Epistemic Authority, Autonomy, and Humility

UC Irvine UC Irvine Electronic Theses and Dissertations Title Epistemic Authority, Autonomy, and Humility Permalink https://escholarship.org/uc/item/99d4t6zq Author Popowicz, Dylan Mirek Publication Date 2019 Copyright Information This work is made available under the terms of a Creative Commons Attribution License, availalbe at https://creativecommons.org/licenses/by/4.0/ Peer reviewed|Thesis/dissertation eScholarship.org Powered by the California Digital Library University of California UNIVERSITY OF CALIFORNIA, IRVINE Epistemic Authority, Autonomy, and Humility DISSERTATION submitted in partial satisfaction of the requirements for the degree of DOCTOR OF PHILOSOPHY in Philosophy by Dylan Mirek Popowicz Dissertation Committee: Professor Sven Bernecker, Chair Professor Annalisa Coliva Professor Duncan Pritchard Professor Karl Schafer 2019 © 2019 Dylan Mirek Popowicz DEDICATION In memory of my father, Miroslaw Popowicz: my first authority and exemplar in all too many ways. For my mother, Alison Toon, who quietly exemplifies the many virtues that I have failed to learn, and who has always put her children first. And for Rebecca Tristan Lacy, who lovingly sacrificed far too much for the sake of my intellectual journey: this dissertation is hers as much as it is mine. ii TABLE OF CONTENTS Page ACKNOWLEDGEMENTS v CURRICULUM VITAE vii ABSTRACT OF THE DISSERTATION viii INTRODUCTION: Social Epistemology and the Importance of Authority CHAPTER I CHAPTER II CHAPTER III 1 Epistemic Authority Thus Far 13 I.1 Zagzebski’s Account 16 I.2 Dormandy’s and Jäger’s Criticisms of Preemption 26 I.3 Dormandy’s Proper-Basing View 41 I.4 Jäger’s Socratic Epistemic Authority 45 I.5 Constantin and Grundmann’s Limited Preemption 52 Epistemic Authority 59 II.1 More than Mere True Belief 60 II.2 Beyond True Belief: A Variety of Epistemic Goods 78 II.3 Expertise and Epistemic Authority Defined 84 II.4 In Contrast: Michel Croce’s Account of Epistemic Authority 107 Epistemic Authority and Autonomy 115 III.1 A Brief Look at Personal Autonomy 124 III.2 Considerations on Epistemic Autonomy 129 III.3 Authority and Minimal Epistemic Autonomy and Responsibility 138 III.4 Cognitive Limitations and the Various Degrees of Epistemic Involvement 150 III.5 Doing What One Can: Epistemic Responsibility and Epistemic iii Laziness in Deferring to Authority CHAPTER IV 156 III.6 Epistemic Autonomy is Not Enough: Epistemic Helplessness in Deference 163 Epistemic Authority and Humility 165 IV.1 Humility in Deference 169 IV.2 Accounts of Humility 175 IV.3 Against Arrogance: Benefits of Epistemic Humility 188 IV.4 Towards Virtuous Epistemic Authority 198 BIBLIOGRAPHY 201 iv ACKNOWLEDGEMENTS The completion of this dissertation, a record of my thoughts on the nature of epistemic authority, would not have been possible if not for the individuals that have guided my philosophical education. As models of the phenomenon in question, they have inspired me. These exemplars of the philosophical tradition have been my epistemic authorities throughout: they have provided me with the guidance, criticism, and insight crucial to my project’s success. My deepest appreciation goes to my committee chair and advisor, Professor Sven Bernecker. His wisdom as an advisor has manifested itself in a plethora of ways. He has always been there to offer his guidance, while providing me with the room to make my own discoveries and encouraging me to develop my own voice as a philosopher. It has been remarkable how he has repeatedly given me precisely the kind of assistance that I needed to continue: whether that be an epistemic prod in the right direction, an inspiring suggestion for a research project, a criticism that has led to growth, a small nod of encouragement, or an act of recognition. Perhaps most importantly, he has given me a genuine ear, and provided human conversation in good faith when things have been personally difficult for me. I would like to thank my committee members, Professor Annalisa Coliva, Professor Duncan Pritchard, and Professor Karl Schafer. All three joined the University of California, Irvine, after my admittance as a graduate student, and I will always be thankful that they did. Their presence has fostered such growth in learning and community, that I cannot imagine life here without them. Prof. Coliva’s transparency and sincere commitment to the future of the department, and particularly to the well-being of its graduate students, has had a profound effect on my life in academia: she has encouraged me to look to the future with more conviction and hope than I previously had. As a philosopher, she has inspired me to pursue my own voice and perspective, and to not be afraid to consider the broader picture, and broader applications, when pursuing my own theories. This dissertation is greatly indebted to discussions that she, Adam Sanders, and I had as part of an independent study focused on social epistemology. I am indebted to Prof. Pritchard for always holding my feet to the fire, albeit in the kindest fashion. I am in awe of the wealth of his knowledge, and always look forward to his criticisms: they teach me both where I have erred in understanding, and direct me to pockets of the vast landscape of epistemological scholarship that I have yet to be properly acquainted with. My dissertation would have been far weaker if he hadn’t shown me the kindness of joining my committee on such short notice, and with no prior experience of my work or person to guide his decision. The consistent manner in which Prof. Schafer provides me with comments and criticisms that in themselves suggest entirely new lines of research, pregnant with the possibility of novel and alternative dissertations, continues to astound me. Where once I thought that all had been done and said, he has shown me just how deep the rabbit hole goes, and how profoundly complex the logical space from which we pluck our ideas really is. I am incredibly thankful for his presence on v my committee. My philosophical education will always be indebted to Prof. David Lopez, of American River College, Sacramento. I spent my late teens and early twenties very much at a loss as to exactly how and where I should apply my passion for scholarly pursuits. Looking back at my indecision, I have to credit Prof. Lopez for inspiring me to finally settle down, and to plant my feet solidly within the philosophical tradition. A very special thanks goes to Prof. G. Randolph Mayes of California State University, Sacramento. Prof. Mayes continues to inspire me as an exemplar of the good life. I am indebted to him for more reasons than I could count. I arrived in his epistemology class as a confused student primarily acquainted with a more ‘continental’ philosophical tradition—he spent more time and care teaching me how to write a philosophy paper than any student could reasonably ask. His sincere commitment to his students’ learning inspired me to seek out the opportunity to serve students in the same capacity; though I had no confidence in my ability to pursue such a future, his encouragement paved the way. His authenticity as a human being—in his role as academic, pedagogue, colleague, and friend—continues to remind me of the better person that I could become. My gratitude goes to my cohort. I will always be thankful for Kourosh Alizadeh, Adam Bobella, Jacob Heim, and Adam Sanders: they inspired me, as examples of the best of scholars, but more importantly as good people. Their virtues are those that I respect the most: they are humble, sincere, open-minded, and conscientious. Though I have too often been perturbed by the lack of such virtue, and a sometimes-cold antipathy, to be found in academia, these individuals have always reminded me of what a more ideal academic community could be like. I am especially thankful for Adam Sanders, who, though he may not have always realised it, has helped me through some of the darkest times in my life. I am incredibly proud to be acquainted with, and a member of, this cohort. I would not have survived my five-year residence as a graduate student if it was not for the luck I had in being part of such a wonderful group. Outside of my cohort, I must especially thank Itzel A. Garcia and Nicholas Smith. Nick led me to the barbell, proffering a healthy coping mechanism in place of old and destructive habits; he kept me company even when my nature led me far from the path of virtue and wisdom. Itzel showed me the deepest form of love and kindness when I most needed it: a shining light at my darkest times. She exemplifies the best of us: in her bravery, sincerity, compassion, and authenticity. My thanks go to all those that have helped me at American River College, California State University, Sacramento, and University of California, Irvine. My time within these communities has been long, and I only regret I cannot individually thank each person that has touched my life. I am thankful to all of the individuals listed above for having been there to support me, when I have all too often failed to be my best, and failed to overcome my own weaknesses. The writing of this dissertation would of course not have been possible without the financial and scholarly support of the University of California, Irvine. vi CURRICULUM VITAE DYLAN MIREK POPOWICZ 2011 A.A.s. in General Education, Social Sciences, and Liberal Arts, American River College, Sacramento 2013 B.A. in Philosophy, California State University, Sacramento 2014-18 Teaching Assistant, Department of Philosophy, University of California, Irvine 2017 M.A. in Philosophy, University of California, Irvine 2018-19 Lecturer, Department of Philosophy, Chapman University 2019 Ph.D. in Philosophy, University of California, Irvine vii ABSTRACT OF DISSERTATION Epistemic Authority, Autonomy, and Humility By Dylan Mirek Popowicz Doctor of Philosophy in Philosophy University of California, Irvine, 2019 Professor Sven Bernecker, Chair As social creatures, we lean on the work of others to epistemically situate ourselves in our complex environments. A particularly important instance of our epistemic reliance on others is our dependence on epistemic authorities: those who hold some normative power over our epistemic behaviour. Often, accounts of such epistemic authority suggest that it is constituted by a connection to a certain set of true beliefs. I provide an alternative account that roots epistemic authority in an individual’s superior ability to participate in a certain kind of epistemic practice. Such a view better accounts for the variety of ways in which we engage with epistemic authorities: as sources of knowledge, as providers of understanding, as intellectual collaborators, and as pedagogues and guides. With this account in hand, I put it to work: first, in providing an account of what is required of an epistemic agent to remain epistemically autonomous, and epistemically responsible, when dealing with an epistemic authority; secondly, in establishing the role of epistemic humility in ensuring that relationships of epistemic authority function in an epistemically successful fashion. viii INTRODUCTION Social Epistemology and the Importance of Authority It is now widely acknowledged that our epistemic success as individuals owes a considerable debt to the epistemic efforts of others. To be epistemically successful is not simply to rely on one’s own faculties. Here, the social epistemologist begins to sink in her teeth: by accounting for how we can know via testimony, and how knowledge can, more generally, pass from one individual (or group) to another. There is, however, far more complexity to the kinds of epistemic relationships that hold between various epistemic agents than a basic model of testimonial exchange accounts for. This dissertation has been motivated by consideration of two such points of complexity in particular. First, there are a diverse range of epistemic statuses that any individual involved in a social-epistemic relationship may hold, and, perhaps more importantly, there may be more or less of a relative discrepancy between the statuses of any two individuals involved in such relationships. Secondly, there are a plethora of epistemic goods that can be, and are, passed from one epistemic agent to another—many of which will require a far more nuanced and detailed theory than many accounts of testimony provide. On the condition that we wish for our epistemological accounts to track the facts of our social-epistemic lives, we will need to consider such complexities. My broader goal in writing this dissertation is to contribute to the project of making sense of the rich fabric of our social-epistemic lives. 1 One particularly interesting component to this epistemic tale is the role played by experts, or, to put it in broader terms, those that are in an epistemically superior position relative to others. There is a question of how the layman can come to gain knowledge from an expert, given that the means of justifying beliefs within a given domain of expertise are often opaque to those without the relevant expertise. In the case that we can make sense of this epistemic relationship, there are still further problems to consider, such as the issue of how a layman can handle disagreement between experts from her own limited vantage, or whether or not such a layman has any epistemic right to disagree with the expert herself. There is a question of whether or not one’s own reasons, as a layman, can have any force in the face of the expert’s supposedly authoritative reason. Though these issues are often discussed through the lens of specific problems regarding expertise and testimony, it has been less common to investigate the role of such epistemic superiors in the broader sense, pertaining to the wider scope of epistemic considerations that may be relevant to the relationship between such agents and their inferiors. We could ask: is there such a thing as epistemic authority, i.e. does it make sense to suggest that someone can tell you how to epistemically behave? If so, does such authority consist in being able to tell us what to believe, or are there other epistemic goods that such authority can furnish us with? Is it rational to completely defer to the reasons held by epistemically authoritative agents? What can we say about what makes an agent a better or worse candidate for such authority? I understand these questions to be of fundamental importance, both theoretically, but also practically. To leave them unanswered would be to leave our epistemic theories thoroughly incomplete, at best describing a perhaps ideal world in which individuals have no need to rely on others for their epistemic success, but failing to track the realities of our quotidian, epistemic customs. More notably, without answering these questions, epistemology misses an opportunity 2 to be practically significant: a central thesis of this dissertation is that a sufficiently detailed account of epistemic authority—one that can account for the wide variety of forms it can take— can put us in a better position to evaluate real life cases of such social-epistemic relationships. Such evaluations can help us understand precisely what is problematic in certain cases, and, I think even more promisingly, can provide us with a means to establish guidelines for how such relationships should be conducted and regulated. My reader may initially wonder why the notion of epistemic authority ought to be seriously considered in the first place, especially given, as we shall see, the controversial nature of philosophical accounts of such authority. The first answer to this question is that there are a plethora of situations in which we epistemically interact with others that I think clearly fall under the explanatory scope of such a notion. Perhaps some of my readers are more cautious than the rest of us, but I think any empirical investigation into the epistemic practice of the average human subject, in normal, everyday circumstances, will reveal a multitude of ways in which we defer to those that my account classifies as epistemic authorities, and that we grant such authority a special dominion over our epistemic behaviour. As a philosophy PhD student, I grant my dissertation committee, comprised of exemplars of the philosophical tradition, a certain authority over me. As a teacher, I understand myself to have a similar kind of authority over the (domainspecific) epistemic behaviour of my students: though the nature of the pedagogical relationship necessitates that I assist my students in adjusting their own epistemic behaviour on their own terms, it does not preclude the possibility of my authoritatively telling them to behave, or not behave, in a certain epistemic fashion. As a patient, my doctor has a somewhat different kind of authority over my epistemic states—but she has such authority nevertheless.1 Importantly, this A caveat in all these cases: so long as I do not have an independent reason to doubt the authority, or epistemic credentials, of such agents. 1 3 kind of authority is not merely one bestowed upon such agents by social convention—there are deeply epistemic roots for such a status.2 If epistemology is to accurately account for the epistemic situation that we generally find ourselves in, then it ought to be able to make sense of these kinds of relationships. My claim is that a well-defined notion of epistemic authority can do just that. My first goal, then, will be to motivate the view that there is a sense of epistemic authority that goes beyond that which is typically discussed in the epistemological literature on experts. In itself, this is not an original point. Linda Zagzebski has already argued for, and provided an account of, such authority, over and above the more basic sense expertise and epistemic authority often discussed by epistemologists: i.e. one merely concerning those that are ‘reliable sources of information in some domain’.3 I am in agreement with Zagzebski that it would be of philosophical benefit to consider and explore the possibility of there being a stronger sense of epistemic authority, one more akin to the kinds of authority more often discussed in the political domain. In exploring this possibility, I follow in Zagzebski’s footsteps, focusing primarily on the notion of Preemption, a condition of authority originally articulated by Joseph Raz in accounting for political authority, which Zagzebski has adopted and applied to the epistemic realm. This condition states that an epistemic authority can provide me with a preemptive reason to believe some proposition p, where a preemptive reason is understood to be one which overrides all my prior reasons pertaining to p. Thus, if an epistemic authority tells me that p is the case, Preemption states that I ought to believe that p, and I ought to believe it solely on the basis of the authoritative statement, disregarding all other previous reasons I have for believing, or not believing, that p. Consider, for example, the possibility that I find myself in a society that for some reason gives an authoritative intellectual status to individuals with what we would think of as possessing a quite questionable epistemic status. In such a society, it would still make sense for me to recognise a non-socially recognised individual as having epistemic authority over me, given that they themselves do in fact have the right kind of epistemic standing. 3 Linda Zagzebski, Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief (New York: Oxford University Press, 2012), 5. 2 4 Preemption, Zagzebski suggests, is central to epistemic authority, in fact necessary for epistemic authority: it is the normative power that grounds such authority, and it is constitutive of such authority.4 As we shall see, the Preemption thesis for epistemic authority, as stated by Zagzebski, has its problems. However, I nevertheless agree with Zagzebski on two points: first, in thinking that there is such a real-life phenomenon as that which her theory attempts to account for; secondly, that a successful theory of such a phenomenon must include a strong normative component, such as her Preemption thesis. Much of the first part of this dissertation will be dedicated to making sense of how exactly we can articulate the nature of this component, without being ensnared by counter-intuitive consequences. I disagree with Zagzebski, however, on two crucial points: first, we disagree on what provides the kinds of agents in question with the relevant epistemically normative (and authoritative) power over others; secondly, we disagree on precisely what the scope of such power is, as I think the relevant individuals have a role to play in normatively adjusting our epistemic behaviour in a variety of ways beyond merely altering what we come to believe about a given question—this is to say that an epistemic authority can authoritatively adjust my epistemic behaviour in other ways that merely telling me what to believe. Contra previous accounts of expertise and epistemic authority, my view is not limited to epistemic agents who have relatively more true beliefs—or a higher chance of success at acquiring more true beliefs—relative to a certain domain of inquiry than we. Roughly, I argue that an epistemic authority is someone who, other conditions being satisfied, has the requisite skills, abilities, and know-how to partake in a certain epistemic practice that is constitutive of inquiry in the domain in question. These individuals have authority over us precisely because we, And, in being constitutive of epistemic authority, such authority is thus argued by Zagzebski to be continuous with other forms of authority: such as moral and political. 4 5 relative to them, have fewer of the requisite skills, abilities, and know-how, or perhaps do not have them at all. Significantly, this status explains why an epistemic authority is someone that can provide me with reason to adjust my epistemic behaviour in other ways than merely changing what I do, or do not, believe. This allows me to argue that epistemic authorities are recognised as such because they traffick in a variety of epistemic goods, many of which are often overlooked in the philosophical literature on expert knowledge and testimony. Thus, I argue that epistemic authority is not grounded merely in considerations of possessing a sufficiently large set of true beliefs, or being a reliable source of such true beliefs, but in possessing the skills, ability, and know-how to partake in a certain kind of epistemic practice constitutive of inquiry in the domain in question. These differences are substantial. My account has the advantage of being applicable to a broader range of cases, all of which I argue have the right to be considered under the scope of an account of epistemic authority. This includes cases in which an authority may not be able to provide us with true beliefs, perhaps even in principle, but can nevertheless improve our epistemic situations in some other fashion (here, I have in mind cases in which an agent may be authoritative relative to a domain of inquiry that some may not consider to deal in truths per se: possible examples being various aesthetic, religious, spiritual, and moral domains).5 More importantly, this includes cases in which a relative non-authority may be seeking something other than true belief in his engagement with an epistemic authority. As I will argue, particular instances of the epistemic-authority-to-non-authority relationship are far more complicated than described by accounts that focus on the mere transmission of true beliefs. These complexities include, but are not limited to: cases in which an epistemic authority furnish us with understanding 5 It should be noted, however, that my dissertation is not wedded to a particular view of truth, or to the assumption that there are no truths in certain domains of inquiry, such as the ones listed. Even if one accepts some pluralist account of truth, I argue that there is reason to think that we are better off accounting for epistemic authority in a manner that is not directly tied to the acquisition of true beliefs. 6 (as Christoph Jäger suggests), cases in which they guide us in our own epistemic endeavours, and cases in which we in fact collaborate with them in such endeavours. With the aim of establishing my own view, I begin in Chapter I by surveying a selection of prior definitions of epistemic authority. I take as my examples definitions by Zagzebski herself, Katherine Dormandy, Jan Constantin and Thomas Grundmann,6 and Christoph Jäger.7 As already mentioned, Zagzebski, modifying Raz’s account of political authority, provides an account of epistemic authority with the Preemption thesis at its centre. Dormandy and Jäger present a series of arguments against Zagzebski’s Preemption thesis, highlighting the counterintuitive consequences of the suggestion that an agent should disregard some of her reasons in the light of authoritative testimony. Dormandy presents an account of epistemic authority that is intended to avoid the issues that Zagzebski faces, while Jäger presents his own account of a certain kind of epistemic authority, which he refers to as Socratic Epistemic Authority. Constantin and Grundmann defend their own, more limited version of Preemption, which, though it avoids some of the pitfalls of Zagzebski’s account, is not sufficient to do justice to the stronger sense of authority that Zagzebski and myself wish to account for.8 Much in the same way, I argue that Dormandy and Jäger’s accounts don’t do justice to this stronger sense of authority either. Jäger’s account, does, however, correctly highlight the need to broaden our considerations regarding epistemic authority to consider epistemic goods other than the mere transmission of true beliefs when considering authorities in the epistemic realm. In Chapter II I argue against the possibility of limiting the extension of the epistemic authority to that of agents with a more reliable sense of the truth than ourselves. I argue that for Jan Constantin and Thomas Grundmann, ‘Epistemic Authority: Preemption Through Undercutting Defeat’, Synthese, forthcoming. 7 Christoph Jäger, ‘Epistemic Authority, Preemptive Reasons, and Understanding’, Episteme 13:2, 167-185. 8 Not that Constantin and Grundmann would deny this point: in fact I think they emphatically accept it, for reasons that we shall see. 6 7 someone to be epistemically authoritative over another agent she is required to be more than a mere possessor of a sufficient amount of true beliefs. To merely be a more reliable gauge of truth does not grant one the necessary dominion over the epistemic actions of another agent that is required to be considered an authority. Epistemic authorities are exemplars of a certain epistemic practice, and possess other virtues than merely knowing truths. Furthermore, limiting our considerations to the acquisition of true beliefs as the epistemic good that we can acquire from epistemic authorities does not make sense of the wide variety and kinds of social-epistemic relationships that we are considering: there are a plethora of epistemic goods that we can get from our epistemic superiors, and some of these may have little to do with the kinds of true beliefs relevant to an epistemic authority’s domain of inquiry per se.9 One example of Jäger’s inclusion of understanding as an epistemic good trafficked by such authorities, which he suggests is an important component of Socratic Epistemic Authority. But there are further epistemic goods that epistemic authorities may provide, or at least can put us in a better position to acquire, and I argue that an account of epistemic authority must be able to make sense of all of these if it is to be an adequate theory in the first place. At bottom, I suggest that to be an epistemic authority, as Zagzebski is right to suggest, is to be an epistemic exemplar of sorts, one that can show an epistemically less well-off agent how to do better epistemically, and can do so in a variety of ways. I suggest that the aspect of such epistemic exemplars that grants them authority over others in the epistemic domain does involve a kind of Preemption, but not of the sort that either Zagzebski, or Constantin and Grundmann, have argued for. Rather, I argue that epistemic authorities are able to pre-empt our higher-order beliefs and reasons—our beliefs about our reasons—and pre-emptively guide our epistemic My account thus can make sense of how one can plausibly treat someone as an epistemic authority even when that authority’s domain of inquiry, or epistemic practice, does not necessarily provide us with true beliefs. 9 8 behaviour, on the basis that such reasons, and behaviour, are constitutive of a certain kind of epistemic practice that defines the domain of which they are authorities. I argue that to be an epistemic authority is to have the requisite skills, abilities, and know-how to successfully partake in an epistemic practice that can provide us with epistemic goods. I argue that, in part, what it means to be an epistemic authority is to, via very the act of practicing a certain discipline, establish the very rules and norms that guide reasoning, and other epistemic processes, in that domain. When we engage with such epistemic authorities, we do so because we recognise that practice as epistemically valuable, and the authority as being better situated to practice it (either because we have none of the requisite skills, abilities, and know-how, or because we have them to a lesser degree). In interacting with such authorities, then, we can be understood much as somewhat addled outsiders being welcomed in to an alien practice or epistemic behaviour—on the assumption that we wish to acquire the fruits of that practice, we have to defer some of our choices in acting, at least in part, to that established practice. Broadening the scope of normative power constitutive of epistemic authority, as my account does, however, presents an obvious problem: it threatens to multiply some of the problematic elements of Zagzebski’s view, and to conflict with what some may intuitively consider to be important components of our individual epistemic agency. If we are to take the existence of such authority seriously, we ought to consider and account for its affect and consequence on the individual epistemic status of the relative non-authority. I will tackle this question in Chapter III, which considers the issue of epistemic autonomy and responsibility, on the side of the non-authority, when engaging with epistemic authority. The first question to be answered here, is whether the notion of epistemic authority does harm to what some may consider a valuable element of being a fully-fledged, epistemic agent: our epistemic autonomy. To answer this question, I first identify key components central to our intuitive notion of 9 autonomy broadly, and evaluate how and why we may initially think that its epistemic counterpart is jeopardised by the idea of epistemic authority. I conclude from this that those who place value on such autonomy have nothing to fear, and articulate a series of condition that we, as non-authorities, must meet in order to be considered autonomous in our dealings with epistemic authority. By doing so I hope to take important steps towards a more urgent task: delineating where, when, and how an epistemic agent ought to be considered epistemically responsible for the instantiation and consequence of a relationship with an epistemic authority, which, in turn, will illuminate when an epistemic authority is alternatively responsible for any consequence. This theoretical delineation of the scope of epistemic responsibility plays an important task in improving our ability to make real-life changes: it allows us to properly focus our attempts to improve the enactment of various social-epistemic customs, by ensuring that we target our efforts to the appropriate individuals within the relevant relationships. Very broadly, this means that we ought to be able to do a better job of educating all members of society how to best identify and engage with those in epistemically superior positions, relative to a certain domain of inquiry, and to understand their own responsibility in doing so; furthermore, it allows us to get a better grasp of what an epistemic authority’s duties are in such a relationship.10 To further understand the roles and responsibilities of all parties in such authoritative relationships, I turn in Chapter IV to the question of epistemic humility. My purpose here is to further elaborate on what is required of each party to make relationships of epistemic authority function correctly—or to their full potential. Though I focus on the role of such humility on the Of course, it is open to question whether an authority has ‘duties’ at all in such a relationship, particularly given that my account of epistemic authority in no way requires that an authority herself recognises the relationship—the establishment of such a relationship is solely on the shoulders of the non-authority. This will have to be discussed in further detail in Chapter IV, but for now I will say this: a key component to my project here is the recognition that our success as epistemic agents is deeply rooted in the fact that we are social creatures. I am motivated by the thought that any honest articulation of such success, and that any truthful self-reflection on ourselves as epistemic agents, will place a serious weight on the fact that we are members of a broader social-epistemic community. Given such membership, it is not a far cry to suggest that we each have duties in respect to it. 10 10 side of the authority herself, I argue that such humility is required both on the part of the authority and the non-authority. First, I argue that consideration of this particular kind of socialepistemic relationship supports a view of intellectual humility roughly akin to what Maura Priest has previously argued for: namely an interpersonal kind, one that places weight on the idea that to be intellectually humble is to hold a certain positive attitude towards the role that other epistemic agents play, even if they are in some sense epistemically inferior. Secondly, I illustrate why such an interpersonal attitude is required to make interactions with epistemic authorities, as a whole, successful. This will be shown to be particular relevant in the more complicated kinds of relationships that we can have with epistemic authorities that I have already alluded to, in which things are not as cut and dry as basic accounts of such authority, and testimony in general, make them out to be. By focusing on the role of such humility on the part of the authority, I aim to at least illustrate one key point: the best of epistemic authorities instantiate certain epistemic virtues. My hope is that my illustration of one such virtue will motivate further research into other such virtues, and lead to a fuller account of what could possibly constitute an ideal authority of this kind. This is an important task because the possession of such an account would allow us to do two things: first, it would allow us to better train, and provide guidelines for, those individuals who are likely to end up in a position of epistemic authority relative to others; second, it would also allow us to better educate the rest of us to identify if and when an individual, whom we may be considering as identifying as an epistemic authority, possesses the kinds of intellectual virtues we would like them to have. The issues I raise in Chapters III and IV are of timely consequence. If popular reports are to be believed, we live in a time in which much intellectual authority, and the general social status of cognitive expertise more broadly, has been undermined by the presence of various forms of acerbic, and especially corrosive, political criticism. Many of my readers may, for example, be 11 troubled by the current prevalence of climate change denial, and the anti-vaxxer movement.11 In place of those once thought of as our intellectual superiors, many have placed their trust in political and popular figures: authoritarian rather than authoritative figures, who’s claim to such status are often grounded on intuitive pull, the allure of the conspiracy theory, and the hook emotional appeal. Though I leave my reader to make their own judgements, I think it a boon that my account and related arguments show us precisely why such figures should not be identified or treated in such an authoritative manner: not only do they lack the epistemic credentials to be epistemic authorities, our engagement with such individuals qua authorities does damage to our epistemic autonomy, and places individuals who do not exhibit the right intellectual virtues in places of epistemic, political, and practical power. Even if one is not persuaded by my examples in themselves, they nevertheless illustrate my point: many of us no longer trust those that are supposedly our epistemic superiors (in some regard, or in some domain of inquiry), and this attitude of distrust can potentially have disastrous consequence. By better understanding the conditions in which we ought to identify someone as an epistemic authority, and how we, and the authorities themselves, ought to act, we can make steps to reverse this trend, and to better our overall social situation. 11 Aptly enough, the last revisions to this dissertation were made on days in which I received what struck me as emails with anachronistically absurd contents: warnings of various cases of a recent measles outbreak in my community, my places of study, and place of work. 12 I. EPISTEMIC AUTHORITY THUS FAR Problems with Preemptive Accounts Much has been written on the epistemic status of experts, though philosophical inquiry into the question generally tends to deal with particular problems related to experts, such as the issue of what a layman ought to do when faced with expert disagreement. What such analyses boil down to, however, is a very basic problem: the epistemic status of the non-expert relative to an epistemic task in which he does not have the appropriate domain-specific skillset, background, or prior knowledge, to be able to reliably judge. Most discussions of this problem are focused on the question of how, we, as non-experts, can have justification in coming to some conclusion, and thus some belief, in a domain of inquiry that we lack sufficient access to—domains for which we do not have the appropriate epistemic tools for accessing alone, at least not without much more education, training, practice, etc. The expert, it is assumed, provides us with such access to these domains. Without testimony from such experts, and without our being able to gain knowledge via such testimony, most of us could not claim to have knowledge about such things as quantum mechanics, the inner workings of the economic process, a wide range of important historical facts, and so forth. Epistemologists have defined expertise in a variety of way. Typically, following Alvin Goldman, accounts have primarily focused on the idea that an expert is someone with more true beliefs (or propositional knowledge), and fewer false beliefs, in a domain of inquiry d, than a non13 expert.12 Others have, still focused on the good of true belief, limited the account to the mere claim that experts are more likely to acquire true beliefs, relevant to questions pertinent to d, relative to the non-expert. Others have focused on the idea that experts are simply those who have engaged in a more sustained inquiry in d compared to the non-experts.13 Though things could be said in favour of these definitions, and I think elements of these will play a role in any fully-fledged account of expertise and epistemic authority, I understand such theories of expertise to be limited. As will be argued in Chapter II, I think the focus on true beliefs is an unnecessary limitation in these accounts, and, in particular, hinder such accounts from being able to account for the full range of phenomenon that I argue they should. As I understand it, when we engage with experts qua epistemic authorities, we engage with them as a source of a wide variety of epistemic goods, and an account of experts that focuses too narrowly on the epistemic good of truth will fail to properly account for these other cases. Furthermore, I argue that an epistemic authority’s reliable connection to the truth cannot be the grounds for the kind of authority appropriate to their epistemic standing. The above discussion of ‘experts’ and ‘epistemic authorities’ is confusing. For reasons that will become clearer as we progress, I assume that there are in fact two distinct concepts at play in these discussions: expertise, and epistemic authority. What confuses matters, however, is that (or at least I argue) our everyday usage of the word ‘experts’ often equivocates between the two. Unfortunately, epistemologists themselves have made the mistake of equivocating in the same fashion. As a preview of the position I will defend in Chapter II, let me say the following: Expertise will be understood in non-relational terms, as a quality, or qualities, 12 Alvin I. Goldman, ‘Experts: Which Ones Should You Trust?’, Social Epistemology: Essential Readings, ed. Alvin I Goldman and Dennis Whitcomb (New York: Oxford University Press, 2011), 109-133. 13 John Hardwig, ‘Epistemic Dependence’, The Journal of Philosophy, 82:7 (July 1985), 338. 14 that an individual epistemic agent may have, by virtue of their training, skills, acquired knowledge, etc. Epistemic authority will be understood in relational terms, as a status that one epistemic agent may have against another: by virtue of having some epistemic status that others lack, by various degrees. Though my own reasons for keeping the two distinct will become clearer as we progress, a few things can be said for such a distinction at this stage: 1. As Constantin and Grundmann have pointed out,14 one can be an expert in some epistemic domain while not being epistemically superior relative to another agent, who may, for instance, also be an expert within the same domain. This is a simple point that an account of expertise and epistemic authority ought to track. 2. The literature on expertise tends to obfuscate the distinction between a properly descriptive notion of ‘expertise’, intended to identify what an expert is per se, and a more appropriately evaluative notion of ‘expertise’, intended to capture the properly normative considerations that determine the relation that non-experts can have with their epistemic superiors. I argue that an account that keeps these two concepts distinct has benefits over one that doesn’t. 3. It is perfectly plausible for an epistemic agent to interact with, and to make use, of someone who is an expert in a given domain, and yet not treat them as authorities per se, in the stronger sense that I will be discussing. To treat an expert merely as a reliable gauge of truth, for instance, would be a case such as this: here, I simply refer to expert testimony much as I would any reading 14 Constantin and Grundmann, ‘Epistemic Authority’, 6. 15 off of a well-calibrated measuring device. To treat someone as an epistemic authority, I will argue, is far more than this. If this is right, then the two concepts ought to be kept distinct. That all being said, the distinction between expertise and epistemic authority will not always be clear in my discussion of the various accounts of epistemic authority presented by other epistemologists: part of my task will be to clearly delineate the difference in my own account of expertise and epistemic authority, and to argue for the benefit in doing so. In I.1 I explain Zagzebski’s account of epistemic authority, focusing on the central notion of Preemption, which she considers to be constitutive of epistemic authority itself. In I.2 I discuss a series of arguments that Dormandy and Jäger have presented against the Preemption thesis. In I.3 I present Dormandy’s alternative account of epistemic authority, which she refers to as the ‘Proper-Basing’ view. I.4 discusses Jäger’s alternative account of Socratic Epistemic Authority. Lastly, I.5 reviews Constantin and Grundmann’s own defence of Preemption, which they make sense of in terms of undercutting defeat. Those aware of the recent literature on epistemic authority will likely notice that I have not here mentioned Michel Croce’s account. This is due to the fact that, though our accounts differ in important ways, they also share important similarities. For this reason, I have chosen to discuss his account in contrast to mine. I refer my reader to Chapter II.4 for more detail. I.1 ZAGZEBSKI’S ACCOUNT Recent interest in the notion of epistemic authority is to be credited to Linda Zagzebski’s pioneering work in her Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. I understand Zagzebski to be motivated by the observation that it is somewhat peculiar that a strong, normative sense of authority, though prevalent in other domains of philosophical 16 discourse, is nowhere to be found in epistemology. As she sees it, there are likely two reasons that such authority is not considered in the literature, namely, that the existence of such authority arguably conflicts with two important epistemic values: epistemic autonomy, and epistemic egalitarianism. Epistemic egalitarianism is, roughly, the view that everyone has sufficiently similar epistemic powers; epistemic autonomy, on the other hand, which Zagzebski suggests is generally equated with epistemic self-reliance.15 On the face of it, neither seems compatible with the notion of epistemic authority. To suggest that another agent be epistemically authoritative over one’s beliefs and reasons would plausibly be to insinuate that they have sufficiently different epistemic powers than oneself. Furthermore, if we understand autonomy as the ‘right or ideal of submitting to nothing but one’s own rational will’,16 then the suggestion that one in any way submit to the rational will of another epistemic agent can quite clearly be understood as undermining one’s autonomy. Though I won’t discuss her arguments here,17 Zagzebski’s nevertheless concludes that the wish to protect these values provides no real challenge to epistemic authority: properly understanding these values, and the nature of epistemic authority, reveals that they are perfectly consistent with one another.18 These worries aside, Zagzebski argues that we have no good reason to ignore the task of providing an account of epistemic authority. Motivated to bring epistemic considerations closer to those of the ethical and political spheres, Zagzebski further suggests that we look at models of Zagzebski, Epistemic Authority, 7. Ibid., 19. 17 I discuss epistemic autonomy in Chapter III, and argue that there is no necessary conflict between epistemic authority and such autonomy. I do not tackle epistemic egalitarianism in much detail in this dissertation. However, certain points of my disagreement with Zagzebski’s view, as my reader will see, will suggest that my own view of epistemic authority is not particularly wedded to the idea of saving epistemic egalitarianism, at least in the simple form suggested here. As I see it, it is perfectly plausible, and quite likely, that there are individuals amongst us who do in fact have greater epistemic powers than we. 18 Those interested in the further details of Zagzebski’s arguments should look to Chapter 1 of her Epistemic Authority. 15 16 17 authority in those areas, and then see if they can be appropriately applied to their epistemic counterparts. Those less incentivised to unite these historically disparate branches of philosophy may wonder why we should follow Zagzebski in step, here. I think the strategy can be justified in a far more straightforward fashion: I would argue that if there is such a thing as authority in the epistemic realm, then it would be hardly surprising to find that it bears a resemblance to authority in other realms. This suggests that it would be prudent to begin with this assumption. In much the same way, I would think that an attempt to define epistemic authority that resulted in a theory that shared no common ground with other forms of authority would look very peculiar indeed, and would suggest that the phenomenon in question was not a kind of ‘authority’ per se. Zagzebski thus takes as her model Joseph Raz’s conditions for political authority.19 She argues that the conditions Raz lays out for political authority can be applied to the epistemic realm. Though I will briefly discuss all such conditions, my primary concern is her Preemption condition, the condition that Zagzebski argues gives authority its properly normative force, and the one condition that is the most philosophically controversial (at least in its application to the epistemic realm). The conditions are as follows: Content-independence: ‘An authoritative utterance gives the subject a reason to follow the directive which is such that there is no direct connection between the reason and the action for which it is a reason’.20 This is to say that the authority could have directed any wide range of actions, and this would still give the receiving subject reason to do that action. It is, in a sense, the fact that an authority has uttered that p that gives one a reason to behave a certain way, rather than the fact p 19 20 Joseph Raz, The Morality of Freedom, Oxford: Clarendon Press, 1988. Zagzebski, Epistemic Authority, 106. 18 has a certain content. Zagzebski thinks this condition applies in the case of epistemic authority: if I defer to the authority of a climate scientist, for example, I would still have reason to believe what she asserts, on authority, even if it had been the case that she believed and uttered something else. The subject’s reason for believing the authority’s utterance is completely independent of the content of said utterance—it is based on who the speaker is, not what she says.21 If we are to take it that such authoritative utterances to give us good reason, or a strong enough reason, to act a certain way, then content-independence may strike some as prima facie implausible in some cases. We can certainly imagine many cases in which the content of an authority’s utterance stands opposed to a variety of other good reasons we may have. Contentindependence, however, only tells us that an authority’s uttering that p gives us a reason to take a certain course of action: but it does not say how we should weigh this reason against all other reasons that we may have. The burden of explaining that component of the epistemic relationship between an authority and a subject falls upon the Preemption Thesis. In the political domain, Raz has formulated it as such: Raz’s Preemption Thesis: ‘[T]he fact that an authority requires performance of an action is a reason for its performance that replaces other relevant reasons and is not simply added to them’.22 Zagzebski thinks this thesis applies to the epistemic domain, and formulates the epistemic version as follows: Zagzebski’s Preemption Thesis for epistemic authority: ‘The fact that the authority has a belief p is a reason for me to believe p that replaces my Though Zagzebski does not directly state it in her discussion of this condition, I assume that her and I agree that content-independence is at least limited in one sense: the content does matter in that an epistemic authority must be speaking about some proposition that is the subject of inquiry in the domain of inquiry that she is an authority over. A climate scientist, for example, does not have authority over my beliefs in regard to God’s existence, my understanding of string theory, etc. 22 Ibid., 107. 21 19 other reasons relevant to believing p and is not simply added to them.’23 Zagzebski understands this to mean that the authority takes the subject’s place in determining that p. In coming to choose between whether to believe that p, or not-p, I disregard any of the various inferences I could make about the case, based on whatever prior evidence I have for either option. If I believe on authority, then my reason for believing that p is simply that the authority has stated that p. It would be useful here to take a close look at why this is so. Part of believing on authority is working on the assumption that the epistemic authority has already done the epistemic work of coming to the conclusion that p. For Zagzebski, this is work that I would myself have done if I had been more conscientious, i.e. if I had spent the time honing the relevant epistemic skills, acquiring the appropriate domain specific knowledge, carefully surveying and analysing the evidence, etc. It is this fact (that the authority has done the work for me) that justifies my epistemic behaviour.24 Note, however, that this reason for believing needn’t be a complicated and explicit inference on the part of the subject in every instance of accepting authoritative testimony, one based on the observation that the authority has done x, y, z in coming to her conclusion that p, with the decision to believe her utterance that p based on a calculated probability of her having gotten it right. Believing on authority is not to make one’s best judgement of an authority’s decision-making process on any particular occasion, because such a strategy would result in a range of contextually dependent credences—in one case an authority’s utterance will seem very likely to be true, in another less so—and this outcome, as will become clear soon, is completely antithetical to Zagzebski’s general defence of the Preemption thesis, and her account of epistemic Ibid. This is why Zagzebski thinks that epistemic authority is perfectly compatible with epistemic egalitarianism: for her, the authoritative testimonial relationship is grounded in the recognition that the authority is sufficiently similar to oneself—the authority has simply had the time and resources to be more conscientious relative to a certain domain of inquiry. Again, as we shall see, I disagree with this assumption. 23 24 20 authority more broadly. To put it another way: once an epistemic authority has been recognised and accepted as such, such inferences and calculations are, according to Zagzebski’s framework, contrary to the relevant epistemic relationship.25 The subject’s believing that p is in fact based on no such evidence, but merely on the fact that she recognises another as an authority, and that the authority has uttered that p.26 The epistemic work, regarding p, is all the authority’s. The subject’s work is merely in recognising the authority as such. Clearly, this condition requires far more explanation, which I will return to below. As we shall see, the Preemption condition is also a highly controversial one: it suggests that, given the utterance of an authority, I should disregard any reason I have for believing p, or not believing p. This may have serious, unintuitive consequences, as Katherine Dormandy and Christoph Jäger have pointed out—these will be discussed in section I.3. Before tackling the Preemption thesis in more detail, however, a few words should be said about the other conditions and theses Zagzebski thinks relevant to epistemic authority, again, as modified forms of Raz’s own theses pertaining to political authority. Raz’s Dependency Thesis: ‘[A]l] authoritative directives should be based on reasons that already independently apply to the subjects of the directives and are relevant to their action in the circumstances covered by the directive’.27 Zagzebski suggests the following epistemic articulation: Zagzebski’s Dependency Thesis for the authority of another’s belief: ‘If the belief p of a putative epistemic authority is authoritative for me, it should be formed in a way that I would conscientiously believe is deserving of emulation’.28 This should not be confused with what Raz and Zagzebski refer to as the No Difference Thesis, which states that an authority’s assertion should make no difference to what a subject My own view suggests this same conclusion, as will be seen in Chapter II. We will have much more to say about the idea of recognising an authority as such later in this chapter. 27 Zagzebski, Epistemic Authority, 108. 28 Ibid., 109. 25 26 21 ought to do, in the sense that the authority’s assertion should guide the subject to do what they ought to do in any event.29 Zagzebski thinks the No Difference Thesis is false in the epistemic domain (as Raz does in the political): if it were true, then the subject would have reason to believe that p without the authority’s assertion that p. As Zagzebski points out, this may not however be the case: the subject’s access to the relevant reasons for believing that p may be incredibly remote, thus leaving her with no reason to believe p at all.30 Raz’s Normal Justification Thesis: ‘the normal way to establish that a person has authority of another person is to show that the alleged subject is likely better to comply with reasons that apply to him if he accepts the directives of the alleged authority as authoritatively binding and tries to follow them, rather than by trying to follow the reasons that apply to him directly’.31 Zagzebski provides two epistemic analogues for this thesis. First: Zagzebski’s Justification Thesis 1 for the Authority of Belief (JAB1): ‘The authority of another person’s belief for me is justified by my conscientious judgment that I am more likely to form a true belief and avoid a false belief if I believe what the authority believes than if I try to figure out what to believe myself’. And, secondly: Zagzebski’s Justification Thesis 2 for the Authority of Belief (JAB2): ‘The authority of another person’s belief for me is justified by my conscientious judgment that I am more likely to form a belief that survives my conscientious self-reflection if I believe what the authority believes than if I try to figure out what to believe myself’.32 What is it then that Zagzebski means by ‘epistemic conscientiousness’? She defines this quality as that of ‘using our faculties to the best of our ability in order to get to the truth’,33 where Ibid. Arnon Keren thinks that Zagzebski has not interpreted the No difference Thesis as Raz’s intended it to be understood, and argues that, if interpreted correctly, it is actually true in the epistemic domain, but not the political one. If correct, this suggests an important lack of symmetry between political and epistemic authority. Arnon Keren, ‘Zagzebski on Authority and Preemption in the Domain of Belief’, forthcoming in European Journal of Philosophy of Religion. 31 Zagzebski, Epistemic Authority, 108. 32 Ibid., 111. 33 Ibid., 48. 29 30 22 a higher degree of conscientiousness goes hand-in-hand with a higher degree of critical selfawareness and consciousness of one’s reasoning. Conscientiousness is the state or disposition to make use of one’s own epistemic powers in a self-reflective, careful, and fully-conscious manner. The basic idea behind JAB1 and JAB2, then, is that an authority’s belief is justified to the subject’s epistemically conscientious conclusion that she is more likely to get to the truth if she believes what the authority tells her. From the subject’s perspective, she is justified in believing on authority by recognising that the authority can do a better job than her at getting to the truth. On the face of it, there may not seem to be anything immediately problematic about Zagzebski’s suggestion. Given that I, as a subject, recognise, or at least have reason to believe, that some authority A has a better chance of getting to the truth in some relevant domain d, I seem to have a very good reason to believe that p, given A’s utterance that p, if I want to have the best chance of getting to the truth. Given that I conscientiously believe that I have less chance of getting to the truth of the matter on my own, what reason could I have not to do just as Zagzebski suggests I am rationally required to? To both understand why Zagzebski thinks the Preemption thesis is correct for epistemic authority, and why others think it is not, we ought to consider two distinct epistemic goods in play. One, quite clearly, is that of Truth, and here I mean this in the somewhat limited sense that we aim, in our epistemic pursuits, to have more true beliefs, and fewer false beliefs. This is not to be confused with a broader goal of getting closer to, or grasping, ‘the truth’ in some form or another. To avoid any confusion in the following discussion, I will refer to this goal as Truth(B) throughout my discussion. Truth(B): The epistemic good of having more true beliefs, and fewer false beliefs. When considering our epistemic standing more broadly, however, other goods and goals come to light. It is generally accepted that our beliefs ought to be responsive to the totality of our 23 evidence, or to the totality of our reasons. In the case that a subject has a variety of reasons for believing that p, and other reasons for believing that not-p, it is thought that she should conclude with a belief that appropriately responds to all such reasons, and the weight that each hold. If there are more, or at least weightier, reasons in favour of p, then she ought to believe that p; if there are more, or at least weightier reasons, in favour of not-p, the she ought to believe that not-p; if the reasons are not decisive in this regard, it may be that she ought to abstain from believing either that p or that not-p. Let us call this the value of Total Reasons. Total Reasons: The epistemic good of having doxastic states that are responsive to, and reflective of, the totality of my reasons and evidence. To preview what is to come: the controversial nature of Zagzebski’s Preemption thesis centres around its being in apparent conflict with Total Reasons.34 Consider the following: a subject S considers a certain proposition p, within some domain d, and tries to decide whether she should believe that p, or that not-p. Before coming across someone whom she considers an authority on the matter, she compiles a series of reasons for and against p, and, on the basis of those reasons, comes to the conclusion that not-p. Now, in a discussion with an authority A in d, she is told, by A, that p is actually the case. If the Preemption thesis is true, then S ought to believe that p, but not because A’s testimony outweighs all other reasons for and against p, but solely because A claims that p. The only reason that factors in to supporting S’s belief is that A has uttered that p is the case. This flies clearly in the face of Total Reasons: S has neglected to hold beliefs that are responsive to, and reflective of, all of her reasons. Now, I think that such disregard for Total Reasons is highly problematic, as do Christoph Jäger35 and Katherine Dormandy.36 I’ll return to considering such criticism, however, in later 34 For a detailed account of this conflict, and more on Truth and Total Reasons in general, see Katherine Dormandy’s ‘Epistemic Authority: Preemption or Proper-Basing?’, Erkenntnis 83:4 (2017), 773-791. 35 Christoph Jäger, ‘Epistemic Authority’. 36 Dormandy, ‘Epistemic Authority: Preemption or Proper-Basing?’. 24 sections. Here I focus on Zagzebski’s approach to solving the conflict between Truth(B) and Total Reasons, and how she thinks that the Preemption thesis survives, and is in fact bolstered by, such conflict. Zagzebski’s initial defence of Preemption boils down, very simply, to the following: Truth(B) is more epistemically valuable than Total Reasons, and given that, in the case of epistemic authority, only one can prevail, we ought to choose Truth(B).37 To see why we cannot have both, Zagzebski invites us to consider the alternative, in which we do not follow Preemption, but we merely consider an authority’s utterance that p as a reason to believe that p that is added to all our other reasons for and against p. The problem, she claims, is that even if we weight the authoritative reason appropriately highly, there is still the risk that there will be cases in which such a reason is outweighed by the other reasons that S may have, resulting in S ending up believing that not-p when A believes that p. Given that, by definition, A is more likely to have a true belief, it follows that S has forsaken Truth(B) in favour of Total Reasons. One can easily think of problematic cases in which what Zagzebski envisages comes to pass: a medical surgeon suggests a course of action that a patient refuses due to a variety of information, from others suffering a similar condition, she has found online; a parent refuses to vaccinate her children in light of the overwhelming amount of testimony she has received from people she trusts, contrary to a medical scientist’s authoritative claims.38 39 Ultimately, if we refuse to abide by the Preemption thesis, then we run the risk of the I will consider various other defences that Zagzebski has presented in answering her critics in later sections. It may be argued that, in such cases, the patient and parent are not in fact recognising the relevant agent as an authority per se. I don’t think this is the case, however. There is at least nothing incoherent in the thought that one can recognise someone as an authority in a certain domain, and even think of such an authority as being more likely, in general, at getting at the truth, and yet having (what one takes to be) an overwhelming large amount of evidence against their testimony in any one given circumstance. It would only be incoherent for such a subject to consider an authority an authority in the case that she consistently turns against the authority’s judgements. 39 There is a further issue here pertaining to the difference between one’s having reasons that are properly within or without the domain d. One may have reasons that are not domain specific, for instance, such as evidence to the fact that authorities in d are under financial influence when it comes to the issue in question. On the other hand, one may be making use of reasons that are properly within the relevant d. This distinction is an important one: it will be discussed in more detail below (section I.5). 37 38 25 good of acquiring more true beliefs, and risk having more false beliefs, in favour of other, potentially less-truth-conducive, reasons. This is, of course, a simplified picture of the kinds of epistemic interactions we are considering. To better come to grips with the Preemptive thesis, and the nature of epistemic authority in general, I think we ought to take a look at the various criticisms it has faced, and the various modifications that other philosophers have suggested to Zagzebski’s account. This will be the task for the next three sections. Before I move on, however, I think it important to emphasise one key point. For Zagzebski, it is clear that being ‘a normative power that generates reasons for others to do or believe something preemptively’ is essential to the very notion of authority.40 It is this kind of reason that it gives me that makes it an authority, not that it can command me per se.41 I think there is something right about this intuition, though I have concerns about the way in which Zagzebski has fleshed out the notion of Preemption. Given that I think it central to the project that Zagzebski is pursuing that we capture this normative force at the centre of epistemic authority, any attempt to modify her view, or to present a contrary account of epistemic authority, will have to make sense of this normative power proper. Of course, one could also argue that there is no such thing as epistemic authority in this stronger sense, that all we have are normative reasons to believe experts, and to give special weight to the evidence that they provide—whether this is the correct conclusion to make, however, is yet to be seen. I.2 DORMANDY’S AND JÄGER’S CRITICISMS OF PREEMPTION Katherine Dormandy approaches the notion epistemic authority by directly tackling the 40 41 Zagzebski, Epistemic Authority, 102 (emphasis my own). Ibid. 26 puzzle that Zagzebski has presented: either an account of epistemic authority must accept Truth(B) at the cost of Total Reasons, or it must accept Total Reasons at the cost of Truth(B). I think that the difficulties of Preemption are best understood, at least broadly speaking, through the lens of this supposed incompatibility of epistemic goals, and I thus follow Dormandy’s example in framing my discussion in this light. Jäger doesn’t frame his discussion of Preemption in quite the same way, but his arguments against it are such that I think they can be fruitfully discussed in tandem with Dormandy’s criticisms, and in light of the tension presented between Truth(B) and Total Reasons. Dormandy presents the apparent puzzle as such: ‘Either belief on expert authority (a) must promote Truth at the expense of Total Reasons, (b) must promote Total Reasons at the expense of Truth, or (c) is not epistemically appropriate, all things considered’.42 She defines ‘Truth’ and ‘Total Reasons’ as follows: Truth: ‘It is epistemically good to form beliefs that are true.’ Total Reasons: ‘It is epistemically good for an agent’s beliefs to be based on all of the reasons which she has supporting them’. Though there are differences,43 my discussion in what follows assumes that Dormandy and I agree on what the goods of Truth(B) and Total Reasons are. For reasons that I have already mentioned, Dormandy’s option (c) is undesirable. This leaves us with having to choose between (a) and (b). That would be on the assumption that the puzzle is correct, and that all options have been exhausted. Dormandy’s own solution, which we will see in the next section, is to show that there is in fact another way, a way to save both Truth(B) Katherine Dormandy, ‘Epistemic Authority: Preemptive Reasons or Total Reasons?’, unpublished draft, 3. Though these definitions capture the general gist of the kinds of epistemic concerns we are dealing with, I think they are too narrowly construed here. Truth(B), I take it, is not solely about having more true beliefs, but also about having fewer false beliefs: it is about avoiding error as much as it is about being right, which means that agnosticism can be an epistemic virtue. Total Reasons should also be thought of as being more than merely basing one’s beliefs on all the reasons for which one has supporting them: our beliefs should be reflective of all the evidence we have, which includes those that may speak against such a belief. 42 43 27 and Total Reasons. In this section, I’ll focus on the arguments that motivate the need to search for an alternative to Zagzebski’s account. Dormandy herself presents three such reasons, or arguments, suggesting that it would be a mistake to simply accept the manner in which the Preemption thesis, or at least Zagzebski’s defence of it, dismisses the good of Total Reasons in favour of the supposedly more important one of Truth(B). I.2.A. EPISTEMICALLY SUBSTANDARD ARGUMENT Why not simply reject Total Reasons? Because, Dormandy argues, there is something ‘epistemically substandard’ about an agent who recognises that a certain reason that she has supports a certain conclusion or belief, and yet does not hold that belief on the basis of that reason. Appealing to Zagzebski’s notion of ‘fit’—the idea that our beliefs, desires, emotions, and other mental states should in some sense fit with reality, or with an appropriate object—she suggests that one’s beliefs being based on all of one’s reasons is simply another way for one’s epistemic state to ‘fit’ with reality: if ‘reality contains epistemic connections between the reasons which one has and the propositions which they support [ . . .] then surely beliefs which do not track these connections fit less well with reality than beliefs which do’.44 By analogy, Dormandy likens this epistemic quality to a certain lack of moral character: ‘there is something morally substandard about an agent who witnesses an outrageous crime, recognizes theoretically that it is outrageous, yet does not feel outrage’.45 I must admit I am not completely sure how one should understand Dormandy’s claim here. It seems intuitively easy to make sense of the idea that one’s doxastic states should match reality, or ‘fit’ with it in some sense, but it isn’t as obvious that all of one’s reasons ought to ‘fit’ 44 45 Dormandy, ‘Preemptive Reasons or Total Reasons?’, 6-7. Ibid., 6. 28 with one’s doxastic states in the same sense. Though I think there is clearly something intuitively unappealing about an agent who fails to exhibit such cohesion, I don’t think such oddness can be merely reduced to claims about a relationship between epistemic states and reality as I take it that Dormandy is suggesting. I think a lot more would have to be said (particularly on the metaphysics of beliefs and reasons) to philosophically articulate and defend the intuitive discomfort one may have in the face of an agent that suffers from such a psychological discontinuity between her reasons and her belief. However, given that I think most epistemologists would generally accept the importance of Total Reasons, and would find much to agree with in Dormandy’s intuition about an agent who fails to recognise that her reasons support a certain belief, I will, for the time being, put the issue aside. My assumption is that Total Reasons is epistemically valuable, in any epistemic context, and thus think that we should at least try to preserve it in the case of epistemic authority, if at all possible. For now, then, I will continue on this assumption, which suggests that there is a prima facie, intuitive motivation to avoid accepting Preemption. I.2.B. META-REASON ARGUMENT Dormandy refers to her second criticism of Preemption as the ‘Meta-Reason Argument’.46 To illustrate, she considers what she refers to as ‘same belief’ cases: these are cases in which a subject, S, prior to finding out an authority A believes that p is the case, already has her own reasons, r, for believing that p. To put it another way: these are cases in which S begins in possession of her own reasons r for believing that p, is then told that p is the case by A, and thus continues to believe that p. In such a case Preemption does not suggest that S should change her 46 Dormandy, ‘Preemptive Reasons or Total Reasons?’, 7. 29 belief that p (clearly, since A states that p is the case), but it does suggest that S should disregard her prior reasons for believing that p in favour of the preemptive reason that A says that p is the case.47 Dormandy’s claim is that this goes against reason: if anything, she states, ‘discovering that the authority believes that p makes it [. . .] more likely that the agent is correct in taking [her prior reasons] to be a good reason to believe that p’.48 Dormandy argues that if S has a reason, r, to believe to believe that p, then she will also rationally believe the following: rm: r is a good reason to believe that p.49 Dormandy then reasons in the following way: given the assumption that A is quite likely aware of a wide variety of reasons one may have for believing that p, she is likely aware of r as being a possible reason for supporting p. A will thus be committed to either believing that r is a good reason for believing that p, or it is not. So, either A believes rm or A does not believe rm. Further, Dormandy assumes that if S were to find out what A’s position was in regard to rm, then this would rationally commit S to raising or lowering her own credence in rm. An authority, after all, is assumed to know what makes for a good reason, given a certain domain of inquiry. Dormandy thinks that S, when faced with A’s testimony that p, is faced with two possible options, either: i. A states that p, and A does not believe that rm. ii. A states that p, and A does believe that rm. If (i) is more likely, then S should lower her own credence in rm, if (ii) is more likely, then S should raise her credence in rm. Dormandy’s claim is quite simply that (ii) is more likely. Or, at least, from the perspective of S, and given the information available to her in such a situation, she A potential motivation for this is that to continue to base one’s belief in p on r would be to put oneself in an epistemically weak position, as r may be far more sensitive to revision than the reasons that A actually has for p. 48 Dormandy, ‘Preemptive Reasons or Total Reasons?’, 7. 49 Ibid. 47 30 has reason to believe that (ii) is true, given that she herself takes rm to be true. Given that S discovers that A thinks p is true, Dormandy thinks this should raise S’s credence in rm. Again, she thinks that there is something ‘epistemically amiss’ about an agent believing that rm is true, and yet does not take A to also believe that rm is true. I take the intuition to be that one’s belief that r is a good reason for believing that p, is not compatible with the belief that someone who is better epistemically situated with regard to p does not think that it is: to believe that an epistemic authority thinks your reason is a bad one would be to no longer believe that one’s reason is a good one. This is all to show that, whereas the Preemptivist may want to suggest that a subject, faced with an authority’s utterance, ought to show a sort of epistemic caution about the strength of one’s own reasons, and that this better matches the epistemic connection between reasons and beliefs, Dormandy thinks an authority’s utterance should in fact support one’s belief that one has good reason for one’s belief, and thus bolster the connection between one’s total reasons and one’s beliefs, rather than hinder such connection.50 To follow Preemption, then, is to do damage to the coherence of our epistemic states, or, as Dormandy puts it: the ‘agent who pre-empts will therefore fail to match his noetic system with epistemic reality’.51 Dormandy considers a possible rejoinder: it is not that S ought to think that A thinks that r is a bad reason, but should rather assign equal possibilities to A’s either believing it is a good reason or a bad reason. Dormandy points out, however, that this would not make a difference: if this was the case, then S would be no less confident in rm than she was previously. In order for S to lower her credence in rm she would have to find out that A in fact thinks rm is false. I think it Dormandy also considers a parallel argument in different belief cases, in which S starts out believing that p, but changes her belief to not-p after hearing that A believes that not-p. In this case, Dormandy thinks that A’s utterance ought to raise S’s credence in any of her previously outweighed beliefs that she had good reason for believing that not-p, and diminish any credence in the reasons she had for believing that p. 51 Dormandy, ‘Preemptive Reasons or Total Reasons?’, 9. 50 31 should be pointed out, however, that the Preemptivist has an alternate means of response here: it is not that S should assign equal probabilities, or think that A believes that rm or not. Rather, given the situation S is in, faced with A’s epistemically superior position, he ought to simply refrain from assigning a credence to rm at all. Of course, one might still think that there seems to be something peculiar about the epistemic agent in this situation: they have changed their doxastic state, from believing that r is a good reason to believe that p, to being completely undecided about r, even though no new information about r has been revealed. However, I think it would be too quick to think that such a change could only rationally occur because of such directly targeted information, or evidence. As I will later argue, part of recognising someone as an epistemic authority is identifying them as being a more capable agent, relative to a certain epistemic practice, and thus better able to recognise good and bad reasons for belief in a specific domain of inquiry. The act of deferring to an epistemic authority, then, could quite feasibly involve a suspension of one’s previous convictions about all manner of reasons and beliefs—in fact, I will argue that this is precisely the case. But this is to get ahead of ourselves. Since I think other objections to Preemption more clearly articulate its flaws in a less questionable fashion, I leave this issue aside for now. I.2.C. EPISTEMIC-SUPPORT ARGUMENT Dormandy refers to her third argument as ‘the epistemic-support argument’.52 Here she criticises the view that S ought to solely base her belief p on an authority’s testimony, a, and not a in conjunction with her own prior reason for believing that p, r. The Preemptivist would claim that to base one’s belief on the conjunction of a and r would be to double-count r, given that, in 52 Ibid. 32 taking A to be an authority on the matter, we assume that she is aware of r as being a (possible) reason for believing that p. Dormandy’s argument is that believing that p on the conjunction of a and r would provide more epistemic support than merely believing it on the grounds of a, and, thus, that if we were to follow Preemption, we would be choosing to have weaker epistemic support for our beliefs. I do not wish to tackle the full complexity of Dormandy’s argument here, but will merely attempt to sketch it in its outline. Let us take r to be a reason that a subject S has for believing that p prior to hearing an authority A’s testimony, where a is the reason for believing that p based on A’s saying that p is the case. 1. Assume that both r and a, taken individually, support that p. Which is to say that S’s probability for r given p is higher than her old/prior probability for r (i.e. her probability for r being the case before she received it as a reason), and to say the same for a: Pr(r|p) > Prold(r), and Pr(a|p) > Prold(a). 2. The conjunction of r and a supports p: Pr(a & r|p) > Pr(a & r). 3. Before S receives either r or a, her probability for either is approximately the same as her probability for the other. 4. S’s old probability for the conjunction of a and r is less than her old probability for a: Prold(a & r) < Prold(a). 5. The conjunction of a and r makes p just as probable as does a by itself: Pr(a & r|p) = Pr(a|p).53 This suffices to show that the conjunction of a and r provides more support for p then a does merely by itself: given that the prior probability for (a & r) is less than the prior probability for a (premise 4), but that the probability for both is equal given p (premise 5), there is a larger 53 Dormandy, ‘Preemptive Reasons or Total Reasons?’, 9-11. 33 difference between Pr(a & r|p) > Prold(a & r), than the difference between Pr(a|p) > Prold(a), and this larger difference means that (a & r) confers a greater degree of support to the belief that p than a does by itself. This means that to follow Preemption would be to choose the path of less epistemic support for one’s beliefs, it would be to choose to relinquish the full force of one’s reasons and evidential basis for a less robust epistemic stance. I.2.D. THE PROBLEM OF COMPETING AUTHORITIES Christoph Jäger presents a further problem for Preemption in the context of the philosophically familiar issue of peer disagreement, specifically, disagreement between epistemic authorities. The problem here is a simple one. Consider the following case: a subject S is unsure about a certain issue, and in weighing the various reasons for or against p, he comes to believe that p, but with only a small degree of confidence. To find better epistemic footing, he approaches someone that he determines to be an epistemic authority on the matter (relative to some domain d for which p is in its purview). This authority agrees with him that p. However, S then finds out that another person that he identifies as an authority in d holds the opposite belief that not-p.54 What is S to do? Preemption, it would seem, has very little to say here: it is formulated to handle one’s relationship with a single epistemic authority, not multiple. Zagzebski has suggested that the problem can be partially solved by assuming that such epistemic authorities would be aware of such disagreement, and that, therefore, one’s chosen epistemic authority’s conclusion should be 54 Jäger, ‘Epistemic Authority’, 171-172. 34 accepted, meaning that S should not bother to seek out a second opinion.55 The problem, of course, is that this doesn’t seem to be empirically accurate.56 As Jäger points out, however, even if we accept the assumption, it isn’t clear how this can solve the problem: often, even if aware of such disagreement with another authority, an epistemic authority will simply continue to believe the same thing, and the disagreement will continue indefinitely.57 It is not as if being an epistemic authority disqualifies you from the plethora of complexities that plague peer disagreement—far from it, it would seem that the problem of peer disagreement becomes all the more troubling! Thus, we are still left with trying to understand what a subject ought to do, following Preemption: 1. Ought she to preemptively believe in favour of the first epistemic authority’s position? 2. Ought she to instead preemptively believe in favour of the second epistemic authority’s position? 3. Or ought she to simply withhold judgment, and refuse to preemptively believe either authority?58 As Jäger argues, all three options seem untenable for the Preemptivist, in that they all fail to stick to the spirit of Preemption in some form (by either refusing to pre-emptively believe on the basis of one, or both, of the authorities’ assertions), and that they seem to fly in the face of the kinds of rational considerations that generally motivate a Total Reasons approach. In her reply to Jäger, Zagzebski has argued that there is no problem here, and that to think otherwise is to misunderstand the very point of appealing to an epistemic authority in the This seems problematic in itself: even if I don’t seek out a second opinion, it doesn’t follow that I will never become aware of such disagreement, and the awareness in itself seems to be enough to suggest the intuitive peculiarity that Jäger is alluding to. 56 Furthermore, if the suggestion is that this is a necessary condition for authority, then we are left with a definition of such authority with what is likely to be a very narrow extension indeed. After all, how many authorities are actually aware of all relevant disagreements in their field? 57 Jäger, ‘Epistemic Authority’, 172. 58 Ibid. 55 35 first place. Zagzebski argues that the whole purpose of seeking an epistemic authority’s opinion is to gain the results of a more epistemically conscientious agent’s investigations: since I cannot do the research (I don't have the time to acquire the relevant skills and background knowledge, etc.) I appoint someone else to do the work for me, all while recognising that they are doing what I myself would do were I more epistemically conscientious.59 Zagzebski’s response then is that it does not matter, and should not matter, that there are other epistemic authorities that disagree with one’s chosen authority: one is relinquishing one’s say on the matter to the epistemic authority one chooses. Zagzebski’s response is deeply unsatisfactory on at least two fronts. First, it flies in the face of what I assume are widely accepted epistemic intuitions regarding epistemically rational behaviour. The intuition that seems to drive the identification of expert disagreement as a significant problem for epistemology is precisely the thought that, when faced with disagreeing experts, I have reason to think that at least one of them, if not both, have made an error, and, possibly, that their very status as reliable testifiers, and experts per se, is in jeopardy (or, at least, their status as equally competent experts). This is to say that I have a reason that is external to the relevant domain of inquiry for thinking that one of the experts is unreliable (I assume that the nature of disagreement and its epistemic consequences are not domain specific, and thus considerations of such facts are external to the domain in the relevant sense). Secondly, I am suspicious of Zagzebski’s response in terms of it seeming to trade on certain facts of practical rationality, rather than epistemic. The thought would be that, though it is perhaps practically rational for me to relinquish some of my epistemic considerations, given that I do not have the time or resources available to me to accomplish certain epistemic tasks myself, this does not suggest that I have epistemic reasons to so completely relinquish my epistemic duty to consider the 59 Linda Zagzebski, ‘Replies to Christoph Jäger and Elizabeth Fricker’, Episteme 13:6 (2016), 188. 36 higher-order evidence of expert disagreement. Though it may sometimes be prudent to simply leave things in the hand of one’s chosen epistemic authority (Zagzebski uses the example of a financial advisor, for instance),60 it is difficult to make sense of how this necessarily provides one with any solid reason to do so across the board, to the same degree in each case.61 Now, clearly, there will be times when practical reasons motivate me, quite reasonably, to less rigidly stick to certain epistemic rules, consideration of which I will return to in Chapter II, but especially in Chapter III. However, even if this is the case, far too much is left unexplained by Zagzebski’s quick response. One particularly poignant question sticks out: even if I defer to a chosen epistemic authority for partly practical reasons, on what basis do I choose which authority of the two disagreeing ones to follow? Clearly, my knowledge of such disagreement greatly undermines my ability to make such a choice in a conscientious and confident fashion. I.2.E. THE SWITCHING PROBLEM To further drive home the issue, Jäger proposes a second problem for Preemption. Imagine the following case in which a subject S follows Preemption: S has her own prior reason, or reasons, r, for believing that p. She then meets an epistemic authority who tells you that p is in fact false, and that S should instead believe that not-p. Given Preemption, S disregards r, and starts to believe that not-p, on the basis of the reason that the authority has told her that not-p is the case (call this reason a). So far, so good. But now imagine that S later finds out that the Ibid., 189. There is another response that Zagzebski alludes to: I trust my epistemic authority, and this trust is a reason to believe what she says. I assume the suggestion is that I will accept the testimony of the authority that I trust, even if I find out that another expert, within the same domain, disagrees. I do not wish to tackle the philosophical literature on epistemic trust, but I think it suffice to point out here that, if having this feeling of trust is part of what it is to recognise someone as an epistemic authority, then we simply have to stipulate that the kind of disagreement we are considering surfaces precisely when two such authorities, both of which you trust, disagree. There is, after all, no reason to think that one should necessarily be limited to one epistemic authority per domain. So, whereas disagreement between experts may not be a problem, disagreement between authorities will be. 60 61 37 authority herself has changed her mind, perhaps based on some new evidence, and now believes that p. Given Preemption, S should once again change her belief, and believe that p. Perhaps this still seems acceptable. However, Preemption requires a lot more than S’s changing her mind for a second time, she must now only believe that p on the basis of a, even though she previously believed that p on the basis of r. The reasons that she previously had in favour of p are not rationally allowed to provide additional support for p, even though the authority agrees. We can, of course, continue this pattern, with the authority switching between p and not-p over various spans of time.62 This case simply highlights the similar kinds of problems we have been discussing above, driving home the counter-intuitive consequences of rejecting Total Reasons even if it is for the sake of adhering to Truth(B). We can make the case even worse, however, by stipulating that the authority, in switching to the belief that p, actually switches back upon re-consideration of the very reasons r that S originally based his prior belief that p! In this case, Jäger argues, to adhere to Preemption would be to still believe that p but without ‘reactivating’ r, even though S knows that the authority believes that p on the basis of, or at least partly on the basis of, p.63 For Zagzebski, as I understand it, to use one’s own reasons is just beside the point: the authority is meant to be doing the epistemic heavy lifting. I take this to highlight a far deeper problem with Preemption that has not yet been made explicitly apparent: Preemption rationally blocks an epistemic subject from the possibility of using her reasons even when she has very good reasons (objectively and subjectively) to do so. Not only does it strike against our more common intuitions about Total Reasons, it seems to paint a picture of epistemic rationality that severs the relationship between our reasons and rationally appropriate doxastic states that we simply take for granted as being fundamentally important to our epistemic 62 Though I think it would be reasonable to think that eventually the authority’s switching her mind would be a reason to think that she is not as reliable as one previously thought, and thus diminish her status as authoritative. 63 Jäger, ‘Epistemic Authority’, 173. 38 well-being.64 In Dormandy’s terms, Preemption does great violence to the fit between our noetic systems and epistemic reality.65 In response to this particular problem, in the form that Jäger has formulated it, Zagzebski has argued—and I have to admit I am somewhat perplexed by her response—that sometimes one is simply better off following the trend of authoritative opinion (such as scientific consensus on medical issues), whereas in other cases one is better off suspending judgment (as in religious and philosophical cases), opting to wait for a time when you can conscientiously judge that the authority’s opinion is likely to provide you with a belief that will survive your future conscientious self-reflection (referring to JAB2, her second justification thesis for believing on authority).66 Again, I find Zagzebski’s response here puzzling. It seems to suggest that I ought to believe on authority only in cases where I—given a an epistemic evaluation of this particular issue, and the particular response on the authority’s part—decide that I am more likely to have a true belief that will survive future conscientious reflection. But this seems problematic for Zagzebski’s view on two fronts: 1. It goes against what she has said regarding the problem of disagreeing authorities, in which, given one’s recognition of someone as being an epistemic authority, and trusting them as such, I can simply accept their authoritative opinions even if the presence of disagreement may give me reason to think that such opinion will not survive my future conscientious selfreflection. Why does an authority’s willingness to change opinion provide me with a better reason to suspend judgement in some cases than does the I think there is a fundamental reason why Zagzebski’s account leads us to such a peculiar end: the first error was in shackling her account of epistemic authority to the idea that epistemic authorities provide us only with true beliefs, and nothing more. As I will argue in Chapter II, and as we will see Jäger similarly suggests, epistemic authorities can do a lot more than that. 65 Dormandy, ‘Epistemic Authority’, 9. 66 Zagzebski, ‘Response to’, 189. 64 39 presence of severe disagreement between experts? 2. By applying her justification principle (JAB2) in this fashion, on a case-by-case basis, Zagzebski severely undermines the picture of epistemic authority that I assumed her to be arguing for. It was my impression that the justification principle was meant to justify why it would be epistemically proper to follow Preemption, broadly speaking. Rather than the justification principle doing all the work in a case-by-case fashion, it was the authority’s ability to provide me with a preemptive reason to believe p once I had recognised her as such an authority. Now the suggestion seems to be that (JAB2) does all the work.67 I.2.F. THE PROBLEM OF UNHINGING PROPER BASES The last problem that Jäger presents against Preemption drives home many of the same points that Dormandy focuses on in her Meta-Reason Argument (I. 2. b.) and Epistemic-Support Argument (I. 2. c.). I only mention it here briefly to summarise the overall problem with Preemption. As Jäger puts it, the problem is basically this: ‘Preemption may require you to cancel adequate epistemic basing relations between [good] reasons and beliefs’.68 Jäger defines a belief as being properly based on a given grounds in the following way: Proper Basing: S’s (graded) belief B is properly based on a given ground (or set of grounds) G iff (i) G is the ground for which S holds B and (ii) S has a true and rational belief to the effect that G sufficiently supports B’.69 It is of course an open possibility that I have misinterpreted Zagzebski on this matter. But this is beside the point. I think it is clear that if we are to capture the strong sense of epistemic authority that I believe Zagzebski has tried to make sense of, then it cannot be the case that the normative power of such authorities is, in each case, derived from an epistemic principle that the non-authority simply applies to each case of authoritative testimony. 68 Jäger, ‘Epistemic Authority’, 174. 69 Ibid., 8. 67 40 First, Jäger argues that it seems psychologically implausible, if not impossible, to suggest that in a case in which I believe that p based on my own good reasons, I could then somehow come to base my belief that p solely on authority, with my prior reasons somehow remaining a part of my cognitive economy, and yet not acting as a ground for my belief that p in any way whatsoever. If he were asked to do just this, Jäger says, he ‘would not know how to proceed’.70 Most importantly, echoing Dormandy, Jäger argues that to unhinge one’s belief from its previous bases in such a way would be a kind of epistemic regress rather than progress. To unhinge one’s beliefs from one’s reasons would be to leave the belief less epistemically stable. I.3 DORMANDY’S PROPER-BASING VIEW Of course, all of the above arguments are to suggest that there is a cost to following Preemption: it side-lines Total Reasons in favour of Truth(B). As we already know, however, Zagzebski has suggested that, since we cannot have both, we ought to choose Truth(B). Dormandy and Jäger’s criticisms have simply shown what the costs of such a move are. The important question is: do we have any other choice? Dormandy thinks we do. She suggests the following view: The Proper-Basing View of Belief on Authority: ‘Believing that p on authority amounts to basing this belief on the authoritative reason and any other reason you may have, where (i) pro-reasons exert a force for p and (ii) contra-reasons do not exert any force against p’.71 This view, she claims, does justice to both Truth(B) and Total Reasons. S supposedly continues to base her beliefs on all of her available reasons, but also avoids the problematic possibility of having her own reasons outweigh the authority’s, which would leave her with a less reliable 70 71 Ibid., 10. Dormandy, ‘Preemption or Proper Basing?’, 11. 41 connection to the truth. I’ll begin by discussing the disjunctive manner in which Dormandy’s account deals with so-called ‘pro-reasons’ and ‘contra-reasons’. ‘Pro-reasons’ are reasons that an agent has in favour of p, prior to receiving an authority’s testimony that p, and are thus reasons in favour of what the authority asserts, independent of the authority’s testimony. ‘Contra-reasons’ are reasons that an agent has against p, prior to receiving an authority’s testimony that p, and are thus reasons opposed to what the authority asserts, independent of the authority’s testimony. The different treatment of pro- and contra-reasons can be seen to solve two problems: i. By allowing for pro-reasons to still exert a force in favour of believing that p, when an epistemic authority asserts that p, the Proper-Basing View solves the kinds of problems that Dormandy has argued that the Preemptive view falls foul to (such as the Meta-Reason Argument, and the Epistemic-Support Argument, in sections 1.2.b. and 1.2.c.). Dormandy’s account of belief based on authority allows for one to use the full force of one’s reasons in believing that p, when an authority asserts that p, rather than forfeiting one’s full doxastic justification, or giving up on Total Reasons. ii. By not allowing contra-reasons to continue to exert a force against one’s believing that p, when an epistemic authority asserts that p, the Proper-Basing View avoids the problem that Zagzebski thinks any attempt to avoid Preemption will: the view does not forfeit one’s more reliable access to the truth, as one will not face the possibility of having one’s contra-reasons swamp out the authoritative reason given in authoritative testimony. This means that one can follow the Proper-Basing View without giving up on Truth(B). As should be clear, these two considerations together entail that the Proper-Basing View suggests 42 an account of believing on the basis of epistemic authority that saves both Truth(B) and Total Reasons. I have my concerns about this disjunctive treatment, which I will return to shortly. For now, however, it is important to make sense of precisely what Dormandy is suggesting in arguing that one can base a belief in p on the totality of one’s reasons, while not having certain of those reasons actually exert force upon that belief. Dormandy suggests that ‘the force of a reason is relative to background information’.72 To use her example (following in Wilfrid Sellars’ steps): ‘a perceptual experience of a green necktie is a forceful reason for believing that the tie is green’ against a certain background of information, but, given another background of information—such as one in which you know that the necktie is in a room that is lit by green lighting, for example—‘this perceptual experience loses all force whatsoever’.73 In much the same way, Dormandy is suggesting that an authority’s assertion that p has the same ‘force-depleting effect’ on contra-reasons.74 This is because such an authority not only has more reasons for believing whether p or not-p is the case, but also is better at evaluating such reasons. We suppose that the authority has considered the kinds of reasons we have access to, and, given that they have still come to believe that p in spite of them, we can conclude that they aren’t good reasons, either in that they don’t support the conclusion that notp, or they are simply false.75 Dormandy suggests that such authoritative reasons are thus rebutting defeaters for one’s belief that not-p.76 77 Unlike a Preemptive view (even a limited Preemptive view that only preempts contra- Ibid., 12. Ibid. 74 Ibid. 75 This isn’t to mean that the same will hold if we have reason to think that he authority has not considered the kinds of reasons we have access to (as will be discussed in more detail when we get to Constantin and Grundmann’s account in section I.3). 76 Dormandy, ‘Preemption or Proper Basing?’, 13. 77 I have to admit I am somewhat perplexed by why Dormandy thinks of these as rebutting defeaters rather than undercutting defeaters as Constantin and Grundmann discuss them: see section I.5) 72 73 43 reasons), Dormandy thinks that her view does not do damage to our psychology by simply removing these contra-reasons from play. She thinks that, instead, the Proper-Basing View achieves the goal of having such contra-reasons unable to swamp out the truth-conduciveness of authoritative reasons simply ‘by letting the natural interplay of reasons do its epistemic work’.78 Though a Preemptive view and her own view may both explain why contra-reasons have no say in whether or not one believes that p, when an authority has told one that p is the case, her view, she alleges, does it without recourse to a psychologically and epistemically questionable procedure. A first problem with Dormandy’s view, is that I have a hard time understanding how the different treatment it suggests for pro- and contra-reasons makes any sense, given how Dormandy herself seems to justify the way contra-reasons are discounted. Given that she assumes that it is central to one’s recognition of another as an epistemic authority that one thinks of the authority as having considered all the reasons that one has for believing (or not believing) that p, it would follow that to continue to count pro-reasons in one’s epistemic calculus would be to double-count reasons in favour of p, just as the Preemptivist would no doubt claim. In order for the different treatment of pro- and contra-reasons to hold, Dormandy would have to provide a more plausible account of how some of one’s reasons are to be treated differently than others. After all, though an epistemic authority’s assertion that p may speak against reasons we have for believe that not-p, this does not necessarily entail that these reasons are, in themselves, bad ones. Though it may be epistemically responsible to lower one’s credence in such reasons upon hearing an epistemic authority’s assertion, this in no way entails that they should not exert any force against p. Clearly, they could quite as well exert such force, and the epistemic authority may be quite aware of this fact herself, just as she is of any pro-reason. 78 Dormandy, ‘Epistemic Authority: Preemption or Proper Basing?’, 13. 44 As it stands, Dormandy’s move strikes me as rather ad hoc on this front. As we shall see, I think Constantin and Grundmann do a better job of articulating a similar view in which pro- and contra-reasons are treated in a different fashion. More importantly, however, I think Dormandy’s view fails to account for the kind of phenomenon that I have in mind. First, her view is not an account of epistemic authority per se (though to be clear, she does not claim it is!); secondly, as will be explained in Chapter II, I think the kinds of relationships one can hold with epistemic authorities are far more varied than what has been discussed here. Even if Dormandy’s view, as a theory about the role of our reasons vis-àvis epistemic authority, is correct, it can only take us so far in defining epistemic authority itself. In order to make sense of the complexities that Dormandy has quite carefully captured with her criticisms of Zagzebski’s account, we need to make sense of how our reasons can play a variety of roles and exert a variety of different levels of force on responsible, epistemic behaviour, when interacting with epistemic authorities. As I will argue in Chapters II and III, the dynamics of our relationships with such authorities will vary, and the extent to which we can incorporate our own epistemic judgement throughout our interactions with epistemic authorities will depend on a variety of factors. I.4 JÄGER’S SOCRATIC EPISTEMIC AUTHORITY As we have already seen, Christoph Jäger has presented a series of arguments against the plausibility of the Preemption thesis in the form presented by Zagzebski. I have yet, however, to discuss Jäger’s own account of what at least one kind of epistemic authority may look like. Given the problems he identifies with Preemption, Jäger suggests that an authority should instead be considered as someone who can actually improve the non-expert’s ability to navigate reasons 45 pertaining to the relevant (supposedly) true proposition provided by expert testimony. In brief, Jäger refers to a Socratic Epistemic Authority as someone who provides the subject with an understanding of why p is the case, or at least an understanding of why their reasons do or do not support a belief in p, rather than merely providing them with a true belief in p, and preempting all of their previous reasons for or against p. Jäger defines this kind of epistemic authority in the following way: ‘[A]n epistemic authority for a given subject is someone who not only succeeds more often in attaining the truth, but who also is able to foster the subject’s overall insight into the problem under consideration’.79 Such an authority not only has more true than false beliefs in a given domain, but: ‘[. . .] also displays superior methodological skills and insights which enable him properly to assess evidence, reasons, methods of thinking and investigation, and so on, and to communicate such insights to others. Engaging with a Socratic authority will thus typically not only (rationally) motivate an epistemic agent to adjust his beliefs if they did not match the authority’s beliefs; it will also enable him to see what was right and what was wrong with his own grounds and reasoning methods, and what is right about the grounds and methods favored by the authority. A Socratic authority, in other words, serves not only as a source of maximizing true beliefs at the object level, but also a source of understanding.’80 I take Jäger to be suggesting that there are at least three necessary conditions for being such a Socratic Authority (SA): 1. She is more likely to have a true belief relative to some domain d. 2. She has superior methodological skills and insights into the epistemic workings of d, and these allow her to assess the relevant evidence, reasons, and methods of investigation relevant to d in a more reliable fashion. 3. She is able to communicate such insights to others. (SA3) is clearly the most novel aspect of Jäger’s account. Built into the very definition of Socratic 79 80 Jäger, ‘Epistemic Authority’, 179. Ibid. 46 epistemic authority is the means to provide a subject with the relevant epistemic standing to be able to both believe the relevant truths, relative to d, while also basing such belief in such truths on the totality of her evidence, specifically evidence afforded one by being communicated insight about the relevant methodology, skills, epistemic workings of d, etc. If the supposed authority cannot guide you towards an understanding of the relevant grounds of reasoning in d, then she simply is not a Socratic Authority. It seems, at first glance, Truth(B) and Total Reasons are thus meant to be respected, given the case that the authority is of the Socratic kind. Such authority ties or severs the connections between your reasons and the belief in p that you ought to hold, but does so in a way that seems internally consistent, psychologically plausible, and epistemically responsible. In a sense that will be cashed out in more detail in Chapter III, such an approach leaves the epistemic subject in such a position that she can actually form the resulting belief in p as ‘her own’—in respect of her own epistemic agency and autonomy. One ought to be careful, however: Jäger’s writes that SA can typically provide you with the rational motivation to believe that p (where p is a relevant (apparent) truth in d) and the understanding of the relevant grounds for reasons in d. Though I cannot be sure of Jäger’s intentions, it seems that he is suggesting that the two come apart: I take this to mean that a Socratic Authority is thus someone who is in the epistemically appropriate position to provide you with understanding, along with her testimony that p, but that she can nonetheless provide you with a rational basis for believing that p even if you fail to gain that understanding. It thus isn’t entirely clear whether or not such a view escapes the issue of Preemption: given a situation a subject has been given the opportunity to understand the relevant reasons for believing that p, but for some reason fails to do so, it would seem that Jäger would still have to tell some story regarding how she is rationally motivated to believe that p. Does she do so simply by weighing the evidence she does have, in the same way that she usually does (lacking an understanding of how one should reason 47 about p, as per the Socratic Authority’s explanation, and thus possibly landing herself in an epistemically unfavourable situation)? Or ought she to preemptively believe that p, given her failure to adopt a more rationally motivated position via the authority’s explanation? One possibility to keep in mind is the thought that Jäger’s additional condition (SA3) on epistemic authority provides some kind of insight into why an authority has the normative power she does over S. As we shall see, I certainly think there is something to the idea that the primary source of an epistemic authority’s normative power is not necessarily grounded in her relation to first-order issues in her domain, i.e. her having more true beliefs about the various subjects matters in her domains, but rather in her relationship to the higher-order issues that Jäger alludes to: the fact that she knows how to navigate the relevant reasons in d, that she can make sense of the methodology and discuss the principles that guide it, that she has the means to evaluate evidence based on certain criteria, etc. Jäger’s position, then, should be considered as a foreshadowing of what is to come in my own. The problem, however, is that Jäger’s account doesn’t give us a full picture of what kind of authority an epistemic authority can have for us in the case that we aren’t able to come specifically to an understanding of whether or not that p. Moreover, even in cases in which an authority can do just this, it isn’t entirely clear what normative power she has over us—a normative power that I agree with Zagzebski in thinking is constitutive of authority. This means that it isn’t exactly clear if Jäger has provided an account of epistemic authority at all. To elaborate on Jäger’s view, I think it would be beneficial to say more about what Jäger takes the relevant kind of understanding to be. Jäger has in mind a kind of understanding that he refers to as the Weight-of-Reasons Principle of Understanding: ‘The degree to which S understands a subject matter is proportional to S’s awareness of the relative epistemic weight of the total available reasons relevant to propositions 48 belonging to that subject matter’.81 This is, as Jäger points out, an account of only one type of understanding (which he considers to be in the vicinity of what Katherine Elgin refers to as an ‘understanding of a topic’ or ‘subject matter').82 Furthermore, Jäger requires two assumptions to hold in order for his account to make sense: first, that understanding can come in degrees, and, second, that understanding ‘involves something like grasping systematic connections among elements of a complex whole, or gaining insights into certain relations between items within a larger body of information’.83 The suggestion then, is that to understand a subject matter, or domain d,84 to degree n is to have some kind of proportional awareness of the relative epistemic weights of reasons within that subject matter, or d—i.e. to see how the reasons are related to, and are connected with, one another as well as with various doxastic attitudes (this is to grasp, to some extent, what Jäger refers to as a ‘web of reasons’).85 Jäger suggests that, in interacting with such Socratic Authority, one of our main rational aims is to acquire such understanding.86 He claims that when we engage with epistemic authorities (of the Socratic bent), we generally do far more than merely adopt the authority’s belief.87 We typically re-evaluate our reasons, and try to align our reasons with the truth, given the authority’s guidance. This, I think, is certainly true in many cases. Nevertheless, given that Ibid., 180. Which, I have to admit I find somewhat confusing: the kind understanding he describes seems to be more akin to the kind we refer to when we say that ‘S understands why p is the case’, given that Jäger talks about having insight into how reasons weight in favour, or relate to, p. 83 Ibid., 179-180. 84 I do not think it right to equate subject matter with domain, necessarily, given the context of epistemic authority. Assume that we will generally define expertise, and authority, relative to a given domain. It may be the case that such a domain can be broken down in to various subject matters and topics. An epistemic authority within a certain domain may be able to provide a subject with the relevant kind of understanding of one of those subject matters (which could be very narrow indeed!), without necessarily providing the same type of understanding of the domain. For this reason, I prefer to keep the two distinct, and thus use the term ‘domain’, rather than ‘subject’, or ‘topic’, as specifically reserved for the purpose of defining the relevant scope of expertise. 85 Jäger, ‘Epistemic Authority’, 180. 86 Ibid. 87 Ibid., 181. 81 82 49 Jäger’s account is presented in a rather limited form—first in that it only isolates a type of epistemic authority rather than epistemic authority per se, and, second, in that even this kind of authority only typically provides one with the ability to better meet Total Reasons—it only seems to suggest that, some of the time, given more ideal circumstances, we can do far better than Preemption suggests. Though even here one can imagine that Zagzebski will simply raise the same concerns she uses to motivate Preemption: if we give the non-expert or layman any room to weigh reasons for p in coming to believe, or not believe that p, we leave ourselves open to the possibility that she will come to a different conclusion than the epistemic authority (in this case, perhaps, because she has such a low degree of understanding), thus forfeiting our grip on Truth(B). This aside, however, I don’t see how Jäger’s view can take us very far in giving us a more satisfying account of epistemic authority as a whole, in that it is too narrowly concerned with one very particular kind of epistemic relationship between an epistemic authority and non-authority. As shall become clear, however, I do think that Jäger’s account puts us on the right path in an important way: it concerns itself with epistemic goods other than the acquisition of true beliefs.88 However, in order to get the kind of account of epistemic authority I am seeking, we will still have to supplement this consideration with further normative force. Again, what Zagzebski does seem to get right is the need for this normative power proper in accounting for epistemic authority. A further problem with Jäger’s account is that I think the notion of understanding is far too strong to be made use of in this context. Understanding, as I see it, is (generally) a more difficult cognitive achievement, arguably more valuable and challenging to obtain than 88 Not that such consideration excludes the acquisition of true beliefs. 50 knowledge.89 To suggest that an epistemic authority can simply provide another with an understanding of a subject via an interaction with them seems contrary to the nature of what we are generally referring to when we talk about understanding. Furthermore, understanding seems to be intuitively something that one gets by means of one’s own cognitive actions, not something that is transferred or passed on to one. As Daniel Wilkenfeld has put it, to understand a subject is importantly constituted by my thinking about the subject—not another agent’s thinking.90 For this reason I think the kind of epistemic good that Jäger is discussing is not necessarily the one that he thinks he has identified. As we shall see, I think a part of the reason Jäger’s account is incomplete (as with the rest), is that it fails to account for the dynamic details of the epistemic relationship in question, focusing too much on the static notion of what an authority can provide for us. An epistemic good similar to that which Jäger discusses will be discussed in Chapter II. Rather than focus on understanding per se, I suggest that we would be better off to think of something that may be more accurately thought of as a component of such understanding. Roughly, one of the kinds of epistemic goods an epistemic authority can provide is the ability to ‘make sense’ one’s own epistemic situation relative to a given domain, allowing one to gain a perspective on one’s own epistemic position regarding the relationship between one’s own reasons and the appropriate doxastic attitudes in that domain, measured by the standards and norms of that domain. This is not to understand a subject matter or domain in any broad sense, but rather to gain insight into certain relationships between reasons and conclusions, and, particularly, one’s own epistemic position in relation to these. ‘Understanding’ one’s own epistemic position in this sense can provide an agent with right kind of perspective to act in an 89 Duncan Pritchard, ‘Knowledge and Understanding’, Virtue Epistemology Naturalized: Bridges Between Virtue Epistemology and Philosophy of Science, edited by A. Fairweather, (New York: Springer, 2014), 322. 90 Daniel A. Wilkenfeld, ‘Understanding as representation manipulability’, Synthese 190:6 (2013), 1002. 51 epistemically responsible way in relation to the relevant domain, and the epistemic authority with which she chooses to engage. More will be said about this phenomenon below. Again, however, this is a small point: the more crucial issue is that Jäger’s account only concerns itself with one small part of the broader phenomenon that I aim to account for. I.5 CONSTANTIN AND GRUNDMANN’S LIMITED PREEMPTION Constantin and Grundmann have suggested an alternative version of Preemption, or a Preemptive account of epistemic authority.91 The account varies from Zagzebski’s both in that it is derived from different epistemic considerations, and that it has a far more limited in its application. Generally, Constantin and Grundmann think that Zagzebski is right to suggest that, when following an epistemic authority, it is not permissible to make use of one’s own reasons in determining whether p is or is not the case, and argue that Preemption can be identified as ‘a special case of undercutting defeat’.92 However, though they think the preemptive view to be ‘basically correct’, they do not think it to be correct ‘across the board’.93 As a preliminary, recall that, as I previously noted, Constantin and Grundmann take epistemic authority and expertise to be distinct concepts, a trend I follow in my own discussion.94 As they understand it, the notion of expertise ignores the specifically relational aspect of epistemic authority: this is to say that expertise does not vary in relation to different persons in the way that authority does.95 An expert, on their account is someone that has a substantial body of evidence, This is partly due to the kinds of criticisms that have been levelled against Zagzebski’s account (see I.3 and I.4), but also, as we shall see, due to other epistemological motivations having to do with the derivation of epistemic principles. 92 Constantin and Grundmann, ‘Epistemic Authority’. 93 Ibid., 2. 94 I in fact owe them both my gratitude for the clarity they brought to me on this issue, though I think I ultimately separate the two concepts even more so than they suggest (see Chapter II). 95 Constantin and Grundmann, ‘Epistemic Authority’, 6. 91 52 has the methods to reliably arrive at true beliefs and adequate credences from that evidence, and is able to use those methods. Such an expert may be an epistemic authority relative to someone who themselves has no expertise or knowledge of the relevant domain, but may not be an authority relative to another expert in the same domain, another expert with the same skills, evidence, etc. Thus, Constantin and Grundmann suggest the following account of epistemic authority: EA: A is an epistemic authority for S with respect to domain D iff S has justification to believe that i. A is an expert about D.96 ii. A is to a significant degree an epistemic superior to S with respect to D.97 The important point is that an epistemic authority is epistemically ‘superior’ to S. Constantin and Grundmann understand such superiority to be cashed out in terms of truth conduciveness, as well as in terms of being able to assign to right credences to the propositions relevant to D: ES: A is an epistemic superior to in relation to S on a given subject matter D iff A is more likely than S both to assign the right credences to propositions in D for a given pool of evidence and to be correct about matters in D.98 Not only must such an agent be more likely than S to be right on these terms, but must also be sufficiently likely—in other words there is some kind of threshold to be decided that, on the one hand, demarcates who is sufficiently more likely in order to be considered an authority, and, on the other hand, decides who is merely only a little epistemically better than S, in D, but not good enough to be considered superior in any meaningful sense.99 Given this account of epistemic authority, we can now take a look at how Constantin and Grundmann attempt to rationally justify Preemption. i.e. has a substantial evidence in D, has methods that reliably lead to true beliefs, and is able to apply those methods. 97 Constantin and Grundmann, ‘Epistemic Authority’, 7. 98 Ibid., 6. 99 Constantin and Grundmann suggest the example of some subject S1 who gets things right approximately 30% of the time in D, and a second subject S2 who gets things right approximately 35% of the time. They do not think that S2 is likely enough to get things right in order to be considered an epistemic superior, relative to S1. 96 53 Constantin and Grundmann claim the following: the ‘distinction between undercutting and rebutting defeaters runs parallel to the distinction between authority beliefs and other evidence’.100 Following Michael Bergmann,101 they define undercutting and rebutting defeaters as follows: Rebutting defeater: d is a rebutting defeater for a subject S’s belief that p iff d is a prima facie reason to believe that not-p. Undercutting defeater: d is an undercutting defeater for a subject S’s belief that p iff d is a prima facie reason for S to believe that, under her present circumstances, it is irrational for her to believe that p. Rebutting defeaters can thus be added to one’s total evidence, and appropriately weighted and balanced with them (and would not necessarily mean that one would be rationally required to stop believing that p). An undercutting defeater, on the other hand, cannot be merely added to one’s previous reasons, as the undercutting defeater informs one that there is something wrong with the connection between one’s evidence and reasons, and the truth or falsity of p. An undercutting defeater thus ‘normatively screen[s] off’ the defeated evidence from playing the role of being an evidential basis of our beliefs.102 Constantin and Grundmann motivate their account by considering the plausibility of doing otherwise in the face of authoritative testimony. Consider the case that you find out that an epistemic authority believes that p with a credence of .75. You have two options: either, you follow Preemption and match your credence with theirs, solely on the basis of your believing that they are epistemic authorities, or, you weigh the authoritative reason along with the rest of your reasons. Similarly to Zagzebski, Constantin and Grundmann point out that, given enough prior contra-reasons, you would end up with a lower credence in p than the authority, meaning that Constantin and Grundmann, ‘Epistemic Authority’, 11. Michael Bergmann, ‘Defeaters and Higher-Level Requirements’, Philosophical Quarterly (July 2005) 55:220, 424. 102 Constantin and Grundmann, ‘Epistemic Authority’, 13. 100 101 54 you may end up failing to track the truth; in the case that you have a lot of prior pro-reasons in favour of p, you would in-fact end up with a higher credence in p than the authority has. Why is this problematic? Because part of recognising that someone is an epistemic authority, on Constantin and Grundmann’s account, is believing that they have access to a wide range of evidence pertaining to their domain, D, and this evidence would supposedly include the kinds of reasons that one has access to. Now, given that another aspect of recognising someone as an epistemic authority is that we think of them as having a variety of methods at their disposal, that they do in fact apply to their reasons and evidence in order to reliably track the truth, it would follow that having any credence in p that is not the same as the authority’s credence in p would suggest that one takes oneself to be a better judge of the evidence on hand, and, even though one has i.) less evidence at one’s disposal, and is ii.) less able to reason from and evaluate such evidence. Unless one has reason to think that the authority has not considered the evidence you have, it would thus be irrational to have any other credence in p than the epistemic authority’s. To repeat, it is not that the authority believes that p that acts as an undercutting defeat to one’s evidence, but rather that, in hearing that the authority believes that p with credence x, you acquire reason ‘to believe that someone who is much better than you in assessing the evidence and who has already taken into account all of your relevant evidence ends up with a credence of 0.75 regarding p’.103 It is the fact that you recognise something about another agent, a necessary property of their being an epistemic authority, that you have reason to match your credence with theirs—this is what grounds the Preemption of your other reasons. Constantin and Grundmann provide four explanations for why a subject’s deviating from the authority’s judgement, and instead aggregating her own reasons and evidence with the authoritative reason, is problematic: 103 Ibid., 14. 55 1. A subject may evaluate her own evidence in a less competent way than the authority, as she is likely to do, by definition of her epistemic status relative to the epistemic authority. 2. A subject may have additional evidence or reasons (above the authoritative reasons) that are, unbeknownst to her, undercut by further evidence that the authority has—evidence that is opaque to the subject. This would be a case of an epistemic authority being in possession of an undercutting defeater. 3. A subject may have additional evidence or reasons (above the authoritative reason) that are, unbeknownst to her, defeated by further evidence that the authority has. In this case the epistemic authority would be in possession of a reason-defeating defeater. 4. A subject, in relying on her own evidence or reason, may in fact be double-counting such evidence, given that the epistemic authority’s own belief may be based upon the very same evidence or reason.104 Importantly, as Constantin and Grundmann are well aware, their account of epistemic authority varies from Zagzebski in that they do not consider epistemic authority as defined by Preemption—i.e. they don’t take Preemption to be constitutive of such authority. They refer to the kind of authority that they account for as grounded authority, in the sense that their account is supposed to ‘identify epistemically relevant properties of persons or institutions in virtue of which we plausibly may have to treat them as “authorities”’.105 It is in virtue of these properties that they think a subject, in recognising that these are authorities as such, has preemptive reasons to believe that p when an authority tells them that p is the case. The kind of authority that Zagzebski takes herself to be accounting for is what Constantin and Grundmann refer to as role authority: accounts of such authority first identify ‘the normative force of preemptive reasons and then 104 105 Ibid., 16-17. Ibid., 8. 56 [define] authorities with respect to it’.106 Constantin and Grundmann suggest that the focus on grounded authority is a useful one when considering the debate between those that defend Total Reasons and those that defend Preemption, since they don’t think that the former are necessarily interested in denying that there is such a thing as epistemic authority. Given that my project is to try to make sense of the kind of epistemic authority that Zagzebski has in mind, a kind of authority that is constituted by a certain normative power, I take myself to be interested in role authority. That aside, however, Constantin and Grundmann’s dismissal of role authority is too quick. Those in the Total Reasons camp have a specific problem with Preemption in the form that Zagzebski has presented it, but it does not follow that we cannot provide an alternate account, either of Preemption, or of some other normative element that could possibly constitute such epistemic authority. Though it is clear that Constantin and Grundmann’s limited Preemption cannot play this role, it could yet be the case that something else could. Furthermore, Constantin and Grundmann are also motivated by the thought that it would be odd to postulate a highly local principle that would make authoritative reasons behave in the way that Zagzebski describes—such a local principle would be, as they put it, ‘mysteriously special’.107 Grounding Preemption in the specific properties of an epistemic authority, and deriving it from broader epistemic principles, as they claim to have done, solves this peculiarity. However, I think we should be cautious: to suggest that Zagzebski’s notion of Preemption involves a purely local principle misses the broader point of her project. Zagzebski has not ‘discovered’ Preemption merely by considering one type of epistemic phenomenon, or relationship, but rather by considering a phenomenon that can be found, though in slightly 106 107 Ibid. Ibid., 11. 57 different forms, throughout a wide variety of contexts: epistemic, political, theological, moral, etc. Preemption is, following Raz’s account, central to the very concept of authority, not solely to the epistemic phenomenon in question. To forget this is to fail to grasp the nature of Zagzebski’s project, one that does not merely deal with the epistemic, but aims to show that there is a continuity between the epistemic realm, and others. Again, I think there is room for an account of epistemic authority that captures the strongly normative component that Zagzebski is trying to capture, while also doing justice to the considerations that the above criticisms have raised. My task is to forge such an account: one that identifies the strongly normative power that Zagzebski and I think are constitutive of authority proper, but also one that makes sense of the wide variety of role such authority can play, the different goods such authority can provide us with, and, quite crucially, all while remaining faithful to the complexities of the dynamic and multifaceted relationships that we can have with epistemic authorities. Ultimately, none of the accounts presented above capture the intricacies of the phenomenon I have in mind. As we shall see, I think there are a wide variety of issues that such discussions of expertise and epistemic authority generally fail to acknowledge. Nevertheless, it remains the case that all of the above accounts reveal something important about epistemic authority. My hope is that my own account can capture the best of all of the above views. This is the task I turn to in Chapter II. 58 II. EPISTEMIC AUTHORITY More than Mere True Belief In this chapter I present my own account of epistemic authority. My account, following in Jäger’s footsteps, puts important weight on the idea that an epistemic authority can play other roles than merely providing us with the epistemic good of acquiring more true beliefs. In II.1 I focus on the suggestion that possession of (a sufficiently large amount of) true beliefs, or the possibility of acquiring such a body of beliefs, is neither necessary nor sufficient for grounding epistemic authority, and, more precisely, that it cannot appropriately explain the nature of epistemic authority given the wider range of roles that such authority can take. I ultimately suggest that what is central to being such an authority is having a certain kind of know-how, and that—one’s stance on the knowledge-that/knowledge-how debate notwithstanding—this cannot be reduced to a simple discussion about the kinds of true beliefs that an authority has possession of, and can transfer to us. This is important because, as I argue, such knowledge-how can explain the variety of roles such authorities have, and the variety of relationships and interactions we can have with them. These roles are explored in II.2, where I look to motivate the thought that the epistemic good of Truth(B)108 should not be considered the only epistemic good that one can receive from an epistemic authority—that, rather, the kinds of epistemic improvements that we seek in dealing with such authorities come in a variety of forms. If this is correct, then an account As discussed in Chapter I, Truth(B) is the good of having more true beliefs, and fewer false beliefs. This is not the same as the goal of being closer to the ‘truth’ in a broader sense, though these goals, of course, often go together. 108 59 of epistemic authority ought to be able to make sense of these different goods, and how such authority provides us with access to such goods. In II.3, I give my account of expertise and epistemic authority, in an attempt to account for these various considerations. Lastly, as mentioned in Chapter I, I briefly discuss Michel Croce’s own account of epistemic authority, in contrast to my own, in II.4. Before I proceed, I should clarify one point. I do not intend to argue here that any of the goods that an epistemic authority provides us with are necessarily completely removed from considerations of ‘truth’ per se. Though I am personally sympathetic to a pluralistic view about epistemic values, the account of epistemic authority I present here is intended to be entirely compatible with either the view that all epistemic goods are ultimately derived from the value of truth per se, or with any opposing view. My only claim is that there is a lot more to epistemic authority than a mere connection to the epistemic good of Truth(B), as described in Chapter I, in the sense that there is more to the epistemic status of being epistemically superior, in the authoritative mode, than possessing a large amount of true beliefs, or being a means for others to directly acquire more true beliefs. As a friendly reminder: where one reads ‘Truth(B)’, I am referring to this epistemic goal, not truth in a broader sense. Ultimately, for those who remain unconvinced of this point, my account of epistemic authority should remain attractive, as shall be seen in II.2: it can still make better sense of the wide variety of ways in which we interact with those that I would call epistemic authorities. This, ultimately, is the crucial point of my argument. II.1 MORE THAN MERE TRUE BELIEF As should be fairly clear from the above discussion, philosophers have generally discussed the notions of expertise and epistemic authority in relation to the epistemic good of Truth(B). This 60 is, of course, not surprising, considering that truth, broadly speaking, is often considered to be the primary epistemic good: that which our epistemic investigations aim to acquire. If there is such a thing as epistemic authority, with the strong sense of normative power that Zagzebski suggests, then it would make sense to account for it in terms of such a good. It is my contention, however, that other epistemic goods than the mere acquisition of true beliefs can play just as important a role in explaining and accounting for the kinds of epistemic relationships that hold between epistemic agents of diverging epistemic status, and can in turn make sense of how certain individual can behave as, and play the role of, epistemic authorities.109 We have already seen that there is reason to be suspicious of Truth(B)’s ability to play the central role in ground epistemic authority—we saw this in the various criticisms levelled at Zagzebski’s account, in Chapter I: by justifying epistemic authority solely on grounds of our access to true beliefs, relative to a domain d, we are faced with a variety of epistemic problems that I presume we would wish to avoid if at all possible. Though I assume that we, as responsible epistemic agents, ought to aim at getting to the truth in the broader sense, I think we can at least say that our having the best possible epistemic standing does not merely entail acquiring any true beliefs, no matter the consequences: to be epistemically successful agents involves, at least, acquiring true beliefs, and not just any true beliefs, in the right way. As we saw in I.2, there are a variety of ways in which we can have access to true beliefs, and be more reliably connected to them via an epistemic authority’s testimony (or our knowledge of their own beliefs), and yet be in (what at least intuitively strike me as) epistemically unfavourable positions. My intuition is that in many of these unfavourable cases, seeking the acquisition of true beliefs alone, regardless of other epistemic considerations, is actually far too costly. If I am right, I think such considerations can be applied to other epistemic relationships as well, those between authorities, or those involving no authority whatsoever. That is, however, a story for another time. 109 61 In this section, to further underline the need to ensure that our notion of epistemic authority is not necessarily wedded to this particular epistemic good, I specifically discuss my various motivations for thinking that the acquisition of true beliefs, potential or actual, is neither necessary nor sufficient for the establishment of epistemic authority. As I discuss in II.1.A., it is not necessary because it is completely plausible to think of an epistemic authority in a domain of inquiry that one may not consider to produce the relevant kinds of true beliefs at all (either for contingent reasons, or in principle). In II.1.B., I argue that it is not sufficient for an individual to provide access to a larger body of true beliefs in order to be considered an epistemic authority. II.1.C., in turn, elaborates on a point that surfaces on multiple occasions in the II.1.A, and II.1.B: the idea that what all epistemic authorities share is not a relation to a body of true beliefs in a particular domain of inquiry but the knowledge-how, or ability, to engage in a certain kind of epistemic practice. This last point will be the focus of my account, in II.3. II.1.A. EPISTEMIC AUTHORITY WHERE THERE ARE NO TRUTHS There is a simple reason to avoid thinking of epistemic authorities merely as possessors, and conveyors, of true beliefs. I would like to defend the perhaps controversial view that there can be such a thing as an epistemic authority in a domain that does not strictly deal in truths at all: i.e. it does not furnish true beliefs as answers to the questions that it seeks out to answer. It follows that, if this is the case, then it would make no sense to talk of such authority as being constituted by the possession of, or acquisition of, true beliefs, relative to that domain.110 It should be noted that I here assume some sort of correspondence theory of truth, and that the 110 Or, specifically, true beliefs about the answers to the primary questions of import to be dealt with in that domain. There will obviously be true beliefs to be had with some questions. Refer to the examples below for further elaboration. 62 kind of true beliefs we are concerned with when discussing epistemic authorities are those that correspond to a certain way in which the world is. Those of my readers in favour of a pluralistic or pragmatic theory of truth may find much of the discussion less motivating. Even so, it should be here noted that if different domains of inquiry do in fact deal in different varieties of ‘truth’, as a pluralist may suggest, then an account of epistemic authority that nevertheless illuminates a core similarity between authorities across all such domains, without having to rely on a specific account of truth, ought to be considered more favourably than one that does not. I think my account does just that. I invite those readers who remain unconvinced by such considerations to appraise my argument in II.2 instead. Here I have in mind domains such as those which deal with issues of aesthetics, for one, but possibly including (and this will of course depend on one’s own views on such issues) domains such as philosophy, and religion. There are at least certain domains of inquiry in which we can question whether the products of inquiry in such domain are ones which we would feel comfortable referring to as facts, or true statements, in the same way as we would in other domains. Note that I am not suggesting that these are domains that simply deal with difficult questions, and, as such, have so far failed to provide us with the true conclusions that they seek out to discover—if this were the case one could still quite plausibly argue that any so-called epistemic authority in such a domain would be at least more likely to discover such truths than the non-authority. My suggestion is that some of these domains may simply not, as their primary focus, deal with such statements of fact at all. Now, I do not wish to get bogged down here with an intimate analysis of the truth-aptness of any of these suggested domains. My hope is that my reader can at least be convinced that it is plausible to think of there being such a domain that does not necessarily traffick in statements that we would refer to as being strictly true or false, and yet, that such domains may nevertheless 63 admit of their being an epistemic authority of their subject matter. Consider, for example, the literary interpreter, or film analyst111—someone whom you may refer to as an ‘expert of American Literature of the 20th century’, or an ‘authority on the films of David Lynch’.112 Being someone who has little experience with post-modern literature, and its interpretation, one might seek such an authority when trying to make sense of Thomas Pynchon’s Gravity’s Rainbow, for instance; being someone who has never strayed far from mainstream cinema, one may similarly seek guidance when trying to make sense of Inland Empire, or Lost Highway. What could such an authority provide us with? I would argue that they could, amongst other things, provide us with a kind of ‘understanding’ of the target piece of art, an interpretation of the pieces (and their components) that provide a coherent explanation of their ‘meaning’ and telos. The kind of understanding in question here could certainly be a ‘merely’ subjective kind—if one holds the view that understanding proper is factive. Nevertheless, such understanding certainly strikes me as a phenomenon that falls squarely within the scope of epistemological analysis: we here have two epistemic agents, interacting via the act of testimony, cognitively and intellectually engaged in adjusting their set of beliefs towards a certain topic. They certainly seem to be engaging in an intellectual inquiry! Let us consider the rejoinder, however: if such domains do not deal with truths, in the way that other domains do, why should it be the case that we refer to the relevant practices in such domains as in any sense epistemic? I think this is a very good, and very difficult, question to answer: I certainly do not wish to claim here that I can answer, and provide an account of, what it means for something to be epistemic if not in some way connected to matters of truth. I would, Another example, which may have more gravitas for some, is that of the interpreter of religious texts, on the condition that one does not believe that such texts are the direct word of God, but are rather ‘living documents’ or religious guides. 112 I apologise to any of my readers who have strong views about the semantics of statements about literature— again, my aim is only to clarify, by example, a possibility that I think most would find plausible in some domain or another. 111 64 however, like to suggest something a little more modest: that it is possible for there to be such things as epistemic practices—in that they provide us with beliefs, intricate views and perspectives on the world, possibly a certain kind of understanding, or even know-how, etc.—and that, even though such practices do not lead to true beliefs about the topics relevant to the domain in question, they nevertheless put us in an intellectually better and richer position. There are a few ways in which this could be the case. I think it would be odd, first of all, to suggest that one could not have a better- or worseformed belief on matters of art interpretation and analysis, even in the case that such beliefs could not be considered to be true or false. Beliefs about the manner in which a passage in Gravity’s Rainbow should be interpreted are not claims that one can make willy-nilly, in an epistemically non-responsible fashion: such beliefs ought to be formed on the basis of, and be responsive to, the evidence—such as the existence of other passages within the text, biographical facts about Thomas Pynchon,113 and so forth; such beliefs ought to be internally coherent and consistent; and such beliefs ought to fit well with other held beliefs about Gravity’s Rainbow, as well as higher-order beliefs about literary analysis, or art, as a whole. In short, they ought to be wellformed and reasoned beliefs. Secondly, it is not as if the beliefs we form about such matters hold no relationship to our broader views on the world of fact, or play no role in how we interpret such facts. My understanding of a piece of art, and its underlying political theme, for example, can greatly influence how I in-turn go on to analyse world events, and how I react to a political party’s policies. People reference fictional characters when making sense of real-life personalities. I would go so far as to argue that the kind of know-how that one can acquire from learning how to interpret various pieces of art can often be applied to the real-world directly, and that such 113 Not that there are many of these to be found, given his notoriously well-guarded privacy. 65 application can be epistemically beneficial. This is to suggest that it is possible for you to learn a certain kind of epistemic behaviour, practicing and honing it, in a domain that does not deal in true beliefs, or truth per se, that nevertheless improves your epistemic status more broadly speaking. For these reasons, it strikes me as implausible to disregard the above kinds of activities from consideration if we are to say that epistemology is in any way interested in making sense of which beliefs, or other cognitive states, are ‘good’ ones to have. And, if we include these considerations in the scope of epistemology, then I see no reason why we ought to find it in any way peculiar to consider the existence of an epistemic authority in such a domain. Perhaps my reader is not particularly motivated by the literary example. An issue closer to home may do the trick: what of the epistemic authority of the philosopher? Clearly, one cannot claim that the philosopher deals in no truths at all: she at minimum holds true beliefs about which philosophers have held such-and-such a position, the status of the current literature on Gettier-style cases, etc., just as the literary scholar has access to similar true beliefs about the lives of authors, publication dates, editorial changes, etc. Again, however, these are not the primary intended product of philosophical investigation: the philosopher is instead tasked with answering the fundamental and difficult questions—metaphysical, epistemological, ethical, or otherwise. Arguably, many of these questions are not of the sort that can provide us with clearcut, ‘true’ answers, answers that correspond to an objective standard and reality. Nevertheless, it seems to me that the well-trained philosopher, relative to those without such training, is in some sense an authority in relation to these questions. To suggest that the scope of the philosopher’s epistemic authority is instead merely limited to questions of publication history, or social facts about the current state of philosophical belief and research, strikes me as particularly odd. We 66 would, after all, think it highly peculiar if someone suggested the same thing about a physicist’s epistemic authority. Rather, I think the philosopher has authority over her domain. She is up to a certain task. When she provides us with a conceptual analysis of knowledge, one could argue that we are provided with an account that is intended to best accommodate for the evidence at hand, the relevant practices of the kinds of epistemic agents that we are, and most coherently deal with objections to previous epistemological theories. Such answers are often meant to prove themselves to be the most useful to us: a conception of what knowledge is that fails to accommodate for our quotidian intuitions and epistemic practices seems may be of very little use to us, and thus of little philosophical significance. These analyses are meant to be tools that best explain our activities, our common use of terms such as ‘knowledge’, ‘rationality’, ‘justice’, and ‘good’. Perhaps there is no fact to the matter as to what knowledge is in the same sense that there is a fact to the matter as to how old the universe is—this does not change the fact that the philosopher can have authority over such inquiry. My suggestion is that an epistemic authority in such a domain as philosophy, that provides us access to such insights as the ones illustrated above, is not one because they hold more true beliefs about the topic at hand: they are considered thus because they have the philosophical skill, the intuitive insights and analytical expertise to best navigate the literature, to consider reasons, to articulate critiques, and garner evidence for or against a certain treatise. They are intellectually more capable than others, relative to a certain domain of inquiry. As we shall see, this is what I take to be the fundamental core of the kind of authority I have set out to account for. My reader’s personal views on meta-philosophy aside: its shouldn’t matter whether the specific domains I have used as examples deal significantly in truths or not—my suggestion is that 67 whether they do or not, we can coherently conceive of them as admitting of epistemic authorities, precisely because what matters first and foremost is that there are individuals who have developed the relevant kinds of knowledge-how in order to be considered as such. We can remain undecided on the status of these disciplines as sources of true beliefs, and yet still acknowledge that they involve a certain kind of epistemic practice, and that they provide us with epistemic goods nonetheless. So long as this possibility remains, I think we have reason to prefer an account of epistemic authority that does not put its full weight on the possession of acquisition of true beliefs, or the epistemic good of Truth(B). II.1.B. KNOWING TRUTHS BUT LACKING AUTHORITY My next point is this: merely possessing an abundance of true beliefs, or having a more reliable connection to such truths, cannot by itself ground the kind of authority we are considering. Now, I do not think this is by any means a surprising or controversial claim. Many accounts of expertise and epistemic authority directly or indirectly allude to other conditions required to establish such authority.114 Nevertheless, I think it is useful for present purposes to underline the point, and to illustrate it further. By showing that such considerations of truthacquisition cannot alone ground the kind of authority in question, in conjunction with my illuminating the fact that other facts can, I hope to give my reader further reason to accept my own account of epistemic authority. Furthermore, though other accounts of expertise/epistemic authority clearly do make reference to such considerations, it is my contention that their failure Alvin Goldman refers to an expert having answers to various second-order questions, i.e. questions about the methodology, existence of evidence, weight of reasons, etc., of certain domain of inquiry, as well as a general cognitive ability to be able to answer knew and future questions in that domain. See Alvin Goldman, ‘Experts’, 114115. In the literature on epistemic authority, as we have seen, Jäger refers to ‘methodological skill and insight’, and Constantin and Grundmann refer to the ability to make use of a methodology, to give but two examples. 114 68 to ground authority on these other conditions is where their accounts go astray, or can at least be shown to be lacking in some regard. To see that a possession of an adequately large body of true beliefs cannot be sufficient, in and of itself, to ground authority (at least the kind we have in mind), I think it suffices to consider the following illustration. Imagine an individual of the sort that we may refer to as a ‘trivia junky’ on some particular topic, or ‘book nerd’ in relation to a certain domain of inquiry. This Trivia Junky knows a lot, for example, about physics. In fact, to make the point more salient, let us assume that this individual, in some respects, knows a lot more than many actual practicing research physicists, particularly when it comes to what Goldman has referred to as the ‘primary questions’ of a domain of inquiry, i.e. the ‘principle questions of interest to the researchers or students of the subject matter’.115 He can recite a wide range of facts about what physicists believe, for example, from the common and well-agreed upon, to more obscure and contentious facts. He can regurgitate passages from a wide range of published research articles in physics, encompassing a broad range of topics. Perhaps he is even better than your average physicist at citing the right experimental evidence when supporting certain conclusions, thus answering questions such as ‘why do physicists believe that p?’ more successfully.116 Could this Trivia Junky play the role of epistemic authority for me, in the strong normative sense that I have in mind? I do not think so. This is because I think there is a fundamental difference between the mere ability to reliably transfer true beliefs (or perhaps even Goldman, ‘Experts’, 115. These questions are in contrast to the secondary questions: those that concern evidence, reasons, and arguments for answers to the primary questions. 116 Note that I do not think it makes a difference if we further stipulate that our Trivia Junky also possesses a wealth of knowledge about what Goldman calls secondary questions in a domain of inquiry: i.e. he has many true beliefs about the available evidence and arguments for certain answers to the primary questions. Such an agent still lacks the relevant authoritative standing: he is only a reporter of facts, and not appropriately situated as a participant in the relevant epistemic practice. I will return to this particular challenge to my argument, however, in II.1.C. 115 69 any other kind of epistemic good) and being in some sense a potential originator of such goods. Only the latter can provide an individual with the kind of authority that we have in mind. To understand why I think this the case, it would may be illuminating to take a brief detour here, and once again contrast epistemic authority with the kind of political authority that Zagzebski uses as a model of her own account. As briefly mentioned in Chapter I, Raz thinks that a political authority’s utterances provide me with a reason to act that hold independent of what the content of the utterance is. What gives the uttered words such power over me is not that they express a certain meaning, but simply that the authority has uttered them.117 Now, we may have good reason to think that this cannot quite be the case in the epistemic domain: if a belief is a good one to have, for instance, then surely the normative power acting upon me to hold such a belief is the fact that it is a good one, not that an ‘authority’ has uttered it! However, I think this is too fast: this would be true, if I had access to the relevant kinds of reasons, understanding, and evidence to come to such conclusions on my own. But the antecedent of this conditional is precisely what we are stipulating to not hold when considering epistemic authority: here our project is to make sense of our social-epistemic position on the assumption that we are often far removed from having such access. Yes, if we were some kind of (supposedly) ideal epistemic agents, with ideal access to the facts of the world, then it would make no sense to discuss our epistemic behaviour being in any sense normatively motivated by another agent. Given that this is not our situation, my argument is that it makes perfect sense to suggest that there are cases in which we epistemically behave in a certain fashion because of authority. When a political authority makes an utterance, and this utterance gives me reason to act a certain way, the authority is in some sense a source of that action’s being proper, or fitting, to the The idea that I am alluding to here involves what Raz and Zagzebski refer to as the Content-Independence Thesis and the No Difference Thesis. See Chapter I.1 for more detail. I choose to discuss the matter in less technical terms here, for clarity’s sake. 117 70 political situation. Much in the same way, I think that when an epistemic authority gives us reason to epistemically behave in a certain fashion, they are also in some sense the source of that behaviour being epistemically apt. In both cases, if we were to defiantly ask the authority why we should behave as such, it would not be a completely empty gesture for her to respond with ‘Because I said so’—the point being that there is something about her agency that establishes her authority, and this is what an account of epistemic authority ought to aim to explain. This element is what the Trivia Junky lacks. I do not think, then, that the kind of relationship we are alluding to when discussing epistemic authority is merely an instantiation of a reliable signalling of, or reliable connection to, a body of facts, or epistemic goods more broadly. The Trivia Junky does not have such authority over us in any meaningful sense, and we cannot have a relationship with him of the sort that we do have with our epistemic superiors, such as teachers, doctors, scientists, religious authorities, etc. This is because such a person, though in possession of many true beliefs relative to d, does not possess certain qualities and characteristics that I take to be essential to the kinds of epistemic status we are trying to make sense of: he does not possess the skills, the abilities, or the know-how, to partake in the relevant kind of epistemic investigations that pertain to d (even if he may be able to state certain facts or beliefs about these processes). An agent who lacks this quality, no matter how reliable a connection to the truth they provide, does not have any kind of authority over our own epistemic practice: they would merely be a tool, much like a reliable thermometer, for us to use in our own epistemic investigations and calculations—not an epistemic agent completing some epistemic work on our behalf, and authoritatively modifying our own epistemic behaviour. I will have more to say about this set of skills, abilities, and know-how, in II.3. Here I would like to briefly elaborate on what I said above about an authority being in some sense an ‘originator’ of the epistemic goods she provides, not merely a mere reporter or bearer of such 71 goods. As will be seen, my account takes seriously the idea that an epistemic authority is someone who has the skills, abilities, and know-how to partake in a certain kind of epistemic practice, the one relevant to inquiry in a certain domain. It is this fact that grounds their authority, because it is this fact that establishes them as more than mere conveyors of true belief (or bearers of any other epistemic good). An epistemic authority is someone who can discover further truths, create new explanations and theories, improve on previous understanding, etc. In this sense they are—at least potentially, if not actually—‘originators’ of the existence of certain epistemic goods in our lives.118 What matters fundamentally is that there is an epistemic practice, one that I do not have the ability (or time, or resources) to participate in, and one that is a source of certain epistemic goods; an epistemic authority is someone who does partake in that practice, and in doing so instantiates that practice, and is a part of it. Epistemic authorities are not merely the tools of our own epistemic agency: they are in some sense important agencies in their own right, agencies with some sort of normative power over our own. There is nothing in the Trivia Junky case that suggests how or why such a creature would have such a normative power. Even Zagzebski, though she defends the notion of Preemption in terms of our acquisition of true beliefs, would, I think, deny such status to such a person. Such agents do not exemplify the kinds of epistemic qualities that we would generally attribute to someone who deserves our epistemic respect: in Zagzebski’s terms, such a creature would not be an epistemic exemplar, someone to emulate, for such a creature would not exemplify the kind of epistemic behaviour that we ourselves would conduct if we were to be more conscientious, better trained, unconstrained by temporal, financial, cognitive, etc., limitations. The intuitive point, I think, is that the Trivia Junky is not in any meaningful sense a better Of course, my language here is intended to be somewhat metaphorical: they do not literally create ‘truth’, in the broader sense. They can, however, play the role of creator when it comes to establishing the social fact that we, as individuals and as a society, believe that p, or understand that p, or hold a certain theory to be true. 118 72 epistemic agent than we are, no matter how many true beliefs he may hold. Again, how we get to the truth, and the way in which we have true beliefs seems to play an intuitively central role in measuring our epistemic success. It is important to note that Zagzebski does herself refer to much more than the possession of true beliefs when discussing epistemic authorities: such an agent is ‘a person who has more of the qualities I trust in myself insofar as I am epistemically conscientious’;119 she appropriately weighs her evidence, tests hypotheses, conducts further research, considers alternatives, remains cautious, etc. It is for this reason that I find it somewhat puzzling that Zagzebski has decided to defend the notion of Preemption, as the constitutive component of such epistemic authority, merely on the grounds that such authorities can act as reliable sources of true beliefs: it is her behaviour, not the results of such behaviour, that provide her with her status (even if we may value the behaviour itself as a means to such results). To reiterate, I take it that there is much more obvious candidate for a grounds for epistemic authority: the very fact that such agents are involved in a certain epistemic practice— that they have the skills, access to evidence, and a knowledge of a certain kind of methodology constitutive of that practice. I think the strongest argument against thinking of Truth(B) as the grounds for epistemic authority is that there is this alternative account available, one that makes sense of our actual social-epistemic practice, and one that does not run afoul of the same counterintuitive consequences as Zagzebski’s account of epistemic authority does. In section II.3. I will present this alternate view.120 Before I do so, however, I return to an objection briefly alluded to above: the idea that the kind of know-how that I am suggesting is partly constitutive of epistemic Zagzebski, Epistemic Authority, 108-109. Again, it should be noted that I am not claiming that other accounts of expertise or epistemic authority have failed to note this. However, my point is that they put too little weight on such skills, abilities, and know-how, and do not identify this know-how as the locus of such authority—identifying instead the good of acquiring true beliefs. 119 120 73 authority can be made sense of in terms of true beliefs, thus falling under the scope of Truth(B). I turn to this response now. II.1.C. EPISTEMIC AUTHORITY & KNOWLEDGE-HOW There is an obvious objection the above argument: of course we don’t think of an epistemic authority as someone who can merely provide us with first-order, true beliefs about issues in a certain domain. As already noted for example, Goldman claims that a central component to being a (cognitive) expert is also possessing what he refers to as more true beliefs about the answers to ‘secondary questions’ in a domain:121 questions about the evidence, arguments, and strength thereof, in favour of answers to the primary questions. What I refer to as possession of a certain skill set, or know-how, one may object, could plausibly be made sense of by talk of such answers to these ‘secondary questions’. Why then, cannot we think of an epistemic authority as someone who has an abundance of true beliefs such as these, as well as answers to the kinds of questions that a certain epistemic practice aims to answer? Why complicate things, when Truth(B) really can make sense of such issues? A first answer to this objection has roughly been given above, in the sense that I think there is a fundamental difference between being the kind of epistemic agent that can merely report, or transfer, certain epistemic goods, and one that is in position to be, in a meaningful sense, the originator of such goods. The latter requires having certain skills, abilities, and know-how, which the former lacks. This, however, requires more discussion. 121 Though Goldman also makes the claim that we would consider someone who has more knowledge of ‘secondary’ questions but less of ‘primary’ questions as less of an expert than one who has more knowledge of ‘primary’ questions than ‘secondary’. This strikes me as completely backwards, for reasons that, if not already clear, will become clearer in the following sections. 74 My intuition is that the kinds of skills, abilities, and know-how under discussion are in no way reducible to the mere propositional attitudes that such epistemic authorities have. I do not, however, wish for my account to require a commitment to either side of the knowledgehow/knowledge-that debate, nor do I aim to argue here for why one can or cannot be reduced to the other. Here I aim to provide reasons to think that, whether or not knowledge-how really is ultimately some form of knowledge-that, we ought not to limit our notion of epistemic authorities in such a way that it only apply to agents that have conscious access to properly formed propositional attitudes on all relevant issues pertaining to a certain kind of epistemic practice, and that someone who behaves in the right epistemic fashion, even if they cannot provide us with as many true beliefs about this practice, may have more right to be considered an epistemic authority, than someone who can recite all the relevant facts. Let me elaborate. When it comes to having a skill, or know-how, there is often going to be a vast difference between the stated beliefs such agents have about the practice that they are involved in, and the very way in which they participate in the practice itself. Someone who has the ability to perform a certain activity does not necessarily have to have the corresponding ability to express precisely what they are doing, nor explain how they are doing it. One may even state false things about one’s own know-how when considering such ability or dispositions122 consciously.123 This is quite evident outside the epistemic domain: ask a professional golfer, for example, how best to strike a golf ball, and not only may they fail to tell you, the very act of thinking about the swing in action may hinder them from being able to properly perform it (this phenomenon is sometimes referred I don’t think it necessary to take a stance on whether knowledge-how is specifically an ability, or a disposition, in the context of this chapter. Nevertheless, I will admit a preference for the latter view of knowledge-how, and think that it better aligns with my general account of epistemic authority. For an illustration of this view, see Gilbert Ryle, The Concept of Mind (Chicago: The University of Chicago Press, 1949), 55. 123 For a review of the relevant evidence in cognitive science, and other reasons to deny the intellectualist thesis about knowledge-how, see: Charles Wallis, ‘Consciousness, Context, and Know-How’, Synthese 160 (2008), 137. 122 75 to as ‘paralysis by analysis’).124 Whether or not there is some sense in which we can understand such knowledge-how in terms of unconscious or ineffable knowledge-that, the point remains that such skilled individuals are not in the position to express the relevant propositions. Even if you want to take the strongly intellectualist stance of claiming that they do have true beliefs about the rules and methods of a certain practice, in whichever sense can be made plausible, it is very difficult to understand how such beliefs could play a role the social-epistemic relationships that are relevant here. What matters to the fact of recognising someone as an epistemic authority is not that they possess some ‘beliefs' about the matter in some convoluted philosophical sense, especially when they are not able to express these to you, but rather that they behave a certain way, that they are skilled and well-practiced, that they act with a certain know-how. I would argue that we recognise agents as epistemic authorities when we in some way attribute to them these skills and abilities—even if this is just through looking at their credentials, resumé, etc. It certainly is not the case that we recognise them in terms of their possessing knowledge-that in a way completely opaque to their own first-personal perspective, never mind our own, far more removed, point of view. In fact, I would argue that the act of recognising someone as holding a series of true beliefs about an epistemic practice is a different question entirely. The latter kind of inquiry is one often relegated to a different kind of individual altogether: the expert-expert, perhaps, the individual whose field of study is precisely the epistemic behaviour of those in other domains (the philosopher or sociologist of science would fall under this category, for instance). Perhaps we ought to be suspicious here, however: could we really treat someone (and be epistemically responsible in doing so) as an epistemic authority if they were not able to consciously access, and clearly articulate the ‘rules of the game’, as it were? Here I think we 124 Readers interested in a particularly interesting case should consider the American golfer Ralph J. Guldahl. Guldahl was considered to be one greatest golfers of the 1930s, but, when commissioned to write a book on how to golf, abruptly found himself no longer playing his best. (Or at least some have argued as much.) 76 should be careful to avoid thinking that such a question could have clear answer that would cut across all cases. I think it certainly the case that we would, at least in most standard cases, be in a bad position to engage with someone qua epistemic authority if they were unable to articulate any such propositions pertaining to the epistemic practice that they were participants of. That being said, there is a large difference between someone who can say nothing about the skills, ability, and know-how they possess, and someone who can say something, but not everything. My suggestion is only that it is implausible to think that an epistemic authority be someone who only trafficks in true beliefs about so-called ‘secondary’ issues within a certain domain. We have to make room for the idea that it is knowledge-how more broadly, effable and ineffable, that is of central importance when it comes to considering those whom we wish to interact with as epistemic authorities. Furthermore, I will go so far as to defend the plausibility of there being an epistemic authority such that they can see very little indeed about the very practice with which they engage: certain epistemic practitioners may engage with such activities on a more intuitive and non-propositional level. Though such an epistemic authority may not be in a position to provide as much for the non-authority, in terms of epistemic goods, she will nonetheless classify as an epistemic authority under the account I present in II.3—and I think rightly so. At risk of belabouring the point, consider the difference between the scientist herself, and the philosopher of science. Arguably there are ways in which the scientist could be argued to not fully ‘know’ precisely what she is doing when she goes about her work: the philosopher of science steps in, attempting to properly articulate and make sense of the scientific methodology, and to analyse the actual behaviour of those that partake in it. There is a clear sense, however, in which the philosopher of science (in most cases) does not have the appropriate kind of knowledge about scientific practice, however, and this is the kind of knowledge I am alluding to: knowledge-how in the strictest sense. There is quite the gap between having a theoretical understanding of a certain 77 practice, expressed propositionally, and having the actual kind of skill, ability, and general knowhow required to partake in that practice. To underline the point: I wager that most would agree that when it comes to matters of science, we are right to treat the scientist as our epistemic authority, not the philosopher of science, though this point will be made clearer in II.3.125 II.2. BEYOND TRUE BELIEF: A VARIETY OF EPISTEMIC GOODS If my reader remains unconvinced by the considerations of the previous section, this section’s argument is intended as motivating a stand-alone reason to adopt the account of epistemic authority I provide in II.3, and should not be considered as relying on the arguments of II.1. Whether or not we consider all epistemic authorities as having such status due to some relationship to a possession or acquisition of more true beliefs (and fewer false ones), my argument here is that we are still better served by an account that focuses on different facets of the authority’s character, given that these facets can better explain the wide plethora of goods the such agents can provide us with, including the relevant true beliefs. Something should be made very clear from the outset, however: my ultimate goal in discussing the range of epistemic goods that we may seek and acquire from an epistemic authority is not to show that they cannot be understood, whatsoever, in terms of truth—this may certainly be possible. My point in doing so, rather, is to illustrate the fact that it makes far more sense to ground these wide-ranging epistemic goods in the practice from which they are a result, a practice that an authority’s know-how allows us to access. If epistemic authorities can provide, This is not to say that the latter has no role to play—in fact, I think philosophy has a fundamentally important role to play, precisely in its ability to elucidate matters of methodology and practice to both the authority and the non-authority. Here, however, the point is that, when it comes to engaging with that practice itself, the scientist holds authority, unless we are perhaps asking specific questions about how to model the rational process of the scientific method, for instance, in which case I think it is fairly easy to ascertain that we are dealing with a different domain of inquiry, and should turn to the philosopher for guidance. 125 78 and be the source of, as diverse a range of epistemic goods as I suggest, then accounts focused on an epistemic authority’s relation to true beliefs have a lot of explanatory work to do in answering how and why such a relation can also furnish other, often more complicated, epistemic goods. As shall be shown, I will argue that my own account does not have this explanatory problem. Thus, I return to elaborate on a point previously made: not only is Truth(B) (the good of possessing more true beliefs, and fewer false beliefs) not the basis for epistemic authority, it is not the sole good that we seek when dealing with epistemic authorities. This is, I take it, one of the key insights of Jäger’s discussion of Socratic epistemic authority. Zagzebski’s suggestion that we accept the Preemption thesis on the grounds that Truth(B) is of more importance than Total Reasons fails to account for the fact that we may seek other epistemic goods in our interactions with epistemic authorities, and that we do well epistemically for doing so. My suggestion is that we do not limit our account of epistemic authority such that it can only make sense of our epistemic relationships with such agents in terms of our desire for one epistemic good: rather, we should seek an account of epistemic authority that can make sense of how we interact with our epistemic superiors in seeking out any sort of epistemic good. I would argue that it would be a limitation of, and a strike against, any account of epistemic authority that failed to account for how we can acquire these other goods via our relationships with such individuals. Zagzebski’s account, as should now be clear, cannot adequately make sense of this, as her view sees epistemic authority as constituted by Preemption, a power narrowly described as concerning our beliefs, and reasons for them, and, furthermore, a power justified by its providing us with a source of more true beliefs, potentially at the cost of consideration of other important epistemic goods. The criticisms levelled at Zagzebski’s account have shown that such social-epistemic relations, if based solely on the goal of acquiring truths, can in fact leave us with a highly 79 undesirable epistemic standing. True belief or not, the agent who preemptively believes may find herself alienated from her belief, her reasons cut off from the supposedly epistemically desirable doxastic state that Zagzebski prescribes for her. Now, this point is clearly open to competing intuitions: perhaps some will simply accept this outcome on the grounds that, all things being equal, truth is the final good of our epistemic inquiries, and thus is the only epistemic good of relevance to the kinds social-epistemic relations we are considering. Before those readers who are attracted to such a solution settle for it, however, allow me to motivate an alternative view that I think can, again, make better sense of our actual social-epistemic practices. Let me do this by illustrating a variety of alternate ways in which we do in fact interact with epistemic authorities— interactions that I think a theory of epistemic authority ought to be able to account for, for failure to do so would reveal any such account as empirically inadequate. I think that it is more broadly the case that we interact with epistemic authorities so as to generally improve our own epistemic positions. Whether the final value of such improvement is an increased ability to reach the truth, inside and outside of the domain in question, it is certainly not always the case that the immediate desired product of such epistemic behaviour is true belief. In the previous section I discussed the possibility of interacting with an epistemic authority for the sake of acquiring a certain interpretive perspective on a piece of art, or for a certain analysis of a philosophical concept. I suggested that these could be epistemically beneficial for us towards a broader epistemic standing towards the actual, real, world, but that they could also be epistemically evaluated without recourse to this broader application: as simply being better or worse beliefs to have given a responsiveness to evidence and coherence with other beliefs. My suggestion is that these are epistemically good beliefs to have, even if we hold the view that they are not ‘true’ in the usual meaning. 80 Let us look at further examples. Consider the plausible claim that we sometimes we interact with epistemic authorities not to acquire a basic, true belief in some relevant proposition p, but rather to acquire the less tangible and more complicated good of making sense of our own epistemic experiences and situation in light of some domain of inquiry. As an epistemic being, I experience myself as having certain kinds of reasons, and think of these reasons as having a certain connection with what I think I ought to believe. Common sense suggests that the way I experience things recommends that I make certain conclusions about how the world is, and my place within it. To simply trust myself as an epistemic agent would likely result in accepting these intuitive conclusions. However, I am also aware of the fact that others, trained in certain epistemic practices, have very different views about how things are. The physicist tells me that objects are not solid ‘all the way through’ as I may experience them, but are composed of vast amounts of space; the mathematician tells me that, contrary to my intuitive grasp of the infinite, there are in fact multiple sizes of infinity; the religious mystic tells me that what I experience of the world, and myself in it, is an illusion, and that there is a deeper reality beyond that of appearance . . . In recognising these agents as my epistemic superiors, as being in some sense epistemic authorities over certain areas of my epistemic behaviour, I wish to recognise something about myself as an epistemic agent. In interacting with these different epistemic practices, I sometimes want to better understand my own epistemic position in relation to these practices, because understanding my own epistemic standing is itself an epistemic good for which I desire: it strikes me as an essential component to my epistemic autonomy and rationality to have this understanding. This reflective awareness of my own epistemic standing—whether or not it is required to acquire the true beliefs relevant to the domain in question, etc.—strikes me as an obvious epistemic good, one desirable in itself, and one that can be attained via our socialepistemic relations with the kind of agents we are here considering as epistemic authorities. This 81 is the kind of good that may be classified under what Jonathan Kvanvig has more broadly referred to as ‘viewpoints that make sense of the course of experience’,126 or ‘sense-making’,127 and a key component of what Stephen Grimm refers to as subjective understanding.128 129 I think it is this kind of epistemic good that Jäger is alluding to in his own account of Socratic epistemic authority—though I think it is a mistake to refer to it as understanding per se, as I have noted above. I may also, for example, desire to partake in a certain kind of epistemic practice, and this requires training, education, and guidance from the relevant kind of epistemic authority. An epistemic authority can provide me with insight into how to behave in order to succeed in a certain kind of epistemic practice—she can train me to be able to come to the relevant kinds of conclusions in a certain domain on my own terms; she can present me with new means for acquiring and evaluating evidence. This is to say that an epistemic authority can provide me not with new reasons and beliefs, but with new ways of making use of reasons to come to beliefs—she can provide me with higher-order reasons, tools, a range of methodologies, etc. Not only do epistemic authorities traffick in true beliefs, or other doxastic states and propositional attitudes, but also in know-how, or a certain kind of epistemic behaviour. I think this is a particularly illuminating kind of situation to consider: here the very component that I will argue is central to epistemic authority, is in fact the epistemic good being sought. Sometimes we interact with epistemic authorities not only in an attempt to become better epistemic agents ourselves individually, but also to collaborate with them in some epistemic project. More importantly (and this is a point I will return to throughout the following chapters): it is Jonathan Kvanvig, ‘Truth Is not the Primary Epistemic Goal’, Contemporary Debates in Epistemology, edited by Matthias Steup and Ernest Sosa (Malden: Blackwell Publishing, 2016), 286. 127 Ibid., 287. 128 Stephen Grimm, ‘The Value of Understanding’, Philosophy Compass 7, 108. 129 Note that I say a component of such a phenomenon—I do not claim that what I am discussing here is subjective understanding, as Grimm means it, per se. 126 82 often a necessary condition of successful inquiry in certain domains that an authority and nonauthority do collaborate in the relevant investigation. Prior accounts of expertise and epistemic authority, being all too focused on the transmission of true beliefs, have failed to recognise that there are such cases. Pedagogy is a clear example: in interacting with a teacher I defer to their authority on a particular subject matter, but I also wish to remain epistemically autonomous in the sense that I am attempting to learn how to be a responsible epistemic agent myself. The teacher-student relationship involves a certain kind of epistemically guided process: epistemic authorities can help us reach our own epistemic conclusions, but under their guidance and correction. Here is another way to put this point: one can place oneself under the supervision of an epistemic authority. This kind of relationship is sufficiently complex, and it involves a complexity that is greatly obfuscated by a simple reduction of expertise and epistemic authority to the role of reliably providing true beliefs.130 Importantly, I may also seek certain kinds of propositional attitudes that fall short of belief, or justified belief, when interacting with an epistemic authority: I may seek rational presuppositions, justified assumptions, a useful conceptual toolkit, an epistemically good working hypothesis by which to pursue further inquiry, etc. I am sure that one could argue that all such goods are in some way tied to the truth, in the broadest sense, even if a true belief is not the immediate product of acquiring such goods. Again, however, my argument does not rest on the plausibility of suggesting that any of these goods fail to relate to the truth, or do not in any way derive their value from a relationship to the truth. Rather, I am simply aiming to show that we Nor do I think this kind of relationship can only be found in the pedagogical context. There will be cases in which experts of different fields interact and collaborate with each other, and each will have to act as an epistemic guide to the other, given certain limitations in each others’ expertise; there will also be cases in which a mere layman may have some kind of special role to play in an epistemic investigation, such as when a patient and medical practitioner interact in order to come to some conclusion about the appropriate course of action to take—here, again, it will not suffice to simply reduce the relationship to one in which the doctor supplies the patient with true beliefs: the two are more intricately involved in an epistemically collaborative project. I will say more about these cases in future chapters. 130 83 do an injustice to the complexity of the phenomenon in question if we merely reduce the role of the epistemic authority to that of a source of true beliefs simpliciter, and that the role of such individuals is especially complex in such a way that grounding such individuals’ normative power on considerations of true belief seems explanatorily deficient. I do not wish to belabour the point. The kinds of relationships that I have briefly illustrated above will be a recurring theme throughout the following sections and chapters. Here it suffices to have shown two things: one, that there are a plethora of epistemic goods that we can at least plausibly consider epistemic authorities as providing us with, some of which at least are arguably not immediately reducible to considerations of mere true belief;131 and, two, to have highlighted the way in which a certain epistemic practice, and an agent’s being involved in that practice, plays a very important role in our discussions of epistemic authority. As will soon become clear, this role is to play a central role in my own accounts of expertise and epistemic authority. I turn to these now. II.3. EXPERTISE AND EPISTEMIC AUTHORITY DEFINED In my discussion above, I have discussed the role of truth both in terms of the grounds for epistemic authority, and as a good acquired through epistemic authority. I have tried to be careful in keeping these two distinct, as I think that those who defend the accounts of epistemic authority presented in Chapter I often fail to do just that: they make the mistake of thinking that the grounds of epistemic authority simply are the goods that we can acquire from such agents, or, in other terms, the fact of that an agent is a reliable source of such goods. Kvanvig, at least, thinks that some of these cannot be so easily reduced to considerations of truth more broadly. See: Kvanvig, ‘Truth is not the Primary’, 294. 131 84 I think this is a mistake. As it turns out, what I understand to ground epistemic authority is not (at least in most cases) the same thing as the epistemic good that one seeks to acquire through interaction with such authority.132 By keeping these issues clearly distinct in our discussion, I think we are afforded with a much more detailed and accurate account of expertise and epistemic authority, as we shall see. By not limiting the defining quality of epistemic authority to the mere acquisition of true beliefs, we open up the concept of epistemic authority in such a way as to make sense of the wide variety of epistemic roles that such agents can play, and the kinds of epistemic goods that they can provide. II.3.A. DEFINITION OF EXPERTISE On my account, what makes someone a ‘cognitive’ expert of some domain d, in the relevant sense, is that they have the relevant skills, abilities, and know-how, to successfully partake in the epistemic practice that defines that domain of inquiry. The degree to which someone is a ‘cognitive’ expert in d is proportional to the degree to which they are able to successfully partake in the relevant epistemic practices that those involved in d involve themselves in. ‘Cognitive’ Expert: An agent S is a ‘cognitive’ expert in a domain of inquiry d iff she has the requisite skills, abilities, and know-how to successfully partake in the relevant kind of epistemic practice, EP, that is used, in d, to answer the kinds of questions that are pertinent in d. Note that my account is, given that we are focused on epistemic cases, limited to experts of the ‘cognitive’ or ‘intellectual’ variety (as Goldman has referred to them). It should be noted, however, that this definition can easily be modified to include other kinds of expertise: one simply This is not, however, to say such epistemic goods do not play a role in defining the appropriate kind of socialepistemic relationship as one of epistemic authority per se. 132 85 has to drop ‘epistemic’ from ‘epistemic practice’ and replace ‘answer the kinds of questions’ with the relevant kind of task for that kind of expertise. Given that I think the relevant weight of epistemic considerations can be generally accounted for by looking at the notion epistemic authority, rather than ‘expertise’ in this sense, I think that it is a boon that my definition of cognitive expertise is not far alienated from the more general notion of expertise. This allows us to easily recognise that ‘cognitive’ or ‘intellectual expertise’ is a sub-class of ‘expertise’. As should be clear, on my account, expertise is not defined in relational terms, as it is in many other accounts. More strikingly, my account does not directly mention an expert’s relation to the truth, or her acquisition of true beliefs (or avoidance of false beliefs), or any other specific epistemic good. This is because I think that to include such a component would be to confuse two distinct epistemic components: the component that makes someone an expert per se, and the reasons we have for interacting with such agents (i.e. what we wish to gain from such interactions), and why we hold certain kinds of social-epistemic relationships with them. As already stated, I think it has been a mistake in previous epistemic accounts of expertise to try combine these two issues. To illustrate this point further, it seems to me that there is: i. A way in which we refer to experts in the descriptive sense, i.e. as being experts in their field simpliciter. This is merely a descriptive claim about what someone is in non-relational terms: an expert is an expert, whether I care about their expertise, or not. ii. A way in which we refer to experts in the evaluative sense, i.e. as being the kinds of agents that we can rely on to provide us with a variety of epistemic goods etc. This not a merely descriptive claim, but a normatively loaded one. Furthermore, this refers to someone in relational terms. 86 By distinguishing between a basic sense of cognitive expertise, and epistemic authority proper, my intention is to keep these two ways of referring to other agents distinct. Many accounts of expertise, by being limited to questions of truth and the possession of true beliefs, immediately disqualify ‘experts’ in domains of inquiry that fail to produce a sufficient amount of true beliefs or any other epistemic good. I think it would go against our common language use to disqualify from our definition experts who do not possess the appropriate, relative to us, epistemic success. I think the advantage of my account of expertise is that it leaves room for considering such agents as experts, and (in conjunction with my account of epistemic authority) still leaves us with the possibility of interacting with them as the source of certain epistemic goods, even if the acquisition of true beliefs in the relevant domain isn’t one of them. If the concerns of II.1.c. are in any way motivating, then this should be a considerable reason in favour of my account. Of course, my definition of expertise is not completely without its connection to the normative. To be an expert, one must in some sense be successful in one’s participation in a certain practice. I do not, however, suggest that such success necessarily entail one’s acquisition of epistemic goods, and it certainly does not entail that such success puts one in a position to have a sufficiently greater success at attaining epistemic goods, such that one could be considered intellectually superior to others in the ways that other accounts of expertise entail. This success is merely one pertaining to one’s ability to partake in an established practice—it remains an entirely open question whether that practice itself is successful in the relevant epistemic sense. What I mean to pry apart here are the senses in which someone can be an expert, of, for example, astrology, and yet (assuming that astrology is not a successful epistemic practice) fails to attain any epistemic standing from doing so. As will be discussed in further detail below, and particularly in Chapter III, I argue that it is crucial to distinguish between these two elements: it matters that I can recognise someone as an expert in a certain field, and yet not engage with 87 them as my epistemic superior, and certainly not as an epistemic authority, due to the fact that I do not recognise their practice as epistemically desirable. Besides clarifying what it is not, what is there to say about the sense of ‘success’ that I am using? Unfortunately, the answer may not be as clean as one would hope. I think the measure of success will vary between different practices, and will be coloured by strictly social considerations, such as the dynamics of intra-disciplinary recognition and accolade, or criticism and rejection. It will also depend on how broadly practiced it is: it is completely possible to think of such a practice as being only instantiated in the actions of one individual, at which point it may be very difficult to verify the ‘success’ of such an individual’s ability to participate in that practice, given that the individual’s behaviour would, in a very direct and immediate sense, constitute the practice in question. In more general cases, to recognise someone as successfully engaging in a certain epistemic practice—on the assumption that I do not have the relevant skills to engage with it myself—will generally require that I rely on other experts in the domain to either directly or indirectly identify someone as successfully engaging with it. Credentials are intended as such identification. Nevertheless, I think there is a way in which we can make sense of the broader idea of success I have in mind: I merely mean that such individuals are successful relative the very standards and rules of the practice in question. Such rules may not always be explicitly provided, but, nevertheless, we can generally make sense of them with a proper and careful analysis of such practice.133 To be an expert then is to simply have the skills, abilities, and know-how to (generally) follow those rules of the game when attempting to answer the questions pertinent to that practice. A ‘good’ astrologer would be one who had the skills, abilities, and know-how to In the case that the rules really do seem to be difficult to pin down in any form whatsoever, we may have to conclude that we are not dealing with a practice of the relevant sort at all. 133 88 follow the practice correctly, as per the rules of practice, whether or not such practices leads to any ‘good’ product; a ‘bad’ astrologer would be one who does not have the requisite skills, abilities, and know-how to properly follow the rules of practice (someone who merely pretends to provide answers to the relevant questions, for example, without the proper training or practice). The ‘bad’ astrologer would fail to be an expert of astrology in the relevant sense, even if they were, arguable, no worse off epistemically speaking.134 Now, it may be objected that such a distinction leaves my account of what it is to be an expert too thin, in the very plausible sense that the ability to partake in a certain kind of epistemic practice does in no way guarantee that such an agent has the epistemic standing to be recognised or appraised as an epistemic expert in the sense that we usually mean when we call someone an expert. As many would point out, it is usually the case that when we call someone an expert, we are using the term in an epistemically evaluative sense: by referring to them as such, we are saying that they ‘know what they are talking about’, that when they say p (where p is a proposition about something relevant to d) p is probably something that we should believe to be true (at least more likely that our own judgement on the matter), etc. This is, I assume, true. My contention, however, is that this evaluative component is an element of our recognition of, and our acting towards someone as, an epistemic authority—an epistemic exemplar that has the ability to furnish us with certain kinds of epistemic goods that we seek. My point is that when we use the word ‘expert’ we do not always use this evaluatively loaded sense of the word: we sometimes refer to someone as an expert without identifying them as an authority of this ilk. This is to say that to be an ‘expert’ in one sense, is completely independent of the kinds of epistemic goods and epistemic relationships that we are concerned with when we choose to interact with At least in some sense—whether it is valuable to us or not, we would have to recognise that the ‘good’ astrologer had a certain kind of know-how that the ‘bad’ astrologer did not. 134 89 ‘experts’ in the sense of being epistemic authorities. I do not, however, intend to do damage to our common, everyday language use of the term ‘expert’—I suspect that what I have in mind when identifying the concept of epistemic authority is simply something that we certainly do identify with the word ‘expert’, in one sense. My point is simply that there are these two different concepts in play, and that clarifying the distinction can help us better understand how and when we are engaging with others in a specific social-epistemic relationship.135 Again, I hope the advantages of keeping these points distinct will become clearer as we progress. To reiterate ideas already stated above, I think the account here presented clarifies an important social-epistemological fact: many of the truths that such experts allow us to access are not truths that were discovered, in particular, by that individual expert’s investigations, but rather by the much broader epistemic project and epistemic practice of which that expert partakes. The beliefs that such individuals possess are not generally important to us because that specific individual holds them, but rather because they represent the results of a certain epistemic practice that we have identified as epistemically beneficial. I value the assertion of a scientist’s conviction, for example, because I have identified scientific practice as an epistemically apt method for inquiring about certain aspects of the world. What an account of expertise focused on true beliefs fails to properly illuminate is that the source of the expert’s supposed success is their ability to involve themselves with that practice, not the fact that they merely have the right beliefs at hand. Again, this is what I thought the Trivia Junky lacked, no matter how good he is as a source of true beliefs: he is not grounded in the very practice that affords humanity such epistemic success. My account of expertise is designed to take into consideration, and to capture that basic point. As we shall see, it is this component that can also make such an expert someone that we would wish I take this to be another advantage of my account: it suggests that, though the idea of an ‘epistemic authority’ may be somewhat alien to our everyday language use, the concept identified is in fact one that we generally make use of, even if we may not have a clear, analytic grasp of its conditions. 135 90 to engage with as an epistemic authority, so long as other conditions are met—not the results of such practice. A last illustration to underline the importance of this point: notice that we, as nonexperts, can be in possession of the results of a valuable epistemic practice, and this in no way makes us experts! II.3.B DEFINITION OF EPISTEMIC AUTHORITY On my account, to be an epistemic authority is relational: one is only an epistemic authority in relation to another agent. An epistemic authority is, first, much as a cognitive expert, someone who has the skills, abilities, and know-how, to partake in a certain kind of epistemic practice. Additionally, however, an epistemic authority is considered an authority relative to another agent who either does not have these requisite skills, abilities, and know-how, or at least has them to a sufficiently lesser extent. An epistemic authority is, furthermore, by means of her partaking in the relevant kind of epistemic practice, a source of some epistemic good that the non-authority would be epistemically better off in having. Lastly, an epistemic authority, by virtue of her skills, abilities, and know-how required to partake in the epistemic practice, has the power to Preemptively amend the non-authority’s epistemic behaviour in relation to the relevant domain of inquiry and epistemic practice (this can come in the form of providing reasons to behave a certain way, but I will also argue that it can simply take the form of commanding someone to behave a certain way when conducting the relevant kind of inquiry). Epistemic Authority: An agent S is an epistemic authority relative to another agent S* iff: 91 1. S has the requisite skills, abilities, and know-how to successfully partake in a certain kind of epistemic practice, EP, relevant to some domain of inquiry d. 2. S has sufficient access to the evidence and resources required for her to properly partake in the EP relevant to d. 3. S* does not have the skills, abilities, and know-how required to partake in the EP relevant to d, or at least has them to a (significantly) lesser extent than S. 4. S* recognises that d and the relevant EP is epistemically valuable for S*, i.e. it pertains to questions that would be epistemically valuable to have answers to. 5. S* recognises S as having a sufficient level of the requisite skills, abilities, and know-how to successfully partake in EP. 6. (From 4 and 5:) S* recognises that S is, by virtue of her ability to partake in the relevant EP, a potential source of some kind of epistemic good that S* would be epistemically better off in having. 7. S has the power to give S* a preemptive reason to behave in a certain epistemic fashion, perhaps by providing higher-order reasons and beliefs about methodological issues relevant to the kind of EP that is pertinent to d, or alternatively by commanding S* to behave in a certain sort of epistemic fashion. In short, S is able to authoritatively tell S* how she, S*, should epistemically behave in order to partake in the EP relevant to d. 92 (7) will obviously require some defence and will be discussed shortly. First, I would like to say more about the other components of epistemic authority. Part (2) of the definition is included simply because there will be cases in which someone is an expert in d, but due to not having access to the necessary resources and evidence, will not be able to make sufficient use of her skills, abilities, or know-how, and thus not able to properly produce the kinds of epistemic goods that such practice would generally furnish. Other accounts of expertise simply discount such agents as experts at all. This, however, strikes me as rather peculiar, given that it is merely a matter of luck that such agents’ skills are left without use, and that they are not provided with the opportunity to produce the appropriate fruits of their labour. Again, I think this shows a fault in confusing of the descriptive sense of expertise, on the one hand, and the evaluative on the other. Under my account, one can succeed in being an ‘expert’ in the first descriptive sense (since condition 2 is not part of my definition of expertise), but nevertheless fail to be an epistemic authority. Part (3) of the definition is simply there to capture the properly relational component of epistemic authority. I think it is simply an intuitive truth that such authority has to be considered in relational terms: to be authoritative, is to be an authority over someone else in the relevant sense. To be an expert is at least distinct from being an epistemic authority merely on this relational level: two experts may not be authorities in relation to each other, for example.136 Parts (4), (5), and (6) of the definition are intended to capture the fact that it is the very domain of inquiry, and the kind of epistemic practice of which the authority has the requisite ability to partake in, that both endows an authority with the relevant kind of epistemic privilege that the non-authority may lack, thus making them epistemically superior, and captures the Again, this point is not a new one: Constantin and Grundmann have already pointed out the same, as I have previously noted. 136 93 motivation for, and the value of, any other epistemic agent as recognising them as such, and interacting with them qua authorities. There are two points to be made here: i. Whereas other definitions here leave it up to the epistemic goal of acquiring true beliefs (in d) to do the work, I have here left the definition open to a variety of options: i.e. anything that could be considered as an epistemic good. The reasons for this should already be clear, as laid out in sections II.1 and II.2: I think it is quite obvious from empirical observation that the kinds of epistemic goods that we aim to acquire by interacting with our epistemic superiors, especially qua epistemic authorities, goes far beyond the mere acquisition of true beliefs, and, arguably, may sometimes not even include a concern for truth per se. Whether we value such goods ultimately for their bringing us closer to the truth, more broadly speaking, is a separate question: what I am concerned with here is making sense of what it is specifically that an epistemic authority herself possesses, and that a non-authority may recognise and seek, that provides her with her normative status. ii. I think that it makes far more sense to identify an epistemic authority’s involvement in an epistemic practice, one that is part of a shared collaborative project with a variety of other suitably skilled individuals, as the source of the epistemic goods with which they can provide us with, rather that limiting the account to a discussion of what epistemic goods the epistemic authority possesses. After all, I take it that a crucial element of what it is that makes such an epistemic authority’s testimony, for example, of such value, is that it gives us access to this domain of inquiry—a domain and practice that we find far 94 more valuable than any one individual’s epistemic position. The fact that the authority has access to these epistemic goods in the first place is simply reducible to the fact that they have the skills, ability, and know how to partake in the relevant kind of epistemic practice. This does not, of course, limit the authority’s role to a mere intermediary: the very fact that she practices and involves herself in that domain means that she is responsible for the proliferation of that epistemic practice, and the production of its results, and all the epistemic goods with which it is furnished. In a very real sense, even if only at a small level, her epistemic behaviour in tackling the questions relevant to that domain in part constitutes the very practice itself, in so much as the practice, as a social whole, is composed of the actions and behaviours of all those that are involved in it. Even if such a practice can be reduced to a series of propositional statements outlining a certain methodology for practice, it is still the fact of the matter that it is the individuals partaking in the practice, by following those rules, that define the practice and constitute it in their actions. Furthermore, it is the epistemic behaviour of such individuals that can change the ‘rules of the game’, so to speak: if a practice is a shared one, the individuals concerned can adjust their behaviour, even without explicit recognition of such a change, and in-turn modify the very practice that they are participants of. Parts (4), (5), and (6) of my definition also disqualify from consideration as epistemic authority any experts in domains of inquiry that furnish no epistemic good whatsoever. As I noted before, I think it important to leave open the possibility of there being epistemic authorities in domains where there are not a sufficient amount of truths produced to generally count such agents as 95 experts under other definitions of expertise, but it cannot be the case, under my definition, that there be epistemic authorities in domains that produce no epistemic good, i.e. that do not improve our epistemic situations in any way whatsoever. One may wonder whether it matters as to whether the non-authority can recognise certain fact about the epistemic authority in order for the latter to be an epistemic authority, or whether she has access to the sorts of facts that make the authority what she is. One may think that there is a difference between what makes someone an epistemic authority per se, recognised as so or not, and what makes one an epistemic authority for someone in a recognised sense. I, however, am fundamentally interested in this latter sense: the notion of epistemic authority I am discussing is intended to clarify the kinds of social-epistemic relationships we have with authorities, and these relationships are going to be ones in which the non-authority recognises the authority for what she is, and interacts with her accordingly. It is a relational notion, pregnant with normative significance: it would be odd to suggest that such a relationship could hold when both relata are completely ignorant of the facts that constitute the foundation of such relationships—namely: the asymmetry of their epistemic standings. Though it certainly makes sense to say that there are individuals that hold the right epistemic standing to be considering ‘epistemic authorities’ in the sense that they would have such normative power over us if we recognised them as such, this conditional element only underlines my point. All that being said, however, it is clear that part (7) of my definition is cut out to do a lot of the important work, given that I have stated my goal as being to capture a sense of what it is to be an epistemic authority that involves the kind of normative power proper that Zagzebski has suggested constitutes authority. I would here like to first explain what I mean by suggesting that an epistemic authority has the power to preemptively amend the non-authority’s epistemic behaviour in relation to a certain epistemic practice, and then to say a few words about why I 96 think such an element of epistemic authority is not as difficult to swallow as Zagzebski’s Preemption thesis, particularly in that it avoids the undesirable consequences of that thesis, and that it actually aligns with how we generally behave in other aspects of our lives when faced with authorities. As I have been discussing it, I understand a domain of inquiry to be governed by a certain kind of epistemic practice, whether or not such practice is explicitly codified, or simply implicitly assumed in various individuals’ behaviour. What this means is that there are certain normative structures in place when it comes to how one ought to inquire into the questions that pertain to d. A scientist follows the scientific method, a medical practitioner may follow guidelines of the Evidence Based Medicine movement, etc. Furthermore, I have claimed that epistemic authorities in such domains, by embodying certain kinds of epistemic behaviour, partly constitute the very practice itself. By being such authorities of a particular epistemic practice, their behaviour constitutes a part of the whole that is the epistemic practice itself, to put the point rather crudely. Now, my claim is that, if one has the skills, abilities, and, specifically, the know-how to partake in a certain practice, then one has the standing to tell others how to behave in order to partake in that practice. This will involve telling others how to behave when engaged in that practice, telling them what are good or bad reasons for taking any individual action in that practice (i.e. providing them with higher-order reasons relevant to engagement with that practice), telling others what the rules of that practice are (if explicitly available), telling others to mimic one’s own behaviour (to act and think like oneself), etc. Sometimes, however, (and I assume this is the most controversial component of my claim) it will simply involve chastising or encouraging a certain kind of epistemic behaviour, without providing an external reason to the non-authority besides the authority’s own reaction to the non-authority’s behaviour. 97 Before I defend this condition, let me make one thing clear: just because an epistemic authority has such power, it is not in any way meant to follow that such an authority would be behaving appropriately in using such power in any given situation, whatsoever. The task of Chapters III and IV will be to outline the conditions for an epistemically responsible manner in which both epistemic authority and non-authority can interact given their particular socialepistemic relationship. I will defend the view that, depending on a variety of conditions (some targeting the authority, some the non-authority, and some both) the specific kind of epistemic relationship such individuals can have will vary. In some more mundane cases, the authority will be required to withhold demanding the non-authority to behave a certain way, and will merely be in a position to explain the correct or incorrect manner in which the non-authority is behaving. In far more richer forms of the relationship between authority and non-authority, such as in a pedagogical context, the authority can rightly engage in the practice of exerting her normative power on a more regular basis (though this will always be limited by certain other considerations).137 The broader point here is that the relationship that holds between an epistemic authority and relative non-authority is always going to be dynamic: the needs of a nonauthority will change, the subject of inquiry will become more or less complex, etc. The point, however, is that, throughout all such dynamic change and difference in circumstance, it remains the case that an authority has the relevant kind of normative power no matter in what form it is exercised, or if it is not exercised at all. This disclaimer aside, let me continue: as a first pass, I think the claim that an authority has such power is intuitively plausible when it considering wide variety of analogous, nonepistemic practices. Consider a basic example in which S teaches S* to ride a bicycle, where S* As the King in Antoine de Saint-Exupéry’s Le Petit Prince would point out: an authoritative command must reasonable, if it is to be obeyed! 137 98 does not yet have the relevant skills, abilities, and know-how required to ride a bicycle, and where S is proficient in doing so, i.e. has a high-degree of the skills, abilities, and know-how required to ride a bicycle. Here, I do not think we would be fazed by the claim that S has the power to amend S*’s behaviour in a certain manner, in order to have S* successfully engage with the practice of riding a bicycle. Told to move her foot in such-and-such a way, S* will do so, even if she may have had prior thoughts, tendencies, inclinations, or intuitions about how to go about the act of riding a bicycle. In fact, if she is being rational, and recognises S as an authority on the matter, she will (at least attempt to) disregard all of these prior thoughts, tendencies, inclinations, etc., given that they are not part of the very practice that she is attempting to partake in. One is, after all, in trying to learn how to do something from one who can, engaging in an attempt emulate the latter’s behaviour. Secondly, I find no reason to think that the very same process should not apply in the epistemic realm as well. In fact, to argue that there is a fundamental difference in the epistemic case would be problematic, given that I think there are clear cases of epistemic relationships much like the ones in the bicycle example above. To teach a student how to think logically, scientifically, philosophically, spiritually, etc., is just to teach a skill set—the fact that these are cognitive skills rather than physical ones makes no great difference. If S* wishes to learn how to perform in a certain cognitive practice from S, and S has the requisite skills, abilities, and knowhow, then it seems intuitively plausible to say that S has the standing to tell S* precisely how to behave. This is, I would argue, precisely how we all learn new cognitive skills in the first place. Here I think it important to consider the plausibility of learning how to engage with a certain epistemic practice without going through the process of emulating the appropriate behaviour. It should be noted that I agree with Gilbert Ryle that there is a vast difference between being good at doing a certain thing, and simply knowing all the rules for how to do that 99 thing. I do not mention this to suggest that knowledge-how cannot in any sense be understood as a species of knowledge-that (as noted above, I wish my account to be somewhat neutral relative to this debate). However, I am assuming that it is necessary, at least with a majority of sufficiently complicated practices (and the ones we are considering here are just that) that one do more than simply learn the rules of such practices in order to be able to properly practice them. Explicit instructions and guidelines provided in propositional form are clearly a fundamentally important component to learning how to properly epistemically behave (relative to some practice and domain of inquiry), but it is nevertheless the case that we also have to practice that behaviour, to impersonate and emulate it, before we can be said to properly know-how to do so. Engaging in a certain epistemic practice is not merely a pattern of following certain rules, applying guidelines to certain problems, etc. If all problems were so easily solved, then we wouldn’t need anything like an epistemic authority, or epistemic exemplar, in our lives. Given that participating in such a practice, in the appropriate sense, requires this learning of new behaviour beyond merely learning a set of rules and methods in propositional form, it isn’t at all surprising to realise that such practice would require guidance, and the overriding hand of one who knows how. I furthermore see no reason to think there is something problematic about such a preemptive power, either. Now, I ought to point out here that though I am defending the preemptive component of my account by reference to a more pedagogical kind of relationship between epistemic authority and non-authority, I am not suggesting that all such relationships of the kind take this form. What I am trying to show is that for a non-authority to be in some sense partake with the relevant kind of epistemic practice, then he needs to at least be in some sense behaving appropriately. Does a non-authority have to have the desire to partake in such a fashion? No, of course not. But my argument is that if the non-authority does not do so, then he is not treating the epistemic 100 authority as an epistemic authority at all. Rather, in this case, he may be treating her merely as a reliable tool for accessing certain conclusions and results in the domain in question. However, as I’ve already noted, I think that Jäger’s insight is particularly on point here: this is not generally how we engage with epistemic authorities. Generally, we seek some access to the practice itself— we do not merely seek propositional utterances detached from the relevant contexts of method, or approach such figures with the very specific aim of being simply told what to believe. I ask my reader to forgive me for putting the point so metaphorically, but it seems to me that we seek out epistemic authorities as means to be ‘invited in’ to partake in the richness of their epistemic practice: we are not mere passive observers, but actively engaged listeners. Where I think Jäger makes a mistake is simply in suggesting that this element has to come in the form of a full-blown understanding—which is clearly not what we all seek, in all cases. Of course, there is a very pressing worry here: if I am suggesting that an epistemic authority does have some preemptive power, how does it manage to avoid the problems that are levelled at Zagzebski’s account, as discussed in Chapter I? If an epistemic authority can preemptively give me a reason to behave a certain way, to tell me that the appropriate way to behave in a certain epistemic practice is such-and-such, or to even go so far as to command that I behave in such-and-such a way, can this not conflict with other reasons I have to do otherwise? And, if so, can I not be left in the same epistemically defective position that we have seen Zagzebski’s account to lead us to? Perhaps surprisingly, my initial response to this may seem somewhat similar to Zagzebski’s own to her critics. However, I think the answer can actually be better defended on the grounds of my account, precisely because it locates the epistemic authority’s normative power in her skills, abilities, and know-how that allow her to participate in the relevant epistemic practice. My general answer is that it is completely inconsistent with treating someone as an 101 epistemic authority, in the way that I have laid out, that I could believe that I ought to behave differently, or that the methodology is incorrect, etc. In recognising someone as an epistemic authority, I am recognising her as someone who knows how to engage with a certain practice: more than that, she is part and parcel of that practice. To suggest that I have a reason to behave differently in the sense of acting as part of that epistemic practice, strikes me as incoherent if I am in the position of the non-authority as defined. Of course, I may disagree with the practice itself, I may even engage with a certain epistemic practice in a less-than-completely serious fashion. I could, for instance, think there is a better practice for answering the kinds of questions I have in mind, but, nevertheless, still engage with an epistemic authority of a different practice to try and make sense of that alternate perspective. This does not, however, provide me with a reason to do otherwise, so long as we are talking within the limits of that specific practice. We should only be concerned when an authority speaks beyond the limits of the very practice which we identify them as authorities of—but nothing about my account of epistemic authority suggests that they have dominion over such things! That all being said, there is a particular criticism of Jäger’s that I think does cause a certain problem for my view—or, I should say, it highlights the complexity of the phenomenon I have in question. Here I have in mind two kinds of disagreement, where the disagreement is centred on issues of epistemic practice, pertaining how one ought to epistemically behave, or what Goldman has referred to as ‘secondary questions’: i. A disagreement between two purported epistemic authorities of the same epistemic practice, in which a non-authority is interested in adjudicating between the opposing views. 102 ii. A disagreement between two individuals, where one has less, and one who has more, of the requisite skills, abilities, and know-how to partake in the epistemic practice in question. Let me consider case (ii.) first, as I think my answer here is rather straightforward, though my reader may find it unsatisfactory prior to venturing into Chapter III. For case (ii.) I simply state that the epistemic authority in this case (of which there is only one, given that we are talking about the relationship between the two individuals, not between the two individuals and a further third party) does indeed have the relevant normative power. She can give the (relative) nonauthority in this case a preemptive reason to believe such-and-such about the relevant methodology and practice—she is in a position to amend and command the other’s behaviour. However, as we will see in the next two chapters, I think there are clear ways in which one can do epistemic harm by exercising such power when not appropriate. To give a rough preview: I will suggest that an epistemic authority who exercises her authority while not being properly sensitive to the epistemic capabilities of the relevant non-authority does her an epistemic harm. This would be a case in which the non-authority’s epistemic autonomy was indeed undermined. I think case (i.) is far more complicated. Unlike Zagzebski, I think that disagreement between authorities (if I recognise both as such) can cause serious problems, and we should not be so hasty to simply disregard this issue. My suggestion is that such cases can tell us at least one of three things about the supposed authorities, or the epistemic practice, involved in this case. a. Hearing of such disagreement, I have reason to believe that, given that both authorities are supposedly partaking in the same epistemic practice, there is something wrong with that practice, given that two individuals, partaking in such a practice, can disagree on what the very rules of that practice are. 103 b. Hearing of such disagreement, I have reason to believe that the two authorities are, in fact, not partaking in the same epistemic practice, leaving me in a position to decide which of their respective epistemic practices is the most desirable to emulate. c. Hearing of such disagreement, I have reason to believe that though there is only one epistemic practice in question, one of the two individuals I have identified as an authority is in fact not an authority (or at least significantly less so, relative to the non-authority). How do we, as non-authorities, decide which of these is most appropriate response in any given situation? Unfortunately, I do not think there is a straightforward answer to this, as specific instances of this problems will vary widely in significant ways. I think this is simply an unfortunate consequence of living in the social-epistemic contexts that we do: there are simply times when we are at the mercy of our own ignorance, and lack of epistemic capabilities.138 That being said, there are certain things that can ameliorate the situation. I may be able to identify reasons to think that one supposed authority or the other has more of the requisite skills, abilities, and know-how, than the other: other individuals involved in the same practice may identify them as such, they may have received more accolades, whereas the other may have received certain criticism that signifies a lesser standing. One of the two supposed authorities may be able to explain to me the nature of the disagreement about epistemic practice, and confer to me an understanding of the disagreement that gives me reason to go one way or another. And, to echo Zagzebski, it may be the case that I simply trust one of the two supposed authorities more Goldman has analysed a variety of ways in which we might go about adjudicating between disagreeing experts, all of which are quite unsatisfactory. However, these may simply be the best that we have. See Goldman, ‘Experts: Which Ones’. Also see David Matheson, ‘Conflicting Experts and Dialectical Performance: Adjudicating Heuristics or the Layperson’, Argumentation 19 (2005), 145-158, for a further defence of some of Goldman’s suggested methods. 138 104 than the other.139 Of course, I don’t expect any of this to be entirely satisfying—as I have already suggested, I simply think this is simply a consequence of such disagreement that we have to accept. What is certainly clear is that it would be beneficial to discover further ways in which we could identify conditions in which one epistemic authority should be trusted over another: this is a task I begin to look at in Chapter IV. All the above considerations aside, on may still find it difficult to accept the possibility that one could be given a preemptive reason for what could be thought of as thinking a certain way, or to be told to go about one’s epistemic inquiry in a specific manner. To suggest that we could seems to fly in the face of our own epistemic autonomy—assuming that such autonomy in some sense requires that I own my epistemic behaviour, that the reasons that motive me or mine in some sense—and may seem psychologically implausible to some. But it is here again important that we note (and I am repeating myself) the limitation of this kind of normative force: it only pertains to a certain kind of epistemic practice, the one relevant to the domain in question. Furthermore, given the definition of epistemic authority, the normative force only applies in the case that S* has recognised d, and thus the relevant epistemic practice, as worthy of epistemic emulation. Again, I think this makes sense of everyday practice: we tend to think of certain kinds of inquiry as ones worth in some sense partaking in, and we reach out to the epistemic authorities of those domains to help us do so. We have already accepted a certain way of thinking as epistemically appropriate: if we come to think otherwise, then we cease to engage with the individuals in question as epistemic authorities. Though there are many who have argued otherwise (Paul Faulkner and Katherine Hawley, for example), I am personally rather pessimistic about the possibility of identifying a properly epistemic reason at the locus of such trust. Nonetheless, we clearly have practical reasons to trust individuals in such circumstances—and, as will become clearer in later chapters, I think these practical elements are crucial to understanding the epistemic relationship between epistemic authorities and non-authorities. 139 105 None of this in any way entails that an epistemic authority can tell you what the best way to think is, tout court. We have to remember that, by definition, we are always dealing with epistemic authorities as practitioners of a certain epistemic practice, within the limits of a certain domain of inquiry. For those offended by the possibility of having one’s epistemic autonomy undermined by the authoritative declaration of a practitioner of the scientific method, for example, one must remember that such a practitioner only has authority relative to the scientific domain (or whichever specific category within). The non-authority remains epistemically free to determine which kind of epistemic inquiry, or which domain of inquiry, he thinks is worthy of their epistemic endeavours. It is precisely because S* has already valued EP that she can be preemptively instructed to behave a certain way. The question of evaluating these domains, and methods of epistemic inquiry overall, is of course a difficult one, and one that I will not be able to answer here.140 Nevertheless, I do think that all the above still leaves us with needing to make better sense of how our relationship with someone such as an epistemic authority can properly coexist with our epistemic autonomy. Is it not epistemically irresponsible to put ourselves in such a relationship? Do we give up our autonomy when we do so? My answer to both of these questions is ‘no’, given that the parties involved appropriately recognise the relationship for what it is, and the appropriate limits of each relata’s epistemic role in such a relationship. Of course, these relationships can go awry, and often do. Understanding how such a relationship can take on a better or worse form will be the topic of Chapters III and IV. By looking at the relationship between epistemic authority and epistemic autonomy, and epistemic authority and epistemic humility, I will further elaborate on the various roles that such epistemic authorities can play. As Of course, S* may recognise S’s domain d, and the relevant kind of epistemic practice, EP, as the best form of epistemic inquiry. In this case it may be the case that S can tell S* what the ‘best’ way to think is, and give preemptive reasons to behave in a certain way. But note, this is again first a result of S*’s own recognition of d and EP as worthy of epistemic consideration and emulation. 140 106 we will see, an epistemic authority can fail to appropriately engage with an non-authority, in such a way as to do the latter epistemic harm; it is also the case that a non-authority can fail to recognise his own epistemic duties in such a relationship, and thus fail to act epistemically responsible. II.4 IN CONTRAST: MICHEL CROCE’S ACCOUNT OF EPISTEMIC AUTHORITY Now that I have presented my own account of epistemic authority, we are in a better position to take a brief look at Michel Croce’s view. I have chosen to do so here, and not in Chapter I, as there are some similarities between Croce’s view and my own (at least in motivation, if not in result), and thus I think it is easier to understand points of differentiation between our views in a more direct contrast. Perhaps most importantly, Croce and I agree upon a thesis that I have, in the preceding pages, spent much ink in defending: ‘the role of epistemic authority can go far beyond the mere transmission of true beliefs to the novice’.141 Much as in my own case, Croce seems to be motivated by consideration of Jäger’s response to Zagzebski. We agree in thinking there is no reason to limit a view of epistemic authority to one only concerned with what he refers to as the ‘authority of belief’, i.e. the idea that we can believe on authority.142 Looking to make sense of our relationships with epistemic authority in the variety of forms that they can take, Croce suggests a multi-tiered account of such authority. He states his view in terms of the following three definitions: Expert: A subject A is an expert in domain D (for a subject S) iff: (1) A has more accurate information than the majority of people do in D; (2) A possess expert-oriented abilities. Michel Croce, ‘Expert-Oriented Abilities vs. Novice-Oriented Abilities: An Alternative Account of Epistemic Authority’, Episteme 15:4 (2017), 479. 142 Ibid., 484. 141 107 Authority of belief (AofB): A subject A is an AofB in domain D for a subject S iff: (1) A is more conscientious than S—who considers her to be an EA—in D; (2) A possesses and makes use of sensitivity to S’s needs. Authority of understanding (AofU): A subject A is an AofU in domain D for a subject S iff: (1) A is more conscientious than S—who considers her to be an EA—in D; (2) A possesses and makes extensive use of novice-oriented abilities.143 By ‘expert-’ and ‘novice-oriented abilities’, Croce means the following: Expert-oriented abilities: virtues that allow an expert or authority to exploit their fund of knowledge to find and face new problems in their field of expertise (e.g. intellectual curiosity, intellectual creativity, open-mindedness, intellectual courage, firmness, autonomy, etc.) Novice-oriented abilities: virtues that allow an expert or authority to properly address a layperson’s epistemic dependency on them (e.g. sensitivity to S’s needs, intellectual generosity, intellectual empathy, sensitive to S’s resources . . . maieutic ability).144 Though we discuss these abilities in somewhat different terms, it is clear that there are some similarities between Croce’s view and my own. Though I deny his first condition for being an expert, his second condition aligns with my own view: an expert is someone who has the relevant skills, abilities, and know-how, to partake in a certain form of epistemic practice. I’ll leave it to my reader to decide whether it makes a crucial difference that Croce’s account is directly formulated in terms of intellectual virtues, where mine is not, but, for now, I assume that our conditions generally refer roughly to the same set of facts. What matters here, however, is Croce’s views on epistemic authority, not expertise. Here I think there are important differences. Two differences in particular are crucial: i.) Croce thinks that what he refers to as expert-oriented abilities are not necessary for epistemic authority (of both sorts: belief and understanding), whereas I do; and, ii.) novice-oriented abilities are necessary for 143 144 Ibid., 495. Ibid., 494. 108 epistemic authority, where I think they are not. However, it should be noted that, as we will see in Chapters III and IV, I think that various considerations that fall under the scope what Croce refers to as novice-oriented abilities certainly do play an important role in defining how we do choose our authorities, and how we go about interacting with them. The most important difference between our accounts, I think, is the first. On the condition that we think of Croce’s ‘expert-oriented abilities’ as akin to my own discussion an expert and authority’s relevant know-how, we disagree on whether or not such know-how is necessary for authority. As I have suggested, I think this element is crucial—Croce seems to think otherwise. To understand why Croce might think this is the case, we only need to pay attention to the fact that he accepts a broader range of individuals under the scope of his account of such authority than I do. He, for example, thinks that a grandmother, teaching her grandson to understand why a boat floats in water, or why fish can breathe in water, is an epistemic authority for that grandson.145 In another example, he describes his sister, who is not an ornithologist, as being an epistemic authority on whether or not the bird he indeterminably observes in the garden is a red cardinal, merely on the grounds that she has better eyesight than oneself.146 It is sufficient, on his account, then, that someone simply is in some broad sense epistemically better off than yourself, in order for them to be considered an epistemic authority relative to you. They are in some general, non-domain-specific manner, better situated to grasp some epistemic good, relative to yourself. I think this is incorrect. First, I think this does far too much damage to our concepts of expertise (not in his Croce’s sense, but the common evaluative sense) and epistemic authority. 145 146 Ibid., 488 and 493. Ibid., 482. 109 Croce’s account dilutes the notion of epistemic authority far too much: if he is right, then I am surrounded, all the time, by a variety of epistemic authority: such as every time I take my glasses off and ask someone with better eyesight to describe what they see. Or consider this example: I ask my dissertation advisor Prof. Sven Bernecker, via email, what the weather is like in Cologne, and he informs me that it is cold. Under Croce’s view, so long as Prof. Bernecker is sensitive to my needs, he has satisfied the conditions of (AofB)! All I wish to point out here is that I think there is a vast difference between the kind of relationship I have with Prof. Bernecker (when asking him about the weather, that is) as I do with a meteorologist (who specifically studies conditions in Germany). The kinds of individuals I have in mind when thinking of epistemic authority are quite far removed from the individuals in Croce’s fish, boat, and red cardinal examples, and I think are deservedly considered as different in kind, precisely because they have a specialised epistemic superiority: they are practitioners of a certain epistemic practice, constitutive of a domain of inquiry. As my discussion this chapter has made clear, this know-how is of fundamental import to my account—I refer my reader to the above discussion for consideration. Secondly, not only does this kind of account incorrectly track the phenomenon in question, I think it fails to describe it correctly. The kind of authority that Zagzebski and I have in mind has a strong normative force and is in fact constituted by such power. Broadening the account of such authority to include the cases that Croce does flies contrary to the goals of my project. Of course, it is up to my reader to decide whether this is a point in favour of my own view, or one against my project. The second point of contention between our accounts can be seen more clearly by considering Croce’s condition that an epistemic authority be sensitive to a novice’s epistemic needs, in conjunction with the fact that he thinks that ‘having a personal relationship with someone epistemically superior us in some domain is a necessary condition for her to be able to 110 acknowledge our dependence and needs’.147 Our disagreement here is fairly clear: I do not think it is a necessary condition that epistemic authority is sensitive to our specific needs, nor, in turn, that we have to have a personal relationship with them. However, I think I have to be careful to ensure that the nuances of this disagreement are clearly laid out: though these may not be necessary conditions on authority, on my account, these factors are nevertheless important elements in describing the dynamics of the relationships we may have with an epistemic authority, and how that relationship can change, blossom, or perhaps come to an end. First, to see why I don’t think it a necessary condition on authority that such an authority be sensitive to my epistemic needs, it suffices to consider cases in which the authority is simply not aware of my existence. Perhaps this may strike my reader, prima facie, as somewhat peculiar. But let us consider a few simple cases: I read a series of books about string theory, and recognise the author of these books as an epistemic authority on the matter; I watch a series of documentaries and films about the evolution of birds, and understand the writer and presenter of the show (an evolutionary biologist) as an epistemic authority on the matter; I observe a mathematician at work, and emulate his strategies and methods in solving mathematical problems, even though he is not aware that he is being so observed; and so forth. In all these cases, I think it perfectly correct to suggest that we are dealing with epistemic authorities, and to suggest that these cases should fall under the purview of my account. My reader may object here, however: how can the individuals in the cases above satisfy condition (7) of my account, viz. the condition that they have the power to give me a preemptive reason to behave in a certain epistemic fashion? On closer inspection, however, this is not at all that troubling: having such power, and even employing it, in no way requires that one is aware of those upon which the power acts. This can be illustrated quite simply. Consider the 147 Ibid., 479. 111 mathematician in the above example: now imagine that he thinks to himself aloud, chastising certain methods, and applauding other uses of reason. If it is the case that the issues he raises apply directly to my own chosen methods, or deployed reasons, I think it makes sense to think of his utterances as giving me a preemptive reason to adjust my own epistemic behaviour. The same can be said of similar cases, I think: imagine that I am reading a book by a famous physicist, in which she, step by step, describes the way she went about hypothesising, experimenting, compiling data, and finally drawing conclusions, establishing important scientific theories in the process. Such a description could surely be authoritative for me in the sense that I have been discussing above, giving me the right kind of authoritative reason to behave in a similar fashion when conducting my own empirical inquiry. I’ll leave my reader to think of other such examples. In case my reader is not convinced of this stronger point against Croce’s view, let me make a somewhat weaker one. For the sake of argument, let us assume that it is a necessary condition of epistemic authority that the authority be sensitive to the non-authority’s epistemic needs, as Croce suggests. I do not think it follows from this that it is necessary that the two have a personal relationship. Consider here the example of a pedagogue, in some domain of inquiry or other, who writes a series of educational books, or produces educational videos. Such an individual may arguably be thought of as being aware of, and sensitive to (at least to a degree), the needs of her audience,148 thus satisfying one of Croce’s conditions of epistemic authority ((AofB) or (AofU)). However, I do not think it the case that such an authority can be considered to have a ‘personal’ relationship with her audience, not in any meaningful sense. To suggest that such a personal element is a necessary condition to such authority I think is far too restrictive, and fails to account for crucially important instances of epistemic authority. Thus, Croce’s account Though even on this point I think the sensitivity may be far too broad and vague to do the work that Croce seems to be suggesting. 148 112 turns out to be, in one sense too inclusive, and, in another, not inclusive enough, ultimately failing to properly demarcate the phenomenon in question. Nevertheless, it should be noted that my view is not completely blind to what Croce is alluding to when he talks about novice-oriented abilities, and the importance of an authority’s possessing such abilities. Let us revisit condition (6) of my account of epistemic authority: 6. (From 4 and 5:) S* recognises that S is, by virtue of her ability to partake in the relevant EP, a potential source of some kind of epistemic good that S* would be epistemically better off in having. A condition on my account of authority is that the non-authority recognises the authority as the source of an epistemic good that she wants. It would follow from this that if a non-authority comes to recognise that an identified epistemic authority actually cannot provide these goods, because she lacks certain of what Croce refers to as novice-oriented abilities, then it is possible that this condition will come to be undermined, and, thus, that the expert will lose her status as epistemic authority. If the good that I seek from an epistemic authority is understanding, and it turns out that she is completely insensitive to this, instead telling me that I ought to believe p, q, r, etc., then I have reason to think that she does not satisfy condition (6), and thus no longer consider her as an authority, as such. Notice, however, that my account allows for more subtlety here. Croce thinks it necessary that an epistemic authority be sensitive to my needs, and that if she fails to be so sensitive, then she is simply not an authority.149 But this can’t be right: if I come to realise that an authority is not particularly sensitive to my specific needs in any given case (perhaps because she is in fact completely oblivious to my existence, or simply does not care), it does not follow that she is not an authority tout court. Perhaps I seek a variety of epistemic goods, some more complicated than Is say this because on Croce’s account, it seems like being an (AoB) is the ‘lowest’ kind of authority one can be—if one fails to be an (AoB), one fails to be an authority. This is because Croce suggests that an (AoB) is sensitive to our needs—all other kinds of epistemic authorities have are required to have this virtue, and others in addition. 149 113 others, and, in realising that an authority lacks the novice-oriented abilities that makes it possible for me to acquire some of these goods, I cease to treat her as a means to those goods. It does not follow from this that she is not an authority in regard to other goods! Here, then, is where I think my account has a particularly important advantage over Croce’s. Criticisms aside, however, Croce’s account brings to light important points in regard to the kinds of virtues and attitudes that we at least ideally want in our epistemic authorities. The task that he has started for us, specifically in identifying the category of ‘novice-oriented abilities’, is an important one, and one that I will do my best to contribute to in the next two chapters, but particularly in Chapter IV. This is the task of identifying the kinds of intellectual virtues, character traits, habits, and general attitudes that we ought to seek and attempt to foster both in epistemic authorities themselves, but also in ourselves as non-authorities. 114 III. EPISTEMIC AUTONOMY IN THE FACE OF AUTHORITY Identification, Reflective Endorsement, and Self-Reliance To behave in accord with another’s authoritative command—to think a certain way, to adopt or disqualify certain reasons over others, or to simply believe that p—stands opposed to the Kantian injunction to ‘think for oneself’.150 At least, this is how it may strike some of us at first glance. To react obediently to such authority, as some accounts of epistemic authority suggest, may intuitively strike us as actions contrary to what it is to be an autonomous person, more broadly, and contrary to what it is to be epistemically autonomous, more specifically. If another person were to directly bring about a change in one’s own epistemic stance, it would seem that one could only be considered epistemically responsible if it involved one’s having one’s own reason to adjust such a stance. To change one’s epistemic stance without such a component would suggest something lacking in the epistemic agent: they would be, to use Kant’s terminology, heteronomous. To be autonomous, very broadly put, is in some sense being one’s own person, directed by ‘considerations, desires, conditions, and characteristics that are not simply imposed externally upon one, but are part of what can somehow be considered one’s authentic self’.151 At a first pass, we could consider the notion of epistemic autonomy to concern one’s ability to reason, to adopt doxastic states, and to generally cognitively engage with the world in such a way that the relevant 150 151 Immanuel Kant, Critique of Judgment, translated by J. H. Bernard (New York: Hafner, 1951), 136. John Christman, ‘Autonomy in Moral and Political Philosophy’, Stanford Encyclopedia of Philosophy (Spring 2018). 115 cognitive states are in some sense considered authentically one’s own, and not merely directed by some external force. Under this broad conception, it becomes intuitively clear how a notion of epistemic authority incorporating some normative, preemptive force may seem contrary to such autonomy. To be guided by an authoritative utterance regarding one’s epistemic behaviour doesn’t immediately fit with the idea of being governed by one’s own reasons, or epistemic methods, i.e. thinking for oneself. This will be particularly obvious and troubling in cases in which we are told that the results of our own thinking ought not to be rationally in play at all! To put the point even more plainly, philosophical discussions of autonomy often refer to an agent’s ability to act on one’s own authority, as being driven by such authority, rather than behaving in accord with the authority of another, or behaving with or by no authority at all—e.g. arbitrarily, on a whim, motivated by emotion, etc. There is obviously much vagueness to such language, and the sense of authority in play may be quite different to the one that I have discussed here. However, I think the intuitive tension between considering one’s autonomous authority over one’s own actions, in contrast to that of another’s authority imposed upon one from the outside—whether in the practical, moral, political, or epistemic domain—illustrates the issue in question here nicely. Given that we may think our autonomy, and thus epistemic autonomy, is a fundamental value, one central to what it is to be authentic, responsible, agents more broadly speaking, we may want to conclude that there simply is no room for a strong sense of epistemic authority. We may strongly object to such a notion on the grounds that to ever defer to such authority (as described by Zagzebski’s or my own account) would be epistemically irresponsible, and contrary to what it is to be an epistemic agent. Of course, we could here simply resist the idea that such autonomy is important, or perhaps at least suggest that it doesn’t carry enough weight to cause any serious worries for an 116 account of epistemic authority. Here there are two things to be said: i.) the import of epistemic autonomy (whatever it is) is an open question, and I think it would at least be a benefit to any epistemic theory that countenances the existence of epistemic authorities to show how such authority can be compatible with such autonomy—this is the more modest task I set myself to accomplish in this chapter; ii) I think it is very difficult, as epistemologists, to deny that at least something like epistemic autonomy plays a quite fundamental role in our epistemic analyses and considerations. We care, I would argue, when formulating our analyses of knowledge, and deciding on the conditions of knowledge ascription, whether a given subject S in some sense ‘owns’ the relevant achievement. This is particularly apparent in discussing virtue epistemology, in which we often appeal to the question of whether a cognitive achievement can be properly accredited to an agent: here I think we are asking whether such achievements are the result of their own epistemic agency. Catherine Elgin, in defending virtue responsibilism, has gone so far as to argue that epistemic autonomy—as expressed through what she refers to as the ‘the epistemic imperative’ (an epistemically-focused modification of Kant’s Categorical Imperative)— underwrites all the epistemic virtues, and vindicates them qua virtues.152 I think there is something right about all of these considerations, and my reader’s personal philosophical commitments aside, I think it is fair to at least suggest that epistemic autonomy is something that many, philosophers and otherwise, certainly care about, and confer upon it a sufficiently weighty value. Here, I assume that there is something to this claim. Though it is not my primary goal, I hope that my analysis of intellectual autonomy, in relation to the phenomenon of epistemic authority, can further shed light on what constitutes such autonomy, and, in turn, perhaps convince my readers of its merit as a serious desideratum in our epistemological investigations. The perceived value of epistemic autonomy is certainly that which Zagzebski herself 152 Catherine Z. Elgin, ‘Epistemic Agency’, Theory and Research in Education 11:2 (2013), 135-152. 117 identifies as a primary reason for why a strong sense of epistemic authority has been generally neglected in epistemology. Zagzebski gives such weight to this apparent conflict that she seems to think that no defence or theorisation of epistemic authority can even be started until the relevant worries are put to rest. She argues, however, that it would be rash to think that there is in fact a conflict between the notions of autonomy and authority, and suggests that the above concerns are merely based on a confusion, viz. the failure to differentiate between the notions of autonomy and self-reliance. Though I agree with Zagzebski (and others)153 that thinking of autonomy merely in terms of self-reliance or self-sufficiency is wrongheaded, I do not think that the problem of epistemic authority vis-à-vis epistemic autonomy simply dissipates when we recognise such error. One can make sense of the tension without identifying the notion of epistemic autonomy with that of self-reliance. If we are to take seriously the idea that epistemic autonomy in some sense requires that my epistemic stance is, to put it loosely, ‘my own’, then even those who accept basic epistemic deference—in the sense of believing that p solely from testimony that p—and who recognise the ideal of epistemic self-reliance as a bad one, may nevertheless think that there is something amiss in cases of agents deferring in the stronger sense suggested by the notion of epistemic authority that I have been defending. At least, one can think of cases where this is clearly so. If such authority can provide me with a preemptive reason to adjust my epistemic behaviour in some way, there is at least a question of how such adjustment can be considered my own, and not heteronomous in the sense of being imposed upon me by some external power. What I will argue in this chapter and the next, however, is that this incongruity between autonomy and authority is not of conceptual necessity but is rather the result of irresponsible social-epistemic behaviour. This is to say that an Jesús Vega Encabo also argues that autonomy should not be considered as being equivalent to self-sufficiency: ‘Epistemic merit, autonomy, and testimony’, Theoria 23:1 (2008), 45-56. 153 118 authority can be irresponsibly deferred to in such a manner as to undermine one’s own epistemic position, via an erosion of one’s epistemic autonomy. It is also to say that such authority can be wielded inappropriately, in such a way as to do an epistemic harm to another agent (i.e. the nonauthority): our ability to properly exercise our own epistemic autonomy partly rests in the hands of those that have such authority over us. It also follows that we cannot be responsible in simply blindly deferring to those we identify as epistemic authorities across the board: if it is the case that I have the capability to engage with a certain question of epistemic import, and refuse to do so without good reason, then I have acted epistemically irresponsibly. I argue that I fail to be autonomous not when I adjust my epistemic behaviour in light of authoritative commands, but when I fail to supplement my engagement with such epistemic behaviour with an attempt to do what I can to involve myself in that practice in a more direct fashion—where such an attempt reaches some threshold of reasonable effort, given the limitations of my epistemic and practical situation. Some of what I argue for stands opposed to what Benjamin McMyler has referred to as ‘the epistemic right of deferral’.154 He suggests that in the case that we defer to (what he calls) a ‘theoretical authority’, we also defer the responsibility to respond to relevant epistemic challenges to any epistemic stance we may have as a result of such deference. If, for example, I believe that p in deferring to the epistemic work of an authority, McMyler thinks that any challenge that I receive for believing that p can be deferred to the relevant epistemic authority.155 I think this is wrong, or, at least, it is far too simple a picture. In actuality, the kinds of relationship we have with such authorities will vary greatly, in terms of the kinds of epistemic goods we seek to gain, as I have discussed in Chapter II, but also in terms of how much or how little of our own intellectual Benjamin McMyler, Testimony, Trust, and Authority (New York: Oxford University Press, 2011). 62. Arnon Keren, though disagreeing on the motivations for such a view, seems to defend a similar conclusion. See Keren, ‘Trust and Belief’. 154 155 119 abilities we are capable of bringing to the table, and in terms of how relevant our own epistemic involvement is to the practice in question. Not all cases of such appeal to authority are simple in the sense that they involve an authority who has the capacity for accessing a complicated epistemic practice, with which the non-authority has no means of access whatsoever. In many cases, the non-authority will have some capacity relevant to the practice; in the more interesting cases, the non-authority may even have special access to certain facts relevant to the inquiry that the authority does not, and in such cases the two will have to collaborate. I suggest that, though the non-authority has a right to defer on matters in which here has no such capacity to involve himself in, he is nonetheless responsible for responding to other epistemic challenges related to elements of the epistemic practice with which he is capable of engaging with in a more direct fashion. To motivate this last point, consider, as a preliminary, the relationship between a doctor and a patient, in which the two are attempting to come to a decision on the best course of treatment for a life-threatening illness. Though the doctor in this case may be the authority when it comes to medical issues, certain components of such decision making can involve (and in many cases even require) the engagement of the patient: considerations of quality of life, personal preferences between varieties of treatment, first-personal access to one’s own suffering, descriptions of an illness’ effect on daily life, etc. In such cases, an authority and/or a nonauthority that fail to recognise both parties’ roles in the relevant investigation hinder the ability for such an investigation to proffer the appropriate epistemic fruits. Such a failure, I argue, is epistemically blameworthy in that it causes epistemic harm to an individual with which one is involved in a social-epistemic relationship with. This will be fairly clear in cases in which the epistemic authority and non-authority are joined in an epistemic task for which they are both partly responsible for. However, I will further argue that a non-authority can behave 120 epistemically irresponsible in this fashion even when his general epistemic position is not a crucial component of the relevant epistemic inquiry. As I understand it, there are two distinct problems to consider when evaluating our ability to be epistemically autonomous while deferring to an epistemic authority. 1. There is the broader question of how we can be considered to be autonomous at all, when the very epistemic practices that drive the authority’s conclusions are beyond our ken. This is to say: how can we authentically identify with a certain course of epistemic behaviour, if we are alienated from its justificatory basis in the appropriate epistemic practice? 2. There is the narrower question of whether a non-authority is failing to be autonomous, and thus shirking his epistemic responsibility, when he fails to engage with certain epistemic practices that he does have, at least partly, the means to engage with. I argue that first question can be dealt with by suggesting that a non-authority is epistemically autonomous, all things being equal, so long as he in some sense reflectively endorses the broad epistemic principles that govern the relevant epistemic practice in question. It should be noted, however, that I will also argue that such reflective endorsement may not necessarily be purely epistemic, or intellectually motivated, in nature: it is often the case that a non-authority will endorse such principles, and the practices they constitute, on the grounds of various pragmatic concerns—certain epistemic principles may bring about desired practical consequences. Such pragmatic concerns can put a limit on how much weight certain epistemic ends may play. To illustrate the second question, one need only think of cases in which I completely defer to an epistemic authority, and the relevant epistemic practice, even though I have the means to 121 at least partly engage with the relevant epistemic task myself. The aim here is to make sense of the question of whether or not I have done something epistemically wrong if I fail to ‘play my part’, so-to-speak. Consider, for example, a case in which I have the capacity to understand (at least in a rough sense) why a doctor has chosen to prescribe a certain course of treatment for me, but I choose not to ask any questions about the basis for such a decision. Would I, first, be the appropriate target of an epistemic reprimand or criticism? And, if so, on what grounds? If it is the case that I have failed to ‘think for myself’, i.e. failed to be epistemically autonomous, what is the nature of this failing precisely? My argument here is that we do indeed have an epistemic responsibility to involve ourselves in such epistemic practices to the best of our ability, where our ‘best’ may be tampered by limits in cognitive capacity, intellectual skill, prior knowledge and training, but also by pragmatic constraints. Prima facie, this may strike one as too strict, and far too arduous a condition to place on responsible behaviour. However, as we will see, given the reality of our social-epistemic situations, my view endorses epistemic behaviour quite in line with our common intuitions about what is, and is not, responsible. If anything, my conclusion is quite weak, and I think this is a good thing: it shows how we can be epistemically autonomous in our dealings with authority without establishing onerous conditions for doing so. It should be noted that not all philosophers are as quick to accept that there is a problem here at all. John Hardwig, for example, in light of various observations about the practice of scientific investigation, responds negatively to Kant’s enlightenment plea, and has concluded that it ‘is sometimes irrational to think for oneself’.156 However, my argument here is completely compatible with such a view: the question is how to make sense of one’s being autonomous in such situations, in spite of the fact that it would be irrational to attempt to (completely) engage with an epistemic practice on one’s own terms. It would be irrational to insist on being entirely 156 Hardwig, ‘Epistemic Dependence’, 343. 122 self-reliant in this situation, yes, but I will argue that the notion of epistemic autonomy is not exhausted by considerations of self-sufficiency or self-reliance. Hardwig’s point, I take it, is that it is important for social epistemology to provide us with some way of identifying the conditions that determine when it is, and when it is not, appropriate to engage in one’s own thinking per se— my discussion here is aimed to at least clarify the limits of such activity in the case of our dealing with an epistemic authority. I begin in III.1. by very briefly surveying various accounts and notions of autonomy as broadly discussed in moral and political philosophy. In III.2 I isolate at least one plausible sense of epistemic autonomy applicable to the investigation at hand. My goal will not be to argue for a single, complete account of autonomy per se, but to rather identify the sense of autonomy that is in play when we consider the threat of epistemic authority vis-à-vis epistemic autonomy. Section III.3. will argue that we can make sense of the minimal level of epistemic autonomy a nonauthority has, in the relevant social-epistemic relationship with an epistemic authority, by reference to two components: the notion of reflective endorsement, roughly as characterised by Catherine Elgin,157 and the idea that a non-authority bears the brunt of the responsibility in bringing into existence the social-epistemic relationship that holds between himself and the epistemic authority. In III.4. I consider the variety of ways in which a non-authority may or may not be able to partake in the relevant epistemic practice when engaging with an epistemic authority. On one extreme will be cases in which the non-authority can have no plausible, responsible epistemic involvement in the relevant epistemic question or investigation; on the other extreme will be cases in which the non-authority can play a very substantial role indeed. Having identified these various cases, I move on, in III.5., to discuss what constitutes responsible, and epistemically irresponsible, behaviour on the part of the non-authority given such varieties of 157 Elgin, ‘Epistemic Agency’. 123 possible intellectual involvement. My argument will be that, we are indeed responsible for engaging where we can, even if this merely means taking steps to acquire some crude, rudimentary understanding of the epistemic bases for adopting the epistemic behaviour that an authority tells us to, or simply sharing with the authority any evidence of our own we have accrued the bears on the questions pursued. Lastly, in III.6., I briefly discuss the ways in which an epistemic authority can cause damage to the social-epistemic relationship between authority and non-authority, by failing to be properly sensitive to, and respectful of, the latter’s epistemic autonomy. Since this last point will be heavily tied into the topic of Chapter IV, viz. epistemic humility, the discussion here will primarily act as a precursor of what is to come. In what follows, I will take it for granted that my reader accepts that there is at least a prima facie weirdness to the kinds of epistemic behaviour that certain accounts of epistemic authority, which include a strong normative power, endorse. Even if my reader does not have this intuition, however, I see no reason why he or she would begrudge the attempt to make sense of precisely how or why a non-authority could be considered epistemically autonomous (or not) in such relationships. Understanding the relevance of epistemic autonomy in this particular social-epistemic context will at least move us closer to a broader and fuller account of epistemic autonomy in general. III.1. A BRIEF LOOK AT PERSONAL AUTONOMY As already stated, I have no intention of providing and defending an account of autonomy, nor even a detailed and complete account of epistemic autonomy more specifically. My aim is instead to try to illuminate the various senses in which, in the specific circumstance that a non-authority interacts with an epistemic authority, he may fail to be epistemically or 124 intellectually autonomous. Here then, I will very briefly discuss a few key philosophical accounts of moral and personal autonomy. I will not discuss criticisms raised against such views, nor will I discuss the variety of subtle differences between various instantiations of the broader categories of views I present. My point is only to capture the key components of such views, in hopes that they can help us better understand the epistemic sense of autonomy relevant to the discussion of epistemic authority specifically, but also to social epistemology more broadly. Though Kant did not invent the notion, I think it quite fair and apt to begin such a discussion with him, given that our contemporary conception of autonomy owes much to his pioneering work. Famously, Kant discusses the moral or practical autonomy of a rational agent whose actions are guided and determined by her own self-imposed law. For Kant, then, autonomy is constituted by one’s acting in accord with a rule, or set of rules, that one is responsible for establishing (or can envision oneself as establishing), and not behaving in accord with a rule that is imposed on one ‘from the outside’, so-to-speak, nor by any other non-rational motivation, such as emotion or appetite. Contrary to what more recent notions of autonomy entail, however, Kant’s view is not strictly individualistic—the very maxims upon which one ought to base one’s actions are those that one can universalise, and that one can imagine as establishing as a creator of laws in the Kingdom of Ends, a community whose members are all rational agents, treated as ends-in-themselves. Central to this historically significant notion of autonomy, then, is the idea that one can endorse, as a rational being, the rational bases for one’s actions, qua universalizable and non-individualistically acceptable rational bases, but also that one can identify oneself, again qua rational being, as the source of such rational motivations, as the legislator of the relevant rules and laws. Hierarchical accounts of personal autonomy also revolve on this notion of identification. But rather than suggest that an autonomous agent identifies with the rules governing her actions, 125 such accounts cash out the notion by reference to a harmony between first-order and higherorder desires. Harry Frankfurt, who I take as my example of such accounts, suggests that an action is autonomous so long as the first-order desires that motivate the action are endorsed by the agent’s second-order desires.158 To take a trivial example: my action of drinking a third cup of coffee this morning would count as autonomous so long as the desire to drink a third cup of coffee was endorsed by my second-order desire, i.e. that I desired to be desire to drink another cup of coffee. This second-order desire, hierarchical accounts of autonomy suggest, more accurately reflect ‘who I am’ as a person. Thus, any first-order desire sanctioned by such higherorder desires will better serve as an expression of my personhood or identity. The intuitive plausibility of this view can be captured by considering the non-autonomous coffee addict: he has a very strong desire to consume another cup of coffee, but has a higher-order desire incompatible with that: he wishes he were not so physically dependent on caffeine, and in fact desires that he be the kind of person that did not drink so much coffee. Again, as with a Kantian conception of autonomy, we find the notion of identification playing a central role in an account of autonomy: an autonomous act is one driven by a motivation that I have identified with. Here, however, the notion takes on a highly individualistic form, given that there is no requirement that my higherorder desires reflect any objective or universalised value or goal. On reasons-responsive accounts of personal autonomy, an agent can only be considered as being autonomous in the case that the motivations that lead the agent to act are themselves responsive to ‘a sufficiently wide range of reasons for and against behaving as she does’.159 On such an account, someone who does not have the capacity to understand why she does what she does cannot be considered to be acting autonomously—she must have the capacity to appreciate Harry Frankfurt, The Importance of What We Care About (Cambridge: Cambridge University Press, 1988). Sarah Buss and Andrea Westlund, ‘Personal Autonomy’, The Stanford Encyclopedia of Philosophy (Spring 2018), ed. Edward N. Zalta. 158 159 126 the reasons that she has, or the lack thereof.160 What matters for such an account then, is not necessarily that one can identify with ones motivations in the more personal sense suggested by the hierarchical accounts of autonomy above (or coherentist below), but rather that one is able to recognise the rational force that gives birth to such motivation through the act of responding to reasons present for and against such a motivation. We could consider an agent in this case reflectively endorsing a motivation via a process of engaging with reasons for and against it. Coherentist accounts of personal autonomy suggest that an action is autonomous so long as it is motivated by a certain preference that coheres with an agent’s ‘true’ character. Such character is to be understood in terms of a coherent set of beliefs, desires, preferences, etc., that has been endorsed through a process of critical, self-reflection.161 Here we find the ideas of identification and reflective endorsement in play: such views suggest that a preference that coheres with my overall set of beliefs and preferences is one that properly represents my self, and is thus something I can identify with (though this may not require a conscious, reflective act of identification); such views also suggest that this set of beliefs and preferences is constituted as it is due to a process of reflection and evaluation, where we could consider any surviving belief or preferences as being endorsed by that process of reflection. The notion of self-identification here is admittedly far more robust than suggested by the types of hierarchical accounts mentioned above, as it comes as the result of serious reflection, and a process of critical evaluation, rather than a mere synchrony between lower- and higher-order states. Similarly, others have defended what has been referred to as a responsiveness-to-reasoning accounts of personal autonomy.162 Here, an act is autonomous so long as it is brought about by a motive that has been evaluated by the acting agent in relation to all of the agent’s other beliefs Buss and Westlund, ‘Personal Autonomy’. Jane Dryden, ‘Autonomy’, The Internet Encyclopedia of Philosophy, accessed November 2, 2018, https://www.iep. utm.edu/autonomy/. 162 Buss and Westlund, ‘Personal Autonomy’. 160 161 127 and desires. These motives are thus the kinds of things that can be adjusted by such evaluative reasoning, and, given that such evaluation happens on the basis of the total set of beliefs and desires that an agent has—a set of beliefs and desires properly central to accounting for ‘who’ the agent is—I take it that they can be considered to properly represent the agent. These accounts differ to the coherentist accounts in they do not think it is sufficient that a motivating force for the action in question merely coheres with one’s other beliefs and desires, but is rather evaluated in relation to them. Of course, there are many variations on the broader types of accounts canvased above, and many more accounts of personal autonomy beyond that. However, I think that the brief survey above suffices to show that there are two components quite common to the general discussion of personal autonomy, both of which, unsurprisingly, seem to play a central role in Kant’s notion of moral autonomy. I will refer to these as: identification, and reflective endorsement. The first is, roughly, the idea that one must be able to identify with the motivations for a certain course of action; the second is the idea that one’s motivations must be able to survive a process of self-reflection, or rational evaluation. Both suggest that one in some sense ‘owns’ or ‘governs’ over the relevant motivation—both identification and reflective endorsement are a means to make them ‘my own’ in the relevant sense. Now, of course, I do not mean to claim that there is a clear, singular sense of either of these notions that plays an identical role in each of the accounts presented, especially given that I have only discussed these types of views in broad strokes. The two notions I have made use of here are merely place-holders, and are certainly vague and of little use to the current discussion as they stand. However, by highlighting these two broader ideas, I hope to provide clarity to any following attempt to map the notion of autonomy onto epistemic considerations. 128 III.2. CONSIDERATIONS ON EPISTEMIC AUTONOMY Even the briefest of surveys of the literature on personal autonomy, as provided above, are enough to bring to light a peculiarity about the intuitive notion of epistemic autonomy one might assume to be at work in many epistemological discussions.163 This peculiarity is that epistemic autonomy is sometimes equated with the notion of epistemic self-reliance.164 As already noted above, it is this assumed synonymy that likely drives many knee-jerk, negative reactions to the notion of epistemic authority: to defer to such authority would be to fail to meet the ideal of ‘doing it for oneself’, to be self-sufficient and self-reliant epistemic agents. The Kantian injunction to ‘think for oneself’ can, at least on its surface, be easily equated with the corresponding negative injunction: ‘don’t let others do the thinking for you!’. This identification of epistemic autonomy with epistemic self-reliance is peculiar in the sense that neither the notion of identification, nor reflective endorsement, as mentioned above, necessarily include any sense of self-reliance. In fact, I would go so far as to suggest that there is no conceptual connection between these three notions at all. One may thus wonder how the epistemic notion of autonomy is in any way related to the personal or moral sense—is it a concept related by name only?165 The view that epistemic autonomy is constituted by epistemic self-reliance would be, roughly, this: Epistemic Autonomy as Epistemic Self-Reliance: An epistemic agent S is epistemically autonomous, in relation to a certain doxastic attitude or epistemic position, so long as she relies solely on her own cognitive faculties when coming to hold that attitude or position. That is, so long as we assume that there ought to be some kind of similarity between the notion of personal autonomy, and that of epistemic autonomy. 164 See, for instance: Elizabeth Fricker, ‘Testimony and Epistemic Autonomy’, The Epistemology of Testimony, ed. Jennifer Lackey and Ernest Sosa (New York: Oxford University Press, 2006), 242. 165 This is, of course, a distinct possibility: perhaps when epistemologists discuss epistemic autonomy, they are merely talking about a specifically epistemic notion that does not map on to other notions referred to by the same word. However, as I hope my discussion will show, I think there is plenty of room for a notion of epistemic autonomy that better relates to autonomy of the personal and moral kind. 163 129 Why would epistemic (or intellectual) autonomy be thought of in terms of epistemic self-reliance, then? Such a position may be attractive to those who are particularly sceptical of the epistemic competency of other agents. Upon reflection, we find that our own reasoning, justifications, appeals to evidence, etc., are arguably more easily accessible as targets of critical evaluation than those of other agents. In the instance that we rely on another agent for epistemic goods, these agents’ epistemic doings become key links in a chain that connects our own epistemic positions with how the world actually is; the fact that we cannot directly evaluate another agent’s actions in epistemic space to the same degree as we can our own presents us with a reasonable sense of uncertainty and risk. The more often we have to rely on such deference to others, and the more acts of such deference are added to the chain that connects our own epistemic positions with the fact of the matter, the more reason we have to think that some error could have been made, or that some act of reasoning could have been led astray, by bias or incompetence. The more we rely on others, it would seem, the more we have reason to think our connection to the fact of the matter is itself faulty. Elizabeth Fricker has put the issue, in terms of knowledge gained via testimony, in the following way: ‘this knowledge from trust in testimony is knowledge at second hand (or third, or fourth . . .), and as such my epistemic position vis-à-vis what I know is in at least one respect inferior to when I know first hand’.166 The suggestion is that first-hand knowledge is far less risky than second-, or third-, etc. Fricker suggests that a superior cognitive being, who was not limited in the same ways that we are, ‘could be epistemically autonomous in a way that no one of us, with our limited research time and processing capabilities, is able to be’.167 Fricker thus suggests that this notion of epistemic autonomy as self-reliance is an ideal, one that we should strive to get 166 167 Fricker, ‘Testimony and Epistemic Autonomy’, 240. Ibid., 240. 130 as close to as possible. Nevertheless, Fricker is quick to point out, as many others have, that aiming to reach this ideal of epistemic autonomy (as epistemic self-reliance) at the cost of all other epistemic considerations would be highly irrational. As she puts it: a ‘refusal to bow to others’ judgement or advice even when they are clearly relatively expert, is pig-headed irrationality, not epistemic virtue or strength’.168 The point has been echoed in much of the literature on social epistemology. John Hardwig has gone so far as to say that, much of the time, thinking for yourself is simply irrational: to not trust in others in many cases would be to deny oneself access to the best evidence available to us to answer a wide range of questions, and thus to the broad range of truths that such evidence would point us to. If ‘I were to pursue epistemic autonomy [as epistemic self-reliance] across the board, I would succeed only in holding relatively uninformed, unreliable, crude, untested, and therefore irrational beliefs’.169 Hardwig goes so far as to suggest that, in cases in which a non-authority engages with an epistemic authority, the non-authority ought to passively and uncritically accept what the authority tells them to believe.170 The necessity for such trust in our social-epistemic relationships, claims Hardwig, is simply at odds with our valuing of intellectual or epistemic autonomy so highly. This point in regard to the tension between the supposed ideal of epistemic autonomy, understood as epistemic self-reliance, and the undeniable facts about our social-epistemic lives, is one that I will not belabour, as I take the point to have been made forcefully and clearly through the literature. Here I simply assume that any worthwhile social epistemological theory, whether pertaining to testimony broadly, or epistemic authority specifically, must accept that any extreme adherence to the idea of self-reliance is irrational, and that the best epistemic practice for Ibid., 239. Hardwig, ‘Epistemic Dependence’, 341. 170 Ibid., 343. 168 169 131 creatures such as ourselves involves at least a suspension of the pursuit of such an ideal. How far we go in denying the relevance of this ideal of self-reliance, however, is a more difficult question to answer. Whereas Hardwig suggests that it is perfectly rational to be entirely passive and uncritical in certain social-epistemic relationships, suggesting that we can trust blindly in the epistemic work of other agents, Fricker maintains that we should hold on to as much of our self-reliance as possible, all others things being equal: ‘[i]t is crucial for the maintenance of epistemic self-governance that our trust in the words of others is given not blindly and universally, but discriminatingly’.171 This can be done because we are all ‘experts’ in one area: that of folk psychology.172 The claim here is that we all have the ability to evaluate the trustworthiness and sincerity of any human knower—so, whereas cases of expert deference may make it impossible for us to evaluate the nature of the appropriate evidence relevant to the question at hand, we can nevertheless evaluate other knowers to some degree. Our ability to evaluate the trustworthiness of other epistemic agents aside, however, it is clear that the case of epistemic authority presents us with a particular challenge. It is one thing to evaluate the trustworthiness of George, a man on the street who provides us with instructions on how to reach the local Post Office, knowing that we at least have the means to evaluate him in a more epistemically thorough fashion (by asking for his credentials, testing his knowledge, verifying for ourselves, etc.); it is another thing to evaluate the trustworthiness of a supposed expert of a domain that we have no means of evaluating, one in which we lack the relevant cognitive processing skills, time, knowledge, etc. to access. As Fricker writes: ‘where my reliance on others depends on an expertise they possess relative to me which is more deep-seated, and I lack the ability to check up for myself if it seems worth it, the existential supposition and 171 172 Fricker, ‘Testimony and Epistemic Autonomy’, 239. Ibid., 242. 132 dependence on others’ epistemic skills and truthfulness is more troubling’.173 Such reliance is troubling in three ways: one, because adjusting our epistemic positions in light of such deference multiplies opportunities for deceit, honest error, etc.; second, because with such epistemic dependence also comes practical dependence; and, third, because such epistemic dependence, ‘while it extends one’s knowledge base so enormously, also lessens one’s ability rationally to police one’s belief system from falsity’.174 I wish to draw attention to this third point mentioned by Fricker—that the kind of deference in question, and epistemic reliance on others in general, ‘lessens one’s ability rationally to police one’s belief system from falsity’. I think this is right, but it also suggests to me that what is at stake here is not self-reliance per se, but rather something that could be considered a potential consequence of being self-reliant: namely the ability to be able to critically evaluate and reflectively endorse a component of one’s own epistemic position. I think it is this that cuts precisely at the heart of the matter: even the threat of introducing more error and deceit (by adding testimony into the chain that connects our own epistemic positions with the fact of the matter) is merely a component of this broader problem of not being able to adopt the proper position of epistemic evaluation from a detached, overseeing standpoint. The fear that those attached to the notion of epistemic self-reliance experience is the fear of having blind spots in one’s epistemic economy, areas in which some of the constituents of one’s epistemic position present themselves as inaccessible black boxes that we merely have to hope and trust house epistemically proper mechanisms and rational or evidentiary support. The fear is that I have beliefs, or other doxastic states, that play a role in constituting my total epistemic position that I do not directly have means for reflectively evaluating and calling ‘my own’ in the same way that I do beliefs that I 173 174 Ibid. Ibid. Emphasis my own. 133 acquire first-hand. Importantly, notice that this fear has no clear, necessary conceptual connection to the notion of self-reliance. Self-reliance is not a necessary condition on one’s being able to reflectively evaluate and endorse a given epistemic attitude or behaviour. Self-reliance is merely one method for alleviating and minimising this fear. What in fact matters is that I able to rationally evaluate and endorse the epistemic methods, doxastic attitudes, and other cognitive components, that I consider and identify with as constituting my own epistemic position. Self-reliance is one means of doing this: if I put in the cognitive work to discover whether or not it is the case that p, rather than relying on someone else to do so, it is generally the case that I then have first-person access to the reasons, methods, arguments, etc., that lead me to include the consequent doxastic attitude (believing that p, for example) as part of my epistemic make-up. With this first-person access I can then reflectively evaluate, and re-evaluate, my belief that p, reflectively endorsing it when I am generally satisfied with my findings. Epistemic deference threatens our ability to do just this, not because we didn’t do the relevant epistemic work, but because we don’t have access to the possibility of reflectively evaluating and endorsing that work. My claim then, is that what matters to us when we consider epistemic autonomy (at least in the sense of the autonomy we fret about when considering epistemic deference and reliance) is that we are able to rationally evaluate and reflectively endorse a certain epistemic position/doxastic attitude/etc., and that such an ability allows us to rationally identify with that epistemic position, making it, to speak in rather metaphorical terms, ‘our own’, as part of our own total individual epistemic position and stance, or viewpoint. Roughly, then, the view would be this: Epistemic Autonomy as Reflective Endorsement and Identification: An epistemic agent S is epistemically autonomous, in relation to a certain doxastic attitude or epistemic position, so long as she is able to critically evaluate, and reflectively endorse, that attitude or position, and thus identify it as partly constituting her own epistemic viewpoint more broadly. 134 Such a notion of epistemic autonomy fits comfortably with what other epistemologists have said about such autonomy elsewhere. Catherine Elgin, for example, suggests something very similar in her own account of epistemic autonomy,175 which she states in the following way: In making and reflectively endorsing commitments, the agent exercises her autonomy. She considers herself justifiably bound by those commitments because she believes that being bound by those commitments will promote her epistemic ends.176 Such endorsement is contrasted by Elgin to other epistemic processes or dispositions that may be heteronomous to the subject, such that subject ‘cannot reflectively endorse or reflectively repudiate his beliefs or dispositions or processes that give rise to them, he is a victim of circumstance’.177 Unlike Elgin, I have not included any reference to ends in my own account, but I assume that any kind of reflective endorsement will be made generally, if not always, relative to some end or another—I certainly see no prima facie reason to deny this. I do not however think we should limit such an account to merely considering epistemic ends, reasons for which will become clearer in III.3. Though she does not provide a clear account of epistemic autonomy, I think that Zagzebski’s discussion of epistemic autonomy, again differentiated from the notion of selfreliance, captures the same general idea as I have presented above. She writes: A self-conscious being reflects. It thinks about what to do, what to believe, whom to trust. A being conscious of being self-conscious also reflects about the process of reflecting about what to do, what to believe, whom to trust, and so on. [. . .] Consciousness of self-consciousness adds higher-order norms to the norms of acting, believing, and desiring—norms of reflecting about the norms of acting, believing, and desiring. ‘Autonomy’ was invented as a name for what we do when we are conscious of being self-conscious in this way. Autonomy is the executive Which she seems to equate with epistemic agency. Elgin, ‘Epistemic Agency’, 144. 177 Ibid., 141. 175 176 135 self’s management of itself, or what is usually called self-governance. In this book I have proposed that the basic norm of self-reflection is what I call conscientiousness, the property of exercising my faculties in the best way I can to make the outputs of those faculties fit their objects—to make my beliefs true, my desires of the desirable, my emotions appropriate to their intentional objects. Conscientiousness is the higher-order norm of self-reflective beings. It is the norm of autonomy.178 Similarly, in speaking of how one’s autonomy can be potentially be undermined, Zagzebski speaks of coercion as a violation of selfhood: ‘[i]t is a violation of my selfhood because it prevents me from exercising the reflective control I have over myself’.179 In the epistemic realm we can understand this as being unable to reflectively evaluate the components of one’s epistemic perspective, and thus losing control over ‘who one is’ as an epistemic agent. What matters to my being epistemically autonomous, then, is that I am able to take a higher-order position on my first-order epistemic attitudes or behaviours. This first requires that I can take a (somewhat detached) position overseeing these attitudes and behaviours, from which I can then evaluate them, and reflect upon whether they are the kinds of epistemic attitudes and behaviours that I wish to endorse as a rational, self-conscious creature, aiming for certain epistemic (as well as practical and moral) goods and ends. For Zagzebski, that is what is required for me to be ‘conscientious’, or to ‘do my best’ as an epistemic agent of the sort I am, so-to-speak. Again, nothing about this requires that the all the cognitive and epistemic work that takes place to reach this end is mine, in the sense suggested by the notion of self-reliance. Why would it be? If we take a moment to bracket out the potential consequences of being self-reliant or not (avoiding error, other people’s biases or deceit, etc.) and focus solely on the idea of self-reliance in itself, we are quickly confronted with a question that strikes at the core of the issue: what does it matter that the epistemic work is ‘mine’ in this sense, how does that fact confer any epistemic value to it? I 178 179 Zagzebski, Epistemic Authority, 230. Emphasis my own. Ibid., 234. 136 think it is quite clear that the answer to that question is that it does not.180 Of course, this in no way means that the notion of self-reliance, or the fact that a certain epistemic stance is the result of one’s own doing, has no relevance to our epistemic judgements of an agent’s cognitive behaviour. My only claim is that there is room for a notion of epistemic autonomy that does not require self-reliance, and that this notion better matches our other notions of personal and moral autonomy; importantly for my purposes here, this notion can also make sense of our intuitions in regard to our epistemic autonomy in relation to our interactions with epistemic authority, and other social-epistemic relationships besides. The notion of selfreliance most certainly has a role to play in epistemology (whether someone deserves merit for the epistemic positions, for example, is a question that can be partly answered by reference to such facts).181 However, I hope to have shown that we should at least be wary of wedding these considerations with the notion of epistemic autonomy. Given this conception of epistemic autonomy as reflective endorsement and identification, I will now move on to a discussion of how such a notion of autonomy relates to our epistemic responsibility when establishing a relationship with an epistemic authority, and when acting as non-authorities deferring to such authorities. To give one more example, Foley seems to provide further credence to the view I am defending here when he writes that he conceives of ‘autonomy as being grounded in our ability to use our existing methods and opinions to examine these very same methods and opinions, the very same ability, not coincidentally, which makes epistemology possible’. See: Foley, Intellectual Trust, 128. Here I further stipulate that it also includes the ability to use our existing methods and opinions to examine other methods and opinions, and to endorse them, even if one is not ‘making use’ of them in exactly the same fashion. 181 It should be noted here that self-reliance may not even be relevant to this task, however. Jesús Vega Encabo, for example, has argued that there is room for another notion of epistemic autonomy, once again distinguished from that of self-reliance, that can make sense of how we can attribute an epistemic success to an agent even when they do not rely on their own faculties (specifically in the case of testimony). See: Jesús Vega Encabo, ‘Epistemic Merit, Autonomy, and Testimony’, Theoria 61 (2008). 180 137 III.3 AUTHORITY AND MINIMAL EPISTEMIC AUTONOMY AND RESPONSIBILITY With a clearer sense of epistemic autonomy in mind, we can now move on to evaluate the case of epistemic authority vis-à-vis such autonomy. Recall that I began this chapter by reference to what I take to be a fairly intuitive worry about accounts of epistemic authority that include some strong normative power (of the sort found in either Zagzebski’s, or my own, conditions of Preemption). In this section I will argue for two things: 1. That an agent deferring to these kinds of epistemic authorities can meet a minimal level of epistemic responsibility by meeting a certain condition of epistemic autonomy: namely that they have reflectively endorsed the epistemic practice with which they identify the authority as being an exemplar of. 2. That an agent deferring to these kinds of epistemic authorities can be held epistemically responsible, and thus blameworthy or praiseworthy, for the consequences of such deferral due to the fact that it is that agent’s reflective endorsement of both the epistemic practice, and the individual identified as an epistemic authority, that precisely establishes the very relationship between non-authority and authority in the first place. The purpose for accounting for a minimal sense of epistemic autonomy and responsibility in (i.) is to assuage worries that involving oneself in a relationship with an epistemic authority is in principle irresponsible or epistemically inappropriate. In III.3.A., I show that, given that epistemic autonomy is not epistemic self-reliance, there is a straightforward way to account for how an agent is not irresponsibly forsaking their own epistemic autonomy or agency in establishing a relationship with an epistemic authority. Point (ii) is pursued for the sake of making sense how we can blame or praise someone for deferring to an authority—this is argued for in 138 III.3.B. Both of these provide an account of how one can have a minimal form of epistemic autonomy and responsibility in dealing with an epistemic authority. Later, in sections III.4 and III.5, I will argue that the diversity of the forms that a relationship with an epistemic authority can take (as seen in Chapter II) add further complexity to the discussion of epistemic autonomy and responsibility beyond this minimal sense. The upshot of this will be that epistemic responsibility does in fact require, in some cases, that a non-authority do more than satisfy the minimal sense of epistemic autonomy presented in this section. I will argue that not to do more than what is minimally require for epistemic autonomy in such cases is akin to being epistemically lazy, and thus epistemically irresponsible. I further suggest that not only are we epistemically autonomous in establishing these relationships with epistemic authorities, such relationships, when responsibly pursued, improve our epistemic autonomy. III.3.A. AUTONOMY IN ENDORSING AN EPISTEMIC PRACTICE Before I provide an account for a minimal sense of epistemic autonomy in an agent’s identifying another agent as an epistemic authority and consequently adopting a relationship with such an authority, I should make it clear that I take the notion of epistemic autonomy to be strongly related to that of epistemic responsibility, and that being epistemically autonomous is necessary condition of being responsible. Though I think there are numerous ways in which epistemologists have understood epistemic responsibility, I mean something like the following: Epistemic Responsibility: An agent is epistemically responsible in relation to some epistemic attitude, behaviour, or position, iff she is an appropriate target of praise or blame for that epistemic attitude or position on the grounds that she is behaving as an autonomous epistemic agent in relation to that attitude, behaviour, position, where being ‘appropriate’ takes into account not only epistemic considerations, but also pragmatic. 139 To be judged ‘epistemically responsible’ in believing that p, would just to be the appropriate target of praise as outlined; to be judged ‘epistemically irresponsible’ would be to be the appropriate target of blame as outlined. Judging an agent’s epistemic stance as being epistemic responsible is thus not the same as attributing rationality or irrationality. It may be the case that an agent’s believing that p, for instance, is irrational because it fails to be properly responsive to all the evidence on hand. However, consideration of the fact that the agent was under duress, and was pragmatically constrained by the fact that she had to make a quick decision to adopt a certain doxastic stance, may entail that we would not want to consider her blameworthy for believing that p. In this sense she is not epistemically responsible for believing that p. This distinction between epistemic rationality and responsibility has been made by Richard Foley, who refers to an agent who holds a belief that is not rational, in the sense that, upon reflection, she ‘would be critical of them, insofar as her goal is to have accurate and comprehensive beliefs, but they may nonetheless be responsible beliefs for her to have, given that it was reasonable, relative to all of her goals, for her not to engage in this reflection’.182 Assuming that the other goals in questions may be pragmatic, I take it that Foley’s account of epistemic responsibility generally echoes my own. Whether an agent is epistemically autonomous relative to a certain epistemic position is a central question in making sense of such judgements of blame and praise. If, for example, an agent is not able to reflectively evaluate and endorse a certain epistemic position, and thus identify it as her own, but instead has such a position thrust on her by some coercive means, then it would seem inappropriate to blame her for it. An agent finds a certain doxastic attitude ‘implanted’ into her general cognitive make-up would be such an example, or an individual 182 Richard Foley, Intellectual Trust in Oneself and Others (New York: Cambridge University Press, 2001), 33. 140 brainwashed into believing a certain doctrine.183 This connection between epistemic autonomy and responsibility should come as no surprise, given that the broader notions of autonomy and responsibility outside of the epistemic domain so clearly come hand-in-hand. Paul Benson refers to autonomous agents as those that treat themselves as ‘sufficiently competent and worthy to speak for their actions’, i.e. as being responsible for them.184 Andrea C. Westlund argues that an agent is autonomous in relation to some action so long as she holds herself interpersonally accountable, in that she can speak ‘on her own behalf’ apropos these actions, and that she sees herself as having endorsed certain ‘values and desires [treating them] as justifying reasons for her actions’.185 My claim is simply that this is the same in the epistemic domain: an agent who reflectively endorses and identifies with a certain epistemic attitude, behaviour, or position can be held responsible for it. In broader terms, Lorraine Code echoes this connection between epistemic autonomy and responsibility when she states that a ‘person can be held [epistemically] responsible or irresponsible only if she/he is clearly regarded as an agent (in this case a cognitive agent) in the circumstances in question’.186 Before I continue, however, I should make two small points clear: Note that an interesting consequence of my view of epistemic autonomy is that such individuals could be blameworthy for such epistemic positions at a later time, so long as they were provided with the opportunity to reflectively evaluate such positions in the meantime. There would be a point in which they would have to be held responsible for such beliefs, attitudes, etc. This is of course a complicated matter: subjects of constant brainwashing and external pressure will likely never have the opportunity to reflect on such beliefs for prolonged periods of time, and there is also the question of how such coercion, brainwashing, etc., could impair the ability to reflectively evaluate and endorse epistemic attitudes and positions, i.e. how they could undermine the possibility of epistemic agency completely. I do not want to make a definitive stance here, but there is certainly a question to be asked here in regard to our responsibility for what we may call our ‘basic’ epistemic methods and practices. It seems to me that there is at least room to criticise even those who have been raised in such manipulative environments, as the ones mentioned above, on the grounds that they have not properly evaluated the very basic means by which they acquire and sustain their beliefs. Such individuals, it would seem, give authority to others, even though those others are their peers when it comes to matters of epistemic method. Though I do not fully agree with his view, I am at least sympathetic to Michele Palmira’s related argument—see Michele Palmira, ‘Expert Deference and the Objectivity of Epistemology’, unpublished manuscript. 184 Paul Benson, ‘Taking Ownership: Authority and Voice in Autonomous Agency’, Autonomy and the Challenges of Liberalism: New Essays, ed. J. Christman and J. Anderson (Cambridge: Cambridge University Press, 2005), 117. 185 Andrea C. Westlund, ‘Rethinking Relational Autonomy’, Hypatia 24:4 (2009), 35. 186 Lorraine Code, Epistemic Responsibility (Hanover: University Press of New England, 1987), 51. 183 141 i. Being held responsible for an epistemic attitude, behaviour, or position, does not necessarily have to be a total affair, in the sense that the agent in question is held solely and completely responsible for that attitude, behaviour, or position. This is especially relevant in the context of this discussion of epistemic authority, given that I will later argue that the authority is responsible for much besides (see III.6, and Chapter IV). ii. An agent can be held epistemically responsible for an epistemic attitude, behaviour, or position that she seemingly does not take steps to reflectively endorse, given that she had the opportunity to reflectively evaluate and endorse said attitude, behaviour, or position, but chose not to.187 In the context of our current discussion on epistemic authority, for example, an agent could be held responsible for willingly not adopting a relevantly reflective position in adopting a position of deference relative to the authority. This point will become clearer in III.3.B., however. Now that we have a clearer sense of what is meant by epistemic autonomy and epistemic responsibility, my first claim here is a simple one: Minimal Epistemic Autonomy and Responsibility: An agent is considered minimally epistemically autonomous, and thus epistemically responsible, as a non-authority in a social-epistemic relationship with an epistemic authority, so long as she reflectively endorses the epistemic practice with which she identifies the epistemic authority as being a successful practitioner of. The point here is that an agent can manifest her epistemic autonomy by endorsing, upon reflective consideration, the very practice that she, via identification of an epistemic authority, defers to. Although, due to the nature of the epistemic authority to non-authority relationship, it 187 I say seemingly due to the fact that an agent, on my account, that did not endorse a certain epistemic attitude, behaviour, or position, would not be counted as autonomous. I would argue that such an agent in fact did endorse a certain position, but did so lazily and incompletely. This will become clearer in the next section, however. 142 may not be possible for her to reflect on the minutiae of the epistemic practice that lead to a certain doxastic attitude or epistemic position (which, in deferring to the authority, she ends up adopting herself), there is nothing stopping her from consideration of the relevant epistemic practice as a whole. In fact, it is epistemically mandatory that she do just that: any agent who fails to do so before deferring to an authority would be aptly criticised for being epistemically irresponsible. Such deference would constitute a severe violation of one’s epistemic autonomy, albeit a self-imposed one, because such an agent would then be at the mercy of an authority who would impose a certain epistemic practice, constituted by a certain methodology and set of rules of which the non-authority would have no means of reflectively evaluating. The act of endorsing the practice as a whole, as a constitutive step towards adopting a position of deference to an authority, is a crucial component to doing so in an epistemically responsible manner. Such endorsement could be extremely crude, as it will often be in cases in which the disparity between an epistemic authority and non-authority’s knowledge and skill level is severe. Consider the case of my own deferral to an expert in sub-atomic physics: I have very little, if any, ability to rationally evaluate the details of the methodology that drives the relevant epistemic practices of that domain. However, I reflectively endorse these practices as a whole: I believe that they have an empirically adequate track record, as exemplified by their successful application to technological advancement, for instance. Though I do not understand most of the answers that such epistemic practices provide, I may see that the answers have been successfully used in application to other domains, or other problems, in which I see that the answers have played a sufficiently important and successful role. If I did not have reason to endorse the practice even in this crude manner, I think it would be perfectly appropriate for me to strongly criticised for adopting a position of deference to the relevant authority identified as a practitioner of that epistemic practice. 143 This is an important point that I think is often overlooked or passed over. Hardwig, for example, in suggesting that deference to an epistemic expert (either by a layman or another expert) is necessarily ‘blind’, fails to acknowledge the components of such relationships that are necessarily reflective and appropriately epistemically perceptive. Yes, we are blind to the particular processes that govern some components of a certain epistemic practice with which we wish to defer to—but this in no way suggests that our choice to defer to such practice lacks the appropriate reflectively evaluation and endorsement. Surely, for some individuals, such evaluation will be very crude (as described above), given their inability to even evaluate an epistemic practice in the simplest of terms, but for others such evaluation and endorsement will be quite sophisticated and intellectually virtuous indeed. My reader may rightly raise an objection at this point: surely the act of endorsing an epistemic practice overall necessarily involves the act of endorsing the particular methods, rules, and so forth that constitute that practice. Given that part of what gives rise to the epistemic authority to non-authority relationship is precisely the latter agent’s inability to directly engage with these methods, it becomes questionable that the kind of reflective evaluation and endorsement I am suggesting would generally be possible. However, I think this criticism fails to mark the crucial distinction between a non-authority’s inability to make direct use of these methods, and his ability to evaluate them on a variety of grounds nonetheless. Surely, I can recognise that it would be epistemically good, in a general sense, to follow a certain method even though I do not have the means to practice it myself, or to evaluate its application in particular instances. I think this still holds even for methods that I cannot even make sense of or comprehend in clear terms: it suffices that I can recognise that there is a method in place and that it has a good track record, for instance, or that it benefits the epistemic practice in a more indirect fashion. Nor do I think it necessary that a non-authority has to be able to reflectively 144 evaluate all the various methods and rules that practitioners of a certain epistemic practice adopt. I can endorse key, constitutive elements of a certain practice, while remaining unable to do much cognitive work in evaluating others due to their being difficult to understand, opaque, etc., but perhaps seeing that they work in tandem with those that I do. Furthermore, as mentioned above, I do not think that the kind of reflective endorsement I have in mind here has to be the kind that solely takes into accounts epistemic considerations. Sometimes we endorse an epistemic practice, and follow through with it, because we consider the practice, in a given situation, as being our best bet when trying to solve a practical problem, or reach a practical end. Much as with any attribution of epistemic responsibility, here pragmatic considerations have a role to play. The practical benefits of an epistemic practice surely play some weight in my decision to reflectively endorse said practice. This does not mean that one can simply endorse any position solely on non-epistemic grounds and be considered epistemically autonomous in the relevant sense. If I find myself in a position in which pragmatic grounds force me to follow an epistemic course of action that I reflectively realise is epistemically bad, then I think I can be considered as being epistemically heteronomous in relation to that course of action. Here I am ‘a victim of circumstance’, as Elgin put it.188 Given these pragmatic considerations, I am not epistemically responsible either, given that I am not an appropriate target of blame for adopting such epistemically bad courses of action. It may be wondered why I have suggested that the minimal condition for epistemic autonomy in one’s relationship with an epistemic authority is a reflective endorsement of the practice, rather than the individual that is identified as the authority. The reason for this is that I think it is precisely the practice that we defer to, not the individual: the individual only has power to be authority given that he or she is a practitioner of this epistemic practice that we find value 188 Elgin, ‘Epistemic Agency’, 141. 145 in. We defer to the authority qua practitioner, as an instantiation of that practice in action, and as an agency that is in some sense constitutive of that practice (at least in a social sense). Elgin notes just this: deference is not ‘primarily to a person: the expert is taken to embody the epistemic commitments of his area of expertise’,189 and it is these commitments that I endorse and defer to.190 By endorsing a certain epistemic practice, an agent opens herself up to certain criticisms. She could be asked, for example, why she thinks that such a practice is worth deferring to. If I am asked why I defer to the practice of astrology in adopting certain doxastic attitudes towards propositions describing future states of affairs relevant to my life, for instance, and I cannot provide any reflectively good reasons for thinking that the methods that constitute that practice are good ones (for whatever ends I have)—assuming that I have in no way been withheld the opportunity to reflect on the matter—I am surely an appropriate target of criticism. This would seem to be especially clear in the case that my critic asks us to compare my chosen epistemic practice with one that she, my critic, can provide good reasons for deferring to. III.3.B. ENDORSEMENT AS CONSTITUTING THE AUTHORITATIVE RELATIONSHIP Our epistemic responsibility vis-à-vis epistemic authority does not end there, however. Not only does our reflective endorsement of a given epistemic practice, i.e. our autonomous deference to the practice, make us epistemically responsible in the sense that we can be the appropriate target of criticism if we have bad reasons for endorsing the relevant methods that constitute that Ibid., 148. It should be noted, however, that Elgin and I disagree on whether or not we can reflectively endorse these commitments themselves. I have already stated that I think it is so. Elgin, however, assumes that given that the practice depends on certain ‘fine-grained commitments that we are not privy to and would not understand even if we were’, we cannot endorse them. Instead we have to place trust in the relevant epistemic authority that they themselves endorse good epistemic commitments. 189 190 146 practice—such endorsement is also fundamentally central to, at least on my account of epistemic authority, the establishing of the relevant relationship between epistemic authority and nonauthority in the first place. Recall my account of epistemic authority: Epistemic Authority: An agent S is an epistemic authority relative to another agent S* iff: 1. S has the requisite skills, abilities, and know-how to successfully partake in a certain kind of epistemic practice, EP, relevant to some domain of inquiry d. 2. S has sufficient access to the evidence and resources required for her to properly partake in the EP relevant to d. 3. S* does not have the skills, abilities, and know-how required to partake in the EP relevant to d, or at least has them to a (significantly) lesser extent than S. 4. S* recognises that d and the relevant EP is epistemically valuable for S*, i.e. it pertains to questions that would be epistemically valuable to have answers to. 5. S* recognises S as having a sufficient level of the requisite skills, abilities, and know-how to successfully partake in EP. 6. (From 4 and 5:) S* recognises that S is, by virtue of her ability to partake in the relevant EP, a potential source of some kind of epistemic good that S* would be better off in having. 7. S has the power to give S* a preemptive reason to behave in a certain epistemic fashion, perhaps by providing higher-order reasons and 147 beliefs about methodological issues relevant to the kind of EP that is pertinent to d, or alternatively by commanding S* to behave in a certain sort of epistemic fashion. In short, S is able to authoritatively tell S* how she, S*, should epistemically behave in order to partake in the EP relevant to d. Though it is not the sole condition, on such an account, the act of recognising the epistemic practice as beneficial, and of recognising the identified authority as a sufficiently talented practitioner of that practice, is precisely what gives the authority her authoritative standing. It is precisely for this reason that a non-authority cannot simply shirk responsibility when things go wrong. It may be tempting to think that, given that the authority in such a social-epistemic relationship has the means to deceive and misguide the non-authority, and that the non-authority does not have the means to directly and reflectively evaluate certain parts of the relevant epistemic practice, that the non-authority cannot be responsible for the end results: she is completely at the mercy of the authority. But this is clearly wrong: it makes perfect sense that we could ask such an agent ‘Why did you treat her as an authority in the first place, then?’. On my account, this criticism is provided with the appropriate force: my identifying someone as an authority (by meeting a variety of conditions) is a necessary part of their attaining the position of epistemic authority in the first place. The strength of this criticism will of course vary, depending on the reasons that a non-authority has for both identifying the epistemic practice as a good one, and for identifying the authority as a sufficiently skilled practitioner of said practice. But the point remains: she is clearly responsible to some degree, partly because she was epistemically autonomous, in the sense that I have outlined here, in bringing it about that the authority was treated as such, and had the relevant normative power over her, the non-authority. As Elgin puts 148 it, an epistemic agent ‘confers epistemic authority on those that she counts as experts’, and thus remains responsible for doing so—furthermore, she always ‘retains the power to revoke’ the conferral of authority. The non-authority remains epistemically autonomous and epistemically responsible precisely because of the fact that she has the power to establish and revoke such relationships, and this power comes prior to the authority’s power to determine the nonauthority’s epistemic behaviour in relation to the epistemic practice in question. Discussing testimony more broadly, Jesús Vega Encabo suggests something similar when he argues that the hearer in the testimonial relationship is responsible, and deserving of merit (in a successful case of testimony) given that the hearer plays a role in constituting that relationship, by recognising his own epistemic position in the exchange.191 Both the speaker and hearer declare, at least in an implicit way, the standings they occupy in the epistemic space. And both of them contribute to generate a situation under which they are obliged to acknowledge certain epistemological requirements [ . . . As such,] success is due to the exercise of relevant epistemic competences of both participants. And merit attribution depends on how each participant puts into play his own epistemic agency.192 When it comes to my account of epistemic authority, things are somewhat different, given that it is not necessarily the case that the authority will recognise the contours of the social-epistemic relationship.193 It may be the case the authority does not contribute to the epistemic situation in the way that Vega Encabo suggests happens in the case of basic testimony. However, Vega Encabo’s views about the hearer’s role are what interest me here, and I think they echo my own For empirical research on how human beings, qua listeners, can be considered as active, and vigilant participants in testimonial exchanges, see Dan Sperber, Fabrice Clément, Christophe Heintz, Olivier Mascaro, Hugo Mercier, Gloria Origgi, and Deirde Wilson. ‘Epistemic Vigilance’. Mind and Language 25:4 (2010). 359-393. 192 Vega Encabo, ‘Epistemic Merit’, 52. 193 For instance, as mentioned in Chapter II, my account makes it possible that I treat someone as an epistemic authority even though they have no idea who I am. Perhaps I treat the author of a book as epistemically authoritative, and allow their expressed views in said book to determine my own epistemic behaviour. Such an author would have little reason to recognise me, in particular, as being part of the social-epistemic relationship. (One could argue that in writing the book, the authority would have accepted and recognised the possibility of being placed in such a position, and thus could be considered as adopting it—however, we can avoid this by simply stipulating instead that the book was never intended for publication or circulation.) 191 149 thoughts. It is the non-authority’s adopting a position of self-aware and reflectively endorsed epistemic inferiority, and of attributing authority to another, that plays a fundamentally important contribution to the establishment of the authoritative relationship. I agree with Vega Encabo in his claim that the non-authority (or ‘hearer’) deserves credit for such an act—this is just to say that he is epistemically autonomous and responsible in the situation. In being a necessary part of establishing the epistemic authority to non-authority relationship, my reflectively endorsing the epistemic practice, and identifying the authority as a practitioner of that practice, makes me epistemically responsible for the fact that I am in a position of deferring to such authority, and under their normative, epistemic guidance. III.4. COGNITIVE LIMITATIONS, AND THE VARYING DEGREES OF EPISTEMIC INVOLVEMENT Much of what I have argued for above has already been said by others, as we have seen, in one way or another. What I would like to do now, however, is suggest that in the context of the expert-to-layman relationship, or more specifically in the case of an epistemically authoritative relationship (as identified by my account), there are further complexities pertaining to questions of epistemic autonomy and responsibility that have not been properly addressed in the literature. The basic point is this: in the relationship between an epistemic authority and a nonauthority, there can be varying degrees of possibility for epistemic involvement, on the nonauthority’s part, in the relevant epistemic practice. This is a simple consequence of what I argued for in Chapter II: that the relationship between epistemic authorities and non-authorities can take on a variety of different forms, depending on the epistemic practice in questions, and the 150 relationship the non-authority has with that practice on various epistemic dimensions, including consideration of the kind of epistemic good that the non-authority is seeking in establishing said relationship. Some of the relevant differences to keep in mind, are the following, which I will refer to as Possibly Significant Differences: i. The kinds of epistemic goods that the non-authority seeks in identifying the authority as a source of such goods can vary, and can be quite different in kind/complexity. ii. The disparity between authority and the non-authority’s level of skill, ability, and know-how required to partake in the relevant epistemic practice, can range from wide to narrow. iii. The kinds of evidence that a non-authority has access to may be more or less accessible for the authority. iv. The kinds of evidence that a non-authority has access to may be more or less relevant to the epistemic practice in question. v. The nature of the questions that the epistemic practice aims to answer may be more or less objective, in the sense that some answers to such questions may be more or less dependent on the subjective perspective of the non-authority who seeks the answers. vi. The nature of the questions that the epistemic practice aims to answer may have more or less practical import, meaning that the non-authority may have more or less to practically risk in accepting the outputs of said practice. There are likely a wide range of other factors that can play a significant role in altering the contours of the social-epistemic relationship, and, in turn, the epistemic role and responsibilities 151 of both parties involved. Points (i) - (vi), however, should sufficiently illustrate the kinds of differences that I have in mind. These differences suggest that there are varying degrees of possible involvement a non-authority can have in the relevant practice. In many cases, none of these possible differences will play a significant role in altering the non-authority’s epistemic responsibilities beyond that of merely deferring to the authority.194 For instance, if I were to identify Peter Higgs as my epistemic authority in the domain of physics (or at least in some sufficiently narrow sub-domain pertaining to questions regarding the nature of sub-atomic particles), it is fairly plausible to assume the following application of Possibly Significant Differences holds. The Epistemic Authority of Peter Higgs (EAPH): i. I am merely seeking true beliefs pertaining to the nature of sub-atomic particles, given that I doubt that I could reach any higher epistemic position in the domain in question (i.e. I don’t think I could understand the relevant physics, specifically given the differences stated below). ii. The disparity between my level of skill, ability, and know-how relevant to partaking in the epistemic practice in question, and his, is significantly wide. iii. The kinds of evidence I have access to about sub-atomic particles and physics more broadly is the kind of evidence that Higgs has access to as well (along with a broad range of evidence that I do not have access to). iv. The kinds of evidence that I have access to are likely completely irrelevant to the epistemic practice that Higgs is a skilful practitioner of. v. The kinds of questions that the epistemic practice in question is aimed to answer are independent of my subjective perspective, and the answers to these 194 I will return to the responsibility of the epistemic authority in the next chapter. 152 questions would reasonably be thought to be completely unaffected by my own subjectivity. vi. The answers to these questions serve very little practical import to me, at least in a direct sense.195 This is the kind of cases that epistemologists typically target when they discuss expertise or epistemic authority. The deference in question is simple: I have no plausible epistemic role to play in the epistemic practice, and all I can responsibly do is simply defer to Higgs’ authority, where any interjection on my part would likely be irrational. Note, however, that if I were to seek understanding rather than mere true belief (just as Jäger has suggested we often do in interacting with an epistemic authority) then things would be slightly different—I would, given my epistemic aims, seek some explanation from Higgs, rather than simply deferring to his belief. However, given (EAPH) (b) - (f) it would be very difficult to see how I could gain such understanding, at least beyond a merely metaphorical or highly abstract sense, and I would argue that it would be highly irresponsible of me to aim for such a high epistemic position if it were at the cost of deferring to Higgs at the level of belief, and thus losing access to such true beliefs. As will be discussed further below, this suggests that an awareness of my own cognitive limitations is crucially important in my establishing the right kind of relationship with an epistemic authority. Now, in contrast to the Higgs case, consider the following kind of case, briefly alluded to above. Here I am a medical patient interacting with a medical practitioner, trying to decide the best course of action to take in treating a serious, possibly fatal, illness. The Epistemic Authority of the Medical Practitioner (EAMP): 195 They of course play a significant role in the sense that other agents can use the answers to questions about quantum mechanics to invent technologies that significantly alter my life, for instance. However, the point here is that my having the answers to these questions is practically irrelevant to my life. 153 i. I am not merely seeking true beliefs pertaining to the answers to basic medical practice—i.e. ‘What disease do I have?’, ‘What is most likely to cure the disease?’, ‘What is the best course of action I can take?’, etc.—but rather seeking a broader, stronger epistemic foundation from which I (in collaboration with my doctor) can make good decisions (epistemically and pragmatically) about my medical future. Arguably I am looking for some understanding of my epistemic position, and how it relates to certain practical considerations. ii. The disparity between my level of skill, ability, and know-how relevant to partaking in the epistemic practice in question, and my doctor’s, is significantly wide. iii. However, the kinds of evidence I have access to about my medical situation, at least from at a first-personal level, are not directly accessible to my doctor. Though, of course, she also has access to a wide array of evidence that I do not (third-personal facts about my physical condition, what my reports of firstpersonal experience generally indicate about a condition, etc.). iv. The kinds of evidence I have special access to are significantly relevant to the epistemic practice in question, and specifically to my goal of acquiring the epistemic goods that I seek. v. The answers I am seeking, and that the epistemic practice (at least in part) aims to provide, are arguably highly dependent on my subjective perspective: if, for example, I value a high quality of life over the longevity of my life, certain courses of medical action may be more preferable to me than they would be to someone who had the opposite preferences. 154 vi. And, of course, the answers to the questions the epistemic practice aims to answer are clearly of grave practical import for me. In this specific situation they may pertain to my continuing to live, or not. Here we have a case that is significantly different from the situation outlined in the Peter Higgs case (EAPH). The medical practitioner case suggests that there is room for my own epistemic involvement in the relevant epistemic practice, in order to get to the epistemic goods I seek. In fact, I think this case is close to the other end of the spectrum, in contrast to the Higgs situation: it would seem epistemically imprudent for me to not involve myself as much as possible. To be epistemically responsible here, I would seek out and share as much of the kinds of evidence that I have special access to; I would ask questions about the authority’s statements, the results and evidence that she shares with me, in order to better understand my own situation, both epistemically and practically; I would reflect on the practical import of the epistemic practice, and the role my own subjectivity (in the form of certain preferences) had to play in clarifying an answer to the question of what was best to do. This all suggests that it would not be enough that I had been epistemically autonomous in establishing the epistemic authority to non-authority relationship: more is required of me beyond this to be epistemically responsible.196 The nature of the epistemic practice, the kinds of questions it aims to answers, and the significant differences or similarities between the epistemic positions of authority and nonauthority, to name a few, all play a role in structuring the role and responsibilities that I have in the face of the individual that identify as my epistemic authority. In the Medical Practitioner case, unlike the Higgs case, we find illustrated a complicated social-epistemic relationship, the contours of which are not so easily captured by a discussion of pure, basic deference. In the next It should be pointed out, however, that none of these actions suggested go against the idea that the authority has a preemptive power over me. Asking questions of an authority is not the same as submitting answers to an authority, contrary to their own answers. 196 155 section, I will argue that such cases present the non-authority with further epistemic responsibilities, above and beyond the minimal sense argued for above in III.2. This suggests that, in such cases, I can succeed or fail to be epistemically responsible, in a more robust sense above and beyond the minimal sense. III.5. DOING WHAT ONE CAN: EPISTEMIC RESPONSIBILITY AND EPISTEMIC LAZINESS IN DEFERRING TO AN AUTHORITY In the last section, I suggested that the kinds of relationships that can fall under the category of epistemic authority can vary on a variety of important epistemic dimensions. These variations on certain dimensions amount to the following: Varying Degrees of Involvement: In the epistemic authority to non-authority relationship, there are varying degrees of (possible) involvement that a nonauthority can take in the relevant epistemic practice. I have suggested that in cases in which there is more opportunity for a non-authority to be involved, he has an epistemic responsibility to do so, all things being equal. In this section, I will explain, and expand on this point by arguing for the following: Epistemic Laziness: It is epistemically lazy to completely defer a certain epistemic practice to an epistemic authority when there is a possible level of significant involvement that the non-authority can take in that practice. I will then show that being epistemically lazy in this sense is detrimental to one’s epistemic autonomy. From this, it will be argued that it is epistemically irresponsible to be epistemically lazy in this fashion. In III.1. and III.2. I articulated a minimal sense of epistemic autonomy, and responsibility, that I think one maintains when establishing a relationship with an epistemic authority. This sense of autonomy and responsibility was intended to illuminate how I, as a non- 156 authority, can behave epistemically appropriately in such deference to authority. This suggests that in satisfying the conditions of such autonomy and responsibility, I can be attributed merit for the success of the social-epistemic exchange, given that I have manifested my epistemic autonomy in that exchange, reflectively endorsing the relevant epistemic practice, and in establishing the epistemically authoritative relationship in the first place, through such reflective endorsement. This might suggest that, given that one has met the conditions for minimal epistemic autonomy and responsibility, one can simply ‘sit back’ and rake in the fruits of the epistemic authority’s epistemic labours. Accounts such as Zagzebski’s, for instance, seem to suggest just this: so long as I think deferring to an epistemic authority will provide me with better access to true beliefs, or beliefs that will survive my future self-conscious critical reflection, I should simply defer.197 But, this is too simple a picture: Jäger is more on point when he suggests that we seek much more, i.e. understanding, from our authorities, and that seeking such understanding would require more epistemic labour on our part, even if this merely involved our asking more clarificatory questions of the authority, seeking deeper explanatory understanding. Once again, I am in favour of Jäger’s view. The various dimensions that I have identified, upon which we can measure the varying degrees of a non-authority’s possible involvement in an epistemic practice, are intended to clarify just how this is so. It would be facile to think that, given that the epistemically authoritative relationship is a direct consequence of the non-authority’s cognitive limitations, and, more broadly, the fact of our limitations as human beings more generally, it is sufficiently responsible for us to accept our I think Zagzebski would balk at this characterisation of her view: as we saw in Chapter I, some of her replies to Jäger’s and Dormandy’s criticisms suggested that a non-authority can simply choose (on good grounds) when and where to defer to an authority. As I noted in Chapter I, however, I do not think that this response tracks the kind of account that she is attempting to formulate. If the kind of Preemption she is endorsing is to be taken seriously, I don’t think a non-authority can simply pick and choose when it comes into effect. That being said, I do not wish to unfairly characterise Zagzebski’s view: what I refer to here is merely one formulation of a view in her vicinity, and my point in mentioning it is not to criticise her directly, but to illustrate my own account. 197 157 limitations and allow others to do the work for us when necessary. Certainly, I agree with this in the abstract—this is what responsible division of cognitive labour amounts to. In the Peter Higgs case discussed above, this is precisely what I would do if I were being epistemically responsible. However, as already shown, this is not how things always are. Let us return to the case of the Medical Practitioner4. Here, I think it fairly easy to accept that a patient, qua non-authority, who did not take certain steps to involve themselves in the epistemic practice would be the appropriate target of certain epistemic criticisms. If the patient were told to take a certain course of medicine, or to submit to a certain medical procedure, and he failed to ask certain questions, it would be perfectly reasonable to think that he had failed to fulfil certain epistemic responsibilities. Such questions could be of the following kind: ‘Why is this medicine the best option for me?’, ‘What are the other options?’, ‘Will this medicine have sideeffects that I should know about?’, ‘Will it improve my quality of life?’. Some of these questions could only be answered with further involvement on the patient’s part, by reference to: his preferences in regard to which side-effects he would consider worth the cost, or not; his own ideas of what makes for a good quality of life, and what makes for a bad quality of life; and so forth. As another example, consider the case in which a doctor tells you that you are suffering from some medical condition x, based on a series of complicated medical tests which you are in no position to directly evaluate. The condition x is such a condition that is known to cause distress in a certain part of one’s body. You, however, feel no such distress in that part of your body. It seems to me that you would be doing something epistemically wrong if you were not to mention this fact to your doctor—not because you should think yourself as having a defeater for the doctor’s claim, but rather because, in the context of the relevant epistemic practice, you have access to a kind of evidence that is both relevant to the practice, and that the doctor does not have the same access to. It may turn out that your specific evidence in this case is not in fact 158 relevant, or at least not substantial enough to change the authority’s judgement—nevertheless, you are still responsible for proffering it for evaluation. It would seem to me that, in establishing the relevant authoritative relationship through reflective endorsement of the relevant epistemic practice, one is then responsible for continuing to reflect on the possible involvement one may have with the epistemic processes, and playing one’s part in such processes when appropriate. One way to say this is in the following way: my reflectively endorsing the epistemic practice involves my committing to behave in such a way as to maximise the success of such practice, especially given that the practice is one that you seek to benefit from, in the form of receiving certain epistemic goods. The collection and presentation of evidence that one has special access to is certainly an important part of playing one’s part. Given the nature of the authoritative relationship, it will be up to the authority to attribute that evidence with the correct level of relevance, and to integrate that into any following epistemic calculation or decision-making process.198 Unless the authority has already told one that such evidence is irrelevant, I see no way to escape the fact that one is responsible for playing one’s evidence-collecting role in such an exchange. There are of course many factors that can make it reasonable for a non-authority to shirk some of these responsibilities. There may be pragmatic limitations: in the medical context, it is often the case that decisions have to simply be made quickly, and not all of the relevant evidence can be collected or accounted for. Perhaps more importantly, there is the issue of the nonauthority’s awareness of both his own cognitive abilities and limitations vis-à-vis the epistemic authority and practice, but also the methodology and ‘rules of the game’ of the epistemic practice with which he is attempting to gain access to. Just because he has reflectively endorsed that practice, it does not follow (in fact, given the nature of the authoritative relationship, it will most likely not follow) that he is aware of all the ways in which he could involve himself in the practice. 198 This is, roughly, what is behind the Preemption condition of my own account of epistemic authority 159 However, it seems to me that such ignorance cannot justify one’s failure to involve oneself when possible. Let me explain: ignorance of one’s ability to be in any way involved with an epistemic practice cannot reasonably be equated with a firm belief that one is not able to be involved. These are two completely different things. Ignorance in this case is tantamount to not knowing either way. To conclude that my evidence is not relevant, or that I have no epistemic role to play in the epistemically authoritative exchange, would in fact be anathema to my identifying someone as my epistemic authority: it would be to suggests that I know better than they do as to whether or not I have a role to play. To conclude as such would be to go against the autonomous decision one has made to treat someone as an authority in the first place. On the face of it, the assumption that one makes when one says that ‘I have no role to play here’ may seem like a humble one—however, in the context of the epistemic-authority to non-authority relationship, it is actually the opposite: it suggests an arrogance that goes against the very epistemic grounds of that relationship. This suggests the following: the default position in such relationships ought to be one in which one attempts to involve oneself, and withholds judgements as to what is or is not relevant to the practice. This means that the default position as a non-authority, reflectively aware of his epistemic position in the context of the relationship, is actually one of epistemic effort—it is not an epistemically lazy one. To refuse to involve oneself in the practice is to make an ungrounded assumption about one’s epistemic situation, an assumption one makes to justify one’s not doing further epistemic work beyond the establishment of the relationship, and telling oneself that this labour is for another to do. This is to be epistemically lazy. I take it that such laziness is opposed to what it is to be a good epistemic agent more generally—it is an epistemic vice. In the specific case of epistemic authority, however, I think it is especially an issue because it undermines our epistemic autonomy severely, both in that it 160 undermines the minimal sense of autonomy, discussed above, but also because it hinders our ability to be epistemically autonomous in a more robust sense beyond that. How does such laziness undermine the minimal sense of epistemic autonomy outlined above? It does so simply because a non-authority acting epistemically lazy in such a fashion behaves in a way that is incongruent with what they reflectively endorsed in establishing the epistemically authoritative relationship. Recall that to recognise someone as an epistemic authority is to recognise them as a more skilful practitioner of a certain epistemic practice, an epistemic practice which one reflectively endorses. Moreover, it is the case one only establishes the authoritative relationship on the grounds that one wishes to partake in the epistemic practice, even if such participants is in the minimal sense of acquiring access to its epistemic outputs. Part of establishing this relationships is taking the authority as having authoritative sway over questions of how one should behave in relation to the epistemic practice in order to partake in it (on my view such authority actually has the preemptive power to command us how to act, relative to the practice). As already noted, it seems to me that one can only justify one’s epistemic laziness on the grounds that one is not able to be epistemically involved in the practice in the sense outlined in III.4. But to assume that this is the case on grounds other than that the authority tells you so is to think of oneself as having the authority to make such judgements, completely contrary to one’s identifying the epistemic authority as having such authority. To behave epistemically lazy in the way that I have identified, then, is to behave in a way that is contrary to what one reflectively endorses—it is to behave in a way that is not in tune with one’s autonomous self. At base, it is to be inconsistent, and to hold incoherent epistemic attitudes. Above and beyond this, however, such epistemically lazy behaviour undermines our ability to be epistemically autonomous in a far richer and more robust sense than the minimal sense outlined in III.3. It seems to me that engaging in a relationship with an epistemic authority, 161 i.e. establishing such a relationship, is a means to improve one’s epistemic position. Part of the value in this epistemic endeavour comes in the fact that such relationships can provide us with the opportunity to be ‘more’ epistemically autonomous in that we can be more self-aware and conscientious of our own position in the epistemic space we find ourselves in. Being aware of my own cognitive abilities and limitations, I can better position myself to reflectively evaluate and endorse certain epistemic behaviours: of my own, and of others. Self-awareness provides me with the opportunity to generate new, and modify my current, higher-order epistemic attitudes, allowing for me to regulate myself as an epistemic agent. Being aware of the fact that I cannot complete certain epistemic tasks on my own is precisely the epistemic attitude that drives me to endorse and participate in crucial social-epistemic practices, most obviously in the case of testimony, and in the specific case of epistemic authority discussed here. Shirking the opportunity to gain a better sense of my abilities and limitations is thus antithetical to my autonomy in that it involves obstructing myself from being able to critically evaluate and reflectively endorse the epistemic activities that I participate in. All this is to say that dependence on others, in particular an epistemic authority, is, rather than a means to limit our epistemic autonomy, in fact the opposite: it is a means to improve ourselves as epistemic agents, and to clarify and to enrich our own autonomy in the epistemic realm. Zagzebski indirectly suggests something in the vicinity when she writes: [B]ut our self-reflective dependence on others also means that other persons can often help us do what we cannot do on our own. They can help us resolve dissonance in many ways—by helping us form higher-order judgment about what ought to change, by influencing us to acquire the proper motivating emotion, by showing us exemplars of harmonious selves.199 None of this can be done without our doing the necessary epistemic labour where necessary, and only the authority can provide us with good reason to think certain epistemic effort is necessary 199 Zagzesbki, Epistemic Authority, 236. 162 or not. III.6. EPISTEMIC AUTONOMY IS NOT ENOUGH: EPISTEMIC HELPLESSNESS IN DEFERENCE I have argued that I am epistemically autonomous, and thus epistemically responsible, at least in a minimal sense, when I identity another agent as an epistemic authority. This is because I reflectively endorse the epistemic practice of which I identify the authority as being a sufficiently skilled practitioner of, but also because I am primarily responsible for establishing the authoritative relationship in the first place—again, by my own volition, and reflective endorsement. Above and beyond this, I have also argued that, in some cases of epistemic authority to non-authority relationships, the non-authority has further conditions to meet to be considered epistemically responsible. These are cases in which, on some dimension or another, the nonauthority can possibly be more involved in the relevant epistemic practice. I have suggested that when this is the case, and a non-authority unreasonable refuses to involve himself in the right way, he is an appropriate target of the criticism of epistemic laziness, which I have understood as pertaining to a diminishing of one’s epistemic autonomy, and thus weakening one’s epistemic position. A crucial component for identifying the possibility of such involvement, and responding appropriately, is one’s having the self-awareness to be aware of one’s own cognitive abilities and limitations. These considerations are intended to provide at least the outline of an account of the nonauthority’s epistemic responsibilities in the epistemically authoritative relationship. Such considerations, however, only tell half of the story. It is, of course, possible for a non-authority to 163 do all that is required of them to be epistemically responsible, and yet for them to be epistemically worse off because of it. This is the case when the epistemic authority behaves epistemically inappropriately, and causes epistemic harm to the non-authority. As is no surprise, being epistemically responsible, and epistemically autonomous (in the sense I have outlined here), is no guarantee of success. The epistemically authoritative relationship can come with great rewards, at the cost of great risk. In the next chapter I argue that, much akin to the non-authority, the epistemic authority has a responsibility to be aware of her own cognitive abilities and limitations, as well as the varying degrees of involvement that a non-authority may have in an epistemic practice. This suggests that a certain level of epistemic humility is required of such authority, and that a lack of such humility can cause great epistemic harm. 164 IV. EPISTEMIC HUMILITY: TOWARDS VIRTUOUS EPISTEMIC AUTHORITY The Significance of our Epistemic Limits As epistemic agents who identify and interact with epistemic authorities on a regular basis, there is much that we can do to ensure that we do so in a responsible fashion. On my account, roughly, a non-authority is epistemically responsible so long as he, on good grounds, reflectively endorses the epistemic principles that regulate the domain of inquiry of which the relevant authority is an authority of, identifies an epistemic authority as such on good evidence, and so long as he does not fail to engage with the relevant epistemic inquiry where epistemically and pragmatically reasonable for him to do so (see Chapter III). Unsurprisingly, however, it is clear that even the most responsible of non-authorities in such authoritative relationships run the risk of being led far astray if the epistemic authority herself does not behave in an appropriate fashion. This is even true for those of us with the practical means to invest much time and energy into the relevant kind of epistemic safe-guarding. To some, this might suggest that reliance on, or deference to, such authority is inherently epistemically dangerous and irresponsible. Such a response, however, fails to account for the epistemic limitations of our real-life situation: we rely on such authorities for much of our epistemic standing, for a wide range of our beliefs, as sources of justification for those beliefs, and so forth. To suggest that we should not rely on such authorities, due to the risk that such reliance engenders, is to suggest that we cannot have the 165 kinds of epistemic goods that we generally take ourselves to possess. In such a direction lies scepticism: a kind that, although not as radical as external world scepticism, nevertheless severely undermines and limits our position as epistemic agents with any basic grasp of the world around us. What is to be done, then, about our situation? I suggest that we have to simply accept this fact about our epistemic situation—a contingent one, yes, but practically inescapable. This means that we ought to refuse the illusory comfort of hiding behind epistemic individualism. We can certainly, however, take steps to ensure that we are better equipped to identify when another epistemic agent is or is not worthy of being considered as an epistemic authority. Some have suggested methods for such discernment, at least when considering expertise if not epistemic authority per se.200 Such methods, however, focus on an individualistic dimension: they pertain to what we can, as individual epistemic agents situated in the world at large, do to protect ourselves and our own epistemic interests. Here I would like to instead focus on the epistemic authority herself, and the characteristics that such authorities should have in order to make the relationship between non-authority and authority as successful as it can be—which is partly to say what one agent can do to improve the epistemic lives of others. Here, then, is the place to discuss the kinds of virtues than an epistemic authority of a certain calibre ought to have. My suggestion will be that such authorities ought to be epistemically humble, where such humility is constituted by an awareness of one’s own cognitive limitations, specifically in relation to the cognitive abilities of other agents that one engages with, but also in relation to the broader epistemic practice that provides crucial content for both the authority and non-authority’s epistemic positions—the practice that one is immediately engaged with when conducting inquiry, See, for instance: Goldman, ‘Experts: Which Ones Should You Trust?’, and David Matheson, ‘Conflicting Experts and Dialectical Performance: Adjudication Heuristics for the Layperson’, Argumentation 19 (2005), 145-158. 200 166 or any other relevant epistemic task. In discussing such virtue, I aim to contribute to a fuller account of the kinds of characteristics that make engagement with an epistemic authority an epistemically fruitful practice. There are two reasons for doing this: first, such an account will take us one step closer to being able to more richly articulate a set of principles that we can all use to better discern when (and when not) to identify someone as an epistemic authority, and thusly treat them as such; secondly, such an account of virtuous epistemic authority can also provide those that are treated as epistemic authorities with the guidance required for them to be the kinds of epistemic authorities that our epistemic situation require. Of course, there would be nothing new in making the claim that our epistemic success, as individuals and as a species, rests on the epistemic actions of multiple parties engaged in our community’s epistemic practices—however, I hope to at least add to the broader social-epistemic project by providing further detail to one fundamental component of a successful social-epistemic structure. I think my account, in application, can provide a fruitful basis from which to better structure certain relationships of epistemic authority as well as a means to organise the very institutions that give birth to such relationships, and the particular interactions between non-authorities and authorities. Michel Croce (see II.4) has suggested that a virtue-theoretic approach to accounting for epistemic authority can make better sense of the phenomena. Though I disagree with his claim that certain virtues are necessary for one to satisfy the conditions of being an epistemic authority per se, my thought is that the account I provide here closely matches Croce’s own views on how the presence of certain virtues can make an authority a more appropriate source of certain epistemic goods, and that there is, beyond this, room to theorise about a most excellent, virtuous type of epistemic authority (such as what Croce refers to as a Supreme Epistemic Authority, who, for one, 167 possess the virtue of being wise).201 This chapter is partly intended to take us one step closer to understanding what such an authority would look like. In IV.1, I briefly discuss the role of epistemic humility on the part of the non-authority, to suggest that the virtue has an important role to play in all relevant parties. The choice to defer to an epistemic authority certainly requires an element of humility, and lack of intellectual arrogance. Given that, on my account, it is a condition on my establishing a relationship with an epistemic authority that I recognise her as having a superior level of skill, ability, or know-how, relevant to the epistemic practice at hand, it requires little argument to show that the possibility of my ever benefiting from such relationships requires that I am, minimally, humble enough to recognise such a fact about other agents. However, given my goals in discussing humility here, which are more squarely pointed towards the epistemic authority herself, my discussion on this topic is rather brief. In IV.2 I survey various accounts of epistemic humility, and defend the need to endorse an inter-personal account of such humility, as has been suggested by Maura Priest,202 who’s account I use as a basis for articulating the kind of virtue that I suggest an epistemic authority ought to have. Though it is not my primary goal, I hope my discussion in this chapter can lend further credence to, if not Priest’s view specifically, some version of an inter-personal account of such humility. In IV.3 I articulate how such humility can be epistemically beneficial in one or more of the following ways: i) it is sometimes required to ensure that a non-authority can acquire the epistemic goods he seeks in identifying an epistemic authority, ii) in cases in which certain evidence or reasons that an epistemic non-authority has is relevant, such humility is sometimes required to ensure that the relevant epistemic practice is fruitful, and iii) it is required to safeguard the non-authority’s epistemic autonomy. In IV.4 I finish by drawing lessons from 201 202 Croce, ‘Expert-Oriented Abilities’, 21. Maura Priest, ‘Intellectual Humility: An Interpersonal Account’, Ergo 4:16 (2017). 168 my observation on epistemic humility and authority, and suggest further avenues of research, in the hope that we can move further towards articulating a full account of virtuous epistemic authority. IV.1. EPISTEMIC HUMILITY IN DEFERENCE At first glance, one may wonder why I am discussing the humility of epistemic authorities, when the humility of any non-authority may be intuitively thought to be a more apt a target for evaluation. Many of us, I’m sure, have observed how a certain lack of humility can manifest itself as a hinderance towards the kinds of epistemic-authority-to-non-authority relationships that I have discussed here. We all know of arrogant individuals who refuse to accept the testimony of those who know better—those who can overwhelm the strength of any evidence proffered to them by the appropriate experts with their own overly-confident evaluations and reasoning. These individuals tend to be epistemic individualists par excellence, and refuse to believe that others may have better reasons for belief, better access to the relevant evidence, better suited to understand the intricacies of some domain of inquiry, or be epistemically better off in any sense at all. Their arrogance in the face of any sort of cognitive expertise, many of us would think, speaks of a severe lack of the kind of epistemic humility we may think is required of us in being good epistemic agents. We hear of such individuals frequently. We perhaps find them in the ranks of climate change deniers, refusing to accept the alleged authority of both the majority of climate scientists themselves, but also supposed ‘meta-experts’, i.e. those who explain the workings of the scientific process itself, why its results should be trusted, and the value of scientific expert testimony. We find them debating against the truth of evolutionary theories, providing their own conclusions drawn from the physical evidence at hand, but also providing their own 169 interpretation of various higher-order issues: such as the meaning of scientific notions—e.g. what a ‘theory’ is—and the evidential merit of certain scientific processes and methods. We may find them in political and economic debates, placing more weight on their (likely morally driven) intuitions rather than the conclusions of careful analysis and evidentially-based research. Such cases, however, by my own account, are not instances of a relationship holding between epistemic authorities and non-authorities at all. In such cases, the layman either considers himself the epistemic peer of the supposed authority, or perhaps even refuses to recognise the relevant domain of inquiry, or epistemic practice, as a legitimate enterprise. Such an individual does not even take the steps necessary to constitute the non-authority-to-authority relationship. Such individuals may be particularly arrogant in their decision to trust their own judgement more generally, rather than those of a broad epistemic community—choosing to trust their own reasons and the limited evidence that they have access to, over the broader swathe of community-sourced evidence, peer-reviewed analyses, and conclusions that a more humble approach would provide. They take their own epistemic methods to be superior to many, if not all, alternatives presented to them. Such arrogance clearly hinders the possibility of there ever being an epistemic-authority-to-non-authority relationship in the first place. It would seem that a condition on being able to benefit from relations with such authority is that I am sufficiently intellectually humble enough to recognise the limitations of my own cognitive abilities, access to evidence, background knowledge, etc. To elaborate, let us see how a lack of epistemic humility on the part of a purported nonauthority could hinder the functioning of any relationship with an epistemic authority. As I see it, there are three possible ways in which an individual S could, in the face of some potential epistemic authority S*, fail to act in accordance with the role of non-authority that they (supposedly) should adopt, due to a lack of epistemic humility: 170 i. S may refuse to recognise the relevant epistemic practice (EP) as a good epistemic practice, but nevertheless recognise that S* has (relative to S) more of the requisite skills, abilities, and know-how to partake in it. ii. S may recognise EP as a good epistemic practice, but may refuse to recognise S* as having (relative to S) more of the requisite skills, abilities, and know-how to partake in it. iii. S may recognise both that EP is a good epistemic practice, and that S* has (relative to S) more of the requisite skills, abilities, and know-how to partake in it, and yet refuse S*’s epistemic conclusions. By my account, neither (i) nor (ii) count as cases in which S and S* have established the relevant kind of relationship in which S* is an epistemic authority. In these cases, S simply does not recognise as S* as an epistemic authority. Notice also that the descriptions in (i) and (ii) are completely compatible with the idea that S has behaved epistemically appropriately: many epistemic practices are bad ones, and there are many individuals that do not have the requisite skills, abilities, and know-how to partake in various epistemic practices, good or bad. Nevertheless, we are here concerned with epistemic humility, or the lack thereof. What I have in mind specifically here are cases in which (i) or (ii) obtain due to S’s epistemic arrogance. From the outside, so to speak, we would criticise S for not recognising the EP a good one, or for refusing to think that S* is better suited for engaging with it. These are the kinds of cases already discussed above: cases in which S arrogantly assumes that well-established epistemic practices, that have a vast amount of evidence in their favour, are not valuable, or at least less valuable than S’s own practices and methods; or, cases in which S, though believing the relevant epistemic practice as a good one, nevertheless arrogantly assumes that he is better-equipped to make use of it than S* (who, as a matter of fact is actually better-equipped). These may be cases in which the individual 171 in question endorses a strong kind of epistemic individualism, or are ‘epistemic egotists’, to use Richard Foley’s term.203 Such individuals may be an apt target for the charge of intellectual arrogance, and may be prescribed a dose of intellectual humility. Alternatively, we may simply think these individuals are basically irrational, for not properly endorsing either the relevant practice, or the relevant individual qua epistemic authority, based on the evidence that they have access to. Perhaps they ignore the plethora of evidence in support of evolutionary biology, for instance. Cases that fall under (i) and (ii) suggest two things to me: a) That a degree of humility is required for us to be able to engage with experts/epistemic authorities qua experts/epistemic authorities in the first place. b) That we ought to be epistemically humble, and aware of our own epistemic standing relative to the epistemic standing of other agents in our societies. I assume (a) is quite uncontroversial. I think that (b) clearly follows for anyone who thinks that epistemically engaging with an epistemic authority, or anyone for that matter, and deferring to such agents, is a good thing, all things considered. Really, the claim here is that we ought to simply be humble enough to accept that we are better off under certain structures of the division of cognitive labour. I don’t think it necessary to exert too much effort defending such a claim here. The more interesting cases, when considered in relation to epistemic authority, may be the kind identified in (iii), above. These are cases in which the epistemic-authority-to-nonauthority relationship is established, and yet the non-authority involved acts arrogantly in the 203 Foley, Intellectual Trust in Oneself, 87. 172 face of authoritative utterances. It is somewhat difficult to imagine what such a case would really look like, however, without immediately painting the imagine non-authority as plainly irrational from the start. But something like the following may do the trick. Consider: there is an individual, let’s call him Samuel, who values the results of psychological research, and who recognises that others have a privileged position in conducting such research and in deriving the appropriate conclusions from it. Specifically, consider the broad body of psychological research that suggests the humans are generally flawed epistemic agents: we fall foul to various biases and errors of reasoning; we allow our moral intuitions to interfere with our ability to properly and rationally evaluate the non-moral facts of various situations;204 we are motivated and influenced by factors that we are often not even consciously aware of;205 perhaps we even present arguments to others, not as a means to reach the truth, or to find the best-supported view, but rather to simply be on the ‘winning’ side of any debate.206 Our arrogant individual, Samuel, who establishes a relationship with an epistemic authority— let’s call her Samantha—recognises her as having the requisite skills, abilities, and know-how to participate in the domain of inquiry in question, which, again, Samuel recognises as a source of some epistemic good or goods.207 Nevertheless, in the case that Samantha applies these results to Samuel himself, by suggesting that his own reasoning falls prey to the same biases and issues, Samuel refuses to accept the results. If we stipulate that Samuel’s denial of the results is selective—i.e. he accepts many, if not most of Samantha’s conclusions, just not those pertaining to himself, and Samantha gives him no reason to think of himself as an exception—the case See, for example: Joshua Knobe, ‘Person as scientist, person as moralist’, Behavioral and Brain Sciences 33 (2010), 315-365. 205 Anyone sceptical of this claim should consider the depth of psychological research that goes in to consumer behaviour, from the layout of your local supermarket, to the presence of ads on the sides of your internet browser. 206 Hugo Mercier and Dan Sperber, ‘Why do humans reason? Arguments for an argumentative theory’, Behavioural and Brain Sciences 34 (2011), 57-111. 207 See my account of epistemic authority, as provided in Chapter II. 204 173 presents itself at least as a plausible example.208 I think it quite accurate to call Samuel intellectually arrogant in this case. However, I think that there is something far worse here than Samuel’s lack of humility: I would argue that he is being straightforwardly irrational. Why? Because his selectively choosing between Samantha’s conclusions is not consistent with the fact that he both recognises her expertise209 in relation to the relevant epistemic practice (where such expertise is one that he lacks) and the value of the epistemic practice itself. To decide to disregard certain conclusions, but not others, would be to wilfully disregard his own beliefs about Samantha’s abilities and superior epistemic position in certain situations, while accepting it in others. If one establishes the appropriate relationship with an epistemic authority, then one cannot reasonably cherry-pick which conclusions to accept, and which not: part of recognising someone as having such authority is realising that they are better off than you in evaluating which claims ought to be accepted, and which not. This is all to say that there is clearly a role for epistemic humility on the non-authority’s side of things. However, I do not think that such humility, or the opposing trait of arrogance, can be interestingly illuminated by discussing epistemic authority, merely beyond stating that an epistemic arrogance that obstructs an individual from recognising another as an authority (where appropriate) is an epistemic detriment to them, given that we acquire so many of our epistemic goods by engaging with such authorities qua authorities. Though intellectual humility is certainly an important component to any epistemically responsible behaviour in this regard, my Consider the real-life example of Donald J. Trump, who, as President of the United States on more than one occasion publicly declared that he disagreed with various conclusions made by US intelligence agencies (such as the F.B.I., C.I.A., N.S.A., etc.) when such conclusions contradicted his own beliefs. When asked whether he had lost faith in these agencies, Trump had nothing but praise for them and their work. See: ‘Trump insists he is on the same page as intel chiefs after insulting them’, CNN, January 31st, 2019, https://www.cnn.com/ 2019/01/31/politics/trump-intelligence-agencies-coats-haspel/index.html. Of course, this case is likely a far cry from the actual kind of case I am considering, given that I doubt that Trump himself accepts the epistemic authority the relevant individuals—however, the case can at least illustrate the kind of selective acceptance that one may identify with the kind of epistemic arrogance in question. 209 i.e. having more of the requisite skills, abilities, and know-how to partake in the relevant epistemic practice. 208 174 observations here are not likely to add much to what has already been said about the issue. The question of epistemic humility on the side of the (supposed) non-authority is part of a much broader question that has to do with epistemic individualism, and the value of making use of social-epistemic relationships in general, rather than one specific to cases of epistemic authority. I leave this question for now, though it should be quite obvious to my reader where I, as a social epistemologist, stand. That being said, consideration of the non-authority and his epistemic humility, or lack thereof, does suggest something of relevance for what is to follow: the kind of humility I suggest a non-authority is required to have, in order to benefit from a relationship with an epistemic authority, is one that is not solely concerned with an evaluation of one’s own epistemic competence. Rather, epistemic humility in this sense is outwardly directed, at least in part, in that it is concerned with the epistemic competence of other epistemic agents. As we will see, I, in agreement with a certain account of humility, think this is crucial to the notion. For the rest of this chapter, however, given my overall aims, I will primarily focus on the role intellectual humility may play in defining the behaviour of the epistemic authority, rather than the non-authority. Here I think far more interesting results await. IV.2. ACCOUNTS OF HUMILITY Before we can begin to discuss epistemic humility’s role in epistemic authority, it is necessary to articulate an account of what such a virtue would look like. It should be noted, however, that this chapter is not intended as a defence of any one account of such humility. I do think that an analysis of epistemic authority can provide us with insight into what an account epistemic humility ought to include, and in fact argue that such analysis provides us with reason 175 to modify the other accounts here discussed. However, my primary aim is to identify an intellectual virtue that I think central to making an epistemic authority a good epistemic authority, a virtue that should be nurtured in society at large in order to improve our collective epistemic position. For this purpose, I choose an account of epistemic humility that best suits the kind of phenomenon I have in mind—namely Maura Priest’s inter-personal account of intellectual humility—and both argue that consideration of the notion of epistemic authority provides further support for this account. At first glance, I think of this virtue as entailing an awareness of one’s own intellectual or epistemic limitations, a recognition which leaves room open to recognise others as being able to fill the space beyond our own limits. As my earlier discussion probably made clear, I also think that this virtue stands opposite a parallel vice: epistemic arrogance, which I think of as an excessive confidence in one’s own cognitive abilities, but also as a involving some kind of disregard for the epistemic competencies of other agents. As the brief discussion in IV.1 illustrated, for example, the intellectually arrogant individual who refuses to recognise another as an epistemic authority, does so partly because they refuse to acknowledge that other agent’s particular epistemic skill set. Given that I will ultimately endorse Priest’s interpersonal account of epistemic humility, I start here by categorising accounts of epistemic humility into two broad categories, following her terminology: A. Self-assessment accounts of epistemic humility B. Interpersonal assessment accounts of epistemic humility A self-assessment account of epistemic humility explains the virtue in question by identifying the manner in which the intellectually humble evaluate certain aspects of their own cognitive and intellectual lives. These views suggest that epistemic humility is a kind of ‘Personal Virtue’, which 176 Priest defines in the following way: ‘If virtue V is a personal virtue, then V can be adequately described while referencing the virtue holder alone’. Such virtue is contrasted to the kind suggested by accounts falling in (B), which speak of an ‘Interpersonal Virtue’: a virtue that can ‘only be adequately described with reference to agents other than the virtue holder’.210 I suggest that we need to endorse some kind of view that falls under (B) in order to make sense of the role of (virtuous) epistemic authority. Roughly, this is because the humility I have in mind is specifically concerned with the epistemic role and position of both the epistemic authority and the non-authority from the perspective of the epistemic authority, as well as the authority’s relationship with a shared epistemic practice, or an epistemic community as a whole. If epistemic humility is merely a virtue concerned with one’s self-assessment, without reference to anything or anyone else, then it is not obviously clear how an epistemic authority’s intellectual humility can be of direct relevance to a non-authority.211 Those strongly wedded to a self-assessment view of humility may simply think this is reason to deny that the virtue I have in mind is epistemic humility. I think, however, that an interpersonal account of humility can both make sense of the phenomena I have in mind, while matching our intuitions about humility more broadly. Furthermore, as Priest argues, such an account seems to have the added benefit of avoiding the problems that affect self-assessment accounts. Let us begin by taking a look at the kinds of views that fall under category (A): selfassessment accounts of epistemic humility. Dennis Whitcomb, Heather Battaly, Jason Baehr, and Daniel Howard-Snyder, discuss three other kinds of views that would fall under the same Priest, ‘Intellectual Humility’, 468. Though this is no to suggest that it cannot be relevant at all. Clearly, recognising an expert as humble, in the ‘personal virtue’ sense, may help us identify them as more epistemically conscientious and capable over all. My point is just that this kind of humility doesn’t necessarily say anything about the relationship such authority has to nonauthority. As I will argue, this is an important element to consider. 210 211 177 category, before presenting their own fourth alternative:212 i. Proper belief views ii. Underestimation of strength views iii. Low concern views iv. Limitations-owning view (Whitcomb et al.) Proper belief views suggest that to be epistemically humble is to either: a) have a ‘proper’ belief, where to have a proper belief is to have a belief with the ‘firmness the given belief merits’,213 or b) have a proper higher-order belief or attitude about one’s beliefs or other doxastic attitudes, and not to ‘overestimate the epistemic status of [one’s] doxastic attitudes’.214 Such views, I assume, are intended to capture the way in which an epistemically humble individual fails to hold beliefs (or higher-order attitudes about those beliefs) that go beyond what is reasonable. To have an improper belief in this way would be to arrogant, in the sense that one would hold especially strong opinions beyond what is warranted, or to have beliefs about one’s doxastic attitudes being of more epistemic merit, than is appropriate. Though these are certainly consequences of epistemic humility, I’m not sure that focusing on these elements really gets at the heart of what it is to be humble per se. Someone who holds ‘proper’ beliefs in this way strikes me as someone who simply manifests the virtue of holding beliefs in accordance with the evidence (including higher-order evidence). More importantly, proper belief views of epistemic humility do not have much to say about the kind of humility I suggest that we ought to seek in our choice of epistemic authority. Certainly, we want our epistemic authorities to have ‘proper’ beliefs, in both of the senses Dennis Whitcomb, Heather Battaly, Jason Baehr, and Daniel Howard-Snyder, ‘Intellectual Humility: Owning Our Limitations’, Philosophy and Phenomenological Research 94:3 (2017), 514. 213 Peter L. Samuelson, Ian M. Church, Matthew Jarvinen, Thomas Paulus, ‘The Science of Intellectual Humility White Paper’, unpublished manuscript, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.370.6006&rep =rep1&type=pdf. 214 Allan Hazlett, ‘Higher-Order Epistemic Attitudes and Intellectual Humility’, Episteme 9:3 (2012), 220. 212 178 discussed above, but this, it seems to me, is simply a consequence of their being skilled and able to partake in the relevant epistemic practice in question. If we found out that an identified authority does not hold proper beliefs, we have reason to reverse the decision of identifying them as such. Furthermore, I am concerned here with epistemic authorities who can provide us with a vast array of epistemic goods, many of which are not merely true beliefs. If an authority is to provide me with understanding, to teach me a certain methodology, or, in some cases, collaborate with me in some joint epistemic project, their having ‘proper’ beliefs in no way guarantees that they will have the right kinds of attitudes towards my epistemic position to guarantee that this relationship goes smoothly. It would seem that the kind of humility we seek in these relationships pertains more so to the authority’s attitude towards a consideration of my doxastic states, my abilities, my epistemic needs and goals, etc. Underestimation of strength views suggest that to epistemically humble is to have a low estimation of one’s own epistemic abilities.215 Julia Driver, for instance, argues that a humble person may be one who is disposed to underestimate her own self-worth, going so far as to suggest that the humble person even does this in the face of evidence to the contrary.216 My problem here is simple: I consider epistemic humility as a virtue, something desirable in an epistemic agent, a trait to be sought out by all of us in seeking the epistemic ideal; yet, the view here suggests something defective about the agent in question—they get things wrong when it comes to holding doxastic attitudes about their own epistemic status. In identifying an epistemic authority, I identify her as an individual epistemically superior to myself, and rely on her to properly conduct herself in certain inquiries, in guiding my epistemic hand, etc. This requires that I trust her to have an appropriate level of confidence in her own epistemic position. If she were to 215 216 Ibid., 512. Julia Driver, Uneasy Virtue (New York: Cambridge University Press, 2001), 114. 179 instead underestimate herself, this would make her less desirable in comparison to an alternative choice of authority: namely one who correctly evaluates her own epistemic strengths, skills, etc.217 In this way the proper belief views of epistemic humility at least seems relatively more promising. Furthermore, even if such a fault could be paired with other conditions to result in a favourable epistemic end,218 it seems that such underestimation is neither necessary nor sufficient for humility, as Whitcomb et al. have pointed out: First, consider someone who is consistently motivated to get epistemic goods and, as a result, is both aware of her intellectual limitations and disposed to respond appropriately to them, e.g. caring how they affect her beliefs and inquiries, admitting them to herself and others when appropriate, regretting them, doing something to change them if she can and when appropriate, etc. She strikes us as an exemplar of [intellectual humility] even if she also accurately estimates her intellectual strengths. Underestimation is not necessary. Second, imagine someone who is disposed to underestimate his strengths, and so on while also being clueless about his intellectual limitations, or being disposed to respond to them inappropriately, e.g. to not care about them, or to be hostile or defensive about criticism of them, etc. Underestimation is not sufficient.219 How about Low concern views? As summarised by Whitcomb et al., these views suggest that to be epistemically humble is to have an ‘unusually low concern for one’s own intellectual status and entitlements’.220 Here, though the relevant kind of assessment is still self-directed, we can see the first hints of an interpersonal element sneaking in to an account of humility. Robert C. Roberts and W. Jay Wood state that humility involves ‘an unusually low dispositional concern for the kind of self-importance that accrues to persons who are viewed by their intellectual Perhaps some underestimation is healthy, but this is just to say that some doubt, when it comes to fallible modes of inquiry, is desirable. 218 Perhaps being wrong about your own epistemic worthiness can lead you to more epistemic goods than being right about it. . . . 219 Whitcomb et al., ‘Intellectual Humility’, 514-515. 220 Whitcomb et al., ‘Intellectual Humility’, 514. Here the authors refer to the views espoused in: Robert C. Roberts and W. Jay Wood, ‘Humility and Epistemic Goods’, Intellectual Virtues: Perspectives From Ethics and Epistemology, ed. Linda Zagzebski and Michael DePaul (New York: Oxford University Press, 2003), 257-279. 217 180 communities as talented, accomplished, and skilled’.221 I think there is something right in this: intellectual humility ought to be understood in relation to the intellectual community that the individual in question is situated within. However, it is not obvious to me that such a lack of concern for status is sufficient for intellectual humility. As a starting point, let us consider Whitcomb et al.’s counter-example case of Professor P: [He] is an extremely talented philosopher who knows he’s extremely talented. He genuinely loves epistemic goods; indeed, his obsession with them drowns out any concern he might have otherwise had for status or entitlement. He simply doesn’t care about impressing others, nor does he take himself to be entitled to special treatment or to disrespecting others. Status entitlement aren’t even on his radar. While extremely talented, Professor P is not perfect. When confronted with his intellectual imperfections or mistakes, his default response is to try to justify, cover up, or explain them away. He is notoriously bad at admitting when he has made a mistake or when one of his argument is vulnerable to serious criticism. Professor P seems to be lacking in [intellectual humility] even though he is disposed to an unusually low concern for status and entitlement.222 I think this is somewhat right: though such individuals may be rare indeed, one can imagine individuals who are motivated purely by the goal of acquiring various epistemic goods while having no interest in their own personal status, and yet continue to both show an overestimation of their own strengths, and underestimation of their weaknesses, while also (and I think importantly) showing a lack of respect for others.223 Such cases show that epistemic arrogance can be aimed at some other goal than mere status and personal gain—even noble goals. Furthermore, it would seem that such arrogance need not be goal-oriented at all: I think it not hard to imagine that one could simply be intellectual arrogant, cock-sure, dismissive of others, without care for status, epistemic ends, or any other end. That is not to say that status cannot play a central role in Roberts and Wood, ‘Humility and Epistemic Goods’, 250. Whitcomb et al., ‘Intellectual Humility’, 515. 223 Priest thinks that Whitcomb et al.’s counter-example is implausible, but her criticism seems to be based on an element of the counter-example that is not part of the final, published paper: ‘Professor P, while having no concerns for his intellectual status, nonetheless “constantly references his own strengths, accomplishments, and publications . . . seizes every opportunity to relate what others say to his own projects and theories . . . [and] is oblivious to his intellectual limitations’. See: Priest, ‘Intellectual Humility’, 466. 221 222 181 making sense of such arrogance, but I shall return to this point shortly. Lastly, let us consider Whitcomb et al.’s own version of a self-assessment account of epistemic humility. Limitations-Owning View of Intellectual Humility: ‘[Intellectual Humility] consists in proper attentiveness to, and owning of, one’s intellectual limitations.224 This account clearly lands close to the crude definition of intellectual humility I began with. Here we have an account of such humility that focuses specifically on the idea that the epistemically humble pay special attention to the limits of their epistemic reach. This is certainly on the right track to make sense of the kind of humility I have in mind: we want our epistemic authorities to know where their epistemic reach ends, what they can speak of assuredly, and what they cannot. Specifically, however, we want our authorities in many cases to be aware of the limitations of their own epistemic role while relatively being aware of our own. As such, this view fails to account for the properly interpersonal element that I think crucial. Priest has also criticised the view on the following grounds: intellectual humility ‘seems to demand much more than just limitation owning’.225 Consider Priest’s own counter-example for illustration: [L]et us imagine a professor who is acutely aware of his own limitations. He also justifiably believes that he is better than most of his students in physics. With this realization in mind, he looks down on them with contempt as his intellectual inferiors. Even when they understand, he lectures patronizingly making sure they recognize his superiority. He acts this way not only toward his students but to all whom he justifiably believes to have less intellectual acumen. Additionally, he jumps at every opportunity to mention his success and prominently displays his awards and accomplishments where and whenever he can. [. . .] It is counterintuitive to think that such a professor is intellectually humble (to say the least).226 Whitcomb et al., ‘Intellectual Humility’, 520. Priest, ‘Intellectual Humility’, 467. 226 Ibid. 224 225 182 Now, it could be argued that Priest’s example confuses the issue by incorporating elements that may speak of a general arrogance, or lack of humility more broadly speaking, rather than epistemic arrogance or humility specifically (though it is not absurd to think that one goes with the other more often than not!). The display of awards and accomplishments, for instance, does not strike me as specifically epistemic in any way, though the awards and accomplishments themselves may pertain to activities with a strongly epistemic component.227 Perhaps this behaviour may be explained by reference to personal pride, and, to return to a view previously mentioned, an excessive concern with status. That being said, I think the case points to an important component: the unnecessarily negative evaluation of, and disdain for, those the arrogant individual considers to be epistemically inferior. As will not surprise my reader, this is the component that I think is of fundamental import in the context of epistemic authority. I now turn to this, through the lens of Priest’s own account of intellectual humility. Priest suggests the following view better explains the above case of the professor: Interpersonal Assessment Intellectual Humility: Agents are intellectually humble just in case interactions with members of their epistemic community creates a tendency to reflect on their own epistemic limitations.228 At the core, I think this view is correct: at the very least, it identifies key components of the phenomenon I have in mind when discussing epistemic authority. Let us take a closer look. Priest argues that her view can make sense of the various intuitions that drive selfassessment views of intellectual humility, while avoiding the problems these views come with. Priest believes that the epistemically humble lack ‘abrasive overconfidence’, for instance, but that I do not mean this as a criticism. My own discussion of epistemic humility in relation to authority does not clearly remain within the scope of purely epistemic considerations. I do not think this is a bad thing: clearly, the boundary between epistemology and ethics will get fuzzy here, as we are dealing with interpersonal relationships, and the question of what we can do for each other. My point is simply that it may not be particularly fair to criticise a view of intellectual or epistemic humility on these moral grounds. 228 Ibid., 475. 227 183 ‘they lack this overconfidence specifically because it is incompatible with other-directed respect’;229 they also own their own limitations, as Whitcomb et al. suggest, but ‘this is because they take advice and criticism seriously, not the other way around’;230 they do not feel a sense of entitlement to accept certain intellectual advantages over those that they deem as being epistemically inferior.231 Priest’s views here map nicely on to the kind of other-directed respect that I think we desire in an epistemic authority. If I seek the help of a medical practitioner, whom I identify as my epistemic authority on medical issues, I may desire that she is epistemically humble in the sense that she respects my own epistemic position: as someone who is trying to make sense of my own autonomous choice on some medical matter based on both the medical facts of the case, but also on practical considerations and preferences that I express; as someone trying to make sense of how certain first-personal evidence (e.g. my experience of the symptoms, medicinal sideeffects, etc.) relates to the epistemic practice of medical inquiry; as someone seeking guidance on various higher-order issues, such as the question of how much evidential weight I should place on the results of certain tests, my own feelings, future-directed preferences; etc. If I adopt the position of a pupil, identifying an advisor or teacher as my epistemic authority in relation to some domain of inquiry, I may seek similar humility: in learning how to navigate and partake in a certain epistemic practice myself, I desire that an authority both tells me how to do so, authoritatively correcting my epistemic behaviour when necessary, but always with a level of respect for my own epistemic position—my own skills, abilities, and know-how, even with a sense of respect for my future epistemic position, i.e. not feeling entitled to abuse my intellectual inferiority in the present, for the sake of improving my position in the future. Ibid., 470. Emphasis my own. Ibid. 231 Ibid., 469. 229 230 184 Central to these considerations, I suggest, is the authority’s awareness of the fact that she is a practitioner of an epistemic practice, a practice that is ultimately responsible for the production of various epistemic goods, that she is merely fortunate enough to be a participant of. Though the skills, abilities, and know-how required to participate in this practice are ones that not everyone may have, this does not mean that she is the ultimate source of the produced epistemic goods—the practice itself is the root such valuable results—and thus these skills, abilities, and know-how do not provide her with a sense of arrogant entitlement to behave as if the goods were solely her product. This means that the humble epistemic authority is aware that she is a member of an epistemic community, the division of cognitive labour it involves, and the fact that other epistemic agents have contributed to the relevant inquiry that she participates in. This is precisely what Priest identifies in her account of epistemic humility. Importantly, the notion of epistemic community in question here is not a narrow one—it is not limited to other similarly able practitioners of the same practice for instance: Priest specifically points out that the intellectually humble take the ideas and criticism of their epistemic inferiors ‘seriously’, respect other epistemic agents of various levels ability, and refuse special intellectual treatment in relation to their inferiors.232 We should be careful, however, in understanding precisely some of the key notions referred to above. What do we mean when we say that an authority ‘respects’ her epistemic inferiors, or that an authority takes their ideas and criticisms ‘seriously’? To unpack these ideas, let us take a closer look at what Priest says about the intellectually humble: [A] person is intellectually humble just in case he: • Respects the intellect of others as his own, and so rarely feels immune to their complaints and criticisms. 232 Ibid., 469-470. 185 • Systematically declines intellectual advantages in interpersonal relations because he feels no sense of entitlement. Persons who meet these criteria tend to behave in ways that signify their intellectual humility. For instance, the intellectually humble: • Rarely demand special intellectual treatment, even when deserving. • Often refuse special intellectual treatment, even when deserving. • Tend to take complaints and criticisms seriously, even when the criticizers are not authority figures and even when the criticism is rude. • Tend to take the ideas (which are not always complaints) of others seriously, even the ideas of intellectual inferiors.233 These interpersonal conditions give way to the kinds of behaviour that the self-assessment views of epistemic humility focus on: ‘[k]nowing that others disagree with her, she can rarely in good faith overestimate the epistemic appeal of her own views’,234 even ‘in instances when the humble agent is quite sure of her beliefs, she does not flaunt this assurance’.235 When considering epistemic authority specifically, some of these statements may raise concerns: one may worry that Priest’s account is suggesting that an authority underestimate her epistemic position because of the beliefs or criticisms of a non-authority (or other epistemically inferior agent). However, this is precisely why we have to be careful in understanding what is entitled by saying that a humble agent takes others’ views ‘seriously’. Priest is not defending the view that a humble agent, or epistemic authority specifically, should doubt herself due to such views, criticisms, etc. She clarifies her view in the following terms: Thoughts like this are not examples of the humble agent doubting her beliefs, but rather examples of the humble agent doubting that her own intellectual brilliance made arriving at those beliefs inevitable. The humble agent’s respect for reasonable and intelligent others tempers not her enthusiasm in p, but rather her enthusiasm in ‘look how special I am to believe p’.236 Here again we can make reference to the notion of an intellectual community: the humble agent Ibid. Ibid., 470. 235 Ibid. 236 Ibid. 233 234 186 does not consider herself entitled to distinguish herself from the community as a whole, she is not special in the sense that she somehow sits outside of the broader community’s overall epistemic workings and history. She recognises herself as part of it, and the epistemic goods that she is privy to as goods that are a result of a broader process and history, and, furthermore, recognises that these goods are goods for all members of the community. What is special, what matters, is the epistemic result (in Priest’s case, a belief in p), which is derived from a certain epistemic practice, not the fact that I accessed or discovered it.237 Thinking of the humble in such a manner can also help us best read Priest’s suggestion that the humble ‘respect the intellect of others as [their] own’.238 At first glance such a suggestion may seem puzzling, particularly when considering the epistemic authority: surely, we may ask, the authority recognises that she is intellectually better in some important sense?239 I think the two thoughts are compatible, however: the epistemically humble respect others’ intellects as their own qua members of the same community broadly aimed at the same goals, both epistemic and non-. To respect another’s intellect as one’s own is not to literally identify the two intellects in question as having the same skills, as both being on par in regard to having a certain kind of expertise in a certain domain of inquiry, as both sharing in the same level of intelligence or epistemic rigour, or anything of the like. Rather, the sense of respect in question entails recognising that others too are of a similar kind: we are epistemic agents, with similar practical needs, seeking out certain epistemic goods, and members of the same, broad epistemic community. Perhaps it is to go so far Of course, this is surely something worthy of praise, and it is quite right that we generally honour and credit those that do such work. The point, however, is that these values are secondary, derivative of the value that is attached to the epistemic good itself. 238 Priest, ‘Intellectual Humility’, 469. 239 Recall that I disagree with Zagzebski, who thinks that epistemic authorities are simply agents who are like us in most important respects, but simply are more conscientious about a certain domain of inquiry (because they have more time, more means, etc., to do so). I think in many cases authorities may have a special skillset that is a result of substantial intellectual differences, whether by nature or nurture, that cannot be merely reduced to having the means to being more conscientious about certain kinds of inquiry. Thus, I think we need to leave room for the idea that (some) epistemic authorities are simply epistemically better than us in more absolute terms. See Chapter I and II for more. 237 187 as to suggest that others deserve to also partake in the epistemic fruits of one’s own labour, or at least to be given the opportunity to autonomously choose to reach for such fruit. To repeat Priest’s point somewhat differently, the suggestion here is that the epistemically humble resist a certain kind of egotistical thought: they remove the ‘I’ from the equation, focused on epistemic community, a certain epistemic practice, and the epistemic goods that result. Here, the epistemic community may arguably be something that the epistemically humble value for its own sake.240 My hope is that the above discussion provides at least an illustration of the kind of epistemically virtuous behaviour that I think beneficial in the epistemic-authority-to-nonauthority relationship. This was my primary goal. Though it is not of fundamental important to the project of this chapter (or this dissertation), I also hope that my discussion here has provided further support for Priest’s account of epistemic humility, or, at least further support for the idea that epistemic humility should be considered in interpersonal rather than personal terms. I will leave it to my reader to decide if I have succeeded in doing just that. IV.3. AGAINST ARROGANCE: THE BENEFITS OF EPISTEMIC HUMILITY With an account of epistemic humility in view, we are now in position to further discuss the use of such virtue in the epistemic-authority-to-non-authority relationship. Though it is probably fairly clear to my reader, I will begin by further elaborating on why such a virtue can play an important role in this relationship, and why, given its benefit to the consequences of such a relationship, it ought to be sought out—both in the sense that a non-authority ought to seek out Though it is beside the point here, I think thinking of humility in such a fashion makes sense of a certain common kind of action that we tend to describe as being humble, such as when the recipient of some prize or honour takes the time to thank all those in their community that have directly or indirectly helped them. The humblest of individuals go so far as to publicly declare that they do not think themselves as special, but rather lucky enough to be the individuals that were in the right position to do the work for the community at large. 240 188 virtuous authorities, but also that authorities should foster such a virtue. Furthermore, I suggest that a lack of epistemic humility on the part of the epistemic authority can undermine the epistemic autonomy of the relative non-authority. As previously noted, I do this for the following two reasons: first, doing so will take us one step closer to being able to more richly articulate a set of principles with which we can all better discern when (and when not) to identify someone as an epistemic authority, and thusly treat them as such; secondly, doing so will also provide those that are treated as epistemic authorities with an, albeit incomplete, set of guidelines for what is required of them in order to be the best kinds of epistemic authorities that our epistemic situation requires (both in the sense of the epistemic situation that we individually find ourselves in, but also in the sense of the epistemic situation our epistemic community finds itself at large). To restate my views on epistemic authority (as discussed in Chapter II), recall that I consider an epistemic authority broadly as an individual that, other conditions being met, can provide us, as relative non-authorities, with certain epistemic goods. These epistemic goods can range from beliefs, to other propositional attitudes such as acceptance and hypothesis, to more complex states such as understanding, and even more complicated goods, such as the possibility of collaborating in an epistemic project, and the goods that we could think of as being born of the pedagogical context: where the epistemic authority can act as a guide, facilitating the nonauthority’s development to the status of an epistemic peer. As I see it, this incomplete list of epistemic results can be placed on a continuum, a scale with certain goods that require little extra work on the side of the epistemic authority on the one end, and far more demanding results that require much more of the epistemic authority on the other end (at least if the relationship is to bear the best fruit it can). Here my view echoes the hierarchy of epistemic authority that I 189 understand Michel Croce to be suggesting,241 with what he calls an epistemic authority of belief on the one end, an epistemic authority of understanding somewhere above this, with the final, ultimate position going to what he refers to as a Supreme Epistemic Authority.242 Though we disagree on precisely what is required of such authority figures, when speaking in terms of their virtues,243 I generally agree with his assessment of this growing complexity. Those of us who simply wish to find out what to believe about a given matter in a domain of inquiry (one opaque to use due to a lack of expertise) need not be too concerned about the various virtues that an identified epistemic authority may possess, beyond those required for her (the authority) to have the requisite skills, abilities, and know-how to partake in the epistemic practice in question. When we seek understanding about a certain issue, however, things are clearly more complicated: we are best served in this instance by an authority that does not arrogantly ignore or disqualify our own epistemic standing, our views, beliefs, intuitions, etc. The same can be said in the case that I seek to understand why my own beliefs, or the evidence that I have at hand, does or does not justify a certain conclusion in a certain domain of inquiry. In the pedagogical context, it is difficult to escape the intuition that a lack of the aforementioned intellectual arrogance is a necessary requirement for the epistemic-authority-to-non-authority relationship being epistemically valuable, and fruitful. I do not think, for instance, that it is an accident that Priest ended up using the example of the professor to illustrate her interpersonal account of epistemic humility. And here is the crux of the matter: to gain these more complicated epistemic results from a relationship with an epistemic authority in these cases, we need the authority to be epistemically humble in her dealings with us. A lack of such humility can, at best, leave us with fewer of the See Chapter II. Michel Croce, ‘Expert-Oriented Abilities’, 21. 243 Most importantly in that I do not think that an epistemic-authority-to-non-authority relationship in which the non-authority is seeking mere beliefs requires that the authority in question possesses what Croce refers to as a sensitivity to the non-authority’s epistemic needs. See Chapter II. 241 242 190 epistemic goods than we could ideally acquire from a relationship with such an authority, and, at worse, could leave us worse off than if we were to have not identified her as our epistemic authority, and not treated her as such. How could we be worse off? I think this could happen in a number of ways. Consider the case that I seek understanding of the relationship between my own reasons, the beliefs they suggest, and the actual fact of the matter in some domain of inquiry. I do this because I not only want to know the answers to certain questions I currently have, that a part of the domain of inquiry in question, but also want to be able to better evaluate my own epistemic position with regard to future questions I may come across. Alternatively, I may seek such understanding simply because I have a hard time accepting the results of the relevant epistemic inquiry— because they are counter-intuitive, for example, or clash with the results of a different domain of inquiry that I take seriously. One can imagine that a bad choice of epistemic authority in such a situation—viz. in which we choose an epistemically arrogant individual—could leave me worse off. The arrogant authority in this case may authoritatively tell me that my reasons are bad reasons, insensitively telling me that I should completely drop the idea of thinking of a particular question in a certain way, but without providing me with any guidance or further explanation: as to how and why my reasons ought to be discarded, why they don’t hold weight in the relevant epistemic practice, or what kind of reasons I should entertain instead. On his authority, I may adjust my epistemic behaviour in the relevant way, and thus leave myself, at least diachronically speaking, in an epistemically worse off position. Let me illustrate with the following tale: Dr. Hoggs: I, knowing no better, am relying on my intuitions about every-day sized objects to make sense of various claims about sub-atomic particles. I already have a vague sense that my reasoning has gone off-course, but, being a complete novice in the relevant domain of inquiry, I have no idea how to rectify this. Given that I would like to avoid making too many mistakes in in 191 the same way in the future, I seek out an epistemic authority, not just to acquire the right answers to the questions that I am considering, but to also get a sense of where I stand—in regard to my reasons, reasoning, and intuitions—in relation to the relevant epistemic practice. For this reason, I find a famous physicist at my local university. His research is focused on various sub-atomic issues, he is credentialed, he has received more awards than one would care to count, and he seems to be well-respected by other physicists. His name is Dr. Hoggs. When I meet him, however, explaining my position, he agrees to correct my errors. However, when I ask my questions, illustrate my line of reasoning, and generally share with him my epistemic perspective on various physics-related questions, he quickly begins to grow irritated. He arrogantly scoffs at my thoughts and concerns, growing tired of considering my queries. He promptly tells me that I am completely wrong, that I should cease and desist in thinking about the questions in the way I have, and, finally, that the answer to the questions I have been considering are p, q, and r, respectively. With a look disdain and contempt, he wishes me adieu, and leaves. Now, presumably, not everything has gone wrong: I can adopt a belief in p, q, and r, simply on the authority of Hoggs’ testimony. However, there are clearly epistemic goods that I have not acquired, goods that I was specifically seeking when I sought out Hoggs’ input. It seems to me that I have, if we are to look at the issue diachronically, been left in a worse position overall: I have not learnt how to properly manage my reasons and reasoning about certain physics-related questions, and, specifically, do not know whether any of my reasons were good ones or not. Because of Hoggs’ arrogance, I am left completely sceptical about my ability to consider certain questions, and thus potentially worse off in the future: having been authoritatively told that I should no longer think in the way that I have been, I may now be left in the position of only being able to accept authoritative decrees on the matter, resisting any desire to engage with the relevant inquiry using my own epistemic powers, because of the authoritative judgements that I ought not to. I cease to have any faith in my reasoning on the subject matter at all. Perhaps my reader does not find the previous case convincing. One might potentially argue that it is actually better that I not try to understand certain complicated questions in the domain of physics without further training. Perhaps one might argue that Hoggs has 192 (albeit inadvertently) helped me by stopping me from engaging in erroneous and naive reasoning. The following case may be more persuasive, of a kind often discussed in relation to issues of epistemic injustice, and medical epistemology: Jane: Jane seeks out a medical professional for advice on a series of symptoms she has been experiencing. She has been suffering from a level of pain during menstruation that she has never experienced before, and often feels nauseous. She makes an appointment with her doctor, and tells him about these symptoms, emphasising the severity of the pain compared to all her previous menstruations. The doctor does not seem to pay much heed to her firstpersonal descriptions of pain, and after briefly checking her over, tells her that it is ‘just her period’, concluding that the pain could be heightened by stress, lack of sleep, or one of any other wide range of possible causes. Jane is sent home without further medical attention. A week later Jane ends up in the emergency room, where it is finally discovered that she has a serious ovarian cyst. Stories like these are unfortunately far too common, and the tendency for medical professionals to take women’s pain less seriously than men’s is well documented.244 Here we have a case in which someone who may be identified as an epistemic authority displays an arrogance that leaves the non-authority in an epistemically unfavourable position (and, here this is compounded by the fact that she is also left in a severely diminished position in practical terms, as a direct result of her impoverished epistemic position). Worse still, the arrogance of the doctor in this case not only diminishes Jane’s epistemic position going forward, it also hinders the doctor’s ability to epistemically function as well as he could. Jane has ended up with a false belief about her current medical state, and further false beliefs about what she ought to do to ensure her health into the future. The doctor in the case also has false beliefs about Jane’s case, but, more importantly, he has stopped the relevant epistemic practice from being able to correctly function, or provide the results it has. The doctor’s arrogance in this case is heavily consequential. See, for instance: Laura Kiesel, ‘Women and pain: Disparities in experience and treatment’, Harvard Medical Publishing (October 9th, 2017), Harvard Medical School, https://www.health.harvard.edu/blog/women-and-paindisparities-in-experience-and-treatment-2017100912562, and: Joe Fassler, ‘How Doctor’s Take Women’s Pain Less Seriously’, The Atlantic (October 15th, 2015), https://www.theatlantic.com/health/archive/2015/10/emergencyroom-wait-times-sexism/410515/. 244 193 As I see it, there are at least two ways in which an epistemic authority’s arrogance, or lack of epistemic humility, could diminish the potential value of an established relationship with a non-authority. First, such arrogance could hinder the non-authority’s ability to acquire certain epistemic goods, specifically when the non-authority does not merely seek true beliefs relative to the relevant domain of inquiry, and could undermine the non-authority’s ability to either properly function, or function to her full potential as an epistemic agent; secondly, in cases in which the nonauthority’s own evidence (or perspective more broadly) is relevant to the inquiry, or in cases in which the relevant inquiry involves a collaboration between epistemic authority and nonauthority, it is possible that arrogance on the authority’s part could even undermine the success of the inquiry itself, as suggested by the Jane case above. In relation to the first point, let me say this: I think there is an important way in which a lack of epistemic humility on the part of the epistemic authority can do the non-authority harm. Such a lack of humility can undermine the epistemic autonomy of the non-authority, of the kind I have discussed in Chapter III. I do not think this is entirely surprising, as the idea that epistemic humility entails a certain level of respect for other epistemic agents, even those deemed epistemically inferior, and it need not be much of a stretch to think that this would include a level of respect for these agents’ epistemic autonomy. A respect for another individual qua epistemic agent seems to require a respect for them as epistemically autonomous. And if having a relationship with an epistemic authority, in the way I have described it here, involves being epistemically dependent on them, it isn’t at all surprising that a lack of such respect can undermine that autonomy. This fits nicely with what I have argued for as being required of the non-authority, in relationship with an epistemic authority, in order for him to retain his epistemic autonomy to the highest degree possible. Recall that I suggested the two following conditions, roughly stated: 194 i. A non-authority is minimally autonomous so long as he reflectively endorses the principles that govern the relevant epistemic practice, and that he has identified an epistemic authority as such on the basis of good evidence. ii. A non-authority is further required (conditional on the nature of the specific details of the epistemic-authority-to-non-authority relationship in question) to involve himself in the relevant epistemic practice and inquiry where appropriate, and where practically reasonable to do so, in order to be considered fully autonomous. To fail to do this, when possible, is to be epistemically lazy, and to diminish one’s own autonomy.245 Though an arrogant epistemic authority may undermine a non-authority’s ability to do (i) to a certain extent (by, for example, obscuring certain facts about the methodology of the relevant epistemic practice on the basis of an arrogant assumption that the non-authority could not understand), I think it more clearly the case that a lack of epistemic humility on the authority’s part could undermine the non-authority’s ability to satisfy (ii) in certain cases. Consider a modified version of the Jane story above. Janette: Janette, like Jane, goes to her doctor to discuss her symptoms—she is suffering from an unusual level of pain during menstruation, and frequently feels nauseous. Her doctor looks her over, and tells that her worries are unfounded: she is probably stressed, and the symptoms will pass. Her feelings of pain and discomfort, she is told, are not reliable indicators of any serious, underlying cause. Luckily for Jannette, her pains eventually pass. However, because of the strength of her doctor’s conviction in regard to the evidence of her first-hand experience, she comes to believe that such evidence is irrelevant to medical practice. In her later interactions with medical professionals, she either minimises the severity of her first-personal observations of various symptoms, or, worse, even fails to mention them at all. When she is given various diagnoses, based on signs rather than symptoms, and offered medical treatment, she accepts these conclusions, even if they do not fit with her first-personal experience of her illness or symptoms. 245 Refer to Chapter III for further detail. 195 Now we can conclude this story tragically, pointing to the practical consequences to Jane’s health. But we need not do so in order to drive home the crucial point here: like in Jane’s case, the process of inquiry into Janette’s health has been undermined, but it is also the case that Janette’s epistemic autonomy, at least in terms of its potential, has been undermined. Janette no longer attempts to involve herself in the epistemic practice where relevant and appropriate, she no longer trusts herself as a competent epistemic agent in relation to certain questions, and certain forms of inquiry. She no longer reflectively endorses and identifies with certain of her prior reasons and claims to evidence. Something certainly seems to have gone wrong for Jannette in this case. If we care about a non-authority’s ability to be epistemically autonomous (to the highest degree possible), then we ought to care about whether or not our epistemic authorities are epistemically humble, i.e. that they are not epistemically arrogant. Arrogant authorities can do harm to our status as epistemic agents, in a variety of ways; such authorities can also do harm to epistemic communities more broadly, the practices they are involved in, and its store and supply of epistemic goods. The connection between one agent’s epistemic humility, and another’s autonomy, should come as no surprise. Priest’s view of epistemic humility already suggests something similar to what I have in mind. In further describing her case of the humble professor, she suggests that such an individual might often act in the following ways: 1. He recognizes his own intellectual superiority. 2. In spite of (1) he listens to each opinion carefully. 3. He is motivated to do (2) because he respects the intellectual autonomy and ability of his students, even in spite of (1). 4. He occasionally revises his own opinion in light of his students’ thoughts. I am sceptical that (4) applies in general across cases of epistemic authority, though I can 196 certainly see it being more likely in the more collaborative context of education, or in cases in which the disparity between the authority and non-authority’s relative level of skill (in regard to partaking in a certain epistemic practice) is not substantially large. I shall leave this aside for now, however. What matters for my purposes is that Priest identifies the intellectually humble as being motivated by a respect for the epistemic autonomy of his inferiors. This seems right, as consideration of cases of epistemic authority specifically seem to show. Before finishing, let us briefly consider here the value of epistemic humility in terms of its benefits to the epistemic authority qua inquirer, without reference to the non-authority.246 The point here is a simple one: research suggests that those we generally consider experts tend to show a diminishing level of skill or ability over time, especially when they are less willing to listen to the criticism of their peers, continue to partake in some form of education, etc. If we understand epistemic humility in the terms that Priest has suggested, then it is easy to see how a lack of such humility can be detrimental to an epistemic authority, since humility is characterised by the tendency to reflect on one’s epistemic limitations when interacting with members of one’s epistemic community.247 In being epistemically humble, then, an authority is provided with opportunity to hone, test, and fine-tune her skillset: she openly receives criticism from her peers; she acknowledges the need to re-educate herself in some areas, where appropriate, and to learn new, updated skills and methods; she becomes aware of important changes in the epistemic practice of which she is a practitioner, rather than being erroneously tied to an ancient form of the practice; and so forth. At minimum, the humble epistemic authority is far more likely to simply practice her abilities, improving upon them or keeping them ‘up to par’; she is more likely to reflect upon the justificatory basis for her methods, and to remind herself of the broader Though even here there will be second-hand benefits for any who choose to identify such individuals as epistemic authorities. 247 Priest, ‘Intellectual Humility’, 475. 246 197 theoretical framework that motivates her epistemic practice. The point, I imagine, is not a very surprising one: being intellectual humble is a virtue with epistemic benefit, in that plays a role in fostering one’s epistemic agency, and, in turn, improving our odds of epistemic success. It should be no less of a surprise that this applies to epistemic authorities, just as much as it does to others. Even if an authority is solely interested in the fruits of their own research, for their own benefit, they would be better served by being epistemically humble. As a social epistemologist, however, I would like to point out the more important consequence here: an epistemic authority who humbly services her own cognitive abilities is of far more benefit to us as a society, and epistemic community. IV.4. TOWARDS VIRTUOUS EPISTEMIC AUTHORITY What are we to conclude from these observations? It seems to me that, besides in the somewhat rarer cases in which the input of a non-authority is an important component of the epistemic practice in question, there isn’t much we can say in terms of what is epistemically required of an epistemic authority. I would not contest this: without recourse to the idea that people in a position of authority are morally obligated to help their inferiors, or stipulating that an authority has made a promise248 to the non-authority to serve him well, it would be hard to suggest that there is any strong normative force pressuring her to do just that (and, of course, even then it would not be an epistemically normative pressure). My account, then, suggests that there are certain normative forces acting on the non-authority, pressuring him to, at minimum, behave in an epistemically autonomous fashion;249 it does not by itself, however, suggest that there are similar forces behaving on the epistemic authority, leaving us with the asymmetry that concerned 248 249 Or signed a contract to, vowed to . . . etc. See Chapter III. 198 us at the start. This is in no way surprising, however, and I think it implausible to think that it could be otherwise: dependence, reliance, trust, and other related phenomena, all involve this asymmetry. It seems to me that social epistemology is concerned with making sense of our epistemic lives given that it is rooted in such asymmetric practices and phenomena, rather than being concentred with trying to avoid such things. Nevertheless, I take myself to have defend this claim: if we care about the epistemic status of non-authorities, and all members of our epistemic communities more broadly, then we ought to take steps to ensure that our epistemic authorities are intellectually humble. This suggests two things: i. As non-authorities seeking out certain epistemic goods, we can better evaluate certain individuals when choosing between candidates for the position of epistemic authority, and considerations of epistemic humility ought to play a part in this. This means that having a clear sense of what epistemic humility entails, and being able to discern various signs of such humility, can help us improve our epistemic situations. To speak idealistically: I would argue that our epistemic educations should include such training, as part of a broader training in identifying experts and epistemic authorities. ii. As potential epistemic authorities motivated by the ‘greater epistemic good’250 we ought to take steps to cultivate this virtue within ourselves. This is especially the case if we are already aware of the fact that others have identified us as epistemic authorities, and are treating us as such. To speak idealistically once more: perhaps the types of education that often churn out plausible candidates for being treated as epistemic authorities should be 250 Perhaps it could be argued that one has a duty to care for such a thing, but I will leave this point aside for now. 199 required to teach certain epistemic virtues (including epistemic humility). This would certainly be prudent in at least some domains, in which it is often the case that epistemic authorities have a more nuanced and complicated relationship with relative non-authorities: such as in medical and pedagogical contexts. The possibility of such practical implications, as well as the possibility of improving our epistemic community’s standing broadly speaking, motivates the following suggestion: we ought to seek out a fuller account of all the epistemic virtues that an epistemic authority ideally ought to have. The above has focused on but one epistemic virtue relevant to epistemic authority, but there are likely many more. Michel Croce has suggested the virtue of being sensitive to others’ epistemic needs as crucial for certain forms of epistemic authority, as well as arguing that the virtue of wisdom characterises the highest form of epistemic authority.251 Perhaps we could look at epistemic fairness, or epistemic altruism for further insight. Beyond identifying these virtues, however, I would argue that there are two other tasks at hand, two that perhaps epistemologists are best suited to accomplish: the task of discovering the most successful methods for identifying these various epistemic virtues in others, as well as how to teach others to use these methods; and the task of discovering the best pedagogical methods for teaching these virtues to those that are like to be treated as epistemic authorities. Only then will we be able to identify the most excellent of individuals to act as our intellectual superiors, and hopefully train our epistemic authorities to be virtuous in a manner that is beneficial to our community at large. 251 Croce, ‘Expert-Oriented Abilities’, 20. 200 BIBLIOGRAPHY Adler, Jonathan E. ‘Testimony, Trust, Knowing’. The Journal of Philosophy 91:5 (1994). 264-275. Benson, Paul. ‘Taking Ownership: Authority and Voice in Autonomous Agency’. Autonomy and the Challenges of Liberalism: New Essays, edited by J. Christman and J. Anderson, 101-126. Cambridge: Cambridge University Press, 2005. Bergmann, Michael. ‘Defeaters and Higher-Level Requirements’. Philosophical Quarterly (July 2005) 55:220. 419-436. Blais, Michel J. ‘Epistemic Tit for Tat’. The Journal of Philosophy 84:7 (1987). 363-375. Broncano, Fernando, and Jesús Vega. ‘Engaged Epistemic Agents’. Crítica, Revista Hispanoamericana de Filosofía 43:128 (2011). 55-79. Buss, Sarah and Andrea Westlund. ‘Personal Autonomy’. The Stanford Encyclopedia of Philosophy (Spring 2018). Edited by Edward N. Zalta. Christman, John. ‘Autonomy in Moral and Political Philosophy’. Stanford Encyclopedia of Philosophy (Spring 2018). Edited by Edward N. Zalta. Church, Ian M. ‘The Doxastic Account of Intellectual Humility’. Logos and Episteme 7:4 (2016). 413-433. Coady, David. ‘When Expert Disagree’. Episteme 3:1-2 (2006). 68-79. Code, Lorraine. ‘Toward a “Responsibilist” Epistemology’. Philosophy and Phenomenological Research 45:1 (1984). — — —. Epistemic Responsibility. Hanover: University Press of New England, 1987. Collins, Harry, and Robert Evans. Rethinking Expertise. Chicago: University of Chicago Press, 2007. Constantin, Jan and Thomas Grundmann. ‘Epistemic Authority: Preemption Through Undercutting Defeat’. Synthese. Forthcoming. Croce, Michel. ‘Expert-Oriented Abilities vs. Novice-Oriented Abilities: An Alternative Account of Epistemic Authority’. Episteme 15:4 (2017). 476-498. — — —. ‘Epistemic Paternalism and the Service Conception of Epistemic Authority’. Metaphilosophy 49:3 (018). 305-327. — — —. ‘On What it Takes to Be an Expert’. The Philosophical Quarterly. Forthcoming. 201 Dormandy, Katherine. ‘Epistemic Authority: Preemption or Proper-Basing?’. Erkenntnis 83:4. 773-791. — — —. ‘Epistemic Reasons: Preemptive Reasons or Total Reasons?’. Unpublished draft. Driver, Julia. Uneasy Virtue. New York: Cambridge University Press, 2001. Dryden, Jane. ‘Autonomy’, The Internet Encyclopedia of Philosophy. Accessed November 2, 2018. https://www.iep.utm.edu/autonomy/. Elga, Adam. ‘Reflection and Disagreement’. Social Epistemology: Essential Readings. Edited by Alvin I. Goldman, and Dennis Whitcomb. New York: Oxford University Press, 2011. 158-182. Elgin, Catherine Z. ‘Epistemic Agency’. Theory and Research in Education 11:2 (2013). 135-152. Faulkner, Paul. ‘The Social Character of Testimonial Knowledge’. The Journal of Philosophy 97:11 (2000). 581-601. — — —. ‘A Genealogy of Trust’. Episteme 7 (2007).305-321. — — —. ‘On Telling and Trusting’. Mind 116:464 (2007). 875-902. Fassler, Joe. ‘How Doctor’s Take Women’s Pain Less Seriously’. The Atlantic (October 15th, 2015). https://www.theatlantic.com/health/archive/2015/10/emergency-room-wait-timessexism/410515/. Foley, Richard. Intellectual Trust in Oneself and Others. New York: Cambridge University Press, 2001. Frankfurt, Harry. The Importance of What We Care About. Cambridge: Cambridge University Press, 1988. Fricker, Elizabeth. ‘The Epistemology of Testimony’. Proceedings of the Aristotelian Society, Supplementary Volumes 61 (1987). 57-83. — — —. ‘Against Gullibility’. Knowing from Words. Edited by Bimal Krishna Matilal and Arindam Chakrabarti. Boston: Kluwer Academic Press, 1994. 125-161. — — —. ‘Testimony and Epistemic Autonomy’. The Epistemology of Testimony, edited by Jennifer Lackey and Ernest Sosa, 225-253. New York: Oxford University Press, 2006. — — —. ‘Doing (Better) What Comes Naturally: Zagzebski on Rationality and Epistemic SelfTrust’. Episteme 13:2 (2016). 151-166. Fircker, Miranda. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press, 2007. 202 Gelfert, Axel. ‘Expertise, argumentation, and the end of inquiry’. Argumentation 25:297 (2011). Goldberg, Sanford. ‘Experts, Semantic and Epistemic’. Noûs 43:3 (2009). 581-598. Goldman, Alvin I. ‘Experts: Which Ones Should You Trust?’. Social Epistemology: Essential Readings, edited by Alvin I. Goldman and Dennis Whitcomb, 109-133. New York: Oxford University Press, 2011. — — —. ‘Expertise’. Topoi 37:3 (2018). 3-10. Grimm, Stephen. ‘The Value of Understanding’. Philosophy Compass 7:2. 103-117. — — —. ‘Epistemic Goals and Epistemic Values’. Philosophy and Phenomenological Research 77:3 (2008). Hardwig, John. ‘Epistemic Dependence’. The Journal of Philosophy 82:7 (1985). 335-349. — — —. ‘The Role of Trust in Knowledge’. The Journal of Philosophy 88:12 (1991). 693-708. Hawley, Katherine. ‘Trust, Distrust, and Commitment’. Noûs 48:1 (2014). 1-20. Hazlett, Allan. ‘Higher-Order Epistemic Attitudes and Intellectual Humility’. Episteme 9:3 (2012). 205-223. Hills, Alison. ‘Moral Testimony’. Philosophy Compass 8:6 (2013). 552-559. Ho, Anita. ‘Trusting Experts and Epistemic Humility in Disability’. The International Journal of Feminist Approaches to Bioethics 4:2 (Fall 2011). 102-123. Jäger, Christoph. ‘Epistemic Authority, Preemptive Reasons, and Understanding’. Episteme 13:2 (2016). 167-185. Kallestrup, Jesper, and Duncan Pritchard. ‘From Epistemic Anti-Individualism to Intellectual Humility’. Res Philosophica 93:3 (2016). 533-552. Kant, Immanuel. Critique of Judgment. Translated by J. H. Bernard. New York: Hafner, 1951. Kappel, Klemens. ‘Believing on Trust’. Synthese 191 (2014). 2009-2028. Kelly, Thomas. ‘Peer Disagreement and Higher Order Evidence’. Social Epistemology: Essential Readings. Edited by Alvin I. Goldman, and Dennis Whitcomb. New York: Oxford University Press, 2011. 183-219. Keren, Arnon. ‘Epistemic Authority, Testimony, and the Transmission of Knowledge’. Episteme 4:3 (2007). 368-381. 203 — — —. ‘Zagzebski on Authority and Preemption in the Domain of Belief’. European Journal of Philosophy of Religion. Forthcoming. — — —. ‘Trust and Belief: A Preemptive-Reasons Account’. Synthese. Forthcoming. Kiesel, Laura. ‘Women and pain: Disparities in experience and treatment’. Harvard Medical Publishing (October 9th, 2017). Harvard Medical School. https://www.health.harvard.edu/blog/women-and-pain-disparities-in-experience-andtreatment-2017100912562. Knobe, Joshua. ‘Person as scientist, person as moralist’, 315-365. Behavioral and Brain Sciences 33 (2010). Kvanvig, Jonathan L. ‘Truth Is not the Primary Epistemic Goal’. Contemporary Debates in Epistemology, edited by Matthias Steup and Ernest Sosa, 285-295. Malden: Blackwell Publishing, 2016. Lackey, Jennifer. ‘It Takes Two to Tango: Beyond Reductionism and Non-Reductionism in the Epistemology of Testimony’. The Epistemology of Testimony, edited by Jennifer Lackey and Ernest Sosa, 160-189. New York: Oxford University Press, 2006. Lane, Melissa. ‘When he Experts are Uncertain: Scientific Knowledge and the Ethics of Democratic Judgment’. Episteme 11:1 (2014). 97-118. Matheson, David. ‘Conflicting Experts and Dialectical Performance: Adjudicating Heuristics or the Layperson’. Argumentation 19 (2005). 145-158 McMyler, Benjamin. Testimony, Trust, and Authority. New York: Oxford University Press, 2011. Miller, Boaz. ‘Why (Some) Knowledge is the Property of a Community and Possibly None of its Members’. The Philosophical Quarterly. Forthcoming. Mercier, Hugo and Dan Sperber. ‘Why do humans reason? Arguments for an argumentative theory’, 57-111. Behavioural and Brain Sciences 34 (2011). Moran, Richard. ‘Getting Told and Being Believed’. Philosopher’s Imprint 5:5 (2005). Mudd, Sasha. ‘Epistemic autonomy: a criterion for virtue?’. Theory and Research in Education 11:2 (2013). 153-165. Munpower, Jeryl L. and Thomas R. Stewart. ‘Expert Judgement and Expert Disagreement’. Thinking and Reasoning 2 (1996). 191-211. Palmira, Michele. ‘Expert Deference and the Objectivity of Epistemology’. Unpublished manuscript. Priest, Maura. ‘Intellectual Humility: An Interpersonal Account’. Ergo 4:16 (2017). 204 Pritchard, Duncan. ‘Knowledge and Understanding’. Virtue Epistemology Naturalized: Bridges Between Virtue Epistemology and Philosophy of Science, edited by A. Fairweather, 315-327. New York: Springer, 2014. — — —. ‘Epistemic Paternalism and Epistemic Value’. Philosophical Inquiries 1:2 (2013). 9-37. — — —. ‘Seeing it for Oneself: Perceptual Knowledge, Understanding, and Intellectual Autonomy’. Episteme 13:1 (2016). 29-42. Quast, Christian. ‘Expertise: A Practical Explication’. Topoi 37:1 (2018).11-27. Raz, Joseph. The Morality of Freedom. Oxford: Clarendon Press, 1988. Roberts, Robert C. and W. Jay Wood. ‘Humility and Epistemic Goods’. Intellectual Virtues: Perspectives From Ethics and Epistemology, edited by Linda Zagzebski and Michael DePaul, 257-279. New York: Oxford University Press, 2003. Ryle, Gilbert. The Concept of Mind. Chicago: The University of Chicago Press, 1949. Samuelson, Peter L., Ian M. Church, Matthew Jarvinen, Thomas Paulus. ‘The Science of Intellectual Humility White Paper’. Unpublished manuscript. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.370.6006&rep =rep1&type=pdf. Schafer, Karl. ‘How Common is Peer Disagreement? On Self-Trust and Rational Symmetry’. Philosophy and Phenomenological Research 91:1 (2015). 25-46. Sellinger, Evan, and Robert P. Crease. The Philosophy of Expertise. New York: Columbia University Press, 2006. Solomon, Miriam. Making Medical Knowledge. Oxford: Oxford University Press, 2015. — — —. ‘Expert disagreement and medical authority’. Philosophical Issues in Psychiatry III. Oxford: Oxford University Press, 2015. 60-72. Sperber, Dan, Fabrice Clément, Christophe Heintz, Olivier Mascaro, Hugo Mercier, Gloria Origgi, and Deirde Wilson. ‘Epistemic Vigilance’. Mind and Language 25:4 (2010). 359-393. Vega Encabo, Jesús. ‘Epistemic merit, autonomy, and testimony’. Theoria 23:1 (2008). 45-56. Wallis, Charles. ‘Consciousness, Context, and Know-How’. Synthese 160:1 (2008). 123-153. Westlund, Andrea C. ‘Rethinking Relational Autonomy’. Hypatia 24:4 (2009). 26-49. 205 Whitcomb, Dennis, Heather Battaly, Jason Baehr, and Daniel Howard-Snyder, ‘Intellectual Humility: Owning Our Limitations’, Philosophy and Phenomenological Research 94:3 (2017), 514. Wilkenfeld, Daniel A. ‘Understanding as representation manipulability’. Synthese 190:6 (2013). 997-1016. Wright, Stephen. ‘Circular testimony’. Philosophical Studies 173 (2016). 2029-2048. Zagzebski, Linda. Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. New York: Oxford University Press, 2012. — — —. ‘Replies to Christoph Jäger and Elizabeth Fricker’. Episteme 13:6 (2016). 187-194. — — —. ‘The Search for the Source of Epistemic Good’. Metaphilosophy 34:1/2 (2003). 12-28. 206