1. Introduction
Much of the literature concerning epistemic injustice has focused on the variety of harms done to persons in their capacities as potential contributors to knowledge projects. For example, recognizing that knowledge generation is a social endeavour, feminist epistemologists and critical race theorists have articulated and analyzed the idea of testimonial injustice arguing that social prejudices and implicit bias can result in members of certain groups suffering credibility deficits that interfere in their ability to participate in joint epistemic projects.Footnote 1 Similarly, arguments have been made that dominant conceptual frameworks favouring the experiences of the privileged can play a role in wronging members of groups through a lack of culturally available hermeneutical resources that would be necessary for the marginalized to come to fully understand their experiences, as well as have those experiences given uptake within society.Footnote 2 These kinds of wrongs, along with others, have been conceptualized as wrongs that thwart one's ability to participate in epistemic projects of generating knowledge.Footnote 3
But the variety of ways in which one's ability to participate in the generation of knowledge can be unjustly thwarted do not exhaust the epistemic injustices that marginalized inquirers can suffer. If we are to understand the full implications of the social nature of knowing, we must confront another dimension of the knowing enterprise: the circulation of knowledge and the capacity of epistemic agents to take up knowledge produced by others and make use of it. Not enough philosophical attention has yet been devoted to understanding ourselves as potential receivers of knowledge and understanding. As epistemic agents, we have the potential both to be responsible contributors to knowledge production and responsible recipients of knowledge. Furthermore, our abilities to act as responsible receivers of knowledge and understanding are crucially linked to our abilities to serve as contributors to knowledge: receiving knowledge well places us in a better position to both generate further knowledge and participate in passing on knowledge to others. To be trustworthy contributors to the endeavours of our epistemic communities, we must be good at judging the trustworthiness of those who offer knowledge claims to us.Footnote 4 The interaction of these two components of epistemic responsibility – contributor and recipient – makes clear the philosophical importance of investigating our epistemic responsibilities as recipients of knowledge, and relatedly, the nature of epistemic injustices that might occur on this front.
In what follows I first examine the important role of epistemic trust given the cognitive division of labour and the presence of communities of expertise in modern society. Although what I argue is relevant to many different contexts of knowing through trust, I focus on the case of scientific knowledge and understanding, examining the relationship between those communities and institutions that generate and communicate such knowledge and situated lay communities who are in a position where, as non-experts, trust is the only way to access such knowledge. Scientific knowledge is an especially important case to examine with respect to epistemic injustices; because it is a dominant and powerful form of knowing in contemporary society, with deep significance for the structure of our social and material lives, it has the potential to pose a greater threat to the epistemic capacities and opportunities of the socially marginalized than other less pervasive epistemic practices. I argue that members of socially marginalized lay communities can suffer epistemic trust injustices when potentially powerful forms of knowing such as scientific understandings are generated in isolation from them, and when the social conditions required in order to generate reasons that would support a responsibly-placed trust in the relevant epistemic institutions fail to transpire. Significant epistemic and other harms can result from such injustices.
2. Epistemic Injustice and Trust
In Miranda Fricker's characterization of an epistemic injustice, an epistemic injustice is a wrong ‘done to someone specifically in their capacity as a knower’.Footnote 5 Others have used slightly different language. Elizabeth Anderson characterizes epistemic injustices as involving impediments to one's capacity ‘as an inquirer’Footnote 6 and Christopher Hookway discusses epistemic injustices in terms of obstacles to activities (my emphasis) that are ‘distinctly epistemic’.Footnote 7 These descriptions are used to specify the characteristics of an epistemic injustice as distinct from other forms of ethical injustice. The emphasis of these theorists is clearly on the activities of inquiry; when Fricker refers to the wrong of an epistemic injustice with reference to one's ‘capacity as a knower’, the term ‘knower’ must be understood as someone engaged in the activities of knowing, not a ‘knower’ in the sense of someone who is already in possession of knowledge. This is an important feature to note because for the idea of epistemic injustice to be taken seriously, it cannot simply amount to the denial of an opportunity to access a piece of knowledge. Since ‘knowing everything’ is not a viable goal of epistemic life, simply missing out on bits and pieces of knowledge cannot carry the seriousness required of the concept of epistemic injustice. This is not to say that being prevented from accessing certain knowledge, or having it hidden from one, could not constitute an epistemic injustice, but the case would have to be made for the seriousness of this particular instance of knowledge. The knowledge would have to be of particular significance to one's well-being overall, or its lack must more generally negatively affect one's epistemic abilities, such as one's capacity to obtain other knowledge extremely relevant to one's life. This is why science is an important case to examine: if in today's society scientific knowledge is a dominant source of significant knowledge relevant to people's lives, and social conditions and historical relations with scientific communities block members of some marginalized communities from having the ability to access such knowledge claims in a responsible way, then epistemic trust injustices and their associated harms result.
Given that epistemic injustices are typically defined as affecting one's capacities as an active knower or inquirer, it is reasonable to understand epistemic injustices as impeding (unjustly) on the exercise of one's epistemic agency. Moreover, we engage our epistemic agency not only when trying to actively generate new understandings or convey them to others. We also engage our epistemic agency when we are on the receiving end of others’ attempts to convey their understandings of the world to us. When we take someone else's word on trust, or when we trust others to produce a certain kind of knowledge for us (via a cognitive division of labour), we are engaging our epistemic agency, and this can be done either well or poorly. In what follows I first address how one's epistemic agency is engaged when we take epistemic matters on trust, and then explain how that epistemic agency can itself be stymied by histories of certain social conditions that result in marginalized groups having reason to distrust dominant knowledge sources.
Many philosophers and sociologists of knowledge who take seriously the social nature of knowing have taken up issues concerning the relationship between experts and non-expertsFootnote 8 and have recognized the fact that a cognitive division of labour requires that epistemic agents trust other knowers to a great extent. However, fewer have worked explicitly with the likes of a situated analysis such as what we see in the epistemic injustice literature specifically, and more generally in many versions of feminist epistemology.Footnote 9 A situated analysis of expert/non-expert trust relations considers the social relations between knowledge providers and knowledge recipients and the histories of interactions between them. The philosophical work that has identified epistemic injustices that impede on one's participation in the generation and communication of knowledge has stemmed from an awareness of the deep ways in which power relations and the social biases that are a part of relations between inquirers have epistemic effects. In arguing the case for epistemic trust injustices, I likewise attend to the power-infused social relations between scientific communities as authoritative generators of knowledge and socially marginalized lay communities.
3. The Role of Trust in Knowing, and Responsible Trust
One of the reasons epistemologists have worried about people's dependence on others for knowing through trust is that trust makes one vulnerable, and may appear to remove epistemic agency from us as individual knowers, or suggest an abdication of epistemic responsibility. In John Hardwig's phrasing, trust is ‘at least partially blind’ and as he notes, for many epistemologists who focus on the need for evidence in order to have knowledge, the problem is not simply that knowledge does not involve trust, but rather that trust and knowledge are conceptualized as antithetical to each other.Footnote 10 But it is far from clear that trusting others for knowledge incapacitates us as epistemic agents or involves an abdication of epistemic agency, and if our epistemic agency is still involved in trusting others in our epistemic pursuits, then epistemic injustices involving trust are possible and need to be examined.
When it comes to a general theory of trust, numerous trust theorists have argued that in addition to its cognitive dimension, trust has an affective and attitudinal dimension that goes beyond simply relying on someone to deliver the goods with which one is entrusting them. To trust someone is to hold an attitude of optimism toward their fulfillment of your expectations.Footnote 11 Trust relations have a moral dimension to them, and the difference between trust and reliance is what explains the feelings of betrayal that can occur under circumstances when a trusted one fails you. Further, we can trust people for many different things, and we rightfully place boundaries on our trust: to trust someone is shorthand for person A trusting person B for Z (where Z is some kind of good that A cares about). I can easily trust a friend to not steal from me, but at the same time, because I know that they also are forgetful, I may not trust them to care for and feed my dog for the weekend. Given this, we need to identify the kind of trust and the boundaries of that trust that are of interest for understanding cases of epistemic trust.
I take it as uncontroversial that in many circumstances, trusting another epistemically is the most epistemically virtuous course of action. The need for trust makes each of us vulnerable to others who claim expertise. I hope it is also uncontroversial that in the case of highly specialized and large scale epistemic endeavours such as those involved in many forms of contemporary scientific knowledge, it would be epistemically unwise to offer a blanket trust to anyone and anything purporting to be scientific or that one takes to be scientific. One can lack discernment in whom to trust for knowledge, but that is not something we should strive for. The issue is not whether we do or do not place trust in others, but rather whether and when such trust can be relatively well-grounded. For ideal epistemic success, trust relations would be such that the degree of trust one grants would always be balanced by the degree of trustworthiness of the source. In the case of epistemic trust, the balance of epistemic trust and trustworthiness would still not guarantee epistemic success; even a trustworthy source who is doing their best can end up failing to deliver the epistemic goods, or can (out of character) betray the trust one has given them, upending their history of trustworthiness. Grounding our trust, to whatever degree that we do, does not mean we escape the vulnerability that we expose ourselves to when we trust. Additionally, the truster is never in a position to fully determine the trustworthiness of their source. Thus, we cannot expect ideally placed trust that is always perfectly balanced with the source's trustworthiness. Even so, it remains possible to distinguish responsibly-placed trust – trust granted in cases in which one has good reason to take one's source as trustworthy – from trust which might fail to be responsible, such as might occur if one ignores the reasons right in front of them for thinking that a source is not trustworthy, but trusts anyways.Footnote 12 If the options are an inability to access the knowledge (by resisting trust), or to trust in the claims irresponsibly (when the reasons direct you otherwise), we might (with further specified conditions) be facing a case of an epistemic trust injustice.
Although trust necessarily makes one vulnerable and one can go wrong in trusting others, building robust trust relations is a very important part of a fruitful epistemic life, and thus participation in well-developed trust relations constitutes an important exercise of epistemic agency. A responsibly-placed trust is not a result of an abdication of one's epistemic responsibilities of assessment, or being closed to re-assessment should further evidence come one's way. Hardwig's note that trust is ‘partially blind’ (my emphasis) is not the same as claiming that trust is fully opaque or that we have turned off our capacities as epistemic agents in resorting to trust. Placing epistemic trust in another is an exercise in epistemic agency, even though these trust placements and the reasons supporting them function quite differently from the ways we engage in the collection of evidence during individual inquiry. Trust relations are formed with others in part through histories of interactions with them, and these histories give a broad set of relevant reasons for trust in particular circumstances, and give reason for doubt, skepticism, and distrust in others. When those reasons inform in whom, for what, and to what degree one places one's epistemic trust, this trust (or distrust) can be said to be responsibly-placed. It is in part because of this breadth of relevant reasons for trust and distrust that histories of social marginalization can result in epistemic trust injustices.
The most simplified version of epistemic trust concerns trusting a single person's testimony. Most straightforwardly, epistemic trust concerns one's willingness to take the word of another – to trust in their testimony, form beliefs on the basis of that testimony, and act in accordance with such beliefs. As many testimony theorists have pointed out, conditions of acceptable trust in testifiers involve judgements that the testifier is competent and sincere.Footnote 13 It only makes sense to rely on someone's testimony if I have reason to think that the person is competent in the area in question – that is, likely to be in possession of the knowledge I am interested in, and sincere – that is, likely to be trying to convey to me what they take to be true beliefs rather than dupe me, lie to me, play a joke on me, or mislead me in my beliefs. While the competency requirement focuses on the testifier's relationship to the knowledge in question, the sincerity requirement emphasizes the ethical dimension of successful testimony practices; it signals a relationship between the speaker and the hearer, and an attitude toward the knowledge recipient. Competency and sincerity are necessary for a testifier to be trustworthy, and a truster can responsibly place their trust in a testifier who appears to satisfy these requirements, relative to the context at hand.
In simple and isolated cases of testimony, the above interpretations of competence and sincerity may do the trick for inquirers determining their placements of epistemic trust. However, in order to understand our ongoing epistemic trust relations with communities of experts, especially in cases where the stakes are more significant, these initial interpretations need to be broadened and I take this up in the next section.
Additionally, although the personal testimony work helps us identify core features of epistemic trust, further issues emerge when we consider trust in communities and institutions of knowledge production, such as scientific communities. Most trust theorists agree that interpersonal trust remains the paradigmatic form of trust, with institutional trust being modeled after it.Footnote 14 Epistemic trust in institutions and communities share core features of interpersonal epistemic trust: vulnerability of the truster, an attitude of optimism toward the institution and its ability to fulfill certain expectations related to providing the truster with reliable knowledge, the possibility of being betrayed, and the possibility of the trust being more or less well-grounded.
What is different about the ‘impersonal’ trust in institutions and communities compared with trust in persons, however, is that the trustworthiness of the specific practices of the institutionFootnote 15 plays a more significant role in determining well-placed trust than in the case of individual testifiers. It may be less important that every participant in the institution be perfectly trustworthy if an institution operates with robustly trustworthy practices that help protect against the undue influence of untrustworthy individual members. But this can work the other way as well; when historically an institution has been failing to operate with trustworthy practices, and there is a track record that demonstrates this, these failures will be more relevant to assessments of the institution's trustworthiness than would be the sincere efforts at trustworthiness exhibited by particular individuals within the institution. Grasping the relevance of the practices of the institutions and communities themselves beyond the epistemic efforts of their individual members in assessments of trustworthiness allows us to appreciate the impact that such institutional features as embedded racism and sexism can have on epistemic trustworthiness. Lay persons and their communities have historical relations with enduring institutions of knowledge production and dissemination that outlive connections with individual members of those institutions, and evidence of histories of untrustworthy practices due to embedded racism and sexism can place responsibly-formed trust out of reach for some marginalized lay persons, resulting in epistemic trust injustices and their associated harms.
4. Complicating the Grounding of Epistemic Trust 1: The Competence Condition as ‘Providing Significant Knowledge’
In anything more than a mundane case of testimony, the relevant epistemic competencies are likely broader than simply that the testifier has the background to ‘know what they are talking about’. The idea in the standard case of testimony was that if a hearer has reason to think that the speaker knows what they are talking about, that counts as a reason (ceteris paribus) to believe them. But one of the problems with using the standard testimony analysis as a model for epistemic trust across expertise, either generally or specifically in cases of scientific experts, is that it focuses on a very narrow form of epistemic trust: simply trusting another's word in a given instance, without taking account of how that version of epistemic trust is embedded in a collection of competency expectations that are crucial to long-standing trust relations that operate in the service of our epistemic goals. In the case of scientific experts (and other large institutions of expert knowledge production) we (rightly) place upon our speakers a set of expectations that cover a broader range of epistemic goods than simply delivering a claim of knowledge.Footnote 16 Their competence as a testifier depends crucially on many other epistemic skills and capabilities.
For example, in his discussion of public trust, David Resnik points out that members of the public trust scientists to do many different things.Footnote 17 Identifying what members of the public trust scientists to do is a way of articulating the public's expectations of scientists. Resnik provides a partial list of public expectations including: trusting researchers with public resources, trusting researchers to provide knowledge and expertise that can inform policy debates, trusting researchers to provide knowledge that will ‘yield beneficial applications in medicine, industry, engineering, technology, agriculture, transportation, communication, and other domains’, and trusting researchers to make informed judgments about new technologies. Many other activities could be added to the list, but two I want to explicitly draw attention to are first, trusting researchers to filter information for us, determining what the best understandings of the day are and omitting poorer quality, less important, or outdated research,Footnote 18 and second, trusting researchers to treat stakeholder populations, including research subjects, ethically and not place them at too great a risk in the pursuit of knowledge. In the first, we expect experts to be able to offer expert judgement on the current status of knowledge, and in the second, we expect them to conform to certain societally-held ethical norms as they make judgements concerning limits in the pursuit of knowledge.
Clearly, not everything laypersons trust scientists to do will be relevant to an assessment of their overall epistemic trustworthiness; the work of scientists involves them in social practices which may have non-epistemic goals as well. For example, the public may expect scientists to use their funds appropriately, not embezzle them and not to use them carelessly or inefficiently. If they fail in this, it is not immediately clear that this failing is relevant to one's assessment of their epistemic trustworthiness, though it certainly is relevant to their financial trustworthiness and potentially their general personal trustworthiness as well. Similarly, many may consider that the point concerning whether or not scientists treat their research subjects ethically may be relevant to their ethical trustworthiness, but not relevant to their epistemic trustworthiness. I will argue against this position shortly, but many will find it an initially plausible position. Even acknowledging that there will be some public expectations of activities that are not epistemically relevant, a strong argument can be made that the epistemic competencies members of the public expect of expert communities include a broad range of activities that need to be done well in order for scientists to serve as trustworthy testifiers of scientific knowledge, and failure in some of those may affect their overall core role as trustworthy providers of knowledge. These activities include skillfully undertaking their research activities, conveying their results in an epistemically responsible manner, avoiding research and publication misconduct throughout their professional activities, and filtering and assessing results within their field of expertise to allow them to convey robust judgements of the current state of knowledge in a field.
Particularly in the case of organized epistemic communities, such as scientific communities, when we trust these communities, we are not just trusting their specific claims. We are in fact (and quite rightly) trusting them to undertake a variety of valuable epistemic work: to engage in inquiry for us, to produce knowledge for us, to orient their research in ways that will be useful, or at least not harmful for us, and to make judgements concerning which claims need communicating. This clustering of epistemic expectations means that when there are reasons to suspect that such communities have failed in some of these expectations, these failures should be relevant to our assessment of their epistemic trustworthiness, in effect lowering the trust levels we grant them.
In determining which activities are relevant for these trust assessments, it is helpful to characterize the overall function of epistemic work of communities of experts in more general terms. Importantly, when we turn to communities of experts, we have an interest in certain kinds of knowledge. It is, of course, significant knowledge that we are after, not just any collection of trivial facts about the world,Footnote 19 and if the knowledge was not important to us, there would not be much motivation to place one's trust in the experts. We turn to experts when and if we consider that they are well positioned to produce such knowledge, recognize its significance for us, and communicate that knowledge to us. This involves expecting them to run robust and sound research programs, be able to filter results such that they can offer us a sound judgement on the best knowledge available, and also attend to our particular epistemic needs (which may not be shared by all) so that what they are providing is actually significant knowledge for us. Furthermore, given that each of us is socially situated, we cannot assume that what is significant knowledge for us will be significant for all other knowers.
In short, a general characterization of the epistemic role of experts is one of being a provider of significant knowledge. In order to succeed in this, many activities are involved, but this overarching characterization can give us a foothold into determining how these various activities contribute to the epistemic function with which lay persons seek to trust them. From there, we will be able to identify when a history of failures in certain activities may be surprisingly relevant to creating a blockage in possibilities for responsibly-placed trust for particularly situated lay persons, especially the socially marginalized.
5. Complicating the Grounding of Epistemic Trust 2: The Sincerity/Care Condition
Similarly, although the sincerity requirement as previously described can work well in identifying the moral component of the trust relation for simple cases where the testimonial trust relationships are quite limited and short-lived (such as when I ask a stranger for directions to the nearest bus stop) it is insufficient for richer relations of epistemic dependence. When the knowledge in question is much more than trivial, the relationship between the speaker and the hearer will need to be more robust in order to sustain the depth and breadth of trust that would support making oneself vulnerable by taking someone's word for it on an issue of grave importance. Consider the case of my physician giving me news of a serious disease, alongside a recommendation for experimental treatment as my best course of action. I have interests in coming to understand both my health condition and the recommended options and their risks so that I can make treatment decisions. My medical interests in pursuing good health include epistemic interests; there are things I need to know. In trusting my physician with my relevant epistemic needs, I will expect (with optimism) that she will have a stronger attitude of care toward me than simply being sincere in her statements (i.e. telling me the truth). I will expect that she has enough care directed towards me and my medical interests that she will attempt to offer me the specific knowledge that I need in order to make reasonable decisions – perhaps outlining the range of available options. For her to be trustworthy, sufficient care for my interests will need to override competing concerns, such as her potential need to feel socially secure by offering a definitive diagnosis (even if she is not completely confident of the diagnosis), or her need to satisfy other interests of hers, such as finding enough research subjects for the experimental treatment. My trust might be bolstered in part by my understanding that the professional standards of medical practice stipulate caring for the medical interests of one's patients, including their epistemic interests of accessing the necessary knowledge to help them take care of themselves and make appropriate treatment decisions. My history of interactions with my physician, and with the medical establishment more generally, can also offer me reasons for trust: have I been well-served in the past by them on different medical issues? Or are there past instances in which either my physician or other doctors failed to display the relevant care (and competency) towards my interests and needs?
In rich and important cases of ongoing epistemic exchange, ‘sincerity’ in itself is insufficient to capture the attitude of care that is needed to support the kind of extensive and lasting trust for satisfying the ongoing epistemic needs of a nonexpert. I propose renaming this general requirement as the ‘sincerity/care’ condition. The sincerity/care condition notes that a trustworthy expert must embody some degree of a moral attitude of care toward the recipient. Sincerity captures a minimal requirement of care, but the degree of care required to form and sustain trust will be context-dependent, in part depending on the stakes of the epistemic situation and the scope of the epistemic trust placed in the expert.
6. Epistemic Trust Injustices
A brief recap is in order. Having begun with a narrow understanding of the sincerity and competence requirements for making reasonable judgements for trust in cases of mundane testimony, I expanded these requirements, arguing that in cases of higher stakes knowing, and cases where we need to trust communities and institutions for expert knowledge on an ongoing basis over time, the relevant competencies are multiple and broad in terms of the activities that must be undertaken to be epistemically trustworthy. I offered a core description of these competencies as consisting in the experts’ ability to provide significant knowledge to the non-expert, which, in the case of science, includes both producing it and conveying it. Additionally, I argued that the sincerity requirement is better understood as a broader ‘sincerity/care’ condition, in recognition that trust relations are ongoing and exist in complex epistemic environments. In order to depend upon an expert community for their ongoing and sometimes high stakes epistemic needs, a lay person needs to trust that in undertaking and conveying the epistemic work, the expert embodies an appropriate degree of care for their interests, most importantly their epistemic interests, though caring for one's other central interests can be crucial to identifying and serving those epistemic interests. One needs to trust that the ongoing judgements of the expert account for one's concerns.
Both of these conditions draw attention to the relevance of differences in social situation. If competencies are understood under the larger heading of being a provider of significant knowledge, attention must be paid to differences in situation that make some knowledge significant to some, but not necessarily all. Differently situated lay persons will differ on at least some epistemic interests, and histories of the provision of knowledge from various institutions and communities may or may not demonstrate patterns of having addressed such epistemic interests. Similarly, the sincerity/care requirement can only be fulfilled if there is an awareness of the relevant interests of the particular non-expert in question. If an expert community is unaware of the needs of a particular lay community and the epistemic dimensions of those needs, it will be challenging for it to offer, let alone demonstrate the appropriate level of care for those needs. To be trustworthy for particularly situated lay persons, an expert or expert community must be able to satisfy these conditions.
From the perspective of the potential truster, what is relevant to determining cases of epistemic trust injustices is whether, through past histories of interaction, there have emerged reasons to distrust these experts, or at least withhold trust.Footnote 20 When there is historical evidence available concerning how well suited a community of experts is to satisfy the competency and the sincerity/care conditions, we need to use it. A history through which an epistemic community has served us well as a provider of significant knowledge leads us to reasonably and responsibly continue in that trust relationship, learning lots of things on the basis of that trust, without further checking.Footnote 21 But in cases where there is evidence that the competency and sincerity/care conditions have not been well satisfied over time, a lay person may fail to have adequate grounds from which they could responsibly place their trust in the expert community on an ongoing basis. This is a situation where they actually have reasons to distrust or withhold trust from the expert community. This may not stop them from trusting, but, if responsibly-placed trust is prevented under such conditions, then regardless of whether they trust or distrust, their capacity to inquire and attain knowledge in a responsible way is stymied.
When faced with such reasons for distrust, there are various possible outcomes. Lay persons may reject the claims, or they may end up merely ‘relying’ on an expert community out of desperation, having no better options than to proceed without any expectation or optimism that the expert will be able to deliver the epistemic goods. They also may consciously take the risk of trusting in the hopes of improving the trust relationship, engaging a long-haul strategy of bettering epistemic relations. However, the relevant point here is that although reaching for the poorly grounded trust or mere reliance strategy might work out for them in a particular case, at this juncture (given that they have reasons to distrust) they are faced with the unsatisfying situation of being unable to partake in a healthy trust relation in an epistemically responsible manner, due to the way the socially embedded epistemic practices have been functioning.
We have now reached a point where we can articulate the idea of an epistemic trust injustice and connect it to a failure to satisfy the conditions of responsibly-placed trust. We will recall that in the case of epistemic injustices that concerned epistemic agents as contributors to the production and circulation of knowledge, it was the thwarting of their capacities as inquirers and epistemic agents that constituted the epistemic nature of the injustice. I have demonstrated that the ways in which we place and sustain trust in others for expert knowledge also involves our epistemic agency; though we trust, our epistemic capacities are not set aside but rather play a role in forming and sustaining our attitudes of trust and distrust. When the conditions are such that the available evidence points in the direction of distrust, making it difficult or impossible to responsibly trust in expert communities such that I could receive knowledge and understandings of the world through responsibly-placed trust, there is a sense in which my epistemic agency is being thwarted.
Of course, one might argue that withholding trust in such a situation is actually a case of well-exercised epistemic agency; after all, I've noted that this person has reasons for distrust. One could say that the epistemically responsible thing to do in such a hostile environment is to withhold trust. However, my point in emphasizing the threat to epistemic agency is that the described circumstances that give reasons for distrust do not just pop up for an isolated case. As a relationship of distrust that has grounding, it will apply to many interactions with this expert community over time. As I pointed out early on, the incapacity to acquire a particular piece of knowledge does not itself threaten one's epistemic agency. But if social conditions are such that this incapacity is either quite far-reaching (by blocking my access to large swaths of knowledge) or concerns understandings central to my self-understanding (blocking knowledge that is of crucial importance to my overall life and ability to continue as an active epistemic agent) at some point we can make sense of this blockage as actually threatening one's epistemic agency. Such blockages in one's ability to form responsibly-placed trust constitutes an epistemic injustice when such thwarting is systemic, and involves my capacity (more specifically, my lack thereof) to access areas of knowledge that are significant for me. Epistemic trust injustices occur in situations of substantial social marginalization by having a significant dampening effect on the abilities of members of subordinated groups to function well epistemically in a social world with widespread divisions of cognitive labour. The epistemic capacities of these inquirers are systematically stymied, even when what the experts are reporting might qualify as cases of reliable knowledge, or useful understanding.
7. Indicators of Epistemic Untrustworthiness
Building on my discussion of the competence and sincerity/care conditions, there are at least three different types of indicators that could suggest a lack of trustworthiness on the part of a community of scientific experts from the perspective of a specifically situated lay person. The first concerns a history of failing to provide significant knowledge for a particular lay community by having gotten things wrong, especially with respect to areas of knowledge that are particularly relevant for a marginalized group. When this becomes evident, the reliability of the community and its research practices are called into question. For example, feminists have drawn attention to the fact that sexist biases and background assumptions have played a significant role in the history of research on women's sexuality, an area of research that could produce significant knowledge for women. The use of such assumptions has resulted in a pattern of mistaken understandings and areas of ignorance that have been damaging to women.Footnote 22 Similarly, in spite of the repeated discrediting of projects of scientific racism that purport to explain away economic disparities between those of European descent and those of African descent by undertaking and referencing poor quality scientific studies, such projects keep popping up throughout the history of the human and biological sciences.Footnote 23 Evidence that scientists have repeatedly produced theories and results that turn out to be mistaken, especially when this pattern occurs more frequently with respect to particularly relevant knowledge for a specific group, offers reasons for the group's distrust of the relevant scientific community, especially on topics of specific concern for the group. Over time, such scientific communities show themselves to be poorly suited as trustworthy providers of significant knowledge for these groups.
Second, there may be an identifiable history of a community of experts ignoring the interests and concerns of a particular lay community. Often, areas of ignorance are not just gaps in knowledge, but are actively constructed to serve the interests of the dominant.Footnote 24 When there are patterns of the research priorities of scientific communities being consistently directed toward the needs of the dominant, without attention to the needs of subordinated groups, again, there is reason for members of such groups to distrust those communities epistemically. Histories of androcentric assumptions in medical science, such as the research on heart disease that was performed on all male subjects rather than researchers attending to the possibility of sex-specific manifestations of the disease offers an example here.Footnote 25 Again, researchers have been failing as successful providers of significant knowledge for the group (women), but this time the failure is due to the inability to generate knowledge that is of particular significance for the group in question.
These first two indicators that might support the distrust of a marginalized community both stem from failures in the core competency condition of being a provider of significant knowledge. The second indicator – ignoring the epistemic needs of a particular group – can also be said to fail to satisfy the sincerity/care condition. Particularly in cases where the history of the expert community's knowledge-making indicates a gross amount of care and attention offered to other (more dominantly situated) people relative to the attention offered to epistemic concerns of the marginalized, the question arises of whether there has been adequate care and attention directed toward the epistemic interests of the marginalized in order to generate the types of knowledge that are important for them.
In contrast, the third type of indicator I suggest primarily involves the sincerity/care requirement. This indicator concerns actual histories of ethical and social injustices involved in the production of knowledge. If there is evidence that members of one's group have been mistreated within the institutions and practices of knowledge production (particularly if there is a historical pattern), this offers some reason to think that these institutions lack a sufficient dose of the moral attitude of care required to sustain making oneself vulnerable by trusting them epistemically on matters of importance. I noted early on that many will find it initially implausible that ethical injustices perpetrated by epistemic institutions (such as the ethical abuse of research subjects, or more generally placing stakeholder populations at risk) should be relevant to the basis one has for forming or maintaining epistemic trust. However, through the articulation and expansion of the conditions of competency and sincerity/care, the connection between a history of ethical injustices and the epistemic case is now more apparent.
Though her work is not framed in terms of epistemic injustices, Naomi Scheman has drawn attention to the ‘systematically trust-eroding effects of various forms of social, political, and economic injustice’Footnote 26 all of which can be identified in a variety of the institutions and practices of science. For Scheman, social injustices within the communities and social practices of knowledge generation that have epistemic relevance can include histories of particular groups having suffered ethical abuses as research subjects, or having being discriminated against in the entry to and participation in the institutions of science. As she writes, ‘the credibility of science suffers, and, importantly, ought to suffer…when its claims to trustworthiness are grounded in the workings of institutions that are demonstrably unjust – even when those injustices cannot be shown to be responsible for particular lapses in evidence gathering or reasoning’.Footnote 27 Her point is not that the ethical injustices she cites imply that scientific institutions cannot serve as reliable truth trackers, but rather that what matters for grounding trust is whether or not variously situated laypersons outside of science can justifiably think they can serve as such.Footnote 28 The cases Scheman describes are cases of epistemic trust injustices, whereby a history of social conditions suggest the epistemic untrustworthiness of an expert community toward a marginalized group, thereby significantly diminishing opportunities for group members to gain important knowledge through a responsibly-formed trust.
8. The Harms of Epistemic Trust Injustices
There is in fact evidence that certain social groups are more distrustful of some scientific communities than others: many indigenous groups distrust geneticists and their research that seeks to identify genetic markers affiliated with their groups,Footnote 29 and African Americans are more distrustful of medical researchers than are whites, often citing the history of research abuses of African Americans as reasons for their distrust.Footnote 30 I have focused on how, from the point of view of epistemological analysis, some of such socially located distrust can be warranted. Where histories of oppression and social marginalization create conditions that do not support a judgement of the expert community's trustworthiness as a sincere and caring provider of significant knowledge for a marginalized group, the possibility of responsibly-placed trust may elude the group, amounting to its members suffering epistemic trust injustices.
The most obvious harm that results from epistemic trust injustices are the epistemic losses that are incurred by members of these marginalized communities when current expert communities actually are capable of providing significant knowledge for them, but where the poor track record of historical interactions leads the marginalized lay persons to dismiss the experts (and reasonably so). Quite simply, such social conditions that result in poor trust relations make it harder for members of such communities to obtain knowledge that is otherwise in circulation.
More prominently, I have argued that compounding those independent epistemic losses is the harm of impediments to one's epistemic agency. When the epistemic losses are systemic, one's epistemic agency can itself be threatened or diminished. As I outlined earlier, epistemic agency itself can only be said to be thwarted when the effects of the warranted distrust prevent significant areas of knowledge acquisition – areas crucial to one's ability to understand oneself and one's situation. But in such cases, these conditions of injustice can begin to undermine one's general capacity to succeed as an independent inquirer within the maladapted social network. Though any expert community might fail to be trustworthy, or could face a situation where a particular lay community lacks reasons to trust it, the harm of thwarting epistemic agency is reserved for cases where such expert communities dominate the social networks of knowledge production and circulation. In modern society, communities of scientific expertise are plausible candidates.
Furthermore, beyond the losses that can occur in one's role as a recipient of knowledge, epistemic trust injustices can circle back, with recurrent effects on the participation of members of marginal groups in the production of knowledge – that is, in their contributory roles. For example, if a member of a marginalized community is distrustful of a scientific research community, they are less likely to be eager to participate in the practices of that community. This will make it more challenging for research communities to correct some of the social and epistemic injustices that played a role in creating the very situation of the epistemic trust injustice in the first place! Those who don't trust medical research communities are less likely to pursue such a research career, and they are unlikely to be eager to take on the role of a research subject. Indeed, as medical researchers have recognized the importance of a diverse human subject pool for robust knowledge generation, they have found it challenging to encourage greater research subject participation across diverse populations, something that is necessary if they are to correct for some of the subject biases in their research.Footnote 31 Distrust of scientific institutions may even lead to a dissociation from educational opportunities (such as high school or college level STEM classes) that could be important for significant epistemic endeavours besides research careers (as teachers, policy makers, or engineers for example). In these and other ways, epistemic trust injustices can interact with the kinds of participatory epistemic injustices I mentioned at the outset, all of which can combine to create quite a difficult situation for the social production and circulation of knowledge, with certain groups bearing the burden of the losses and harms more than others.
Though I have focused on epistemic harms, there are of course innumerable social harms that can also result from epistemic trust injustices. When trust in knowledge-producing institutions is low within one's community, the practical benefits of putting those knowledge results into use are lost. If one does not trust the knowledge claims being circulated by the medical community, or one rejects the claims of government environmental regulators concerned about pollutants in local drinking water, one could even be putting one's life at risk if in fact these expert communities are reliable sources. Such social harms offer another reason why trust relations in the epistemic realm need to be taken seriously; epistemic trust injustices can re-inscribe social inequities when the opportunities that knowledge acquisition can foster are not taken advantage of by the victims of epistemic injustices.
9. Addressing Epistemic Trust Injustices
As epistemic agents, we inherit our places within a network of social epistemic practices that have histories of successes and failures – histories that shape our own epistemic capacities and opportunities. The various epistemic injustices that result from these practices make epistemic success harder on some than others, and epistemic success is often tied to other forms of flourishing.
The variety of harms associated with epistemic trust injustices, and the feedback loops between recipient-based epistemic trust injustices and contributory epistemic injustices create a challenging situation for those who seek to correct both epistemic and broader social injustices. Yet the same connections between the different forms of epistemic injustice that make the problem so difficult also offer pathways forward. Though historical patterns of institutional failures as providers of significant knowledge for particular groups serve as evidence against offering up epistemic trust to these institutions, evidence is defeasible, and there other sources of evidence could arise that would push in a different direction. What is needed to counter warranted distrust is evidence that the epistemic practices have somehow changed, or that the previously untrustworthy epistemic community has taken on new commitments that suggest a greater epistemic accountability to one's marginalized communities. For example, on the participatory side, many scientific communities are beginning to take seriously the need to diversify both their practitioners, and their pools of research subjects. If members of marginalized groups see improvement in these institutions in terms of participation by similarly situated persons, or if they see increased efforts to involve their community in determining the direction and parameters of its research goals, this could mark a difference in the institution's practices that could be relevant to trust relations. A research community that increases diversity within its ranks could plausibly develop a greater awareness of the epistemic needs of relevant groups, and demonstrate greater care, concern and motivation for addressing these epistemic needs. No doubt, the burden is on the institutions themselves to not only make the changes to their practices, but to communicate new evidence that their practices have shifted so as to be better positioned to serve specifically situated communities.
One thing that should be clear from my discussion of epistemic trust injustices is that they are highly contextual. I have not argued that any and every marginalized group in society suffers from epistemic trust injustices. In fact, my argument depends on the social conditions that produce the reasons for distrust being fairly systemic and wide-ranging, but they can be quite specific to a particular group. We must look closely to the specific historical context of the relations between particularly situated lay communities and specific epistemic institutions, seeking out the failures and trust betrayals that have resulted in epistemic trust injustices and made it challenging to turn those trust relations around to foster healthier and more epistemically fruitful networks of epistemic trust. This is difficult work. Yet given that for each of us, the vast majority of knowledge that is significant to us can only be accessed through trust, it should be clear that working against epistemic trust injustices is an important part of achieving epistemic justice for all, and social justice for all.
