Epistemic Entitlement

The Right to Believe

H. Matthiessen

• Publisher: Palgrave Macmillan
• Year: 2014
• ISBN:
  9781137414984
• Topic: Philosophy
• Subtopic: Epistemology in Philosophy

What entitles you to claims about your perceivable environment? Matthiessen suggests that it is neither your experience, nor the reliability of your cognitive processes, but rather your being in the right kind of perceptual situation.

1

The Right to Believe

The starting point of this study is the idea that positive epistemic status simply depends on a subject being an epistemic situation of the right sort. If my girlfriend takes a look into the cupboard she is thereby in a position to claim that there are plenty of onions in it, and this position does in no way depend on her ability or willingness to offer a justification for her belief. Two traits of the epistemic status that feature in our little story are worth mentioning: on the one hand, the epistemic status concept that we are concerned with seems to be one that applies to subjects rather than to mental states, like beliefs. It is my girlfriend, and not her mental state, that is primarily affected by the look into the cupboard. Traditionally, however, it has been more common to talk of positive epistemic properties as properties of beliefs instead of subjects. In the following section, I argue that it makes more sense, in general, to regard subjects as the primary bearers of epistemic status.

On the other hand, because the positive epistemic status that someone has confers an advantage to that person – the advantage of being allowed to hold a belief without being required to justify it – it seems appropriate to speak of an epistemic right, or entitlement. The notion of epistemic entitlement has played a role in contemporary epistemology over the past few years, and I will discuss various recent conceptions of epistemic entitlement in the remainder of this chapter. In doing so, I try to distil the core of the contemporary notion of epistemic entitlement. Although – as we shall see – philosophers employ entitlement for quite different aims; nevertheless these conceptions share something like a common core, which I will use as a starting point for my own view. I will limit my discussion for the most part to perceptual entitlement. There are two reasons for this: on the one hand, the recent debate on epistemic entitlement has focused mainly on perception. If the conception of entitlement that I want to develop is meant to build on this debate, it makes sense, therefore, to begin with the same sort of restriction. On the other hand, perceptual knowledge seems to be the type of knowledge for which the idea of an entitling condition has the greatest initial plausibility: it is easier to describe situations under which we ascribe epistemic rights to people who hold a perception-based belief than, for example, to figure out which conditions need to be in place in order to be entitled to a complex philosophical belief. I will discuss the prospect of applying this social externalist account to other areas of our epistemic practice briefly in the concluding section.

1.1 Personal and doxastic epistemic status

Epistemic entitlement is a sort of epistemic right. Therefore, the epistemic status concept entitlement describes a status of subjects, rather than of mental states. However, since the main concern of recent epistemology has been with knowledge, and knowledge has been – with few notable exceptions¹ – understood as a special case of true belief, epistemologists have often talked about epistemic status as a property of beliefs rather than of persons.

But is there really any issue here? Is there any interesting difference between the epistemic properties of beliefs and of persons? The easy answer would be that talk about doxastic versus personal epistemic properties really refers to the same thing. That is, a subject has an epistemic right if her belief qualifies for a certain epistemic status, and a belief has a positive epistemic property if the believer has an epistemic right to it. Indeed, many philosophers handle the distinction in ways that suggest that they are sympathetic to such reasoning. However, I will try to show that this way of thinking about the matter is mistaken, and make the claim that epistemic status should primarily be understood as attaching to persons.

An illuminating discussion of the distinction between epistemic properties of persons and epistemic properties of beliefs can be found in Mylan Engel’s paper ‘Personal and Doxastic Justification in Epistemology’ (1992). He introduces the concept of personal justification as justification that attaches to persons (‘S is epistemically justified in believing that p’, p. 133) and doxastic justification as a property of beliefs (‘S’s belief that p is epistemically justified’, ibid.). Engel observes that many epistemologists tacitly (and some explicitly)² embrace what he calls the ‘equivalency thesis’ according to which ‘S is epistemically justified in believing that p iff S’s belief that p is epistemically justified’. These epistemologists speak of justified persons and justified beliefs interchangeably without being aware of any significant difference between the two.

Engel observes, however, that epistemological internalists typically use the term justification in the personal sense, whereas epistemological externalists tend to use justification to refer to a property of beliefs. This is quite plausible if one understands that ‘personal justification is a normative notion in terms of which persons are evaluated from the epistemic point of view’ and ‘doxastic justification is a normative notion in terms of which beliefs are evaluated from the epistemic point of view’ (Engel 1992, p. 136). Since the epistemic evaluation of persons can plausibly be understood as an evaluation in terms of epistemic responsibility or rationality (classical internalist notions) and the epistemic evaluation of beliefs in terms of (objective) probability (a classical externalist idea), Engel’s idea that an interest in personal justification is characteristic of internalism, whereas doxastic justification is more characteristic of externalism, has some initial plausibility.

Michael Williams distinguishes between two ways in which we can talk about justification that resembles Engel’s treatment. When we ask whether ‘a belief has been responsibly formed or is responsibly held’ or ‘whether, in forming a certain belief, I have negligently ignored important counter-evidence’, we are talking about ‘“epistemic responsibility” or “personal justification”’ (Williams 2001, p. 22). By contrast, if we are interested in the justificatory status of the belief, what we are asking for is an account of ‘“adequate grounding” (“grounding” for short)’ (ibid.). Like Engel, Williams thinks that the doxastic epistemic status – grounding – is present only if ‘my epistemic procedure, in the circumstances in which it was executed, was in fact reliable’ (ibid.).

Thus Engel and Williams agree that doxastic epistemic statuses are concerned with the probability of a belief being true, and personal epistemic status is concerned with epistemically responsible behaviour on the part of the subject. However, they have two quite different ideas about the relation between these two kinds of epistemic status. Engel thinks that justification basically comes in two varieties, doxastic and personal, and suggests that being aware of this fact can help us to avoid philosophical conundrums. For example, it resolves the dispute between externalists and internalists about the nature of justification. Classical counter-examples to externalist accounts of justification may lose their force once we distinguish between doxastic and personal justification. As an illustration, Engel cites Bonjour’s (1980) story about Norman, the reliable but doxastically irresponsible clairvoyant, who comes to believe – as a result of his reliable clairvoyant capacity – that the president of the United States is in New York City. If he has no evidence whatsoever for the president being in New York City and is not aware that he is clairvoyant, Norman is not justified in this belief, although the belief itself is justified (cf. Engel 1992, p. 146). Once the distinction between personal and doxastic justification is in place, this example no longer functions as a counter-example to externalist theories of justification tout court. All it shows is that externalist (doxastic) justification is not sufficient for knowledge. Consequently, Engel suggests that both varieties of justification are necessary for knowledge (ibid., p. 148).

Williams agrees with Engel that knowledge requires both kinds of justification (cf. Williams 2001, p. 23), but he sees an important asymmetry between them. He argues that personal justification is the basic variety of justification and offers two considerations in favour of this claim.

Sometimes a belief is based on evidence that appears strong but is in fact misleading or defective. Is such a belief justified? Williams suggests that there are two different kinds of reactions to this question, each of which is plausible. First, ‘We might say: “In a way, yes, but in another way, no”’ (ibid., p. 22) In this case, we treat doxastic and personal justification as two independent epistemic statuses, one of which is present and the other of which is absent. But we might as well say (and this, according to Williams, is the more plausible response), ‘My belief was certainly justified at the time, though the evidence I was relying on turned out to be not as strong as it appeared’ (ibid.). By responding in this way, we give preference to the personal aspect of justification, since we imply that the belief was justified as long as we didn’t know about the unfavourable circumstances – the belief in question was justified but it was not adequately grounded. Justification is understood here as something that depends on the epistemic subject’s point of view and concerns the agent’s rationality.

Williams also has a second argument for personal justification as the more basic of the two notions. Personal justification implies epistemic responsibility, which is a form of rationality. Adequate grounding, on the other hand, is primarily connected to truth (‘makes our beliefs likely to be true’, ibid., p. 24). But truth and rationality do not exist independently of epistemic goals, according to Williams. Quite the contrary, rationality is the more fundamental notion, since our interest in truth is intelligible only in the light of our search for epistemic rationality. ‘Why are we so interested in truth? Because it is irrational – epistemically irresponsible – not to be.’ I am not sure whether this argument is convincing. Couldn’t one say with equal plausibility that truth is the most basic epistemic goal and epistemic rationality is basically a means to that end?³

Williams contends that a typical mistake in traditional epistemology is to conflate epistemic rationality or responsibility with adequate grounding. Most epistemologists have thought that epistemic rationality consists in possessing grounds that make it likely that my belief is true. (This is what Williams calls the ‘Prior Grounding Requirement’.) Once we give up this idea and consider personal justification as an epistemic status that does not necessarily depend on the subject’s cognitive access to her belief’s grounding, we can see the relation between personal justification and adequate grounding in a new light. We are then in a position to recognize that the two may fall apart, although in most cases they go together.

Although Williams and Engel hold different views on the relation between personal and doxastic epistemic status, they share the assumption that personal epistemic status is a question of agentive responsibility, whereas doxastic epistemic status is a matter of reliability. What speaks in favour of this view? It seems uncontroversial that epistemically responsible behaviour is connected to personal epistemic status. But why suppose that non-accidentality should be primarily connected to properties of beliefs? Some personal statuses depend on what I have done. My right to drive a car depends on my having learned the traffic rules and passed an examination. But my status as a citizen of France just depends on my being born there. There is, therefore, nothing astonishing in personal statuses that depend on factors that do not underlie the bearers’ control.

Williams recognizes both types of epistemic assessment as part of our epistemic practice. He suggests that we ‘sometimes … focus on the person’s entitlement to hold a certain view. But sometimes we are interested in whether the grounds on the basis of which he holds it are objectively adequate, whether they establish its truth, irrespective of whether he would be culpable for any defects’ (2001, p. 22). The idea is, thus, that being adequately grounded contributes to doxastic epistemic status because we can be interested in it in cases where we are not interested in the epistemic subject’s behaviour. There is no additional argument why we should attribute an epistemic property to the belief simply because epistemic responsibility is not under discussion. Engel’s view is similar: he notes that ‘objective probabilities [of a belief’s truth] provide us with … a means of evaluating beliefs’ (1992, p. 137). He says that since objective probability is desirable from an epistemic point of view, positive epistemic status is attributed to a belief – but he provides no argument for this attribution. Why can’t we say that holding a belief that is not simply true by accident confers an epistemic advantage to the believer? Maybe the idea is that the source of the epistemic status – the high objective probability of a belief being true – is not connected in the right way to the person who holds the belief. As long as considerations of the epistemic agent’s responsibility are not in view it doesn’t matter who holds the belief. But there is reason to doubt that attributions of a normative status to a subject require some kind of activity or responsibility on the part of the subject. And Williams is well aware of this fact. Sometimes people just happen to have a certain normative standing, such as a legal right or a citizenship. So there seems to be nothing wrong, in principle, with attributing a normative status to a person where this does not depend upon that person’s active initiative. Apart from there being no obvious reason to attach epistemic properties to beliefs instead of persons just because we focus on the objective probability of the beliefs being true, there may also be independent reasons not to attach epistemic properties to beliefs. In the remainder of this section, I will be concerned with two kinds of considerations, namely with matters of ordinary language on the one hand, and with a problem that arises from defining the concept of knowledge in terms of a doxastic notion of justification, on the other.

Note, to begin with, that we do not ordinarily refer to a belief as justified in everyday discourse. We do not say things like ‘Her belief was justified, hence she knew it’ or ‘You shouldn’t believe that. The belief is not justified.’ Rather, talk of justification has its natural home in reference to actions. Our response to someone who doesn’t understand why we did something or (perhaps more typically) thinks that we did something wrong (or silly) is to offer a justification of our action. I justify an action by making it intelligible, by explaining my reasons for doing it, and this may be required of me, especially in cases where it looks like I did something wrong (or imprudent, dangerous or silly). When the agent is able to provide such a justification, we are sometimes warranted in saying that the action itself is justified.

Justificatory vocabulary can then also be applied to beliefs. When someone makes claims that sound false or silly (such that we may ask: ‘Why do you think so?’), or when there are doubts as to whether a person is in a position to know what she claims to know (such that we may ask: ‘How do you know?’), then the subject is often required to make her belief intelligible by explaining why she believes it, or how she knows it. Such an explanation may perhaps be described as ‘justifying why one holds a belief’ or even as ‘justifying one’s belief’, although such formulations sound much more technical and artificial here than they do in the case of actions.

Here, as in the case of practical justification, we may say that a belief for which a justification is – or, according to some, can be – given is justified. Note that, although the cases of action and belief have something in common, talk of justified belief is extremely rare in ordinary discourse.⁴ Although we may naturally say something like ‘Her reaction to the offer was justified’, we hardly ever say things like ‘His belief that they stole the money was justified.’

Suppose for a moment that establishing talk of justified belief was just a harmless stretching of ordinary language. Even so there would remain a huge step to get to a notion of doxastic justification according to which a belief is justified iff it has a high objective probability of being true. For many epistemologists nowadays this idea sounds like plain common sense. Being justified – in the sense that the believer is able to make the belief intelligibl...

Table of contents

• Cover
• Title
• Copyright
• Contents
• Introduction
• Chapter: 1 The Right to Believe
• Chapter: 2 The Social Character of Entitlements
• Chapter: 3 A Default and Challenge Model of Perceptual Entitlement
• Chapter: 4 Perceptual Knowledge
• Chapter: 5 Perceptual Knowledge and the First-Person Perspective
• Chapter: 6 Concluding Remarks
• Notes and References
• Bibliography
• Index