Believing Against the Evidence

Agency and the Ethics of Belief

Miriam Schleifer McCormick

• Editore: Routledge
• Anno: 2014
• ISBN:
  9781136682759
• Edizione: 1
• Argomento: Filosofía
• Categoria: Epistemología en filosofía

The question of whether it is ever permissible to believe on insufficient evidence has once again become a live question. Greater attention is now being paid to practical dimensions of belief, namely issues related to epistemic virtue, doxastic responsibility, and voluntarism.

In this book, McCormick argues that the standards used to evaluate beliefs are not isolated from other evaluative domains. The ultimate criteria for assessing beliefs are the same as those for assessing action because beliefs and actions are both products of agency. Two important implications of this thesis, both of which deviate from the dominant view in contemporary philosophy, are 1) it can be permissible (and possible) to believe for non-evidential reasons, and 2) we have a robust control over many of our beliefs, a control sufficient to ground attributions of responsibility for belief.

Part I
Doxastic Norms

1 Conceptual Defenses of Evidentialism

When Bernard Williams first introduced the idea of the truth-aim, it was the first of five features discussed to illuminate the nature of belief, and to show “how far, if at all, believing something can be related to decision and will.” Williams actually says of this first feature that it can be “roughly summarized” or “vaguely summed up” as “beliefs aim at truth.”¹ His employment of the truth-aim in an argument for why we cannot believe at will is as follows: If truth is the aim of belief, any states that I can achieve at will would not recognizably be beliefs. For if what I believe were up to me, seemingly, I could form a belief regardless of whether I thought it true—but if I knew this, then I would know that there is no reason to think that this “belief” accurately represents reality. But if to have a belief is to be committed to its truth, then believing entails viewing the belief as representing reality. Believing at will is, thus, incoherent, because it entails one viewing a belief as 1) necessarily representing reality and 2) not necessarily representing reality. Thus, believing at will is not only a psychological impossibility, but the very concept of belief also renders the idea incoherent.

It does not seem that my involvement in bringing about a mental state means that I could not view such a state as representing reality. An obvious counter-example, and one of those used to show that there is a trivial sense in which we can control what we believe, is that I can make it be true that the lights are on by flipping the light switch to turn on the lights. I could, conceivably, have a belief that the lights are off and then decide to believe the lights are on, and make this decision effective by flipping the on switch. Williams does consider the possibility of manipulating oneself in various ways to get oneself to believe something that one wants to believe through processes such as taking drugs or hypnosis, but the beliefs he considers are momentous—such as whether one’s child is alive or dead. Even if such projects of self-manipulation are conceptually possible, Williams argues that they must be irrational and, in any case, none of them are a direct result of a decision to believe.

I will return to the question of whether certain projects of self-manipulation that result in belief can be counted as decisions to believe.² What is most important about Williams’s discussion in the context of defenses of evidentialism is that he was the first to clearly argue that one’s inability to believe at will is not a contingent fact but is rather inconceivable. Thus, he is the first to argue that the idea that beliefs aim at truth expresses a conceptual truth about belief that can help explain other belief-related phenomena. While Williams was not centrally concerned with defending evidentialism, a number of theorists in recent years have turned to Williams as inspiration in thinking that a proper understanding of the concept of belief can reveal why evidentialism is necessarily true.

One may wonder if it makes sense to talk about belief having an aim at all. As Ralph Wedgwood has pointed out, beliefs do not literally aim at anything, as does an archer or a hitman, so this talk is, in a sense, metaphorical.³ There are, however, many activities that obviously have aims where there is no literal “taking aim” at anything. It is common, for example, to ask what the aim of a board game is.⁴ We shall see in the discussion that follows that when many theorists posit the truth-aim for belief, they do not think it is necessary to tie this aim in any way to desires, or conscious goals of agents.⁵ The aim of chess is to checkmate one’s opponent, and this is the case independent of any chess player’s desire to do so. It is even the case if one is aiming to let one’s child win so as to encourage her. Understanding the meaning of the game of chess tells us what its aim is. The conceptual defenses of evidentialism with which we are concerned in this chapter make an analogous claim. Once we understand what it means to have a belief, we can know that its aim is truth and, in turn, appealing to this aim reveals that the norms of beliefs are strictly evidential.

To say that it is a conceptual truth that beliefs aim at truth or that belief is subject to a norm of truth can be understood in a many different ways, and the implications of such a view varies depending on its characterization. In what follows, I will consider a number of different ways that the truth-aim is thought to express a conceptual truth about belief. I begin with what I take to be the most extreme (and ultimately the most implausible) position. I will then consider progressively less extreme versions of the view and will argue that the only plausible characterization of the truth-aim cannot show that evidentialism is true.

1. The Extreme View: A Belief Not Grounded in Evidence is Not a Belief at All

According to one way of understanding the relationship between belief, evidence, and truth, the entire notion of an ethics of belief is problematic. For, normally “ought” statements only apply when it is possible to deviate from what such a statement dictates.⁶ If believing something for which I lack evidence is incoherent, then there is no question about what I ought to believe beyond what I must believe. If it can be shown that it is conceptually impossible “to maintain a belief in open defiance of the evidence,” then evidentialism quickly follows. On such a view, evidential “norms” are not really normative; they do not offer direction on how to be a good believer. Rather, they are descriptive; they merely help elucidate what it means to believe something. According to this view, most extensively argued for by Jonathan Adler, for me to really believe p, I must take myself to have adequate evidence (or epistemic reason) for p. Adler argues for the conceptual necessity of evidentialism mainly by showing that the incoherence of a number of assertions (inspired by Moore’s paradox) can only be understood if evidentialism, as a conceptual truth, is correct. Adler’s argument is as follows: If I believe something, then I must be willing to assert what I believe, but to assert something is to claim that it is true, so to say I believe p is just another way of stating p. This can be illustrated by the incoherence of the following assertion: “It is raining, but I do not believe it is raining.” Adler would add that the following is equally incoherent: “It is raining but I lack sufficient evidence that is raining,” which (if assertion expresses belief) would also show that the following assertion is incoherent: “I believe it is raining, but I lack sufficient evidence that it is raining.” So Adler concludes, “We cannot recognize ourselves as believing p while believing that our reasons or evidence are not adequate to its truth and conversely.”⁷ And the “cannot” is conceptual, not psychological.

Adler’s “incoherence tests” do not settle the matter of what is conceptually possible. All they do is tell us that certain assertions sound very strange. Adler admits that when one leaves the realm of relatively simple beliefs, assertions of a similar kind may not seem so obviously incoherent. He considers, for example, the self-acknowledged anorexic who may believe she is overweight, despite recognizing evidence to the contrary. Adler maintains that any seeming conjunction of one believing p while maintaining that one lacks sufficient evidence for the belief will either be a case where one is not “fully aware” of having both beliefs or there will be a temporal distinction between the two conjuncts; they are not held in “a single consciousness.” He says if this were not the case, then one would not have to seek “esoteric cases, like those afforded by thoughts of the mentally disturbed” to refute evidentialism. But it seems that if one can generate counter examples (no matter how esoteric) then what he is deeming “full belief” is only one species of the genus “belief.” Adler says, “If there is no compelling connection between the concepts of belief, truth and evidence, then counter examples . . . should be plentiful. The need to search beyond the simple, blunt cases concedes the connection even as it tries to refute it.”⁸ I am not denying that there are such connections; the question at issue is the nature of the connection specifically, is it one that is unbreakable?

What then accounts for the incoherence of these Moorean assertions if evidentialism is not true? I think Adler’s diagnosis is largely correct, because beliefs often do behave the way Adler says they do. If I sincerely assert that I believe p, then it seems I am committed to acknowledging that I have some evidence for this belief. To say, “I believe Tom is in the bar but I have no evidence that he is,” seems incoherent because my consciously believing it would normally entail that I have evidence for it. But it does not seem that beliefs must behave this way. What if a vicious blow to the head caused me to have the belief that Tom was in the bar and I was not aware that this was the cause?

It would seem, in such a case, I can recognize myself (and even assert) that I have a belief while also seeing I have no reason for it. It is possible that I may take the fact that I believe it to itself indicate that I have some reason for it, even if I cannot access or remember such reasons. I may have forgotten the reasons I had for initially forming many beliefs, but I take the fact that I now have them as a clue that I once had a good reason for believing them. Maybe my belief about Tom would seem just like my belief that gold doesn’t decay the way other metals do. The difference with beliefs for which one is unable to access the reasons and the belief about Tom is that, in the former case, I can remember having reasons; I just cannot access their content. I may even be able to remember roughly the context in which I first formed these beliefs and I can imagine ways to recover and evaluate the reasons for it.

Even if the simple holding of a belief is some evidence in its favor, it certainly seems that Adler’s stronger claim doesn’t hold, namely that I regard my evidence as adequate for the truth of p. This belief may be irrational and false, but Adler’s theory is about belief, not only about rational belief. We are charitable to one another, and assume rationality, which is why these Moorean assertions sound so strange; one cannot be fully rational and make such assertions. Adler is emphatic that he wants to preclude discussions of rationality in the ethics of belief. He thinks that once we allow that the question of what we ought to believe be “determined by criteria external to belief, most prominently rationality,” we have gone “onto the wrong track.”⁹ His intrinsic approach, which asks what the concept of belief demands, is supposed to reveal why these more traditional, normative discussions of the ethics of beliefs are misguided.¹⁰

To show that something is not a conceptual impossibility, all one needs is a single counter-example. Is it possible for one to have a belief, think that the evidence dictates one ought not to have it, and still retain it? In developing the idea of the truth-aim, Williams says, “If a man recognises that what he has been believing is false, he thereby abandons the belief he has.”¹¹ As we see, Adler agrees that this is so.¹² I think it is possible for a gap to exist between this recognition and abandonment. In some situations, one may find oneself compelled to belief against one’s better judgment in a sense analogous to compelled action when one acts against one’s best judgment. An action is compelled only if the agent could not avoid performing it even if he were to believe that, all things considered, he ought to refrain from performing it. It seems, analogously, a belief is compelled only if the believer could not abandon it even if he were to believe that, all things considered, he ought not to believe it. We can thus define a compelled belief (or CB) as follows:

CB: A belief b is compelled for Subject S during time t if and only if:

1. S’s best judgment during t, all things considered, dictates that he abandon b and
2. S cannot abandon b during t.¹³

What must be better understood in order to assess the possibility of compelled belief is the relationship between the recognition that a belief is not evidentially supported and its abandonment. What happens after such recognition takes place? Is it plausible that at times such recognitions will be entirely ineffectual, where one cannot give up a belief even if one thinks that one ought to? What could cause a gap between the appreciation of a belief’s illegitimacy and its abandonment? In John Heil’s discussion of these issues, he is quite vague on the causes, saying only that there is some deficiency in the believer’s total psychological state that allows him 1) to fail properly to integrate his appreciation of certain facts and so 2) to continue to harbor beliefs that are at odds with his better epistemic judgment.¹⁴

Alfred Mele makes use of his discussion of akratic action to say more about the nature of this breakdown. Mele has pointed out that akratic action occurs when evaluation and motivation diverge. One’s evaluations of which desires should be acted on do not always correspond to their motivational strength. One of Mele’s examples is the following: Someone with a severe fear of flying may judge that his flying would be better than his not flying on a particular occasion, and yet be so anxious t...

Indice dei contenuti

• Cover
• Title
• Copyright
• Dedication
• Contents
• Preface
• Introduction
• PART I Doxastic Norms
• PART II Doxastic Responsibility
• Conclusion
• Bibliography
• Index