Summary
Epistemic injustice is what happens when someone is wronged specifically in their capacity as a knower: when their testimony is dismissed, their credibility is downgraded, or the concepts they need to make sense of their own experience do not exist. Miranda Frickerās Epistemic Injustice (2007) gave this phenomenon a name and a structure. The framework has since become one of the most productive philosophical tools for understanding what goes wrong between neurodivergent people and the institutions that claim to serve them.
This is not an abstract concern. An autistic woman tells her GP that fluorescent lights cause her physical pain, and the GP attributes it to anxiety. A late-diagnosed man spends thirty years knowing something is different about him but lacking any framework to understand what. An autistic childās account of being bullied is treated as less reliable than the neurotypical childrenās because the autistic child ādoesnāt understand social situations.ā These are not isolated failures. They are systematic patterns, and epistemic injustice theory explains why they persist.
The framework
Fricker (2007) identified two primary forms of epistemic injustice.
Testimonial injustice occurs when a speaker is given less credibility than they deserve because of a prejudice held by the hearer. The hearer does not consciously decide to disbelieve; the credibility deficit operates through identity prejudice, an automatic downgrading of the speakerās word based on who they are rather than what they are saying. A woman is believed less than a man reporting the same symptoms. A Black patientās pain reports are taken less seriously. An autistic personās account of their own experience is dismissed as lacking insight.
Hermeneutical injustice occurs when someone lacks the shared conceptual resources to make sense of a socially significant experience. The experience is real, but the concepts needed to understand and communicate it do not exist in the collective interpretive toolkit, because the group that has the experience has been excluded from the processes through which shared meanings are produced. The result is not confusion. It is something worse: the inability to understand what is happening to you, in a way that others can recognise.
Frickerās framework has been extended by others. Dotson (2011) identified contributory injustice, where a knower has the hermeneutical resources to understand their experience but the dominant group refuses to uptake those resources. Kidd and Carel (2017) applied the framework to healthcare, showing how illness systematically produces both forms of injustice: patients are disbelieved (testimonial) and lack concepts for their experience (hermeneutical). Their work on what they call āpathocentricā epistemic injustice, the injustice that arises specifically because someone is ill or disabled, maps directly onto the neurodivergent case.
Testimonial injustice and neurodivergence
The most visible form of epistemic injustice in neurodivergent life is the routine dismissal of neurodivergent testimony.
In clinical settings. Uisma et al. (2026) documented how autistic young adults experience healthcare interactions as sites of systematic testimonial injustice. Their accounts of their own bodies, their pain, their sensory experience, their mental states were routinely minimised, reinterpreted, or overridden by clinicians working from a narrow medical model. The study found that autistic patients were not simply disbelieved on occasion. The structure of clinical interaction itself, its time constraints, its reliance on neurotypical communication norms, its assumption that the clinicianās assessment outranks the patientās testimony, produced a consistent credibility deficit.
Kidd and Carel (2017) described this pattern in broader terms: when someone is perceived as ill or disabled, their epistemic authority is deflated. Their reports are treated as symptoms rather than testimony. An autistic person who says āthis environment is painfulā is heard not as reporting a fact about their experience but as exhibiting a feature of their condition. The statement is medicalised rather than credited.
In education. An autistic child who says āthe classroom is too loudā is told they need to learn to cope. The child is the expert on their own sensory experience; the teacher is the expert on classroom management. When these conflict, the childās testimony loses. Not because the teacher is malicious, but because the institutional structure assigns credibility based on role and age, and the childās neurodivergent status further reduces whatever credibility their role affords.
In daily life. āBut you donāt look autistic.ā This common response to disclosure is testimonial injustice in miniature. The speaker has decided, on the basis of appearance, that the autistic personās claim about their own neurotype is not credible. The autistic personās self-knowledge is weighed against the neurotypical personās expectation of what autism looks like, and the expectation wins.
Masking (see Masking and camouflaging) intensifies this dynamic. The better someone masks, the less credible their claim to be autistic becomes. The very skill that allowed them to survive neurotypical environments is used as evidence against their own testimony. This is a particularly cruel variant of testimonial injustice: the cost of being believed is being visibly impaired, which means the price of credibility is the abandonment of the coping strategy that kept you functional.
Hermeneutical injustice and neurodivergence
If testimonial injustice is about not being believed, hermeneutical injustice is about not being able to understand what is happening to you in the first place. For many neurodivergent people, this is the deeper wound.
Late diagnosis as hermeneutical injustice. Dinishak (2021) argued that the experience of undiagnosed autism constitutes a textbook case of hermeneutical injustice. The person has the experience: the sensory overwhelm, the social exhaustion, the feeling of being fundamentally out of step. What they lack is the concept. They do not know they are autistic. The word, the framework, the community of people who share the experience: none of these are available to them. So they reach for the concepts that are available, āIām too sensitive,ā āIām bad at people,ā āsomething is wrong with me,ā and those concepts are not just inadequate. They are actively harmful, because they locate the problem in personal failure rather than neurological difference.
Late-diagnosed adults (see Late diagnosis) describe this with striking consistency. The decades before diagnosis were not decades of ignorance. They were decades of having an experience and not possessing the interpretive resources to make it intelligible. The diagnosis does not create the experience. It provides the hermeneutical framework that makes the experience, at last, comprehensible.
The concept gap. Before terms like āautistic burnout,ā āmasking,ā āsensory overload,ā and āmeltdownā entered wider usage, the experiences they describe existed without names. An autistic person in 1990 could experience burnout, could spend decades masking, could melt down from sensory overload, and there was no concept in the shared interpretive toolkit that captured what was happening. The medical vocabulary available (ābehavioural disturbance,ā āsocial skills deficit,ā āinappropriate affectā) described the neurotypical observerās perception, not the autistic personās experience.
This is precisely Frickerās mechanism. Hermeneutical injustice occurs because the group with the experience (autistic people) has been systematically excluded from the epistemic practices (research, clinical taxonomy, educational theory) through which shared concepts are produced. The concepts that do exist were made by and for neurotypical observers. The autistic vocabulary, autistic burnout, stimming, shutdown, spoon theory, is community-generated. It emerged from autistic people talking to each other, often online, outside the institutions that should have produced these concepts but did not.
Interoceptive hermeneutical injustice. Chapman and Carel (2022) identified a form of hermeneutical injustice that operates at a particularly intimate level. If interoceptive processing differs in autism (see Interoception in autism), and if those differences mean that standard emotion vocabulary does not map onto autistic internal experience, then the injustice is not just social. It is personal. The person lacks the concepts to understand their own inner life, not because they are insufficiently introspective, but because the available concepts were built for a different kind of body.
Epistemic injustice in research
Pellicano and den Houting (2022) brought epistemic justice concerns directly into the methodology of autism research. Their review argued that the conventional research paradigm, in which non-autistic researchers study autistic people as subjects, reproduces both forms of epistemic injustice at scale. Autistic testimony is subordinated to behavioural observation (testimonial). The research questions are framed in terms that presuppose deficit rather than difference (hermeneutical). The result is a body of knowledge that reflects the neurotypical researcherās umwelt, not the autistic personās.
Den Houting et al. (2021) examined what happens when researchers attempt to correct this through participatory methods. The findings were sobering. Autistic participants and researchers valued participatory approaches but reported that power relations persisted beneath the participatory rhetoric. āNothing about us without usā is an epistemic justice claim, a demand that the people whose experience is being studied be treated as knowers, not just as subjects. But saying it is easier than doing it. Pickard et al. (2022) confirmed this: participatory autism research requires not just including autistic people but restructuring the epistemic hierarchy of the research process.
This is where epistemic justice meets the Socratic method described in Socratic inquiry and neurodivergent knowledge. The paradigm shift in autism research was not just a change in what was studied. It was a change in who counted as a knower. The autistic self-advocacy movement, the autistic academics who challenged the deficit model, the community that generated the vocabulary of burnout and masking and shutdown: these were epistemic justice interventions. They were acts of people insisting on their status as knowers in a field that had treated them as objects.
The catch-22 of autistic credibility
Chapman and Carel (2022) identified a structural problem that goes beyond individual instances of testimonial or hermeneutical injustice. The dominant framework for understanding autism creates a catch-22: autistic people are seen as either autistic or credible, but rarely both.
If a person presents as visibly autistic (stimming, atypical prosody, difficulty with eye contact), their credibility is downgraded because their autistic presentation triggers identity prejudice. If a person masks effectively and presents as neurotypical, their claim to be autistic is not believed. If an autistic person reports that they are flourishing, the report is dismissed because the framework assumes autism precludes flourishing. If they report suffering, the suffering is attributed to autism-as-pathology rather than to the environment or to the injustice itself.
The catch-22 is not a bug in the system. It is what happens when the dominant interpretive framework (the medical model, the deficit narrative) structures the conditions under which testimony is evaluated. Correcting individual instances of prejudice is not enough. The framework itself needs to change, because the framework determines what counts as credible before any individual interaction begins.
Open questions
How does epistemic injustice operate for autistic people with intellectual disability, who may be unable to provide verbal testimony in standard forms? The framework depends heavily on the concept of a āknowerā who can articulate their experience. For people whose knowing takes non-verbal or non-standard forms, the theory needs extension, not abandonment. Their epistemic marginalisation is arguably the most severe, precisely because the usual remedies (amplifying voices, participatory research, self-advocacy) are hardest to apply.
Can institutions be redesigned to reduce epistemic injustice, or does the injustice inhere in the structure of institutions themselves? Clinical settings, schools, and workplaces are built around neurotypical communication norms. Adjusting those norms is possible but requires recognising that the norms are norms, not neutral defaults, which is itself a hermeneutical achievement that many institutions have not made.
What is the relationship between epistemic injustice and masking? One reading: masking is a rational response to testimonial injustice, a strategy for avoiding the credibility penalty that visible autism attracts. If this is correct, then reducing epistemic injustice should reduce the pressure to mask. The evidence on masking decreasing in autistic-majority spaces (see Masking and camouflaging) is consistent with this, though the causal mechanism has not been established.
Implications for practice
If you assess or support neurodivergent people: the person in front of you is the primary authority on their own experience. Their testimony about what they feel, what overwhelms them, what helps and what does not, should be the starting point, not something to be overridden by clinical judgement or behavioural observation. This does not mean accepting every statement uncritically. It means beginning from a presumption of credibility rather than a presumption of deficit.
If you design research involving neurodivergent people: participatory methods are necessary but not sufficient. The question is not just āare autistic people involved?ā but āwhose knowledge counts, and who decides what questions are worth asking?ā
If you are neurodivergent and have spent years feeling that your experience was not taken seriously: epistemic injustice theory says you were right. The problem was not that you failed to communicate clearly. The problem was that the systems you were communicating with were structured to discount your testimony and to lack the concepts your experience required.
Key sources
- Fricker, M. (2007). Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press.
- Chapman, R. and Carel, H. (2022). Neurodiversity, epistemic injustice, and the good human life. Journal of Social Philosophy, 53(4), 614-631. doi: 10.1111/josp.12456
- Kidd, I.J. and Carel, H. (2017). Epistemic injustice and illness. Journal of Applied Philosophy, 34(2), 172-190. doi: 10.1111/japp.12172
- Dinishak, J. (2021). Autistic autobiography and hermeneutical injustice. Metaphilosophy, 52(2), 239-258. doi: 10.1111/meta.12514
- Dotson, K. (2011). Tracking epistemic violence, tracking practices of silencing. Hypatia, 26(2), 236-257. doi: 10.1111/j.1527-2001.2011.01177.x
- Pellicano, E. and den Houting, J. (2022). Annual research review: shifting from ānormal scienceā to neurodiversity in autism science. Journal of Child Psychology and Psychiatry, 63(4), 381-396. doi: 10.1111/jcpp.13534
- Den Houting, J., Higgins, C., McLeod, K., Pellicano, E. and Stears, M. (2021). āIām not just a guinea pigā: academic and community perceptions of participatory autism research. Autism, 25(4), 1098-1113. doi: 10.1177/1362361320951696
- Pickard, H., Pellicano, E., den Houting, J. and Crane, L. (2022). Participatory autism research: early career and established researchersā views and experiences. Autism, 26(5), 1247-1260. doi: 10.1177/13623613211019594
- Uisma, R., Pellicano, E., Skafle, I. and Dahl, T.H. (2026). Misunderstood, minimised, misrepresented: autistic young adultsā experiences of epistemic injustices in healthcare interactions. Sociology of Health & Illness. doi: 10.1111/1467-9566.70124
- Kadodia, Z. and Krueger, J. (2026). Epistemic injustice, niche construction, and neurodiversity. Philosophical Psychology. doi: 10.1080/09515089.2026.2657453