sign in sign up
<<
Home , Fallibilism, Pragmatism

, Kant, and Transcendental

Edited by Gabriele Gava and Robert Stern

6244-678-1pass-0FM-r02.indd 3 8/21/2015 4:17:40 AM First published 2016 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2016 Taylor & Francis The right of the editor to be identiied as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identiication and without intent to infringe. Library of Congress Cataloging-in-Publication Data

[CIP data]

ISBN: 978-1-138-79191-6 (hbk) ISBN: 978-1-315-76244-9 (ebk) Typeset in Sabon by Apex CoVantage, LLC

6244-678-1pass-0FM-r02.indd 4 8/21/2015 4:17:40 AM To Mario

6244-678-1pass-0FM-r02.indd 5 8/21/2015 4:17:40 AM Contents

Acknowledgments ix

Introduction 1 GABRIELE GAVA AND ROBERT STERN

1 German , Classical Pragmatism, and Kant’s Third Critique 22 SEBASTIAN GARDNER

2 The of Kant’s Architectonic 46 GABRIELE GAVA

3 A Kant-Inspired Vision of Pragmatism as Democratic Experimentalism 67 DAVID MACARTHUR

4 Peirce, Kant, and What We Must Assume 85 CHERYL MISAK

5 Peirce and the Final : Against Apel’s Transcendental Interpretation of the 94 DANIEL HERBERT

6 Forms of Reasoning as Conditions of Possibility: Peirce’s Transcendental Concerning Inductive 114 JEAN-MARIE CHEVALIER

7 Kant and Peirce on 133 MARCUS WILLASCHEK

6244-678-1pass-0FM-r02.indd 7 8/21/2015 4:17:40 AM viii Contents 8 Round Kant or Through Him? On James’s for Freedom, and Their Relation to Kant’s 152 ROBERT STERN

9 in Kant and 177 GRAHAM BIRD

10 of Objects as Prescribing : A Kantian and Pragmatist Line of Thought 196 JAMES R. O’SHEA

11 Subjectivity as Negativity and as a Limit: On the and of the Transcendental Self, Pragmatically Naturalized 217 SAMI PIHLSTRÖM

12 A Plea for Transcendental Philosophy 239 WOLFGANG KUHLMANN

13 Transcendental , Epistemically Constrained , and Moral Discourse 259 BORIS RÄHME

Contributors 287 Index 291

6244-678-1pass-0FM-r02.indd 8 8/21/2015 4:17:41 AM 7 Kant and Peirce on Belief Marcus Willaschek

According to Peirce, pragmatism (or “”) is with- out things in themselves: “The present writer was a pure Kantist until he was forced by successive steps into Pragmaticism. The Kantist has only to abjure [. . .] the that a thing-in-itself can, however indirectly, be conceived; and then correct the details of Kant’s accordingly, and he ind himself to have become a Critical Common-sensist” (CP 5.452)1 (where “Critical Common-sensism” is an aspect or consequence of what Peirce had previously called pragmatism). As this and many other state- ments by Peirce show, and as has been repeatedly stressed in the literature on the of pragmatism, Kant was a major inluence on Peirce.2 His inluence can be seen in many aspects of Peirce’s philosophy: his conception of thought as , his speciic brand of realism, his account of the cat- egorical structure of thinking, just to name a few. In this paper, I will take a closer look at an aspect of Kant’s inluence that has occasionally been noticed (cf. Murphey 1968, 177; Thayer 1981, 138), but to my knowledge never closely investigated. It concerns one of the fundamental concepts of Peircean pragmatism: the of belief. My interest, however, will not be primarily historical. Rather, I hope to bring into relief an insight Peirce shares with Kant that still is philosophically important today. Moreover, I will argue that Kant’s account of belief is in a certain sense more radically pragmatist than Peirce’s in that it allows belief to be rationally justiied by recourse to the role it plays in action. While this point moves Kant closer to James than to Peirce, he also differs from James in that the practical justii - cation of a belief is not constituted by its practical consequences, but rather by the belief’s a of the action’s being rational. Thus, Kant’s account of belief can be seen as a genuine and original contribution to the pragmatist , sharing features with Peirce’s and James’s ac- counts, but also differing from them in key respects. I will approach my topic somewhat indirectly by irst looking at the Kan- tian origin of the label “pragmatism,” which will take us to Kant’s distinc- tion between pragmatic and moral belief. I will then present Kant’s account of belief in the Canon section of the Critique of Pure and, after sketching Peirce’s conception of belief, highlight the many differences, but

6244-678-1pass-007-r02.indd 133 8/21/2015 4:19:12 AM 134 Marcus Willaschek also the much more important commonalities, between them. Finally, I will very briely compare Kant’s account with James’s in “” in order to bring into view both the genuinely pragmatist of Kant’s account of belief and the way in which it differs both from Peirce’s and James’s.

1 PRAKTISCH AND PRAGMATISCH

As Peirce recollects in 1905, the term “pragmatism” derives from Kant inso- far as labels such as “practicalism” or “practicism” appeared inappropriate to Peirce because of the moral and metaphysical connotations of the term “practical” in Kant: “for one who had learned philosophy out of Kant [. . .] and who still thought in Kantian terms most readily, praktisch and pragma- tisch were as far apart as the two poles, the former belonging in a region of thought where no of the experimentalist type can ever make sure of solid ground under his feet, the latter expressing relation to some deinite human purpose” (CP 5.412). Peirce was surely correct in thinking that from a Kantian perspective the label “pragmatism” was much more itting for his own philosophy than “practicalism” or “practicism.” However, this is not for the reason Peirce himself offers, since “praktisch” and “pragmatisch” in Kant are by no means opposites, related as the North Pole is to the South Pole. Accord- ing to Kant’s most general deinition of “practical,” which we ind in the Canon of the Critique of Pure Reason, “ is practical that is pos- sible through freedom” (CPR A800/B828), where “freedom” is freedom of . Thus, in this wide sense, the practical concerns the realm of human , broadly conceived, and there is no apparent reason why a “mind of the experimentalist type” should shy away from it. It is within this prac- tical realm that Kant goes on to distinguish between “pragmatic laws of free conduct for reaching the ends recommended to us by the senses” and “pure practical laws whose end is given by reason completely a priori,” the latter being “the moral laws” (CPR A800/B828). Thus, in the Canon we encounter a contrast not between the practical and the pragmatic, but rather between the pragmatically practical, which is empirical, and the morally practical, which is a priori or “pure.”3 Similarly, in a passage from the Groundwork for the Metaphysics of Morals, Kant distinguishes between “technical,” “pragmatic” and “moral” imperatives (KGS 4:416f.), all of which are “practical ” (cf. KGS 4:414–5). Moral imperatives are supposed to hold a priori, whereas prag- matic imperatives—although they presuppose an end of which we know a priori that every human being pursues it, namely one’s own (KGS 4:415)—are empirical insofar as what will or will not contribute to some particular person’s happiness is an empirical question (KGS 4:418).4 So again, the contrast is not one between the practical and the pragmatic,

6244-678-1pass-007-r02.indd 134 8/21/2015 4:19:13 AM Kant and Peirce on Belief 135 but rather one within the practical between the pragmatic and the moral.5 In , Kant never draws a contrast between the practical and the pragmatic, but consistently treats the pragmatic as one species of the practical.6 So it seems that what Peirce must have had in mind with his reference to Kant’s use of “praktisch” and “pragmatisch” is rather the distinction between the moral (as the realm of “pure ”) and the prag- matic. Since the moral realm, according to Kant, is based on the moral , which holds a priori for all rational and requires us to think of our- selves as part of an intelligible world (KGS 4:452), the moral in the Kantian sense is indeed a “region of thought” that is necessarily suspect to the exper- imentalist . Kant draws a distinction between the pragmatic and the moral in various different contexts, and he applies it to different kinds of objects: to laws (CPR A800/B828), imperatives (KGS 4:416–7; 6:391), ends (KGS 6:354), forms of friendship (KGS 6:472), (KGS 6:281; 9:455) and (KGS 9:470), just to name a few. While the precise con- notations of the distinction may vary with and subject , its core remains the same: While the pragmatic is directed at the real- ization of empirical ends (which are variable, arbitrary, and contingent), the moral realm is constituted by the moral law and directed at realizing moral ends (which are a priori and necessary). Among the various applications of this contrast, there is one that stands out as being particularly close to Peirce’s own concerns, namely the dis- tinction between pragmatic and moral belief drawn by Kant in the Canon of the Critique of Pure Reason.7 There, Kant deines “pragmatic beliefs,” which concern “arbitrary and contingent ends,” as “contingent beliefs, which however ground the actual use of the means to certain actions” (CPR A824/B852) and contrasts them with “moral beliefs,” which are “neces- sary” (CPR A824/B852) and have ends that are “inescapably ixed” (CPR A828/B856) by the moral law. Kant’s deinition of pragmatic belief as “ground[ing] the actual use of the means to certain actions” is obviously a close cousin of the famous -deini tion of belief as “that upon which a man is prepared to act,” attributed to , of which Peirce said that pragmatism was a mere corollary to it (CP 5.12). In the next section, I will take a closer look at Kant’s con- ception of belief and his distinction between pragmatic and moral belief. As we will see later, although the Peircean concept of belief is particularly close to Kant’s account of pragmatic belief, the afinities extend also to Kant’s conception of belief in general and, thus, even to his conception of moral belief (as a special case of belief in general).

2 KANTIAN BELIEF

In the section entitled “Canon of Pure Reason” in the Critique of Pure Rea- son, Kant distinguishes between opinion, belief, and knowledge as three

6244-678-1pass-007-r02.indd 135 8/21/2015 4:19:13 AM 136 Marcus Willaschek kinds or “stages” of “taking (something) to be true” (Fürwahrhalten) (CPR A822/B850). A “taking to be true” (or, as it is also translated, an assent), is a theoretical propositional , that is, an attitude towards a prop- osition (or, in Kantian terms, a judgment), which proposition is taken to represent some state of affairs truly. It is, therefore, what today in philoso- phy is commonly called a “belief,” which has been deined as “the attitude we have, roughly, whenever we take something to be the case or regard it as true” (Schwitzgebel 2014). Kant’s concept of belief, by contrast, is nar- rower than the current notion, since it is only one species of assent alongside opinion and knowledge. Here is Kant’s oficial deinition of the three kinds of assent: “Having an opinion is taking something to be true with the con- sciousness that it is subjectively as well as objectively insuficient. If taking something to be true is only subjectively suficient and is at the same held to be objectively insuficient, then it is called believing. Finally, when taking something to be true is both subjectively and objectively suficient it is called knowing” (CPR A822/B850). Kant goes on to describe “subjec- tive suficiency” as “conviction (for myself)”8 and “objective suficiency” as “ (for everyone)” (CPR A822/B850). Unfortunately, it is not entirely clear what he means by this, and various divergent interpretations have been proposed.9 For present purposes, I suggest that “subjective suficiency” or conviction be read as concerning the irmness with which the subject is committed to the truth of the judgment in question.10 In particular, subjective suficiency concerns the irmness of assent vis-à-vis (a) possible counterevidence and (b) rising stakes. For instance, if I assent to the judgment that Fred was in Frankfurt yesterday, which is what he himself has told me, how might I react to your claim to having seen him yesterday in Munich? And how might I react if I were to learn that Fred was suspected of having committed some crime in Munich yesterday? The more I am disposed to give up my as- sent under such circumstances, the lower its degree of conviction or subjec- tive suficiency. Thus, subjective suficiency concerns how deeply entrenched an assent is in the subject’s cognitive system.11 An assent is subjectively suf- icient to the highest degree only if there is no possible counterevidence and no practical consequence in light of which the subject would give it up. (As we will soon see, even though Kant’s deinition may seem to demand this, not all beliefs are subjectively suficient to this degree.) By contrast, “objective suficiency” or “certainty” concerns the relation between the subject’s for taking the judgment to be true and the actual truth of the judgment. An assent is objectively suficient if the reasons on which it is based in fact guarantee that the judgment is true and thus, if the subject is aware of this, result in certainty, that is, in the awareness that a mistake is impossible (cf. KGS 9:66).12 Belief in the Kantian sense, there- fore, is the attitude towards a judgment in which one is irmly convinced of its truth even though one is aware of the fact that the reasons on which one bases one’s assent fall short of guaranteeing that the judgment is in fact true.

6244-678-1pass-007-r02.indd 136 8/21/2015 4:19:13 AM Kant and Peirce on Belief 137 This raises the question whether it can ever be rational to believe some- thing in Kant’s sense. As Hume famously says with regard to belief in mira- cles, “[a] wise man . . . proportions his belief to the ” (Hume [1748] 1999, sect. 10, § 4). This can be read as saying that the weaker the evidence for some claim, the weaker the conidence one should place in its truth. Kantian belief violates this evidentialist , because it allows the irm- ness of conviction (degree of subjective suficiency) to be disproportionate to the available evidence (degree of objective suficiency): In particular, in a Kantian belief, a irm conviction may go hand in hand with weak evidence. How can we ever be justiied in holding such an attitude? Before we try to answer this question, it may be helpful to be more pre- cise about what is meant here by “evidence” and “evidentialism,” since these terms have been used in many different ways. In what follows, I will take “evidence” for some person A’s belief that p to be a fact that q, which minimally satisies the following two conditions: (a) A is epistemically jus- tiied in believing that q and (b) the fact that q is in some way indicative of the fact that p. (Thus, “evidence” involves at least a three-place relation: q is evidence for A’s belief that p.) Now, evidence is “suficient” with respect to A’s belief that p, if the irmness of the belief corresponds to the strength of the evidence. For instance, if A is irmly and fully convinced that p, then suficient evidence must guarantee that p, or at least make p highly likely. By contrast, if A only that p very tentatively, the evidence is suficient if, on balance, p is slightly more likely, or more plausible, than its negation. (All this is obviously extremely vague, but it will sufice for the purposes at hand.) A strong form of evidentialism would then hold that it is irrational to believe anything without suficient evidence. This is meant to capture the gist of a famous line from Clifford, whom James quotes as saying: “It is wrong always, everywhere, and for every one, to believe anything upon insuficient evidence” (James [1897] 1979, 18).13 Therefore, according to this strong form of evidentialism, Kantian belief must appear to be irratio- nal, since in this case the irmness of the belief does not correspond to the strength of the evidence. Kant’s response to this worry consists in limiting belief to theoretical attitudes that are related to action: “Only in a practical relation, however, can a theoretically insuficient taking to be true be called believing” (CPR A823/B851, translation altered).14 Kant’s is that a judgment that does not warrant full conviction on purely theoretical grounds (because the avail- able objective reasons fall short of guaranteeing its truth) can nevertheless be something one can be irmly convinced of—without violating any norms of —if the judgment stands in the right kind of relation to one’s actions and ends. But what relation is that? In answering this question, I will rely on some insights from Thomas Höwing’s excellent analysis of Kant’s concept of pragmatic belief (Höwing forthcoming).15 Kant explains: “Once an end is proposed, the conditions for attaining it are hypothetically necessary” (CPR A823/B851). Höwing calls this “the

6244-678-1pass-007-r02.indd 137 8/21/2015 4:19:13 AM 138 Marcus Willaschek Canon Principle” and argues convincingly that it is a fundamental principle of rational agency. Now belief according to Kant consists in “my presupposi- tion and taking to be true of certain conditions” (CPR A824/B852), namely those conditions that, according to the Canon Principle, are hypothetically necessary for attaining my end through some action of mine (assuming that I cannot be certain that these conditions in fact obtain). For instance, if I want to save money for my retirement (= end) by making a deposit in a bank (= action), I must believe (in Kant’s sense) that the bank will still be active, and able to pay out the money, when I retire. Of course, I cannot be certain of this; thus, my taking it to be true is not objectively suficient and hence not knowledge. But nevertheless, I must be suficiently convinced that the bank will repay my money if I am to be motivated to make the deposit. This means that there is an additional sense in which the assent, in the case of belief, must be “suficient,” namely suficient for action. As Kant puts it in the Jäsche-Logik: “This is a taking to be true that is suficient for action [genug zum Handeln], i.e. belief” (KGS 9:67n, translation altered). I take this to show that by deining belief as “subjectively suficient,” Kant does not want to say that only an assent that is maximally irm counts as belief (although he does require that for a special kind of belief, namely moral -be lief).16 Rather, a belief is an assent that lacks certainty but is still suficiently irm to serve as a basis for one’s actions.17 To understand what this means, we must next explain what it is to act on the basis of a given belief. Recall the Canon Principle according to which the conditions for reaching one’s end are necessary relative to that end. Fol- lowing Höwing, I will call these conditions “enabling conditions of practical success.” I will assume that they include the relevant instrumental relation between my action and its end as well as the of a of background conditions that are suficient for the action to realize the end. In the case of saving for retirement, the enabling conditions thus comprise (i) the fact that regularly depositing small amounts of money in a bank is a means to having a larger sum available in the , as well as (ii) the supposed fact that the bank will still exist and be able to pay out the money once I have entered retirement. For an action to be based on a belief thus means that the latter is a belief in the obtaining of the enabling conditions for the action’s being successful. The Canon Principle, as a fundamental of rational agency, thus comes to this: it is rational for some agent S to do A in order to achieve end E only if S takes it to be true that the enabling conditions for A’s being successful in realizing E obtain. In all cases where this taking to be true is not certain (objectively suficient, knowledge), it is a belief in the Kantian sense.18 One might worry that the Canon Principle is too demanding. Do I really have to irmly believe that the bank will be able to pay out the money in order to make it rational for me to deposit my money in it? What if I live in politically unstable and, therefore, am rather skeptical that I will get my money back, but still think that this is my best bet? Wouldn’t it

6244-678-1pass-007-r02.indd 138 8/21/2015 4:19:13 AM Kant and Peirce on Belief 139 still be rational to deposit the money in the bank, even if I don’t believe, but only hope that the bank will be able to pay it back? This worry can be overcome by keeping in mind the practical character of Kantian belief. As we have seen, a belief is an assent that is suficiently irm to serve as the basis for one’s actions. Now, the degree of conviction that will be suficient in a particular case will depend on the alternative courses of action that are open to me. If there are many other ways to realize my end, the degree of conidence in the obtaining of the enabling conditions of my action’s being successful will have to be fairly high in order to make it rational for me to act on it. But, if there is no alternative way to realize my end, my belief that the enabling conditions obtain need not be very irm at all. It need only be suficiently irm, relative to the alternatives. Consider Kant’s example of the doctor who doesn’t know what illness his patient suffers from, but who has got to do something to cure the patient (CPR A824/B52). Since appearances indicate that the patient might suffer from consumption, and the doctor can’t think of any alternative diagnoses, he treats the patient for consumption. Assuming the patient would most probably die if the doctor did , and considering that he doesn’t see any alternative treatment, the doctor’s degree of conidence in his diagnosis needn’t be very high in order to make it rational to act in accordance with it. Nevertheless, the doctor cannot rationally act that way if he takes it to be false that the patient suffers from consumption or if he doesn’t take any stance on this issue at all. Some kind of positive theoretical attitude towards the patient’s from consumption is required, and this seems to be suficient for Kantian belief (in this kind of case). This means that the degree of conidence required for (pragmatic) belief can sometimes be quite minimal. If all competing diagnoses are ruled out, but consumption is not, the best bet is to treat the patient for consumption. In such a case, one “believes,” according to Kant, that the patient suffers from consumption. What is characteristic of Kantian belief, though, is that even this minimal degree of conidence goes beyond what is warranted from a purely theoretical point of view (according to which we would have to suspend judgment). Thus, a belief in the Kantian sense is an assent to an objectively less-than- certain judgment that the enabling conditions for some action the agent has decided on obtain, where the assent is suficiently irm to serve as a basis for action. Since Kantian belief is limited to cases of assent that serve as the basis of action, its anti-evidentialist consequences are somewhat mitigated. A moderate evidentialist might admit that even if a Kantian belief such as the doctor’s belief in Kant’s example may not be epistemically justiied, it can nevertheless be rationally justiied in some non-epistemic sense.19 Kant’s account of belief is compatible with this moderate kind of evidentialism. But, it is not compatible with the more radical version of evidentialism, introduced above, according to which it is always and in every respect ir- rational to believe in light of insuficient evidence. According to Kant, the

6244-678-1pass-007-r02.indd 139 8/21/2015 4:19:13 AM 140 Marcus Willaschek lack of alternative ways of realizing one’s end can outweigh weakness of evidence and make it rational to believe something on the basis of insufi- cient evidence.20 On the basis of this deinition of belief, Kant then goes on to distin- guish between contingent and necessary belief on the one hand and between pragmatic and moral belief on the other (CPR A823f./B851f.). A belief is necessary, relative to a given end, if the subject knows that there is no al- ternative course of action available that would realize the end; otherwise, the belief is contingent. Now, pragmatic belief is doubly contingent: it is a contingent belief in the sense just deined, but it is also contingent in the ad- ditional sense that the end in question is not rationally necessary.21 By con- trast, moral belief is necessary in the twofold sense that the action is without alternatives and the end (because it is given by the moral law) is rationally necessary.22 Pragmatic belief admits of degrees; moral belief does not. While Kant’s example of pragmatic belief is the doctor who believes that his patient suffers from consumption, his example of a moral belief is be- lief in the . Kant had argued in the preceding section of the Canon that, as rational agents, we are morally required to realize the highest good, which consists in a world in which everyone is just as happy as they morally deserve to be. But, according to Kant, we can realize this end only with divine assistance (cf. CPR A814/B842). Thus, the existence of God is an enabling condition for realizing a rationally necessary end and is, therefore, the of a moral belief—even though, as Kant had argued in the , the question whether God exists cannot be decided on purely theoretical grounds. We do not need to discuss the plausibility of Kant’s ar- gument for moral belief in God. What matters for our purposes is only that, apart from the rational necessity of the end, moral belief does not differ in structure from the case of pragmatic belief. Just as with pragmatic belief, moral belief is rendered rational even in the absence of suficient evidence, because it concerns the enabling conditions of some of our actions (namely those aiming at realizing the highest good).

3 BELIEF IN PEIRCE

Let me now turn, much more briely, to Peirce’s conception of belief. In a well-known passage from “How to make our Clear,” Peirce names three essential characteristics of a belief: “First, it is something that we are aware of; second, it appeases the irritation of doubt; and, third, it involves the establishment in our of a rule of action, or, say for short, a habit” (CP 5.397). The irst feature, awareness, follows from the fact that belief, according to the Peirce of 1878, has a characteristic qualitative aspect to it, a “sensation of [. . .] believing” (CP 5.370). 23 In particular, this sensation differs from the one connected with doubting. While doubt is “an uneasy and dissatisied state,” belief is “calm and satisfactory” (CP 5.370).

6244-678-1pass-007-r02.indd 140 8/21/2015 4:19:13 AM Kant and Peirce on Belief 141 This takes us to the second feature. Belief, for Peirce, is essentially an element in the dynamic process of thinking that comprises, as its other basic elements, doubt and inquiry. Put very schematically, thinking starts from a state of belief, which is followed by doubt, which motivates inquiry, which leads to a new belief. Since this new belief can also be displaced by further doubt, belief is essentially a transitory state: “at the same time that it is a stopping-place, it is also a new starting-place for thought” (CP 5.397). But, why can’t we just be content with the beliefs we happen to have? The reason derives from the third feature, which is that belief establishes a rule for action or, as Peirce often puts it, a habit. This means, irst, that belief is essentially practical: “Our beliefs guide our desires and shape our actions” (CP 5.371). And second, it means that any given belief is related not only to individual acts, but also to a general type of action. Having a belief, for Peirce, means being ready to act in a certain way under speciied circumstances: “Belief does not make us act at once, but puts us into such a condition that we shall behave in some certain way, when the occasion arises” (CP 5.373). A belief, or the rule of action associated with it, tells us what to do in a speciictype of situation. And this is why beliefs can, and often will, be unsettled by doubt, since every so often there will be situations of the relevant type where acting in the way speciied by the belief will either be impossible or unsuccessful. If this happens, the belief will be replaced by doubt, which initiates the search for a solution, which Peirce calls inquiry, until a solution has been found and a new belief is established. This is not to say that doubts can arise only from practical dificulties. But even a merely “theoretical” doubt (concerning a belief that is not at present practically relevant) will have to be, according to Peirce, a “real and living doubt” (CP 5.376) (requiring more than the admission of the of false- hood) in order to shake any of our current beliefs. Obviously, there are many more aspects to Peirce’s conception of belief than this thumbnail picture can convey, but for present purposes I will have to restrict myself to three further remarks. First, Peirce’s point is not just that belief as a kind of is characterized by its functional role in the process of thought and its relation to action. Rather, and even more importantly from Peirce’s perspective, the content of a speciic belief is also ixed by its functional role: “The of belief is the establishment of a habit; and different beliefs are distinguished by the different modes of action to which they give rise. If beliefs do not differ in this respect, if they appease the same doubt by producing the same rule of action, then no mere differences in the manner of consciousness of them can make them different beliefs, any more than playing a tune in different keys is playing different tunes” (CP 5.389). Second, the kind of action that is to be performed in a given situation depends not only on one’s belief and what kind of situation one is in, but also on one’s motives and ends: “[Readiness] to act in a certain way under given circumstances and when actuated by a given motive is a habit; and

6244-678-1pass-007-r02.indd 141 8/21/2015 4:19:13 AM 142 Marcus Willaschek a deliberate, or self-controlled, habit is precisely a belief” (CP 5.480; em- phasis added). Thus, the role of belief is to mediate between one’s ends and a possible course of action. Consider Peirce’s example of The Assassins, a medieval religious sect in the Middle East: “The Assassins, or followers of the Old Man of the Mountain, used to rush into death at his least com- mand, because they believed that obedience to him would insure everlasting felicity. Had they doubted this, they would not have acted as they did” (CP 5.371). Nor, one might add, would they have acted this way if they hadn’t cared so much about everlasting felicity. So, where a belief serves as a “rule for action,” at least three factors are involved: a speciic situation, an end, and a possible course of action. The belief then states the conditions under which that course of action will, in that particular situation, successfully contribute to realizing one’s end. As Peirce puts it most succinctly in a text from 1904: “A belief in a proposition is a controlled and contented habit of acting in ways that will be productive of desired results only if the proposi- tion is true” (EP 2.312). Third, concerning the epistemological aspect of beliefs, Peirce acknowl- edges their essential fallibility: “Hence, the sole object of inquiry is the settle- ment of opinion [which term Peirce here uses interchangeably with “belief”]. We may fancy that this is not enough for us, and that we seek, not merely an opinion, but a true opinion. But put this fancy to the test, and it proves groundless; for as soon as a irm belief is reached we are entirely satisied, whether the belief be true or false” (CP 5.375; emphasis added). 24 Therefore, a belief does not need to be based on evidence that rules out its falsity in order to serve as the basis for action. What is required is only the absence of “a real and living doubt.” Without such doubt, “all discussion is idle.” Once everyone is fully convinced of a certain point, “no further advance can be made” by arguing for it. I take these remarks from “The Fixation of Belief” to make not just a psychological, but also an epistemological point: We are fully justiied in our beliefs as long as they are “perfectly free from all actual doubt” (CP 5.376); providing additional evidence for a belief which is not subject to a living doubt does not improve its epistemic standing. This allows for rational belief without any positive evidence for its truth, as long as the belief is free from doubt. Thus, Peirce’s conception of belief is anti-evidentialist in dissociating rational belief from the availability of ev- idence.25 Moreover, it is “conservative” in that it allows us to continue to hold whatever beliefs we may have, even in the complete absence of evidence for their truth, the latter being required only in the case of living doubts.

4 COMPARING KANT’S ACCOUNT OF BELIEF WITH PEIRCE’S (AND WITH JAMES’S, TOO)

A comparison of Kant’s and Peirce’s conceptions of belief reveals both deep similarities and important, though perhaps less deep, differences. Among

6244-678-1pass-007-r02.indd 142 8/21/2015 4:19:14 AM Kant and Peirce on Belief 143 the things they have in common is, irst, the basic of what kind of thing a belief is. Both Kant and Peirce think of belief as a conscious theoretical that is essentially linked to action. For both, the link between belief and action consists in the belief’s serving as a basis for action in a twofold sense. On the one hand, a belief is a basis of action in a psychological sense in that, without the belief (or a relevantly similar one), the agent would not have performed the corresponding action. On the other, a belief is the rational basis of an action in that, without the belief (or a relevantly similar one), performing the action would be irratio- nal. (This latter aspect is less prominent in Peirce than in Kant.) Second, for both Kant and Peirce, the link between a belief and the action based on it is that the action can be successful only if the belief is true.26 Third, both Kant and Peirce insist that beliefs do not have to be certain, or known to be true, in order to be rationally justiied; both are fallibilists about belief and both build fallibility into their concepts of belief. And inally, Kant’s and Peirce’s accounts of belief are both anti-evidentialist in that they allow that one can be rationally justiied in irmly believing something on theoretically “insuf- icient” evidence. While it seems that for Peirce there can be rational belief even in the absence of any evidence at all, Kant at least allows for rational belief without suficient evidence (i.e., evidence adequate to the irmness with which the belief is held). But, of course, there are also important differences. First, Kant considers belief to be a highly speciic form of assent, to be distinguished from opin- ion and knowledge as the other forms of “taking something to be true” (the latter being roughly equivalent to belief in the current sense), whereas Peirce tends to see belief as the fundamental kind of theoretical attitude that includes opinion and knowledge as well as other theoretical attitudes as its species. Second, where Kant thinks of beliefs as more or less static mental states, Peirce understands them as “thought at rest” (CP 5.396), that is, as part of a dynamic process that includes doubt and inquiry. Third, at least the Peirce of 1878 attributes to belief an experiential character (the “sensa- tion of [. . .] believing,” which is a “calm and satisfactory state” (CP 5.370), which Kant does not. Fourth, Kant distinguishes between two important kinds of belief, pragmatic and moral, whereas Peirce has no use for the concept of moral belief (because an “experimentalist mind” treats all ends as contingent). Fifth, where Kant seems to think of the relation between belief and action primarily as one between a particular belief and an indi- vidual action, Peirce regards beliefs as related to habits, and thus to types of action.27 Sixth, and relatedly, Kant’s and Peirce’s accounts of belief differ in the order of explanation. While Kant, in his account, starts from an action that is supposed to realize a given end and then speciies the content of the belief required for the action to be rational, Peirce starts with the distinc- tion between belief and doubt and then identiies belief with a disposition to act in a speciic way. Seventh, according to both Kant and Peirce we can specify the content of a belief by inquiring into the enabling conditions of

6244-678-1pass-007-r02.indd 143 8/21/2015 4:19:14 AM 144 Marcus Willaschek our actions; however, unlike Peirce, Kant does not assume that the content of the belief is ixed exclusively by its functional role with regard to agency. Rather, the content of the belief, which Kant calls “judgment,” is ixed by its representational content, that is, by the concepts employed in it. No doubt this list could be extended to include both further similarities and further differences. Instead, however, I would like to highlight, in clos- ing, an aspect of Kant’s and Peirce’s conception of belief that I ind philo- sophically particularly important. It concerns the anti-evidentialism which Kant and Peirce share on a general level, but on which they also differ in important respects. Neither Kant nor Peirce would allow that a belief can be rational in the face of undefeated evidence to the contrary. Thus, it is only because the doc- tor in Kant’s example, after “looking at the appearances,” “doesn’t know anything better” (CPR A824/B852) that he is entitled to believe that the patient suffers from consumption. For cases of moral belief, Kant explicitly requires that neither their truth nor their falsity can be known or even at- tributed a probability (KGS 9:67). For Peirce, awareness of counterevidence would lead to doubt, which replaces the belief and initiates inquiry. But, even if the available evidence thus constrains what we can rationally believe, according to both Kant and Peirce the evidence does not uniquely determine what it is rational for us to believe and to what degree. Rather, for Peirce, we are generally entitled to hold on to the beliefs we happen to have. Only when there is a “real and living doubt” are we rationally required, and psychologically compelled, to give them up. To be sure, one has to be open-minded and critical in one’s belief-forming processes, but mere lack of theoretical evidence, according to Peirce, is no good reason to give up a belief that works practically. Note, however, that Peirce does not seem to hold that a belief can be rationally justiied by non-theoretical factors exclu- sively. As Christopher Hookway has remarked: “The distinctive character of Peirce’s pragmatism depends upon a view about the scope of relective thought: it offers a clariication which is to be valuable where relection has a fundamental role and is not intended for use in other areas of life. The only ‘consequences’ of a concept or proposition which are ‘pragmatically relevant’ will then be those which are pertinent to ‘relective inquiry’; -as pects of meaning which are of importance in attempting to answer ‘vital’ questions need not be taken into account” (Hookway 2012, 190). In this respect, at least, Kant can be seen as more radically pragmatist than Peirce. Peirce thought of the relation between belief and action as moving primarily in one direction, namely from the belief to the act. Kant, by contrast, sees the relation as moving in both directions: While the belief is the psychologi- cal basis for the action, the action (or more precisely: the fact that an action of an appropriate kind is rationally required) can be the rational warrant for the belief. As Kant is aware, this transmission of rational warrant works only under highly speciied conditions; in particular, it presupposes that the end served by the action is suficiently important, the action without known

6244-678-1pass-007-r02.indd 144 8/21/2015 4:19:14 AM Kant and Peirce on Belief 145 alternatives, and the belief not defeated by evidence to the contrary. But, once these conditions are given, Kant is willing to admit that an assent that would otherwise be rationally unjustiied can turn out to be rational, even in the absence of suficient theoretical evidence in its favor. In this respect, Kant is closer to James than to Peirce, since it is William James who argues that beliefs that lack suficient theoretical evidence can still be rationally warranted if adopting them contributes to their turning out to be true, as is the case when someone will be able to jump over a chasm only if she believes that she will succeed (cf. James [1897] 1979, 29, 53–4, 56, 80). As James argues in “The Will to Believe,” certain kinds of religious belief can be seen as special cases of this kind of rational justiication of a belief by its practical consequences (James [1897] 1979, 29–33). Peirce disagreed. Here is how Christopher Hookway characterizes the views of Peirce and James on this point: “Peirce and James would agree that where ‘ methods’ (relection and the method of ) cannot settle a live or vital question, then we should rely upon sentiment (or the passions). But where we can read this as consonant with James’s pragmatism, the answer being justiied by the effect it has on conduct, Peirce argues that such issues are outside the scope of the sort of rational logical self-control within which his pragmatism inds a home” (Hookway 2012, 194n). Thus, while James claims that in vital is- sues belief without suficient evidence can be rationally justiied by its effect on conduct, Peirce only admits that nothing is wrong with such a belief inso- far as it falls outside the scope of rational inquiry altogether. In this dispute among fellow pragmatists, Kant would have sided with James, since for Kant, as for James, a belief can be rationally justiied by its role in action. But, note that Kant’s account is much more general, and I think also more profound, than James’s. Kant’s point is independent of the expected positive effects of holding the belief in question. What matters is not the expected outcome, but rather that the belief is a rational presuppo- sition of one’s action. Therefore, Kant’s point is not about an entitlement or, as James later put it, a “right to believe” (James [1907] 1975, 124), but rather about a rational commitment one undertakes by acting in a certain way. What one is committed to is the belief that the enabling conditions of one’s action in fact obtain. This does not just hold for Jamesian beliefs with the character of a self-fulilling prophecy, but rather for all our actions and the beliefs they are based on. Kant’s argument for a moral belief in God is meant to show that religious belief can be seen as a special case of this general point, namely the one in which belief in God is a rational presupposition of realizing the highest end, which in turn is a requirement placed on us by reason itself. One may doubt that Kant’s argument for the moral belief in God is convincing. But, one can still acknowledge that Kant is right to insist, irst, that we are rationally committed to believing that the enabling conditions of our actions’ being successful obtain, and second, that this includes enabling conditions the ob- taining of which is “metaphysical” in the sense of going beyond what can

6244-678-1pass-007-r02.indd 145 8/21/2015 4:19:14 AM 146 Marcus Willaschek be empirically established.28 While the irst point anticipates some import- ant features of Peirce’s account of belief, the second can be seen as making a point similar to James’s in “The Will to believe,” while at the same time avoiding what in James can look like facile . One inal remark. Even though Peirce does not distinguish between prag- matic and moral belief and seems to have rejected the very idea of moral belief in the Kantian sense, this does not affect the important similarities between Kant’s and Peirce’s accounts of belief. And the reason for this is not that Peirce’s account is similar to, and perhaps inluenced by, Kant’s account of pragmatic belief (although that may well be true). Rather, the reason is that Kant’s generic account of belief, of which pragmatic and moral beliefs are species, is importantly similar to Peirce’s. As we have seen, beliefs in the Kantian sense are cases of taking something to be true that stand in a “prac- tical relation,” where the “practical” concerns the sphere of human agency broadly conceived. Thus, insofar as Peirce had the distinction between prag- matic and moral belief in mind when he chose the label “pragmatism,” he might just as well have chosen “practicalism” or “practicism,” since it is not pragmatic belief in particular, but rather belief in general—which for Kant is “practical”—that is importantly similar to his own account of belief.29

NOTES

1. References to Peirce are given according to volume and paragraph number in the Collected Papers (CP) and to volume and page number in The Essential Peirce (EP); page-numbers for Kant refer to the A/B-numbering of the Cri- tique of Pure Reason (CPR) and, for all other works, to the Academy Edition (KGS) (Kant 1900-). When not differently indicated, translations in English are given according to The Cambridge Edition of the Works of . 2. Kant’s inluence on Peirce has been emphasized, among others, by Murphey (1968); Haack (2007); Pihlström (2010). 3. Thayer, in his history of pragmatism, refers to this passage as the background for Peirce’s statement, but falsely claims that Kant himself distinguishes be- tween “practical” and “pragmatic laws”; cf. Thayer (1981, 138). Similarly, Pihlström claims that Peirce accepts Kant’s distinction between the pragmatic and the practical, where the latter is the realm of a priori moral laws (Pihl- ström 2004, 41). 4. Kant explains his calling these imperatives “pragmatic” in a footnote, some- what confusingly, by claiming that “pragmatic” designates a relation to “general welfare”: “It seems to me that the authentic signiication of the word ‘pragmatic’ could be determined most precisely in this way. For those sanctions are called ‘pragmatic’ which really low not from the rights of states, as necessary laws, but from provision for the general welfare. A his- tory is written ‘pragmatically’ when it makes us prudent, i.e., teaches how the world could take care of its advantage better than, or at any rate at least as well as, the world of antiquity has done” (KGS 4:417). This is confusing because pragmatic imperatives, as Kant deines them, are not directed at gen- eral, but rather at individual welfare.

6244-678-1pass-007-r02.indd 146 8/21/2015 4:19:14 AM Kant and Peirce on Belief 147 5. In one place in the Groundwork, though, Kant calls the moral or categorical imperative “the practical imperative” (4:429). 6. Even in the Critique of Judgment, where Kant expels the “technically-practical” (which there includes the pragmatic) from the realm of proper, he still treats it as a species of the practical (and, again, contrasts it with the moral). 7. In later writings Kant sometimes, although rarely, uses the terms “practical belief” (KGS 5:647) or “pure practical belief of reason” (KGS 5:144) for what in the Canon he calls moral belief. Mostly, he continues to speak of moral belief. 8. A reading of section 3 of the Canon is complicated by the fact that Kant implicitly distinguishes between objective and subjective conviction (cf. CPR A824//B852, where Kant speaks of “subjective conviction”) without mark- ing the distinction in a consistent way. Thus, at A820/B848 Kant talks about objective conviction (“valid for everyone”), while at CPR A822/B850 he speaks both of objective conviction (“which is at the same time objectively valid”) and subjective conviction (“for myself”) in the same paragraph, call- ing both “conviction” without further qualiication. 9. Cf. e.g., Chignell (2007), Pasternack (2014), Stevenson (2003), Höwing (forthcoming). There are at least three reasons why there is little consensus about Kant’s account of opinion, belief and knowledge: (i) Kant’s published and unpublished pronouncements on this topic are rather brief and leave many questions unanswered (partly because Kant does not seem to have thought of his views as original); (ii) Kant offers divergent accounts in different places, primarily the Canon and the Jäsche-Logik (for instance, in the latter, Kant seems to restrict belief to moral belief, thus ruling out the possibility of prag- matic belief, which nevertheless seems to come up in some of his examples); and (iii) Kant’s use of central terms such as “conviction” (Überzeugung) in the Canon seems to be, at least at the surface of the text, either ambiguous or contradictory. For these reasons, there is considerable interpretative lee- way here, so that none of the proposed interpretations may be clearly correct or incorrect. In what follows, I will base my interpretation primarily on the Canon section, drawing on the Jäsche-Logik and the transcripts from Kant’s lectures only where these help to clarify the text in the Canon. 10. Kant emphasizes irmness in the context of subjective suficiency or convic- tion at CPR A824/B852 (“subjective conviction, i.e., irm belief”); also cf. CPR A828/856 (“I am certain that nothing can shake this belief) and KGS 9:72. 11. Even though subjective suficiency admits of degrees, it cannot be identiied with what present-day Baysian epistemologists call degree of belief, since subjective suficiency in the Kantian sense may often, but need not always, depend on the probability the subject attributes to the judgment’s being true. 12. Kant also requires that one is conscious of the reasons’ being objectively suficient; cf. KGS 9:58. 13. Consider the weaker formulation of evidentialism by Conee and Feldman in their inluential paper entitled “Evidentialism” from 1985 (as reprinted in Conee and Feldman (2004)): “Doxastic attitude D toward proposition p is epistemically justiied for S at t if and only if having D toward p its the evidence S has at t” (Conee and Feldman 2004, 83). Conee and Feldman are not very explicit about (i) what evidence is, (ii) what it means for S to have evidence, and (iii) what it means for a doxastic attitude to “it” evidence. However, they emphasize that although it may be epistemically irrational to hold beliefs that do not “it” one’s evidence, it can be rational in other (e.g., moral) respects.

6244-678-1pass-007-r02.indd 147 8/21/2015 4:19:14 AM 148 Marcus Willaschek 14. I owe this point to Thomas Höwing. That Kant limits belief exclusively to attitudes with some practical import is obscured by the fact that he, in the relevant passages in the Canon, also discusses something he calls “doctrinal belief” (CPR A825/B863), which is a purely theoretical attitude. However, Kant explicitly points out that this attitude is called belief only in a derivative sense and in analogy with practical belief proper (CPR A825/B863). 15. I should note that my interpretation also differs from Höwing’s on many points, most importantly on what the subjective suficiency of an assent consists in. 16. This would it with the fact that in the Jäsche-Logik, too, Kant’s oficial dei- nition of belief holds only for moral belief, but is then tacitly extended so as to include pragmatic belief. 17. Cf. KGS 9:68n, where Kant says that a merchant who enters a deal must not only opine that he will make a proit, but really believe it, “that is, that his opinion is suficient for an uncertain enterprise.” The essential between opinion and belief (that the latter is subjectively suficient, while the former is not) does not concern degree of belief in a merely “theoretical” sense (such as an estimation of probability, given one’s evidence), but rather a practical commitment. This is conirmed by a footnote in the Jäsche-Logik, where Kant says that belief is distinguished from opinion “not by the degree, but by the relation it, as a , has to action” (KGS 9:67n, translation altered). Since both opinion and belief are objectively insuficient kinds of as- sent, but differ in their subjective suficiency, Kant can be read as saying here that even where a given belief and a given opinion do not differ in the degree to which they are conirmed by the available evidence (both are “objectively insuficient”), they differ in that only belief can serve as a basis for action, and is thus “subjectively suficient.” (As mentioned before, though, this is not the only way to read this and other related passages.) 18. Note that the Canon Principle cuts both ways: If I decide to do A, I am rationally required to believe that the enabling conditions ofA hold. But if I don’t believe that the enabling conditions of A hold, I cannot rationally decide to do A. 19. This is the view, e.g., of Conee and Feldmann, cf. Conee and Feldman (2004). 20. In his contribution to the present volume, Robert Stern argues that Kant and James should not be read as anti-evidentialists. In particular, Kant’s “practical” arguments for freedom of the will, both that from Groundwork III (“acting under the idea of freedom”) and that from the second Critique (“fact of reason”), should be seen not as dismissing the need for evidence, but rather as being based on evidence of a particular kind, namely evidence provided by practical, rather than theoretical reason. But note, irst, that this reading does not touch on Kant’s concept of belief and thus does not un- dermine the anti-evidentialist reading of that concept offered here. Second, Stern himself notes that Kant’s arguments for the “postulates of pure practi- cal reason” (God, freedom, and immortality) differ in structure from Kant’s practical arguments for freedom from the Groundwork and the Analytic of the second Critique. Since belief, according to Kant, is the adequate atti- tude towards the postulates, an evidentialist reading of the latter arguments does not rule out an anti-evidentialist reading of the former. Third, Stern’s conception of evidence seems to be less demanding than the one used in the present paper. “Evidence,” on the use of that term adopted here, requires good epistemic standing plus truth-indicativeness. On my reading, the basis of Kant’s practical arguments for freedom (roughly: in Groundwork III our awareness that we have to make up our own about what to do, and in the second Critique our awareness of unconditional ) is either not

6244-678-1pass-007-r02.indd 148 8/21/2015 4:19:14 AM Kant and Peirce on Belief 149 in good epistemic standing (Kant admits that whether we’re transcendentally free cannot be decided on purely theoretical grounds) or (if one focuses only on how things present themselves to rational agents, whether true or not) is not indicative of our “really” being free (as opposed to our necessarily taking ourselves to be free). Hence, the sense in which Kant’s practical arguments for freedom are evidentialist arguments (that is, arguments that offer reasons for taking ourselves to be free) is compatible with them not being eviden- tialist in the more demanding sense of evidence adopted here. I was greatly helped in revising this paper for publication by an exchange with Robert Stern on the issue of evidentialism. Stern does agree with the irst and second point made in this footnote, but not with the third. 21. Kant anticipates Ramsey and present-day Baysians by claiming that the “touchstone” for how irmly an agent is committed to some belief is bet- ting. Although Kant himself doesn’t put it that way, betting, on the Kantian account of belief, can be understood as artiicially turning some judgment into a judgment about enabling conditions. If I opine that it’s going to rain tomorrow, there may be nothing at all hanging on whether I’m right. So my assent cannot be a case of belief. But if you challenge my opinion and pro- pose a bet, my accepting that bet with the aim of winning some money will be successful only if it rains tomorrow. By betting on it, tomorrow’s raining has become an enabling condition for my winning the bet and thus the con- tent of a belief of mine. 22. This means that there are two further types of belief besides pragmatic and moral for which Kant, however, does not reserve a special label, namely contingent beliefs that have necessary ends and necessary beliefs that have contingent ends. I deviate here from Höwing’s interpretation, which claims that Kant identiies contingent with pragmatic and necessary with moral be- lief, but also notes that this seems unwarranted. 23. In later writings, Peirce denies that belief has to be conscious (cf. CP 5.417; 5.480). I have beneited from the online Commens Dictionary of Peirce’s Terms, edited by Mats Bergman and Sami Paavola, 2003-, which offers a collection of some relevant passages by Peirce on belief (http://www.helsinki. i/science/commens/dictionary.html). 24. Even though Peirce often uses the terms “opinion” and “belief” interchange- ably, he sometimes distinguishes them in a way that, although different from the Kantian distinction, is not altogether dissimilar: “We believe the prop- osition we are ready to act upon. Full belief is willingness to act upon the proposition in vital crises, opinion is willingness to act upon it in relatively insigniicant affairs” (CP 1.635). 25. Recall that evidence forp needs to be positively correlated to the truth of p so that the mere absence of doubt does not, as such, count as evidence. 26. More precisely: Without the belief’s being true, the action can be successful only accidentally, by sheer luck. 27. Kant discusses “habit” in the context of judgment and prejudice at KGS 9:76. Also note that there are some interesting similarities between Peirce’s conception of belief and habit and Kant’s concept of a , which is a rule for action and has a habitual aspect, but differs from belief in being a “de- termination of the will,” that is, a practical attitude, not a theoretical one. 28. One may wonder whether it might not be suficient for rational agency to hope, rather than believe, that the enabling conditions for one’s action’s being successful obtain, where hoping that p would be compatible with not believing that p. In another paper, I argue that rationally trying to do A re- quires, minimally, that one does not believe that doing A cannot be success- ful (Willaschek forthcoming). In cases of straightforwardly doing something

6244-678-1pass-007-r02.indd 149 8/21/2015 4:19:14 AM 150 Marcus Willaschek that one takes oneself to be able to do, this doesn’t seem to be suficient, though. If I go to turn on the lights (and do not just take myself to be trying to do so), I must take myself to be able to turn them on, which means that I believe that all enabling conditions for my turning on the lights obtain and all that is further required depends on my own decision to turn them on. If that is correct, Kant’s account of moral belief could be defended in this re- spect (since according to Kant we can be morally certain of our ability to act as the moral law requires), while his account of pragmatic belief may need to be revised (since the doctor, in Kant’s example, can at best try to heal the patient by treating him for consumption, which would not require that he believe that the patient has consumption, but only that he not believe that he does not have consumption). 29. Thanks to Gabriele Gava, Thomas Höwing, Andreas Kemmerling, and Rob- ert Stern for very helpful comments on earlier versions of this paper.

REFERENCES

Bergman, M., and Paavola, S., eds. 2003. Commens Dictionary of Peirce’s Terms. http://www.helsinki.i/science/commens/dictionary.html Chignell, A. 2007. “Belief in Kant.” Philosophical Review, 116: 323–60. Conee, E., Feldman, R. 2004. Evidentialism. Oxford: Oxford University Press. Haack, S. 2007. “The Legitimacy of Metaphysics: Kant’s Legacy to Peirce, and Peirce’s to Philosophy Today.” Polish Journal of Philosophy, 1: 29–43. Höwing, T. Forthcoming. “Kant on Pragmatic and Moral Belief.” In The Highest Good in Kant’s Philosophy, edited by T. Höwing. Berlin/Boston: de Gryuter. Hookway, C. 2012. “Logical Principles and Philosophical Attitudes: Peirce’s Re- sponse to James’s Pragmatism.” In The : Essays on Peirce and Pragmatism, 182–96. Oxford: Oxford University Press. Hume, D. (1748) 1999. Enquiry Concerning Human Understanding. Edited by T. L. Beauchamp. Oxford: Oxford University Press, 1999. James, W. (1897) 1979. The Will to Believe and Other Essays in Popular Philoso- phy, edited by F. H. Burkhardt, F. Bowers and I. K. Skrupskelis. Cambridge, MA: Harvard University Press. ______. (1907) 1975. Pragmatism: A New Name for Some Old Ways of Thinking. In Pragmatism, edited by F. Bowers and I. K. Skrupskelis, 1–144. Cambridge, MA: Harvard University Press. Kant, I. 1900-. Kants gesammelte Schriften. Berlin: De Gruyter, Reimer. Quotations in English are given according to: The Cambridge Edition of the Works of Im- manuel Kant, edited by P. Guyer and A. Wood. Cambridge: Cambridge Univer- sity Press, 1992. Murphey, M. G. 1968. “Kant’s Children: The Cambridge Pragmatists.” Transac- tions of the Charles S. Peirce Society, 4: 3–33. Pasternack, L. 2014. “Kant on Opinion: Assent, Hypothesis, and the Norms of Gen- eral Applied Logic.” Kant-Studien, 105: 41–82. Peirce, C.S. 1931–58. Collected Papers of . Edited by C. Hartshorne, P. Weiss (vols. 1–6), and A. Burks (vols. 7–8). Cambridge, MA:- Har vard University Press. ______. 1992–98. The Essential Peirce: Selected Philosophical Writings. Edited by The Peirce Edition Project, 2 vols. Bloomington: Indiana University Press. Pihlström, S. 2004. “Peirce’s Place in the Pragmatist Tradition.” InThe Cambridge Companion to Peirce, edited by C. Misak, 27–57. Cambridge: Cambridge Uni- versity Press.

6244-678-1pass-007-r02.indd 150 8/21/2015 4:19:14 AM Kant and Peirce on Belief 151 ______. 2010. “Kant and Pragmatism.” Pragmatism Today, 1: 50–61. Schwitzgebel, E. 2014. “Belief.” In The Stanford Encyclopedia of Philosophy (Spring 2014 Edition), edited by E. N. Zalta, . Stevenson, L. 2003. “Opinion, Belief or , and Knowledge.” Kantian Review, 7: 72–100. Thayer, H. S. 1981. Meaning and Action: A Critical History of Pragmatism. Second edition. Indianapolis: Hackett Publishing. Willaschek, M. Forthcoming. “Must We Believe in the Realizability of Our Ends? On a of Kant’s Argument for the Postulates of Pure Practical Reason.” In The Highest Good in Kant’s Philosophy, edited by T. Höwing. Berlin/Boston: de Gruyter.

6244-678-1pass-007-r02.indd 151 8/21/2015 4:19:14 AM