Epistemic Agency and the Value of Knowledge and Belief Feminist Philosophy Quarterly Volume 8 | Issue 1 Article 2 2022 Epistemic Agency and the Value of Knowledge and Belief Aron Edidin New College of Florida edidin@ncf.edu Recommended Citation Edidin, Aron. 2022. “Epistemic Agency and the Value of Knowledge and Belief.” Feminist Philosophy Quarterly 8 (1). Article 2. mailto:edidin@ncf.edu Edidin –Epistemic Agency and the Value of Knowledge and Belief Epistemic Agency and the Value of Knowledge and Belief Aron Edidin Abstract This paper develops an account that connects the value of knowledge to socially situated agency twice over. We value the exercise of agency in connection with outcomes we value, and true belief is such an outcome. But true belief is valuable to us because belief is the form of information-possession most directly suited to feed into the interpersonal dimensions of our agency, both epistemic and otherwise. In the first place, the credit-worthiness approach to the value of knowledge locates that value in the broad domain of epistemic agency. An account in terms of the exercise of epistemic agency puts us in position to address criticisms of the credit- worthiness account by mobilizing insights from feminist work on relational autonomy. Meanwhile, epistemic agency is an essential component of agency tout court. Effective action requires information. If our agency is paradigmatically exercised in socially embedded contexts, our information must be in substantial measure communicable in sentences—that is, it must take the form of beliefs. Keywords: knowledge, value of knowledge, belief, agency, epistemic agency, relational autonomy This paper develops an account that connects the value of knowledge to socially situated agency twice over. We value the exercise of agency in connection with outcomes we value, and true belief is such an outcome. But true belief is valuable to us because belief is the form of information possession most directly suited to feed into the interpersonal dimensions of our agency, both epistemic and otherwise. My first task will be to work out the connection between feminist thought on relational autonomy and mainstream views of the value of knowledge. Focusing on the socially embedded nature of human agency allows us to develop one mainstream approach in ways that respond to existing objections, deepen our understanding of the values engaged by the approach, and connect the approach to a second, putatively antithetical one. More specifically, the credit-worthiness approach to the value of knowledge locates that value in the broad domain of epistemic agency. An account of the value of knowledge in terms of the exercise of epistemic agency puts us in position to address criticisms of the credit-worthiness account by mobilizing Published by Scholarship@Western, 2022 1 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 insights from feminist work on relational autonomy. Locating the exercise of epistemic agency in the context of a community of interacting agents further connects this development of the credit-worthiness approach with the apparently contrasting approach that connects the value of knowing with the value of being a good informant in social practices of information-sharing. Within feminist epistemology, the value of knowledge has also been questioned. In “What Knowledge Is and What It Ought to Be: Feminist Values and Normative Epistemology,” Sally Haslanger (1999) develops a different “value of knowledge” question than the “mainstream” one. But the connection to socially embedded agency can be extended to Haslanger’s question, which ends up focused on the value of belief. The extension develops a hint that Haslanger herself leaves at the end of the article, connecting the value of belief with the exercise of agency. Agency must be informed. A propositional attitude like belief involves the possession of information in propositional form, the form most directly connected to the possibility of sentential expression. And possessing information in sententially communicable form is particularly valuable in light of the socially embedded nature of human epistemic agency and of human agency more generally. My second task will be to develop this response to Haslanger’s challenge and so to connect the mainstream “value of knowledge” questions to Haslanger’s. Two Questions about the Value of Knowledge and Belief Epistemology is traditionally focused on conditions that are deemed valuable: knowledge, justification by evidence, truth. In “What Knowledge Is and What It Ought to Be,” Haslanger (1999) raises concerns about taking these valuations for granted. The role of gender in our practices of ascribing or denying these conditions might indicate that the values involved are seriously flawed. At the end of the article, Haslanger suggests that traditional analyses of knowledge do indeed capture a condition that we should value. But she only hints at a positive account of this value. Her main purpose is to establish that a substantial positive account is needed. The key, she thinks, will be to establish the value of belief. The value of knowledge may lie in its relation to justified true belief, and the value of justification may lie in its relation to truth; and true belief will be valuable if belief is, because truth is the internal end of belief. That entails that true beliefs are better than false ones, but it doesn’t entail that true beliefs are better than no beliefs at all. So for the whole chain of conditional value-defense to get going, a case for the value of belief is needed. Haslanger is willing to grant, at least tentatively, the further steps in the chain. She focuses on ways that the question of value can keep getting pushed back from knowledge to justification to truth without resolving the worries that motivate the question for her. And she’s particularly keen to distinguish the project of identifying Published by Scholarship@Western, 2022 2 Edidin –Epistemic Agency and the Value of Knowledge and Belief valuable epistemic or doxastic conditions from the project of characterizing as exactly as possible the currently established concept of knowledge. (Part of the achievement of the article is a taxonomy of philosophical projects that situates this distinction.) So she has nothing at all to say about complications that might be introduced by the fact that justified true belief isn’t sufficient for knowledge, as indicated by our intuitive denial of the status of knowledge in Gettier cases. But those complications connect to a second literature on the value of knowledge. This latter discussion has proceeded without reference to the feminist work that provides much of the context for Haslanger’s paper. But there’s a connection to be made, and making it can provide both an improved response to the Gettier-related problem of the value of knowledge and a related response to Haslanger’s challenge concerning the value of belief. In what follows, I’ll begin by briefly describing the Gettier-related problem of the value of knowledge and one prominent category of response to it. I’ll then explain how that direction can be strengthened by connecting it to feminist work on related values. Finally, I’ll make the connection back to Haslanger’s challenge. Epistemic Agency, Relational Autonomy, and the Value of Knowledge First, then, the Gettier-related problem of the value of knowledge. To go briefly over familiar ground, the Gettier problem concerns how to respond to the recognition that a person can believe something true, and have perfectly good reasons for the belief, and yet not know the thing (Gettier 1963). A simple example is this: I arrive in my office and look at the trusty old clock to check the time. The clock reads 10:00, and so I believe that it’s 10:00. And so it is. But unbeknownst to me, the clock stopped at 10 the previous night. I just happened to look at it exactly twelve hours later. So a further condition must be added to those of justified true belief to get a full account of what’s required to know: a succession of proposed conditions and counterexamples focused on our willingness or unwillingness to ascribe knowledge in a wide variety of arcane and artificial conditions. But patterns of ascription don’t seem to be the end of the story here. One question that the clock case raises is: What’s wrong with my situation vis-à-vis my belief that it’s 10:00? We may agree that the verb “to know” doesn’t apply to me in this situation. But that’s not just a neutral fact about what words we use for what situations. It reflects a deficiency in our situation vis-à-vis something we care about. Something important to our cognitive relation to the world seems to be missing in these cases. But what is that? My belief corresponds to the way things are, and I’m assessing the evidence as I should. In this case, it’s just a lucky coincidence that my cogent assessment leads me to true belief, but good luck is good, right? (Concern with luck as a feature of true belief that isn’t knowledge goes Published by Scholarship@Western, 2022 3 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 back to Plato; the Gettier problem extends the concern to cases of justified true belief.) One prominent response is to note that my good luck may be good for me, but it doesn’t reflect any credit on me. To say that I lucked out with respect to some outcome pretty much1 just means that I don’t deserve credit for the outcome. And deserving credit is a good thing. If Gettier cases are ones in which a believer just lucks out in the truth of her justified belief, credit-worthiness for the truth of her belief is something valuable that’s missing in those cases (Pritchard 2007, especially 89, 98– 100, and references cited there). This seems to be on the right track (both in terms of what’s missing in Gettier cases and of what’s valuable about what’s missing), but objections have been advanced that suggest that the focus of the credit-worthiness account isn’t quite in the right place. One objection is that knowledge sometimes involves the “easiest” of true-belief formation—for example, knowledge of easily perceived features of the knower’s immediate surroundings. Duncan Pritchard notes that agents can sometimes know even while having a true belief that is of very minimal credit to them. A question to be answered, however, is this: if the credit involved is this minimal, then in what sense is the credit reflecting a cognitive achievement on the part of the agent? (Pritchard 2007, 99) One response to this sort of objection is to replace the vocabulary of credit and accomplishment with that of attributability: knowledge is valuable because when we know, the truth of our belief is attributable to us. But why is that valuable? It’s easy enough to see why I might like the condition of deserving credit for some good thing. But if attributability is divorced from credit-worthiness, where’s the good in it? One direction, developed by John Greco (2012), is to look is away from value- to-the-knower and toward value-to-the-attributor. Attributing outcomes to sources is part of a process of orienting ourselves toward using the sources to obtain the outcomes and toward intervening with sources to modify outcomes. This also provides guidance in determining just what contribution to the production of the outcome makes the outcome attributable to you: the outcome is attributable to me if focusing on my contribution is a good way of organizing the relevant orientation toward sources and outcomes. Attributing knowledge to someone can play a role in our orientation toward potential sources of information. 1 The two come apart in cases discussed below: very easy accomplishments and ones that result largely from aid from others. I may be worthy of little credit in these cases, but they’re not really cases of lucking out either. Published by Scholarship@Western, 2022 4 Edidin –Epistemic Agency and the Value of Knowledge and Belief But this is at best only a part of the picture. We value knowledge itself, not just correct attributions of knowledge. We take knowledge to be valuable to the knower, not just to others who might include her in their orientation toward potential sources of information. Related to the cases of “easy” knowledge that Pritchard raises are ones in which a knower is aided by technological or interpersonal assistance in ways that suggest the assignment of credit to the source of assistance rather than to the knower.2 One version of this phenomenon involves simple cases of knowledge by testimony. Jennifer Lackey describes the following case: Having just arrived at the train station in Chicago, Morris wishes to obtain directions to the Sears Tower. He looks around, approaches the first adult passer-by that he sees, and asks how to get to his desired destination. The passer-by, who happens to be a Chicago resident who knows the city extraordinarily well, provides Morris with impeccable directions to the Sears Tower by telling him that it is located two blocks east of the train station. Morris unhesitatingly forms the corresponding true belief. (Lackey 2007, 352) The considerations raised by Pritchard and Lackey suggest that we might do better to shift the focus from credit-worthiness to the broader category of agency. The assignment of credit and the attribution of accomplishments to persons identify their targets as agents.3 Whatever value we assign to agency will be tracked by this sort of assignment and attribution. Framing the question this way, we might say that the value of knowing is the value of exercising a certain kind of agency—epistemic agency. Thinking about characteristics of agency in general, and about how we value agency, is likely to help to illuminate what we might value about knowledge. Here’s one example: Agency extends to the doing of very simple, easy things that we accomplish with little or no explicit thought. (Consider walking and talking, eating and drinking, grasping and holding small objects, etc.) There may be little credit-worthiness in these actions, but they can be integral parts of fancier actions that might be more credit-worthy. And the absence of agency in these simple matters can constitute a significant kind of helplessness. All of this seems to apply directly to the epistemic agency involved in perceiving features of our immediate surroundings. In many cases this can be easy and not particularly credit-worthy, but it’s an integral 2 For technological examples, see Vaesen (2011). 3 Sosa (2015) comes very close to this formulation, connecting his account directly to Gettier examples (and Gettier-like examples in connection with other kinds of agency) in chapters 1 and 3. Published by Scholarship@Western, 2022 5 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 part of our epistemic agency more generally, playing a key role in fancier epistemic accomplishments. And conditions that interfere with perceptual access to our surroundings can produce an intense sort of cognitive helplessness. Again, and more centrally to what will follow, we perform both simple and fancier feats of agency by engaging with others. If I ask you to pass the salt and you do so, you may have played a larger role than mine in getting the salt to me; but an environment of cooperative fellows is a favorable domain for getting things done, one in which I can act effectively, not one in which my action is moot. If I ask you where a particular building is and you tell me, I’ve informed myself of the location by engaging your cooperation. Your participation increases the scope of my epistemic agency. Attending to the ways that agency can be facilitated by engagement with others helps to put Lackey’s example into perspective. But it isn’t just an incidental fact about some exercises of agency that they involve engagement with others. Human agency is interpersonally facilitated in most of its many instances. This is brought out dramatically in feminist work on the nature of autonomy, a dimension of agency that has often been conceived in terms of independence from social support on one hand and from influence on the other. For example, Diana Meyers (1989) emphasizes the role of socialization in facilitating the development of the competencies required to make my actions most genuinely my own. Andrea Westlund (2009), reflecting a long tradition of concern about how we can be bound by our own inflexible inclinations, argues that those competencies require the ability to engage dialogically with others whose perspective differs from one’s own. Marina Oshana (2006) notes the ways in which leading a life of one’s own choosing is a matter of gearing into social structures and arrangements that enable, and often actually constitute elements of, the chosen way of life. The lesson of all of this is that the actions that most deeply reflect our authentic individual selves also draw deeply on our relations to others. The social facilitation of epistemic agency in Lackey’s example is an instance of a pervasive feature of human agency. We can get a vivid sense of the depth of social embedding by considering a really homely example of something that most of us do alone—say, brushing our teeth. When I brush my teeth nobody else is helping me. But, first of all, I’m availing myself of tools and substances that were made by others. I couldn’t brush my teeth the way I do if others didn’t make toothbrushes and toothpaste and mirrors and eyeglasses, and engage in practices of distribution, transportation, and sale that put those things in my possession. Meanwhile, I was taught how to brush my teeth by parents and schoolteachers and dental hygienists, and my choice to brush is informed by information I’ve received from all of those. I brush partly for broadly medical reasons, to preserve dental and oral health. But I also brush to make my appearance and the smell of my breath more palatable to people around me, conforming to standards that are in some respects society-wide and in others deeply inflected by Published by Scholarship@Western, 2022 6 Edidin –Epistemic Agency and the Value of Knowledge and Belief gender and class and other dimensions of social positioning. Brushing my teeth is something I do myself; I’m the agent of the activity. But my agency is enabled by and entangled with social (including economic and medical) structures and practices and phenomena of tremendous extent and variety. The profoundly social character of our agency suggests that if knowledge is a matter of agency in a particular (epistemic) domain, it too will be profoundly social. Again, this will come as no surprise to readers of feminist epistemology. Among others, Lynn Hankinson Nelson (1990), Helen Longino (1990), Heidi Grasswick (2004), and Gaile Pohlhaus Jr. (2012) have explored the idea that we inform ourselves by gearing into communal structures and arrangements that contribute concepts, standards of evidence, acceptable presuppositions, and opportunities for testimonial sharing of information and dialogic exploration and questioning. We can get a vivid sense of the depth of social embedding by considering a homely example of firsthand—as opposed to testimonial—knowing. If I recognize the make and model of the car in front of me in traffic, nobody is, on that occasion, actively assisting me in my identification of the car. But first of all, I’m availing myself of tools that were made, distributed, and sold by others. If I weren’t wearing my glasses, I wouldn’t see the car well enough to identify it. And, of course, it’s my car that enables me to get close enough to the other one to identify it. Meanwhile, my ability to recognize vehicles was inculcated and developed in my childhood, first by picture books with titles like Cars, Trucks, and Things that Go, and later by magazines, more detailed books, and schoolyard conversation. All of this was informed by social constructions of gender: these epistemic capacities were particularly valued for boys. So boys who read car magazines and talked about cars on the playground and so got to be good at recognizing makes and models were going with the socially sanctioned flow, whereas girls who did the same were transgressing gender expectations. And in addition to all of this, car models are designed to be recognizable; my epistemic capacity is in this way tied up with marketing strategies in the auto industry. If the value of knowing is the value of exercising epistemic agency, then given the way human agency and human epistemic agency work, that’s going to be the value of participating in appropriate ways in social practices and structures and systems surrounding the acquisition, propagation, and mobilization of true beliefs. Feminist work both in social epistemology and on relational accounts of agency and autonomy connects the value of agency (epistemic and otherwise), and thus the value of knowledge, to participation in such practices and structures and systems. This contributes to our understanding of the value of knowledge in a way related to the “credit-worthiness” approach, by moving from credit-worthiness itself to the more general category of agency. But it also allows us to connect that approach to another that is typically presented as a rival to the credit-worthiness approach. This second approach focuses on the place of knowers in interpersonal practices of Published by Scholarship@Western, 2022 7 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 information-sharing, characterizing knowers as potential good informants and the value of knowledge as deriving from that of being a good informant.4 (Greco’s emphasis on the value of knowledge-attribution fits this pattern.) But being an informant is another way of participating in social practices, structures, and systems surrounding the acquisition, propagation, and mobilization of true beliefs. And inclusion as a full-fledged participant in such practices is a social good for the one included, so this elaboration of the value of knowledge-attribution does identify a value of knowledge to the knower and not just to others. At this point, the two approaches seem to focus on two different kinds of relation to this one overall structure. As socially supported agents, knowers draw on the structure. As potentially valuable informants, they contribute to it. So the two approaches look like two sides of the same coin. But the sides are interconnected. The draws help to inform the contributions, and the contributions enrich the source of the draws. We might think of interconnected mutual enrichment along the lines of productive conversation, in which the roles of contributor and recipient are not really distinct. In that case, the two approaches are more like one side of the same coin, or perhaps two sides of the same Möbius strip. One way or the other, they’re closely related. And thinking about the details of the relation will be a matter of sorting out dimensions of social embedding of epistemic agency. So we can think of the fundamental value of knowledge in terms of its relation to epistemic agency. But it can’t quite be that the value of knowledge is that it entails the exercise of epistemic agency with respect to the true belief involved. It’s possible to know in cases in which the true belief in question isn’t the result of the knower’s successful exercise of epistemic agency.5 Suppose there’s something I don’t want to know: I’ve recorded a televised contest, and I don’t want to know who won until I have the chance to watch. You know the winner. I ask you not to tell me. I cover my ears so I can’t hear what you say. But you pull my hands away from my ears and tell me who won. Now, to my disappointment, I know. But it would be perverse to describe my true belief as the result of successful exercise of my epistemic agency. It’s important to note first of all that in this case, it seems reasonable to say that the characteristic value of knowledge is absent. This is, after all, a case in which I take myself to be worse off for knowing. And that’s certainly related to the fact that my knowing is contrary to my agency. To that extent, the example remains a good fit for the agency-focused account of the value of knowledge. But even with respect to values based on agency, cases of knowledge-against-the-knower’s-will raise instructive complications. 4 See Pritchard (2007, 97–98, and sources cited there). 5 I owe the following example to Jon Matheson. Published by Scholarship@Western, 2022 8 Edidin –Epistemic Agency and the Value of Knowledge and Belief Regarding my agency in the case I’ve described, the picture is mixed. Something has gone badly wrong: I’ve been informed against my will. But the case remains closely related to ones in which I successfully inform myself by drawing on the knowledge of others. As we’ve already seen, learning from others is an important domain of exercise of my epistemic agency. When I learn successfully in this way, I seek information from others, and they tell me what they know. In this case, we have the second part without the first. Can we make sense of this combination in a way that continues to center on agency? One way to do so is suggested by Diana Meyers’ idea of autonomy competencies. Meyers (1989) characterizes autonomous actions as ones that exercise a variety of underlying competencies. Something similar seems to apply to agency quite generally. We can further characterize the variety of relevant competencies by noting that autonomy and successful agency require the exercise of competencies relating both to the making of decisions (understood very broadly as the sense of self- direction) and to the execution of decisions we’ve made. Suppose I bake a delicious cake. My agency in producing the result can be compromised in either of two ways. On one hand, I might be an incompetent baker. I make a well-considered decision to bake a cake, but the successful result is sheer accident given my deficiencies in the kitchen. My agency in producing the cake is compromised in a different way if I am forced to bake it, bypassing the exercise of my capacities in the dimension of deciding what to do. For the baking of the cake to constitute a successful exercise of agency on my part, the outcome must be appropriately related to my exercise of both decision- making competencies and decision-enacting competencies.6 In the contest-winner case, it’s the sidelining of my self-directing competencies that compromises my agency. The belief is a product of my competent uptake of the testimony of a reliable informant, but my exposure to the testimony is forced; my decision-making is bypassed.7 So one possibility is that knowledge is present when the true belief in question was acquired through the successful exercise of decision-enacting competencies relevant to agency in the acquisition of the belief, even if the exercise of the relevant decision-making competencies is compromised. The special importance of decision-enacting competencies to the presence or absence of knowledge suggests a more nuanced account of the value of knowledge as a component of agency. Successful exercise of agency must of course be informed; 6 Note that the social dimensions of agency that I discussed above apply to both sorts of competency. 7 Note that decision-making competencies can be engaged negatively, by not doing anything to block ambient sources of information and so remaining open to their deliverances. But in this case, my actual attempt to isolate myself from the information is frustrated. Published by Scholarship@Western, 2022 9 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 my agency may be compromised if I’m misinformed or lack relevant information, even if the outcome of my action is in accord with my decision.8 (This requirement applies to both decision-making and decision-enactment.) In Gettier cases, where knowledge is absent, the use of the (true) belief in question to inform further action and inference seems to involve this sort of compromise. If I truly believe that it’s 10:00 because I’ve seen a clock that stopped at 10:00 twelve hours before, and I act on that belief by rushing to a 10:00 meeting, my action has the desired result by accident; I get to the meeting in spite of the fact that my information is compromised. But if I act in ways that are informed by my true belief about the winner of the contest (perhaps making the best of the fact that I wasn’t able to avoid hearing about it), there is no such problem. True beliefs that result from the successful exercise of the enactment component of epistemic agency competencies feed forward into further agency, epistemic and otherwise. Even if I wish I didn’t know the winner of the contest, my knowledge still has this value to me. This special “forward-feeding” role of the exercise of enactment competencies seems to apply more generally than to epistemic competencies alone. If I’m forced to bake a cake and competently do so, and then find uses of my own for the cake I’ve baked, my agency in the further use of the cake isn’t compromised by the fact that I was forced to bake it. My enactment competency allowed me to forge a tool that (as it turned out) I could use for my own purposes. Indeed, you may have forced me to bake the cake in order for me to have it available for further use. (Parents do this sort of thing a lot.) But if the decision to bake was my own, but my incompetent baking meant that it was just by accident that I wound up with an edible cake, the accidentalness carries forward to further use of the cake, even if I’m wholly competent in that use. (If I’ve baked the cake to serve for your enjoyment, and I serve you the cake competently and you enjoy it, my agency in relation to your enjoyment remains compromised by the fact that what makes the cake enjoyable is the accidental product of my incompetent baking.) Meanwhile, the forward-feeding capacity that’s tied to the successful exercise of the enactment component of epistemic agency competencies also connects the agency-focused approach to the value of knowledge to the good-informant-focused approach. One way of putting the forward-feeding is to say that in the cases in question, even though my agency has been compromised in its decision-making components, the outcome still leaves me a good informant to my future self in the matter in question. My future actions and inferences aren’t less well informed because of the compromise. And this carries over to my capacity as an informant to others. 8 This is discussed in considerable detail in Sosa (2015, chapters 6 and 7). Published by Scholarship@Western, 2022 10 Edidin –Epistemic Agency and the Value of Knowledge and Belief If we identify the characteristic value of knowledge with the exercise of the enactment component of epistemic agency competencies and its consequent positioning with respect to future exercise of agency, we get an account that applies not just to typical cases but to cases of knowledge against the knower’s will. The relevance of social dimensions of agency, and the relation to practices of information sharing, carry over directly from the simpler identification of the value of knowledge with the exercise of epistemic agency with respect to the information known. And we needn’t give up the idea that this value typically takes a form that involves the knower’s exercise of epistemic agency with respect to that information. We keep ourselves informed about the state of our surroundings in large part by maintaining a comportment of perceptual openness to potentially salient features. Part of what makes the easy cases of perceptual self-informing so easy is that the relevant self- direction needn’t extend beyond the casual maintenance of this comportment. With respect to information that might engage my epistemic enactment-competencies, the default situation is one in which the relevant decision-competencies are (at least thus minimally) exercised. Only if I actively attempt to disengage my normal epistemic openness will the possibility arise of knowledge without the exercise of epistemic agency. This won’t be a vanishingly rare occurrence; I will, for example, sometimes act to shield myself from unwanted distractions. But this is itself a response to the default condition of willing openness, and in the default condition, my epistemic self- direction is engaged in a way that makes any consequent acquisition of knowledge an exercise of epistemic agency. Social Dimensions of Agency and the Value of Belief So much (for now) for the part of the project that relates some of the preoccupations of feminist epistemology and autonomy theory to mainstream, Gettier-related discussions of the value of knowledge. To make the connection back to Haslanger, we can note to begin with that the approaches I’ve discussed to the Gettier-related problem presuppose the value of true belief. When we know, we’re typically the agents of our own cognitive good in the form of true belief. At minimum, we’ve exercised the enactment competencies involved in such agency, in ways that can feed into subsequent exercises of (epistemic and other) agency. This agency has value beyond the value of the goods of which we’re the agents. We value the agency along with the outcome. But the value of agency isn’t independent of the value of the outcome. If we don’t care about true belief, then we won’t care in the same way about being the agents of our true belief. So the value of true belief is centrally connected to the solution I’ve proposed to the Gettier-related problem. Haslanger is sympathetic to the idea that truth is a value essentially connected to belief itself, that belief essentially aims at truth. But she notes that this confers value on true belief only if belief itself is valuable. Truth is the internal goal of belief. Published by Scholarship@Western, 2022 11 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 That is a reason to prefer true belief to false belief, but it isn’t a reason to prefer true belief to no belief. So Haslanger proposes that we approach the question of the value of true belief by way of the value of belief. This all happens near the end of a paper that doesn’t go on to a detailed treatment of that question. But it does include a hint: Haslanger suggests that the value of belief stems from its essential role in agency, for beings like us. Indeed, in a closely related paper, Haslanger (2000) suggests that attending to many of the social features of knowledge and agency, and to the feminist literature exploring them, might help to shed light on the issue. To develop the hint, one starting point is the familiar fact that agency must be informed. I can effectively control what’s going on only if I can tell what’s going on and track the effects of my interventions. This much seems built into the very idea of agency. And it carries with it the idea that agency with respect to the relevant kinds of information is part and parcel of agency tout court. My agency is strengthened by the ability to inform myself in relevant respects and weakened by the inability to do so. This role distinguishes belief from some potential alternatives. Haslanger proposes accepting propositions as a possible alternative to believing them. But acceptance that doesn’t involve belief—say, accepting a proposition as a working hypothesis—doesn’t do this work of tracking our actions and their outcomes. It doesn’t in the appropriate way involve the possession of information needed for effective agency. One can, of course, act on a working hypothesis; that’s how working hypotheses work. But when we act on a working hypothesis, the action, like the hypothesis, is in a sense provisional. We try acting on the hypothesis and see how that works out. “Seeing how the hypothesis works out” itself involves tracking outcomes in a way that must go beyond the generation of further working hypotheses about how the initial one is working. This needn’t involve anything like certainty, but it requires a verdict rather than a further hypothesis. “So far, so good” can be one legitimate verdict, though if no evaluation on any level ever gets beyond “so far, so good,” that will constitute a limitation on one’s agency in the situation. But the “so far, so good” must itself be a verdict, the object of belief-like commitment. The tracking of outcomes can be tentative, but it needs to be tracking (however tentative) of what’s actually going on, not just the generation of further hypotheses (which may be accepted as working hypotheses) about what’s going on.9 9 A rather different version of acceptance is Catherine Elgin’s (2004) notion of accepting a proposition as “true enough.” This form of acceptance is a verdict, of a kind that can inform action. But Elgin’s case for the appropriateness of this form of acceptance for some propositions assumes belief in others (she doesn’t propose acceptance as a universal replacement for belief), and acceptance as true enough involves a belief-like essential aim of being true enough, which leads to a notion of Published by Scholarship@Western, 2022 12 Edidin –Epistemic Agency and the Value of Knowledge and Belief This doesn’t by itself mean that agency requires beliefs, because belief is only one way of possessing information. So what distinguishes beliefs from other carriers of information? A familiar formulation characterizes beliefs as propositional attitudes; the information captured in beliefs is propositional in form. And this in turn suits the contents of belief to sentential expression. Discussions of belief often emphasize the grammatical form of belief attributions, which involve “that clauses” with sentential complements: “So-and-so believes that p,” where what replaces “p” is a sentence. Paul Churchland (2013) takes this as the key to understanding belief attribution: beliefs are states whose content is closely related to the content of sentences, so belief content can typically be reported in sentential form. What sets beliefs apart, on Churchland’s understanding, is precisely their suitability to sentential expression. On other approaches than Churchland’s, the relation between propositional attitude and sentential expression may be more indirect.10 Even on Churchland’s approach, it doesn’t follow that every belief is sententially expressible. But one way or another, propositional attitudes are the form of information storage most directly linked to sentential expression. Why should beings like us want our agency to be informed in ways that are expressible sententially? Well, for beings like us, agency (including agency with respect to information) is deeply social. Our agency in general is supported by social connections that depend on sharing of information. Our ability to inform ourselves itself depends heavily on structures and practices of information-sharing. And the primary way beings like us share information is with language, with sentences. The connections that support our agency depend on an extensive ability to say what we know. We certainly possess information in forms that we can’t effectively express in sentences. But it’s essential that lots of the information we possess can be told. For beings like us, who depend on social resources for our agency and on language for the constitution and sharing of those resources, belief—that is, information in tellable form—is essential to the agency we value. Churchland contemplates the possibility of agents who communicate differently, say by way of direct brain-to-brain links, without the need of language. For such beings, even if their agency required social support as ours does, there might be no particular value to having beliefs. Churchland supposes that this sort of communication might be vastly superior to our sentential kind, and he presents it as an exciting possibility for further human development. But for agents like us present- day humans, language, and therefore sentences, and therefore beliefs, are essential “true enough” knowledge to which the considerations in this paper apply mutatis mutandis. 10 Though both language-of-thought accounts like Fodor’s (1978) and attributability accounts like Dennett’s (1971) keep the connection very close indeed. Published by Scholarship@Western, 2022 13 http:indirect.10 Feminist Philosophy Quarterly, 2022, Vol. 8, Iss. 1, Article 2 to the sharing of information, which is itself an essential part of the ways in which we act socially to support our own agency and that of others. So belief, and thus true belief, is valuable to beings like us. And agency with respect to this valuable outcome is valuable. But the connection is tighter than just that: the outcome is valuable in part because of its potential contribution to the agency. True belief is valuable because it’s the form of information best suited to sharing in support of our collective epistemic endeavors, and of our collective endeavors more generally. It’s the social dimension of epistemic agency that in this second way connects the value of knowledge with the value of belief itself. The value of knowledge is that of epistemic agency and of epistemic contribution to agency more generally, and our agency is exercised in an interpersonal context that depends on sentential communication of information, and thus the possession of information in the form of beliefs. References Churchland, Paul. 2013. Matter and Consciousness. 3rd ed. Cambridge, MA: MIT Press. Dennett, Daniel C. 1971. “Intentional Systems.” Journal of Philosophy 68, no. 4 (February 25): 87–106. Elgin, Catherine Z. 2004. “True Enough.” Philosophical Issues 14:113–31. https://doi .org/10.1111/j.1533-6077.2004.00023.x. Fodor, Jerry. 1978. “Propositional Attitudes.” Monist 61, no. 4 (October): 501–23. Gettier, Edmund. 1963. “Is Justified True Belief Knowledge?” Analysis 23, no. 6 (June): 121–23. https://doi.org/10.1093/analys/23.6.121. Grasswick, Heidi. 2004. “Individuals-in-Communities: The Search for a Feminist Model of Epistemic Subjects.” Hypatia 19, no. 3 (Summer): 85–120. https://doi.org /10.1111/j.1527-2001.2004.tb01303.x. Greco, John. 2012. “A (Different) Virtue Epistemology.” Philosophy and Phenomenological Research 85, no. 1 (July): 1–26. Haslanger, Sally. 1999. “What Knowledge Is and What It Ought to Be: Feminist Values and Normative Epistemology.” Philosophical Perspectives 13:459–80. https://www.jstor.org/stable/2676113. ———. 2000. “Defining Knowledge: Feminist Values and Normative Epistemology.” In The Proceedings of the Twentieth World Congress of Philosophy, Volume 8: Contemporary Philosophy, edited by Daniel Dahlstrom, 41–55. Bowling Green, OH: Philosophy Documentation Center. Lackey, Jennifer. 2007. “Why We Don’t Deserve Credit for Everything We Know.” Synthese 158, no. 3 (October): 345–61. https://doi.org/10.1007/s11229-006- 9044-x. Published by Scholarship@Western, 2022 14 https://doi.org/10.1007/s11229-006 https://www.jstor.org/stable/2676113 http:https://doi.org https://doi.org/10.1093/analys/23.6.121 https://doi Edidin –Epistemic Agency and the Value of Knowledge and Belief Longino, Helen. 1990. Science as Social Knowledge: Values and Objectivity in Scientific Inquiry. Princeton, NJ: Princeton University Press. Meyers, Diana T. 1989. Self, Society, and Personal Choice. New York: Columbia University Press. Nelson, Lynn Hankinson. 1990. Who Knows: From Quine to a Feminist Empiricism. Philadelphia: Temple University Press. Oshana, Marina. 2006. Personal Autonomy in Society. New York: Routledge. Pohlhaus, Gaile, Jr. 2012. “Relational Knowing and Epistemic Injustice: Toward a Theory of Willful Hermeneutical Ignorance.” Hypatia 27, no. 4 (Fall): 715–35. https://doi.org/10.1111/j.1527-2001.2011.01222.x. Pritchard, Duncan. 2007. “Recent Work on Epistemic Value.” American Philosophical Quarterly 44, no. 2 (April): 85–110. https://www.jstor.org/stable/20464361. Sosa, Ernest. 2015. Judgment and Agency. Oxford: Oxford University Press. Vaesen, Krist. 2011. “Knowledge Without Credit, Exhibit 4: Extended Cognition.” Synthese 181, no. 3 (August): 515–29. https://doi.org/10.1007/s11229-010- 9744-0. Westlund, Andrea. 2009. “Rethinking Relational Autonomy.” Hypatia 24, no. 4 (Fall): 26–49. https://doi.org/10.1111/j.1527-2001.2009.01056.x. ARON EDIDIN is a professor of philosophy at New College of Florida. He works primarily in epistemology (most recently, social epistemology) and in the philosophy of music (focusing on characterizing the artistic collaboration of performers and composers in Western classical music), but he has also published in philosophy of science and mathematics, philosophy of mind, metaphysics, and metaphilosophy. Published by Scholarship@Western, 2022 15 https://doi.org/10.1111/j.1527-2001.2009.01056.x https://doi.org/10.1007/s11229-010 https://www.jstor.org/stable/20464361 https://doi.org/10.1111/j.1527-2001.2011.01222.x 13641 Edidin title page 13641 Edidin final format