Oxford Scholarship Online

Transcription

1 University Press Scholarship Online Oxford Scholarship Online Religious Faith and Intellectual Virtue Laura Frances Callahan and Timothy O'Connor Print publication date: 2014 Print ISBN-13: Published to Oxford Scholarship Online: June 2014 DOI: /acprof:oso/ Trust, Anti-Trust, and Reasons for Religious Belief Linda Zagzebski DOI: /acprof:oso/ Abstract and Keywords This chapter distinguishes between theoretical or thirdpersonal reasons for believing religious propositions (facts that are logically or probabilistically connected to the truth of those propositions, typically communicable) and deliberative or first-personal epistemic reasons (having an essential connection to the person who holds them, including her experiences and crucially her self-trust). Attacks on the intellectual virtue of a faith commitment admit of different kinds of defenses depending on the category of reasons impugned and have different consequences if successful. Attempts to undermine a believer s confidence, for example by showing how evolutionary biology or psychology might explain away belief, should be seen for what they are: attacks on one s first-personal reasons that should purport to have a pervasive impact on one s beliefs, yet do not bear on third-personal arguments. Keywords: first-person, third-person, reasons, self-trust, virtue, faith 10.1 Introduction Page 1 of 22

2 What I mean by an epistemic reason to believe p is something on the basis of which a reasonable person can settle for herself whether p. An epistemic reason need not be sufficient to settle the question whether p, but it is the sort of thing that can do so, normally in conjunction with other epistemic reasons. I will argue first that there are two kinds of epistemic reasons one irreducibly first personal, the other third personal. Epistemic self-trust is an irreducibly first personal epistemic reason, and it is the most basic epistemic reason we have. Attacks on religious belief are sometimes attacks on third person reasons, but they are sometimes attacks on epistemic self-trust or trust in religious communities. Attacks on self-trust need to be handled in a different way than attacks on third person reasons for belief The Distinction Between First Person and Third Person Reasons I call the kind of reasons that are irreducibly first personal deliberative reasons, and the kind of reasons that are third personal theoretical reasons. My use of the terms deliberative and theoretical is not essential to the distinction I am making, but these terms draw attention to the different functions of the two kinds of reasons in our psychology. By theoretical reasons for believing p I mean facts that are logically or probabilistically connected to the truth of p. They are facts (true propositions) about states of the world that, taken together, give a case for the fact that p. 1 Reasonable persons care about (p.232) getting the truth, and so they care about getting indicators that what they believe (or a candidate for their belief) is true. Theoretical reasons are not intrinsically connected to believing, but they are reasons because a reasonable person who comes to believe them and grasps their logical and probabilistic relations to p will see them as indicating the truth of p. Theoretical reasons can be shared with others laid out on the table, so they are third personal. They are relevant from anyone s point of view. In fact, they do not require a point of view to be reasons. The connections between theoretical reasons and what they are reasons for are among the facts of the universe. What we call evidence is most naturally put in this category of reasons, although the notion of evidence is used in many different Page 2 of 22

3 senses, and I do not insist that what we mean by evidence is limited to theoretical reasons. In contrast, what I mean by deliberative reasons has an essential connection to me and only to me in my deliberations about whether it is the case that p. Deliberative reasons connect me to getting the truth of p, whereas theoretical reasons connect facts about the world with the truth of p. Like theoretical reasons, deliberative reasons are reasons because a reasonable person a person who cares about the truth, takes them to be indicators of truth. Deliberative reasons provide me reasons for p that are not simply weightier than the reasons they provide others. They are not reasons at all for other persons. They are irreducibly first personal. To see the distinction I have in mind, consider experience as a reason for belief. If I have an experience, the fact that I have it is a theoretical reason that supports a variety of propositions. I can tell you about my experience, and if you believe what I tell you, you can then refer to the fact that I had the experience as a reason to believe whatever it supports. So suppose I have the experience of seeming to see a scissor-tailed flycatcher near my home. You and I can both refer to the fact that I had that experience as a reason to believe that the flycatchers have not yet migrated, and so can anybody else who is aware of the fact that I had the experience. The fact that the experience occurred is therefore a theoretical reason. It is on the table for anyone to consider, and anyone can consider its logical and probabilistic connections to other facts about the world. However, I am in a different position with respect to my experience than you are because I not only grasp the fact that I had the experience, but in addition, I and I alone had the experience. I am the one who saw the flycatcher with its long split tail. That visual experience can affect my reasoning processes, emotional responses, and the way I come to have or give up certain beliefs directly, and that is perfectly normal for human beings. In contrast, the fact that I had the experience is something you and I and any number of other persons can come to believe. So my experience of seeing the scissor-tailed flycatcher gives me a reason to believe that the birds have not yet gone south. You cannot have my experience, but you can believe that I had the experience. When you do so, you are not accessing my experience: you are accessing the fact that the Page 3 of 22

4 experience occurred. Of course, I can access the same fact, but my having a reason to believe that scissor-tailed flycatchers have not yet migrated does not depend upon my accessing the fact that I had the experience of seeing one. The seeing itself (p.233) gives me a reason to believe they are in the neighborhood. Perhaps the visual experience is a stronger reason than the fact that the experience occurred, but I think that is a misleading way to put it. The experience is not the same kind of thing as the fact that the experience occurred. Both provide reasons for belief, but they are reasons of a different kind. Another type of deliberative reason is what are often called intuitions in one of its senses. What I mean by an intuition is, roughly, something internal to the mind that responds with an answer to a question, often about a concrete case. For example, I have the intuition that it is not morally permissible to directly kill an innocent person to save five others, but someone else might have a different intuition. Most philosophers have the intuition that a Gettier case is not an instance of knowledge, but we probably have encountered students who do not have that intuition. I have no position on the strength of an intuition as a reason to believe what the intuition supports. Maybe it is strong, maybe it is not. But in so far as it is a reason at all, it is a deliberative reason. My intuitions are mine alone, and they give me but not you a particular kind of reason for certain beliefs. But again, the fact that I have an intuition can be put out on the table. I can tell you that my intuition is such and such. When I do so, I give you a theoretical reason supporting some position. The fact that many people have the same intuition can also be used to support a position. So the fact that a large majority of those persons who have carefully thought about the nature of knowledge have the intuition that a Gettier case is not an instance of knowledge supports the position that a Gettier case is not an instance of knowledge. Intuitions, then, are like experiences. An intuition and an experience provide the agent with first person reasons to believe something, but the fact that the experience occurred or that the intuition is what it is can be treated as evidence, as a theoretical reason for the truth of some proposition. Page 4 of 22

5 Experience and intuition reveal an important feature of deliberative epistemic reasons: They are psychic states of a person that seem to her to indicate the truth of some proposition p. Human beings are constituted in such a way that certain states are like that. We would expect, then, that other psychic states can have the same function, for instance, memories and certain emotions. My favorite example of an emotion that can be a reason for a belief is admiration. My admiration for an epistemic exemplar can be a reason to believe what the exemplar believes in the domain of her exemplarity. That is a deliberative reason, not a theoretical one. Of course, it is not a good reason unless my admiration for the epistemic admirability of the exemplar satisfies certain conditions, including my reflection on the responses of other persons. But when it is a reason, it is a reason only for me, not others. I think also that a belief state can be a deliberative reason. I have argued elsewhere that a state of believing p can be a deliberative reason to continue to believe p. 2 But it is (p. 234) not necessary for present purposes to defend that claim or to give an exhaustive list of deliberative reasons. My general claim is that reasonable persons take certain of their conscious states to be indicators of the truth of some proposition. I propose that these states include experiences, memories, intuitions, some emotions, and epistemic states. What makes these states reasons is that a reasonable person takes them to be truth-indicators. What makes these reasons deliberative is that they are reasons only for their possessors. Theoretical and deliberative reasons have features that make it important to distinguish them. For one thing, theoretical and deliberative reasons do not aggregate. As far as I know, no one has yet figured out how to reduce the first person perspective to the third person perspective or vice versa, nor has anyone identified a perspective that is common to both perspectives. What is a reason from the first person perspective is not a reason from the third person perspective because conscious states are not facts. Likewise, a reason from the third person perspective is not a reason from the first person perspective. The fact that p is not at all the same thing as my conscious states. Earlier I compared a visual experience of seeing a certain species of bird with its third person counterpart the fact that the experience occurred, and I said that they function differently as reasons to believe the birds have not yet gone south. I said it is misleading to say that the experience is a Page 5 of 22

6 stronger reason than its third person counterpart because they differ in kind. This is not to deny that frequently conscientious persons treat their experience as a stronger reason, but I deny that there are rules that determine that whenever someone has an experience of a certain kind, it has a certain weight in comparison with its third person counterpart for example, it is x percent stronger. For the same reason, there are no rules that determine the way any conscientious person ought to weigh the support for p given by her experience against third person reasons she can access for or against p. When I say that theoretical and deliberative reasons for p do not aggregate, I am not denying that they can both raise my confidence in the truth of p. If I have theoretical reasons for p and then get a deliberative reason for p, that often (although not always) increases my confidence that p. Similarly, getting theoretical reasons when I already have deliberative reasons can increase my confidence. But there is no formula that determines whether and how much confidence a given combination of theoretical and deliberative reasons gives a conscientious person. The second important difference between theoretical and deliberative reasons is that I relate to my deliberative reasons as an agent, but my agency is independent of theoretical reasons. I am passive with regard to the facts. My only job when I reflect about the facts is to figure out what they are and what they support. In contrast, my conscious (p.235) states are parts of myself and, because I am self-reflective, I can manage those states. Self-reflection is the process by which the self guides itself. One goal of self-reflection is to get the truth. A reasonable person reflects upon her memories, intuitions, emotions, and beliefs because she thinks that reflection makes it more likely that she will get true beliefs and avoid false ones. Sometimes reflection can make the state itself change for example, reflection on someone I admire and the responses of others to that person can change my emotion of admiration. More commonly, reflection upon a set of my beliefs, emotions, memories, etc. can change what I take to be the relations of support between some of my conscious states and certain beliefs. Deliberative reasons, then, respond Page 6 of 22

7 to self-management. I call them deliberative because they respond to deliberation The Primacy of Self-Trust Suppose I am considering the reasonableness of believing some proposition p. Perhaps I already believe p, but reflection leads me to look for reasons for p. As I have said, a reasonable person wants reasons for her beliefs. Or perhaps I am not sure whether p, so I want reasons in order to determine whether p. I know I might get deliberative reasons, but deliberative reasons are often the sorts of things I cannot simply look for. Theoretical reasons are facts, and I can always look for facts. So if I want reasons for p, I will typically look for theoretical reasons for p. But since a theoretical reason is a fact independent of me, it does not operate as a reason for me to believe anything until I take it on board. I must be aware of the facts that bear on whether p if those facts are reasons for me to believe p. However, my taking a certain set of theoretical reasons for p as reasons for me to believe p is not sufficient to make it likely that p is true. That is because my taking something to be a set of theoretical reasons for p is irrelevant to the actual connection between those reasons and p unless I am taking them properly have accurately identified the appropriate facts, have figured out the correct logical and probabilistic relations between those facts and p, have appreciated the significance of individual facts, and have not left anything out. Hopefully all of that is a fact about me. But that fact depends upon the more basic fact that my epistemic powers are conducive to getting me the truth the fact that I have the kind of powers that enable me to find out facts about the world. But now the question arises: How can I access that fact? It has been pointed out by many others for example, Richard Foley (2001) and William Alston (1986, 2005) that any reasons I have to believe that my powers connect me to the truth are circular. I cannot tell that my powers get me to the truth without using those powers. In fact, I cannot tell in a non-circular way that my epistemic powers ever get me to the truth, much less that they get me to the truth reliably. It would take a perspective outside of my mind to identify the quality of the relationship between my mind and a world outside of it, Page 7 of 22

8 but I am a being who can never ascend to a perspective outside of my own mind, save in imagination. (p.236) Notice that I face the same problem for deliberative reasons as for theoretical reasons. Just as I have no way of telling in a non-circular way that my attempt to access the theoretical facts gets me to the truth, I also have no way of telling that my sense experience, memory, or intuition get me to the truth without using some of my own powers. However, there is a difference between the problem as it arises for my deliberative reasons and the problem for my theoretical reasons. For theoretical reasons, I have two problems: (a) accurately identifying the theoretical reasons, and (b) accurately identifying a relationship between those reasons and some proposition p such that that relationship makes it likely that p is true. I cannot do either (a) or (b) unless my powers are truth-conducive. When, upon reflection, I ask myself whether I have reason to believe that I can do (a) and (b), I realize that I cannot do either one unless my powers are conducive to getting me the truth. But I do not have reason to think that my powers are conducive to getting me the truth without assuming (a) and (b). My situation with respect to my deliberative reasons is somewhat different. Since my deliberative reasons are conscious states, their existence is often (although not always) self-evident. So for at least some deliberative reasons, I do not have a problem parallel to (a). Nonetheless, I have problem (b) for deliberative reasons as well as for theoretical reasons. I cannot tell that my deliberative reasons connect me to the truth without using powers the truth-conduciveness of which I must assume, but I cannot tell that the powers that give me deliberative reasons are generally truth-conducive without appealing to particular outputs of those powers. I do not think it is an exaggeration to say that various forms of this problem have dominated epistemology since Descartes. Not all reasonable persons notice the problem, but I think it is significant that those who do notice it rarely respond with a degree of doubt that prevents them from having beliefs. Reasonable persons have beliefs even if and when they realize that they have no guarantee that their total set of theoretical reasons for p and the deliberative reasons of the sort we have Page 8 of 22

9 discussed so far are sufficient to indicate the truth of p for a reasonable person. What is the reasonable response to this problem about reasons? Both Foley and Alston maintain that it is self-trust, but neither of them tells us what self-trust consists in. I agree that reasonable persons trust their faculties both before and after they notice the problem of the circularity of reasons. But I think that if self-trust serves the function of being a response to our epistemic predicament, rather than just a statement of it, self-trust must have certain features. First, self-trust must be more than the belief that my epistemic powers are generally conducive to getting me the truth. That belief precedes the awareness of epistemic circularity and is not sufficient as a response to it. What is important about the awareness of circularity is that it raises the issue of doubt, and once that issue is raised, I must either succumb to doubt or dispel it. I have already said that reasonable persons rarely succumb to generalized doubt. But doubt is not a belief. In particular, it is not the belief, My beliefs might be mostly false. A crucial feature of doubt is its affective element. (p. 237) Doubt is in part a feeling that leads one to weaken or give up the beliefs that are the object of doubt. Since doubt is partly an affective state, it takes an affective state to dispel it. I think, then, that trust is the response of reasonable persons to doubt only if it has an affective component that dispels doubt. I propose, then, that epistemic self-trust includes at least two components: (a) believing or taking for granted that my epistemic powers are suited to getting me the truth that they are trustworthy, and (b) having a feeling that dispels doubts about their trustworthiness, or holds doubt at bay. I take it, then, that self-trust has an affective element as well as an epistemic element. 3 Notice next that if, as I proposed at the beginning of this chapter, a reason for p is something on the basis of which a reasonable person settles for herself whether p, then self-trust is a reason, but it is a deliberative reason, a reason in the same category as experiences, emotions, and intuitions as a ground for belief. Furthermore, it follows from what I have said that epistemic self-trust is more basic as a reason to believe any proposition p than any other reason I can have, whether theoretical or deliberative. 4 It is in virtue of a state that is partly affective that I take it that what I identify as reasons for Page 9 of 22

10 Page 10 of 22 some proposition p point to the truth of p without succumbing to doubt. Self-trust is a reason for me and me alone. It is the most basic reason I can have, and it is partly affective. 5 I have mentioned a number of deliberative reasons for belief: experience, intuition, memory, admiration, epistemic self-trust. I have argued elsewhere that epistemic trust in others is a commitment of self-trust. In so far as I see no relevant difference between myself and others, I am committed to trusting them in the same way I trust myself. Consistency leads me to trust others when I trust myself because trust in myself for the goal of getting truth includes trust in certain powers of my self, powers shared by others. 6 Epistemic trust in others is also a deliberative reason for belief. It is a state directed towards the epistemic powers of others that is partly epistemic and partly affective. My trust in others is a state only I can have. It can function as a reason for my beliefs in the same way my self-trust and admiration and intuition function as reasons. I trust myself in particular when I am being epistemically conscientious using my powers as carefully as I can to get the truth. I trust others in particular when the conscientious exercise of my powers reveals that they are conscientious. My acceptance of epistemic norms of all kinds derives from trust in myself and others. It is because of trust in self and others that I trust the rules of reasoning that have been devised by persons I trust. The intellectual virtues are the qualities of conscientiously reflective persons who attempt to get truth (or some other epistemic good such as understanding). (p.238) It is reasonable for me to attempt to acquire the intellectual virtues only because it is reasonable to have a more basic trust in the epistemic powers of cognitive agents. There would be no point in rules for the conscientious exercise of powers that are devised by the conscientious exercise of those same powers unless we reasonably trusted the powers. Similarly, there would be no point in advocating open-mindedness, intellectual carefulness, courage, etc., each of which is a trait that we identify as desirable for epistemically conscientious agents by the conscientious exercise of the powers of conscientious agents unless it was reasonable to trust those powers. 7 It is reasonable to trust the powers of conscientious epistemic agents and the norms and virtues that conscientious agents endorse because reasonable persons do so. The trust that undergirds epistemic norms and the identification of the

11 intellectual virtues is first personal. There is therefore a deliberative epistemic reason that is not only more basic than theoretical reasons and other deliberative reasons, but it is more basic than the norms of belief-formation. Ultimately, to be reasonable is to do what conscientious persons persons I conscientiously trust do. Trust in others leads to trust in communities, some of which consist of living persons (e.g., members of one s academic profession), and some of which extend far into the past. Religious communities are almost always in the latter category. There are deliberative reasons for the beliefs of a community as well as theoretical reasons, and the way the community identifies theoretical evidence often depends upon trust in the community s ability to get the truth, just as it does for individuals. Of course, members of a community believe many things that do not arise out of the community per se, but communities often function like a self: They have norms of belief formation and shared background beliefs that derive from trust in the community that operates the way self-trust operates for an individual. I have said that epistemic self-trust is reasonable because reasonable persons have it. Reasonable persons also have basic trust in others. It might appear that I am suggesting that reasonable persons trust out of blind hope, which reasonable persons themselves ought to take to be unreasonable. But that is not correct. It is not blind hope to manage the self as all reasonable persons do. The self manages itself by reflecting upon its conscious states. That is what a self does in a world that is much larger than the self and which includes many other selves. Perhaps we wish that that were not the case. We might prefer to be able to go outside our self in order to determine what states of the self ought to survive and which ought to be given up. But we know that is impossible. A self just is a being that can only manage itself from the inside. I think this tells us something about the object of self-trust. To trust the self is to trust the self s ability to manage itself by exercising its power to conscientiously reflect upon states of the self. Some of these states have aims, such as truth, in the case of beliefs. So to trust myself epistemically is to trust my ability to manage my beliefs for the goal of (p.239) truth by conscientious reflection on my total set of psychic states. Since trust has an affective as well as an epistemic component, this Page 11 of 22

12 means that I take for granted that I am likely to get the truth when I conscientiously reflect on myself in this way, and I have a feeling that dispels doubt about my ability to reach the truth in this way. The ultimate test that I have reached the truth in any given case is survival of conscientious reflection. We reflect upon all our conscious states, evaluate them according to norms that we have adopted as the result of previous reflections, and adjust our beliefs in an attempt to make those states survive future conscientious self-reflection, given that we expect there will be changes to the self with new experience, new memories, changes in what we feel and what we trust, and changes in other beliefs. That is all a reasonable person can do. Page 12 of 22

13 10.4 Theoretical Attacks on Religion Reasonable persons want reasons for their beliefs, and attacks on their reasons are indirectly attacks on their beliefs. They may, of course, have more than one set of reasons for a given belief, and they may have more than one kind of reason. So an attack on one set of reasons is not necessarily sufficient to rationally undermine their belief. But given what I have argued in the previous section, there is a difference between an attack on one s theoretical reasons and an attack on one s deliberative reasons. There have been historically important attacks on religious belief in both categories. An attack on one s self-trust or trust in others is an especially virulent form of attack because it undermines a vast portion of one s epistemic structure. I will discuss attacks on self-trust in the next section. In this section I will discuss attacks on religious belief that proceed either by attacking one s theoretical reasons for a religious belief R, or by offering theoretical reasons to believe not-r. Many, but not all, religious believers have theoretical reasons for their beliefs. Those who do are usually people who are used to discussion and debate with others. Given certain conditions, they will think it important that a religious belief can be supported by reasons that they can put out on the table for all to consider. If their own reasons for belief are primarily deliberative personal religious experience, trust in their religious community, admiration for the Gospel message, etc. and if their acquaintances are people who have similar deliberative reasons, it might not occur to them that theoretical reasons are important or worth the effort to discover and examine. And they might be right about that. But sometimes their own reflections will lead them to notice incongruities in their beliefs, and that can be a motive to look for evidence for theoretical reasons. The traditional arguments for theism provide theoretical reasons for theism, whereas the traditional problem of evil provides theoretical reasons for the denial of theism. So a reasonable person looking for theoretical reasons regarding theism will usually find both theoretical reasons for theism and theoretical reasons against it. Her job with respect to these reasons is to attempt to identify them accurately and to accurately identify their logical or probabilistic connection to the thesis of theism. Page 13 of 22

14 (p.240) Since theoretical reasons are facts, they are accessible in principle to anybody, but theoretical reasons are not necessarily facts that many people believe, nor need they even be facts of which many people are aware. So the following could be among a person s theoretical reasons for believing in God or believing some particular religious proposition: (1) The authors of the Bible were reliably reporting the experiences of Abraham and the prophets. (2) Most of the people in human history have believed in God. (3) The world view of the Christian religion is more coherent as a view of the world taken as a whole, including human consciousness, moral reality, aesthetic features, etc, than a naturalistic world view. (4) My neighbor J had a mystical experience after which he changed his life and became the kind of person I admire. Of course, since theoretical reasons are just facts, there are uncountably many other facts that have logical or probabilistic relations to the thesis that God exists. But the theoretical reasons any one person can access are limited, and they differ from person to person. Many people do not believe (1), (2), or (3), and we can easily imagine that almost nobody believes (4) because very few people in the world have even heard of neighbor J. But these people have access to other facts about the world that support theism or support atheism. How does a reasonable person handle her theoretical reasons for and against theism or other religious beliefs? As I have argued, theoretical reasons are derivative from basic self-trust in our faculties, and the force of these reasons for or against a given proposition for any given person is no stronger than that person s trust in the faculties she uses in coming to accept the premises and the rules of logical or probabilistic inference used in the argument. But of course she may have that trust. If so, she may be able to weigh the probability that theism is true, given all the theoretical facts of which she is aware. I would not suggest it is easy to do this, but perhaps it is possible in principle. Page 14 of 22

15 Page 15 of 22 It is important to notice, however, that a person s total epistemic situation with respect to belief in God may include much more than her theoretical reasons because she may have deliberative reasons to believe there is a God. In this category would be included such things as her personal religious experiences, her admiration for the Gospel message, a moral intuition that supreme goodness is possible, or her trust in her religious community. In the last section I argued that theoretical and deliberative reasons do not aggregate. When a person considers only her theoretical reasons for or against the existence of God, she can weigh the evidence that there is a God, given the facts as she sees them. What she cannot do is to add her deliberative reasons to her theoretical reasons to get a total case for or against the existence of God since her first person reasons do not aggregate with reasons from a third person perspective. (p.241) What does a reasonable person do when she has both theoretical reasons and deliberative reasons for and against the same proposition? As I said in the last section, there is only one thing she can do, and that is to reflect as conscientiously as she can on her total set of beliefs, emotions, and memories of past experiences, as well as anything else in her psyche that responds to reflection. Typically, a reasonable person s confidence in the truth of one of her beliefs decreases once she becomes convinced of a theoretical reason against it of which she was previously unaware. Likewise, her confidence increases when she becomes aware of a new theoretical reason for the belief. But there is no formula that dictates how a reasonable person manages her psychology. In fact, it follows from what I have said that it is impossible that there is such a thing, given that all our norms of reasoning and belief-formation are derived from our basic ability to manage our psychic economy. What we do when we reflect conscientiously dictates the rules, not vice versa. Reasonable persons do not all respond the same way to theoretical reasons even when they are in possession of all the same theoretical facts, including facts about the experiences, memories, emotions, and degrees of trust each person has. The reason for differences in response is differences in their first person reasons. An experience gives me a different kind of reason to believe something than awareness of the fact that the experience occurred. Since deliberative and theoretical reasons do not operate the same way in our psychic economy,

16 there is no reason to expect that the level of support given by an experience is the same as the level of support given by awareness of the fact that the experience occurred. To generalize this point, a being with no inner life no experiences or personal memories, no emotions, no feelings of trust, no intuitions, but who is aware of all the facts about the inner life of some other being, would not possess the same epistemic reasons as that other being. Our deliberative reasons are ours alone, and although awareness of the fact that those reasons exist is also an epistemic reason, it is not a reason of the same kind, and it is not capable of conferring the same level of support as deliberative reasons. This is not to say that deliberative reasons are always stronger. No doubt they are not. But they are the kind of reason that is capable of being so. There will always be differences between reasonable persons in the way they handle theoretical reasons for and against religious belief because there will always be differences in their deliberative reasons bearing on the same beliefs Debunking Self-Trust Beliefs are not always attacked via attacks on theoretical reasons for the belief, or by giving theoretical reasons for the denial of the belief. There are also attacks on one s deliberative reasons. The most extreme attack of this kind is an attack on epistemic self-trust. Given that self-trust is a more basic reason than any other, a general attack on self-trust is an attack on my norms of reasoning, the intellectual virtues I endorse, and my higher order belief that most of the beliefs I have already acquired are true. (p.242) A successful attack on self-trust undermines my entire epistemic structure. Most attacks on self-trust, however, are not designed to be so radical. They are intended to be surgical strikes against a particular epistemic community, or a specific subset of the belief-forming faculties of individuals. There are historically important attacks on religious belief in this category. Freud s attack on religion is a classic example. If religious belief would satisfy our need for safety and the exorcism of our fears, it is claimed that the belief is not caused by faculties aimed at truth, and so we cannot trust the connection between the dispositions in us that produce the beliefs and the truth. Another well-known example is the claim that evolutionary Page 16 of 22

17 Page 17 of 22 theory reveals that our beliefs arise from mechanisms that were selected for survival, not correspondence to reality. Attacks of this kind aim at destroying self-trust, and sometimes they succeed. 8 It is important to see that these attacks offer theoretical reasons aimed at undermining self-trust. So, for example, the Freudian hypothesis offers evidence that the causes of religious belief include wishful thinking and other motives that are not reliably truth-related (Freud [1928] 2011). Similarly, evolutionary attacks on religious belief offer the evidence of evolutionary theory with natural selection as an explanation of the origin of human epistemic faculties. If the hypothesis is true, it would have the consequence that the cause of beliefs produced by these faculties is independent of truth. Such a consequence undermines self-trust in the general reliability of epistemic faculties. Notice that these are attacks on aspects of the self used in self-reflection. As I have argued, ultimately our only way to tell that any belief is true is that it survives conscientious selfreflection, but self-reflection requires a substantial amount of trust in oneself. Evidence we trust that indicates that conscientious reflection is not apt to produce true belief in broad areas of belief formation undermines trust in the reflective capacities needed for believing anything, including the hypotheses generating the problem. Even if the attack is only directed at one category of belief, for example moral, political, religious, or philosophical beliefs, loss of trust in the faculties and dispositions that lead to beliefs in that category can easily lead to a loss of trust in the faculties that lead to beliefs in many other categories because they are usually the same faculties. We do not, after all, have very many beliefforming faculties, and if we have reason to think these faculties are untrustworthy in their deliverances in one area, there is reason to suspect their deliverances in general. At the worst, the evidence casts doubt on the trustworthiness of the norm of conscientious self-reflection on the connection between following the norm and getting the truth. But we cannot act rationally if we doubt the connection between conscientious self-reflection and success at getting the truth. This should lead us to be skeptical that a surgical attack on the faculties that lead to religious beliefs can succeed. An attack on the powers that produce religious beliefs undermines (p.243) more than religious beliefs. I do not deny

18 the possibility that there are mystics who have distinctive religious powers, but by far the greatest proportion of religious beliefs, even among mystics, arise from epistemic powers that they use in other domains. It is doubtful, then, that this approach can undermine religious belief without thereby undermining a vast area of beliefs in other categories, including beliefs in the Freudian hypothesis or the evolutionary hypothesis. Let us assume, however, that a reasonable person reflecting upon the Freudian hypothesis is not led to give up trust in the faculties that produced belief in the Freudian hypothesis. As I have already argued, what she must do, and the only thing she can do is to conscientiously reflect upon her total set of beliefs and other psychic states. In doing so she can tell whether her religious beliefs are harmonious with her deliberative reasons and theoretical evidence, including, we ll assume, the Freudian hypothesis, or whether instead there is dissonance between her total reasons and her religious beliefs. Her religious beliefs may or may not survive conscientious selfreflection. If they do, it does not matter whether they were originally acquired through desire, instinct, testimony, personal experience, deduction from some other belief, or by some other mechanism. Survival of conscientious selfreflection is the ultimate test of truth. The same point applies to belief in the Freudian hypothesis itself. That also will not be adopted by the conscientiously reflective person unless the hypothesis survives conscientious reflection. It might or it might not. If it does survive, it should not reduce trust in those religious beliefs that also survive conscientious reflection. That would be a case in which the conscientious person judges that her religious beliefs are true even though she judges that the Freudian hypothesis is also true. The evolutionary hypothesis about religious beliefs is in the same situation. In each case we reflect in more than one way. We reflect upon theoretical and deliberative reasons for and against the hypothesis, and we reflect upon the conjunction of the hypothesis with our total set of beliefs, emotions, memories, and other psychic states. Conscientious reflection on a hypothesis includes reflection on higher-order beliefs about the hypothesis, such as reflection about the way persons we trust when conscientious respond to the hypothesis. If such Page 18 of 22

19 persons either reject the evolutionary hypothesis about the origin of religious beliefs, or accept it but do not respond by mistrusting religious beliefs, that gives us a deliberative reason to think that such a hypothesis does not undermine trust in religion for the conscientious person. But I will not prejudge how this will be handled within the psyche of an individual person. We also cannot say how much credence any individual person can put in the theoretical reasons that undermine self-trust without the reasons becoming selfdefeating. But I think we can say with confidence that there is a limit to the credence any conscientious person can have in a belief that undermines self-trust in a vast network of her beliefs. There is a more direct attack on epistemic self-trust than the Freudian hypothesis and evolutionary theory. Many people who live for a time in another country, or study the wisdom literature of another culture in depth, find that their trust in their own beliefs is undermined. It is common to think, I would have had different beliefs if (p.244) I had grown up in a different place, and it is an accident of history that I have the beliefs I have. I could have been Hindu, Muslim, Buddhist, Christian, atheist, or many other things. The same line of thought applies to philosophical positions and attitudes about political arrangements. I am a believer in libertarian free will but I could have been a determinist. I am a believer in Western democracy, but I could have believed in Islamic theocracy. This is more serious than the debunking arguments we have already considered because it is not just reason to believe that people in general form beliefs in certain ways. It is an argument that I could have developed a very different self than the one I have, one in which a different set of beliefs and perhaps values would direct my life. Furthermore, if this line of thought arises from my trust in other people whom I believe to be similar to me, it gives me a deliberative reason to weaken my self-trust. It is not simply a theoretical reason like Freudian psychology or evolutionary biology. The attack directly affects people s conscientious reflection upon their own beliefs, and unlike the hypotheses discussed above, conscientious persons do respond to this line of thought by feeling a loss of trust in their beliefs, although I cannot say how many people actually give up beliefs because of these experiences. Page 19 of 22

20 I think this problem reveals a deep dilemma about the nature of the self. On the one hand, I realize that persons whom I trust because of their similarity to me have very different beliefs and values, and that creates a problem for me within the things I trust. But on the other hand, I realize that I am not a different person. When I look back at my life and the way my self has developed through a long sequence of experiences, emotions, and the acquisition of beliefs, some of which change over time, I realize that I could have gone off on a different track at various points. But I am not what I could have been; I am what I am. I must admit that I have the ability to change, and it is possible to radically change my beliefs. Conversion can be the conscientious thing to do for some people, but only when there is something they do trust (not might have trusted), that they trust more than all the beliefs they must give up. But we always change because of something in the self, not because of something that might have been in the self. The awareness that conscientious persons can radically differ from me and from each other in their beliefs is humbling because it shows that a multitude of selves with the same natural faculties and the same ends, and with the same conscientious use of those faculties, end up with different results. That can easily lessen my confidence in my self in particular, even though I will continue to have confidence in human selves in general. Humility is fundamentally the realization of what a self is. We all engage in the task of creating a harmonious self, and we want it to be in contact with reality. Herein lies a puzzle for the self: We think there is something wrong about a clash of beliefs, emotions, or values with those of others, and yet we are also sure there is something right about it too. If I am right that each of us has deliberative reasons for belief that differ from those of any other person, in fact, necessarily differ from those of any other person, then it should not be too surprising that reasonable persons have different beliefs. But this realization does not (p. 245) dissolve the dilemma as long as we also think that there are facts about the universe, facts that are not relative to the selves attempting to discover them. As long as both first person and third person perspectives exist, and as long as both perspectives put rational demands on us, this problem of the self will not be resolved. Religious beliefs are not unique in this respect, but they force us to confront the difference Page 20 of 22

21 between first person and third person reasons for belief in a way that cannot be ignored, and they reveal how much we do not know about the nature of the self and what it means to trust it. References Bibliography references: Alston, William (1986). Epistemic Circularity, Philosophy and Phenomenlogical Research 47: (2005). Beyond Justification: Dimensions of Epistemic Evaluation. Ithaca, NY: Cornell University Press. Barrett, Justin (2000). Exploring the Natural Foundations of Religion, Trends in Cognitive Sciences 4:1: Boyer, Pascal (2003). Religious Thought and Behaviour as By- Products of Brain Function, Trends in Cognitive Sciences 7:3: Foley, Richard Intellectual Trust in Oneself and Others. Cambridge: Cambridge University Press. Freud, Sigmund [1928] (2011). The Future of an Illusion. Connecticut: Martino Fine Books. Pinker, Steven (1997). How the Mind Works. New York: W.W.Norton & Co. Zagzebski, Linda (2012). Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief. New York: Oxford University Press. (2013). Powers and Reasons, in Ruth Groff and John Greco (eds.), Powers and Capacities in Philosophy, New York: Routledge, (2014). Trust, in Kevin Timpe and Craig Boyd (eds.), Virtues and their Vices. New York: Oxford University Press. Notes: (1) In this chapter I do not distinguish facts from true propositions. If there is a difference, the argument of this chapter can be easily amended. Page 21 of 22

22 (2) In Zagzebski (2012: chapter 10), I argue that the popular problem of reasonable disagreement is easier to resolve if we use the distinction between theoretical and deliberative reasons. If we focus only on theoretical reasons, my beliefs are no different than anyone else s. That explains the temptation to say that the reasonable response to peer disagreement is skepticism. However, I argue that my belief state gives me a deliberative reason to continue to have the belief, a reason that nobody else has. Trust in my previous states of self gives me a deliberative reason to continue those same states of self. Of course, this reason can be lost as states of the self change with new experiences and further reflection. (3) I discuss the reasonableness of epistemic self-trust and its components in Zagzebski (2012: ch. 2). I argue there that trust in standard cases includes a behavioral component as well as an epistemic component and an affective component. (4) Of course, self-trust is not more basic than theoretical reasons, which are facts of the universe. But it is more basic than my use of theoretical reasons in my attempt to figure out what the truth is. (5) I argue that powers are more basic than reasons in Zagzebski (2013). (6) I develop this argument in Zagzebski (2012), chapter 3. (7) I argue in Zagzebski (2014) that most of the intellectual virtues either restrain or enhance epistemic self-trust or trust in others, and they all presuppose trustworthiness. (8) See, for example, the focus on religious beliefs in Pinker (1997). Barrett (2000), and Boyer (2003). Page 22 of 22