Epistemology Factualized: New Contractarian Foundations for Epistemology
Epistemology Factualized: New Contractarian Foundations for Epistemology
Many of my beliefs cannot now be reasonably challenged. For instance, I believe that I am a human being, that I have two hands, and that I am now on or near the surface of the Earth. It is unreasonable to challenge me on any of these points. And unless something strange or unexpected were to happen, it would continue to be unreasonable to challenge me on any of these points. Following G.E. Moore, I'll use the phrase 'common sense' to refer to the totality of such ordinarily unchallengeable beliefs.[1] In this paper, I will not be interested in the question whether the beliefs of common sense are true: I will simply assume that they are true. Nor will I be interested in the question whether the beliefs of common sense enjoy some epistemologically favored status, i.e., whether we know them to be true, or are justified in believing them, or are rational to believe them, or are warranted in believing them, and so on. Again, I will simply assume all of this. That is, I will assume that the beliefs of common sense are justified, rational, warranted, and so on. What I will be examining in this paper is the possibility of giving a substantive account of what confers these positive epistemological properties upon the beliefs of common sense.
The attempt to give such a substantive account is widely, but not universally, regarded as one of the primary tasks of epistemology. To understand why it is widely regarded in this way, let’s begin by thinking about the various ways in which epistemologists have been provoked to respond to skeptical arguments. Skeptical arguments typically purport to show that the beliefs of common sense do not enjoy (one or another of) the positive epistemological properties I have listed above. They purport to show that the beliefs of common sense are not justified, or not rational, or not warranted, or what have you. Of course, different forms of skepticism target different regions of such‘common sense’: there is skepticism about the external world, about the past, about meaning, about other minds, about values, about unobservables, and so on. What all such forms of skepticism have in common is that they begin by targeting the epistemological credentials of a particular belief that seems not to suffer from any epistemological shortcomings peculiar to it. The targeted belief is taken to be representative of all of the beliefs within a region of common sense in this way: if it suffers from epistemological shortcomings at all, then so too do all beliefs within that region of common sense.[2]
Epistemologists who respond to arguments for skepticism do so in at least one of the following four (non-exclusive) ways:
(1) We might concede that the skeptic successfully shows that common sense does suffer from one or another epistemological shortcoming. But then we might go on to say that this shortcoming does not impugn our ordinaryclaims or practices, which are more modest in their epistemological purport than the skeptic takes them to be.The purest version of this strategy is to be found in contextualist theories of the semantics of knowledge attributions.[3] But, even apart from any particular theory, it is quite commonly thought that the skeptic “raises the standards” for knowledge, or for justification, or for warrant, or what have you.
(2) We might attempt to show that the skeptic's argument is somehow self-defeating: i.e., one is epistemically entitled to accept the premises only if the conclusion is not true or not justifiable.[4] We may think of this as the “Kantian” strategy for responding to skeptical arguments, since it is commonly pursued by means of some transcendental argument. One’s success in pursuing this strategy does not depend upon one’s identifying any particular premise of the skeptical argument as false, or any particular inference in the skeptical argument as invalid.
(3) We might attempt to identify a false premise, or a bad inference, in the skeptic's argument.[5] This is perhaps the most common strategy for responding to skeptical arguments. Of course, one’s success in pursuing this strategy does not depend upon one’s finding the skepticalargument to be self-defeating. The skeptical argument might be unsound without being self-defeating.
(4) Finally, we might attempt to explain what it is that makes the skeptic's conclusion false. That is, we might attempt to spell out what it is that confers positive epistemological status upon common sense.[6] We can do this even without finding anything in particular wrong with the skeptical argument at issue. In pursuing this strategy, we are not looking for problems in the skeptical argument. Rather, we are assuming that the conclusion of the skeptical argumentis false, and we are trying to explain what it is that makes it false. Given that common sense does enjoy some favored epistemological status, what makes this the case?
Many epistemologists today pursue some version of strategies (1), (2), and/or (3). But while strategies (1) – (3) might all be fruitfully pursued, none of them can give us everything that we want from a response to skeptical arguments. In order to see why this is so, let’s suppose that a skeptical argument were directed against a child’s belief in a fairy tale. Pursuing strategies (1) – (3), we might show that this skeptical argument was, in one or another way, a bad argument (unsound, self-defeating, or directed against an unrealistically immodest opponent). But still, theconclusion of this argument would be – no thanks to the argument – true. It’s a bad argument for a true conclusion. For the child’s belief in the fairy tale does not enjoy any positive epistemological status. The child’s belief is not warranted, justified, rational, highly confirmed, or what have you. So the skeptical conclusion concerning the child’s belief in the fairy tale is true, but the analogous skeptical conclusion concerning common sense is false. And so we might ask: what is it that makes the difference between the two cases? Of course, we might answer: common sense enjoys high positive epistemological status, whereas the child’s belief in the fairy tale does not. But what does this difference amount to? What makes it the case that the child’s belief in the fairy tale does not enjoy any favor epistemological status, but our common sense beliefs do enjoy such status? We want not only to see what's wrong with the skeptic's argument against common sense, but also to understand what makes the conclusion of that argument false. How does common sense manage to achieve the positive epistemological status that the skeptic argues (however badly) it can't possibly achieve?[7] What makes common sense epistemologically better off than the child’s belief in the fairy tale? We would like answers to these questions.
Of course, to answer these questions is not to argue that common sense manages to achieve some positive epistemological status: we can explain how it manages to achieve that status without arguing that it does so. (Compare: evolutionary theory might enable us to explain how humans came into being. But it does not enable us to argue that humans came into being. If someone wants an argument to show that humans did in fact come into being, it will not help to appeal to evolutionary theory.) While we may be pessimistic about the prospects of finding a cogent, non-question-begging argument that common sense enjoys positive epistemological status, we may still reasonably hope for an explanation of how common sense manages to enjoy such status. Many epistemologists have tried to provide such an explanation, either in terms of reliability, coherence, tracking, inference to the best explanation, or what have you. These epistemologists typically take it to be one of the primary tasks of epistemology to explain what confers positive epistemological status upon the beliefs of common sense.
Taken out of context, this constructive anti-skeptical enterprise may seem very odd. Kaplan 1991 tells a story that reminds us of its seeming oddity:
Some years ago, an acquaintance complained to me about her ... philosophy course. Her instructor had devoted a substantial portion of the course to the question of what - if anything - justified his students' belief that there was a lectern at the front of the classroom. To my acquaintance, the entire exercise had seemed a sham. It was obvious to her that there was a lectern in the front of the room and it seemed, as far as she could tell, that it was equally obvious to everyone else in the classroom, the instructor included. And it seemed to her intellectually dishonest of the instructor, and those students whose interest he had managed to engage, to pretend to throw into question the propriety of a belief when, in fact, the truth of that belief was entirely evident to them.[8]
Many epistemologists have heard this kind of complaint from their students, and some philosophers have developed views that would legitimize such complaints. Consider, for instance, the view that the beliefs of common sense serve as the 'bedrock', from which our epistemological activity proceeds, and that this bedrock neither needs nor has anything to confer positive epistemological status upon it:
94. ...I did not get my picture of the world by satisfying myself of its correctness; nor do I have it because I am satisfied of its correctness. No: it is the inherited background against which I distinguish between true and false. ...
96. It might be imagined that some propositions, of the form of empirical propositions, were hardened and functioned as channels for such empirical propositions as were not hardened but fluid... .[9]
This Wittgensteinian view is not popular today, and very few contemporary epistemologists would take seriously the kind of complaint issued by Kaplan's acquaintance. But, in a Wittgensteinian spirit, Kaplan argues that this complaint exposes a fundamental unclarity in a currently widespread conception of the aims and methods of epistemology. Specifically, on Kaplan's view, questions concerning what justifies the beliefs of common sense have no true, non-stipulative answer: there is nothing about our epistemological practices that could make any such answer correct, and so there is nothing at all that could make any such answer correct. Kaplan's argument for this thesis can be equally well used, mutatis mutandis, to support a more general thesis, namely, that questions concerning what confers positive epistemological status (e.g. justifiedness, reasonableness, warrant, confirmation, or what have you) upon the beliefs of common sense have no true, non-stipulative answer. I propose to use the phrase 'Kaplan's thesis' to refer to this more general thesis, and to use the phrase 'Kaplan's argument' to refer to the argument that would support this thesis. In short, I will reply to Kaplan as if he had argued for the more general thesis.
Kaplan's argument poses a serious challenge to strategy (4) above, and so to the very project of constructive anti-skeptical epistemological theorizing. If Kaplan’s challenge cannot be met, then all contemporary attempts to execute strategy (4) are bound to fail: we can offer no explanation at all -- coherentist, reliabilist, foundationalist, or otherwise -- of what confers positive epistemological status upon the beliefs of common sense. Asking what confers positive epistemological status upon the beliefs of common sense would, in that case, be like asking “What makes today Wednesday?” The question itself is defective. Is this the conclusion that we must draw? Must we reject the very question what confers positive epistemological status upon the beliefs of common sense? Must we regard that epistemological question as defective, just as we regard the question “what makes today Wednesday?” as defective?
I will argue that we needn’t reject the epistemological question, for we can meet Kaplan’s challenge: we can find something about our ordinary epistemological practices that can serve as data for constructive anti-skeptical epistemological theory. But to do this we have to look not just at intrinsic features of those practices, but also at the relation that those practices bear to the rest of our lives. That is, we have to explain how it is that our epistemological practices matter to us. On the basis of that explanation, we can give an account of what it is that makes a norm epistemological -- i.e. what it is for a norm to be part of a practice that matters in the distinctive way that our epistemological practices matter. I will attempt to show that giving such an account enables us both to legitimize and to execute strategy (4). In other words, not only will this account enable us to rebut Kaplan’s challenge to the very project of explaining what confers positive epistemological status upon common sense, but it will also enable us to offer precisely such an explanation.
I should stress at the outset that the dispute between Kaplan and me does not concern the issue of whether the beliefs of common senseenjoy some positive epistemological status. Both parties to this dispute grant that the beliefs of common sense doenjoy positive epistemological status: they are justified, reasonable, warranted, or what have you. Thus, Kaplan and I would both disagree with those philosophers who, like C.S. Peirce and John Dewey, regard it as a mistake to think of the beliefs of common sense as enjoying any epistemological status whatsoever. According to those philosophers, a belief enjoys some epistemological status only in so far as it is defended (successfully or unsuccessfully) by appeal to the beliefs of common sense, but the beliefs of common sense themselves are not proper targets of epistemological evaluation. I will not address that view in the present paper; I simply assume that it is false. I will devote my attention exclusively to the issue that divides me and Kaplan. Where Kaplan and I disagree is with regard to the question of what confers positive epistemological status upon those beliefs. It’s not that Kaplan would answer this question in one way and I would answer it in another way. It’s rather that Kaplan would reject the question itself as defective, whereas I want to defend the traditional enterprise of answering the question. I want to show that it’s a good question.
I should also note at the outset that I will not be addressing the complicated and important issue of whether it is beliefs, degrees of confidence, or changes in view that are the proper or primary targets of epistemological appraisal. For the purposes of this paper, I will simply assume that beliefs can be epistemologically appraised, and so are among the proper targets of such appraisal. (This assumption has been challenged, but I will not attempt to rebut these challenges here.) I leave it open whether the epistemological appraisal of beliefs is derivative, i.e., whether the epistemological status of a belief somehow derives from the epistemological status of the change in view that led to it, or from the epistemological status of the believer’s degree of confidence in the truth of the belief. Some epistemologists assume that the epistemological appraisal of beliefs is derivative in one or another of these ways. Other epistemologists assume that the epistemological appraisal of degrees of confidence or of changes in view is derivative from the epistemological appraisal of beliefs. Perhaps the ugly but complicated truth is that none of these forms of epistemological appraisal is derivative from any of the others: there are simply three different targets of appraisal. For the purposes of this paper, I leave it open whether or not this is so.
In section 1, I will explain and motivate the traditional enterprise of explaining what confers positive epistemological status upon the beliefs of common sense. In section 2, I will explain Kaplan’s challenge to that enterprise. In section 3, I will begin to develop a response to Kaplan’s challenge by drawing upon, and generalizing, Edward Craig’s account of the usefulness of our practices of knowledge attribution. In section 4, I will argue, in a schematic way, that the envisaged generalization of Craig’s account has much more explanatory power than has been so far realized. Finally, in sections 5 and 6, I will show how this generalization of Craig’s account can satisfactorily respond to all of Kaplan’s objections.
1. The Chisholmian Enterprise
Kaplan's argument is directed against strategy (4). Let’s begin, then, by trying to understanding why someone would think that strategy (4) is a reasonable strategy to pursue in rebutting skeptical arguments. Specifically, let’s consider how one of the most influential practicioners of that strategy– namely, Roderick Chisholm – would respond to the student's complaint described above. According to Chisholm[10], the student's complaint is based on a misunderstanding. As ordinarily understood, questions of the form 'What justifies the belief that p?' are challenges to the belief that p. If the challenge is reasonable and the believer cannot answer it, then she is not epistemically entitled to persist in that belief. To see this point, imagine the following dialogue.
A: Jack won't be coming to the party tonight.
B: Really? He told me that he was coming. What justifies your belief that he's not coming?
A: I have no idea what, if anything, justifies my belief. Still, he's not coming.
In issuing this reply to B's challenge, A is being unreasonable. That's not just because A is speaking unreasonably. It is because, if A cannot answer B's challenge, then A is not epistemically entitled to persist in her belief that Jack won't be coming to the party tonight. In this case, then, B's question 'What justifies your belief that he's not coming?' is to be understood in such a way that A's epistemic entitlement to persist in her belief depends upon her ability to answer the question. More generally, questions about what confers positive epistemological status upon someone's belief that p are ordinarily to be understood in such a way that the belief's positive epistemological status is at stake in the believer's ability to answer the question.