Research spotlight: Michael Magid on self-concept and language learning

A student who can picture herself arguing a point in English at a seminar table in Manchester has something a student who draws a blank does not. Call it a rehearsal. The ones who can rehearse keep showing up to class. The ones who can't tend to stop.

That, in plain terms, is the argument behind The Impact of Self-Concept on Language Learning, the edited volume Kata Csizér and Michael Magid assembled for Multilingual Matters in 2014. Csizér teaches in the Department of English Applied Linguistics at Eötvös Loránd University in Budapest. Magid is an English Language Pedagogy Specialist at the English Language Institute of Singapore, where he runs professional development for English teachers and conducts action research on language learning motivation and on drama as a classroom tool. He has taught English in Canada, England, China and Japan. Their book runs to 21 chapters in five parts, with contributors based on four continents and studies drawn from Central Europe, Canada, Asia and Australia.

The question the volume keeps circling is the one every language department now faces. When a free app will handle a menu, a contract or a conference talk in seconds, why should anyone spend years learning to do it alone? Csizér and Magid's answer is that the motive was never purely transactional. Learners persist because of who they imagine themselves becoming.

Three selves and a mental picture

Most of the chapters work with a framework Zoltán Dörnyei proposed in 2005, the L2 Motivational Self System. L2 here simply means a second or foreign language. The system has three parts. The ideal L2 self is the fluent version of you that you would like to be. The ought-to L2 self is the version other people expect, such as the relative who wants a doctor in the family who speaks German. The L2 learning experience covers how the class itself feels, week to week.

The theory borrows from Markus and Nurius's work on possible selves and from Higgins's self-discrepancy theory, and its central claim is narrow enough to test. A future self only motivates if it behaves like a memory of something that hasn't happened yet. Dörnyei's own chapter in the book, on future self-guides and vision, puts that sensory quality at the centre and argues that teachers have to convert vision and imagination into practical action, or nothing follows.

Magid's contribution to the framework is a checklist of conditions. In his writing on the L2 Motivational Self System he sets out what has to be in place before an imagined self does any work: the image must be distinct from the current self, vivid and detailed, plausible without feeling guaranteed, and attached to a set of specific action plans. Two further conditions shape whether it survives contact with a classroom. The ideal and ought-to selves should not pull in opposite directions, and the learner should feel that effort actually moves the needle.

Think about the difference between a New Year's resolution and a dress rehearsal. A resolution is a sentence you say in January. A rehearsal is a scene you can walk through, with a script and a sense of where the light falls. The intervention studies in this book are attempts to turn resolutions into rehearsals.

What the classroom studies found

Several chapters look at learners who go quiet. Jim King's contribution examines the reticence of Japanese university students learning English through the lens of clinical social anxiety, specifically the cognitive model David Clark and Adrian Wells published in 1995. Drawing on interviews, stimulated recall and classroom observation from a large mixed-methods study, King describes students who monitor themselves so closely in class that speech stops, reading a stumble in English as evidence of a defective self rather than a grammar slip. The self-concept is doing the damage, not the vocabulary.

Nihat Polat's chapter connects the same system to something measurable in a recording: accent. He examines how the ideal L2 self interacts with learner beliefs and with patterns of socialisation and identification with the target-language group to shape how close learners get to native-like pronunciation. Elke Stracke, Jeremy Jones and Nicolette Bramley followed adult migrants in Australian English programmes, where a bicultural identity, comfortable in the old country and the new one at once, seemed to support persistence. Sarah Mercer's theoretical chapter argues that the self is not a private possession but a network of relationships, which means a learner's self-concept can be rebuilt by changing the people around them.

On the teaching side, Gabriella Mezei tested motivational strategies against students' self-related beliefs about their own competence. Yuzo Kimura tracked a single English teacher in Beijing longitudinally, using complex dynamic systems theory to show how a teacher's motivation rises and falls with the institution around him. The full volume is available through De Gruyter, including the editors' closing chapter on future research directions.

Photo by Ling App on Unsplash

The programme Magid built

The fourth part of the book moves from description to intervention, and Magid's own chapter sits in it. The sequence he describes is unglamorous: students build a detailed image of themselves as English speakers, break that image into goals, then write action plans for reaching each goal, and rehearse the image again.

In a Singapore secondary-school study reported in his doctoral work and later papers, students who went through the complete version were compared with a group that built the images without the goal-setting and action-planning steps. About nine in ten of the first group said their desire to learn English had strengthened, against roughly half of the comparison group. The outcome measure is self-report, which is a real limit rather than a footnote. Still, the direction is consistent with Magid's earlier comparative work with Letty Chan on two motivational programmes for Chinese learners in England and China, which reported gains in the strength of the ideal L2 self, in motivation and in confidence.

Chan's own chapter in the volume takes the imagery technique to Chinese university students. Jessica Mackay's applies the same logic in a Spanish EFL setting, working with a class in a Barcelona language school over a twelve-week programme. Read together, the intervention chapters make a modest case: the technique is teachable, it fits inside ordinary class time, and it does not require new technology.

What changed after the book came out

The volume appeared in August 2014, two years before neural machine translation cut error rates sharply enough to make the practical case for language learning feel less obvious. The Guardian reported in November 2025 that nearly 50 UK centres of French, German, Italian and Iberian studies have closed or shrunk in five years, with A-level French entries falling from close to 30,000 to under 10,000 over three decades and German from 10,000 to about 2,000 a year.

Translators have felt it too. A survey by the Translators Association of China reported in April 2025 found 97.1 percent of translation firms using AI tools, with a quarter of projects delivered as pure machine translation. Middlebury College announced in August 2025 that it would close its Monterey graduate programmes in translation and interpretation by 2027, citing enrolment decline.

Here the book's framework becomes more useful, not less. If a learner's goal is transactional, the ideal self is thin and gets outsourced to a phone. If the goal is relational, the imagined self has to carry weight no app can carry: being the person who gets the joke, who handles the negotiation without an intermediary, who is not standing outside the conversation waiting for a translation.

Where the evidence strains

Most of the book's data are self-report, and that matters. A learner who says her ideal self is vivid may simply be a confident questionnaire-taker, and the outcome measures are often intended effort rather than demonstrated proficiency. A 2024 critique by Al-Hoorie and colleagues questioned whether the standard ideal L2 self scale captures future-self vision or quietly captures self-efficacy, and whether the effort items overlap too much with the predictors to prove anything. MacIntyre argued in 2022 that the ought-to L2 self leaves the process of internalising other people's expectations unspecified, which is a problem in a model built on the idea that external pressure can become personal.

There is also the matter of scale. The chapters are mostly small studies in single contexts, and a chapter collection is not a trial. Muhammad Shahbaz, reviewing the volume in the Canadian Journal of Applied Linguistics, called it valuable for researchers and practitioners alike and highlighted the range of methods across chapters, framing the practical payoff as help with motivating the unmotivated. Ema Ushioda of the University of Warwick said the editors had drawn together the leading thinkers in the psychology of language learning. Neither endorsement is the same as evidence that the interventions work at scale. The book closes with Magid and Csizér on future research directions, which is an admission that the questions outnumbered the answers.

Photo by Ling App on Unsplash

What a department can take from it

None of this requires buying anything. The techniques described in the intervention chapters are paper, pen, talk and time.

• Ask learners to write the scene rather than the goal: where they are, who is in the room, what they say first.
• Separate the image from the plan. A vivid picture with no steps attached tends to produce a pleasant afternoon and no homework.
• Watch the quiet students for self-monitoring, not just for shyness. King's chapter suggests some silence is a self-protective strategy that a teacher can interrupt with smaller stakes.
• Treat teacher morale as part of the motivational system, as Kimura's longitudinal case does, rather than as a separate management problem.

For programme directors, the implication is blunter than it looks. If enrolments are driven by an imagined future self, then the job advert, the course description and the first week of class are all part of the motivational apparatus, and a department that cuts them to a list of credit hours is working against its own recruitment.

Where does this leave language teaching over the next decade? Probably with fewer students who arrive for transactional reasons and more who arrive for identity reasons, and with departments under pressure to justify themselves in economic language that the evidence no longer fully supports. The book offers a different justification. A learner persists because of a version of themselves they want to live in, and the teacher's job is to make that version detailed enough to rehearse. That claim was plausible in 2014. As translation gets cheaper, it may be the one that holds.