On this page

Synthesis: Corbin, Tai and Flenady argue that feedback works only inside mutual recognition — each party granting the other standing to participate in a practice — so the same comment can guide one student and vanish for another. Recognition supplies the architecture the relational turn lacked, names three ways it breaks down, and makes repair a matter of rebuilding standing rather than sharpening wording.

This is a theoretical paper with no dataset of its own; findings it cites are attributed to the studies that produced them.

Key Findings

  1. Recognition is standing, not approval — it treats a person's words as counting in a shared practice, which is what lets information become a reason to act.
  2. Feedback's power is relational — identical comments carry different force depending on who may say them and who is seen as able to answer.
  3. Trust, care, respect and belonging are one structure — facets of recognition, not separate ingredients.
  4. Feedback forms identities both ways — it can place a student inside a discipline, and ignored comments reduce a teacher to a functionary.
  5. Breakdown takes three forms — misrecognition, under-recognition and conditional recognition, each making the next encounter harder.
  6. Automated feedback is extra-recognitive — accuracy and safety come from the absence of judgment that leaves it unable to move a learner.

What recognition adds to feedback theory

Feedback research already knows that information alone does not change practice: students discount comments from providers they doubt, and teachers find generic comments inadequate. The relational turn answered with trust, care, respect and belonging, but those terms arrived separately. Dai and colleagues found the vocabulary fragmented, and even relational accounts keep an asymmetry: teachers supply comments, and only students are shaped by the relationship.

Recognition, borrowed from social philosophy, is the reciprocal attribution of standing within a shared practice — being entitled to speak, be answered, and hold others to account. It is constitutive, not decorative: it turns utterances into a conversation and comments into feedback. Respect, belonging, vulnerability, trust and care name facets of one relation: equal entitlement to speak, shared standing, dependence on others' judgment, and what that dependence obliges. The vocabulary does not disappear; it becomes one structure with several names.

Recognition also runs both ways. A student who dismisses every comment withholds standing, and a teacher whose comments go unread loses a professional identity.

Why identical comments land differently

Three students receive one sentence: "Your argument lacks sufficient evidence to support the claim." One recognizes a disciplinary expert and revises toward disciplinary standards; one recognizes the grade-controller and reads "add sources to pass"; one grants neither and disengages. The information is constant; the standing is not; the force varies.

Recognition is how identity is built. "You are starting to think like a psychologist" does more than describe a draft: it places the student inside a discipline, answerable to its standards and entitled to its responsibilities. Corrective comments do the same work when the relationship supports it. In reverse, stock comments can make a student feel like an anonymous case, and a student persistently misgendered by a teacher may hear accurate feedback as noise.

Learning needs two recognitions: the student must recognize the provider as entitled to guide, and the provider must recognize the student as capable of responding. Without that, feedback literacy has nothing to work on.

When feedback loses force

Automated and generative AI feedback is extra-recognitive: mutual recognition is structurally impossible, because a system cannot recognize a learner's trajectory or be recognized as a responsible agent. Students often seek it for that reason — it carries no judgment, so it cannot wound a vulnerable identity. That safety is the limit: it does not matter what a model thinks of my idea, so its comments give me no reason to change. It can refine performance; it cannot cultivate the person.

Within recognitive relationships, failure takes three forms. Misrecognition actively misrepresents identity. Under-recognition is perfunctory, as in generic comments that never acknowledge a particular learner. Conditional recognition grants standing only on terms, such as being heard if you conform to dominant cultural or linguistic norms. Each is a harm, and each is recursive: the student withdraws, the teacher stops investing, and the withdrawal confirms both parties' worst reading.

Technical fixes therefore underperform: detailed rubrics, faster returns and directive language can improve measured quality while leaving the relation untouched.

What repair requires

Repair is the re-establishment of mutual recognition after breakdown, not better phrasing: restoring the standing that makes feedback consequential. The paper follows Ella, a second-year psychology student whose tutor's marginal comment lands with force, then a semester in which marking is outsourced to casual markers she does not know. The new advice is sound, even excellent, but it comes from someone with procedural authority and no recognizable place in the discipline. Correct information becomes decontextualized advice.

Repair means redesigning the encounter so both parties become visible: markers introduce themselves, discuss exemplar reports, hold short follow-ups. These are structural moves, not sentimental ones. The burden is not the teacher's alone: students reaffirm teachers' standing by engaging, and institutions decide whether relationships can form at all.

The authors treat recognition as feedback theory's foundation rather than an addition to it.

What this means for practice

  • Instructors: Address the student as a participant in the discipline, not just the owner of a draft: name what the work is starting to do well, and frame corrections inside that.
  • Feedback designers: Protect continuity between whoever comments and whoever teaches; where marking is distributed, design introductions and exemplar discussions rather than longer comments.
  • Program leaders: Treat outsourced marking by staff with no continuing relationship to the cohort as a pedagogical cost, and fund the contacts that let recognition form.
  • Faculty developers: Ask "whose standing is being withheld?" and sort complaints into misrecognition, under-recognition or conditional recognition before reaching for a template.

Limitations

  • Conceptual argument, no empirical test: the Ella scenario comes from the authors' teaching experience, not data.
  • It imports Honneth and Brandom into higher education without saying how recognition would be measured, so it diagnoses more readily than it evaluates.
  • The claim that AI feedback cannot be recognitive rests on a strong account of agency that would need revisiting as systems gain memory and persona.
  • Cohort size, workload and institutional power stay background conditions rather than analyzed constraints.

Citation

Corbin, T., Tai, J., & Flenady, G. (2026). The pedagogical power of recognition in feedback. Teaching in Higher Education, 31(5), 1034–1049. https://doi.org/10.1080/13562517.2026.2633634

Embed this page

Copy the code below to embed this page in a learning management system or other website. The embedded version shows only the page content — it hides the site header, navigation, and footer.