Miranda Fricker's excellent monograph occupies some relatively uncharted philosophical territory, being ‘neither straightforwardly a work of ethics nor straightforwardly a work of epistemology’, but instead seeking to ‘[renegotiate] a stretch of the border between these two regions’ (p. 2). Fricker wants to release us from the sceptical impasse induced by a strain of late twentieth-century postmodernist thought which, having arrived at an ostensibly critical consciousness of ‘reason's entanglements with social power’ (p. 3), proceeds to incapacitate itself for any genuine critique through its suppression of the ‘distinction between what we have a reason to think and what mere relations of power are doing to our thinking’ (ibid.) By salvaging this distinction, we can reconnect the project of ‘socially situated’ epistemology with the guiding values of the critical theory tradition, and equip ourselves to investigate in concrete terms ‘how our epistemic conduct might become at once more rational and more just’ (p. 4).

One of the book's leading ideas is that ‘epistemic psychology’ can usefully shift its focus from the successful acquisition and sharing of knowledge to a study of some of the things that tend to go wrong in the course of these activities, thereby moving into the ‘negative space’ (p. viii) defined by certain kinds of injustice. Attention can then be given to the (corrective) virtues needed to resist such injustice, a resistance that will take a variety of socially determinate forms, depending on the context of the knowledge-gathering enterprise. Our incentive to cultivate the relevant virtues is an ethical one, since it concerns the avoidance of wrongdoing towards other persons, or the promotion of their welfare, in their capacity as knowers; but it is also epistemic, since it concerns the care of our own powers of arriving at truth.

Fricker recognizes two subdivisions of epistemic injustice, the ‘testimonial’ and the ‘hermeneutical’. The former, which is discussed at considerably greater length, is a wrong ‘inflicted individual to individual’ (p. 168), whereas the latter is a ‘purely structural notion’ which ‘involves no culprit’ (p. 159). Testimonial injustice (TI), in its central case, consists in the infliction on certain people of an ‘identity-prejudicial credibility deficit’ (p. 4), meaning that because of entrenched prejudices about the sort of person they are (e.g. ‘identity prejudices’ concerning women or ethnic minorities, which affect the way members of these groups are perceived as individuals), the community at large systematically undervalues their testimony as an aid to acquiring new knowledge; that is, it is less disposed than it ought to be to believe what they say, or to take them seriously as informants. (Such prejudices may operate more at the imaginative - or imaginary - level than at that of conscious belief.) ‘Hermeneutical injustice’ (HI), by contrast, is ‘the injustice of having some significant area of one's social experience obscured from collective understanding owing to a structural identity prejudice in the collective hermeneutical resource’ (p. 155) - such obscurity being of the kind addressed, for example, by the women's liberation movement of the 1960s, to which we owe the (now thoroughly familiar) concept of ‘sexual harassment’. Fricker's point here is that one of the disadvantages attaching to membership of a subordinate group is an expressive poverty resulting not just from cultural deprivation in the abstract, but from the vested interest of the powerful in not having a name put to certain quite specific social phenomena. Again, the interest of this thought lies in the way it can direct our attention to determinate areas of conflict in the representation of social experience. ‘Hermeneutical lacunas are like holes in the ozone - it's the people who live under them that get burned’ (p. 161), and who therefore have an incentive to try to remedy the (ideologically charged) obscurantism of ‘normal’ social discourse.

Several of the case-studies used to exemplify TI and HI are drawn in highly engaging style from literary texts: for TI, Harper Lee's To Kill a Mockingbird (1960) and Anthony Minghella's screenplay The Talented Mr Ripley - Based on Patricia Highsmith's Novel (2000); for HI, Ian McEwan's Enduring Love (1998) and Edmund White's A Boy's Own Story (1983). The texts for TI are concerned with refusal to grant an individual the credit due to him or her as a source of insight into some question of fact (what actually happened in a certain judicially contested incident; what has become of a certain missing person?) Philosophers will note that by a reasonably smooth extension of the basic account, we may be able to bring into view the prejudiced reception of ‘testimony’ which is not (or not simply) about questions of this sort, but also involves the organization of factual information into discursive argument or theory. Fricker envisages this extension in a footnote (p. 50, note 22), but does not elaborate on it - an omission that might be regarded as disappointing, given that the accompanying text describes what sounds like a deplorable bout of student-on-student intellectual bullying suffered by Simone de Beauvoir at the hands of Sartre. However, her more fundamental concern is with events or processes occurring further upstream in the activity of knowledge-gathering. Drawing inspiration from the work of Edward Craig and Bernard Williams, she tells a ‘state of nature’ story designed to lay bare the collective needs attributable a priori to rational agents in their capacity as enquirers: a need for correct practically relevant information, obviously, but also (and consequentially) for participation in ‘an epistemic practice whereby information is shared or pooled’, and hence for the encouragement of ‘dispositions in individuals that will stabilize relations of trust’ (pp. 110, 111). Through apprenticeship in a social practice directed towards these ends, we may hope to develop (as part of a larger stock of epistemic virtues) a ‘testimonial sensibility’ analogous to the ethical sensibility of the virtuous agent - in fact, to qualify as ‘virtuous hearers’ attuned to ‘epistemically salient features of the situation and the speaker's performance’ (pp. 71, 72), namely those features relating to the speaker's probable sincerity and competence with regard to the matter at hand.

But a job that can be done well can also be done badly, and this is where we encounter the vice of TI - an attitude which is unjust both to the objective value of the information on offer, and to the person who offers it; for in TI ‘one person undermines another's status as a subject of knowledge’, just as ‘in Kant's conception of immorality, one person undermines another's status as rational agent’ (p. 136). There is no equivocation here on the word ‘injustice’, for ‘the two sorts of wrong are closely intertwined’ (ibid.); indeed, testimonial justice can properly be regarded as a ‘hybrid’ virtue, dependent on context to determine whether its intellectual (truth-directed) or its ethical aspect is paramount. Either way, its individuating motive is ‘to neutralize prejudice in one's credibility judgement’ with regard to others (p. 122). Both testimonial and hermeneutical justice are essentially corrective virtues (pp. 118, 169), enabling us to supplement the normal ‘unreflective yet critical alertness’ of the good-enough interlocutor (p. 80) with a ‘reflexive social awareness’ (p. 170) of likely misjudgements or blind spots.

Fricker's literary case-studies provide ample reason for gloom at the reflective level, but she steers clear of any stern moralism over past failures of testimonial sensibility. One may even feel she is inclined to be over-charitable in her assessment of when individuals are to blame for their defective ‘credibility judgements’. This question turns, for her, on a contrast between those judgements which are ‘routine’ and those which are ‘exceptional’ in a given historical setting: wishing, understandably, to avoid ‘the hubris of deeming [people] blameworthy for actions not routinely regarded as wrong in their culture’, she claims that ‘One cannot be blamed for making a routine moral judgement’ (p. 105). Thus Herbert Greenleaf in The Talented Mr Ripley - unsuspecting father of the unfortunate Dickie Greenleaf, who has been murdered and had his identity stolen by Tom Ripley - is not to blame for his condescending attitude to the so-called ‘feminine intuition’ of Dickie's girlfriend Marge Sherwood; the reason being that this (obviously flawed) sexist attitude was ‘routine’ for the 1950s, the historical setting of the novel, though less benighted views on gender were ‘just around the corner’. It is hard to disagree with Fricker's guiding principle in the allocation of blame for defective credibility judgement, which is a version of the ‘ought implies can’ doctrine: that is, we must allow that not all morally necessary ideas are equally accessible to everyone at every moment and that our guilt, if any, for failing to grasp them depends on the extent of their accessibility to us in our own time. On the other hand, it is important to recognize the contentiousness of the relevant historical claims about accessibility. Fricker, for example, suggests that although Greenleaf senior is at fault in his attitude to Marge, he and people like him ‘were not culpably at fault until the requisite critical consciousness of gender became available to them’ (p. 100 - emphasis in original); and that although Greenleaf has a reason to doubt his distrust of Marge's word, ‘his historical situation is such that he cannot access that reason’ (p. 101). But here we may ask: isn't this to let Greenleaf off too lightly for his failure to be influenced by the huge advance in ‘critical consciousness of gender’ that took place in, say, the hundred years before the events of the novel, to look no further back? (How would he account for the fact that American women now have the vote and can go to university?) Paradoxically perhaps, this leniency may mask the effect on our thinking of older strata of politically motivated TI; for our conception of the 1950s as a period when the ‘requisite critical consciousness of gender’ was unavailable to someone like Greenleaf may reflect a long-term tendency to dismiss the feminism of the late nineteenth and early twentieth centuries as something lightweight, extremist, and (despite the achievement of female suffrage!) not fully integrated into the ethical mainstream. Here as elsewhere, the extent to which we are disposed to accept past instances of moral stupidity as forgivable or understandable in their historical context will presumably depend on how we draw up the inventory of voices to which, at the time, it was intellectually obligatory to listen.

Fricker concedes that ‘even Greenleaf could have judged exceptionally’ (p. 105), but argues that his failure to do so does not warrant anything as strong as blame (since his moral outlook was ‘routine’ for his own cultural setting). But here the contrast between ‘routine’ and ‘exceptional’ moral judgements seems to be taking on a clarity that it does not really display in practice. To change the example, someone today impressed by the perils of global warming and general environmental degradation might decide that ‘it is no longer morally legitimate to travel by air unless absolutely necessary’. Statistically, I suppose this would be an ‘exceptional’ judgement (in that few people seriously entertain it); the ‘routine’ judgement, which of course receives massive support from government policy, would be that flying is perfectly all right. But it would be misleading to say that the ‘more exceptional alternative is, as a matter of historical possibility, just around the corner’ (p. 105 again), because actually it is already here (since all you have to do is read some reasonably intelligent newspaper: after all, Fricker warns us that we need to ‘resist thinking of a culture's moral discourse … as a finite monolith’, p. 104). And although this issue gives rise to well-known hazards of priggishness (or ‘hubris’, as she puts it), I still think we should hesitate to conclude that whatever is morally ‘routine’ is, as such, not blameworthy. One might argue, rather, that we need to learn a greater willingness to blame ourselves for collaborating with ‘routine’ wickedness or craziness, and to change our behaviour accordingly; and if I take that view of myself I can hardly think it just doesn't apply to others.

This reservation, however, has to do with the application rather than the substance of Fricker's theory. Taken as a whole, her discussion is outstandingly lucid and persuasive; it is also much richer than I have been able to indicate here, for the book is an admirable reminder of what can be accomplished in under two hundred pages of crisp yet free-flowing philosophical prose. It deserves, and will surely command, widespread attention.