Epistemic Injustice Through a Social Lens — Epoche C1
What the disagreement is about When Carmita Wood resigned from her job at Cornell University in 1975 because of a supervisor's conduct and was refused unemployment benefit, the obstacle was not that nobody believed her but that there was no available word for what had happened to her; the term "sexual harassment" was coined at a meeting she helped organise shortly afterwards. Miranda Fricker uses that case in Epistemic Injustice: Power and the Ethics of Knowing (2007) to introduce a kind of wrong done to someone specifically in their capacity as a knower, and this essay is about whether her account of what puts that wrong right is adequate. The claim to be defended is that it is not — that Fricker's diagnosis is social while her remedy is individual, and that the gap between them is not a matter of emphasis but of what testimonial capacity is. Fricker distinguishes two forms. Testimonial injustice occurs when a hearer assigns a speaker less credibility than the speaker's word deserves, because of a prejudice attaching to the speaker's social identity: the hearer discounts the testimony not on the evidence but on the identity. Her illustrations are literary — Tom Robinson in Harper Lee's To Kill a Mockingbird , whose testimony an all-white jury cannot hear, and Marge Sherwood in Patricia Highsmith's The Talented Mr Ripley , whose accurate suspicions are put down to female intuition. Hermeneutical injustice occurs when a gap in the collective resources for interpreting experience leaves someone unable to make sense of a significant part of their own life, and that gap exists because they and people like them were excluded from the practices in which social meanings get made — what Fricker calls hermeneutical marginalisation. Wood's case is of the second kind. One feature of Fricker's framework worth noticing at the outset, because it shapes everything after, is that she denies that credibility is a good in finite supply. Nobody's credibility has to fall for another's to rise. This is why she treats epistemic injustice as something other than a problem of distribution, and why the corrective she proposes is a change in how hearers judge rather than a reallocation of a scarce resource. What Fricker actually says, and what the criticism can fairly target Before pressing the objection, the target has to be described accurately, because the version of it that circulates overstates the case and is easy to refute. It is not true that Fricker frames both injustices as failures of individual virtue. Hermeneutical injustice, on her own definition, has no perpetrator at all: the gap in interpretive resources is a structural fact, and she is explicit that it can wrong someone without anyone doing anything wrong. Nor does she treat testimonial injustice as merely incidental. She distinguishes incidental cases, where a prejudice happens to bite once, from systematic ones, where the same identity prejudice tracks a person through employment, law, medicine and politics, and it is the systematic kind she takes to be central. The diagnosis, then, is thoroughly social. What is individualistic is the remedy. Fricker's corrective for testimonial injustice is the virtue of testimonial justice: a hearer who has cultivated it notices the prejudicial influence on their own credibility judgement and revises upwards to compensate. This is a demanding and genuinely useful proposal, and the criticism to be made of it is not that individual virtue is worthless. It is that a remedy pitched at the level of the individual hearer's judgement cannot reach a wrong whose mechanism operates at the level of what hearers have available to judge with. Elizabeth Anderson made this argument directly in 2012, holding that epistemic justice must be understood as a virtue of social institutions and not only of persons, on the grounds that many credibility deficits arise from institutional arrangements — segregated schooling, unequal professional networks, the absence of certain groups from the bodies that certify expertise — which no amount of individual vigilance dissolves. A hearer cannot correct for a prejudice by revising a judgement upwards if the arrangement that produced the deficit has already determined which speakers reach them at all. Why testimonial trust cannot be an individual achievement The stronger version of the objection is not about remedies but about constitution, and it rests on an argument in the epistemology of testimony that predates the epistemic injustice literature. The undergraduate position is reductionism : my belief on someone's say-so is justified because I have inductive evidence that testimony of that sort is generally reliable, so testimonial justification reduces to ordinary empirical justification. C. A. J. Coady's Testimony: A Philosophical Study (1992) shows that this cannot work as stated. To acquire the inductive base — to have checked that reports of a given kind tend to be true — an individual would have to establish a great many correlations between what people said and what was the case, and no individual has done anything of the sort. Worse, the checking itself relies on testimony at almost every step: on what was said, on when, on what the instruments recorded, on the reliability of memory as reported by others. The general reliability of testimony cannot be established from a standpoint that does not already presuppose it. The step up from this to the present argument is short. If the entitlement to rely on others is not something an individual assembles from private evidence, then the conditions under which that entitlement holds are not conditions inside the individual. Sanford Goldberg develops this in Relying on Others (2010), arguing that the reliability of a testimonial belief is a property of a process that extends beyond the hearer's mind to include the speaker's cognitive processes and the social arrangements that make speakers reliable — credentialing, correction, sanction for