Epistemic Angst offers a completely new solution to the ancient philosophical problem of radical skepticismthe challenge of explaining how it is possible to have knowledge of a world external to us.
Duncan Pritchard argues that the key to resolving this puzzle is to realize that it is composed of two logically distinct problems, each requiring its own solution. He then puts forward solutions to both problems. To that end, he offers a new reading of Wittgenstein's account of the structure of rational evaluation and demonstrates how this provides an elegant solution to one aspect of the skeptical problem. Pritchard also revisits the epistemological disjunctivist proposal that he developed in previous work and shows how it can effectively handle the other aspect of the problem. Finally, he argues that these two antiskeptical positions, while superficially in tension with each other, are not only compatible but also mutually supporting.
The result is a comprehensive and distinctive resolution to the problem of radical skepticism, one that challenges many assumptions in contemporary epistemology.
About the Author
Duncan Pritchard is professor of philosophy at the University of Edinburgh, where he is the director of Eidyn: The Edinburgh Centre for Epistemology, Mind and Normativity. His books include Epistemic Luck and Epistemological Disjunctivism.
Read an Excerpt
Radical Skepticism and the Groundlessness of our Believing
By Duncan Pritchard
PRINCETON UNIVERSITY PRESSCopyright © 2016 Princeton University Press
All rights reserved.
Radical Skepticism and Closure
1. The Contemporary Radical Skeptical Paradox
My focus will be on the broadly Cartesian understanding of the problem of radical skepticism that appears in the contemporary epistemological literature. As we will see, we can find in this contemporary discussion of radical skepticism — at least with a little digging and reworking — the essential elements of a particular kind of skeptical challenge that is of profound philosophical importance, and which is thus a genuine source of epistemic angst.
Radical skepticism of this form specifically concerns our knowledge of a world external to us, and it proceeds by attempting to undermine the possibility that we might have knowledge of this world. It pivots on the use of radical skeptical hypotheses, where these are scenarios that are subjectively indistinguishable from a paradigm case of perception, but where one is in fact massively deceived. For our purposes, we can take the so-called brain-in-a-vat (BIV) radical skeptical scenario as representative, where this concerns an agent who from her point of view reasonably supposes herself to be in paradigm perceptual conditions, but who is in fact not perceiving a world around her at all, her beliefs being instead in response to fake "perceptual" stimuli offered by supercomputers wired up to her brain (which is floating, disembodied, in a vat of nutrients).
The initial plank in the case for skepticism comes from the contention that one cannot know that one is not a BIV. Such a claim seems entirely compelling. After all, since the BIV scenario is ex hypothesi subjectively indistinguishable from normal perceptual conditions, it is hard to see how one might come to know such a thing. What kind of rational ground might one have for such a belief, given that there is no subjective basis on which one can discern that one is not in a radical skeptical scenario?
We thus have (S1 1):
(S11) One cannot know that one is not a BIV.
The idea now is to demonstrate that this claim is in tension with our conception of ourselves as perceptually knowing a great deal about the external world. We can bring this out by considering a paradigmatic case of "everyday" perceptual knowledge, the kind of perceptual knowledge such that, if one knows anything about the external world, then one knows this. In my case, for example, this might be that I am presently sitting at my desk, typing on my computer. Call this proposition "E":
(S13) One knows that E.
If I do not have the perceptual knowledge at issue in (S13), then it is hard to see how I could know anything much (at least perceptually), as my epistemic basis for this perceptual belief is rarely, if ever, bettered.
On the face of it, of course, there is no immediate tension between (S11) and (S13), in that there seems no obvious reason why it cannot be the case both that one lacks the knowledge at issue in (S11) and that one possesses the knowledge at issue in (S13). The task in hand for the skeptic is thus to motivate the following connecting claim that puts (S11) in direct conflict with (S13):
(S12) If one cannot know that one is not a BIV, then one cannot know that E.
If we grant (S11) and (S12) to the skeptic, then it will follow that we must reject (S13) and so embrace the skeptical implications of that move. We've already seen that there is a prima facie case for (S11), so the issue thus turns on the skeptical case for (S12). The problem the skeptic faces is that (S12) is not intuitive in the way that (S11) is, and since allowing it in conjunction with (S11) calls (S13) into question, which is intuitive, there is a strong prima facie case for rejecting it.
So how is the radical skeptic to motivate (S12)? The most common way of doing this in the contemporary literature is by appeal to some form of "closure" principle for knowledge. For example, the principle that knowledge is "closed" under entailments would suffice for our purposes. Here is an initial formulation:
THE CLOSURE PRINCIPLE: FIRST PASS
If S knows that p, and p entails q, then S knows that q.
So, given that being seated at one's desk entails that one is not a BIV (since this scenario concerns a disembodied brain that is floating in a vat of nutrients), it follows that if one did know the former then one would know the latter. Conversely, if one is unable to know the latter, as (S11) alleges, then one would be unable to know the former.
While this formulation of the closure principle will suffice to motivate (S12), it is not itself very convincing. There are all kinds of propositions that are entailed by propositions that, plausibly, one knows, but where it does not seem at all credible that one should know all these entailed propositions. Indeed, there are entailed propositions that one is completely unaware of, and so one cannot know because one is not even in a position to form a belief about them.
At the very least, then, we ought to restrict our attention to those propositions that one knows are entailed by what one knows, such that one is at least in a position to form beliefs about the entailed propositions and so be in the market to have knowledge of them. We thus get the following adapted formulation of the closure principle:
THE CLOSURE PRINCIPLE: SECOND PASS
If S knows that p, and S knows that p entails q, then S knows that q.
This formulation of the closure principle would equally suffice to motivate (S12), at least insofar as we make the reasonable assumption that anyone considering this argument will know that the relevant entailment holds (henceforth we will treat this assumption as granted). But even this formulation has its problems. For example, there is nothing in this principle that demands that the subject believes the entailed proposition on the basis of the relevant entailment, and yet the idea that the subject has knowledge of the entailed proposition in such cases is surely dependent on it being so based. For instance, if a subject believed the entailed proposition on a completely independent basis, then there would be no inherent reason why her knowledge of the entailing proposition and of the entailment should incline us to regard this belief as amounting to knowledge. Instead, it would depend on the epistemic credentials of this independent basis for belief.
This highlights a deeper point about why we might find these closure-style principles compelling. This is that such principles attempt to codify how one might legitimately extend one's knowledge via competent deduction from what one already knows. That this ought to be possible looks undeniable, but we haven't quite captured this thought in either of the formulations of the closure principle just offered. This is why most commentators in the epistemological literature now tend to formulate the closure principle diachronically, as opposed to synchronically, along the following lines, such that the competent deduction element of the principle is made explicit:
THE CLOSURE PRINCIPLE
If S knows that p, and S competently deduces from p that q, thereby forming a belief that q on this basis while retaining her knowledge that p, then S knows that q.
With the closure principle so formulated it is built into the principle that the subject is acquiring her belief in the entailed proposition on the basis of her undertaking a competent deduction from her existing knowledge. Moreover, since competent deductions are diachronic processes, it is also important to specify that the subject retains her knowledge in the entailing proposition throughout. For if the knowledge in the entailing proposition is lost during this process (perhaps as a result of the process itself), then clearly there is now no longer the same intuition that the entailed proposition should be known.
With the closure principle so formulated, it is hard to see how it could be denied. How could one draw a competent deduction from one's knowledge (modulo the caveats just noted) without thereby coming to know the deduced conclusion? As Keith DeRose (1995) has remarked, denying such a principle seems to commit one to endorsing the possibility of "abominable conjunctions" — for example, that one knows that one is presently seated at one's desk, but that one has no idea whether one is a bodiless BIV floating in a vat of nutrients (even though it is quite obvious the former entails the denial of the latter). Accordingly, henceforth when we refer without qualification to the "closure principle" we will have this highly compelling articulation of the principle in mind.
With this formulation of the closure principle in play, it follows that if one did know that one is seated at a desk (E), then one could, via closure, come to know that one is not a BIV. Conversely, if one cannot know that one is not a BIV, it follows that one does not — indeed cannot — know that one is seated at a desk. So while (S12) isn't in itself very compelling, it can be motivated via the incredibly plausible closure principle, and with (S12) and (S11) in play, the anti-skeptical (S13) is under threat.
Of course, the negation of (S13) is some logical distance away from a radical skeptical conclusion, since it directly concerns only a single proposition which has been shown to be unknown. Even so, it is not hard to see how one could derive the radical skeptical conclusion from this initial skeptical victory. For the general pattern of argument on display could be repeated any number of times to call specific propositions into play — there might be a need to vary the radical skeptical hypothesis employed on a case-by-case basis, but other than that the mechanics of the skeptical argument would stay the same. More generally, insofar as the skeptic can call into question our knowledge of something so straightforward as that one is currently seated at one's desk in (what appear to be) otherwise normal conditions, then the potentially devastating power of the skeptical argument is manifest.
What we have here is thus a putative paradox, in that we have a series of claims that have been shown either to be intuitive, or to be immediate consequences from intuitive claims (like the closure principle), but which are in fact in logical tension with one another, such that one of them must be denied:
THE RADICAL SKEPTICAL PARADOX (I)
(S11) One cannot know that one is not a BIV.
(S12) If one cannot know that one is not a BIV, then one cannot know that E.
(S13) One knows that E.
More generally, notice that while we have focused on a specific radical skeptical error possibility and a particular instance of everyday knowledge, the paradox in play does not trade on these details. What we have determined is that the following three claims appear to be inconsistent:
THE INCONSISTENT RADICAL SKEPTICAL TRIAD
(I) One is unable to know the denials of radical skeptical hypotheses.
(II) The closure principle.
(III) One has widespread everyday knowledge.
We can usefully represent this inconsistent triad in terms of (S11–3), in that these three more specific claims present us with a concrete instance of the triad. In doing so, (S11–3) makes the inconsistency at issue in (I–III) manifest.
Granted that this is an inconsistent triad, it follows that at least one of the claims that make up this triad must be false, since they cannot all be true on pain of contradiction. But given the intuitiveness of each claim, this means that radical skepticism appears to call on us to claim something deeply counterintuitive. While we can think of radical skepticism as a philosophical position, such that it involves the denial of (III) — and all that this entails — it is in fact more interesting to conceive of it rather as a paradox, in the sense that we are confronted with a deep tension within our own folk epistemological concepts, one that does not appear to be amenable to any obvious solution (including the skeptical solution of denying (III), which is surely the least palatable option of those available).
In particular, if we conceive of the skeptical problem as a paradox, then we will not be tempted to try to deal with this problem by aiming to convict an actual skeptical adversary of some dialectical error, such as by claiming that they are being incoherent in explicitly advancing their position. For suppose it were true that there is something seriously amiss with someone trying to actively advance the radical skeptical conclusion, via a rejection of (III), that we lack widespread perceptual knowledge of the world around us. Why would this offer us any intellectual comfort? After all, we knew already that the denial of (III) — of (S13), say — is implausible, so it's not as if finding out that its rejection leads to incoherence is unexpected. Moreover, the logical conflict among (I), (II), and (III) remains, even though these are three claims that we appear to be individually committed to. Far from resolving the skeptical problem (qua paradox), discovering that the skeptical position is incoherent seems to leave it entirely untouched.
2. Undercutting versus Overriding Anti-skeptical Strategies
With the skeptical challenge so understood, how might one go about responding to it? Notice that by conceiving of the problem as a putative paradox we thereby impose some constraints on philosophically adequate responses to this problem. In particular, while denying one (or more) of the claims that make up the inconsistent triad is obviously a necessary ingredient in any adequate response to the problem, merely denying one of these claims will not suffice to resolve it, at least not in a philosophically satisfying way. In particular, what is further required is some diagnostic story to explain the intuitive appeal of the claim in question even despite its falsity.
We can think of these diagnostic responses to radical skepticism as being of two main varieties. The first, an overriding anti-skeptical strategy, will offer a revisionary diagnostic story, one on which we have an independent theoretical basis for disregarding the relevant intuition in play. On this way of dealing with the problem, the skeptical paradox is bona fide, in that there is indeed a deep tension in our epistemological concepts, pretheoretically understood, which is being exposed by this puzzle. But it is also resolvable, in that we recognize that there are independent theoretical grounds for revising our epistemological concepts in fundamental ways that avoid the puzzle.
Compare an overriding anti-skeptical strategy with an undercutting anti-skeptical strategy, which is a much more ambitious response to the problem. On this proposal the diagnostic story on offer is meant to demonstrate that the skeptical "paradox" in play is in fact illusory. More specifically, the aim of this approach is to show that although the skeptical problem seems to be arising out of a tension in our epistemological concepts, pretheoretically understood, in fact it is the product of specific theoretical commitments that are revealed to be dubious. The skeptical puzzle is thus shown to be the product of faulty philosophical theory and not the natural manifestation of prephilosophical common sense.
While both undercutting and overriding anti-skeptical strategies can be adequate ways of dealing with the skeptical problem, undercutting anti-skeptical strategies are clearly to be preferred, all other things being equal. For if the skeptical paradox is bona fide, in that it arises out of an authentic tension within our pretheoretical epistemological concepts, then even if we can supply a sound theoretical basis for rejecting one of the claims that make up this paradox it will remain the case that the skeptical problem will generate intellectual unease. In contrast, if we can show that the skeptical "paradox" is in fact illusory, such that no theoretical revisionism is called for, then clearly there is likewise no need to regard this problem as generating any cause for genuine intellectual discomfort.
Excerpted from Epistemic Angst by Duncan Pritchard. Copyright © 2016 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Part 1. Epistemic Angst 9
Chapter 1. Radical Skepticism and Closure 11
1. The Contemporary Radical Skeptical Paradox 11
2. Undercutting versus Overriding Anti-skeptical Strategies 16
3. An Overriding Anti-skeptical Strategy (I): Nonclosure 17
4. An Overriding Anti-skeptical Strategy (II): Epistemic Externalism 19
5. Radical Skepticism about Rationally Grounded Knowledge 22
6. An Overriding Anti-skeptical Strategy (III): Abductivism 25
7. Concluding Remarks 28
Chapter 2. Radical Skepticism and Underdetermination 29
0. Introductory Remarks 29
1. Radical Skepticism and the Underdetermination Principle 29
2. An Overriding Anti-skeptical Strategy: Epistemic Externalism 32
3. Attributer Contextualism as an Overriding Anti-skeptical Strategy 36
4. Attributer Contextualism as an Undercutting Anti-skeptical Strategy 40
5. Comparing the Two Forms of Radical Skepticism 46
6. The Source of Underdetermination-Based Radical Skepticism 49
7. Two Sources of Radical Skepticism 54
8. Anti-skeptical Desiderata 58
Part 2. Wittgenstein and the Groundlessness of Our Believing 61
Chapter 3. Wittgenstein on the Structure of Rational Evaluation 63
0. Introductory Remarks 63
1. Wittgenstein on the Structure of Rational Evaluation 63
2. Wittgenstein contra the Skeptical “Paradox” 66
3. A Core Problem for the Wittgensteinian Account of the Structure of Rational Evaluation 70
4. Epistemic Ways of Developing the Wittgensteinian Account of the Structure of Rational Evaluation (I): The Externalist Reading 73
5. Epistemic Ways of Developing the Wittgensteinian Account of the Structure of Rational Evaluation (II): The Entitlement Reading 77
6. A Nonepistemic Way of Developing the Wittgensteinian
Account of the Structure of Rational Evaluation: The Nonpropositional Reading 84
7. Concluding Remarks 87
Chapter 4. Hinge Commitments 89
0. Introductory Remarks 89
1. The Nonbelief Reading 90
2. Hinge Commitments 94
3. Anti-skeptical Contrasts (I): Inferential Contextualism 103
4. Anti-skeptical Contrasts (II): Strawsonian Naturalism 110
5. Anti-skeptical Contrasts (III): Davidsonian Content Externalism 112
6. Wittgensteinian Anti-skepticism and Underdetermination-Based Radical Skepticism 113
7. Epistemic Priority and Underdetermination-Based Radical Skepticism 116
8. Concluding Remarks 118
Part 3. Epistemological Disjunctivism 121
Chapter 5. Epistemological Disjunctivism and the Factivity of Reasons 123
0. Introductory Remarks 123
1. Epistemological Disjunctivism in Outline 123
2. Three Core Problems for Epistemological Disjunctivism 127
3. Epistemological Disjunctivism qua Anti-skeptical Strategy 132
4. Radical Skepticism and Favoring/Discriminating Epistemic Support 136
5. Concluding Remarks 142
Chapter 6. Epistemological Disjunctivism and Closure-Based Radical Skepticism 144
0. Introductory Remarks 144
1. Anti-skeptical Contrasts (I): Rational Support Contextualism 144
2. Anti-skeptical Contrasts (II): Contrastivism 153
3. Anti-skeptical Contrasts (III): Dogmatism 157
4. A Weakness in Epistemological Disjunctivism 160
5. Epistemological Disjunctivism and Its Competitors 163
6. Concluding Remarks 166
Part 4. Farewell to Epistemic Angst 167
Chapter 7. Farewell to Epistemic Angst 169
0. Introductory Remarks 169
1. Recap: The Problem of Radical Skepticism 169
2. The Biscopic Proposal: Epistemic Angst Avoided 173
3. Some Anti-skeptical Contrasts 179
4. Concluding Postscript: Epistemic Vertigo 184