Contemporary Epistemology BLACKWELL PHILOSOPHY ANTHOLOGIES

Each volume in this outstanding series provides an authoritative and comprehensive collection of the essential primary readings from philosophy’s main fields of study. Designed to complement the Blackwell Companions to Philosophy series, each volume represents an unparalleled resource in its own right, and will provide the ideal plat- form for course use.

1 Cottingham: Western Philosophy: An Anthology (second edition) 2 Cahoone: From Modernism to Postmodernism: An Anthology (expanded second edition) 3 LaFollette: Ethics in Practice: An Anthology (third edition) 4 Goodin and Pettit: Contemporary Political Philosophy: An Anthology (second edition) 5 Eze: African Philosophy: An Anthology 6 McNeill and Feldman: Continental Philosophy: An Anthology 7 Kim and Sosa: Metaphysics: An Anthology 8 Lycan and Prinz: Mind and Cognition: An Anthology (third edition) 9 Kuhse and Singer: Bioethics: An Anthology (second edition) 10 Cummins and Cummins: Minds, Brains, and Computers – The Foundations of Cognitive Science: An Anthology 11 Sosa, Kim, Fantl, and McGrath Epistemology: An Anthology (second edition) 12 Kearney and Rasmussen: Continental Aesthetics – Romanticism to Postmodernism: An Anthology 13 Martinich and Sosa: Analytic Philosophy: An Anthology 14 Jacquette: Philosophy of Logic: An Anthology 15 Jacquette: Philosophy of Mathematics: An Anthology 16 Harris, Pratt, and Waters: American Philosophies: An Anthology 17 Emmanuel and Goold: Modern Philosophy – From Descartes to Nietzsche: An Anthology 18 Scharff and Dusek: Philosophy of Technology – The Technological Condition: An Anthology 19 Light and Rolston: Environmental Ethics: An Anthology 20 Taliaferro and Griffiths: Philosophy of Religion: An Anthology 21 Lamarque and Olsen: Aesthetics and the Philosophy of Art – The Analytic Tradition: An Anthology 22 John and Lopes: Philosophy of Literature – Contemporary and Classic Readings: An Anthology 23 Cudd and Andreasen: Feminist Theory: A Philosophical Anthology 24 Carroll and Choi: Philosophy of Film and Motion Pictures: An Anthology 25 Lange: Philosophy of Science: An Anthology 26 Shafer‐Landau and Cuneo: Foundations of Ethics: An Anthology 27 Curren: Philosophy of Education: An Anthology 28 Shafer‐Landau: Ethical Theory: An Anthology 29 Cahn and Meskin: Aesthetics: A Comprehensive Anthology 30 McGrew, Alspector‐Kelly and Allhoff: The Philosophy of Science: An Historical Anthology 31 May: Philosophy of Law: Classic and Contemporary Readings 32 Rosenberg and Arp: Philosophy of Biology: An Anthology 33 Kim, Korman, and Sosa: Metaphysics: An Anthology (second edition) 34 Martinich and Sosa: Analytic Philosophy: An Anthology (second edition) 35 Shafer‐Landau: Ethical Theory: An Anthology (second edition) 36 Hetherington: Metaphysics and Epistemology: A Guided Anthology 37 Scharff and Dusek: Philosophy of Technology – The Technological Condition: An Anthology (second edition) 38 LaFollette: Ethics in Practice: An Anthology (fourth edition) 39 Davis: Contemporary Moral and Social Issues: An Introduction through Original Fiction, Discussion, and Readings 40 Dancy and Sandis: Philosophy of Action: An Anthology 41 Fantl, McGrath, and Sosa: Contemporary Epistemology: An Anthology Contemporary Epistemology An Anthology

Edited by Jeremy Fantl Matthew McGrath Ernest Sosa This edition first published 2019 © 2019 John Wiley & Sons, Inc.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

The right of Jeremy Fantl, Matthew McGrath, and Ernest Sosa to be identified as the authors of the editorial material in this work has been asserted in accordance with law.

Registered Office John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA

Editorial Office 111 River Street, Hoboken, NJ 07030, USA

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.

Limit of Liability/Disclaimer of Warranty While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all ­warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements­ for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential­ source of further information does not mean that the publisher and authors endorse the information or services the organization,­ website, or product may provide or recommendations it may make. This work is sold with the ­understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers­ should be aware that websites listed in this work may have changed or disappeared between when this work was ­written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial ­damages, including but not limited to special, incidental, consequential, or other damages.

Library of Congress Cataloging‐in‐Publication Data Names: Fantl, Jeremy, editor. | McGrath, Matthew, editor. | Sosa, Ernest, editor. Title: Contemporary epistemology : an anthology / edited by Jeremy Fantl, Matthew McGrath, Ernest Sosa. Description: First edition. | Hoboken, NJ : John Wiley & Sons, 2018. | Series: Blackwell philosophy anthologies ; 41 | Includes index. | Identifiers: LCCN 2018025092 (print) | LCCN 2018035283 (ebook) | ISBN 9781119420781 (Adobe PDF) | ISBN 9781119420798 (ePub) | ISBN 9781119420774 (pbk.) | ISBN 9781119420804 (hardback) Subjects: LCSH: Knowledge, Theory of. Classification: LCC BD143 (ebook) | LCC BD143 .C658 2018 (print) | DDC 121–dc23 LC record available at https://lccn.loc.gov/2018025092 Cover Design: Wiley Cover Image: © Lorna J Webb/Moment Open/Getty Images

Set in 9.5/11.5pt Bembo by SPi Global, Pondicherry, India

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1 Contents

Preface vii

Part I The Ethics of Belief 1 1 Deontological Desiderata 3 William Alston 2 Voluntary Belief and Epistemic Evaluation 17 Richard Feldman

Part II Practical Reasons for Belief ? 29 3 The Wrong Kind of Reason 31 Pamela Hieronymi 4 No Exception for Belief 44 Susanna Rinard 5 Promising Against the Evidence 58 Berislav Marušic ́

Part III Reliance 75 6 Evidentialism and Pragmatic Constraints on Outright Belief 77 Dorit Ganson 7 Alief and Belief 91 Tamar Gendler 8 Can It Be Rational to Have Faith? 110 Lara Buchak 9 Assertion and Practical Reasoning: Common or Divergent Epistemic Standards? 126 Jessica Brown vi contents

Part IV Epistemic Dysfunctions 147 10 Testimonial Injustice 149 Miranda Fricker 11 Cognitive Penetrability and Perceptual Justification 164 Susanna Siegel

Part V Virtue Epistemology 179 12 The Search for the Source of Epistemic Good 181 Linda Zagzebski 13 Why We Don’t Deserve Credit for Everything We Know 192 Jennifer Lackey 14 A (Different) Virtue Epistemology 205 John Greco 15 Knowledge and Justification 220 Ernest Sosa

Part VI Disagreement 229 16 Epistemology of Disagreement: The Good News 231 David Christensen 17 The Epistemic Significance of Disagreement 249 Thomas Kelly

Part VII Permissivism About Belief ? 265 18 Epistemic Permissiveness 267 Roger White 19 Permission to Believe: Why Permissivism Is True and What It Tells Us About Irrelevant Influences on Belief 277 Miriam Schoenfield

Index 296 Preface

Contemporary epistemology is a vibrant and wide‐ ­further questions: how does evidence about other’ ranging field within philosophy. Some anthologies, beliefs make a difference to what one should believe? including our earlier effort, Epistemology: An Anthology, How does evidence about what oneself should believe aim at giving the reader more or less comprehensive make a difference? access to influential works across the field. This anthol- Action, not belief, is what is most naturally subject to ogy takes a different approach. It covers a set of interre- norms. Therefore, the ethics of belief is often developed lated topics which, taken together, comprise the materials with an eye to how belief and justified belief compare for a unified course in epistemology, suitable either for and relate to action and the ethics of action. It follows graduate students or upper‐level undergraduates. from this that the ethics of belief cannot be fruitfully These topics in this volume fall under the broad pursued without some engagement with ethics proper. heading the ethics of belief; they all concern, in some way Likewise, some conclusions about the ethics of belief or other, questions about what one may or should depend on considerations about the metaphysics of believe. William Clifford, who first used the term “eth- belief. Therefore, some of the readings in this volume ics of belief,” thought of these topics as genuinely ethi- could have found their way into ethics or philosophy cal: in asking what one should believe, we are asking of mind anthologies as well as an epistemology anthol- what one morally should believe. Over time, the term ogy. We think this is as it should be. “ethics of belief” has acquired a broader reference; Editors of anthologies must also decide whether to it now covers any distinctively normative questions include primarily influential articles and chapters, or about belief – any questions concerning the shoulds, important recent publications which have not yet had oughts, musts, and mays of belief. The ethics of belief time to steer the field. This is a particularly difficult asks whether there are indeed “norms” for belief and, conundrum when one is anthologizing contemporary if there are, what they are and what makes them pos- epistemology. Our policy was to create a balance. We sible. What sorts of factors bear on what one should, have included a number of influential articles, though may, etc., believe: is it only a matter of evidence, or do none published before 2000, as well as several very the costs and benefits of believing for your life and recent articles we expect to have influence on the field. others’ lives help determine what one should believe? We have also included several articles which we think Within the category of evidence, one might ask deserve more attention from epistemologists.

Part I

The Ethics of Belief

­Introduction blameworthy, because it doesn’t look like believers are in control of their beliefs. In 1877, the mathematician William Clifford published The view that people can control their beliefs is “The Ethics of Belief” in which he argued that “it is sometimes referred to as “doxastic voluntarism.” A wrong always, everywhere, and for any one, to believe majority of philosophers think that doxastic volunta- anything upon insufficient evidence.” Clifford’s con- rism is false. They point, for example, to the obvious tention concerned the morality of belief on insufficient difficulty of, while situated in a well‐lit room, choosing evidence; just as actions can be morally wrong, beliefs to believe that the lights are off. Furthermore, some can be morally wrong, as well. Furthermore, it was argue, a plausible control condition on blameworthi- because actions could be morally wrong that beliefs ness has it that you can only be blamed for something could be morally wrong. Belief, Clifford thought, was if you can voluntarily choose to do it. It follows that partly characterized by a willingness to act in ways sug- believers can only be blamed for their beliefs if they gested by those beliefs. So for the actions based on can choose what to believe. those beliefs to be morally wrong, the beliefs would This is the central contention of William Alston have to share in that wrongness. If the believer is blame- in the selection reprinted here. Alston argues against worthy for the actions based on the beliefs, they can what he calls a “deontological conception of justifica- also be blameworthy for the beliefs themselves. tion.” According to the deontological conception, When contemporary epistemologists talk about the ­justified belief is a matter of believing blamelessly – ethics of belief, they not only have in mind the moral of believing as one ought. But, argues Alston, on the wrongness of belief, but an epistemic wrongness, as deontological conception of justification, we only well. While Clifford thought that beliefs (and believers) have justified (or unjustified) beliefs if we are in con- could be morally blameworthy, many contemporary trol of those beliefs. And, he argues, we aren’t. We lack epistemologists think that beliefs and believers can be direct (or, in his word, “basic”) immediate control over epistemically blameworthy – blameworthy in a distinc- our beliefs – the kind we have over simple movements, tively intellectual or epistemic way. But whether it’s like raising our arm. And the kinds of long‐term, non‐ epistemic or moral blameworthiness at issue, the view basic control we might have over our beliefs is not that beliefs and believers can be blameworthy is subject sufficient to ground a robust deontological conception to a central worry: it doesn’t look like beliefs can be of justification.

Contemporary Epistemology: An Anthology, First Edition. Edited by Jeremy Fantl, Matthew McGrath, and Ernest Sosa. © 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc. 2 part i: the ethics of belief

Richard Feldman is more sympathetic toward a don’t do them, then we don’t do as we ought, regardless deontological conception of justification, though he of whether we could have. does not think that we have the kind of control over According to Feldman, one of our roles is believer. our beliefs that Alston seems to require. Instead, Feldman As believers, there are certain obligations we accrue. argues that there is a natural deontological conception Though Feldman doesn’t want to take a stand in this of justification that doesn’t require satisfaction of a con- paper on the nature of those obligations, we know trol condition. According to Feldman, there is a range of from his other work what he thinks the primary – obligations that accrue to us in virtue of our occupying indeed, the only – doxastic obligation is: to believe certain roles. For example, parents might have certain what fits the evidence. This is an obligation whether obligations to take care of their children, etc. Feldman or not you’re able to believe what fits the evidence or calls these obligations “role oughts.” Importantly, we can control whether or not you believe what fits the might fail to satisfy role oughts even though we can’t evidence. If you can’t believe what fits the evidence, help but do so. This, argues Feldman, does not get us off then you can’t believe in accord with your duty, and the hook; if our role demands certain acts of us, and we you have failed to fulfill the duties of a believer.

Further­ Reading

Clifford, William. (1886). The ethics of belief. In: Lectures and Ryan, Sharon. (2003). Doxastic compatibilism and the ethics Essays. Macmillan. of belief. Philosophical Studies 114: 47−79. Firth, Roderick. (1978). Are epistemic concepts reducible to Wolterstorff, Nicholas. (1996). John Locke and the Ethics of ethical concepts? In: Values and Morals (ed. Alvin Goldman Belief (No. 2). Cambridge: Cambridge University Press. and Jaegwon Kim), 215−29. Dordrecht: Kluwer Academic Publishers. James, William. (1979). The Will to Believe and Other Essays in Popular Philosophy (Vol. 6). Cambridge, MA: Harvard University Press. 1 Deontological Desiderata

William Alston

i. Preliminaries* the act. My use is broader. I use it to range over any kind of requirement, not restricted to moral obligation, I now begin the detailed treatment of the items on my and not excluding requirements that are based on initial list of alleged epistemic desiderata. I will be con- ­consequences of what is required. And I identify deon- cerned with clarification of the nature of each desid- tological considerations as having to do with the triad eratum, how it should be construed. Where there are of statuses – required, forbidden, and permitted. Thus any serious questions as to the viability of an item, those way in which it would be epistemically desirable (desir- will be addressed. I discuss the deontological group first able from the standpoint of an aim at true belief) for a because it gives rise to crucial problems about viability, belief to be required or permitted (i.e., not forbidden) as a result of which I postponed consideration of it in would count as a deontological desideratum in my Chapter 3 until and unless they can be resolved. terminology. Here are the deontological candidates for epistemic Back to the above list, I think it will suffice to con- desiderata of belief (B) that were listed in Chapter 3. centrate on 9 and 11. Each of these can be construed as focusing on something’s being permitted, not being in 9. B is held permissibly (one is not subject to blame for violation of any intellectual requirements. Desideratum doing so). 9 is matter of the having or the acquiring of the belief 10. B is formed and held responsibly. being permitted. Desideratum 11 is a matter of the 11. The causal ancestry of B does not contain viola- permissibility or lack thereof of what one did that led tions of intellectual obligations. to the acquisition of the belief. Although 10, the for- mation in terms of responsibility, is familiar in the First a word about my terminology. ‘Deontology’ and ­literature, I think it is ambiguous between 9 and 11 and ‘deontological’ come from the Greek deon – ‘what is so does not require separate treatment. The basic differ- binding’ or ‘duty’. In ethics, deontology is the study of ence between 9 and 11 is what is said to be permitted – duty or obligation, and a deontological theory of ethics either the believing itself or what led up to it. Thus, to is one that takes duty or obligation to be the most basic foreshadow a major point in the ensuing discussion, ethical concept and treats it as an intrinsic ethical value 9 gives rise to problems about voluntary control of of an act rather than in terms of the consequences of belief whereas 11 does not.

Original publication details: William Alston, “Deontological Desiderata.” Reprinted from Beyond “Justification”: Dimensions of Epistemic Evaluation, by William P. Alston, pp. 58−80. Copyright © 2006 by Cornell University. Used by permission of the publisher, Cornell University Press.

Contemporary Epistemology: An Anthology, First Edition. Edited by Jeremy Fantl, Matthew McGrath, and Ernest Sosa. © 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc. 4 william alston

I have already pointed out […] that it is plausible to forbidden to do A if S lacks an effective choice as to suppose that ‘justified’ came into epistemology from its whether to do so. Therefore, the most fundamental more unproblematic use with respect to voluntary issue raised by the claim of 9 to be an epistemic desid- action. I am justified in doing something, for example, eratum is whether believings are under effective volun- appointing someone to a Teaching Assistantship on my tary control. If they are not and hence if deontological own, provided my doing so is in accordance with the terms do not apply to them, alleged epistemic desiderata relevant rules and regulations, provided it is permitted by like 9 do not get so far as to be a candidate for an epis- those rules and hence that I could not rightfully be temic desideratum. It suffers shipwreck before leaving blamed or held to account for it, and was acting responsibly port. I will argue that believings are not subject to vol- in doing so.1 The rules could be institutional, as in the untary control. But before that, there are some prelimi- above example, or legal or moral. Thus I would be nary points to be made. morally justified in failing to make a contribution to a First, if I considered the possibility of deontological certain organization provided my doing so doesn’t vio- ED for beliefs to be a live one, I would need to con- late any moral rule. Because of this provenance it is sider a belief’s enjoying the stronger deontological sta- natural to think of believing, when taken to be subject tus of being a case of complying with an epistemic to being justified or unjustified, as subject to require- obligation, doing what is required, as well as the weaker ment, prohibition, and permission. We say things like status of merely being something that is epistemically “You shouldn’t have supposed so readily that he would permitted. But since I hold that no deontological status not return”, “You have no right to assume that”, “You is possible for beliefs, I will not need to go into the dif- shouldn’t jump to conclusions”, and “I ought to have ferent statuses separately. And since justificationists of trusted him more than I did”. Locutions like these a deontological bent have concentrated on a belief’s seem to be interchangeable with speaking of a belief as being epistemically permitted, I will go along with being, or not being, justified. These considerations were that focus. introduced in this book prior to the abandonment of a Second, although the discussion […] thus far has justification‐based epistemology of belief, and in the been solely in terms of belief, we need to include con- new dispensation they have no force. Since we are sideration of other prepositional attitudes that are con- thinking of 9 and 11 simply as states of affairs that are, trary to belief. Chisholm (1977, chap. 1) speaks in terms or may be thought to be, important goals of cognition, of a trichotomy of ‘believe’ (or ‘accept’), ‘reject’, and the fact that they have often been thought to constitute ‘withhold’ that p. Since rejecting p is identified with a belief’s being justified, with all the associations that believing some contrary of p, at least not‐p, it brings in brings from talk of the justification of actions, has lost no new kind of prepositional attitude, but withholding whatever meta‐epistemological significance it had p, believing neither it nor any contrary, does. The basic under the old dispensation. The idea of a belief’s being point here is that one has control over a given type of required, permitted, or forbidden will have to swim or prepositional attitude only if one also has control over sink on its own, without support from the etymology some field of incompatible alternatives. To have effec- of ‘justified’. I will now enter onto the elucidation of 9 tive control over believing that p is to have control over and a critical discussion of its credentials as an epis- whether one believes that p or takes on some alterna- temic desideratum. The criticism will mostly hinge on tive thereto. Therefore, to be strictly accurate we should whether we have effective voluntary control of believ- say that our problem about 9 concerns voluntary con- ings. I will argue that we do not. trol over intellectual prepositional attitudes generally. It seems clear that the terms of the deontological Though my formulations will mostly be in terms of triad, permitted, required, and forbidden, apply to some- belief, they should be understood as having this more thing only if it is under effective voluntary control. By general bearing. the time‐honored principle “Ought implies can”, one Third, something must be said about the relation can be obliged to do A only if one has an effective between the voluntary control of actions and of states choice as to whether to do A. It is equally obvious that of affairs. Thus far I have been oscillating between the it makes no sense to speak of S’s being permitted or two. A belief is a more or less long‐lived state of the deontological desiderata 5 psyche that can influence actions and reactions of the testimony; I had no choice” – also strongly suggest that subject so long as it persists. And the same holds for belief is under voluntary control. Else why could we other propositional attitudes. Thus, in speaking of vol- speak of what beliefs one should or shouldn’t form, untary control of beliefs, we have been speaking of the or that one did or did not have a choice as to whether control of states. But couldn’t we just as well speak of one forms a certain belief? Though this view is dis- the voluntary control of the action of bringing about tinctly out of favor today, it still has its defenders.2 such states: accepting, rejecting, or withholding a prop- Such locutions also naturally suggest not only that osition? If the two are strictly correlative, we could believing is under voluntary control but that this con- equally well conduct the discussion in terms of either. trol is of the maximally direct sort that we have over the Whenever we are responsible for a state of affairs by motions of our limbs, the voluntary movements of virtue of having brought it about, we may just as well which constitute basic actions. A basic action is one that speak of being responsible for the action of bringing it we perform “at will”, just by an intention, volition, about. There are reasons, however, for proceeding in choice, or decision to do so. It is something we “just terms of states. do”, not by doing something else. Let’s call the kind of The main reason is this. If we hold that beliefs are control we have over states of affairs we can bring about subject to deontological evaluation because they are by basic actions basic voluntary control. If we do have vol- under voluntary control, we need not restrict ourselves untary control of beliefs, we have the same reason for to beliefs that are formed intentionally by a voluntary supposing it to be basic control that we have for suppos- act. I could be blamed for believing that p in the ing ourselves to have basic control over movements of absence of adequate evidence, even if the belief was our limbs, namely, that we are hard pressed to specify formed automatically, not by voluntarily carrying out any action by doing which we get the limbs moved or an intention to do so. Provided believing in general is the beliefs acquired. Hence it is not surprising that the under voluntary control, it is enough that I could have basic voluntary control thesis has had distinguished rejected or withheld the proposition by a voluntary act proponents throughout the history of philosophy. had I chosen to do so. Augustine, Aquinas, Descartes, Kierkegaard, and many The final preliminary note is this. Our issue does not others have usually been read this way.3 And discussions concern free will or freedom of action, at least in any pro and con of the voluntary control of beliefs have sense in which that goes beyond one’s action being mostly focused on the basic control version. under the control of the will. On a “libertarian” con- Nevertheless, as the subsequent discussion will show, ception of free will this is not sufficient; it is required there are other forms of voluntarism about belief that also that both A and non‐A be causally possible, given need to be taken into account in a complete treatment. all the causal influences on the agent. A libertarian will, But for now I am concerned to give a critical exami- no doubt, maintain that if deontological concepts are nation of the basic voluntary control thesis. Those who to apply to believings in the same sense in which they have attacked it are divided between those who hold apply to overt actions, then all the libertarian condi- that believing at will is logically impossible and those tions will have to apply to believings. Here, however, I who hold that it is only psychologically impossible, a am concerned only with whether believings are under capacity that we in fact lack though one we conceivably voluntary control. could have had.4 I cannot see any sufficient reason for the stronger claim, and I shall merely contend that we are not so constituted as to be able to take prepositional attitudes at will. My argument for this, if it can be called ii. Basic Voluntary Control that, simply consists in asking you to consider whether of Believing you have any such power. Can you, at this moment, start to believe that the Roman Empire is still in control of Locutions like the ones cited earlier as encouraging the western Europe, just by deciding to do so? If you find it application of deontological terms to believing – “You incredible that you should be sufficiently motivated to shouldn’t jump to conclusions”, “I had to accept his even try to believe this, suppose that someone offers 6 william alston you $500 million to believe it, and that you are much And it is perfectly clear that in this situation I have no more interested in the money than in believing the power at all to refrain from that belief. So it is with eve- truth. Could you do what it takes to get that reward? rything that seems obvious to us. We have just as little Remember that we are speaking of believing at will. No voluntary control over ordinary beliefs formed by intro- doubt, there are things you could do that would increase spection, memory, and simple uncontroversial inferences the probability of your believing this, but we will get to from uncontroversial premises. that later. Can you switch prepositional attitudes toward The above discussion may suggest to the voluntarist that proposition just by deciding to do so? It seems clear that he can still make a stand on propositions that do to me that I have no such power. Volitions, decisions, or not seem clearly true or false and hold that there one choosings don’t hook up with propositional attitude often has the capacity to adopt whatever propositional inaugurations, just as they don’t hook up with the secre- attitude one chooses. In religion, philosophy, history, tion of gastric juices or with metabolism. There could and high‐level scientific inquiry it is often the case that, conceivably be individual differences in this regard. so far as one can see, the relevant arguments do not Some people can wiggle their ears at will, but most of definitively settle the matter one way or the other. I us cannot. However, I very much doubt that any of us engage in a prolonged study of free will or causality. I are endowed with the power of believing that p, for any carefully consider arguments for and against various given p, at will. The temptation to suppose otherwise positions. It seems to me that none of the positions may stem from conflating that power with others that have been decisively established, though there are are clearly distinct. If I were to set out to bring myself weighty considerations that can be urged in support of into a state of belief that p, just by an act of will, I might each. There are serious difficulties with all the posi- assert that p with what sounds like conviction, or dwell tions, though, so far as I can see, more than one con- favorably on the idea that p, or imagine a sentence tender is left in the field. So what am I to do? I could expressing p emblazoned in the heavens with an angelic just abandon the question. But, alternatively, I could, so chorus in the background intoning the Gloria of Bach’s it seems, simply decide to adopt one of the positions. Is Mass in B Minor. All this I can do at will, but none of it that not what I must do if I am to make any judgment amounts to forming a belief that p. It is all show, an on the matter? elaborate pretense of believing. Having gone through There are also practical situations in which we are all this, my propositional attitudes will remain just as confronted with incompatible answers to a certain they were before; or if there is any change, it will be as question, none of which we see to be clearly true or a result of these gyrations.5 false. Here we often do not have the luxury of leaving Don’t suppose that our inability to believe at will is the field; since we must act in one way rather than restricted to what is obviously false. It also extends to another, we are forced to form and act on some belief beliefs that are obviously true. I have already made the about the matter. It would be a good idea for me to point that voluntary control attaches to sets of contraries. plant these flowers today iff it will rain tomorrow. But it To take the simplest case, if the sphere of my voluntary is not at all clear to me whether tomorrow will be rainy. control does not extend both to A and to not‐A, then it I must either plant the flowers today or not, and if I just attaches to neither. If I don’t have the power to choose ignore the issue, that will be equivalent to assuming that between A and not‐A, then we are in no position to say it will not rain tomorrow. Hence the better part of wis- that I did A at will, rather than just did it, accompanied dom would be to make a choice between the alterna- perhaps by a volition. Thus, even if I willingly, or not tive predictions. On a larger scale, a field commander in unwillingly, form perceptual beliefs in the way I do, it by wartime is often faced with questions about the current no means follows that I form those beliefs at will, or that disposition of enemy forces. But often such information I have voluntary control over such belief formation. It as he has does not tell him just what that disposition is. would have to be true that I have voluntary control over In disposing his own forces he must act on some whether I do or do not believe that the tree has leaves on assumption about the enemy’s forces. Hence he is forced it when I see a tree with leaves on it just before me in to decide on a hypothesis as to that disposition and act broad daylight with my eyesight working normally. on that basis. What else can he do?6 deontological desiderata 7

Despite the intuitive appeal of the idea that beliefs matter. He is, self‐consciously, proceeding on an are formed at will in these cases, there are several alter- assumption concerning the truth of which he has no native construals, one or another of which is a better belief at all. One may also make an assumption for the- reading of each. Begin with the philosopher who really oretical purposes, in order to see how it “pans out” in does come to believe the libertarian account of free the hope that one will thereby obtain some additional will or the epiphenomenalist position on the mind‐ reasons for believing it to be true or false. A scientist body question. Where that happens it is presumably can adopt “as a working hypothesis” the proposition because at least for the moment the considerations in that the atomic nucleus is positively charged, draw var- favor of the position seem to be conclusive, even ious consequences from it, and proceed to test those though previously they did not. And at that time the consequences. He need not believe that the atomic belief follows automatically from that momentary nucleus is positively charged in order to carry out this seeming of conclusiveness, just as it does in cases where operation. Indeed, he would be doing this because he it always seems obvious what the truth of the matter is does not yet know what to believe about the matter. whenever one turns one’s attention to it. At that Likewise a philosopher might take materialism as a moment, S is no more able to accept a compatibilist working hypothesis to see how it works out in applica- account of free will or a hard‐nosed materialism on the tion to various problems. mind‐body problem than he would be if the positions Working hypotheses may also be involved in activi- he comes to believe had seemed obviously true from ties that are a blend of the theoretical and the practical. his first consideration of the problem. If, at a given One may accept the existence of God, or some more time, it still seemed to the philosopher that libertarian- robust set of religious doctrines, as a guide to life, trying ism and compatibilism were approximately equally to live in accordance with them, seeking to act and feel well supported, how could she simply decide to believe one’s way into a religious community, in order to one rather than another? How could we do that any determine how the doctrines work out in the living of more than, lacking any reasons at all for one alternative them, both in terms of how satisfactory and fulfilling a rather than the other, we decide to believe that the life they enable one to live and in terms of what evi- number of ultimate particles in the universe is even dence for or against them one acquires. Again, at least rather than odd? in early stages of this process, one does not yet believe The above account in terms of a momentary sense the doctrines in question. of conclusive support for one alternative could also There are other possibilities as well. S may be seek- apply to our practical cases. It could be that the military ing, for whatever reason, to bring herself into a position commander, at a certain point in his deliberations, of believing p, and she, or others, may confuse this comes to think the reasons for a particular hypothesis activity, which can be undertaken voluntarily, with concerning the disposition of enemy forces are conclu- believing the proposition to be true. Or S may align sive. But I believe that there are other construals for herself with some group – a church, a political party, a both the theoretical and practical cases. For one thing, group of thinkers – that is committed to certain doc- the subject may be resolving to act as though it is true that trines, and this, which can be done voluntarily, may be p, adopting it as a basis for action without actually confused with coming to believe those doctrines. believing it. This could well be a correct description of Finally, there is the distinction between acceptance and the military commander. He may have said to himself: belief that was briefly mentioned earlier. The basic dis- “I don’t know what the disposition of enemy forces is. tinction is that belief is something that one finds oneself I don’t even have enough evidence to consider one with, something that springs into consciousness spon- hypothesis much more likely than any other. But I have taneously when the question is raised. Whereas accept- to proceed on some basis or other, so I’ll just assume ance of a proposition is, at least in the first instance, a that it is H and make my plans accordingly”. If that’s deliberate voluntary act of accepting a proposition as the way the land lies, it would be incorrect to describe true. It differs from the “working hypothesis” or the commander as believing that the disposition of “assuming that p as a basis for action” in that, unlike enemy forces is H or having any other belief about the these cases, S does commit himself to p’s being true. 8 william alston

He “takes it on board” as one of the things he acts on it, or put some other part of my body into suitable and draws consequences from. It is, we might say, just contact with it (assuming that I lack telekinetic pow- like belief except that the commitment to p’s being ers), and this must result in the door’s coming to be true doesn’t arise spontaneously but, at least at the out- open. In order to inform II that p, I must produce vari- set, has to be kept in activation by a deliberate volun- ous sounds, marks, or other perceivable products, and tary act. Thus the philosopher and the religious seeker the product in question must fall under linguistic rules might accept, in this sense, a position on the free‐will in such a way as to constitute a vehicle for asserting that issue or the mind‐body problem or various religious p. Thus actions like these are not immediately conse- doctrines. The philosopher, even though libertarianism quent on a volition and are not strictly done “at will”. does not seem to him to be conclusively established, Nevertheless, I might be blamed for my failure to open might accept it – take it as his position on the issue, the door when it was my obligation to do so and I was defend it, draw various consequences for it, while seek- not prevented from performing basic bodily move- ing for conclusive evidence pro or con, and not yet ments sufficient to bring it about that the door was finding himself believing it. And there is an analogous open. In typical cases we take the extra conditions for possibility for religious doctrines.7 success for granted. We suppose that if the agent will Thus I take it that the analysis of a wide variety of just voluntarily exert herself in a way that is open to supposed cases of believing at will reveals that in each her, the act will be done. Here we can say that the case coming to believe that p may well have been con- action and its upshot are subject to the immediate volun- fused with something else. Hence I think that there is a tary control of the agent (more strictly, nonbasic immedi- considerable case for the position that no one ever ate voluntary control), even though more than an act of acquires a belief at will. will is required. I call this control “immediate” since the agent is able to carry out the intention right away, in one uninterrupted intentional act, without having to return to the attempt a number of times.10 I will use iii. Other Modes of Voluntary the term ‘direct control’ for both basic and nonbasic Control of Believing immediate control. If beliefs were subject to one’s direct control in either way, that would suffice to ren- The demise of basic control of belief is by no means the der them susceptible to deontological evaluation. end of voluntarism about belief. Many deontologists, But are beliefs always, or ever, within our immediate after avoiding any commitment to what they call “direct nonbasic voluntary control? As in the discussion of voluntary control of belief” (what I have called “basic basic control we can first exempt most beliefs from voluntary control”), insist that beliefs are subject to consideration. Where it is perfectly clear that a certain what they term “indirect voluntary control”.8 They proposition is true or false, as with typical perceptual, generally use this term in an undiscriminating fashion introspective, memory, and simple inferential beliefs, it to cover any sort of voluntary control that is not basic. is absurd to think that one has any such control over Hence they fail to distinguish the three kinds of nonba- whether one accepts, rejects, or withholds the proposi- sic control I will proceed to enumerate.9 Some of their tion. When I look out my window and see rain falling, examples fit one of my three types and some another. water dripping off the trees, and cars passing by, I no First, note that we take many nonbasic overt actions more have immediate nonbasic control over whether I and their upshots to be under voluntary control in a accept those propositions than I have basic control. I way that is sufficient for their being required, permit- form the belief that rain is falling willy‐nilly. There is ted, or prohibited. Consider opening a door, turning no way I can inhibit this belief or acquire a contrary on a light, and informing someone that p. Succeeding belief. At least there is no way I can do so on the spot, in any of these requires more than a volition; in each in carrying out an uninterrupted intention to do so. case I must make one or more bodily movements, and What button would I push? I could try asserting the these movements must have certain consequences. In contrary in a confident tone of voice. I could rehearse order for me to open a door, I must pull it, push it, kick some skeptical arguments. I could invoke the Vedantic deontological desiderata 9 doctrine of Maya. I could grit my teeth and command proposition is true or false. After all, they say, that is myself to withhold the proposition. But none of these what inquiry is for, to resolve such issues. One certainly will have the least effect on my doxastic condition. has voluntary control over whether to keep looking for Since cases in which it seems obvious to the subject evidence or reasons, and voluntary control over where what is the case constitute an enormously large pro- to look, what steps to take to find relevant considera- portion of propositional attitudes, the above considera- tions, and so on. It is suggested, in effect, that since we tions show that immediate nonbasic voluntary control have voluntary control over these intermediate steps, cannot be the basis for the application of deontological this amounts to what I call long‐range voluntary control of concepts to most of our beliefs and withholdings. a propositional attitude. Chisholm, for example, says: But what about situations in which it is not clear whether a proposition is true or false? Here I can sim- If self‐control is what is essential to activity, some of our ply refer the reader back to the last section, in which I beliefs, our believings, would seem to be acts. When a man argued with respect to basic control that the cases in deliberates and comes finally to a conclusion, his decision which it may look as if one comes to believe a proposi- is as much within his control as is any other deed we tion at will are best construed in other ways. In those attribute to him. If his conclusion was unreasonable, a cases involving the philosopher, the general, and the conclusion he should not have accepted, we may plead gardener, it is, I claim, implausible to suppose that the with him: “But you needn’t have supposed that so‐and‐so subject acquired a belief voluntarily, whether by a mere was true. Why didn’t you take account of these other facts?” We assume that his decision is one he could have act of will or by a series of basic or more nearly basic avoided and that, had he only chosen to do so, he could actions that led right away to the intended result. Here, have made a more reasonable inference. Or, if his conclu- as with the obviously true or false cases, we are at a loss sion is not the result of a deliberate inference, we may say, to think what button to push, what bodily movements “But if you had only stopped to think”, implying that, had to make so as to bring about the formation of an he chosen, he could have stopped to think. We suppose, as intended belief. Until some plausible story can be told we do whenever we apply our ethical or moral predicates, as to what one can do voluntarily to result in a belief’s that there was something else the agent could have done being formed immediately, we can ignore the possibil- instead. (1968, 224) ity of treating voluntary control of beliefs on the model of nonbasic but immediate voluntary control of doors To be sure, the mere fact that one often looks for evi- being open and lights being on. dence to decide an unresolved issue does not show that This brings us to a second grade of what is com- one has voluntary control over one’s propositional atti- monly called “indirect voluntary control”, what I will tudes. That would also depend, at least, on the inci- call long‐range voluntary control. It will be noted that the dence of success in these enterprises. And sometimes types of voluntary control I am considering are one finds decisive evidence and sometimes one doesn’t. arranged in an order of increasing indirectness, increas- But let’s ignore that complexity and just consider ing distance from the most immediate control. Here, as whether there is a case for long‐range voluntary con- with immediate nonbasic voluntary control, we think trol of belief in the successful cases. of the belief as being produced by the carrying out of No, there is not, and primarily for the following rea- an intention by one or more actions that are designed son. Claims like those in the quote from Chisholm to produce the belief rather than as being produced by ignore the difference between doing A in order to a mere act of will, choice, or decision. But unlike the bring about E, for some definite E, and doing A so that last case, the belief production is not carried out in one some effect within a certain range will ensue. In order uninterrupted action. It involves a series of actions that the phenomenon of looking for more evidence spread out over a greater or smaller period of time, the would show that we have voluntary control over prop- smallest period of which is too extended to accom- ositional attitudes, it would have to be the case that the modate a single uninterrupted act.11 A number of vol- search for evidence was undertaken with the intention untarists seem to be thinking in these terms of the cases of taking up a certain attitude toward a specific proposi- in which it is not immediately obvious whether a given tion. For only in that case would it have any tendency 10 william alston to show that we have exercised voluntary control over is where the only voluntary control I have over my what propositional attitude we come to have. Suppose propositional attitudes is to enter onto an investigation that I can’t remember Al Kaline’s lifetime batting aver- that will eventuate in some propositional attitude or age, and I look it up in the baseball almanac. I read other on what is being considered. there the figure .320, and I thereby accept it. Does that Or consider propositions concerning what is visible. show that I have voluntary control (of any sort) over I have the power to voluntarily open my eyes and look my belief that Kaline’s lifetime batting average was about me, thereby putting myself in a position, when .320? Not at all. At most it shows that I have long‐ conditions are favorable, to reliably form propositions range voluntary control over whether I take up some about the visible environment. Again, with respect to propositional attitude toward some proposition ascrib- past experiences, I can “search my memory” for the ing a lifetime batting average to Kaline. So this is not at details of my experiences of the middle of yesterday, all parallel to cases where we definitely do have some thereby usually putting myself in a good position to (albeit fallible) long‐range voluntary control over other form beliefs reliably about my experiences at that time. sorts of affairs. Suppose that I can perform voluntary No one, I suppose, would take these facts to show that actions that will result, subject to the usual chances that I have voluntary control over what I believe about the infect all human endeavor, in my losing twenty pounds. visible environment or about my remembered experi- Here there is a completely definite and unique result ences. What I can control voluntarily is whether I form toward which my voluntary efforts are directed, and (or am in a position to form) some accurate beliefs or success, or at least repeated success, will show that I do other about my current visible environment or about have long‐range voluntary control (within limits) of my experiences of yesterday. And yet this is the same my weight. sort of thing as the search for additional evidence of What the situation described by Chisholm is closely which Chisholm speaks, differing only in the type of analogous to is the following. I am a servant, and I am belief‐forming mechanisms involved. motivated to bring the door into whatever position my I suspect that those who take positions like the one employer chooses. He has an elaborate electronic sys- in the passage just quoted from Chisholm secretly sup- tem that involves automatic control of many aspects of pose that the additional evidence, rather than “auto- the household, including doors. Each morning he matically” determining the propositional attitude, leaves detailed instructions on household operations in simply puts the subject in a position to make an a computer. Doors can be operated only through the informed choice of an attitude. That is, they really computer in accordance with his instructions. There is locate the voluntary control in the moment of attitude no way I can carry out an intention of my own, no formation rather than in the preliminary investigation, matter how long range, to open or to close a particular thereby in effect taking the (basic or nonbasic) imme- door at a particular time. All I can is to actuate the rel- diate‐control position. But then, faced with the implau- evant program and let things take their course. Since sibility of those positions, they think to save the the employer’s instructions will be carried out only if I application of deontological concepts to beliefs by actuate the program, I am responsible for the doors’ pushing the voluntary control back to the preliminary assuming positions he specified, just as in the Kaline search for decisive considerations. But their undercover case I was responsible for taking up some attitude or attachment to the immediate‐control thesis prevents other toward some proposition within a given range. them from seeing that voluntary control of the investi- But I definitely am not responsible for the front door’s gative phase has no tendency to ground the deonto- being open rather than closed at a particular time, nor logical treatment of propositional attitudes themselves. can I be said to have voluntary control over its specific Despite the above arguments against false preten- position. Hence it would be idle to apply deontological sions to the title of “long‐range voluntary control of concepts vis‐à‐vis the specific position of the door: to belief”, I have no intention of suggesting that there forbid me or require me to open it, or to blame or could not be legitimate claimants. Let’s take a fresh start reproach me for its being open. I had no control over and lay out what it takes for a genuine case of such that; it was not subject to my will. And that’s the way it control in general (not restricted to beliefs). It requires deontological desiderata 11 the capacity to bring about a state of affairs, C, by vol- constituted that the most I can do with respect to my untarily doing a number of different things over a con- irritability is to make it slightly less likely that it will siderable period of time, typically interrupted by exceed a certain high average threshold, I can hardly be activity directed to other goals. One has this sort of blamed for being irritable. control, to a greater or lesser degree, over many things: It is very dubious that we have a reliable long‐range one’s weight, cholesterol concentration, blood pressure, voluntary control over any of our beliefs, even in the and disposition; the actions of one’s spouse or one’s off- most favorable cases, such as beliefs about religious and spring. One can, with some hope of success, set out on philosophical matters and about personal relationships. a long‐range project to reduce one’s weight, improve Sometimes people succeed in getting themselves to one’s disposition, or get one’s spouse to be more believe (disbelieve) something. But I doubt that the friendly to the neighbors. The degree of control one is success rate is substantial. I know of no statistics on this, likely to have varies markedly among these examples. but I would be very much surprised if attempts of this But all these examples and many more illustrate the sort bore fruit in more than a small proportion of the point that one can have long‐range control over many cases. In thinking about this, let’s first set aside cases in things over which one lacks immediate control. I can- which the attempt succeeds because the subject hap- not markedly reduce my weight right away by the pens onto conclusive evidence that would have pro- uninterrupted carrying out of an intention to – by duced the belief anyway without deliberate effort on ­taking a pill, running around the block, or saying his part to produce that belief. Thus we need to con- ‘Abracadabra’. But that doesn’t nullify the fact that I sider cases in which the subject is swimming against have some degree of long‐range control. either a preponderance of contrary evidence or a lack To return to our main concern, it does seem that we of sufficient evidence either way. That is, S is fighting have some degree of long‐range voluntary control over very strong tendencies to believe when and only when at least some of our beliefs. People do set out on long‐ something seems true to her. Whether these tendencies range projects to get themselves to believe a certain are innate, or engendered and reinforced by socializa- proposition, and sometimes they succeed in this. tion, they are deeply rooted and of great strength. To Devices employed include selective exposure to evi- combat them one must exercise considerable ingenuity dence and deliberate attention to supporting consid- in monitoring the input of information and in expos- erations, seeking the company of believers and avoiding ing oneself to nonrational influences. These are tricky nonbelievers, self‐suggestion, and more bizarre meth- operations, and it would be very surprising if they were ods like hypnotism. By such devices people sometimes successful in a significant proportion of cases. I am not induce themselves to believe in God, in materialism, in suggesting that it is unusual for people to form and communism, in the proposition that they are loved by retain beliefs without adequate grounds. That is all too X, and so on. Why doesn’t this constitute a kind of common. But in most such cases the proposition in voluntary control that grounds deontological treat- question seems clearly true to the person, however ill ment of beliefs? supported. The typical case of prejudice, for example, is Well, it would if we do have sufficient control of this not one in which S manages to believe something con- sort. People could properly be held responsible for trary to what seems to him to be the case or something their attitudes toward propositions in a certain range concerning which he has no definite impression of only if those who set out to intentionally produce a truth or falsity. It is a case in which his socialization has certain attitude toward such a proposition and made led it to seem clearly true to him that, for example, sufficient efforts were frequently successful. For only if blacks are innately inferior. we were generally successful in bringing about goal G Thus the possibility of long‐range voluntary control when we try hard enough to do so, do we have effec- of beliefs does not provide significant grounding for tive control over whether G obtains. And if I don’t have deontologism, even for the sorts of propositions people effective control over G, I can hardly be held to blame do sometimes try to get themselves to believe or disbe- for its nonoccurrence. This is a generally applicable lieve. Much less is there any such support for deontolo- principle, by no means restricted to beliefs. If I am so gism for those propositions with respect to which 12 william alston people don’t normally even try to manipulate their the effect that Jim is trying to undermine Susie’s position attitudes. We have already noted that most of our as departmental chair. If I had done what I should have beliefs spring from doxastic tendencies that are too done by way of checking into this matter, I would not deeply rooted to permit of modification by deliberate have formed that belief or would not have retained it effort. In such cases the project of deliberately pro- for as long as I did. Hence I could be blamed for hold- ducing belief or disbelief is one that is never seriously ing the belief. To take a case where I am blameless in envisaged. Thus, even if we were usually successful holding a belief, consider a visual belief where my when we set out to produce a propositional attitude, vision and my belief‐forming mechanisms are working the voluntary control thus manifested would not normally. There is nothing relevant to that belief for- ground the application of deontological concepts to mation that I should have done but didn’t, and so I am beliefs generally. not subject to blame in forming the belief. Note that other deontological terms like ‘ought’ and ‘should’ are also applicable to states of affairs not themselves under direct voluntary control in this derivative way. Thus we iv. Indirect Voluntary Influence can say that I ought to have a lower cholesterol count on Believing and that I should not have believed that he did it. Note that this kind of application of the deontologi- Up to this point I have been considering various ways cal categories of blameworthiness or the reverse to in which believing, rejecting, and withholding proposi- believings is a derivative one. It is the (actual or possi- tions might be themselves under effective voluntary ble) voluntary acts in the causal ancestry of the belief to control. We have seen that for most of our beliefs we which blameworthiness and other deontological terms have no such control and that for the others we have at of evaluation apply in a primary way. My checking, or most some spotty and unreliable control of the long‐ failing to check, on the accuracy of the gossip is some- range sort. But this is not the end of the line for the thing that is directly, underivatively blameworthy or prospects of an epistemic desideratum of the deonto- the reverse. The propositional attitude that eventuates is logical kind. There is still one more way in which sub- blameworthy or the reverse only by derivation from jects might be held responsible for their believings, for the voluntary acts that give rise to it. Strictly speaking, believings to be required, forbidden, or permitted, even in thinking of the matter in this way we should not though they themselves are not under effective volun- think of the forming of the propositional attitude itself tary control. To see this, consider the general point that as required, forbidden, or permitted since it itself is not we can be blamed for a state of affairs F, provided some- under effective voluntary control. Its blameworthiness thing we voluntarily did (didn’t do) and should have or blamelessness stems from the required, forbidden, or not done (done) was a necessary condition (in the cir- permitted voluntary acts in its causal ancestry. Harking cumstances) of the realization of F. That is, F would not back to the initial list of candidate deontological desid- have obtained had we done (not done) something we erata, we have now moved from should have done (not done). If my cholesterol buildup would have been prevented had I regulated my diet in 9. B is held permissibly (one is not subject to blame the way I should have done, but didn’t, I can be blamed for doing so). for that buildup, whether or not I have direct effective to voluntary control of my cholesterol level. 11. The causal ancestry of B does not contain violations Applying this general point to beliefs, we can say of intellectual obligations. that even though believings are not under effective vol- untary control, we can be blamed for holding a certain I will call this kind of impingement of the voluntary on belief, B, if there are things we can voluntarily do such belief indirect voluntary influence.12 that we should have done (not done) them and if we I will now proceed to put a bit of flesh on the idea of had done (not done) them we would not have held an indirect influence of voluntary actions on proposi- that belief. Suppose that I accept some idle gossip to tional attitudes. First, let’s note that we do have voluntary deontological desiderata 13 control over many actions that can influence our believ- Thus it will sometimes be the case when I believe ings, rejectings, and withholdings. These can be divided that p that I would not have done so had I done various into two groups: (a) activities that bring influences to bear, things in the past that I could and should have done but or withhold influences from, a particular candidate, or failed to do, and it will sometimes be the case that I field of candidates, for belief, and (b) activities that affect would not have believed that p had I not done various our general belief‐forming habits or tendencies. There are things in the past that I could and should not have many examples of (a). With respect to a particular issue, I done but did. In either of these cases there is a failure have voluntary control over whether and how long I con- of obligations in the causal ancestry of the belief that sider the matter, over whether and where I look for rele- renders me blameworthy for having the belief. And if vant evidence or reasons, reflect on a particular argument, neither of these is the case, then I am blameless, not seek input from other people, search my memory for properly held to blame for the belief. (All this applies analogous cases, and so on. Here we come back to the equally to rejectings and withholdings). Hence the activities we saw Chisholm wrongly classifying as the indirect voluntary influence on the formation of prop- intentional inauguration of an attitude toward a specific ositional attitudes does have an evaluative bearing on proposition. Group (b) includes such activities as training those attitudes, either positive or negative. myself to be more critical of gossip, instilling in myself a I must pause to refine the above formulation. There stronger disposition to reflect carefully before making a are certain ways in which dereliction of intellectual judgment on highly controversial matters, talking myself duty can contribute to belief formation without ren- into being less (more) subservient to authority, and prac- dering S blameworthy for forming that belief. Suppose ticing greater sensitivity to the condition of other people. that I fail to carry out an obligation to spend a certain It is within my power to do things like this or not, and period in training myself to look for counterevidence. when I do them with sufficient assiduity I make some I use the time thus freed up to take a walk around the difference to my propositional attitude tendencies, and neighborhood. In the course of doing so I see two dogs thus indirectly to the formation of such attitudes. fighting, thereby acquiring the belief that they are There would be no harm in including in (a) attempts fighting. There was a relevant intellectual obligation I to bring about a particular attitude to a specific proposi- didn’t fulfill, which is such that if I had fulfilled it I tion. For these too would be things that influence our wouldn’t have acquired that belief. But if that is a per- propositional attitudes and over which we have volun- fectly normal perceptual belief, I am obviously not to tary control. The point of stressing other things is that blame for having formed it.13 since the earlier discussion provided reason for thinking Here the dereliction of duty contributed to belief such attempts are rarely successful, I want here to empha- formation simply by facilitating access to data. That is size the point that even if we are never successful in not the kind of contribution we had in mind in the voluntarily bringing about a belief that p, there are still above formulations. The sorts of cases we had in mind many things we can do voluntarily that do have a bear- were those most directly suggested by the two sorts of ing on what propositional attitudes are engendered. voluntary activities that affect belief formation: (a) The next question is whether the deontological triad those that involve looking for considerations relevant of concepts applies to activities like those canvassed in to the belief in question, or not doing so, and (b) those the next‐to‐last paragraph. Is it ever the case that we that affect our general belief‐forming habits or tenden- ought or ought not to engage in an activity of these cies. By revising 11 so as to make this explicit, we can sorts, such as searching for new evidence or critically avoid counterexamples like the above. examining the credentials of gossip? Is it ever the case that we ought or ought not to strive to make ourselves 11A. S is intellectually to blame for believing that p iff more (less) sensitive to contrary evidence? Deontologists if S had fulfilled all her intellectual obligations, typically hold that we have intellectual obligations in then S’s access to relevant considerations, or S’s such matters, obligations rooted in our basic intellectual belief‐forming habits or tendencies, would have obligation to seek the true and avoid the false in belief. changed in such a way that S would not have I accept this view, which seems eminently plausible. believed that p.14 14 william alston

It follows from the above that 11A is a genuine intel- intellectual obligation. What makes this difficult is that lectual desideratum, a desirable feature of belief outputs attempts to conform to an obligation might or might of cognition. And it also follows from the above discus- not be successful, and where they are not they might be sion that, unlike 9, 11A does not fail to qualify as an more or less close to achieving it. On a sufficiently rig- epistemic desideratum because it is not a real possibility orous interpretation we are almost always failing in for human beings. But though it is a cognitive desid- some intellectual obligation or other. This in turn eratum, it will still not be an epistemic desideratum if it depends on just how it is specified what one is intel- is not connected in the right way with the truth goal. lectually obliged to do. Consider the obligation to look And how do we determine whether it is? Since we for relevant considerations pro and con when it is not have identified three groups of cognitive desiderata that clear whether the proposition in question in true. How are, in their several ways, related to the truth goal so as long and how assiduously does one have to look to to qualify as epistemic desiderata, an obvious way to fulfill the obligation? To require that every conceivably proceed is to consider whether 11 is related to the relevant consideration must be taken into account truth goal in one of those ways. would be a counsel of perfection that is beyond any of I think we can straightaway eliminate the Group III our powers, not to mention the fact that we couldn’t and the Group V ways from consideration. As for III, it know whether that limit had been reached. It seems is obvious that where there are no violations of intel- that to make the notion of fulfilling intellectual obliga- lectual obligations in the ancestry of a considerable tions usable we have to build in a limitation to what number of S’s beliefs, this in no way provides S with could reasonably be expected of a subject, and that is resources for determining under what conditions a itself is a very imprecise notion and one that is subject belief is likely to be true and thereby putting S in a to varying interpretations. Moreover, there is the point good position to restrict belief formation to true that what could be reasonably expected along this line beliefs, in the way higher‐level epistemic knowledge or will vary for different people in accordance with their the capacity for such does. Nor does 11A presuppose abilities, experience, education, propensities, and so on. such a capacity as 8 does. As for Group V, although it So the question whether a given subject has done as could be argued that fulfillment of intellectual obliga- much as could be expected of him or her is beset with tions has an intrinsic cognitive value that is independ- uncertainties, imprecision, and disagreements. ent of truth, as 12–15 do, value is in no way dependent Keeping all this in mind, let’s do the best we can by on being associated with a preponderance of true proceeding on the basis of some sense of what could be beliefs, as is the case with the Group V desiderata. Even expected of a given subject and on the basis of some if S’s beliefs are preponderantly false, it is a good thing plausible construal of the content of intellectual obliga- for S to fulfill her intellectual obligations, insofar as this tions. I now want to suggest that there are very many is possible in the face of something less than an out- sorts of cases in which one does as much as could be standing record of achieving the truth. reasonably expected of one in the way of voluntary acts So this leaves us with Group II, the directly truth‐ leading up to a given belief without the belief’s thereby conducive desiderata. Though I will argue that 11A is acquiring any considerable likelihood of truth. not related to the truth goal in this way either, I do not First, there are cases of cognitive deficiency. Consider think that this is obvious on the face of it, as is the case one who forms the belief that socialism is contrary to with the Group III and Group V desiderata. Hence this Christianity for the reasons that are often given for this suggestion deserves more serious consideration. It is view by the Christian Right, and is intellectually inca- prima facie conceivable that being formed in a way pable of figuring out how bad these reasons are, cases that does not depend on violations of intellectual obli- that I fear are all too common. Such a person may have gations should be a way of rendering a belief probably done as well as could be expected of him in coming to true. Nevertheless, there are many counterexamples to this belief, but that fact does nothing to make the belief such a claim, some of which I shall now make explicit. likely to be true. (Thoroughly bad reasons are not truth‐ Before we embark on this, something needs to be conducive). Or consider a college student who doesn’t said about what would constitute failure to fulfill an have what it takes to follow abstract philosophical deontological desiderata 15 exposition or reasoning. Having read parts of Bk. IV of belief tendency, I can hardly be expected to do so. But Locke’s Essay Concerning Human Understanding, he takes some irresistible beliefs are formed in an unreliable it that Locke’s view is that everything is a matter of fashion in such a way as not to make them likely to be opinion. He is simply incapable of distinguishing true. The most obvious examples concern strong emo- between that view and Locke’s view that one’s knowl- tional attachments that are unshakable. For many peo- edge is restricted to one’s own ideas. There is nothing ple their religious or irreligious beliefs have this status, he could do that would lead him to appreciate the dif- as do beliefs concerning one’s country, one’s close rela- ference. Hence he cannot be blamed for interpreting tions, or one’s political party. Such beliefs are often not Locke as he does; he is doing the best he can to fulfill formed in a truth‐conducive way that would render his intellectual obligations. But his belief about Locke’s them likely to be true. But S cannot be blamed for view is outrageously ill‐grounded, based as it is on his doing or having something she can’t help doing or dim‐witted impression of Bk. IV of the Essay. having. Here too the believing’s not depending on any Second, consider the innumerable beliefs each of us violation of intellectual obligations is no (even fallible) forms on testimony or authority. Practically everything guarantee of the beliefs likelihood of being true. we believe about science, history, geography, and current Of course, showing that 11A is not related to the affairs is taken on authority. Ideally, we would check out truth goal in any of the ways the desiderata already each source to make sure that it is reliable before accept- approved as epistemic are is not a proof that it is not an ing the testimony. But who has time for that? We can do epistemic desideratum. Perhaps it is related to the truth it in special cases where the matter is of special impor- goal in some other way that qualifies it as epistemically tance, but it is not a real option for such beliefs generally. desirable. That is an abstract possibility. Why shouldn’t If we tried to do so, our doxastic structure would be so there be a fourth way, as different from the first three as impoverished that we would not be able to function in they are from each other? But I must confess that I society. Moreover, even if we had time to check up on have found no fourth way. And until I do, I must rest each authority, in most cases we lack the resources for with the conclusion that 11A does not qualify as an making an informed judgment. Thus in most cases in epistemic desideratum. which I uncritically accept testimony I have done as Thus none of the deontological candidates makes much as could reasonably be expected of me. Now con- the grade. The first three (alternative versions of 9) fail sider those cases in which the authority is incompetent through the failure of the version of voluntary control or the witness is unreliable. There we are forming a belief of belief presupposed by each to be a real possibility for on an objectively unreliable basis and hence the belief human beings. And the last (11A) fails through not is not probably true, even though no dereliction of being connected with the truth goal in the right kind intellectual duty is in the background. of way. Thus they will receive no further attention in Next consider irresistible beliefs and belief tenden- the development of the ED approach to the epistemol- cies. If it is impossible for me to alter a certain belief or ogy of belief.

Notes

* Editors’ note: in Chapter 3 of the book from which this despite the rules. When I speak of violating a rule as selection is taken, Alston lists fifteen candidate “features of being blameworthy, it is presupposed that there is no belief that could be claimed, with some considerable plau- such excuse. sibility, to be desiderata from an epistemic point of view” 2 See, e.g., Ginet 1985 and Meiland 1980. (40). Those that pass muster he calls “epistemic desiderata.” 3 On the basis of a distinction between believing that p Here he considers those candidate desiderata he terms and “accepting” that p, according to which the latter “deontological.” but not the former is a voluntary action, I have argued 1 I don’t suggest that doing what is not permitted by the that these philosophers and others are best construed as rules is coextensive with being subject to blame for ascribing voluntary control to accepting, not believing. doing it. One might have a valid excuse for doing it See Alston 1996. 16 william alston

4 The best‐known defense of the logical impossibility is But there are enough clear cases on either side of the Bernard Williams’s “Deciding to Believe,” in Williams distinction to make it usable. 1972. It has been criticized in, inter alia, Govier 1976 12 This kind of relation of voluntary action to belief is and Winters 1979. given much less attention in the literature than the kinds 5 A bit later in the discussion I will present other tempting I have judged to lack viability. Nevertheless, as an impor- conflations. tant aspect of cognitive endeavors, it deserves more 6 Even if beliefs can be formed at will in these kinds of attention than it receives. I should also mention that in cases, there still remain vast stretches of our beliefs, Alston 1988 I argue that a concept of epistemic justifica- including all the cases discussed above, where it seems tion based on this kind of voluntary influence on belief obvious what is the case, where we have already seen is not adequate, not because there is no such influence, believing at will not to be a possible move. And so it but rather because it is not closely enough related to would still be true that believing permissibly would not the goal of true belief. Though I am not concerned be generally viable as an epistemic desideratum. here with how to construe epistemic justification, I will 7 The latter application is explored in detail in Alston make a similar point about treating 11 as an epistemic 1996. desideratum. 8 See, e.g., Alvin Goldman 1980, Plantinga 1983, 13 I am indebted to Emily Robertson for calling this prob- Wolterstorff 1983, Moser 1985, Steup 1988. lem to my attention. 9 Even the extended treatment in Pojman 1986 fails to 14 Another fine‐tuning point has to do with the “absolute- make any distinctions within “indirect control.” ness” of the counterfactual involved. Desideratum 11A 10 Of course, opening a door or turning on a light may, in says S would not have believed that p under these condi- special cases, require repeated attempts with intervals tions. But perhaps S is also blameworthy for believing between. In the above, I was speaking of the simple that p even if it is only much less likely that S would have unimpeded cases. believed that p under these conditions. I am inclined to 11 Obviously, there is no precise boundary between a accept this weaker interpretation, but I will not press the ­“single uninterrupted act” and a “series of temporally matter here since I will go on to reject 11 as an epistemic extended acts” where the temporal extension is small. desideratum anyway.

References­

Alston, William. 1988. “The Deontological Conception of Meiland, Jack. 1980. “What Ought We to Believe? The Ethics of Epistemic Justification.” Philosophical Perspectives 2:257−99. Belief Revisited.” American Philosophical Quarterly 17: 15−24. Reprinted in Alston 1989. Moser, Paul. 1985. Empirical Justification. Dordrecht: D. Reidel. Alston, William. 1996. “Belief, Acceptance, and Religious Plantinga, Alvin. 1983. “Reason and Belief in God.” In Faith Faith.” In Faith, Freedom, and Rationality, ed. Jeff Jordan and Rationality, ed. Alvin Plantinga and Nicholas Wolterstorff. and Daniel Howard‐Snyder. Lanham, Md.: Rowman & Notre Dame: University of Notre Dame Press. Littlefield. Pojman, Louis. 1986. Religious Belief and the Will. London: Chisholm, Roderick. 1968. “Lewis’ Ethics of Belief.” In The Routledge & Kegan Paul. Philosophy of C. I. Lewis, ed. P. Schilpp. La Salle, Ill.: Open Steup, Matthias. 1988. “The Deontic Conception of Epistemic Court. Justification.” Philosophical Studies 53:65 − 84. Chisholm, Roderick. 1977. Theory of Knowledge, 2d ed. Williams, Bernard. 1972. “Deciding to Believe.” In Problems of Englewood Cliffs, N.J.: Prentice‐Hall. the Self. Cambridge: Cambridge University Press. Ginet, Carl. 1985. “Contra Reliabilism.” The Monist Winters, Barbara. 1979. “Believing at Will.” Journal of 68:175−87. Philosophy 76:243 − 56. Goldman, Alvin. 1980. “The Internalist Conception of Wolterstorff, Nicholas. 1983. “Can Belief in God Be Rational Justification.” Midwest Studies in Philosophy 5:27−53. if It Has No Foundations?” In Faith and Rationality, ed. Govier, Trudy. 1976. “Belief, Values, and the Will.” Dialogue Alvin Plantinga and Nicholas Wolterstorff. Notre Dame: 15:642−63. University of Notre Dame Press. 2 Voluntary Belief and Epistemic Evaluation

Richard Feldman

I. Introduction ­“natural” to understand epistemic justification in a “deontological” way. By this he means that it is to Our talk about epistemic matters parallels our talk be understood in terms of “obligation, permission, about ethical matters in noteworthy ways. Among the requirement, blame, and the like.”1 Apparent examples ethical judgments we make are judgments that a person of deontological analyses include one derived from A. ought to perform a certain action, that someone should J. Ayer’s proposal that one has knowledge when one 2 not do a certain thing, that people have obligations to act has a true belief and one has a “right to be sure.” On in some ways, that they are permitted or required to do this view, one is justified in believing a proposition certain things, that they have a right to do one thing and when one has a right to be sure that it is true. Carl a duty to do another, and that sometimes they deserve Ginet has proposed that praise or blame for what they have done. We make seem- One is justified in being confident that p if and only if it is ingly analogous judgments about beliefs and believers. not the case that one ought not be confident that p; one We say that a typical well‐informed contemporary could not be justly reproached for being confident that p. 3 American ought to believe that the Earth revolves around the Sun and should not believe that the Earth is Roderick Chisholm appeals to the notion of an flat. A person wrongly accused of a crime might say his “intellectual requirement” in an effort to clarify the accusers have no right to believe that he’s guilty, since key concepts of epistemological theory.4 Analyses of no evidence of his wrongdoing has been brought forth. epistemic justification in deontological terms are In such a case, we might say that believing the person common.5 is not guilty is permitted or perhaps even required. We Recently, Alvin Plantinga and Alvin Goldman have sometimes praise those who believe the things they independently argued that the viability of a deonto- should, and we criticize those who fail in their believ- logical conception of epistemic justification is crucial ings. We can describe all these judgments as deontological to the debate between internalists and externalists judgments about beliefs. about epistemic justification. Goldman thinks that a Much work in recent epistemology has been about central, but mistaken, line of support for internalist epistemic justification. As William Alston says, it is theories begins with the assumption of a deontological

Original publication details: Richard Feldman, “Voluntary Belief and Epistemic Evaluation.” Reprinted from Knowledge, Truth, and Duty: Essays on Epistemic Justification, Responsibility, and Virtue, edited by Matthias Steup, pp. 77−92. Copyright © 2001 by Oxford University. Used by permission of the publisher, Oxford University Press.

Contemporary Epistemology: An Anthology, First Edition. Edited by Jeremy Fantl, Matthew McGrath, and Ernest Sosa. © 2019 John Wiley & Sons, Inc. Published 2019 by John Wiley & Sons, Inc. 18 richard feldman account of justification.6 Plantinga also argues that epistemic deontologism begins with the claim that internalism derives much of its support from a deonto- believing is not a voluntary activity. This is used to sup- logical view of justification.7 Whether deontological port the proposition that deontological judgments judgments about beliefs are ever true thus is of consid- about beliefs cannot be true. And from this it is inferred erable epistemological significance. that the deontological conception of epistemic justifi- A central problem that both Plantinga and Alston cation is not viable. My concern in this essay will be find with deontological judgments about beliefs is that primarily with the first inference, the move from invol- they presuppose that we have voluntary control over untarism about belief to the falsity of all deontological what we believe. Yet, reflection on our mental lives sug- judgments about beliefs. My goal in what follows is gests that we have no such control. Alston says: to argue that deontological judgments about beliefs can be true even if beliefs are involuntary. I will not this conception of epistemic justification is viable only if be defending deontological analyses of epistemic beliefs are sufficiently under voluntary control to render justification. such concepts as requirement, permission, obligation, reproach, We can formulate the issue in terms of the following and blame applicable to them. By the time honored princi- argument: ple that “Ought implies can,” one can be obliged to do A only if one has an effective choice as to whether to do A.8 The Voluntarism Argument 1. People do not have voluntary control over their He goes on to argue that we don’t have an effective beliefs. choice over what we believe. In the process of object- 2. If deontological judgments about beliefs are some- ing to Chisholm’s views about justification, Plantinga times true, then people have voluntary control over says of a particular proposition that “whether or not I their beliefs. accept it is simply not up to me; but then accepting this 3. Deontological judgments about beliefs are not proposition cannot be a way in which I can fulfill my sometimes true. obligation to the truth, or, indeed, any obligation.…”9 Thus, according to Plantinga, our lack of control over Epistemologists have three kinds of response to this beliefs implies that they are not the sort of thing that argument open to them: (i) They can argue that we do can be a matter of obligation, and this undermines have the requisite sort of control over our beliefs, Chisholm’s deontological conception of epistemic jus- thereby rejecting premise (1); (ii) they can argue that tification. Matthias Steup presents a similar argument, the deontological judgments do not have voluntarist though he goes on to defend the deontological con- implications, thereby rejecting premise (2); or (iii) they ception on the grounds that belief is voluntary.10 can accept the argument and admit that the familiar For the purposes of the discussion that follows, it deontological terms of epistemic appraisal really are will be helpful to distinguish two steps of the argu- inapplicable.11 This in itself is a surprising conclusion, ments just presented. Their target is a deontological whatever the merits of deontological analyses of epis- conception of epistemic justification, a conception temic justification. according to which epistemic justification is to be I will discuss response (i) in section II and response understood or analyzed in terms of the deontological (ii) in section III. I will not discuss response (iii) except concepts of obligation, requirement, and the like. This in passing and by implication. conception is “viable,” in Alston’s terms, only if belief is sufficiently under our voluntary control. Presumably, the deontological conception is viable only if it can be II. Voluntarism About Belief true that we are required to believe things, that we ought not believe other things, and so on. That is, In this section, I will discuss premise (1) of The the deontological conception of epistemic justification Voluntarism Argument. I will eventually argue that we is “viable” only if deontological judgments about do have voluntary control over some of our beliefs, beliefs are sometimes true. Thus, the argument against but I will also argue that this fact is of absolutely no voluntary belief and epistemic evaluation 19 epistemological significance and that it does nothing to Since we in fact have control over some beliefs, it is help resolve the real puzzle concerning voluntarism possible to have control. So, the conceptual impossibility and deontological judgments about beliefs. thesis is false.

The Contingent Inability Thesis Alston has given the most Arguments for the Conceptual thorough defense of the contingent inability thesis, the Impossibility of Voluntary Belief thesis that as a contingent matter of fact, people are not able to acquire beliefs voluntarily. Alston’s paper includes The philosophical literature contains two quite differ- an excellent survey of a variety of notions of voluntary ent kinds of defenses of the claim that people do not control. For each type except one, he argues that we lack have voluntary control over their beliefs. One defense that sort of control over beliefs. Alston admits that there is relies on the idea that voluntarily formed beliefs are a one very weak notion of control that does apply to belief. conceptual impossibility, and the other claims merely But he contends that this sort of control does not provide that it is a contingent fact that we are unable to believe the basis for a good response to theVoluntarism Argument. voluntarily. I will discuss each sort of defense, but most Alston begins by discussing basic voluntary control.17 We of my attention will be directed on the second one. have basic voluntary control over those actions that we The Conceptual Impossibility Thesis According to the can “just do.” Simple bodily motions are the prime conceptual impossibility thesis, the nature of belief examples. I can just raise my hand, close my eyes, and somehow rules out the possibility that it is a state vol- bend my knee. Some people, but not I, can wriggle their untarily entered into. Arguments for this surprising ears and curl their tongue. Alston correctly says that thesis have found their supporters, notably Bernard forming a belief is not like that. We can’t just do it at will. Williams12 and more recently, Dion Scott‐Kakures.13 When we have decisive evidence for a proposition, we Jonathan Bennett has also expressed support for the typically can’t help but believe it. When we have decisive thesis, though, in a thoroughly admirable confession he evidence against a proposition, we typically can’t believe admits that the arguments for it are all no good.14 The it. When our evidence is not conclusive, we typically heart of the argument is that belief is by its nature can’t help but believe what our evidence seems to sup- directed at the truth. If beliefs could be acquired at will, port. I think that Alston is entirely right about this. Of they could be acquired independently of all truth con- course, we don’t always follow our evidence in the way siderations. But any state entered into independently of just described. Sometimes our hopes or our fears get the truth considerations is by definition not a belief. best of us and we believe contrary to our evidence. But The argument is far more complex than the sketch we don’t believe at will in these cases either. The contrast just given. However, I won’t pursue it here. I think between simple actions and beliefs is striking. In my cur- Williams’s argument has been effectively criticized by a rent, relatively normal circumstances, if I want to raise number of writers, as has the more recent variation my hand, I can just do it. In contrast, if I want to believe offered by Scott‐Kakures.15 Alston remarks that he that it will start raining shortly, I cannot just do it. The “cannot see any sufficient reason for the” conceptual same is true of other potential beliefs. I assume that I am impossibility claim.16 Neither can I. It seems to me that not unusual in this respect. We don’t have basic voluntary someone, or something, could in principle will himself control over what we believe. or herself into a state that would share many features of Alston turns next to nonbasic immediate voluntary con- beliefs, including the state of affirming the truth of the trol.18 One has this sort of control over the things one proposition. Obviously, if we don’t typically just will can do right away by doing something else, typically our beliefs, then willed beliefs would not be caused in something over which one has basic voluntary control. the same way our beliefs typically are. So much the Standard examples are opening doors and turning on worse for the idea that to be a belief, a state has to be lights. We can, in typical circumstances, do these things caused in the way our beliefs typically are. simply by moving our bodies in the appropriate ways. Furthermore, I will argue below that we do have There’s vagueness here concerning what counts as “right voluntary control over at least some of our beliefs. away” but that vagueness is in no way problematic. 20 richard feldman

This is because the boundary between nonbasic imme- move in a certain way to get the lights to be on or off. diate voluntary control and the next weaker kind of And I can do this. I assume that the proposition that I control, long‐range voluntary control, is acceptably impre- have control is not refuted either by the fact that the cise.19 The sorts of things over which we have long power could fail so that the lights don’t come on when range voluntary control are the sorts of things we can I turn the switch or by the fact that the switch could do over time by doing other things. Perhaps painting fail keeping them on no matter what I do. These mere my house is an example. Or, more precisely, I have long possibilities don’t undermine control, though if they range voluntary control over what color my house is were actualities perhaps they would. because I can do things like paint it. The next step of the argument notes that my belief Finally, there is indirect voluntary influence.20 This is about whether the lights are on tracks the actual state the kind of control we have when we can undertake a of the lights almost perfectly. As a result, I have control course of action that may affect some condition over over whether I believe that the lights are on. All I have the long term. Perhaps a person has indirect voluntary to do is move a certain way, and then I’ll have the rel- influence over the condition of her heart, since diet evant belief. More generally, when I have control over and exercise, courses of action she can more directly a state of the world and my beliefs about that state track control, can affect it. that state, then I have just as much control over my Consider my belief that the Earth is not flat. Suppose belief about the state as I have over the state itself. Thus, that there were some reason why I’d prefer not to have we have nonbasic immediate voluntary control over this belief. There’s nothing I can do to rid myself of this our beliefs about states of the world over which we belief right away. It’s not like moving my hand or turn- have control, provided our beliefs are responsive to ing out the lights. I don’t have basic or nonbasic imme- those states. diate voluntary control over whether I have this belief. Here’s the argument spelled out in detail as it applies And there’s nothing much I can do long range to con- to a single case. trol it either. Changing my belief is not relevantly like painting my house. I can’t simply set out on a course of An Argument for Voluntarism About Belief action that will almost surely result in my belief being 1. You can causally determine whether the lights are changed. I might enroll in the Flat Earth Society, read on by moving or failing to move your hand in a conspiracy literature asserting that satellite photos are certain way. all phony, and so on. Perhaps this will help rid me of 2. You can similarly causally determine whether you my belief. Alston would agree that it might. But this believe that the lights are on by moving or failing gets us, at most, indirect voluntary influence, and this is to move your hand in that same way. not the sort of effective voluntary control required to 3. If the truth of (1) is sufficient for you to have non- refute The Voluntarism Argument.21 basic voluntary control over whether the lights are I believe that, for most part, what Alston says is right. on, then the truth of (2) is sufficient for you to However, I will argue that we have considerably more have nonbasic voluntary control over whether you control over some of our beliefs than Alston acknowl- believe that the lights are on. edges. Still, this control does not undermine the basic 4. The truth of (1) is sufficient for you to have nonba- idea behind The Voluntarism Argument, although it sic voluntary control over whether the lights are on. may show that the argument needs reformulation. 5. The truth of (2) is sufficient for you to have non- basic voluntary control over whether you believe An Argument for Voluntarism About Belief that the lights are on. (3), (4) 6. You have nonbasic voluntary control over whether My argument for voluntarism begins with the assump- you believe that the lights are on. (2), (5)22 tion that there are states of the world over which ­people have nonbasic voluntary control. For example, I think that this is a sound argument. I also think I have nonbasic voluntary control over whether the that similar arguments establish that each person has lights in my office are on or off. All I have to do is voluntary control over lots of beliefs. Roughly, we have