Varieties of Testimonial Injustice

Authored by: Jeremy Wanderer

The Routledge Handbook of Epistemic Injustice

Print publication date:  April  2017
Online publication date:  March  2017

Print ISBN: 9781138828254
eBook ISBN: 9781315212043
Adobe ISBN: 9781351814508




Suppose that my testifying at a disciplinary tribunal leads to my unfair dismissal. This would be an instance of injustice that is the result of an act of testimony, but not a case of testimonial injustice in the sense under investigation in this essay. Why not? Because we are interested here in investigating the possibility that there is a categorical connection between certain instances of injustice and the social practice of testimony, where the connection is such that an understanding of the injustice in these specific instances emerges from a consideration of the general practice. It seems clear, in contrast, that one can understand the injustice of my unfair dismissal without any consideration of the social practice of testimony whatsoever.

 Add to shortlist  Cite

Varieties of Testimonial Injustice

Suppose that my testifying at a disciplinary tribunal leads to my unfair dismissal. This would be an instance of injustice that is the result of an act of testimony, but not a case of testimonial injustice in the sense under investigation in this essay. Why not? Because we are interested here in investigating the possibility that there is a categorical connection between certain instances of injustice and the social practice of testimony, where the connection is such that an understanding of the injustice in these specific instances emerges from a consideration of the general practice. It seems clear, in contrast, that one can understand the injustice of my unfair dismissal without any consideration of the social practice of testimony whatsoever.

A glimpse of what is meant by the proposed categorical connection can be had through a preliminary characterization of its key terms, one that will be further developed (and modified) in the ensuing discussion.

In the broadest terms, a social practice is a norm-governed pattern of meaningful interaction between people. Participants within a social practice interact with each other through their socially significant performances and hold each other responsible for performing according to the norms integral to such a practice. Such practices are internally diverse and typically include a wide variety of performance-kinds that are recognized by its practitioners. From an explanatory point of view, not all performance-kinds within a practice need be equal; some performances can be seen as more central to the practice, in that they are capable of explaining other performances within the practice. For example, a performance-kind within a practice could be viewed as deviant if it can be seen as subverting other performance-kinds thereby deemed more central to the practice.

The social practice of interest here is that of testimony, an internally diverse pattern of norm-governed interactions that includes the speech act of telling someone something close to its explanatory center. Related performances-kinds within this practice include acts such as querying, challenging, listening and conjecturing, as well as ‘pathological’ performance-kinds such as lying, gossiping and rumor-mongering. 1 (A fuller characterization of this practice, especially its relationship to epistemic matters, will emerge in the ensuing discussion). As with all social practices, the social practice of testimony will take on thicker and culturally-specific forms in different places, times and contexts, including their different functioning in regulated institutional public fora (such as legal settings) and more relaxed, informal everyday exchanges.

It is the social practice of testimony as a whole that stands in a categorical connection to certain cases of injustice, i.e. cases of testimonial injustice. The term injustice here is used very broadly to include any instance in which a person is maltreated, and is not limited to just those cases involving the unfair distribution of goods or capacities, nor to cases in which someone is denied what is their due. Our focus is thus on those cases in which a person is maltreated and where a full understanding of what makes this an instance of maltreatment will invoke the social practice of testimony. In such a case, one can say there is a categorical connection between the social practice and the maltreatment. Think away the social practice, perhaps by imagining a social grouping that has no such practice, and one thinks away the possibility of cases of testimonial injustice arising within such a social grouping.

It is possible to describe the social practice of testimony from a variety of different perspectives, with each perspective bringing a different variety of injustice into view. In what follows, I distinguish three such perspectives and suggest that each sheds light on a different variety of testimonial injustice – transactional testimonial injustice, structural testimonial injustice and testimonial betrayal. This should not be taken to imply that these varieties exhaust the category of testimonial injustice, nor is the order of presentation of these varieties intended to convey their relative importance. Rather the suggestion under investigation here is that these three are distinct varieties of testimonial injustice, in that they are distinguishable from each other in ethically significant ways, while still standing in a categorical connection with the social practice of testimony. We will begin by outlining a much-discussed example of testimonial injustice (Part 1), and explore this same example from three differing perspectives in the remaining sections (Parts 24).

1  Fricker’s account of testimonial injustice

Let us begin with what has become the locus classicus for contemporary Anglophone discussions of testimonial injustice, Miranda Fricker’s path-breaking book Epistemic Injustice, 2 with the aim of relating her work to the general characterization of testimonial injustice just noted.

According to Fricker’s seminal account, testimonial injustice occurs when, following an act of telling someone something, a Speaker is accorded insufficient credibility by a Hearer due to a prejudicial stereotype held by the Hearer. 3 More specifically, the Hearer subscribes, at some cognitive level and in a manner that is relatively resistant to counter-evidence, to a widely-held belief that links a salient aspect of the Speaker’s social identity to a disparaging attribute, as a result of which the Hearer accords the Speaker less credibility than she deserves. 4 One of Fricker’s signature examples is a critical moment towards the end of Anthony Minghella’s screenplay ‘The Talented Mr. Ripley’, where Herbert Greenleaf dismisses the testimony of Marge Sherwood by saying “Marge, there’s female intuition, and then there are facts.” 5 Drawing on a prejudicial stereotype that female opinion is more often the result of emotion rather than intellect, Greenleaf accords Sherwood insufficient credibility and thus does not take her testimony seriously.

What makes this a case of testimonial injustice, as opposed to an injustice that is the result of an act of testimony? This query needs to be handled with some care as Greenleaf’s prejudicial orientation could indeed lead to many instances of maltreatment of Sherwood and others for which we would not need to invoke the social practice of testimony by way of understanding. For example, we could view this as an instance of the prejudice unfairly thwarting Sherwood’s ambitions or simply failing to respect her, neither of which need involve appeal to the practice of testimony per se. What then does consideration of the social practice of testimony add to our understanding of the injustice in this case?

Understanding Fricker’s response requires a detour through the distinctive account of the relationship between the social practice of testimony and the concept of knowledge that she derives from the work of Edward Craig. 6 Craig tells a State of Nature story involving a stripped-down social community, whose agents lack our epistemic concepts such as ‘knowledge’ but share certain needs and cognitive resources with us. Specifically, they need true beliefs about the environment, which can be had either through their own on-board sources (e.g. via perception) or from other informants in a community who have a positional advantage over themselves. Imagine a community member, an inquirer, who does not yet have a belief about p but wants to acquire one from a prospective informant, and consider what general properties this inquirer would want their informant to have. On Fricker’s retelling, the desired properties are threefold: the inquirer wants an informant who: i) is likely enough in the context to be right as to whether p, ii) is willing and able to convey whether p to the inquirer, and iii) bears indicator properties such that the inquirer can recognize the presence of the first two features.

In such a minimal epistemic community, we can easily imagine inquirers introducing a concept whose function it is to flag good informants, i.e. informants with just that trio of properties indicated. On the one hand, the concept of a good informant is not the same as our concept of a knower: we can, for example, easily think of cases in which we would treat a person as a knower but not as a good informant – such as my son who damn well knows where the remote control is but will not divulge. On the other hand, we can see the concept of a knower emerging from the concept of a good informant via an inevitable process of objectification. The notion of a good informant is initially tied to the inquirer’s own powers of recognition; it is a label attached to someone who is not only capable of, and willing to, answer the inquirer’s query whether p, but whom the inquirer can recognize as so capable and willing. The process of objectification then works to extend the idea of a good informant beyond something that the inquirer can recognize in the here and now. The inquirer will come to realize that there can be good informants whose indicator properties they cannot identify, and that there can be a good informant as to whether p, even though they have no current interest in answering the question or even if no one has any current interest in answering the question. The result is an objectivized conception of the good informant, the knower, who is now no longer tied to the particular relationship of inquiry between inquirer and informant, yet is involved in the social practice of pooling information.

According to Craig, then, the concept of a good informant is a response to certain identifiable epistemic needs in the State of Nature, and the concept of a knower is a response to further identifiable needs and everyday processes operating on the concept of a good informant. The notion of being a knower, and with it the social practice of testimony, thus emerges, of necessity as it were, from the practice of identifying good informants. On this way of thinking, there is a close conceptual tie between knowledge and testimony, one that rules out the possibility of being able to comprehend what knowledge is but yet have difficulty in grasping how one may come to acquire that knowledge via the say-so of another. As it has been provocatively put elsewhere, “the concept of knowledge contains an account of the possibility of knowing from others”. 7 The concepts of the knower and the testifier appear as one in the newly emerged practice of pooling information.

An actual performance within the social practice of pooling information will involve agents who are socially-situated, including each having specific social identities and standing in various power relations to one another. For some theoretical purposes, however, it is useful to abstract away from socially-situated subjects and to consider the practice as involving agents who are not conceived as standing in relations of social identity and power. Since Fricker’s focus is on cases of prejudice, it may seem obvious that the relevant description of the social practice for bringing out an understanding of testimonial injustice is the situated one. This suggestion, however, is only partly right. What is right is the contention that the cases of injustice that we hope to understand through a consideration of the practice of testimony will only come into view if we consider agents who are socially-situated. It does not, however, follow that the insight into these cases cannot be provided by a consideration of the social practice of testimony in the abstract. 8

As Fricker’s extended discussion of Craig’s State of Nature reveals, an important feature of her account is the contention that an understanding of the distinctive character of testimonial injustice emerges from a consideration of the social practice of pooling information, abstractly conceived. As she puts it, “[g]iven Craig’s account, what the recipient of a prejudicial credibility deficit is excluded from is the single practice that dramatizes the origin of what it is to be a knower at all”. 9 So, when Greenleaf’s prejudices lead him to dismiss Sherwood’s testimony, he maltreats her in various ways – including excluding her from the social practice of pooling information. The categorical connection between the social practice of pooling information and the injustice in this instance is what makes this a case of testimonial injustice as opposed to an injustice connected to an act of testimony. Think away the social practice of pooling information, such as by imagining a State of Nature community that has no such practice, and one thinks away this injustice.

Following Fricker’s lead, the case of Sherwood and Greenleaf will provide a useful focus for our own investigation into the possibility of there being a categorical connection between certain instances of injustice and the social practice of testimony. That said, as with all the primary examples of testimonial injustice that Fricker considers in her book, the Sherwood-Greenleaf example is complex, involving the interplay between different features worth distinguishing. One such feature is that the case involves a direct linguistic interaction between Speaker and Hearer, with one explicitly rejecting the testimony proffered by the other. A second feature is that, by invoking sexist prejudicial stereotypes, the case involves reference to socio-historically extended patterns of oppression and injustice that range far beyond this particular interaction alone. A third feature is that the case involves an intense personal relationship between the protagonists. Rather than lump these three features together, our investigation here will consider each in turn (in the next three sections). This is because each feature provides a different perspective on the social practice of testimony, and it is by considering these perspectives separately that the possibility of there being different varieties of testimonial injustice comes into view.

2  Transactional testimonial injustice

Theorists interested in the notion of justice have distinguished between two different ways of approaching their subject matter, either by focusing on the normative character of acts of exchange between people or on the distribution of relevant goods. 10 According to the former, an exchange is just when it lives up to certain normative standards governing exchanges of that type, including both generic norms governing all such exchanges and/or norms specific to the particular kind of exchange under consideration. According to the latter, a structure is just when each person in the structure gets their due in terms of an appropriate distribution of relevant goods. In both cases, injustice involves a certain arbitrariness, either the arbitrary rule of one party of the exchange over the other in the former case (the proverbial ‘foot on the neck’) or an arbitrary distribution of the relevant good in the latter (the proverbial ‘disproportionate slice of the pie’). Let us call the former transactional injustice and the latter structural injustice. Although one could debate the possible relations of dependence between the transactional and structural, I propose here to treat these as two potentially independent varieties of injustice, varieties that come into view depending on what aspects of the abstract social practice one focuses on: transactional episodes between practitioners or the distribution of goods within the system at a given time.

As the exchange between Greenleaf and Sherwood exemplifies, Fricker’s primary examples of testimonial injustice all involve exchanges between people, leaving the strong impression that her main focus is on the transactional variety of testimonial injustice. Granted, she explicitly elects to concentrate on what she dubs ‘systematic testimonial injustice’, i.e. on “those prejudices that ‘track’ the subject through different dimensions of social activity – economic, educational, professional, sexual, legal, political, religious, and so on”. 11 Such systematicity, however, merely serves to ensure that testimonial injustice is not isolated by linking it to other kinds of actual or potential injustice and does not involve a structural variety of testimonial injustice. 12 In this section, I will develop the idea of transactional testimonial injustice in a manner that is inspired by Fricker’s use of Craig’s framework as outlined above; in the next section I will evaluate Fricker’s reasons for the apparent focus on the transactional variety.

A transaction, as used here, is a special kind of social act. 13 Not only does the very idea of the act implicate more than one person, but the act has a structure that serves to place those persons in a particular relation to one another, such that the act is done from one (agent) towards the other (patient). Any act with this structure can be described in two ways, one from the perspective of the agent and one from the perspective of the patient; self-conscious awareness of this structure by both parties is part of the very idea of the act itself. A successful instance of a transaction serves to yoke the two parties together in a distinctive order of justice, such that an inappropriate response of either party to the act counts as an injustice done towards the other. 14 In our case, the transaction that Fricker focuses on is that of testimony, i.e. the act of telling someone something. The act implicates two people, such that in the successful case, one person is self-consciously aware of telling someone something, and the other is self-consciously aware of being told something by someone. In a successful case of testimony, self-conscious recognition by both parties of the transaction creates an order of justice between them, one delineated by a certain set of appropriate actions in response by both parties under certain conditions, and testimonial transactional injustice occurs when one party responds inappropriately under some of those conditions.

The relevant transaction-type involved in testimonial transactions has thus far been described interchangeably as the act of ‘testimony’ or ‘telling someone something’. Insight into this transaction-type can be had by extending Craig’s account of the emergence, via objectification, of the practice of pooling information. In the minimal epistemic community of the State of Nature, inquirer and informant are acquaintances who both recognize each other and are responsive to each other’s immediate needs. The same process of objectification that extends the idea of a good informant beyond the inquirer’s immediate recognitional powers (as described by Craig) also serves to create an addressed act that extends beyond the informant’s immediate recognitional powers as well, such that the relationships between participants in the emerging social practice of pooling information cease to be that between acquaintances. Testifying is an addressed speech act (as opposed to the mere verbal expression of thought), but the audience to whom it is addressed is potentially anyone and everyone. Thus, the process of objectification that brings about the concept of a knower from the concept of a good informant also gives rise to a distinctive kind of impersonal yet interpersonal speech act, the speech act of testifying, and it is the presence of this act that ensures that the practice in question can be characterized as that of pooling information.

This characterization of the speech act of testimony is a departure from Fricker’s own characterization, which limits the audience to whom the act of testimony is addressed to those acquaintances intended by the Speaker. 15 Neither characterization denies that a Speaker can target her speech act towards fulfilling the perceived epistemic needs of an already familiar audience (Hearers) and not others (Over-Hearers). Fricker takes this notion of targeting as defining the scope of those addressed by the act. On our characterization, in contrast, the scope of those addressed depends on the structure of the act of which both Speaker and Hearer are self-consciously aware, and does not depend on those targeted by the Speaker. Of course, even on our characterization, acts of testimony typically take place in the context of the perceived epistemic needs of extant relationships, and the Speaker is not absurdly portrayed as making testimonial pronouncements to the universe as it were. Nevertheless, in thinking of the transaction as testimonial, both parties are sensitive to the fact that the scope of those addressed extends beyond those targeted by the Speaker. Whilst it may be contextually impolite for an Over-Hearer to query or challenge the Speaker, the Over-Hearer too is included in the transaction, and this fact is sometimes exploited to create a heightened sense of intimacy between Speaker and Hearer by, for example, whispering in a way that extrudes the Over-Hearer from fully participating in a testimonial transaction to which she is party (on our characterization).

Fricker’s characterization, one that focuses on a relationship of acquaintance between informer and inquirer, is surprising given her own understanding of the social practice of testimony as pooling information. ‘To pool’ used as a verb denotes the cooperative activity of putting some good into a common fund. If the common fund into which the information is pooled by an act of telling is limited to that between acquaintances alone, as implied by Fricker’s characterization of the act of telling, we lose the advertised link between knowing and telling. We will thus persist with our characterization of the act of telling, one that treats informant and inquirer as bound together in an order of justice established by a transaction, all the while remaining possible strangers to each other, in that both potentially stand outside each other’s sphere of recognition.

However the act is best understood, Fricker’s own discussion of the social practice of pooling information as it is relevant to testimonial injustice concentrates on this one transaction alone, the speech act of testifying. This limited focus, as I understand it, stems simply from the explanatory centrality of the act of telling someone something in characterizing the social practice and does not reflect a parallel limitation of the notion of testimonial injustice to cases involving testimony alone. The social practice of pooling information extends beyond the speech act of telling someone something, and includes an array of different activities, including those related to the activities of inquiring, questioning, discussing, speculating and deliberating, as well as the myriad of thicker and culturally-specific forms that such activities take in differing contexts. Consider the following example from Christopher Hookway, in which a teacher, due to a prejudicial stereotype, does not take seriously the questions put forward by a particular student in the course of a classroom discussion. 16 Hookway argues that here we have a case of injustice akin to those Fricker discusses, even though the teacher’s maltreatment of the student does not involve the attribution of insufficient credibility to the student following an act of telling. On the reading proffered here, this is a case of what Fricker would call testimonial injustice even without an act of testimony, since it involves maltreatment of someone that can be explained by appeal to the social practice of testimony, broadly construed to include the activity of participation in discussion from which the student is excluded.

This is admittedly an expansive conception of the social practice of testimony. It is not, for example, limited to what Hookway calls an “informational perspective” in which the relevant “interactions are all concerned with the possession and transmission of information or knowledge from one person to another” but also includes what he terms a broader “participant perspective” that is “involved when we think of [practitioners] as trying to participate in activities such as discussion, inquiry, deliberation, and so on.” 17 That said, the conception is not so broad as to conflate the testimonial with the epistemic, and there are forms of epistemic inquiry involving the say-so of others that do not fall within the social practice of testimony even on this expanded conception.

Consider forms of inquiry in which the Hearer gains knowledge from the Speaker by treating their say-so as evidence of a belief in much the same ways as he may treat a yelp as evidence of predator-proximity. The Hearer may indeed gain knowledge via such an inquiry, but this is not achieved by treating the Speaker’s say-so as the undertaking of a socially significant performance towards the Hearer in the context of a norm-governed transaction. 18 Craig’s way of marking this contrast is in terms of a distinction between somebody’s functioning as a ‘good informant’ and their functioning merely as a ‘source of information’. As Craig explains, “What I have in mind is the special flavor of situations in which human beings treat each other as subjects with a common purpose, rather than as objects from which services, in this case true belief, can be extracted”. 19 There are many ways of developing this pregnant contrast, 20 but one way of understanding the special flavor of treating someone as a good informant is in terms of the Hearer entering into an order of justice with the Speaker through transacting with her, as opposed to merely treating her utterance as evidence for something. Contra Hookway, the key contrast is thus not between the informational and participant perspectives, but between the deliberative perspectives of the potentially solitary inquirer extracting information from objects and the participant in the social practice of testimony via engagement in transactions such as telling/being told something to/by someone.

Here, then, is our first variety of testimonial injustice: transactional testimonial injustice. The injustice is brought into view by focusing on a breach of the order of justice established between the parties to a testimonial transaction. If the source of the breach is one that yields a systemic pattern of maltreatment of people across an array of different transactions and contexts, such as the widespread maltreatment that results from the sexist prejudices exhibited by Greenleaf, then the injustice may well be termed ‘systemic transactional testimonial injustice’.

3  Structural testimonial injustice

At the beginning of the previous section we distinguished between transactional and structural varieties of injustice and noted that Fricker’s work on testimonial injustice tends to focus on the former. It is not that Fricker is unaware of the possibility of there being a structural variety. Indeed, she begins her book by flagging a way of thinking about epistemic injustice in terms of the “distributive unfairness in the distribution of epistemic goods such as information or education. In such cases we picture social agents who have an interest in various goods, some of them epistemic, and question whether everyone is getting their fair share”. 21 This suggestion is, however, flagged by Fricker only to be dismissed, since “there is nothing very distinctively epistemic about [this way of thinking of injustice] for it seems largely incidental that the good in question can be characterized as an epistemic good”.

What does Fricker here mean by labelling this ‘incidental’? Two possibilities suggest themselves. A more substantive reading takes this as dismissive of the very idea of structural testimonial injustice: even if one can talk of injustice in terms of an inappropriate distribution of goods such as information or education, the injustice does not stand in a categorical connection to the social practice of testimony. A less substantive reading is to take this as conceding that there is a structural variety of testimonial injustice but to dismiss its relevance to her overall project. In what follows, I aim to reject the more substantive reading and to offer a qualified defense of the less substantive one. 22

To see why one should reject the more substantive reading, consider the following example, due to Elizabeth Anderson, in which a Hearer accords insufficient credibility to the Speaker’s say-so on a matter that requires an educated judgment, and the Hearer’s primary reason for according a low level of credibility is the Speaker’s use of non-standard grammar in formulating their judgment. 23 The context here is crucial: Speaker and Hearer live in a society in which certain social groups are systematically deprived of decent educational opportunities, and the Hearer takes the use of standardized grammar as a marker of education, and hence of credibility. As Anderson points out, there is indeed a significant injustice here, but it is not a transactional testimonial injustice. Assuming the Hearer is not drawing on a prejudicial stereotype that is resistant to counter-evidence in making his credibility judgment, then it is possible that the Hearer is following a sound epistemic procedure that could even be justified to (and perhaps shared by) the Speaker herself. The injustice here is thus not transactional but structural, viz. the unfair distribution of certain epistemic goods within the society. As Anderson puts it, “[a]n original structural injustice – denial of fair opportunities for education – generates additional structural inequalities in opportunities for exercising full epistemic agency, which is an injustice to the speakers”. 24

A proponent of the more substantive reading need not deny that there is an injustice here, though will reject the contention that insight into the injustice is afforded by the social practice of testimony. Anderson’s point, however, is that an understanding of the injustice here, as well as ways of overcoming the injustice, requires broadening our conception of the social practice of testimony itself, from a narrower focus on transactions alone to a broader one that incorporates the institutions within which such transactions have their home. In the example above, unfair distribution of institutional markers of credibility undermines the possibility of some to participate in the practice of pooling information, a maltreatment of those persons that would remain out of view if we merely focus – as Fricker does – on the normative standing of testimonial transactions between participants. In other words, insight into the injustice in this case is afforded by the social practice of testimony, since the injustice of a skewed distribution of credibility markers can only be understood by reference to “the single practice that dramatizes the origin of what it is to be a knower at all”. 25 Thus it is a case of testimonial injustice in the sense under investigation in this essay, albeit one that requires a focus on structural aspects of the social practice, as opposed to a focus on transactions alone. We should therefore reject the more substantive reading and endorse the claim that there is a structural variety of testimonial injustice.

Earlier we floated a less substantive reading of Fricker’s dismissal of structural testimonial injustice as incidental, one that concedes the possibility of a structural variant but provides principled reasons for focusing on the transactional variant alone, independent of structural considerations. What principle could motivate such a narrowed focus, especially when it is conceded that structural features play an important role in a full understanding of the injustice involved in Fricker’s own examples of unjust testimonial transactions, such as that between Greenleaf and Sherwood?

One possible response points to the following key difference between the transactional and structural varieties: only the transactional variety of testimonial injustice need be available to the self-understanding of the maltreatment by the practitioners themselves. We have said that a transaction is a social act for which the self-conscious awareness by both agent and patient of their respective and complementary roles that comprise the act’s structure is part of the very idea of the act itself. This is not to say that such awareness always accompanies any actual instance of the transaction, but that both parties to the transaction have available an understanding of the structure as part of their participation in the relevant social practice, which includes such a transaction. This means that recognition of the order of justice established between them by the transaction, as well as the possibility of injustice through breaking the order, is part of the self-understanding of the transaction that is available to both agent and patient. The experience of transactional testimonial injustice when it occurs is thus internal to the self-understanding of the transaction of the participants themselves. There is simply no parallel to this in the case of structural testimonial injustice. That is, it is perfectly possible to fully and reflectively participate in the social practice of testimony and have no understanding of the structural injustice whatsoever. Unlike the transactional variant, structural testimonial injustice is not internal to the perspective of the participants in the practice themselves.

A corollary of this difference is that if anyone is interested in understanding testimonial injustice as it is intelligible from within the perspective of those involved in the practice, then there is good reason for limiting oneself to the transactional, and not the structural, variant. There are some indications in Fricker’s own writings that her overall concern accords primacy to such an internal perspective. For example, she begins her discussion of epistemic injustice pleading for a parallel reorientation in epistemology to that in ethical theorizing where “a moribund region of philosophy was revivified by a closer attention to lived experience” by “renewed attention to what we may broadly call ethical psychology – that is, to human beings’ real experience of ethical value”. 26 Although this is not fully developed in the remainder of the book, a parallel attention to what could be called ‘epistemic psychology’ pervades many of her subsequent discussions of testimonial injustice, paying close attention to the ways in which such injustice is experienced and understood by the agents themselves. Whether this accurately reflects Fricker’s own motivations or not, there remains an important sense in which taking seriously the epistemic psychology of testimony, construed as taking seriously the participant’s self-understanding of the social practice of testimony, provides a reason for limiting one’s theorizing to the transactional variant of testimonial injustice, and not the structural one.

Earlier I called this a ‘qualified’ defense of the less substantive reading. What makes this defense ‘qualified’ is just how limited the conclusion of this line of thought is, even if successful. The upshot of the less substantive reading is merely to affirm that not every explanation of an instance of testimonial injustice need appeal to both structural and transactional varieties, and that there may be principled reasons (such as an interest in epistemic psychology) in any case for limiting one’s focus on just one of them. The less substantive reading does not, however, imply that the transactional variant is somehow more pervasive or important than the structural one, or that it is more central in an overall account of testimonial injustice. Far from it: in rejecting the more substantive reading, we have affirmed the need to take seriously the possibility of structural testimonial injustice, since the institutional structures that facilitate participation in the practice of pooling information can lead to injustices within the practice that both extend far beyond the framework of transactions between participants, as well as alter the character of the transactions themselves. As a result, even one sympathetic to the less substantive reading would concede that a full understanding of many cases of testimonial injustice, including many of Fricker’s own examples, requires consideration of both structural and the transactional varieties, as well as the interplay between them.

4  Testimonial betrayal

Perhaps you – like me – harbor the nagging suspicion that our investigation thus far has missed something central to the agent’s self-understanding of the maltreatment experienced in central cases of testimonial injustice. The relationships forged by the transactional interactions that we have focused on thus far are thin, in the sense that they involve recognition of each other solely in terms of the interpersonal roles played by their participation in the practice, and the maltreatment involves abrogation of role-responsibilities in these formal relationships. The nagging suspicion is that many key cases critically involve thick relations of intimacy between acquaintances, and the maltreatment involves a rupture of, or disloyalty within, those thick relationships. 27 Put differently, the suspicion is that whilst we have been portraying the injustice as a breach, it is often far better portrayed as a betrayal.

Take Fricker’s signature case, Greenleaf’s dismissal of Sherwood’s say-so. Sherwood’s world is rapidly collapsing around her. The death of her fiancée, and the sense of uncertainty surrounding it, leaves her grief-stricken, vulnerable and isolated. Part of her isolation stems from the disparity between Sherwood’s belief that her fiancée was murdered by a mutual acquaintance (Ripley) and everyone else’s belief that this was a suicide. As her would-be father-in-law, Greenleaf provides a direct link to her deceased fiancée, and as a wealthy male benefactor in that society, Greenleaf provides the security and stability she craves, as well as a source of power to act and investigate that she lacks. Greenleaf’s sexist attitudes towards Sherwood’s testimony do not just exclude her from participation in the pooling of information; it constitutes a betrayal of their (proxy) relationship and serves to highlight Sherwood’s humiliating powerlessness in the social sphere. The viewer’s final glimpse of Sherwood in the screenplay highlights these themes. Sitting in a launch, accompanied by Greenleaf, she lunges at Ripley, accusing him of murdering her fiancée. Sherwood is physically restrained by Greenleaf’s henchman, who exchanges a knowing glance with Ripley. (“Ripley looks at him as if to say: what can you do, she’s hysterical”). The last shot sees Sherwood escorted away in Greenleaf’s protective company, passive and alone, all too aware – as we are – of Greenleaf’s erroneous complicity in siding with Ripley over herself.

The term ‘betrayal’ does not appear in Fricker’s book-length treatment of epistemic injustice, although it does feature prominently in her more recent work, especially in the context of exploring the possible role played by trust in the epistemology of testimony. 28 At the center of this more recent discussion is a distinction that Fricker makes between second- and third-personal stances of trust. Applied to the testimonial context, her suggestion is that there are two ways of treating someone as an informant as opposed to a source of information. One way (adopting a ‘second-personal stance of trust’) involves the Speaker and Hearer’s mutual awareness of, and acting upon, the Hearer’s manifest trust in the Speaker on this occasion. Another way (adopting a ‘third-personal stance of trust’) involves the Hearer trusting the Speaker, albeit without assuming the Speaker is herself treating the Hearer’s trust as the reason for striving to be a sincere and competent testifier towards him. In both cases, the Hearer trusts the Speaker, and that trust is directed towards the Speaker as a person – an active informant – as opposed to cases of mere reliance where the Hearer treats the Speaker as a passive source of information. A clear difference between these three attitudes (reliance, second- and third-personal stances of trust) can be seen when the attitude turns out to be misplaced. As Fricker puts it: “second personal trust creates the possibility of betrayal; third personal trust creates the possibility of ethical let-down; and the ethical attitudes involved in both kinds of trust are still to be contrasted with mere reliance, whose disappointment incorporates no ethical dimension”. 29

I concur that “[t]here is a difference of perspective here that wants philosophical capture”, 30 and it seems that our discussion thus far provides a way of doing just that. 31 Earlier we distinguished between the deliberative perspectives of the potentially solitary inquirer extracting information from objects, and the participant in the social practice of testimony via engagement in transactions. Transacting with someone involves adopting a thin stance of trust towards them. Although the person thereby recognized is conceived second-personally (“I am telling/being told by you”), the stance of trust (and the subsequent relationship) is thin in that it need not involve any recognition of the distinctiveness of the second-person beyond that formally required by the transaction. This is why such thin recognition on its own does not create rivalry: my recognizing you in a transaction on its own does not in any way detract from my recognizing any other person in such a transaction. 32 The order of justice created by the transaction can be culpably breached, and a ‘kind of ethical resentment’ towards the informant may be properly experienced, unlike the mere disappointment properly experienced in a case of misplaced reliance on a source of information. The breach in the transactional case will not, however, be properly experienced as betrayal, which requires a thicker stance of trusting recognition between persons that creates rivalry. The thin stance of trust adopted towards each other in a transaction thus differs from the thick stance of trust adopted between intimates (in the very broad sense of non-strangers), where the thick stance involves recognition of distinctive features of the personality of the one trusted in a way that creates and maintains rivalries.

Should we therefore admit a third variety of testimonial injustice, in addition to the transactional and structural variants already considered, viz. a betrayal of the intimate relationships established by a thick stance of trust between Speaker and Hearer in certain testimonial encounters? An affirmative response requires further elaboration, as betrayal does not appear to fall under our overall definition of testimonial injustice. Testimonial injustice, we have said throughout, is not merely an injustice associated with an act of testimony, but an injustice that stands in a categorical connection with the social practice of testimony. As we have seen, the act of testimony at the explanatory center of the practice of pooling information does not solely operate between intimates or acquaintances, and thus is not one that necessarily involves a thick stance of trust between Speaker and Hearer. It seems possible to fully understand the sting of betrayal experienced when a Hearer appears to be disloyal to a supposed special relationship with the Speaker without any recourse whatsoever to insight afforded by reflection on the practice of testimony. 33

This concern is deepened when one considers the difference in the standing accorded to the speech performance itself in establishing the thin trust of transactions, when compared to the thick trust of intimacy. In a transaction, the speech performance plays an ineliminable role in establishing the transactional relationship: even if the Hearer has good reason for being sure what it is that the Speaker thinks on a particular matter, there is no order of justice established between them unless the Speaker expresses this in a testimonial speech act (that is recognized as such by the Hearer). Further, the relationship in the transactional case is fixed at the moment of the transaction: subsequent failure by the Speaker to perform as she ought to perform given the norms established by the transaction is treated by the Hearer as a breach of the transaction, and gives no reason on its own for the Hearer to query whether such a transaction took place to start off with. 34 In contrast, establishing thick relations of intimacy and acquaintance do not require a specific speech performance by one party, and subsequent failures to live up to the expectations of that thicker relationship may well leave one party questioning whether such a thicker relationship was actually in place in the first place. These differences cast further doubt on the suggestion that we should treat betrayal of the relationships involved in the thick stance of trust as an additional variety of testimonial injustice, as the differences appear to reveal the manner in which the ethical intrigue of such thicker relationships potentially floats free from the speech act of testifying.

One response to these concerns is to concede that though betrayal is a central feature of the experience of testimonial injustice in many key cases, it is best construed as an injustice that is associated with an act of testimony rather than a testimonial injustice. A more ambitious response explores the interplay between thick and thin trust relations in the context of a transaction so as to reveal a distinctive notion of betrayal.

Let us provide a thumbnail sketch of the more ambitious response. 35 In intimate relationships forged by thick relations of trust, parties seek each other’s approval, often including recognition of their opinions on matters arising in the course of their interactions. In the context of such relationships, seeking approval need not take the form of an explicit request for approval; one party may just assume that the other party will take their opinion seriously given their purported relationship. Failure by one party to accord the recognition sought implicitly may be experienced by the other party as a betrayal of that thick relationship. Matters are subtly different when approval is sought via a transactional engagement between the parties, such as by one party telling the other something. A transaction, we have said, includes the self-conscious awareness of the structure by both parties as part of the very idea of a transaction itself. As a result, once the transaction has taken place, either party can try to ignore the normative implications of the performance, but to ignore these implications in the context of a transaction is to actively reject them. 36 A transaction between intimates thus has the effect of turning a failure to accord the recognition sought into a rejection of the requested approbation. It is not just that one party does not provide the recognition that the other party seeks, but that the one party throws the request for recognition back to the other party. The experience of betrayal following an unsuccessful transaction between intimates thus takes on the distinctive form of the sting of rejection. When Sherwood tells Greenleaf of her suspicions about Ripley, she seeks out his approbation. His failure to believe her is experienced by her as rejection of their relationship, a rejection that exacerbates her own sense of isolation and powerlessness.

Although it needs to be developed further, this more ambitious response opens the possibility for recognition of yet another variety of testimonial injustice – testimonial betrayal. It is admitted that if one thinks of the social practice of testimony abstracted away from agents who are socially-situated, then the sting of betrayal will simply not feature in cases where there is a breach in testimonial transactions. But the actual social practice of testimony does involve socially-situated agents, and this means not just that they stand in varying forms of power relations with one another, but they also stand in varying relations of intimacy and acquaintance with each other. In such a socially situated practice, people seek out recognition and/or approval as believers and testifiers from certain individuals and not others, and experience the sting of betrayal when this recognition is perfidiously accorded to others instead. When seeking approval takes a transactional form, such as through telling someone something, then betrayal takes the form of rejection. The social practice of testimony thus allows us to understand the distinctive kind of betrayal experienced when one has one’s testimony rejected in the context of intimate relationships.

5  Summary

This essay has been an investigation into the possibility of there being a categorical connection between the social practice of testimony and certain instances of injustice. By focusing on different aspects of the social practice, we have managed to identify three potential varieties of testimonial injustice. The first variety – transactional testimonial injustice – emerges when one focuses on the character of the self-conscious, interpersonal transactions that take place within the practice, most centrally the act of telling someone something. The injustice here involves a breach of the order of justice established between the parties by the transactions that fall within the social practice of testimony. The second variety – structural testimonial injustice – emerges when one focusses on the social institutions within which the practice of testimony operates. The injustice here stems from structural inequalities within such institutions that lead to diminished possibilities of participation in the social practice of testimony. More tentatively, a third variety – testimonial betrayal – emerges when one focuses on the thick trust relations between acquaintances within the practice. The injustice here includes the experience of humiliating rejection felt as part of the sting of betrayal.

We have used the seminal work of Miranda Fricker as both inspiration and foil throughout our investigation. Whilst broadly endorsing her understanding of testimonial injustice, we have made three suggested emendations. The first is to extend the scope of those addressed by the speech act of testimony beyond those targeted by the Speaker. The second is to read her earlier work as focusing on a transactional testimonial injustice, whereas her more recent writings on trust as focusing on testimonial betrayal. The third is to view her relative inattention to structural testimonial injustice as stemming from a principled commitment to the importance of ‘epistemic psychology’.

The discussion here has been limited to the diagnostic dimension of Fricker’s work on testimonial injustice, the attempt to make theoretical sense of the very idea of testimonial injustice itself. This leaves open the task of exploring the therapeutic dimension of her work, the attempt to consider ways to mitigate and ameliorate such injustice. Further work in both dimensions will no doubt be enhanced by greater sensitivity to the varieties of testimonial injustice brought into view by considering differing aspects of the social practice of testimony, including at least some of the variants distinguished here. 37

Related chapters: 1, 3, 4, 5, 6, 20, 21, 23


The trio, and the pathological tag, is due to Gelfert (2014: 193–214).

As per convention, I have used terms Speaker and Hearer as labels for the two parties to a testimonial transaction, even though there need be no actual speaking or hearing going on. Further, unless context dictates otherwise, the Speaker is treated as female and the Hearer as male to aid comprehension.

Fricker (2007: 20–21) argues for limiting testimonial injustice to those cases involving a credibility deficit alone and not extending this analysis to cases involving an excess of credibility. For a convincing argument against such a limitation, see Medina (2011).

Rödl (2014: 292).

This is clearest in Fricker (2008).

Fricker (2007: 145).

In addition to Anderson (2012) discussed below, see – for example – Young (2007), Forst (2012) and Laden (2013).

Fricker (2007: 27).

Coady (2010) and Anderson (2012) are particularly clear on this point.

The notion of a transaction is more fully developed in Wanderer (2014).

Cf. Descombes (2014: 243).

Hookway (2010: 156).

This point is developed more fully in Moran (2005).

Craig (1990: 36).

See, esp., Fricker (2012).

Fricker (2007: 1).

Fricker herself rejects the more substantive reading in Fricker (2010), in response to related concerns raised by David Coady (2010). In subsequent work (Fricker 2013), she explicitly marks the difference between these two varieties under the headings ‘discriminatory’ and ‘distributive’ epistemic injustice.

Anderson (2012: 169).

Fricker (2007: 145), cited above.

Fricker (2007: vii).

Maitra (2010) raises a concern along these lines – see especially her comparison of Marge Sherwood with the case of Zara on pages 198–200. In response, Maitra proposes a modification of Fricker’s account, one that supplements it with explicit reference to obligations to believe the say-so of certain others that arise in the context of special interpersonal relationships. As will emerge in the ensuing discussion, such a modification is rejected here as it blurs the difference between transactional testimonial injustice and testimonial betrayal.

Fricker (2012: 257).

Fricker (2012: 257).

Fricker’s own way of capturing the difference invokes her own understanding of the speech act of testimony as personal, rejected above.

Cf. Markovits (2011: 304) for a similar point about promising.

Given limitations of space, I am relying here on an assumed familiarity with the sting of betrayal and a tacit understanding of why it is experienced as a form of maltreatment. I have found Simmel (1950: 333–334) and Shklar (1984: 138–191) helpful in this regard.

I am drawing here on the discussion of promises in Markovits (2011), which has shown me ways of developing an earlier – and cruder – version of the distinction put forward in Wanderer (2013).

My thinking in this paragraph is deeply indebted to discussions with Byron Davies, and to a paper he presented at UMass Boston in April 2015 on ‘The Insult in Not Being Believed’.

Cf. Wanderer (2012).

An earlier version of this chapter was presented at the ‘Between Ethics and Epistemology’ workshop on Miranda Fricker’s work at UMass-Boston in April 2015, and I am indebted to the comments of my copresenters and members of the audience on that occasion. In addition, I have benefitted from generous feedback from Byron Davies, Miranda Fricker, Henry Lara-Steidel, Richard Moran, Lisa Rivera and Leo Townsend, as well as from the editors of this volume.


Anderson, Elizabeth . 2012. ‘Epistemic Justice as a Virtue of Social Institutions’. Social Epistemology 26 (2): 163–173.
Coady, David . 2010. ‘Two Concepts of Epistemic Injustice’. Episteme 7 (2): 101–113.
Craig, Edward . 1990. Knowledge and the State of Nature. Oxford: Clarendon Press.
Descombes, Vincent . 2014. The Institutions of Meaning. Cambridge, MA: Harvard University Press.
Forst, Rainer . 2012. The Right to Justification: Elements of a Constructivist Theory of Justice. New York: Columbia University Press.
Fricker, Miranda . 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press.
Fricker, Miranda . 2008. ‘Scepticism and the Genealogy of Knowledge: Situating Epistemology in Time’. Philosophical Papers 37 (1): 27–50.
Fricker, Miranda . 2010. ‘Replies to Alcoff, Goldberg, and Hookway on Epistemic Injustice’. Episteme 7 (2): 164–178.
Fricker, Miranda . 2012. ‘Group Testimony? The Making of a Collective Good Informant’. Philosophy and Phenomenological Research 84 (2): 249–276.
Fricker, Miranda . 2013. ‘Epistemic Justice as a Condition of Political Freedom?’. Synthese 190 (7): 1317–1332.
Gelfert, Axel . 2014. A Critical Introduction to Testimony. London, UK: Continuum.
Hookway, Christopher . 2010. ‘Some Varieties of Epistemic Injustice: Reflections on Fricker’. Episteme 7 (2): 151–163.
Laden, Anthony . 2013. ‘Learning to Be Equal: Just Schools as Schools of Justice’, in D. Allen and R. Reich (eds.), Education, Justice, and Democracy, 62–79. Chicago, IL: University of Chicago Press.
Maitra, Ishani . 2010. ‘The Nature of Epistemic Injustice’. Philosophical Books 51 (4): 195–211.
Markovits, Daniel . 2011. ‘Promises as an Arms-Length Relation’, in H. Sheinman (ed.), Promises and Agreements: Philosophical Essays, 295–326. Oxford: Oxford University Press.
Medina, José. 2011. ‘The Relevance of Credibility Excess in a Proportional View of Epistemic Injustice: Differential Epistemic Authority and the Social Imaginary’. Social Epistemology 25 (1): 15–35.
Minghella, Anthony , and Patricia Highsmith . 2000. The Talented Mr. Ripley. London: Methuen.
Moran, Richard . 2005. ‘Getting Told and Being Believed’. Philosophers’ Imprint 5 (5): 1–29.
Rödl, Sebastian . 2014. ‘Testimony and Generality’. Philosophical Topics 42 (1): 291–302.
Shklar, Judith N. 1984. Ordinary Vices. Cambridge, MA: Belknap Press of Harvard University Press.
Simmel, Georg . 1950. The Sociology of Georg Simmel (trans. K.H. Wolff ). London: Free Press.
Wanderer, Jeremy . 2012. ‘Addressing Testimonial Injustice: Being Ignored and Being Rejected’. The Philosophical Quarterly 62: 148–169.
Wanderer, Jeremy . 2013. ‘Testimony and the Interpersonal’. International Journal of Philosophical Studies 21 (1): 92–110.
Wanderer, Jeremy . 2014. ‘Alethic Holdings’. Philosophical Topics 42 (1): 63–84.
Young, Iris Marion . 2007. Structural Injustice and the Politics of Difference. Cambridge: Cambridge University Press.
Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.