1 Introduction
Radical skepticism won’t go away. Despite the fact that it is widely dismissed, the best arguments for it are still strangely seductive. Even though it raises philosophical questions that are, by now, completely familiar and perhaps, in some ways, old hat, it remains highly intriguing. It is not a problem that worries many of us and yet we can’t agree on how best to deal with it.
Radical skepticism is extreme insofar as it involves serious doubts about large swaths of beliefs that almost everyone takes for granted. Perceptual skepticism, which questions all of our perceptual beliefs, is one kind of radical skepticism. Memory skepticism, which questions all of our memory beliefs, and a priori skepticism, which questions all of our a priori beliefs, are other kinds. These kinds and more besides feature prominently in the pages that follow.
This is a book about various kinds of radical skepticism and different ways of responding to them. Two of its initial goals are to identify the best arguments for radical skepticism and to reject responses to them that don’t pass muster. But its main task is to develop and defend an account of what, in my view, is the best response to radical skepticism—one that is inspired by the great eighteenthcentury commonsense philosopher, Thomas Reid, and that consciously relies heavily on epistemic intuitions.1 In carrying out this main task, this book offers something that is at once both familiar and new.
On the one hand, a Reidian-style response to radical skepticism is, as a matter of actual practice, one of the most common responses to such skepticism, among philosophers and non-philosophers alike. For this reason, the position I will defend fits well with what many people already do. On the other hand, this kind of response to radical skepticism has never before been developed in this overtly epistemic-intuition-based manner, nor has it ever been presented with as much detail or defended as extensively against so many objections, in a way that is friendly to both internalist and externalist perspectives.2 While it is fairly common for those reflecting on skepticism to respond to it in something like the way
1 Epistemic intuitions are intuitions about the requirements for and the presence or absence of epistemic goods, such as knowledge and rationality. Although I would say that Reid himself also relied heavily on epistemic intuitions, he did not (as far as I can tell from his writings) do so consciously or explicitly.
2 However, Noah Lemos (2004a) provides an excellent book length treatment of similar themes, unpacking and defending the views held by the pillars of the commonsense tradition: Thomas Reid, G. E. Moore, and Roderick Chisholm.
Radical Skepticism and Epistemic Intuition. Michael Bergmann, Oxford University Press (2021). © Michael Bergmann. DOI: 10.1093/oso/9780192898487.003.0001
I advocate in this book, this typically occurs without much understanding of what this sort of response involves. I suspect that, in at least some cases, this is partly due to a reluctance to subject this very natural response to critical scrutiny, perhaps over concerns that it won’t stand up well in the cold light of day. But the Reidian has no reason to be bashful or timid about her view. So I will be working to expose and make explicit what is (or should be) going on in commonsense responses to radical skepticism and then to consider and reply to a wide array of challenges that responses of this kind face. The result will be an answer to radical skepticism that is, in important respects, both unfamiliar and fresh.
In this opening chapter, I will do three things. First, I’ll explain why I think it is worthwhile to devote our time to thinking about radical skepticism. Second, I’ll provide an overview of the book’s contents. And third, I’ll clarify a few key concepts that will play an important role in the chapters that follow.
1. The Importance of Radical Skepticism
Despite its familiarity, perceptual skepticism is rarely embraced. Few people, even among professional philosophers, find it plausible or even consider it a live option. It is sometimes simply dismissed as being unworthy of serious attention, even if it might provide an interesting premise for a novel or a movie. So why spend time, as I will be inviting readers to do in this book, thinking about radical forms of skepticism such as perceptual skepticism?
Three reasons. The first is that, because it is a paradigm case of a kind of skepticism that is both tempting and yet very commonly resisted, it provides us with an instructive pattern for thinking about other kinds of skepticism that are tempting and yet possibly (we aren’t always so sure) worth resisting. All of us are skeptical about the moral, political, philosophical, and religious views of at least some people, while holding positions of our own on these same topics—positions about which others are skeptical. How should we think about our skepticism of the views of others on such topics? How should we respond to their skepticism regarding our own views on these matters? These are questions of broad interest that matter a great deal to a significant percentage of the world’s population. A helpful starting point for our reflections on all such queries is the example provided by a careful and well-considered reaction to radical skepticism. Radical skepticism is widely recognized—even by those whose views on more controversial matters differ markedly from each other—to be, on the one hand, seductive and yet, on the other, well worth resisting. Because of this shared attitude to radical skepticism, discussions of it enable us jointly to make substantial progress on getting clearer about good and bad ways of defending and resisting radical skepticism; and this, in turn, provides us with helpful models of good and bad ways of
defending and resisting skepticism about more contentious views (e.g. in morality, politics, philosophy, and religion). Even if we conclude that our thinking about skepticism on these controversial matters should differ from our thinking about radical skepticism, that in itself (if plausibly defended and explained) can yield illuminating insights into how we should think about moral, political, philosophical, and religious skepticism.
A second reason it is worthwhile to think about radical skeptical challenges and how best to respond to them is that so much of contemporary epistemology is, to one degree or another, influenced and shaped by perspectives taken on radical skepticism. Even though epistemologists are quick to resist radical skepticism, they differ in their views about the best way to do so.3 And these differences have myriad implications for other positions within epistemology and, to some extent, in other areas of philosophy as well. Perspectives on radical skepticism have played a significant role in the shaping and development of views on knowledge, justification, contextualism, closure principles, “relevant alternatives” theories, disjunctivism, contrastivism, the internalism–externalism debate, and the foundationalist–coherentist controversy. Views on radical skepticism have also influenced attitudes among epistemologists towards knowledge-first epistemology, evidentialism, phenomenal conservatism, the epistemology of disagreement, and experimental philosophy, to name just a few other important connections within epistemology. Positions on radical skepticism have also become entwined with discussions in other areas of philosophy, such as the philosophy of perception, realism and anti-realism, and the nature of mental and propositional content. As a result, having a clearer understanding of the good and bad options for defending and responding to radical skepticism will help one arrive at what amounts to a pivotal position within one’s worldview—a position with numerous consequences for other elements of one’s overall perspective on reality.4
A third reason for focusing on radical skepticism is just the intrinsic interest of this perennial philosophical issue. The question “How do you know you are not being tricked—by illusory perceptual experience—into thinking you are perceiving a physical world?” is fascinating. On the one hand, it’s tempting to think that there must be some way we can know this—otherwise the very foundations of our ordinary perspective on the world would be completely undermined, a prospect that is very hard to take seriously. (Perhaps no one takes this seriously enough to refrain from holding any of the usual commonsense beliefs we tend to have
3 Thus, although there is more agreement that radical skepticism (vs., say, moral or religious skepticism) should be resisted, disagreement remains over how best to resist it. (This sort of disagreement is also emphasized in the next paragraph where I lay out a third reason it is worthwhile to focus on radical skepticism.) It’s important to keep in mind that this lack of unanimity about how best to respond to radical skepticism is perfectly compatible with the point made in the previous paragraph about there being significant agreement that radical skepticism should be resisted.
4 Dodd and Zardini (2014: 1–2) make some points similar to those made in this paragraph.
about the physical world around us and about other people.) On the other hand, it has proven to be exceedingly difficult to give a satisfying non-skeptical answer to that question (not just to skeptics, but even to ourselves). Each of the answers that have been given face their own challenges. The attraction of digging into the debates on this very stimulating question has been the siren call that has drawn many to philosophy in the first place, and it has inspired many ingenious ideas by mature philosophers both in recent decades and in centuries past. Coming up with a plausible response to radical skepticism is a goal worth pursuing for its own sake.
2. An Overview of the Book
So what is the best response to the challenge of radical skepticism? We can divide responses to it into three categories (not including those that embrace it). First, there are concessive responses, which concede that we don’t know that we aren’t radically deceived but insist that (contrary to radical skepticism) much of our assumed ordinary knowledge of the external world remains intact.5 Second, there are inferential anti-skeptical responses, which are non-concessive responses (i.e. ones that do not concede that we don’t know we aren’t radically deceived) that say our ordinary beliefs about the external world are (often) justified even though such justification requires that these beliefs are defensible inferentially via good arguments. Third, there are noninferential anti-skeptical responses, which are non-concessive responses that say our ordinary beliefs threatened by the challenge of radical skepticism (including our perceptual, memory, and introspective beliefs) are justifiedly held noninferentially, even if they aren’t based on or defensible via any available good arguments.6 I’ll be defending a Reid-inspired version of noninferential anti-skepticism and objecting to competing responses.
What is distinctive about the Reidian version of noninferential anti-skepticism presented in this book? Two main things. First, it explicitly and consciously relies on epistemic intuitions—which are seemings about epistemic value or goodness— in a way that is introspectively plausible and fully defensible. Most or all of the usual approaches to radical skepticism, including attempts to defend it, also rely on epistemic intuitions. The difference is that they rarely show any recognition of this reliance.7 Second, the Reidian approach of this book is ecumenical across the internalist–externalist divide in epistemology. What this means is that, in general
5 Concessive responses include contextualism, contrastivism, closure denial, and certain Wittgensteinian responses.
6 More careful accounts of these three types of response will be given in Chapter Two.
7 Further discussion of this wide but often unacknowledged reliance on epistemic intuition will be postponed until Chapters Six and Seven.
outline, the main ingredients of the anti-skeptical position proposed and defended in this book can be adopted by both internalists (including evidentialists, mentalists, and phenomenal conservatives)8 and externalists like myself. For this reason, I hope the view will have wide appeal: it is moderate and adaptable to various perspectives within epistemology.
The book is divided into three parts. In Part I, Underdetermination and Inferential Anti-Skepticism, I focus on underdetermination arguments for radical skepticism and inferential anti-skeptical responses to them. (These underdetermination arguments highlight the fact that our evidence underdetermines the truth of the beliefs based on it and conclude from this that, apart from good arguments showing that the evidence in question makes these beliefs true or at least probable—arguments we seem not to have—these beliefs are not justified.) I begin by explaining why I will be setting aside certain skeptical arguments (i.e. all those other than underdetermination arguments—including closure arguments) and certain responses to skeptical arguments (especially the concessive responses noted above and those—given by disjunctivism and knowledge-first epistemology— that deny the phenomenon of underdetermination). In each case, the reason for setting aside these arguments and responses is that they either underestimate or overestimate the appeal of radical skepticism. I then develop a series of underdetermination arguments for radical skepticism about not only perception and memory but also, more surprisingly, about a priori intuition and introspection (and even in support of global skepticism9), and I argue that the inferential antiskeptic’s responses to these skeptical arguments are unsuccessful.
This leaves us with noninferential anti-skeptical responses, which are taken up in Part II. I begin in Chapter Six by explaining the particularist tradition, starting with Thomas Reid in the eighteenth century and continuing through G. E. Moore and Roderick Chisholm in the twentieth century.10 Particularism embodies the methodology I employ in working out my own noninferential anti-skeptical response to skepticism. Then, in the remainder of Chapters Six through Eight (the core chapters of the book), I do two things. First, I lay out my favored version of this particularist method of dealing with radical skepticism—a version I call ‘intuitionist particularism’ because it is spelled out in terms of epistemic intuitions (both those employed by skeptics in support of their skeptical
8 I am here going along with the standard categorizations of these views and ignoring disputes over whether all versions of mentalism and evidentialism count as internalist views. For some discussion of these disputes, see Bergmann (2006: ch. 3) and Bergmann (2018).
9 Global skepticism is skepticism about everything. Local skepticisms—such as perceptual skepticism or memory skepticism—are narrower in scope.
10 In brief, particularists in epistemology (not to be confused with moral particularists) use our natural assumption that most of our ordinary beliefs are rational and instances of knowledge as a starting point in evaluating skeptical arguments and epistemic principles about what is required for rationality and knowledge.
arguments and those employed by anti-skeptics in support of ordinary knowledge claims). Second, I use this intuitionist particularist methodology to develop a noninferential anti-skeptical response to the underdetermination arguments for radical skepticism from Part I. In presenting this response to radical skepticism, I highlight its advantages over other responses as well as the ways in which it can be adopted by both internalists and externalists alike. I conclude Part II by defending this response against several challenging objections.
In the last part of the book, Part III, I take up skeptical challenges to epistemic intuition—the belief source that plays such a significant role in my particularist noninferential anti-skeptical response in Part II to the underdetermination arguments for radical skepticism developed in Part I. In addition to facing some of the same challenges that are directed at my response to radical skepticism about perception and memory, my response to radical skepticism about epistemic intuition also faces challenges from disagreement and from experimental philosophy. All of these challenges to epistemic intuition will be dealt with in the final part of the book.
The overall narrative of the book can thus be summarized succinctly as follows. In Part I we learn two things. First, the tempting intuitions supporting underdetermination arguments for radical skepticism tell us that we are forced to endorse radical skepticism unless inferential anti-skepticism provides an effective means of escape. Second, although inferential anti-skepticism seems to offer a means of escape, that way turns out to be a hopeless dead end. In Part II we see that particularist epistemic intuitions rationally assure us that radical skepticism is a mistake and, therefore (in light of Part I), that (i) the seductive intuitions in support of underdetermination arguments are (surprisingly) incorrect and that (ii) the right response to those arguments is noninferential anti-skepticism. In addition, we see that various alleged problems with this noninferential anti-skeptical response turn out not to be problems after all. In a similar way, we discover in Part III that skeptical concerns about epistemic intuition as a belief source are overblown and, in the end, not a sufficient basis for any lingering worries about the conclusions of Part II.
3. Clarifications of Some Key Concepts
One final preliminary task to be handled in this chapter is that of clarifying three crucial concepts that I’ll be employing frequently throughout this book—the concepts of skepticism, justification, and evidence.11
11 Because the concept of epistemic intuition, briefly introduced earlier in this chapter, features heavily in Parts II and III of the book, but is not emphasized in Part I, it will not be unpacked further until Chapters Six and Seven.
3.1 Skepticism
The topic of this book is radical skepticism of various kinds: perceptual skepticism, memory skepticism, global skepticism, and others as well.12 But what is skepticism? Skepticism about the existence of the physical world is the view that we can’t or don’t know (or hold justified beliefs about) whether there exists a physical world. Skepticism is different from the view that denies the existence of a physical world. Instead of saying it isn’t there, skeptics say we don’t know whether it’s there. Similarly, atheists say there is no God and moral error theorists say there are no moral facts, whereas religious and moral skeptics say we don’t know whether God exists or whether there are moral facts.13
Knowledge-skepticism can be distinguished from justification-skepticism. Knowledge-skepticism about the external world says we don’t or can’t know whether there is an external world; justification-skepticism about it says we don’t or can’t justifiedly believe that there is (or that there isn’t) an external world. If, as most epistemologists think, knowledge entails truth and justification doesn’t (i.e. we can justifiedly believe what is false though we can’t know what is false), then it’s easier to have a justified belief that p than it is to know that p, in which case justification-skepticism is a bolder position than knowledge-skepticism. Or at least this is so if knowledge requires justified belief (another thing that most epistemologists think). For then justification-skepticism entails knowledgeskepticism: if you aren’t justified in believing that p, you don’t know it either. The focus of this book will mainly be on justification-skepticism, mostly because it
12 It’s true that global skepticism includes local skepticisms such as perceptual skepticism or memory skepticism. The point here is just that global skepticism as well as these more local skepticisms are kinds of radical skepticism, and that I’ll be discussing each of these kinds (and others) in this book. For a helpful and concise overview of the varieties of local skepticism, categorized conveniently and neatly along four dimensions, see Hudson (2014: 113–4). The four dimensions of local skepticism that he highlights are:
• propositions: singling out—by subject matter or (sole) route to belief—a subclass of propositions subject to skeptical challenge,
• properties: focusing on one or another epistemic property that is under threat, such as rationality or knowledge,
• subjects: identifying a subgroup of beings confronted with skepticism, such as humans or any actual or possible reasoning beings who perceive spatiotemporally, and
• modality: saying whether the epistemic property in question merely hasn’t been exemplified or, more strongly, that it couldn’t—in some sense or other of ‘couldn’t’—be exemplified.
13 One might think that strong reasons for thinking there are no physical objects or moral facts are strong reasons for skepticism about physical objects or morality. After all, given that knowledge entails truth, strong reasons for the falsity of p count as strong reasons for thinking we don’t know p. (Sinnott-Armstrong (2006: 11) endorses this way of thinking about moral skepticism.) But this line of reasoning is plausible only if not knowing that p counts as skepticism about p. And that is doubtful. All of us agree that we don’t know simple mathematical falsehoods such as 1+1=3. Does that make us skeptics about these simple and obvious mathematical falsehoods? That doesn’t seem right. What’s more plausible is that we are skeptics if we don’t know whether p.
will simplify our discussion (since lack of knowledge might be due not to lack of justification but to other problems instead).
Is skepticism about p the view that we can’t justifiedly believe that p or that p is false? Or is it the view that we don’t justifiedly believe p (or that p is false), even when we do believe p? Some skeptical arguments aim to establish that we can’t justifiedly believe certain propositions (which, of course, implies that we don’t justifiedly believe them). But other skeptical arguments aim to establish only that we don’t justifiedly believe these propositions; they refrain from saying that we are incapable of doing so, or that doing so is literally impossible. Because it is worth including both kinds of justification-skepticism in our discussion, I will be thinking of skepticism about p as the view that either we can’t justifiedly believe that p (or that p is false) or we don’t justifiedly believe that p (or that p is false), even if we can.
3.2 Justification
Justification-skepticism denies that our beliefs are justified. But what is it for a belief to be justified? There is a huge literature on this topic and much controversy. As much as I can, I want to steer clear of the details of those controversies. My hope is that the discussion of justification-skepticism in this book can be as non-partisan as possible in what it says about justification. It might seem overly optimistic or naïve to think that we can talk about justification and justificationskepticism in a book like this while ignoring the major disputes among contemporary epistemologists concerning the nature of justification. After all, some philosophers think that opponents in such disputes are not even talking about the same thing when they talk about justification. I think that matters are not quite so hopeless. I grant that it will be helpful for me to say more about these disputes, and I will do some of that in the remainder of this subsection (and elsewhere in the book). But I will try to keep it to a minimum.
Perhaps the main divide separating theories of justification is the divide between internalism and externalism. Internalists think a belief is justified only if a person is (or can, on reflection, easily become) aware of what the belief has going for it; externalists deny that such awareness (or easy access to it) is required, thinking that a belief is justified so long as it is, in fact, appropriately formed. I agree that there is a genuine and important disagreement between internalists and externalists over the nature of justification, and my sympathies lie with externalists.14 Nevertheless, internalists and externalists can talk together fruitfully about justification (and justification-skepticism) while bracketing
14 See the first two chapters of Bergmann (2006).
some of the things about which they disagree.15 In what follows, I will explain how I’m thinking of justification and I will aim to do so in a way that is amenable to both internalists and externalists.
The first thing to say is that the sort of justification under discussion in this book is doxastic justification, not propositional justification. Doxastic justification is a property that certain beliefs (understood as token mental states)16 have, and it is importantly different from propositional justification, which is a property propositions have relative to a person. In order for a particular belief to be doxastically justified, it must be formed or held in the right way. But whether a belief is formed or held in the right way is irrelevant to propositional justification. A proposition truthfully saying which books are on my desk right now is propositionally justified for me but not for you, because I have good evidence for believing it but you don’t. In fact, that proposition can be propositionally justified for me even if I don’t believe it (perhaps because I’m too distracted to form a belief on that topic). But only an actually held belief can be doxastically justified. Moreover, it may be that, although I have good evidence for a proposition, I believe it for some silly reason. In that case, my belief’s content would still be propositionally justified for me, because of the evidence I have for it, even though that belief isn’t doxastically justified, because it isn’t formed or held in the right way—it is based on a silly reason instead of on the good evidence that I have.
When I say a belief is (doxastically) justified, I’m saying, roughly, that it is reasonable or rational; I will try to steer clear of making fine distinctions between these types of positive epistemic appraisal. Thus, when I speak of justificationskepticism with respect to a certain class of beliefs, I have in mind the view that those beliefs are not justified or reasonable or rational. I am also thinking of justification as something that is compatible with falsity, in the sense that false beliefs can be justified. Justification is, therefore, distinct from and insufficient for knowledge (which entails truth). Moreover, as I’m thinking of it, justification is required for knowledge and—as was noted in Gettier (1963)—justified true belief is insufficient for knowledge. Internalists and externalists can easily agree on all of the points about justification mentioned in this paragraph and the previous one.
A further feature of justification as I’ll be thinking of it in this book is that a belief is (doxastically) justified when it is a fitting response to the input leading to that belief. Even on this point, internalists and externalists can agree, although they will disagree about what fittingness depends on and perhaps also about what counts as a belief input. Consider first the different perspectives that epistemologists of various stripes have on belief inputs. Evidentialists speak here of evidence,
15 In fact, they can talk together fruitfully about justification even if they explicitly focus on these points of disagreement, which is what often happens when they are arguing against each other’s views on the nature of justification.
16 My focus throughout the book will be on ordinary belief, not on credences or degreed belief.
saying that justification requires that the belief is a fitting response to one’s evidence. If the evidentialist is a mentalist (thinking that justification is determined by one’s mental states), she will say that justification requires that the belief is a fitting response to one’s mental states. If the evidentialist is an internalist (thinking that justification is determined by one’s accessible mental states), she will say that justification requires that the belief is a fitting response to one’s accessible mental states.17 Externalists will think that what matters for justification is that the belief is a fitting response to some circumstance or trigger, but that this circumstance or trigger needn’t be a mental state, accessible or otherwise (although it could be). It might be some other event in the body of the believer, or perhaps even a triggering circumstance in the believer’s environment. As for fittingness, externalists would say that a belief is a fitting response to belief input if it is a proper functioning response to it or a reliable (truth-promoting) response to it, in which case this relation of fittingness holds contingently between a belief response and belief input, depending on the relevant facts about reliability or proper function.18 Internalists, on the other hand, would tend to say that the fittingness relation is an epistemic relation (possibly a sui generis one) that holds of necessity between a belief response and one’s evidence, in virtue of some of the intrinsic features of these relata.19
Moreover, if internalists and externalists can agree that a belief response is justified when it is a fitting response to belief input, then they can also agree on the following point, which also applies to justification as I will be thinking of it in this book: being subjectively epistemically blameless is not sufficient for justification. This is because it’s possible for a person innocently and helplessly to believe something that isn’t a fitting response to her belief input. Perhaps—due to no fault of her own but rather to some brain malfunction or to the intervention of some malevolent powerful being with control over her mind—a person is prevented from believing what fits her belief input, no matter how hard she tries to believe in an epistemically respectable manner. In such a case, she would (plausibly) be epistemically blameless in holding this belief even though the belief would remain unjustified because it does not fit her belief input.
3.3 Evidence
The last concept that I want to clarify is one already used numerous times in this chapter. It is perhaps more in need of clarification than the concepts of skepticism and justification given that there are a few significantly different understandings
17 For more on internalism and mentalism, see the first three chapters of Bergmann (2006).
18 See the fifth chapter of Bergmann (2006) for a comparison of proper functionalist and evidentialist accounts of fittingness. For a reliabilist account of fittingness, see Alston’s (1985: 105–11) discussion of Jeg
19 For more discussion, see Conee and Feldman (2004: 101–2) and Bergmann (2006: 50–61, 111–13).
of evidence employed in the contemporary epistemological literature and yet these differences often go unacknowledged or unrecognized. The most distinctive (though, of course, not the only) feature of evidence as I’m thinking of it is that it is what our paradigmatic justified beliefs are based on. These justified beliefs are based on things such as other beliefs, memory impressions, and perceptual experiences.20 Thus, while evidence (so understood) can consist of beliefs, it needn’t; it can consist of other nondoxastic mental states as well.
Unfortunately, as I’ve already indicated, the term ‘evidence’ is ambiguous and there are other related uses of the term at work in epistemology. For example, ‘evidence’ can refer to an object or event, such as a bloody knife or expert testimony presented at trial in a court of law. The term ‘evidence’ is also very commonly used to refer to propositions that confirm or best explain hypotheses for which they are evidence (Williamson 2000: 194–7). Although those are perfectly acceptable ways to use the term, the sense of ‘evidence’ that I am employing in this chapter—where it refers to the grounds (or basis) of our beliefs, especially our justified beliefs—is also a standard one.21 Importantly, evidence thought of as propositions and evidence thought of as objects or events presented at trial in a court of law are not what our beliefs are based on. This is because part of what it is for a belief to be based on something is for the belief to be responsive, in the relevant sense, to that thing.22 And our beliefs don’t seem to be responsive to propositions, especially if propositions are necessarily existing abstract objects. Nor do they seem to be responsive, in the relevant sense, to objects or events presented in a court of law.23 So these other kinds of evidence are not the sort of thing I’m
20 This isn’t to say that unjustified beliefs can’t also be based on evidence. After all, not all evidence is good evidence.
21 It’s what evidentialists such as Richard Feldman and Earl Conee seem to have in mind when saying that justification is determined by one’s evidence. See Conee and Feldman (2004: 2–4, 92–3, 104–5) where they make it clear that they think of evidence as including experiences and feelings, and as being that on which justified beliefs are based.
22 It’s no easy task to give an illuminating full analysis of the basing relation and I won’t try to do so here (see Korcz (1997), Evans (2013), and McCain (2012b) for some discussion). But as a start, we can say that if S’s belief B is based on E, then E is a mental state of S’s that causally contributes to the formation of B in virtue of S (or S’s belief-forming mechanisms) treating E as evidentially supportive of B. Note that S’s belief can be based on E (in this way) even if S has no thoughts at all about E, not even thoughts about E being evidentially supportive of B (and even if “treating E as evidentially supportive” is not something S intentionally does).
23 To see this point, notice that if in place of perceiving the object presented in court, the believer experienced a perfectly vivid hallucination as of the object in question (a hallucination that was identical phenomenologically to a non-hallucinating believer’s normal perceptual experience of the object), then the believer would, presumably, still hold the belief, even in the absence of the object. Likewise, if, while hallucinating a scene that appeared not to include the object in question, the object were placed in the person’s vicinity (but the person’s perceptual faculties were prevented from working normally during the hallucination, thereby keeping the person from having a sensory experience of the object), then the person would not hold the belief that the object was present, despite the fact that light waves from the object would causally affect the person’s retinas. These considerations suggest that the belief is responsive, in the relevant sense, not to the object but to the experience as of such an object (whether that experience is a genuine perception or a vivid hallucination). For more on this point, see Section 1.3 in Chapter Two.
speaking of when I speak of evidence in this book. That said, these other kinds of evidence can be understood in terms of the kind of evidence I have in mind. For example, evidence in the sense of propositions that confirm or best explain a believed hypothesis consists of propositions that either (i) are the content of beliefs that are evidence in my sense or (ii) describe things, beliefs about which or experiences of which can be evidence in my sense. And evidence in the sense of objects or events that can be presented in a court of law consists of things such that beliefs about them or experiences of them can be evidence in the sense I have in mind in this book.
So skepticism, in this book, is justification-skepticism. And the points on which internalists and externalists can agree—the points highlighted in the last four paragraphs of Section 3.2—capture something of the way I’m thinking of justification. Moreover, evidence is the sort of thing on which justified beliefs are typically based (and it doesn’t consist of propositions or ordinary physical objects or events). With these clarifications in mind, we can now turn our attention to arguments for perceptual skepticism and responses to them.
