Kayi-Aydar, H. (2015). Teacher agency, positioning, and English language learners: Voices of pre-service classroom teachers. Teaching and Teacher Education, 45, 94–103. https://doi.org/10.1016/j.tate.2014.09.009

Kidwell, T. (2025). “I shouldn’t have to fumble my way around”: An investigation of the current state of TESOL teacher trainers’ professional development in the United States. TESOL Journal, 16(1), e907. https://doi.org/10.1002/tesj.907

Kılıç, M., & Cinkara, E. (2020). Critical incidents in pre-service EFL teachers’ identity construction process. Asia Pacific Journal of Education, 40(2), 182–196. https://doi.org/10.1080/02188791.2019.1705759

Kiss, T. (2012). The complexity of teacher learning: Reflection as a complex dynamic system. Journal of Interdisciplinary Research in Education, 2(1), 17–35.

Kubanyiova, M., & Crookes, G. (2016). Re-envisioning the roles, tasks, and contributions of language teachers in the multilingual era of language education research and practice. The Modern Language Journal, 100(S1), 117–132. https://doi.org/10.1111/modl.12304

Kubota, R. (2016). Critical language teacher identity. In G. Barkhuizen (Ed.), Reflections on language teacher identity research (pp. 218–222). Routledge. https://doi.org/10.4324/9781315643465-37

Larsen-Freeman, D. (2018). Task repetition or task iteration?: It does make a difference. In M. Bygate (Ed.), Learning language through task repetition (pp. 311–329). John Benjamins. https://doi.org/10.1075/tblt.11.12lar

Larsen-Freeman, D., & Cameron, L. (2008). Complex systems and applied linguistics. Oxford University Press.

Le Cornu, R. (2013). Building early career teacher resilience: The role of relationships. Australian Journal of Teacher Education, 38(4), 1–16. https://doi.org/10.14221/ajte.2013v38n4.4

Lengeling, M. M., & Pablo, I. M. (2016). Reflections on critical incidents of EFL teachers during career entry in Central Mexico. HOW Journal, 23(2), 75–88. https://doi.org/10.19183/how.23.2.158

Li, L. (2013). The complexity of language teachers’ beliefs and practice: One EFL teacher’s theories. The Language Learning Journal, 41(2), 175–191. https://doi.org/10.1080/09571736.2013.790132

Li, L. (2020). Language teacher cognition: A sociocultural perspective. Springer. https://doi.org/10.1007/978-981-15-0321-4

Ludlow, L., Ell, F., Cochran-Smith, M., Newton, A., Trefcer, K., Klein, K., Grudnoff, L., Haigh, M., & Hill, M. F. (2017). Visualizing teacher education as a complex system: A nested simplex system approach. Complicity: An International Journal of Complexity and Education, 14(1), 36–79. https://doi.org/10.29173/cmplct26053

Maistre, C. L., & Pare, A. (2010). Whatever it takes: How beginning teachers learn to survive. Teaching and Teacher Education, 26, 559–564. https://doi.org/10.1016/j.tate.2009.06.016

Mason, S. (2017). Foreign language teacher attrition and retention research: A meta-analysis. NECTFL Review, 80, 47–68. https://www.nectfl.org/wp-content/uploads/2017/09/NECTFL-Review-September-2017.pdf

McGarr, O., & McCormack, O. (2016). Counterfactual mutation of critical classroom incidents: Implications for reflective practice in initial teacher education. European Journal of Teacher Education, 39(1), 36–52. https://doi.org/10.1080/02619768.2015.1066329

Megawati, F., Mukminatien, N., Anugerahwati, M., Indrayani, N., & Unsiah, F. (2020). Critical incidents: Exploring EFL prospective teachers’ teaching experiences. European Journal of Educational Research, 9(3), 943–954. https://doi.org/10.12973/eu-jer.9.3.943

MEXT. (2014). Guidelines for responding to misconduct in research. https://www.mext.go.jp/a_menu/jinzai/fusei/__icsFiles/afieldfile/2015/07/13/1359618_01.pdf

Miller, E. R., Morgan, B., & Medina, A. L. (2017). Exploring language teacher identity work as ethical self-formation. The Modern Language Journal, 101(S1), 91–105. https://doi.org/10.1111/modl.12371

Morales, H. S., & Mena, R. G. (2017). Critical incidents, reflective writing, and future teachers’ professional identities. Revista de Lenguas Modernas, 26, 233–255. https://repositorio.una.ac.cr/bitstreams/a58083fc-c557-4dcd-b1a4-8f9b08dd21c5/download

Musanti, S. I., & Pence, L. (2010). Collaboration and teacher development: Unpacking resistance, constructing knowledge, and navigating identities. Teacher Education Quarterly, 37(1), 73–89.

Nall, M., & Hiratsuka, T. (2023). “Indulge in a little chaos”: Assistant language teachers’ cognition and pedagogical knowledge development through the lens of complex dynamic system theory. Teaching English as a Second Language Electronic Journal (TESL-EJ), 27(2). https://doi.org/10.55593/ej.27106s3

Nazari, M., & De Costa, P. I. (2021). Contributions of a professional development course to language teacher identity development: Critical incidents in focus. Journal of Teacher Education, 73(4), 366–380. https://doi.org/10.1177/00224871211059160

Nazari, M., De Costa, P. I., & Karimpour, S. (2023). The role of institutional policy in English language teacher autonomy, agency, and identity: A poststructural perspective. Language Teaching Research, 30(2), 885–907. https://doi.org/10.1177/13621688221143476

Nejadghanbar, H. (2021). Using critical incidents for professional development with Iranian novice English teachers. TESOL Journal, 12(1), e00521. https://doi.org/10.1002/tesj.521

Norton, B. (2013). Identity and language learning: Extending the conversation. Multilingual Matters. https://doi.org/10.21832/9781783090563

Norton, B., & De Costa, P. I. (2018). Research tasks on identity in language learning and teaching. Language Teaching, 51(1), 90–112. https://doi.org/10.1017/S0261444817000325

Rahman, M. M., & Singh, M. K. M. (2021). Complex dynamic systems of language teacher cognitions: A case study from Bangladesh. Issues in Educational Research, 31(1), 241–254.

Romano, M. E. (2006). ‘‘Bumpy moments’’ in teaching: Reflections from practicing teachers. Teaching and Teacher Education, 22, 973–985. https://doi.org/10.1016/j.tate.2006.04.019

Ruohotie-Lyhty, M. (2018). Identity-agency in progress: Teachers authoring their identities. In P. A. Schutz, J. Hong, & D. Cross Francis (Eds.), Research on teacher identity (pp. 25–36). Springer. https://doi.org/10.1007/978-3-319-93836-3_3

Schön, D. A. (1987). Educating the reflective practitioner: Toward a new design for teaching and learning in the professions. Jossey-Bass.

Schutz, P. A., Nichols, S. L., & Schwenke, S. (2018). Critical events, emotional episodes, and teacher attributions in the development of teacher identities. In P. A. Schutz, J. Hong, & D. C. Francis (Eds.), Research on teacher identity: Mapping challenges and innovations (pp. 49–60). Springer. https://doi.org/10.1007/978-3-319-93836-3_5

Serafini, E. J., & Roca-Ramirez, S. I. (2024). The role of critical experiences, positioning, and agency in the dynamic, emergent construction of heritage speaker selves. Studies in Second Language Learning and Teaching, 14(1), 97–120. https://doi.org/10.14746/ssllt.42373

Shapira-Lishchinsky, O. (2011). Teachers’ critical incidents: Ethical dilemmas in teaching practice. Teaching and Teacher Education, 27, 648–656. https://doi.org/10.1016/j.tate.2010.11.003

Sigstad, H. M. H., & Garrels, V. (2018). Facilitating qualitative research interviews for respondents with intellectual disability. European Journal of Special Needs Education, 33(5), 692–706.

Sisson, J. H. (2016). The significance of critical incidents and voice to identity and agency. Teachers and Teaching, 22, 670–682. https://doi.org/10.1080/13540602.2016.1158956

Song, J. (2016). Emotions and language teacher identity: Conflicts, vulnerability, and transformation. TESOL Quarterly, 50(3), 631–654. https://doi.org/10.1002/tesq.312

Strauss, A., & Corbin, J. (1990). Basics of qualitative research. Sage.

Strong, G. (2023). Supporting part-time ELT faculty in a Japanese university. ELT Journal, 77(2), 241–244. https://doi.org/10.1093/elt/ccac008

Sulis, G., Babic, S., Mairitsch, A., Mercer, S., Jin, J., & King, J. (2022). Retention and attrition in early-career foreign language teachers in Austria and the United Kingdom. The Modern Language Journal, 106(1), 155–171. https://doi.org/10.1111/modl.12765

Sulis, G., Mercer, S., Babic, S., & Mairitsch, A. (2023). Language teacher wellbeing across the career span. Multilingual Matters. https://doi.org/10.21832/9781800412811

Talbot, K., & Mercer, S. (2018). Exploring university ESL/EFL teachers’ emotional well-being and emotional regulation in the United States, Japan and Austria. Chinese Journal of Applied Linguistics, 41(4), 410–432. https://doi.org/10.1515/cjal-2018-0031

Taşdemir, H., & Seferoğlu, G. (2024). Language teacher identities and identity tensions of early career English teachers. TESOL Journal, 15(3), e794. https://doi.org/10.1002/tesj.794

Thornberg, R. (2012). Informed grounded theory. Scandinavian Journal of Educational Research, 56(3), 243–259. https://doi.org/10.1080/00313831.2011.581686

Trent, J. (2017). Discourse, agency and teacher attrition: Exploring stories to leave by amongst former early career English language teachers in Hong Kong. Research Papers in Education, 32(1), 84–105. https://doi.org/10.1080/02671522.2016.1144215

Tripp, D. (1993). Critical incidents in teaching: Developing professional judgement. Routledge.

Tsui, A. B. M. (2007). Complexities of identity formation: A narrative inquiry of an EFL teacher. TESOL Quarterly, 41(4), 657–680. https://doi.org/10.1002/j.1545-7249.2007.tb00098.x

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge University Press. https://doi.org/10.1017/CBO9780511803932

Wesely, P. M., Vyn, R., & Neubauer, D. (2021). Teacher beliefs about instructional approaches: Interrogating the notion of teaching methods. Language Teaching Research, 28(1), 29–51. https://doi.org/10.1177/1362168821992180

Williams, J., & Grierson, A. (2016). Facilitating professional development during international practicum: Understanding our work as teacher educators through critical incidents. Studying Teacher Education, 12(1), 55–69. https://doi.org/10.1080/17425964.2016.1143812

Yazan, B. (2018). A conceptual framework to understand language teacher identities. Journal of Second Language Teacher Education, 1(1), 21–48. https://doi.org/10.1558/slte.24908

Yu, S., Xu, H., Jiang, L., & Chan, I. K. I. (2020). Understanding Macau novice secondary teachers’ beliefs and practices of EFL writing instruction: A complexity theory perspective. Journal of Second Language Writing, 48, 100728. https://doi.org/10.1016/j.jslw.2020.100728

Yuan, R., & Wang, K. (2025). “Passing the torch” to language teachers: A transformative, action-oriented perspective on language teacher identity research. Journal of Language, Identity and Education, 24(2), 483–489.

Zheng, H. (2013). Teachers’ beliefs and practices: A dynamic and complex relationship. Asia-Pacific Journal of Teacher Education, 41(3), 331–343. https://doi.org/10.1080/1359866X.2013.809051

Copyright of articles rests with the authors. Please cite TESL-EJ appropriately. Editor’s Note: The HTML version contains no page numbers. Please use the PDF version of this article for citations.