263 Studies in Second Language Learning and Teaching Department of English Studies, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz University, Kalisz SSLLT 9 (2). 2019. 263-285 http://dx.doi.org/10.14746/ssllt.2019.9.2.2 http://pressto.amu.edu.pl/index.php/ssllt A drama of selves: Investigating teacher identity development from dialogical and complexity perspectives Alastair Henry University West, Trollhättan, Sweden https://orcid.org/0000-0001-7789-9032 alastair.henry@hv.se Abstract Identity is of increasing interest in teacher education. Crucial for resilience, the development of a coherent professional identity has been characterized as emerging from tensions between multiple and sometimes conflicting concep- tions of what it means to be someone who teaches (Akkerman & Meijer, 2011). While light is being shed on these often antagonistic relations, less is known about the dynamics of identity formation and transformation. Providing a con- tribution to work on language teacher identity, in this single case study Hermans’ (2008) concept of the dialogical self is combined with complexity principles in an investigation of changes in the emerging professional identity of a pre-service Eng- lish teacher during a practicum. Drawing on intra- and inter-personal data, expe- riences of learning to become a person who teaches English are conceptualized as a drama that is played out between different and sometimes unaligned selves. Analyses show how this inner drama maps onto the landscape of an emerging teacher identity, how tensions can be understood systemically, and how a teacher identity system can have a signature dynamic. Keywords: language teacher identity; complex dynamic systems; dialogical self; practicum learning Alastair Henry 264 1. Introduction In learning to teach, identity construction takes place in the midst of complex systems of different relationships. It is a process characterized by tension and struggle (Alsup, 2006; Beauchamp & Thomas, 2009; Beijaard, Meijer, & Verloop, 2004). The development of a professional identity begins in the earliest stages of preservice education (Geijsel & Meijers, 2005; Walkington, 2005). Through educational experiences in formal studies and practicum learning, students “de- velop a more sophisticated understanding of their work as teachers” (Suther- land, Howard, & Markauskaite, 2010, p. 456). Often, the practicum is the most influential of all aspects of a preservice teacher’s education. It can also be the most difficult, demanding and frustrating (Ferrier-Kerr, 2009; Trent, 2013). No- where in a teacher’s preservice education is the negotiation of shifting relation- ships as emotionally demanding as in early periods of classroom-based learning: Moving from a university community of student teachers to the community of a school as a new teacher implies multiple tensions as adaptations and adjustments to identity are necessitated or provoked. We understand this period as an intense identity experi- ence, a time when the new school context causes a beginning teacher to question and perhaps reframe her developing identity. (Beauchamp & Thomas, 2011, p. 6) In the practicum, multiple systems of relations exist in overlapping and often conflic- tive constellations. It is through the negotiation of relationships with university faculty, classroom mentors, pupils and preservice teaching colleagues that preservice teachers learn the work of teaching and begin to develop professional identities (Smagorinsky, Cook, Moore, Jackson, & Fry, 2004). These multiple and complexly intersecting rela- tions create tensions that can make identity formation a highly challenging process. It is against this backdrop that the current investigation of the process of the formation of a professional identity during a practicum is carried out. Con- ceptualizing teacher identity as a dialogical accomplishment where develop- ment occurs as a consequence of shifts between different identity positions that map onto the topography of a teacher identity system, the study seeks to show how identity tensions can be understood systemically. It shows how a coherent language teacher identity can be understood as emerging from conflicting per- ceptions of what it means to be a teacher, and how a teacher identity system can have a particular signature dynamic. 2. A dialogical perspective The development of a coherent professional identity can be understood as a process of struggle, and a constant “search for meaning” (Britzman, 2006, p. ix). A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 265 Teacher identities are multi-faceted and dynamic (Beijaard, Meijer, & Verloop, 2004). Under continual construction and reconstruction, development takes place during interactions with others. Identity development is thus both an in- dividual and a social process (Akkerman & Meijer, 2011; Rodgers & Scott, 2008). While the struggles that take place between competing and sometimes irrec- oncilable identities are well-recognised, most studies on teacher identity develop- ment rely on retrospective self-report data and investigate identity dynamics over longer timescales (often semesters or entire programs). Far fewer focus on devel- opment across shorter timescales or are structured around data collection points that are closely-spaced. Even fewer investigate teacher identity development using dialogical data or attempt to map “the dynamics of a teachers’ identity by describ- ing in more detail the self-dialogue that is pursued by a teacher in striving to main- tain a coherent and consistent sense of self” (Akkerman & Meijer, 2011, p. 316). Grounded in the recognition that teaching identities are constructed within relationships, and that development involves tensions between identities that can be conflictive, a dialogical approach offers a useful way of conceptual- izing teacher identity development (Akkerman & Meijer, 2011). In research that frames teacher identities as dialogical accomplishments, the theory of the dia- logical self (Hermans, 1996, 2001, 2003) provides a valuable framework for map- ping development. This is because it offers a systemic approach to the concep- tualization of what it means to be someone who teaches (Akkerman & Meijer, 2011). Like other theories with a focus on self-concept development, dialogical self theory emphasizes the complexity and multifacetedness of the self. Drawing on James’ (1890) conceptualization of the self as comprising both the knower and the known, and Bakhtin’s (1981) notion of the polyphonic novel where multiple voices offer sometimes complementary and sometimes conflict- ing narratives, Hermans (2008) conceptualizes the inner world of the individual as a cast of characters who interact within a mesh of dialogical relations. Each of these characters has a “voice,” which is also related to the voices of other charac- ters. As in a play, where exchanges take place between the people on stage, in the dialogical self interactions occur between “voiced positions” (Hermans, 2008). Emerging from these interactions is a complex and narratively structured self. As situations change, the drama of the self unfolds in ways where shifts take place in the relative dominance of the voices that speak. Thus, at any particular moment, some voices/positions will be dominant, while others are subdued (Hermans, 2008). It is in this way, and through the facility of imagination, that a person can act agentically as if they were another person, or as if they found themselves in another setting (Hermans, Kempen, & Van Loon, 1992). With its emphasis on space and the framing of identity as a landscape where different aspects of the self are related to one another like points on a Alastair Henry 266 map, Hermans’ (2008) theory of the dialogical self provides a way of dividing the self into functional sub-parts. Each of these parts – “I-positions” in Hermans’ theorizing – represents a particular aspect of identity. Systemically, the self can be understood as an ever-shifting conglomeration of potentially autonomous I- positions which, at any point in time, can re-locate to a different part of the sys- tem’s space-time field (Valsiner, 2004). Following this conceptualization, the process of developing a teacher identity can be understood as one that involves interactions between I-positions within the dialogical landscape of “being some- one who teaches” (Akkerman & Meijer, 2011, p. 315). Often, this space is a con- tested one, identities emerging through the constant need to make sense of di- verging conceptions of what it means to be a teacher. Identity development is thus a product of negotiation. It is a product of attempts to interrelate I-posi- tions in ways that can lead to a sense of self that is more or less coherent, and which can be sustained in the everyday work that takes place in classrooms. 3. A complexity perspective The development of a teacher identity is a dynamic process (Henry, 2016; Kaplan & Garner, 2018). It involves “the formation and restructuring of rela- tions” within and between identities and takes place “through intra- and inter- personal processes” (Kaplan & Garner, 2017, p. 2036). The investigation of de- velopmental processes involves pursuing the questions of why, when and how identities emerge. Because complexity principles are focused on the mecha- nisms of development, they offer a means through which these central ques- tions can be addressed. By focusing on the conditions under which identities evolve, a complexity approach can provide a framework for conceptualizing the integrative nature of teacher identity (Kaplan & Garner, 2017, 2018). From a complexity perspective, the dialogical self can be understood as a bounded yet open system where, in the ongoing drama of self-definition, interac- tions between I-positions are situationally influenced. To investigate and make sense of these conflictive relations, and to understand the dynamics of identity de- velopment, complexity theories offer a range of conceptual tools (Bell & Das, 2011). Although traditionally connected with the natural sciences, complexity theories are now firmly established in the social sciences (Byrne & Callaghan, 2014). Complexity theories have made particular inroads in research into second language develop- ment and, more recently, second language education (Larsen-Freeman, 2017; Or- tega & Han, 2017). They provide conceptualizations of change that enable phenom- ena relating to language development to be viewed dynamically, and principles that can be applied in the study of these processes (Hiver & Al-Hoorie, 2016; Kostoulas, Stelma, Mercer, Cameron, & Dawson, 2017; Larsen-Freeman & Cameron, 2008). A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 267 The dynamics of a complex system can be understood in terms of the sys- tem’s movement across its state space. The system’s state space comprises the total number of possible states that a system can occupy. Regions representing enduring periods of stability are conceptualized as attractor states. Regions rep- resenting (temporary) instabilities are repeller states. Transitions between attrac- tor states can arise as a result of perturbations (very specific, and often very sud- den changes in the context). They can also be triggered through changes in the system’s control parameters (dimensions that constitute the state space and which have a “controlling” function). Transitions to a new area of the state space involve a phase shift. New system behaviors are understood as emergent properties. These are aspects of functioning not existing previously which originate in a spon- taneous manner from the system’s own internal interactions and self-organiza- tion (Byrne & Callaghan, 2014; Larsen-Freeman & Cameron, 2008). Using these dynamical conceptualizations, research conducted from a com- plexity perspective has the aim of representing particular systems at particular scales of description, identifying dynamical patterns of change and their emergent outcomes, and modelling the mechanisms that give rise to change (Hiver & Al- Hoorie, 2016). Because the aim of complexity-informed research is to develop un- derstanding of the parameters that influence a system, and in so doing to identify opportunities for effecting positive changes, a complex system is often described in terms of distinct dynamical characteristics. These are the system’s signature dynamics (Byrne & Callaghan, 2014; Dörnyei, 2014; van Geert & Lichtwarck- Aschoff, 2005). Identifying a system’s signature dynamics involves mapping the system’s state space, identifying the attractor (and repeller) states, and plotting the movement between these states. It involves developing an understanding of the types of movement that take place, the events that prefigure such changes, and, at the system level, the nature and the effects of changed behaviors. 4. Study and purpose In a previous study in which a dialogical model of identity was combined with a complexity perspective, Henry (2016) examined the identity dynamics of a pre- service English teacher undertaking her first extended practicum. Drawing on data from the same project, a similar approach is taken in the current study. In this previous study, the participant found herself consistently questioning the practices of her mentor and the other teachers at the school and, as a conse- quence, also her own motives for wanting to be a teacher. For this student, the struggles she experienced in developing a coherent teacher identity were char- acterized by ambivalence. With tensions between two competing I-positions in Alastair Henry 268 her teacher identity system remaining unresolved, at the end of the practicum she was uncertain about whether she wanted to continue her education. While struggles that take place during the practicum can lead to doubt and career reorientations (Bloomfield, 2010), they can also be productive and lead to emerging identities that are oriented to future practice (Trent, 2011). In contrast to the participant in this previous study, the participant in the current study found herself in an environment that supported her professional develop- ment. Here, in mapping a different set of identity dynamics, and focusing on the ways in which a coherent language teacher identity can emerge within the “in- side world” of a student teacher’s imagination (Hermans, 2001, p. 249), the study seeks to show how identity tensions can be understood systemically and how a teacher identity system can have a particular signature dynamic. 5. Method 5.1. Design Studies in which complexity principles are used to investigate identity develop- ment need to focus on (i) an identity system, (ii) the contexts in which it is em- bedded, and (iii) changes that take place over time (Kaplan & Garner, 2017). Case studies of identity systems will therefore often involve the triangulation of data that is derived from multiple sources. Following the design described by Klimstra et al. (2010), analyses with a focus on identity dynamics occurring across shorter timescales (specific events in the classroom) were combined with analyses designed to reveal evolutionary patterns over longer timescales (across the practicum as a whole). 5.2 The participant The participant was a preservice English teacher undertaking a 4-week practi- cum at a school in the western part of Sweden. She was the study partner of the participant in a previous single case study on teacher identity dynamics (Henry, 2016). The participant and her colleague were selected to take part in the re- search on the basis of three considerations. First, when carrying out teaching practice, concerns relating to self-disclo- sure and the handling of sensitive topics and events can have constraining ef- fects on a student’s engagement in “in-public” activities, such as posting contri- butions on online forums (Chu, Kwan, & Warning, 2012; Deng & Yuen, 2013). These two students had established a trusting relationship. This meant that they were comfortable sharing experiences on an externally monitored forum. Second, in A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 269 studies using complexity methodologies, data rich in detail needs to be collected across points in time that are closely spaced (Kaplan & Garner, 2017). For this reason it was necessary to select students who viewed the practicum as an im- portant space for developing a professional identity and who were willing to discuss their day-to-day experiences. Both students had demonstrated ambition and conscientiousness during their education and were selected for this reason. Finally, given the purpose of investigating the dynamics of identity develop- ment, it was important to select participants who, to some extent, displayed uncertainty about the embarked-upon career. Both students viewed the practi- cum as a testing ground for their career choice and hoped to gain an early indi- cation that teaching was right for them. 5.3. The data Intra- and inter-personal data (Jasper, Moore, Whittaker, & Gillespie, 2012) were generated. The intra-personal data comes from two semi-structured interviews conducted by the author immediately before, and immediately after the practi- cum period. The inter-personal data comes from two sources: (i) postings made by the participant on an online forum (n = 22), where she and the other student maintained a conversation during the practicum, and (ii) a 40-minute, video-rec- orded, stimulated recall discussion between these two students immediately sub- sequent to a lesson that the participant had carried out. This combination of intra- and inter-personal data offers a window into the participant’s processes of sense- making, and her authoring of a dialogical self. While the intra-personal data can provide insights into beliefs and cognitions, the inter-personal data can cast light on the internal dialogue between voiced positions in the dialogical landscape of becoming someone who teaches. All of the data generated was in English. 5.4. Analytical procedures A discourse analytical approach was used (Coyle, 2006; Potter & Wetherell, 1995). This included a strategy specially developed for investigating inner dy- namics within the dialogical self (Bell & Das, 2011; Duarte & Gonçalves, 2007). In a first stage, I carefully read the forum postings and the interview tran- scripts. This provided me with opportunities to experience these texts as a reader. It enabled me to develop an understanding of what the text was doing, and how this was accomplished. In the second stage I coded the text. Doing this, my aim was to identify instances where an identity (I-position) was articulated. On each occasion when an identity was voiced, I shaded the text segment, and copied the extract into an adjacent comment box (using the “comment” function Alastair Henry 270 in Word). Doing this, my aim was to be as inclusive as possible. Thus, even bor- derline examples of identity articulations were included. In a third stage, I examined these extracts in more detail. My aim here was to identify the function of an utterance. Reading the text in a situated manner, I at- tempted to relate articulations of identity to discourses associated with teacher ed- ucation, practice learning, and to teaching secondary level English in the Swedish context. This involved focusing on the linguistic construction of the text. Using the comment boxes, I made notes about the rhetorical function of discourse features. Here, I paid particular attention to features in the discourse that might reveal an identity position that was currently foregrounded, and to variability within the dis- course that could indicate whether a shift in identity position might have taken place. In a final stage, and for each identified I-position, I considered whether this was the same I-position as that narrated immediately previously, and if not, whether it constituted either an entirely new I-position or an I-position previously identified in the data and which was foregrounded again. These shifts were similarly noted in the comment boxes. Working in this way, I was able to plot the movement between I-positions across each of the three types of data. 5.5. Ethical considerations Information about the research was provided to the participant and her student colleague. Assurances of confidentiality were given, together with information that participation was voluntary and that withdrawal was possible at any time. Written consent was obtained from both students. 6. Results and discussion Prior to starting teacher education, the participant, Sara, had spent 15 years in a service sector job. Although changes at her workplace provided the impetus to embark on a teaching career, in the initial interview she made clear that the idea of teaching English had “always been there since leaving school.” It was not, she said, “something I’ve come up with like a whim . . . It’s always been there. I wasn’t even applying for anything else when I applied. I knew what I wanted to apply for.” 6.1. I-positions at the start of the practicum In the interview before the start of the 4-week practicum, Sara had the oppor- tunity to talk about her feelings. She spoke of how she was looking forward to an extended period in the workplace, how she valued the opportunity to expe- rience the work of teaching English, and how she saw it as “my chance to get A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 271 proper experience of the job.” She talked about hoping to feel confident in the classroom and was adamant that she wanted to contribute to the work taking place, and of not “feeling that I am being in the way” or of being “a burden to my LUV.”1 Asked how she viewed herself at this stage in her education and how she hoped to develop over the coming weeks, she talked of wanting to experi- ence confirmation of her ability to teach, and said that the work of teaching would feel right for her: “That the more time it gets, the more confident I will become. That I will feel that I have chosen the right thing. And that it will be that it is not just something I think I can do, but will come up as a thing that I can do.” She talked also of being “curious about the students” and said that a com- mitment to young people’s development was another reason “why I want to do the job. For the students.” For Sara the story she tells is both about who she is as a person aspiring to be a teacher and, through the exercise of imagination, who she hopes to become as a person who teaches. Narratives of the self not only provide opportunities for reflection on who one has been in the past and who one is in the present; they also enable the telling of who one will be in the future (Henry, 2019). In projecting forward and describing how she hopes the practicum will be, Sara expresses a desire for validation (“that it will be that it is not just something I think I can do”) and articulates the hope that she will be perceived as a resource (“I don’t want to be a burden to my LUV”). In this way, she positions herself as an emerging practi- tioner. Even though the practicum is the first extended period she will spend in a school, the role she imagines for herself is not that of a bystander; rather it is of someone who is engaged in the practice of teaching English. In imagining the na- ture of this work, Sara describes herself as someone who can teach English, and someone who will be actively involved in facilitating students’ learning. 6.2. Initial conditions and system resilience In Hermans’ (1996) theory of the dialogical self, identities are accomplishments traceable to interactions between I-positions. As in a story, in the internal dia- logue of the self the responses of the characters (the I-positions) relate to both current and previously occurring events. In this ongoing story of the self, imme- diately preceding events can be understood as “initial conditions.” In a dynamic system, initial conditions represent the state of the system at the point in time that an observation sequence begins. Initial conditions have a determining in- fluence on the trajectory that the system takes across its state space (Larsen- Freeman & Cameron, 2008). Consequently, at the time that observations begin, 1 Mentor Alastair Henry 272 it is important to ascertain whether the system is in a more generally stable or a more generally instable state (Verspoor, 2015). A system that is in a more gen- erally stable state is likely to display high resilience to perturbations. In the state space topography, this would be represented by an attractor state with a broad basin and deep-sided valley walls (Nowak, Vallacher, & Zochowski, 2005). If per- turbed, the return of the system to an attractor state of this type would be rapid. However, if the system is in a less stable state (i.e., the basin of attraction is shallower), it can be more easily perturbed. Phase shifts – the movements of the system to a new area of the state space – would be more likely to occur. At the beginning of the practicum, Sara’s teacher identity system is in a generally stable state. It can be understood as lodged within an attractor state that represents the emerging practitioner I-position. This, for example, is re- vealed in her opening contribution to the forum dialogue at the end of the practicum’s first day: Excerpt 1 (forum post, Day 1) My LUV made a real effort to make sure I got introduced to all of her colleagues, in the “program team” she is a part of, and they all made me feel very welcome. We have had some really good conversations about some of the approaches she uses to teach and why. And she has started to give me copies of material she works with and makes herself. She tries to mix the textbook and own-made material, so I got to help her out with that. “She put me to work”, made me feel useful which I really appreci- ated. Even in classroom she through me straight in to help the students with their tasks that they were working with, and I have to admit it was a confidant boost. That I actually could help, but also that she believed in/trusted me to do so. In articulating the experience of how her mentor “put me to work,” Sara voices the experience of being the person she imagined herself to be before the start of the practicum, someone who is a resource in classroom practice. This is sim- ilarly the case when she tells how “we have had some really good conversations about some of the approaches she uses to teach and why.” Even though Sara is an early-program student, and she and her mentor have not previously met, as a discourse marker the pronoun ‘we’ functions to indicate that her knowledge and interests are recognized by an experienced teacher who finds herself having to explain and justify her pedagogical strategies. That such conversations have been identified as valuable to mentors has been noted by Akkerman and Meijer (2011), who point out that student teachers are increasingly perceived as ‘bro- kers’ “who open up possibilities for experienced teachers and schools to learn, for example, by asking critical questions” (p. 315). Voice is also given to the sense of being someone who can facilitate stu- dents’ learning and who is able to teach English. This is also part of the emerging A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 273 practitioner identity. For example, Sara writes that “even in classroom she through me straight in to help the students with their tasks that they were work- ing with, and I have to admit it was a confidant boost. That I actually could help.” The foregrounding of this I-position at the beginning of the practicum is further confirmed when, in the second interview (at the end of the practicum), Sara reflects on these early days. She says that her mentor “was good at showing me and appreciating me,” that her presence in the classroom was valued, that “she was not just having me for me, she’s having me for her as well,” and that this was something her mentor “continuously showed me.” However, even when the basin of attraction of a governing attractor state is deep-sided and broadly-stretched, the system is never entirely stable. A dy- namic system is always prone to perturbations (Larsen-Freeman & Cameron, 2008). Micro-level events that have a perturbing effect on the system and which temporarily dislodge it from the attractor state representing the emerging prac- titioner I-position form the subject of posts written on the two subsequent days: Excerpt 2 (forum post, Day 2) In a split second! Something came up today and my LUV had to leave the classroom for a few minutes to give some instructions. We were having a class in her second subject and I was asked to get the lesson started, to read aloud to the class, and for a split of a second I felt like I was 9-10 again and forced to do something that I’d been repeatedly told I was bad at. It is strange how a thing like that can kind of throws one off guard even if it just for a short while. Anyhow, I’m there to learn, and couldn’t really say no now could I, so I started to read. You know what, about 16 boys sat totally quiet and just listened, no talking or laughing just sat there quiet and listening. Good feeling maybe about time after what, 20 some years or so, to start realizing I’m not that 10 year old girl anymore when it comes to reading, and that I actually can read aloud as good as anyone else. I got asked again of my LUV if it felt ok to read and if I could do it again, so I did with a lot more confident the second time. Excerpt 3 (forum post, Day 3) It was my LUV’s second subject again. I got to go through a thing on whiteboard today, something that she kind of just asked if I could do when we were in class and not before. It kind of threw me off guard a bit making me nervous, not that I didn’t know anything about the subject just that I was so totally unprepared and had to improvise. I have never really gone through something like that before, not even written on the white- board for that matter. Of course I told her I was ok, and that she had to fill in if she felt I had missed something, not wanting her to think I was nervous about it. Hmmm. I did mention it after class though, that I’d never done anything like that before. Maybe she’ll give me more notice next time. Mind you a strange feeling I have now looking back at it is that maybe I shouldn’t get to much time to prepare, not have time make a big thing about it, and learn to trust myself, trust my instincts, and believe that I can do this. Not having time to get nervous about it I mean. Not saying that I don’t need to plan Alastair Henry 274 ahead for bigger assignments, but little once like today and yesterday. Got to read for class aloud today again, and guess what it wasn’t even close to be as “frightening” to do it today as I felt yesterday. Third time is a charm In the first of these excerpts, Sara describes a situation where, unexpectedly asked to read aloud to the class, she experiences anxiety. Instantly, she recalls pre- vious unpleasant experiences of having to read for others, and she describes how she momentarily loses her self-confidence. In the second situation an equally unex- pected request generates a similar response. Asked to explain something on the whiteboard, Sara describes how this request “kind of threw me off guard a bit mak- ing me nervous.” In both cases the system is rapidly dislodged from its attractor state and transitions into a repeller state. Repeller states are non-enduring periods of instability which, in the state space topography, form the ridges between the val- leys of attractor states, as shown in Figure 1. When a cognitive/affective system en- ters a repeller state, negative emotions are generated. Consequently, the system rarely remains there for any extended period of time (MacIntryre & Serroul, 2015). Following a shift to a repeller state, a system is likely to return to its previous attractor state. However, this passage may not always be direct. Evidenced in both of these forum entries, movement back to the attractor state representing the emerging practitioner I-position takes place via a newly emergent I-position, that of a student-apprentice. This is someone who is expected to follow the instruc- tions of a superior and to carry out allocated tasks. While Sara says that she had “never really gone through something like that before,” she also recognizes that she has little choice but to carry out the instructions: “I’m there to learn, and couldn’t really say no.” Although, as an attractor state, the newly emergent stu- dent-apprentice I-position is not one in which the system remains for any length of time, it is here that the system gravitates whenever Sara loses confidence, when she is reminded of knowledge or experiences that she does not yet possess, or when she becomes frustrated by aspects of her mentor’s classroom practice. However, as seen in Excerpt 2, the positive experience of successfully ac- complishing a challenging task – “16 boys sat totally quiet and just listened” – functions to quickly shift the system back to the emerging practitioner attractor state. On the subsequent occasions when she reads to the class, Sara’s confi- dence in her abilities grows. On the third occasion she describes this as “a charm .” This leads to the reflection that in the future she should maybe not be given too much time to prepare for tasks like this. Reflections of this sort also repre- sent newly emergent system behavior. Recognition of being a valued resource in classroom practice, and having the ability to facilitate students’ learning, func- tion to strengthen her sense of self as an emerging practitioner. As the basin of attraction of the system’s governing attractor becomes deeper and broader, the A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 275 system becomes more securely anchored. This has the consequence that, in the face of potential perturbations, it will demonstrate greater resilience. When a new position within an identity system has been narratively cre- ated, it also becomes more readily accessible in the future (Lewis & Ferrari, 2000). Moving further into the practicum, the I-position of the student-apprentice is an attractor state to which the system gravitates with greater frequency. In the final week, Sara’s posts on the online forum reflect a deeper frustration in working alongside her mentor. In the post-practicum interview, she explains how she be- gan to experience being restricted by her mentor’s approach and her methods: Excerpt 4 (post-practicum interview) I did feel after the third week that I wanted to do my own thing. Not because I had had enough of the school, because I loved it. The days went so fast. But I started to realize that I wanted to do it my way. And when you are in a VFU2 you just you are trapped into following everybody else’s footsteps. 6.3. Signature dynamics Dynamic systems can be described in terms of their signature dynamics (Dö- rnyei, 2014). Signature dynamics are “the robust causal mechanisms within a system” (Hiver, 2017, p. 672). Viewed over the practicum period, the signature dynamics of Sara’s teacher identity system reveal a shifting back and forwards between these two attractor states and reflect a pattern where the two I-posi- tions (the emerging practitioner and the student-apprentice) are variously fore- grounded and backgrounded. As revealed in Excerpts 3 and 4, on occasion shifts between these attractor states could take place via a repeller state where, mo- mentarily, the system lacks anchoring. Periodic shifts between these two attrac- tor states reflect a period of “multi-stability,” a pattern of instability that is gen- erally predictable (Vallacher, van Geert, & Nowak, 2015). 6.4. The emergence of a new I-position Being asked to read aloud (Excerpt 2), and to use the whiteboard (Excerpt 3), are perturbations that affect the system across shorter timescales (each a matter of seconds). Dynamic systems also demonstrate fluctuations over longer timescales (Larsen-Freeman & Cameron, 2008). Following the initial period of adjusting to the conditions of learning to teach in her mentor’s classroom, and around the third week of the practicum, a new I-position in Sara’s teacher identity system appears to emerge. The self-narratives that create I-positions have a powerful imaginative 2 Practicum Alastair Henry 276 dimension, meaning that the dialogical self is always oriented to the future. In- deed, in the process of the self’s becoming, the present exists merely as an un- stable moment (Hermans, 2008). As Bento, Cunha, and Salgado (2012) explain, it is imagination that enables the individual “to construct what is not yet present and project something into the future” (p. 428). Self-narratives about the future are constructed in the contexts of stories told about the present, and it is through ongoing processes of interrelations between I-positions that relate to the past, the present and the future that a coherent and consistent sense of the self can emerge (Akkerman & Meijer, 2011). Figure 1 The state space topography of the participant’s teacher identity system. The two deeper indentations (top-left and bottom-right) represent the system’s governing attractor states and correspond to the emerging practitioner and stu- dent-apprentice I-positions. The ridge between these indentations represents a repeller state, a moment in time when the system temporarily lacks anchoring. Sara’s growing self-confidence in the classroom means that the emerging prac- titioner I-position is more frequently foregrounded. However, in the following ex- cerpts, we see how the emergence of a new I-position disrupts this development. In a dynamic system, a phase shift represents a sudden change where the system self- organizes into a new pattern of behavior. Downstream from a phase shift, behavior is qualitatively different from that previously pertaining. It possesses a distinct and rec- ognizable wholeness (Larsen-Freeman & Cameron, 2008). Similarly to the emerging practitioner I-position, this new I-position is constructed from narratives that encom- pass imaginal elements and which involve projections into the future: A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 277 Excerpt 5 (stimulated recall, Day 12) Here because I’ve not gone through what I’ve planned I’m actually just trying to speed things up. So I’ve cut so much out so I kind of realise I need to do this [motions with hands at the screen] to get to the next part. Urm. [pause] And I’m thinking how I’m gonna connect to the next. This is how I’m thinking there. I need to get from here to the next part. Though I’m thinking they kind of need to [pause] So that’s going through my mind when I’m, I’m doing this. Erm. Well I’m actually thinking about it the whole time when I’m [motions with hands towards the screen] I’m going to have to do the next thing and connect it to it. . . . I, I think that it’s something I’m going to come across quite a lot and I think that, I think that, I don’t know, obviously I need to practice on it and maybe it gets better, erm, but I do think that sometimes you never know what’s going to happen in a classroom and I think that I’m going to come across it a lot, but I just need to work the time better, and hopefully that’s something that comes with, comes with experience. Excerpt 6 (forum post, Day 18) Sad thing and a little depressing in all this though is that it here shows that my LUV doesn’t have sufficient the time to spend with each and every student in the class- room that is needed to give the individual student the time they need. Take for exam- ple student 2, that now will get a grade in this subject, and most likely a C, and most probably wouldn’t gotten a grade at all if it hadn’t been for the fact that we now for 4 weeks have been two people in the classroom instead of 1. Not that my LUV isn’t trying to “see” all the students but if you have a student that you over and over again tell that you need them to work and hand things in, and the student doesn’t what is there to do? . . . I know this is the back side of this profession; that time and money sadly limits the teacher to do all that they want to do for their students and that I one day also will be faced with these problems. However, the feeling, the satisfaction I feel after today still over powers that. How good it feels to see how proud the students become of themselves when they have achieved a goal/task, and how good it feels to have been a part of their achievement. In both of these excerpts, Sara demonstrates an awareness of the con- straints and challenges faced in the context of carrying out teaching in her men- tor’s classroom. As Bloomfield (2010) observes, during the practicum pre-service teachers embark on journeys of learning. These journeys involve conflict and struggle; learning to teach requires continual compromise, and preservice teach- ers can become embroiled in constant processes of accommodation. While in the first of these excerpts Sara recognizes how she needs to cut back on the content of her lesson in order to complete it within the stipulated timeframe, in the sec- ond she reflects on the needs-resources equation and how the teacher’s limited time impacts negatively on students’ opportunities for growth. However, it is important to be aware that these challenges are perceived not only in the context of the current practice in which Sara is involved. In sharing Alastair Henry 278 her experiences with her preservice teacher colleague in the stimulated recall discussion (Excerpt 5) and in the discussion forum (Excerpt 6), she projects into an imaginary future, framing the problems as challenges to be addressed in the imaginal spaces of future practice. While in Excerpt 5 the challenge she envis- ages involves structuring activities in ways that can enable students to create meaning (“I think that it’s something I’m going to come across quite a lot”), in Excerpt 6 a future working life is envisaged in the recognition that resource con- straints can prevent students from fully developing their potential (“I know . . . that I one day also will be faced with these problems”). Unlike the emerging practitioner and student-apprentice I-positions, which more generally involve orientations to conditions currently prevailing, the I-posi- tion that emerges here involves an undetermined language teaching future. In di- alogical self theory, the self is not only “here,” but also “there” (Hermans, 2008). It is through the power of imagination that the person can act “as if he or she were the other and the other were him- or herself” (Hermans, 2001, p. 250). This newly emergent I-position – a challenged practitioner – constitutes a new attractor state in the system’s state space. However, it is not just the sense in which Sara frames her present experiences in terms of imagined future challenges that is character- istic of this new I-position. The challenged practitioner also differs from the emerg- ing practitioner in the manner in which it is narratively constructed. In voicing con- cerns about the challenges she expects to face in the future, Sara narrates a self- identity as a member of a community of practice of other teaching professionals: “. . . if it hadn’t been for the fact that we [emphases added] now for 4 weeks have been two people in the classroom instead of 1,” “but I do think that sometimes you never know what’s going to happen in a classroom” (Excerpt 5), and “not that my LUV isn’t trying to ‘see’ all the students but if you have a student that you over and over again tell that you need them to work and hand things in, and the student doesn’t what is there to do?” (Excerpt 6). Pre-service teachers look to the future, often foreseeing different self-identities, such as for example being an inspiring teacher. Providing a source for identity formation, imagined future practice ena- bles the pre-service teacher to transcend immediate situations and to create visual representations of themselves in broader contexts of professional work (Trent, 2011). It is through acts of imagination of this sort, and from the voices that are articulated in these representations, that professional identities develop. 7. Conclusion Comprised of I-positions oriented towards the past, the present and the future, the dialogical self is temporally configured. Mapping the processes of this par- ticipant’s teacher identity development, three I-positions were identified. While A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 279 the student-apprentice I-position is oriented to presently pertaining conditions in learning the work of teaching, the emerging practitioner and the challenged practitioner involve narrative interpretations of current events from a vantage point in an imagined future. In common with possible selves (Markus & Nurius, 1986) and McAdams’ (1985) notion of the imago, I-positions have a function in guiding current behavior (Henry, 2019). The dialogical self is both “here” (in the present) and “there” (in the future). As Hermans (2001) explains, it is through the power of imagination that a “person can act as if he or she were the other and the other were him- or herself” (p. 250). As revealed in the exploration of identity dynamics in this study, experience and imagination combine in the de- velopment of an identity of being someone who teaches English. The practicum is a place of struggle, and the process of developing a teacher identity is rarely linear (Beauchamp & Thomas, 2009). In this study, ten- sions are in evidence in momentary periods where the system enters a repeller state, and when shifts take place between attractor states that represent I-posi- tions which are contradictory in nature. While together the forum posts offer evidence of identity development over time – specifically the emergence of the I-position of the challenged practitioner which is more frequently foregrounded in the posts towards the end of the period – examinations across shorter time- scales (i.e., within a particular post) reveal how changes in the immediate con- text of learning to teach can trigger changes in the identity system. In this study, investigation of the identity narratives of the pre-service teacher participant has shed light on the dynamical processes through which a professional identity develops. Focusing on the complexity principles of self-organization and emergence, analyses show how a language teacher identity is dialogically con- structed, how I-positions are foregrounded and backgrounded, and how new I-po- sitions come into being. As evidenced in the inter-personal data used in the study, it is through the dynamical positioning and repositioning of voiced identities within the dialogical self of being someone who teaches that innovation takes place. In everyday interactions in language classrooms, teachers are constantly involved in identity work; everything that a teacher does and knows is in some way implicated in the continual process of identity formation (Miller, 2009). To- gether with a complexity-framing, the conceptualization of language teacher identity as a collection of narratively constructed I-positions makes it possible to shed light on identity development as it takes place. Because it provides a means of understanding the in-situ mechanics of identify shifts, the complexity approach adopted in the study has enabled identity development to be investi- gated as a process in motion that is operative across varying timescales. Further- more, in taking a perspective in which I-positions are understood as attractor states in a complex dynamic system, it is possible to understand how the foregrounding Alastair Henry 280 of a particular I-position can have effects not only at the system level but also on other interpenetrating systems. As seen here, and in a previous study (Henry, 2016), a contextual change will often coincide with a shift in a mood. In Sara’s teacher identity system, the attractor state representing the emerging practitioner I-position has a wide ba- sin of attraction for positive emotions; she feels good when things go well, and when students respond in anticipated ways. Equally, the attractor state repre- senting the student-apprentice I-position has a wide basin of attraction for neg- ative emotions. Having to work in ways contrary to her beliefs about effective teaching, she experiences negative emotion. Importantly, when an emotional experience is sufficiently incongruent with the system’s currently governing at- tractor – for example experiencing a positive response from a student when the system is in the student-apprentice attractor state – movement to a different part of the state space can be triggered. Revealing how shifts in the teacher identity system have consequences for other systems is also an insight of im- portance that emerges from this study. In periods of learning such as the practicum where processes of identity development can be particularly intense, the conceptualization of teacher iden- tity as dialogical and dynamic can provide a valuable pedagogical tool in the ed- ucation of pre-service teachers. In the same way that Johnston (1997) highlights the theoretical value of Bakhtinian notions of identity in “detecting and unrav- eling the multiple, competing discourses present in the teachers’ speech” and in conceptualizing “teachers’ lives in EFL/ESL in a way that captures the com- plexities and contradictions of those lives without a general descent into inco- herence” (p. 707), from a practice perspective I-positions can be similarly used to identify, explore and untangle developing professional identities and their ef- fects. For pre-service teachers, the process of learning to work can be supported through the development of insights into the complexities of identity work (Henry & Tynkkinen, 2017). Models that enable students to conceptualize iden- tity as multivoiced, and which fame identity tensions as struggles between an- tagonists in a play who have competing opinions, can provide a practical means of investigating the dramas of the self that play out when learning to teach. Working with identity work in this way can play an important role in student- mentoring (Henry, 2019). Of course, it is impossible to predict the exact nature of the struggles and the types of tension that teachers have to confront in developing a professional identity (Pillen, Den Brok, & Beijaard, 2013). However, by enabling pre-service teachers to recognize that the development of a professional identity necessarily involves con- flicts between competing conceptualizations of what it means to be a teacher, and by providing them with models through which identities can be mapped and A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 281 developmental processes can be unraveled, dialogical conceptualizations of the self can play an important role in the mentoring process. Acknowledgements I received insightful and important advice from the anonymous reviewers of this article, and from the reviewers of a previous article in Journal of Teacher Educa- tion. I would also like to thank Bo Svensson, my colleague at University West, and Tammy Gregersen, who got me started with this work. Most of all, I would like to thank “Sara” and her preservice colleague “Lina” for agreeing to take part in the study and for sharing their experiences with me. Alastair Henry 282 References Akkerman, S. F., & Meijer, P. C. (2011). A dialogical approach to conceptualizing teacher identity. Teaching and Teacher Education, 27, 308-319. Alsup, J. (2006). Teacher identity discourses. Negotiating personal and profes- sional spaces. Mahwah: Lawrence Erlbaum. Bakhtin, M. (1981). The dialogic imagination: Four essays by M. M. Bakhtin (C. Emerson & M. Holquist, Trans.). Austin: University of Texas Press. Beauchamp, C., & Thomas, L. (2009). Understanding teacher identity: An over- view of issues in the literature and implications for teacher education. Cambridge Journal of Education, 39(2), 175-189. Beauchamp, C., & Thomas, L. (2011). New teachers’ identity shifts at the bound- ary of teacher education and initial practice. International Journal of Edu- cational Research, 50, 6-13. Beijaard, D., Meijer, P. C., & Verloop, N. (2004). Reconsidering research on teach- ers’ professional identity. Teaching and Teacher Education, 20, 107-128. Bell, N. J., & Das, A. (2011). Emergent organization in the dialogical self: Evolution of a ‘‘both’’ ethnic identity position. Culture & Psychology, 17(2), 241-262. Bento, T., Cunha, C., & Salgado, J. (2012). Dialogical theory of selfhood. In J. Valsiner (Ed.), The Oxford handbook of culture and psychology (pp. 421- 438). Oxford: Oxford University Press. Bloomfield, D. (2010). Emotions and “getting by”: A preservice teacher navigating pro- fessional experience. Asia-Pacific Journal of Teacher Education, 38, 221-234. Britzman, D. P. (2006). Foreword. In J. Alsup (Ed.), Teacher identity discourses: Negotiat- ing personal and professional spaces (pp. ix-xii). Mahwah: Lawrence Erlbaum. Byrne, D., & Callaghan, G. (2014). Complexity theory and the social sciences: The state of the art. Abingdon: Routledge. Chu, S. K. W., Kwan, A. C. M., & Warning, P. (2012). Blogging for information management, learning, and social support during internship. Educational Technology & Society, 15(2), 168-178. Coyle, A. (2006). Discourse analysis. In G. M. Breakwell, S. Hammond, C. Fife- Schaw, & J. A. Smith (Eds.), Research methods in psychology (pp. 366-387). London: Sage. Deng, L., & Yuen, A. H. K. (2013). Forums in preservice teacher education: Ex- ploring the participation issue. Technology, Pedagogy and Education, 22(3), 339-356. Dörnyei, Z. (2014). Researching complex dynamic systems: “Retrodictive qualitative modelling” in the language classroom. Language Teaching, 47(1), 80-91. Duarte, F., & Gonçalves, M. M. (2007). Negotiating motherhood: A dialogical ap- proach. International Journal for Dialogical Science, 2, 249-275. A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 283 Ferrier-Kerr, J. (2009). Establishing professional relationships in practicum set- tings. Teaching and Teacher Education, 25, 790-797. Geijsel, F., & Meijers, F. (2005). Identity learning: The core process of educational change. Educational Studies, 31(4), 419-430. Henry, A. (2016). Conceptualizing teacher identity as a complex dynamic system: The inner dynamics of transformations during a practicum. Journal of Teacher Education, 67(4), 291-305. Henry, A. (2019). Working with identity work: Supporting the exploration of profes- sional identity development in teacher education. Manuscript in preparation. Henry, A., & Tynkkinen, M. (2017). Becoming a process researcher of one’s own development: Using an identity mapping model to make sense of trans- formation dynamics during the practicum. In T. Gregersen & P. D. Mac- Intyre (Eds.), Exploring innovations in language teacher education: Trans- formational theory and practice (pp. 205-228). New York: Springer. Hermans, H. J. M. (1996). Voicing the self: From information processing to dia- logical interchange. Psychological Bulletin, 119(1), 31-50. Hermans, H. J. M. (2001). The dialogical self: Toward a theory of personal and cultural positioning. Culture & Psychology, 7(3), 243-281. Hermans, H. J. M. (2003). The construction and reconstruction of a dialogical self. Journal of Constructivist Psychology, 16(2), 89-130. Hermans, H. J. M. (2008). How to perform research on the basis of dialogical self theory? Introduction to the special issue. Journal of Constructivist Psychol- ogy, 21(3), 185-199. Hermans, H. J. M., Kempen, H. J. G., & Van Loon, R. J. P. (1992). The dialogical self: Beyond individualism and rationalism. American Psychologist, 47, 23-33. Hiver, P. (2017). Tracing the signature dynamics of language teacher immunity: A retrodic- tive qualitative modeling study, Modern Language Journal, 101, 669-690. Hiver, P., & Al-Hoorie, A. H. (2016). A “dynamic ensemble” for second language research: Putting complexity theory into practice. Modern Language Jour- nal, 100, 741-756. James, W. (1890). The principles of psychology (Vol. 1). London: Macmillan. Jasper, C., Moore, H., Whittaker, L., & Gillespie, A. (2012). Methodological ap- proaches to studying the self in its social context. In H. Hermans & T. Gie- ser (Eds.), Handbook of dialogical self theory (pp. 319-334). Cambridge: Cambridge University Press. Johnston, B. (1997). Do EFL teachers have careers? TESOL Quarterly, 31(4), 681-712. Kaplan, A., & Garner, J. K. (2017). A complex dynamic systems perspective on identity and its development: The dynamic systems model of role identity. Developmental Psychology, 53(11), 2036-2051. Alastair Henry 284 Kaplan, A., & Garner, J. K. (2018). Teacher identity and motivation: The dynamic systems model of role identity. In P. A. Schutz., J. Hong, & D. Cross Francis (Eds.), Research on teacher identity: Mapping challenges and innovations (pp. 71-82). New York: Springer. Klimstra, T. A., Luyckx, K., Hale, W. W., Frijns, T., van Lier, P., & Meeus, W. H. J. (2010). Short-term fluctuations in identity: Introducing a micro-level approach to iden- tity formation. Journal of Personality and Social Psychology, 99(1), 191-202. Kostoulas, A., Stelma, J., Mercer, S., Cameron, L., & Dawson, S. (2017). Complex systems theory as a shared discourse space for TESOL. TESOL Journal. Ad- vance online publication. Larsen-Freeman, D. (2017). Complexity theory: The lessons continue. In L. Ortega & Z. Han (Eds.), Complexity theory and language development: In celebration of Diane Larsen-Freeman (pp. 12-50). Amsterdam: John Benjamins. Larsen-Freeman, D., & Cameron, L. (2008). Complex systems and applied linguis- tics. Oxford: Oxford University Press. Lewis, M. D., & Ferrari, M. (2000). Cognitive-emotional self-organization in per- sonality development and personal identity. In H. A. Bosma & E. S. Kunnen (Eds.), Identity and emotion: Development through self-organization (pp. 177-198). Cambridge: Cambridge University Press. MacIntryre, P. D., & Serroul, A. (2015). Motivation on a per-second timescale: Examining approach-avoidance motivation during L2 task performance. In Z. Dörnyei, P. D. MacIntyre, & A. Henry (Eds.), Motivational dynamics in language learning (pp. 109-138). Bristol: Multilingual Matters. Markus, H. R., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954-969. McAdams, D. P. (1985). The “imago”: A key narrative component of identity. In P. Shaver (Ed.), Review of personality and social psychology (pp. 115-141). Beverly Hills: Sage. Miller, J. (2009). Teacher identity. In A. Burns & J. C. Richards (Eds.), The Cam- bridge guide to second language teacher education (pp. 172-181). Cam- bridge: Cambridge University Press. Nowak, A., Vallacher, R., & Zochowski, M. (2005). The emergence of personality: Dy- namic foundations of individual variation. Developmental Review, 25, 351-385. Ortega, L., & Han, Z. (2017). Complexity theory and language development: In cele- bration of Diane Larsen-Freeman. Amsterdam: John Benjamins. Pillen, M. T., den Brok, P. J., & Beijaard, D. (2013). Profiles and change in begin- ning teachers’ professional identity tensions. Teaching and Teacher Edu- cation, 34, 86-97. Potter, J., & Wetherell, M. (1995). Discourse analysis. In J. Smith, R. Harré, & L. Van Langenhove (Eds.), Rethinking methods in psychology (pp. 80-92). London: Sage. A drama of selves: Investigating teacher identity development from dialogical and complexity. . . 285 Rodgers, C. R., & Scott, K. H. (2008). The development of the personal self and professional identity in learning to teach. In M. Cochran-Smith, S. Feiman- Nemser, D. J. McIntyre & K. E. Demers (Eds.), Handbook of research on teacher education (pp. 732-755). New York: Routledge. Smagorinsky, P., Cook, L. S., Moore, C., Jackson, A. Y., & Fry, P. G. (2004). Tensions in learning to teach: Accommodation and the development of a teaching identity. Journal of Teacher Education, 55(1), 8-24. Sutherland, L., Howard, S., & Markauskaite, L. (2010). Professional identity creation: Examining the development of beginning preservice teachers’ understanding of their work as teachers. Teaching and Teacher Education, 26, 455-465. Trent, J. (2011). “Four years on, I’m ready to teach”: Teacher education and the construction of teacher identities. Teachers and Teaching: Theory and Practice, 17(5), 529-543. Trent, J. (2013). From learner to teacher: Practice, language, and identity in a teach- ing practicum. Asia-Pacific Journal of Teacher Education, 41(4), 426-440. Vallacher, R. R., van Geert, P., & Nowak, A. (2015). The intrinsic dynamics of psy- chological process. Current Directions in Psychological Science, 24(1), 58-64. Valsiner, J. (2004, August). Temporal integration of structures within the dialog- ical self. Keynote lecture at the 3rd International Conference on Dialogical Self, Warsaw, Poland. van Geert, P., & Lichtwarck-Aschoff, A. (2005). A dynamic systems approach to family assessment. European Journal of Psychological Assessment, 21(4), 240-248. Verspoor, M. (2015). Initial conditions. In Z. Dörnyei, P. D. MacIntyre, & A. Henry (Eds.), Motivational dynamics in language learning (pp. 38-46). Bristol: Multilingual Matters. Walkington, J. (2005). Becoming a teacher: Encouraging development of teacher identity through reflective practice. Asia-Pacific Journal of Teacher Educa- tion, 33(1), 53-64.