an example of skeptical hypothesis

7.4 Skepticism

Learning objectives.

By the end of this section, you will be able to:

Define skepticism as it is used in philosophy.
Compare and contrast global and local skepticism.
Offer and explain a skeptical hypothesis.
Outline the general structure of argument for global skepticism.

Philosophical skepticism is the view that some or all knowledge is impossible. A skeptic questions the possibility of knowledge—particularly justification—in some domain. A global skeptic rejects the possibility of knowledge in general. But one need not reject the possibility of all knowledge. A local skeptic questions the possibility of knowledge only in particular areas of study. One can be a local skeptic about moral knowledge or scientific knowledge. This section will first look at global skepticism and the arguments offered in support of it and then will briefly look at local skepticism.

Global Skepticism

Global skepticism is a view that questions the possibility of all knowledge. To make their case, global skeptics point to the lack of the possibility of certainty in our beliefs. Because we cannot know that our beliefs are true, we cannot know in general. Usually, global skepticism attempts to undermine the possibility of forming justified beliefs. Global skeptics target all beliefs, or all beliefs about the external world (which amounts to most beliefs). Most beliefs tacitly or explicitly assume the existence of an external world. When I have the experience of seeing a bird in a tree and think, “There is a bird in that tree,” I assume that there is an actually existing physical bird in an actually existing physical tree in an actually existing real world outside of me. There is means “there exists.” I believe the bird, tree, and world all exist independently of my thoughts. The global skeptic questions beliefs such as these.

The Dream Argument

How many times have you realized that you were dreaming while you were dreaming? Most people believe that whatever they are dreaming is real during the dream. Indeed, the fact that people think dreams are real while dreaming is what makes nightmares so terrible. If you knew the content of a nightmare was a dream, then it would not be nearly as scary. Zhuang Zhou (c. 369–286 BCE) was a Chinese Taoist philosopher who argued that for all we know, we could currently be dreaming while thinking we are awake. Imagine dreaming that you are a butterfly, happily flitting about on flowers. When you wake, how can you determine whether you have just woken from dreaming you are a butterfly or you are a butterfly who has just started dreaming that you are human? Zhuang Zhou explains:

While he is dreaming he does not know it is a dream, and in his dream he may even try to interpret a dream. Only after he wakes does he know it was a dream. And someday there will be a great awakening when we know that this is all a great dream. Yet the stupid believe they are awake, busily and brightly assuming they understand things, calling this man ruler, that one herdsman—how dense! Confucius and you are both dreaming! And when I say you are dreaming, I am dreaming, too. (Zhuangzi 2003, 43)

Zhaung Zhou puts forward the possibility that all of what we take to be conscious experience is actually a dream. And if we are dreaming, then all our beliefs about the external world are false because those beliefs take for granted that our current experience is real.

The Evil Demon Argument

Nearly two millennia after Zhuang Zhou, René Descartes also proposed a dream hypothesis. Descartes argued that because dreams often incorporate experiences we have in real life, it is impossible to distinguish between dreaming and waking life (Descartes 2008). But Descartes eventually concludes that even if he could be dreaming, there are still some beliefs he can know, specifically arithmetic. Even in dreams, 1 + 1 = 2, and a square will always have four sides. And so, Descartes devises an even stronger skeptical hypothesis: what if we are being tricked by an evil demon?

Descartes’s evil demon is powerful. It can make you believe things, and it can trick you by controlling your experience. The evil demon can make you believe you are currently eating a sandwich by directly feeding you the sensory experience of eating a sandwich (the sight, the smells, the taste, the feel). Under this scenario, you cannot tell the difference between actually eating a sandwich and merely believing you are eating one because the evil demon is tricking you. If we cannot reliably tell the difference between experiences caused by reality and experiences caused by an evil demon, then we cannot know anything. We can represent Descartes’s argument as follows:

If I cannot rule out the possibility that an evil demon is tricking me, then I do not have any knowledge of the external world.
I cannot rule out the possibility that an evil demon is tricking me.
Therefore, I do not have knowledge of the external world.

Why does Descartes claim we can’t have knowledge if we cannot rule out the evil demon hypothesis? If an evil demon is tricking us, then all our beliefs are wrong. And if we cannot rule out the possibility that we are wrong, then we are not justified. And if we are not justified in our beliefs, then we cannot have knowledge of them.

Putnam’s Brain in a Vat

If you don’t like evil demons, then consider a more modern version of a skeptical hypothesis: the “ brain in a vat ” conceived of by American philosopher and mathematician Hilary Putnam (1926–2016). Imagine that while you were asleep last night, a group of scientists kidnapped you and took you to their lab. There, they surgically removed your brain and placed it in a vat of nutrients. The scientists then hooked up your brain to a sophisticated new computer system. They were able to download your memories so as to create new experiences. The result is a seamless experience of consciousness between yesterday and today. When you woke this morning, your life seemed to proceed without disruption. Can you prove that you are not a brain in a vat? No, you cannot. The scenario stipulates that your experience will seem exactly the same whether you are a brain in a vat or not. Other, similar skeptical scenarios are easy to come up with. Consider the possibility that you are caught in a virtual reality world or that you are trapped in the Matrix.

General Structure of Global Skeptical Arguments

Skeptical hypotheses and the arguments that they inspire all have a similar structure:

If I cannot rule out the possibility of SH, then I cannot be justified in believing that P.
I cannot rule out the possibility of SH.
Therefore, I cannot be justified in believing that P.

SH is a skeptical hypothesis. P is any proposition about the external world. Premise 1 is the skeptic’s challenge—that you must rule out skeptical hypotheses. Premise 2 relies on limitations within your perspective. The skeptic claims that you can rule out the possibility of whatever skeptical hypothesis is at hand only if you are able to construct an argument that defeats that hypothesis using the evidence you have (and a priori knowledge). As demonstrated, this is difficult to do. The nature of the skeptical hypotheses used for global skepticism limits your evidence to the contents of your thoughts. What you take to be evidence of the external world (that you perceive things that seem to be separate from yourself) is effectively neutralized by the possibility of a skeptical hypothesis.

Responses to Global Skepticism

The philosopher who wishes to overcome philosophical skepticism must find reasonable grounds for rejecting the skeptic’s argument. The different skeptical arguments reveal a specific conception of the level of justification required for knowledge. Skeptical arguments rely on the existence of doubt. Doubt exists when we cannot rule out a possibility. If we have doubt, we are not certain. We cannot be certain that we are not, say, a brain in a vat. And if we cannot be certain, then we cannot know anything that implies we are not a brain in a vat. Certainty is a very strict measure of justification. One clear possible response is to simply deny that one needs certainty in order to be considered justified. This section looks at some of the classical responses to the skeptic’s argument that we cannot know anything.

British philosopher G. E. Moore (1873–1958) presented an argument against skepticism that relies on common sense. In his famous paper “Proof of an External World,” Moore begins by raising his right hand and claiming, “Here is one hand,” then raising his left hand and claiming, “Here is another hand” (Moore 1939). Therefore, he concludes that skepticism is false. At first glance, this argument may seem flippant. It is not. Moore means to replace the second premise in the skeptical argument with his own premise: I know I have hands . The skeptical argument starts with the premise that if you cannot rule out a skeptical hypothesis, then you do not have knowledge of some proposition pertaining to the external world. Moore uses “I have two hands” as his proposition about the external world. In effect, he accepts the skeptic’s first premise, then uses his commonsense belief in the truth of “I have two hands” to defeat the skeptical hypothesis . Here is the argument’s structure:

I am justified in believing that P.
Therefore, I can rule out the possibility of SH.

In claiming that he has two hands, Moore claims that he is justified in believing propositions about the external world. And if he is justified, then he can rule out the skeptical hypothesis. The skeptic’s argument takes the form of what is called modus ponens , meaning a valid inference where the antecedent of a conditional is affirmed. Moore’s argument takes the form of what is known as modus tollens , meaning a valid inference where the consequence of a conditional is denied.

But notice that the two arguments contradict each other. If we accept the first premise, then either Moore’s or the skeptic’s second premise must be false. So why did Moore think his second premise is better? The choice is between thinking you are justified in believing that you have two hands and thinking you are justified in believing the skeptical hypothesis might be true. Moore thinks he has better reason to believe that he has two hands than he does for believing the skeptical hypothesis is true. For Moore, it is just common sense. You have reason to believe that you have two hands—you can see them and feel them—while you have no reason to believe the skeptical hypothesis is true.

Many philosophers remain unconvinced by Moore’s argument. Any person who accepts the possibility of the skeptical hypothesis will disagree with his premise 2. The possibility of the skeptical hypothesis effectively undermines justification in the belief that you have two hands.

Contextualism

As we just saw, some theorists reject the notion that you must be certain of a belief—that is, rule out all possible defeaters—in order to have knowledge. Moore thinks he has more justification to believe he has two hands than he does that there’s an evil demon tricking him. And in determining whether I am justified in believing in the bird outside my office window, I rarely consider the possibility that I could be a brain in a vat. I’m more likely to focus on my poor vision as a defeater. In the context of bird identification, wild skeptical hypotheses seem out of place. Indeed, we often adjust how much justification we think is needed for a belief to the task at hand. Contextualism is the view that the truth of knowledge attributions depends on the context. Contextualism is a theory about knowledge and justification . When we attribute knowledge to a subject S, the truth of the knowledge claim depends on the context that S is in. The context of S determines the level of justification needed for a true belief to count as knowledge. Contextualism comes from the observation that the level of confidence needed for justification changes depending on what the belief is as well as its the purpose and its importance, among other things. We expect a high degree of justification from physicians when they diagnose disease but less justification from friends recalling the title of a movie because there’s much more at stake in medical diagnoses.

Contextualism deals with skepticism in a unique way. Rarely are we in situations where we must rule out skeptical hypotheses to consider ourselves justified. Indeed, it is generally only when a skeptical hypothesis has been explicitly raised that we think we need to rule it out to be justified. And in our daily lives, the skeptical hypothesis just does not seem relevant. Yes, the possibility that we are brains in a vat technically still exists; we just do not think of it.

Skepticism in Specific Domains

As explained above, local skepticism questions the possibility of knowledge only in particular areas of study. People can accept that knowledge of the external world is possible while also questioning whether knowledge is achievable in more specific domains. A common form of local skepticism focuses on religious belief, specifically knowledge of the existence of God. Another form of local skepticism concerns the ability to ever have moral knowledge. Skepticism in these domains does not entail that there is no God or that all moral claims are false. Rather, skepticism means that we can never be sufficiently justified in believing that there is a God or that moral claims are true. We simply can never know either way whether, for example, God exists.

Skepticism about morality arises due to the nature of its subject. Moral claims are normative, which means that they assert claims about what ought to be the case rather than what is the case. But moral claims are difficult to prove, given their normative nature. How can you prove what ought to be the case? Usually, moral claims are grounded in value claims. An ethicist may say that we ought to help a stranger because well-being is morally valuable. But the skeptic will point out that we cannot prove that something is valuable. We do not have sensors that can confirm moral value. Moral claims instead rest on arguments. The problem, as Scottish Enlightenment philosopher David Hume (1711–1776) explained, is that no amount of description can ever help us logically derive a normative claim (Hume 1985). This leaves room for doubt, and therefore skepticism.

Skeptical positions about God also focus on the lack of sufficient evidence. A skeptic can reasonably ask, What sorts of evidence would show the existence of God? Certainly, if God unambiguously appeared right now to everyone in the world simultaneously, then we would have reliable evidence. But God has not done so. The most we have is testimony in the form of religious texts. And testimony, particularly a chain of testimony stretching back hundreds and hundreds of years, is not necessarily reliable. Why believe, for example, the Christian Bible? Blaise Pascal (1623–1662), himself a devote Catholic, argued that the very nature of God—having no limits and existing beyond time—precludes the possibility of ever comprehending the full true nature of God or God’s existence. He states, “Who then can blame the Christians for not being able to give reasons for their belief, professing as they do a religion which they cannot explain by reason. . . . It is in lacking proofs that they do not lack sense” (Pascal 1973, 93). Pascal contends that not attempting to give proof of God is the sensible thing to do. A person can simply rely on faith, which is belief based on insufficient evidence.

Think Like a Philosopher

In your view, what is the relationship between reason and faith? Some theologians say that reason can establish the existence of a supreme being. Others think that reason can only partially justify religious belief and that full belief requires faith, or belief without reason. Reason for some is antithetical to faith, which requires blind obedience. For example, in the biblical story of the sacrifice of Isaac, Abraham is willing to sacrifice his only son to God as an act of faith. How do you think we should understand the role of reason in religious belief?

As an Amazon Associate we earn from qualifying purchases.

This book may not be used in the training of large language models or otherwise be ingested into large language models or generative AI offerings without OpenStax's permission.

Want to cite, share, or modify this book? This book uses the Creative Commons Attribution License and you must attribute OpenStax.

Access for free at https://openstax.org/books/introduction-philosophy/pages/1-introduction

Authors: Nathan Smith
Publisher/website: OpenStax
Book title: Introduction to Philosophy
Publication date: Jun 15, 2022
Location: Houston, Texas
Book URL: https://openstax.org/books/introduction-philosophy/pages/1-introduction
Section URL: https://openstax.org/books/introduction-philosophy/pages/7-4-skepticism

© Dec 19, 2023 OpenStax. Textbook content produced by OpenStax is licensed under a Creative Commons Attribution License . The OpenStax name, OpenStax logo, OpenStax book covers, OpenStax CNX name, and OpenStax CNX logo are not subject to the Creative Commons license and may not be reproduced without the prior and express written consent of Rice University.

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

4 Skepticism

Daniel Massey

Chapter Learning Outcomes

Upon completion of this chapter, readers will be able to:

Differentiate global skepticism and external-world skepticism by the scope of their doubts.
Explain the role of skeptical hypotheses in arguments for external-world skepticism.
Construct original skeptical hypotheses modeled on those discussed.
Analyze the Moorean and contextualist responses to skepticism, recognizing how Moorean responses are defiant while contextualist responses are concessive.

Skepticism and Its Scope

Skeptics , as we will use the term, deny knowledge. A skeptic need not deny that free will is real or that God exists but will deny that anyone knows whether either of these is the case.

Skeptics are a variety. Some have limited targets. Some deny only that we have knowledge of what is right and wrong. Others deny only that we have knowledge about the future. But some skeptics are more ambitious, casting the skeptical net more widely. Global skeptics deny that we can have any knowledge at all, [1] and may even deny that we know whether skepticism is true. Such a radical skepticism is bound to elicit derision. It seems unrealistic, impractical, and perhaps even self-refuting. In fact, very few philosophers have been global skeptics. But even anti-skeptics have been enamored of the view, seeing it (and potential refutations of it) as a way of better understanding what it is to have knowledge in the first place.

Our attention will be on one particular kind of skepticism—one that falls short of global skepticism while still denying that we know much of what we think we know—along with one particularly prominent style of argument in its favor. [2] We will then consider two influential responses to this form of skepticism.

External-World Skepticism

You likely believe a great number of things—that you are presently on or near the surface of the earth; that you are human; that there are plants, animals, and other humans; and that their lives have likewise unfolded in close proximity to the earth. These unremarkable beliefs have something in common: you believe them based on sensory experience. You have likely seen, heard, or felt dogs, people, and even planets. On the basis of such experiences, you have come to believe a great number of things about such objects. You know of their existence through your experiences, but their existence continues unabated even when you are not experiencing them. They are, philosophically speaking, external objects , objects that exist in the external world (the world external to our minds). The skeptic we will consider denies that we can have knowledge of any such objects because all the available evidence of sense experience is compatible with no such objects existing . This form of skepticism is called external-world skepticism (hereafter “skepticism” for short).

Let’s start with the fact that experiences of an object are compatible with the non-existence of that object. We often have experiences of things that turn out not to exist. If nothing else, our dreams consist in experiences of objects, many of which do not exist. I may dream of a puppy and be sad to find when I wake that all my experiences of the puppy were figments of my dream. Experience alone does not entail that the things we experience exist.

Now consider the following possibility. You have had a set of experiences for some years, all of which are perfectly coherent. When you have had experiences of puppies, you at least sometimes have further experiences of those same puppies, which have grown and developed in a way consistent with their being real animals. All your experiences perfectly mesh together, and nothing about those experiences suggests anything other than that they are the experiences of external objects. But what guarantees this to be so? What guarantees that you are not having a long, perfectly coherent dream rather than the life you take yourself to have led? French philosopher René Descartes (1596–1650) famously despaired of finding some evidence that could distinguish being awake from having such a dream ([1641] 1985, First Meditation). We will call this hypothesis and others like it skeptical hypotheses . The skeptic contends that if you cannot eliminate such possibilities (the possibilities entertained by skeptical hypotheses), that if you cannot be absolutely certain of what you know, then you cannot know that the external world exists.

A properly crafted skeptical hypothesis is devious. You might conjecture that if your life to this point had been a dream, then you would not have felt pain and kept dreaming, knowing that pain typically wakes a sleeping human. But this alleged fact about the sleeping habits of human beings is itself something you have “learned” through experience. And it is experience itself that is being thrown into doubt. For all you know, this fact about the sleeping habits of human beings is no fact at all, but merely another figment of your dream. The skeptic claims that since experience is our only available source of evidence, and since experience cannot distinguish between being awake and dreaming, we cannot know we are awake. That is, we cannot know that the external world exists. Appeals to facts about the external world “learned” through experience will be powerless to help (Stroud 1984, 8).

Let’s represent this skeptical argument as follows:

Nothing in experience can eliminate the possibility that your life has been a long, perfectly coherent dream.

Therefore, you do not know that the external world exists.

This is a common style of skeptical argument. Other arguments in favor of skepticism have been put forward, and a staggering variety of responses to this argument (and other skeptical arguments) can be found throughout the history of philosophy. We will now turn our attention to two such responses from recent philosophers.

Box 1 – Famous Skeptical Hypotheses

Skeptical hypotheses are scenarios compatible with all possible evidence yet inconsistent with our ordinary beliefs. The possibility that all our experiences have been those of a long, perfectly coherent dream is one skeptical hypothesis, but others have occurred throughout the history of philosophy and even in popular culture.

Zhuangzi’s butterfly dream.

The Chinese Daoist philosopher Zhuang Zhou, also known as Zhuangzi (ca. 369–286 BCE), dreamed that he was a butterfly. Upon awakening, he considered the possibility that instead, he was in fact a butterfly dreaming he was Zhuangzi (2013, 18). No available evidence seems to be able to distinguish between his being a human dreaming he is a butterfly and his being a butterfly dreaming he is a human.

The evil demon.

Another skeptical hypothesis considered by Descartes ([1641] 1985, First Meditation). Here an evil demon with godlike powers uses those powers to deceive you in all your beliefs. What could you know if a near-omnipotent being wanted you to know nothing?

Brain in a vat.

A skeptical hypothesis formulated by the contemporary American philosopher Gilbert Harman (1973) and popularized by the American philosopher Hilary Putnam (1926–2016). In this scenario, you are a disembodied brain suspended in a nutritive inside a vat while powerful computers overseen by scientists feed you experiences of an ordinary human life (Putnam 1981). The 1999 science-fiction movie The Matrix adapted the brain-in-a-vat scenario, with the protagonist managing (due to the intervention of others) to escape.

Moore against Skepticism

One important response to this style of argument comes from the British philosopher G. E. Moore (1873–1958), known for his commonsense approach to philosophical problems. [3] Among his many philosophical accomplishments was to give several powerful responses to skepticism based on an appeal to common sense. We will look at two forms of this Moorean response to skepticism.

In his “Proof of an External World,” Moore claims to be able to demonstrate (contrary to the skeptical view) the existence of the external world—to prove that the world of ordinary objects and people has an existence outside of our experiences. He gives two such “proofs” to illustrate his general approach. Holding up a hand, he claims to know that he has a hand. Holding up his other hand, he claims to also know that he has a second hand. Since hands are ordinary objects existing beyond our experiences of them, Moore infers he has proven the existence of at least two external objects. Our knowledge of the external world is secured! (Moore [1939] 1963, 144).

This may seem to have an air of trickery about it. The skeptic challenges us to prove that we are not dreaming, that our entire life has not been a long yet perfectly coherent set of experiences corresponding to nothing beyond them. Moore’s holding up of his hands and declaring himself to know he has hands seems to entirely ignore the challenge put before him by the skeptical hypothesis, and so to miss the point. However, Moore was not missing the point. His aim is to challenge a sometimes-unstated assumption in the skeptic’s argument. Consider our presentation of the skeptical argument again. Moore does not challenge the premise (P) but instead challenges the inference from the premise (P) to the conclusion (C). The skeptical argument, as presented, assumes that if you cannot prove that your life has not been a long, perfectly coherent dream, then you do not know all those things you normally take yourself to know on the basis of your experiences. That is, it assumes that if you cannot definitively eliminate the possibility that your life is a dream, then you do not know that you are awake or that anything beyond your experiences really exists.

Moore makes the following contentions: (a) that he can know that external objects exist even if he cannot definitively eliminate the possibility that he is dreaming; (b) that our ordinary practices of taking ourselves and others to know things do not require that we meet such skeptical challenges; and (c) that ordinary standards are the only ones that matter, and so they are the standards of evidence in play whenever we are faced with doubts (Moore [1939] 1963, 145). Moore advocates fallibilism about knowledge , the view that you may know something despite the possibility of being wrong. The fallibilist does not accept the requirement that skeptical hypotheses be definitively ruled out before we know that the external world exists. It may suffice if we can manage to establish that they are improbable .

Consider again Moore’s claim that he has a hand. He holds up his hand and claims that this is sufficient proof of its existence. He knows he has a hand because, well, there it is. Someone may doubt that he has a hand. Some such doubts may be reasonable or at least intelligible. If someone were to doubt that Moore had a hand because the light in the room was dim and perhaps Moore had held up a small statue rather than his hand, then Moore would accept the need to respond to this doubt. He could not count as knowing he has a hand if he is unable to show that what he held up was not a small statue. But there are clear ways of making such a distinction. He can look more closely at his hand in better lighting. He can touch his hand and see whether it feels like flesh or stone. There are any number of ways he could respond to this doubt and thus retain his title to knowledge (Moore [1939] 1963, 145).

But in the case of the skeptic’s doubts, no such responses are possible. This is by the skeptic’s design. The skeptical hypothesis is described so that there is no way to distinguish it from being awake. Many philosophers, of course, have tried to find such a distinction. Moore’s response, however, is simply to deny that such a doubt requires a response, and to deny that we must be able to distinguish the dream hypothesis from the ordinary hypothesis. We can know there is an external world because we can know we have hands. We can know we have hands because we can produce them when and if our title to knowledge of their existence comes into question. We cannot prove they exist in the strong sense demanded by the skeptic, but there is no obvious reason to think we must do so to count as knowing they exist (Moore [1939] 1963, 148). Fallibilism must be considered.

Moore is also known for another response to skepticism (Moore [1959] 1963). Here he considers what a skeptical argument does: it presents a series of claims which (according to the skeptic) force us to accept a skeptical conclusion. To illustrate his point, Moore borrows an example from fellow British philosopher Bertrand Russell (1872–1970):

I do not know immediately that the pencil exists.
Its existence does not follow deductively (i.e., with certainty) from anything known immediately.
My belief in it then is based on “analogical or inductive” (non-deductive) reasoning.
What is so based is not “certain knowledge.”
Therefore, I do not know the pencil exists.

This pattern of argument is a familiar one. Premises (1)–(4) tell us that our knowledge of a pencil cannot be based directly on experience, or deduction from anything known on the basis of experience, since we might be dreaming all our pencil experiences. It is at best based on some other (non-deductive) type of reasoning, and beliefs based on these other types of reasoning do not count as “certain knowledge.” Hence, we cannot know that the pencil exists.

Moore notes that this argument does not establish its conclusion in the sense of showing that accepting it is the reasonable course. Arguments generally force us to choose between either accepting the premises (and therefore the conclusion) or rejecting the conclusion (by rejecting one or more premises). The choice, then, is between accepting some number of philosophical claims (given by the premises) or accepting our ordinary claim to knowledge (i.e., rejecting the conclusion). Moore maintains that, considered in this light, the reasonable choice is to reject the conclusion, hence to reject one or more premises, and to maintain our ordinary claim to knowledge.

Indeed, this understates the failure of the argument to make acceptance of the conclusion reasonable. You may not be able to identify exactly what is wrong with the premises, nor marshal any strong arguments against any one of them in particular, and yet you may reasonably maintain that some-premise-or-other – I-know-not-which is false rather than accept the conclusion that you do not know that pencils exist. It is quite reasonable to feel certain that pencils exist, more reasonable than to maintain that each and all of premises (1)–(4), a set of controversial philosophical claims, are true. If Moore is right, then a skeptical argument can only pit a set of philosophically motivated premises against our ordinary, commonsense claims to knowledge, but it can never declare victory. It will always be more reasonable to believe we know that pencils and other external objects exist than accept the difficult, abstruse, controversial premises of skeptical arguments. [4]

Box 2 – To Modus Tollens a Modus Ponens

Moore’s second response to the skeptical argument employs a general argumentative technique that is very useful to learn. Take any propositions p and q . Consider the common argument form that logicians (and philosophers generally) call modus ponens (Latin for “mode that affirms”): [5]

Modus ponens If p then q . p Therefore, q .

This is a (deductively) valid argument form. Logicians use the term “valid” differently from its everyday usage. To say that an argument is valid (in the logician’s sense) means the following: under the assumption that all of the premises are true, it follows with certainty that the conclusion is also true. Apply this idea to modus ponens : assuming that p is true, and that if p then q , it follows with certainty that q is true.

However, it is possible to convert modus ponens into another (deductively) valid argument form by swapping the positions of p and q in the second and third lines, then negating them both (leaving the “if-then” premise untouched). Logicians call this form modus tollens (Latin for “mode that denies”):

Modus tollens If p then q . Not- q Therefore, not- p .

So, upon encountering any modus ponens argument, one logically acceptable option is to oppose it with a corresponding modus tollens argument. As philosophers often say, you can always “ modus tollens a modus ponens .” Of course, only one of the two arguments can be “sound” in the logician’s sense (i.e., valid plus true premises—features which jointly guarantee a true conclusion). Since both arguments are valid and share a premise (“If p then q ”), choosing which argument to accept (if either) comes down to the premise on which they differ ( p vs. not- q ). The deciding factor then, is whether p or not- q is the more plausible proposition to take as a starting point. This is where Moore’s “commonsense” strategy comes into play.

Challenge exercise: Try recasting the skeptic’s argument in modus ponens form. Then see if you can put Moore’s second response into a corresponding modus tollens form. After doing so, identify the premises on which the two arguments differ, then reflect on which one of them you find more plausible and why.

The Contextualist Concession

Moore’s responses are defiant. He seeks to undermine skepticism by rejecting a key premise implicitly employed in the skeptical argument and even by undermining the power of such arguments to ever make it reasonable to accept their conclusion. Many philosophers, however, have found Moore’s responses unpersuasive. Canadian philosopher Barry Stroud (1935–2019) notes that much of the ordinary practice of taking ourselves and others to know things requires that those who know something be able to eliminate contrary possibilities (1984, 18). For instance, if I know that the yellow bird I see outside my window is a canary, then I must know it is not a goldfinch. This is how we normally regard claims to knowledge. Someone who claims to know that the yellow bird outside her window is a canary but who also admits that it might be a goldfinch would not be regarded as someone who knows that the bird is a canary. Likewise, when Moore says that he knows he has a hand but also admits he cannot eliminate the possibility that he is dreaming, many philosophers regard him as confused as well. In the normal case, we require that we be able to eliminate alternatives compatible with our current evidence. Why should things be any different when the alternative is a skeptical hypothesis?

The contextualist response attempts to split the difference. This response admits that, by our ordinary standards of knowledge, the dream hypothesis does not need to be ruled out. In ordinary life, we can know perfectly well that tables, chairs, trees, other people, and even G. E. Moore’s hands exist. But, at the same time, when skeptical doubts are raised, as they are in a philosophy classroom or while reading a philosophy book, then the standards in play shift and become more demanding. Indeed, they become demanding enough to vindicate skepticism. We can have knowledge of the external world, but only as long as some skeptic does not shift the standards of evidence and thereby deprive us of that knowledge. This is the contextualist response to skepticism. It concedes that skepticism is correct when we are operating under standards that would have us eliminate skeptical hypotheses but is incorrect when we are operating under the laxer standards of day-to-day life.

Contextualists differ in their understanding of when and how the standards for knowledge shift, and even what it means for there to be shifting standards for knowing. We will turn our attention to just one way of working out contextualism as found in the work of the American philosopher David Lewis (1941–2001). In his view, the standards in play are those in a conversation; shifts in those standards are due to changes in the conversation. The thought that conversations can set epistemic standards may seem strange, but before digging into the details of Lewis’s views, consider a simple example unrelated to knowledge. Amy, looking at a baby elephant standing next to a dog, may say, “The baby elephant is big.” She will be right. Compared to the dog, the baby elephant is big. But if she were to say again, “The baby elephant is big,” while it was standing next to an adult elephant, she would be wrong. Compared to the adult elephant, the baby elephant is not big. Whether or not the baby elephant counts as “big” depends on the context in which one makes the claim.

Of course, as the conversation moves along, what we should say about the baby elephant may as well. If Amy is at the zoo and her friend says, “The baby elephant is small,” while pointing out both the baby and adult elephants, then her friend says something true. Once the friend brings the focus of the conversation to the adult elephant, it would be daft of Amy to say again that “The baby elephant is big.” The conversation has changed what is needed for the sentence “The baby elephant is big” to count as true, and given the change, subsequent claims that “The baby elephant is big” will be false, even as prior such claims were true. (We will return to this example of Richard Feldman’s shortly.)

In a somewhat similar way, Lewis believes that the truth of attributions of knowledge varies with shifts in the conversation. In a conversation, a speaker may attribute knowledge to herself or to some other person (including someone who is not part of the conversation). It will be perfectly proper, given the state of the conversation, to ignore a wide range of possibilities during the conversation (Lewis 1996, 544). If you and I have a conversation about the safety of the drinking water in our hometown, we properly ignore many possibilities even as we must attend to others. If I claim to know the water is safe to drink, I must be able to eliminate the salient (that is, easily noticeable) possibilities, including the possibility that the water contains unsafe quantities of lead. Yet we also presuppose that a number of possibilities are false, for instance the possibility that the fluoride in our hometown’s drinking water may have been added by a secret cabal of aliens intent on stealing our bodily fluids. Still, my utterance of “I know the drinking water is safe” will express a truth if my evidence rules out the conversationally salient possibilities contrary to that claim. I must be able to eliminate, among other things, the possibility that there are unsafe levels of lead in that water. I need not eliminate the possibility of an alien conspiracy.

But exactly which possibilities may we properly ignore in a conversation? Lewis’s idea is that the range of possibilities that may be ignored shifts as the conversation progresses according to a variety of conversational rules. For the sake of exposition, I will consider only one such rule:

The rule of belief forbids conversational partners to ignore any possibility that one or more of those partners believe. (Lewis 1996, 555–6)

It is a poor conversationalist who simply ignores what her conversational partners believe. So, while you and I may ignore the possibility that aliens have put fluoride into the drinking water of our hometown, if another person joined our conversation—someone who believed precisely that—then we would no longer be able to ignore that possibility. We would now need to eliminate that possibility using the available evidence to know that our town’s drinking water is safe. Once our conspiratorially minded friend joined the conversation, our knowledge might dissipate unless we could eliminate his far-fetched theories .

In Lewis’s view, the basic contextualist response to skepticism is sound. In many conversations, skeptical hypotheses will be properly ignored. Suppose we are trying to determine whether the drinking water in our hometown is safe to drink. In that case, we may properly ignore the possibility that our town and the rest of the external world is merely a figment of a long, perfectly coherent dream. We may then take ourselves to know the water is safe to drink (or not safe to drink) without attending to that far-fetched possibility. Yet if a skeptical epistemologist were to join the conversation, one who took seriously skeptical hypotheses, we could no longer properly ignore that possibility, and thus we could no longer know that our hometown and the rest of the external world exists. Or at least we could not know such things unless we can use the available evidence to eliminate the various skeptical hypotheses that have become by the rule of belief a salient part of our conversation. We have already seen how challenging that can be. The outcome is that skepticism may yet be vindicated, but only when we take it seriously. In most of life, we take neither skepticism nor the supporting skeptical hypotheses seriously. Moreover, absent some shift in our conversation, it is entirely proper that we not take them seriously. Our knowledge of the external world is secure absent such shifts (Lewis 1996, 561).

One objection to this approach points to the way that contextualism treats the word “know” (Feldman 2003, 152–5). The contextualist response to skepticism holds that in contexts where skepticism is not an issue, proper use of the word “know” does not require us to eliminate skeptical hypotheses. Once the skeptical interlocutor challenges our title to knowledge, however, the standards have changed, and we must respond. But Richard Feldman notes that this is different from other words whose use varies from context to context. Return to the example of the baby elephant. If Amy calls a baby elephant big while comparing it to a dog, Amy will not see a need to revise her views when someone notes the baby elephant is not big compared to an adult elephant. Anyone who asked her to recant her view on those grounds would be misguided. Baby elephants are big compared to dogs but not big compared to adult elephants . The latter fact does nothing to challenge the former. But the skeptic does challenge our ordinary knowledge claims. I cannot simply dismiss the skeptical challenge, even if it is presented in response to my ordinary context claims about knowing. The skeptic maintains I am mistaken, and so I must respond. The person who claims Amy is wrong about the size of baby elephants is confused and can safely be ignored. Contextualism requires that the word “know” works like the word “big,” but it is not clear that it does.

As we have seen, skepticism is a considerable challenge, although there are also considerable responses. In closing, it is worth remembering that there are further arguments on both sides of external-world skepticism, not to mention many other forms of skepticism beyond the scope of this chapter. It is also worth noting that, while skepticism has a pessimistic connotation, those who count themselves skeptics of one sort or another tend to find significant value in it. The ancient Greek philosopher Pyrrho of Elis (ca. 360–270 BCE), for example, advocated global skepticism as a cure for the dogmatic certainty which he located at the root of the ills of life. [6] Socrates, though (arguably) not a (global) skeptic, shared the conviction that true wisdom lies in recognizing what one does not know (Plato [ca. 390 BCE] 2009). In the modern era, Russell echoes similar sentiments when he writes of a “liberating doubt” delivered through the vehicle of philosophy ([1912] 2013). Let us close, then, with a passage from Russell linking the very purpose of philosophy to the value of doubt and the improvement of the self.

Box 3 – The Value of Doubt

The value of philosophy is, in fact, to be sought largely in its very uncertainty. The man who has no tincture of philosophy goes through life imprisoned in the prejudices derived from common sense, from the habitual beliefs of his age or his nation, and from convictions which have grown up in his mind without the co-operation or consent of his deliberate reason. To such a man the world tends to become definite, finite, obvious; common objects rouse no questions, and unfamiliar possibilities are contemptuously rejected. As soon as we begin to philosophize, on the contrary, we find, as we saw in our opening chapters, that even the most everyday things lead to problems to which only very incomplete answers can be given. Philosophy, though unable to tell us with certainty what is the true answer to the doubts which it raises, is able to suggest many possibilities which enlarge our thoughts and free them from the tyranny of custom. Thus, while diminishing our feeling of certainty as to what things are, it greatly increases our knowledge as to what they may be; it removes the somewhat arrogant dogmatism of those who have never travelled into the region of liberating doubt, and it keeps alive our sense of wonder by showing familiar things in an unfamiliar aspect. (Chapter XV)

— Bertrand Russell “The Value of Philosophy” In The Problems of Philosophy

Questions for Reflection

As noted in the introduction to this chapter, some philosophers think that global skepticism is self-refuting. What is a self-refuting view? Why might global skepticism be subject to this charge? How might global skeptics respond?
Review the theories of epistemic justification discussed in Chapter 2 (e.g., coherentism, modest foundationalism, and explanationism). Given the role of justification in knowledge (see Chapter 1 , how might each theory inform one’s response to external-world skepticism? Does any such theory provide an adequate response to the skeptical challenge? Why or why not?
Do you think that the Moorean treatment of skepticism is satisfactory? Why or why not?
Do you think that the contextualist response to skepticism is satisfactory? Why or why not?
Revisit the lottery problem from Box 2 of Chapter 1 . Some philosophers have proposed contextualism as a solution to this problem. What might such a solution look like?

Share This Book

school Campus Bookshelves
menu_book Bookshelves
perm_media Learning Objects
login Login
how_to_reg Request Instructor Account
hub Instructor Commons

Margin Size

Download Page (PDF)
Download Full Book (PDF)
Periodic Table
Physics Constants
Scientific Calculator
Reference & Cite
Tools expand_more
Readability

selected template will load here

This action is not available.

7.4: Skepticism

Last updated
Save as PDF
Page ID 162174

Nathan Smith et al.

\( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

\( \newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\)

( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)

\( \newcommand{\Span}{\mathrm{span}}\)

\( \newcommand{\id}{\mathrm{id}}\)

\( \newcommand{\kernel}{\mathrm{null}\,}\)

\( \newcommand{\range}{\mathrm{range}\,}\)

\( \newcommand{\RealPart}{\mathrm{Re}}\)

\( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

\( \newcommand{\Argument}{\mathrm{Arg}}\)

\( \newcommand{\norm}[1]{\| #1 \|}\)

\( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\AA}{\unicode[.8,0]{x212B}}\)

\( \newcommand{\vectorA}[1]{\vec{#1}} % arrow\)

\( \newcommand{\vectorAt}[1]{\vec{\text{#1}}} % arrow\)

\( \newcommand{\vectorB}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

\( \newcommand{\vectorC}[1]{\textbf{#1}} \)

\( \newcommand{\vectorD}[1]{\overrightarrow{#1}} \)

\( \newcommand{\vectorDt}[1]{\overrightarrow{\text{#1}}} \)

\( \newcommand{\vectE}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash{\mathbf {#1}}}} \)

Learning Objectives

By the end of this section, you will be able to:

Define skepticism as it is used in philosophy.
Compare and contrast global and local skepticism.
Offer and explain a skeptical hypothesis.
Outline the general structure of argument for global skepticism.

Global Skepticism

The Dream Argument

While he is dreaming he does not know it is a dream, and in his dream he may even try to interpret a dream. Only after he wakes does he know it was a dream. And someday there will be a great awakening when we know that this is all a great dream. Yet the stupid believe they are awake, busily and brightly assuming they understand things, calling this man ruler, that one herdsman—how dense! Confucius and you are both dreaming! And when I say you are dreaming, I am dreaming, too. (Zhuangzi 2003, 43)

Chinese ink drawing depicting a seated man, who appears to be asleep, with a butterfly hovering above his head.

The Evil Demon Argument

If I cannot rule out the possibility that an evil demon is tricking me, then I do not have any knowledge of the external world.
I cannot rule out the possibility that an evil demon is tricking me.
Therefore, I do not have knowledge of the external world.

Putnam’s Brain in a Vat

If you don’t like evil demons, then consider a more modern version of a skeptical hypothesis: the “brain in a vat” conceived of by American philosopher and mathematician Hilary Putnam (1926–2016). Imagine that while you were asleep last night, a group of scientists kidnapped you and took you to their lab. There, they surgically removed your brain and placed it in a vat of nutrients. The scientists then hooked up your brain to a sophisticated new computer system. They were able to download your memories so as to create new experiences. The result is a seamless experience of consciousness between yesterday and today. When you woke this morning, your life seemed to proceed without disruption. Can you prove that you are not a brain in a vat? No, you cannot. The scenario stipulates that your experience will seem exactly the same whether you are a brain in a vat or not. Other, similar skeptical scenarios are easy to come up with. Consider the possibility that you are caught in a virtual reality world or that you are trapped in the Matrix.

Sketch of a brain floating in a liquid-filled beaker, connected to a computer console by several electrodes. A thought bubble rising from the brain reads “I’m walking outside in the sun!!”

General Structure of Global Skeptical Arguments

Skeptical hypotheses and the arguments that they inspire all have a similar structure:

If I cannot rule out the possibility of SH, then I cannot be justified in believing that P.
I cannot rule out the possibility of SH.
Therefore, I cannot be justified in believing that P.

Responses to Global Skepticism

I am justified in believing that P.
Therefore, I can rule out the possibility of SH.

Contextualism

As we just saw, some theorists reject the notion that you must be certain of a belief—that is, rule out all possible defeaters—in order to have knowledge. Moore thinks he has more justification to believe he has two hands than he does that there’s an evil demon tricking him. And in determining whether I am justified in believing in the bird outside my office window, I rarely consider the possibility that I could be a brain in a vat. I’m more likely to focus on my poor vision as a defeater. In the context of bird identification, wild skeptical hypotheses seem out of place. Indeed, we often adjust how much justification we think is needed for a belief to the task at hand. Contextualism is the view that the truth of knowledge attributions depends on the context. Contextualism is a theory about knowledge and justification. When we attribute knowledge to a subject S, the truth of the knowledge claim depends on the context that S is in. The context of S determines the level of justification needed for a true belief to count as knowledge. Contextualism comes from the observation that the level of confidence needed for justification changes depending on what the belief is as well as its the purpose and its importance, among other things. We expect a high degree of justification from physicians when they diagnose disease but less justification from friends recalling the title of a movie because there’s much more at stake in medical diagnoses.

Skepticism in Specific Domains

Think Like A Philosopher

Your complimentary articles

You’ve read one of your four complimentary articles for this month.

You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please

Sceptical Hypotheses and Transcendental Arguments

Jonathan barfield presents a way to beat the sceptics..

What if the world doesn’t really exist? Perhaps what I think is the real world is actually an illusion, or a computer simulation? Maybe I am really a brain in a vat of liquid goo, connected up by an evil scientist to electrodes producing nerve impulses that pump a simulation of a real world directly into my brain (cf. The Matrix )? Or what if nothing exists outside my mind? Perhaps the only thing that exists is my mind? Is anything I think about the world actually true?

Such deep questions and worries about what we can know about the world are called sceptical hypotheses . A sceptical hypothesis challenges our everyday assumptions about what is real and how we can know it, pushing us to accept the possibility that we might not know anything , or at least that we don’t have very much justification for some of the things that we usually assume we know.

One way of trying to overcome a sceptical hypothesis is to construct a theory of knowledge which sets out the criteria that have to be met in order for me to know something. I can then test any given claim to see if meets the standards I have set myself. The aim in constructing this theory is that we will then be able to demonstrate that we can in fact know that the world exists – that it isn’t an illusion or computer simulation – and that any other radically sceptical hypothesis is also false.

This strategy can be tricky: if you make your theory of knowledge too strict, then no claim will meet your criteria, in which case it will show you that you know nothing; if you make your theory of knowledge too slack, then it will indicate that you know some things that you otherwise don’t think you know. Plus, this approach encourages lengthy debate about exactly what the criteria for knowledge should be, how we should apply the criteria, and how we could be certain that the criteria have been met.

A different way to overcome a sceptical hypothesis is to use a transcendental argument . Transcendental arguments try to show that there is no way that the sceptical hypothesis could be true. That is, they try to show that a sceptical hypothesis can only be one of two things: it is either false, or it doesn’t make any sense in the first place. In either case, the hypothesis cannot be true. And if the sceptical hypothesis cannot be true, then we no longer need to worry about it! A transcendental argument does not beat the sceptical hypothesis directly by accepting the challenge to try and disprove it by providing a theory in response to it. Rather, a transcendental argument overcomes the sceptical hypothesis indirectly , by undermining its basis. It does this by showing that the hypothesis cannot be formulated without relying on the hypothesis being false.

Let’s look at two transcendental arguments which try to overcome sceptical hypotheses: Hilary Putnam trying to overcome the hypothesis that the external world might be a computer simulation; and Saul Kripke trying to overcome the hypothesis that perhaps nothing exists outside my mind.

Putnam’s Brain In A Vat

Perhaps what I think is the world is actually a computer simulation or other illusion? Maybe I am really a brain in a vat of goo connected up to electrodes that pump a simulation of a real world directly into my brain?

In Reason, Truth and History (1981), Hilary Putnam offers a transcendental argument to overcome the sceptical hypothesis that I might be a brain in a vat, or more generally, that what I think is the external world is actually some sort of simulation. Putnam’s transcendental argument is based on a theory about how our words work.

First, let’s suppose I am not a brain in a vat. In this situation, my words refer to real objects, and this reference to real objects is part of what those words mean. For example, my word ‘vat’ (usually) refers to real vats external to me, and this referring to real vats is part of what the word ‘vat’ means. So if I am not a brain in a vat, I can meaningfully say or think the hypothesis ‘I might be a brain in a vat’; but of course, in this situation we have stipulated that I am not a brain in a vat, so the hypothesis is false.

Now let’s suppose instead that I am a brain in a vat. In this situation, my words don’t mean what I think they mean: they don’t refer to the objects that I think they refer to, and my language doesn’t work the way I think it does, so I am consistently wrong about what I think my words mean. For example, in this situation – in the brain-in-a-vat setup – my word ‘vat’ can’t refer to any real vat external to me because I have never encountered any such object, only simulations or illusions of vats. Indeed, in this situation, my experiences of what I take to be vats are not of vats at all, but rather the result of calculated electrical stimulations sent directly into my brain. In this situation, my word ‘vat’ must then refer to some set of the electrical stimulation impulses sent into my brain. In fact, all of my words would be referring to such electrical impulses, and would never refer to real objects external to me, such as vats, because that wouldn’t be how my words could refer. So according to Putnam, in a situation where I was a brain in a vat, I could not truly or meaningfully say or think ‘I am a brain in a vat’, because my words only refer to electrical impulses, and not to real brains or real vats.

So in the first alternative, I am not a brain in a vat, but I can meaningfully, and falsely, say I am one. In the second alternative, I cannot meaningfully say or think that I am a brain in a vat because I cannot refer to real brains or vats, because my language doesn’t work in the way necessary for it to be meaningful. So, either I am not a brain in a vat, or I cannot coherently say or think that I am a brain in a vat; so either the sceptical hypothesis is false or the sceptical hypothesis doesn’t make any sense. Either way, the sceptical hypothesis that I might be a brain in a vat isn’t true. Putnam has therefore shown that the sceptical hypothesis cannot be asserted without relying on the hypothesis being false, as the sceptical hypothesis must be false in order for it to make sense. The general form of Putnam’s argument is that in order for our words, and therefore the sceptical hypothesis itself, to make sense, the sceptical hypothesis cannot be true: we cannot question whether the external world is real without assuming that the external world is real. If it makes sense for me to wonder whether I am a brain in a vat, then I am not one.

Kripke’s Language Checking

Perhaps nothing exists outside my mind? Maybe the only thing that exists is my mind?

In Wittgenstein on Rules and Private Language (1982), Saul Kripke constructs a transcendental argument inspired by the later Ludwig Wittgenstein to overcome the sceptical hypothesis that there might be nothing that exists outside my mind, and that the only thing that really exists is my mind.

Kripke poses the Wittgensteinian question: How can I be sure that I am using my language consistently? For example, how can I be sure that I am using a word, such as ‘table’, the same way today as I was last week, and so be sure that my language use has a consistent meaning? Kripke then uses the Wittgensteinian argument that one of the main ways we can be sure of consistency in our language use is that other speakers of the language correct us if we use words incorrectly or inconsistently. In effect, the people I communicate with provide a checking service to make sure that I am using my words in the same way now as in the past, by ensuring that I am using words in the standard and consistent communal or public manner.

You might object that a far more useful way of checking that I am using my words in the same way now as in the past is to use my memory. However, as the Kripkean/Wittgensteinian argument goes, if I use my memory to check my word use, I also need a way of checking the reliability of my memory; and I can hardly check the reliability of my memory by using my memory. I will need another method. For example, if I want to catch a train, I might bring to mind a mental image of the train timetable and look at this image in my mind to find out what time the train will arrive. My memory of the train timetable might be reliable or it might not be: but the only way I can check the reliability of that memory is not to consult another memory, which may be equally fallible, but to check with a real physical train timetable. Or let’s suppose I wanted to keep a record of how frequently I experience a particular emotion – for example, a really specific feeling of loving affection. I resolve to write the letter ‘S’ in my diary whenever I experience this particular loving feeling. Having now done this for a few weeks, how can I be sure that each ‘S’ that I recorded refers to the same particular emotion? How can I be sure that some of the feelings weren’t significantly different? The only way I can check that each emotional experience was the same is to use my memory – but how am I to be sure that my memory is accurate and that I have indeed recorded ‘S’ consistently? I need a method of checking that my memory is reliable, but there does not seem to be a way of checking that my memory of using ‘S’ is reliable like I had in the train timetable example, so if I am only relying on my memory I have no way of checking I am using ‘S’ consistently.

Although I may use my memory to remember how I use words, the only way I can check that my memory is reliable (and hence that I am using my words in the same way now as in the past) is by consulting something outside my memory, such as other speakers of the language. Ultimately, I need some method of checking that I am using my words consistently that doesn’t rely solely on my memory.

How does this discussion about ensuring that I am using my language consistently help to overcome the sceptical hypothesis that there might be nothing that exists outside my mind, and that the only thing that really exists is my mind?

Well, either there is a world outside my mind including other language-speakers, or I have no means of ensuring that I am using my language consistently. However, in the latter case, I don’t know for sure the meaning of my own words – which, among other things, means that I cannot coherently talk or think about whether there is no world outside my mind. So either the sceptical hypothesis that there might not be a world outside my mind is false, or the sceptical hypothesis doesn’t make any sense, due to my not having a way of being sure that I know the meaning of my words. Either way, the sceptical hypothesis that there is nothing outside my mind isn’t true, which is what we were trying to prove. Thus Kripke also shows that this sceptical hypothesis cannot be made without relying on the hypothesis being false: I cannot question whether there is a world outside my mind without assuming that there is a world outside my mind, in order for my words, and therefore the sceptical hypothesis itself, to make sense. The sceptical hypothesis must be false in order for it to make sense. Or, if it makes sense for me to wonder whether there might not be a world outside my mind, then there is.

Kripke’s solution might perhaps inspire a further sceptical hypothesis: maybe there really is no way of checking that I am using my words consistently, and so I don’t know for certain the meaning of my own words; and so maybe I really am just talking nonsense. But we can construct a further transcendental argument to overcome this objection. I cannot meaningfully question whether I am talking nonsense without assuming that I am talking sense; so the further sceptical objection that I am talking nonsense must be false in order for it to make sense. Or, if it makes sense for me to wonder whether I am talking sense or not, then I am.

Jonathan Barfield is Head of Theory of Knowledge at Brentwood School, Essex, UK. Please visit mrjbarfield.wikispaces.com or tweet @mrjbarfield .

This site uses cookies to recognize users and allow us to analyse site usage. By continuing to browse the site with cookies enabled in your browser, you consent to the use of cookies in accordance with our privacy policy . X

Search Menu
Browse content in Arts and Humanities
Browse content in Archaeology
Anglo-Saxon and Medieval Archaeology
Archaeological Methodology and Techniques
Archaeology by Region
Archaeology of Religion
Archaeology of Trade and Exchange
Biblical Archaeology
Contemporary and Public Archaeology
Environmental Archaeology
Historical Archaeology
History and Theory of Archaeology
Industrial Archaeology
Landscape Archaeology
Mortuary Archaeology
Prehistoric Archaeology
Underwater Archaeology
Urban Archaeology
Zooarchaeology
Browse content in Architecture
Architectural Structure and Design
History of Architecture
Residential and Domestic Buildings
Theory of Architecture
Browse content in Art
Art Subjects and Themes
History of Art
Industrial and Commercial Art
Theory of Art
Biographical Studies
Byzantine Studies
Browse content in Classical Studies
Classical History
Classical Philosophy
Classical Mythology
Classical Literature
Classical Reception
Classical Art and Architecture
Classical Oratory and Rhetoric
Greek and Roman Papyrology
Greek and Roman Epigraphy
Greek and Roman Law
Greek and Roman Archaeology
Late Antiquity
Religion in the Ancient World
Digital Humanities
Browse content in History
Colonialism and Imperialism
Diplomatic History
Environmental History
Genealogy, Heraldry, Names, and Honours
Genocide and Ethnic Cleansing
Historical Geography
History by Period
History of Emotions
History of Agriculture
History of Education
History of Gender and Sexuality
Industrial History
Intellectual History
International History
Labour History
Legal and Constitutional History
Local and Family History
Maritime History
Military History
National Liberation and Post-Colonialism
Oral History
Political History
Public History
Regional and National History
Revolutions and Rebellions
Slavery and Abolition of Slavery
Social and Cultural History
Theory, Methods, and Historiography
Urban History
World History
Browse content in Language Teaching and Learning
Language Learning (Specific Skills)
Language Teaching Theory and Methods
Browse content in Linguistics
Applied Linguistics
Cognitive Linguistics
Computational Linguistics
Forensic Linguistics
Grammar, Syntax and Morphology
Historical and Diachronic Linguistics
History of English
Language Evolution
Language Reference
Language Acquisition
Language Variation
Language Families
Lexicography
Linguistic Anthropology
Linguistic Theories
Linguistic Typology
Phonetics and Phonology
Psycholinguistics
Sociolinguistics
Translation and Interpretation
Writing Systems
Browse content in Literature
Bibliography
Children's Literature Studies
Literary Studies (Romanticism)
Literary Studies (American)
Literary Studies (Asian)
Literary Studies (European)
Literary Studies (Eco-criticism)
Literary Studies (Modernism)
Literary Studies - World
Literary Studies (1500 to 1800)
Literary Studies (19th Century)
Literary Studies (20th Century onwards)
Literary Studies (African American Literature)
Literary Studies (British and Irish)
Literary Studies (Early and Medieval)
Literary Studies (Fiction, Novelists, and Prose Writers)
Literary Studies (Gender Studies)
Literary Studies (Graphic Novels)
Literary Studies (History of the Book)
Literary Studies (Plays and Playwrights)
Literary Studies (Poetry and Poets)
Literary Studies (Postcolonial Literature)
Literary Studies (Queer Studies)
Literary Studies (Science Fiction)
Literary Studies (Travel Literature)
Literary Studies (War Literature)
Literary Studies (Women's Writing)
Literary Theory and Cultural Studies
Mythology and Folklore
Shakespeare Studies and Criticism
Browse content in Media Studies
Browse content in Music
Applied Music
Dance and Music
Ethics in Music
Ethnomusicology
Gender and Sexuality in Music
Medicine and Music
Music Cultures
Music and Media
Music and Religion
Music and Culture
Music Education and Pedagogy
Music Theory and Analysis
Musical Scores, Lyrics, and Libretti
Musical Structures, Styles, and Techniques
Musicology and Music History
Performance Practice and Studies
Race and Ethnicity in Music
Sound Studies
Browse content in Performing Arts
Browse content in Philosophy
Aesthetics and Philosophy of Art
Epistemology
Feminist Philosophy
History of Western Philosophy
Metaphysics
Moral Philosophy
Non-Western Philosophy
Philosophy of Language
Philosophy of Mind
Philosophy of Perception
Philosophy of Science
Philosophy of Action
Philosophy of Law
Philosophy of Religion
Philosophy of Mathematics and Logic
Practical Ethics
Social and Political Philosophy
Browse content in Religion
Biblical Studies
Christianity
East Asian Religions
History of Religion
Judaism and Jewish Studies
Qumran Studies
Religion and Education
Religion and Health
Religion and Politics
Religion and Science
Religion and Law
Religion and Art, Literature, and Music
Religious Studies
Browse content in Society and Culture
Cookery, Food, and Drink
Cultural Studies
Customs and Traditions
Ethical Issues and Debates
Hobbies, Games, Arts and Crafts
Lifestyle, Home, and Garden
Natural world, Country Life, and Pets
Popular Beliefs and Controversial Knowledge
Sports and Outdoor Recreation
Technology and Society
Travel and Holiday
Visual Culture
Browse content in Law
Arbitration
Browse content in Company and Commercial Law
Commercial Law
Company Law
Browse content in Comparative Law
Systems of Law
Competition Law
Browse content in Constitutional and Administrative Law
Government Powers
Judicial Review
Local Government Law
Military and Defence Law
Parliamentary and Legislative Practice
Construction Law
Contract Law
Browse content in Criminal Law
Criminal Procedure
Criminal Evidence Law
Sentencing and Punishment
Employment and Labour Law
Environment and Energy Law
Browse content in Financial Law
Banking Law
Insolvency Law
History of Law
Human Rights and Immigration
Intellectual Property Law
Browse content in International Law
Private International Law and Conflict of Laws
Public International Law
IT and Communications Law
Jurisprudence and Philosophy of Law
Law and Politics
Law and Society
Browse content in Legal System and Practice
Courts and Procedure
Legal Skills and Practice
Primary Sources of Law
Regulation of Legal Profession
Medical and Healthcare Law
Browse content in Policing
Criminal Investigation and Detection
Police and Security Services
Police Procedure and Law
Police Regional Planning
Browse content in Property Law
Personal Property Law
Study and Revision
Terrorism and National Security Law
Browse content in Trusts Law
Wills and Probate or Succession
Browse content in Medicine and Health
Browse content in Allied Health Professions
Arts Therapies
Clinical Science
Dietetics and Nutrition
Occupational Therapy
Operating Department Practice
Physiotherapy
Radiography
Speech and Language Therapy
Browse content in Anaesthetics
General Anaesthesia
Neuroanaesthesia
Clinical Neuroscience
Browse content in Clinical Medicine
Acute Medicine
Cardiovascular Medicine
Clinical Genetics
Clinical Pharmacology and Therapeutics
Dermatology
Endocrinology and Diabetes
Gastroenterology
Genito-urinary Medicine
Geriatric Medicine
Infectious Diseases
Medical Toxicology
Medical Oncology
Pain Medicine
Palliative Medicine
Rehabilitation Medicine
Respiratory Medicine and Pulmonology
Rheumatology
Sleep Medicine
Sports and Exercise Medicine
Community Medical Services
Critical Care
Emergency Medicine
Forensic Medicine
Haematology
History of Medicine
Browse content in Medical Skills
Clinical Skills
Communication Skills
Nursing Skills
Surgical Skills
Browse content in Medical Dentistry
Oral and Maxillofacial Surgery
Paediatric Dentistry
Restorative Dentistry and Orthodontics
Surgical Dentistry
Medical Ethics
Medical Statistics and Methodology
Browse content in Neurology
Clinical Neurophysiology
Neuropathology
Nursing Studies
Browse content in Obstetrics and Gynaecology
Gynaecology
Occupational Medicine
Ophthalmology
Otolaryngology (ENT)
Browse content in Paediatrics
Neonatology
Browse content in Pathology
Chemical Pathology
Clinical Cytogenetics and Molecular Genetics
Histopathology
Medical Microbiology and Virology
Patient Education and Information
Browse content in Pharmacology
Psychopharmacology
Browse content in Popular Health
Caring for Others
Complementary and Alternative Medicine
Self-help and Personal Development
Browse content in Preclinical Medicine
Cell Biology
Molecular Biology and Genetics
Reproduction, Growth and Development
Primary Care
Professional Development in Medicine
Browse content in Psychiatry
Addiction Medicine
Child and Adolescent Psychiatry
Forensic Psychiatry
Learning Disabilities
Old Age Psychiatry
Psychotherapy
Browse content in Public Health and Epidemiology
Epidemiology
Public Health
Browse content in Radiology
Clinical Radiology
Interventional Radiology
Nuclear Medicine
Radiation Oncology
Reproductive Medicine
Browse content in Surgery
Cardiothoracic Surgery
Gastro-intestinal and Colorectal Surgery
General Surgery
Neurosurgery
Paediatric Surgery
Peri-operative Care
Plastic and Reconstructive Surgery
Surgical Oncology
Transplant Surgery
Trauma and Orthopaedic Surgery
Vascular Surgery
Browse content in Science and Mathematics
Browse content in Biological Sciences
Aquatic Biology
Biochemistry
Bioinformatics and Computational Biology
Developmental Biology
Ecology and Conservation
Evolutionary Biology
Genetics and Genomics
Microbiology
Molecular and Cell Biology
Natural History
Plant Sciences and Forestry
Research Methods in Life Sciences
Structural Biology
Systems Biology
Zoology and Animal Sciences
Browse content in Chemistry
Analytical Chemistry
Computational Chemistry
Crystallography
Environmental Chemistry
Industrial Chemistry
Inorganic Chemistry
Materials Chemistry
Medicinal Chemistry
Mineralogy and Gems
Organic Chemistry
Physical Chemistry
Polymer Chemistry
Study and Communication Skills in Chemistry
Theoretical Chemistry
Browse content in Computer Science
Artificial Intelligence
Computer Architecture and Logic Design
Game Studies
Human-Computer Interaction
Mathematical Theory of Computation
Programming Languages
Software Engineering
Systems Analysis and Design
Virtual Reality
Browse content in Computing
Business Applications
Computer Security
Computer Games
Computer Networking and Communications
Digital Lifestyle
Graphical and Digital Media Applications
Operating Systems
Browse content in Earth Sciences and Geography
Atmospheric Sciences
Environmental Geography
Geology and the Lithosphere
Maps and Map-making
Meteorology and Climatology
Oceanography and Hydrology
Palaeontology
Physical Geography and Topography
Regional Geography
Soil Science
Urban Geography
Browse content in Engineering and Technology
Agriculture and Farming
Biological Engineering
Civil Engineering, Surveying, and Building
Electronics and Communications Engineering
Energy Technology
Engineering (General)
Environmental Science, Engineering, and Technology
History of Engineering and Technology
Mechanical Engineering and Materials
Technology of Industrial Chemistry
Transport Technology and Trades
Browse content in Environmental Science
Applied Ecology (Environmental Science)
Conservation of the Environment (Environmental Science)
Environmental Sustainability
Environmentalist Thought and Ideology (Environmental Science)
Management of Land and Natural Resources (Environmental Science)
Natural Disasters (Environmental Science)
Nuclear Issues (Environmental Science)
Pollution and Threats to the Environment (Environmental Science)
Social Impact of Environmental Issues (Environmental Science)
History of Science and Technology
Browse content in Materials Science
Ceramics and Glasses
Composite Materials
Metals, Alloying, and Corrosion
Nanotechnology
Browse content in Mathematics
Applied Mathematics
Biomathematics and Statistics
History of Mathematics
Mathematical Education
Mathematical Finance
Mathematical Analysis
Numerical and Computational Mathematics
Probability and Statistics
Pure Mathematics
Browse content in Neuroscience
Cognition and Behavioural Neuroscience
Development of the Nervous System
Disorders of the Nervous System
History of Neuroscience
Invertebrate Neurobiology
Molecular and Cellular Systems
Neuroendocrinology and Autonomic Nervous System
Neuroscientific Techniques
Sensory and Motor Systems
Browse content in Physics
Astronomy and Astrophysics
Atomic, Molecular, and Optical Physics
Biological and Medical Physics
Classical Mechanics
Computational Physics
Condensed Matter Physics
Electromagnetism, Optics, and Acoustics
History of Physics
Mathematical and Statistical Physics
Measurement Science
Nuclear Physics
Particles and Fields
Plasma Physics
Quantum Physics
Relativity and Gravitation
Semiconductor and Mesoscopic Physics
Browse content in Psychology
Affective Sciences
Clinical Psychology
Cognitive Psychology
Cognitive Neuroscience
Criminal and Forensic Psychology
Developmental Psychology
Educational Psychology
Evolutionary Psychology
Health Psychology
History and Systems in Psychology
Music Psychology
Neuropsychology
Organizational Psychology
Psychological Assessment and Testing
Psychology of Human-Technology Interaction
Psychology Professional Development and Training
Research Methods in Psychology
Social Psychology
Browse content in Social Sciences
Browse content in Anthropology
Anthropology of Religion
Human Evolution
Medical Anthropology
Physical Anthropology
Regional Anthropology
Social and Cultural Anthropology
Theory and Practice of Anthropology
Browse content in Business and Management
Business Ethics
Business Strategy
Business History
Business and Technology
Business and Government
Business and the Environment
Comparative Management
Corporate Governance
Corporate Social Responsibility
Entrepreneurship
Health Management
Human Resource Management
Industrial and Employment Relations
Industry Studies
Information and Communication Technologies
International Business
Knowledge Management
Management and Management Techniques
Operations Management
Organizational Theory and Behaviour
Pensions and Pension Management
Public and Nonprofit Management
Strategic Management
Supply Chain Management
Browse content in Criminology and Criminal Justice
Criminal Justice
Criminology
Forms of Crime
International and Comparative Criminology
Youth Violence and Juvenile Justice
Development Studies
Browse content in Economics
Agricultural, Environmental, and Natural Resource Economics
Asian Economics
Behavioural Finance
Behavioural Economics and Neuroeconomics
Econometrics and Mathematical Economics
Economic History
Economic Systems
Economic Methodology
Economic Development and Growth
Financial Markets
Financial Institutions and Services
General Economics and Teaching
Health, Education, and Welfare
History of Economic Thought
International Economics
Labour and Demographic Economics
Law and Economics
Macroeconomics and Monetary Economics
Microeconomics
Public Economics
Urban, Rural, and Regional Economics
Welfare Economics
Browse content in Education
Adult Education and Continuous Learning
Care and Counselling of Students
Early Childhood and Elementary Education
Educational Equipment and Technology
Educational Strategies and Policy
Higher and Further Education
Organization and Management of Education
Philosophy and Theory of Education
Schools Studies
Secondary Education
Teaching of a Specific Subject
Teaching of Specific Groups and Special Educational Needs
Teaching Skills and Techniques
Browse content in Environment
Applied Ecology (Social Science)
Climate Change
Conservation of the Environment (Social Science)
Environmentalist Thought and Ideology (Social Science)
Natural Disasters (Environment)
Social Impact of Environmental Issues (Social Science)
Browse content in Human Geography
Cultural Geography
Economic Geography
Political Geography
Browse content in Interdisciplinary Studies
Communication Studies
Museums, Libraries, and Information Sciences
Browse content in Politics
African Politics
Asian Politics
Chinese Politics
Comparative Politics
Conflict Politics
Elections and Electoral Studies
Environmental Politics
European Union
Foreign Policy
Gender and Politics
Human Rights and Politics
Indian Politics
International Relations
International Organization (Politics)
International Political Economy
Irish Politics
Latin American Politics
Middle Eastern Politics
Political Behaviour
Political Economy
Political Institutions
Political Sociology
Political Methodology
Political Communication
Political Philosophy
Political Theory
Politics and Law
Public Policy
Public Administration
Quantitative Political Methodology
Regional Political Studies
Russian Politics
Security Studies
State and Local Government
UK Politics
US Politics
Browse content in Regional and Area Studies
African Studies
Asian Studies
East Asian Studies
Japanese Studies
Latin American Studies
Middle Eastern Studies
Native American Studies
Scottish Studies
Browse content in Research and Information
Research Methods
Browse content in Social Work
Addictions and Substance Misuse
Adoption and Fostering
Care of the Elderly
Child and Adolescent Social Work
Couple and Family Social Work
Developmental and Physical Disabilities Social Work
Direct Practice and Clinical Social Work
Emergency Services
Human Behaviour and the Social Environment
International and Global Issues in Social Work
Mental and Behavioural Health
Social Justice and Human Rights
Social Policy and Advocacy
Social Work and Crime and Justice
Social Work Macro Practice
Social Work Practice Settings
Social Work Research and Evidence-based Practice
Welfare and Benefit Systems
Browse content in Sociology
Childhood Studies
Community Development
Comparative and Historical Sociology
Economic Sociology
Gender and Sexuality
Gerontology and Ageing
Health, Illness, and Medicine
Marriage and the Family
Migration Studies
Occupations, Professions, and Work
Organizations
Population and Demography
Race and Ethnicity
Social Theory
Social Movements and Social Change
Social Research and Statistics
Social Stratification, Inequality, and Mobility
Sociology of Religion
Sociology of Education
Sport and Leisure
Urban and Rural Studies
Browse content in Warfare and Defence
Defence Strategy, Planning, and Research
Land Forces and Warfare
Military Administration
Military Life and Institutions
Naval Forces and Warfare
Other Warfare and Defence Issues
Peace Studies and Conflict Resolution
Weapons and Equipment

< Previous chapter
Next chapter >

10 Live Skeptical Hypotheses

Bryan Frances is associate professor of philosophy at Fordham University.

Published: 02 September 2009
Cite Icon Cite
Permissions Icon Permissions

This article examines some examples of live forms of skeptical hypotheses. It highlights some objections to the live skeptic's argument. These objections fall into two classes. First, there are reasons for thinking that we do not have to be able to rule out the error theories in order to know truths obviously inconsistent with them, and second, one can admit the ruling-out requirement but argue that it is actually pretty easy to do the ruling out. This article also discusses the liveness-mortality premise, the nature of live skepticism, and the real and academic threats of skepticism.

1. Skeptical Threats, Real and Merely Academic

Those of us who take skepticism seriously typically have two relevant beliefs: (a) it is plausible (even if false) that in order to know that I have hands, I have to be able to epistemically neutralize, to some significant degree, some skeptical hypotheses, such as the brain‐in‐a‐vat (BIV) one; and (b) it is also plausible (even if false) that I cannot so neutralize those hypotheses. There is no reason for us to think also (c) that the BIV hypothesis, for instance, is plausible or probably true . In order to take skepticism seriously, it is sufficient to hold (a) and (b); one need not hold (c). Indeed, philosophers who accept (a) and (b) never endorse (c). Show me a philosopher who suspects that he is a brain in a vat, and I will show you someone who is deranged.

That is one thing that bothers undergraduates in philosophy. They object: why on earth do some philosophers take the BIV hypothesis to pose any threat at all to our beliefs, given that those very same philosophers think that there is no real chance that the BIV hypothesis is true? Sure, the BIV hypothesis is formally inconsistent with my belief that I have hands, so if the former is true, then my belief is false. But so what? Why should that bare inconsistency matter so much? Is this strange attitude among philosophers the result of some logic fetish infecting the philosophical community? It is sometimes said that the skeptical hypotheses not only are inconsistent with our beliefs but also are explanatory of our experiences, which is supposed to make them more of a threat. But students are not fooled: although the skeptical hypotheses may attempt to explain why our experience is as it is, it is the kind of attempt appropriate for science‐fiction movies that are all special effects and virtually no plot. No one with any sense of reality will take the evil‐demon hypothesis to be even tenuously explanatory.

The students would understand the fuss over the BIV hypothesis if there were some decent reason to think that the BIV hypothesis was really true. If you believe P , a contrary hypothesis Q has some reasonably good backing—perhaps endorsement by legitimate experts in the relevant field—and you are quite familiar with Q 's good status, as well as the conflict between P and Q , then the Q possibility does seem to mount a significant threat to one's belief in P , where a threat is significant just in case when left unneutralized, it ruins one's chance at knowledge of P 's truth. If the BIV or evil‐demon hypotheses were like Q , then we would have a real threat to our belief that we have hands.

Here is an example to which I will appeal later in this chapter as well. When Jo was a teenager, she learned the theory that a huge meteorite wiped out the dinosaurs. She learned this theory in the usual way, hearing it from her parents and teachers and reading about it in books. Now pretend that at the time she was told the meteor story as a child, say, at the age of eight, the scientific community was sharply divided on the issue of what caused the demise of the dinosaurs. Although most scientists accepted the meteor hypothesis, many others subscribed to the idea that their death was caused by some enormous solar flare. A significant number of other scientists thought that it was not a solar flare or a meteor but a particularly nasty series of supervolcanos. These latter two classes of dissenters had decent evidence concerning the sun and supervolcanos that the meteor advocates took seriously. Both the solar‐flare theorists and the supervolcano theorists were professors who were highly respected by the meteor theorists and at the top of their profession. Whole book series, conferences, and doctoral dissertations were devoted to these competing hypotheses. Suppose further that upon going to her university, Jo found out about the rival and highly respected hypotheses. She did not understand all the reasons why they were so well respected and endorsed, but she was well aware that they were well respected and frequently endorsed by the experts, even the best among them. Even so, she kept her meteor belief.

As it turned out, the meteor hypothesis was the right one; lucky Jo! But although some experts may have known that fact, surely Jo was in no position to know it after she found out about the eminent status of the rival hypotheses. She could do little or nothing to defeat the rival hypotheses. Even after studying the issues as an undergraduate, she could not know that a meteor wiped out the dinosaurs, for from the moment she first studied the dinosaurs until she graduated with a bachelor's degree in paleontology, she was perfectly aware that there were two “live” contrary hypotheses that she was in no position to rule out—and they had not been ruled out for her, either, for instance, by her teachers. And if she was aware of these alternative hypotheses, as well as their liveness and how they conflict with the meteor hypothesis, then as long as the competing hypotheses were not ruled out, she could not know that the meteor hypothesis was correct. Perhaps in order to know the meteor truth, she would not have needed to rule out the possibility that she was a brain in a vat. And perhaps someone mostly outside the scientific community and thus not exposed to the ultimately misleading supervolcano and solar‐flare evidence could know that the meteor hypothesis is true. Still, in order for Jo to know the meteor fact, she does have to be able to rule out, at least to some degree, the solar‐flare and supervolcano possibilities, for, unlike the crazy philosophical hypotheses, these are “real, live possibilities,” and she is perfectly aware of their existence, live status, and inconsistency with her belief. Perhaps she does not have to completely demolish the supervolcano and solar‐flare hypotheses in order to know that the meteor story is true, but she certainly has to knock them down a few epistemic notches.

The heart of the dinosaur argument is simple: because the supervolcano and solar‐flare hypotheses are real, live possibilities inconsistent with the meteor hypothesis, Jo is aware of all that, she is nothing even approaching an expert or genius on these matters, and those hypotheses are not ruled out, she does not know that the meteor story is true. There are not many ideas in philosophy that cannot be coherently and cleverly challenged, but it is a brave philosopher who thinks that Jo's story is misguided. Surely, one would think, in those circumstances Jo's true and partially justified belief in the meteor hypothesis did not amount to knowledge.

Let us return to our students' skeptical attitude toward skepticism. If we could find some skeptical hypotheses that were false but live in something like the way the supervolcano and solar‐flare hypotheses were, then we would think that they pose a significant threat to our beliefs that are obviously inconsistent with them: if we could not epistemically neutralize those skeptical hypotheses (whatever neutralization amounts to), then our beliefs obviously inconsistent with those false but live hypotheses would fall short of knowledge (even if they were true beliefs). At least, this lack of knowledge would follow if the people in question were well aware of the live status of the skeptical hypotheses, as well as the inconsistency between their beliefs and the live skeptical hypotheses. For in that case, we would have the threat to our beliefs staring us right in the face, as it were. The threat would be out in the open. Our students would understand our skeptical worries better.

Are there any such skeptical hypotheses—live hypotheses inconsistent with huge portions of our commonsensical beliefs? It is easy to adjust the BIV story so that it becomes live in another possible world. BIVs become technologically and fiscally possible; a team of mad scientists ends up ruling the world; certain top secret documents detailing plans for massive kidnapping and BIVing are discovered and leaked to the public; impenetrable 100,000‐square‐mile complexes of laboratories are set up in the wilds of Canada, Brazil, and elsewhere; thousands of people start mysteriously disappearing in the middle of the night to become BIVed. You can see how at least one BIV hypothesis—“I have become a BIV as the result of these crazy mad scientists”—could become something close to a real, live possibility for almost any person (not the mad scientists themselves, provided we tell the story right). Even if you are never BIVed in that world, does your true belief that you have a bicycle in your garage amount to knowledge?

The students will not be satisfied with this skepticism, not fully, anyway. For we have yet to find any real skeptical threat. No actual version of skepticism is on the horizon. None of our students' actual beliefs are threatened by any skeptical hypothesis. Skepticism remains merely academic and highly dubious. They are still skeptical about skepticism.

However, there are skeptical hypotheses that are actually live; others are at least very nearly actually live. In this chapter , I will list some of these hypotheses and show why they do generate a “real” skeptical threat, one quite different from the usual, merely academic threats. The skeptical theory that results I call live skepticism . 1

2. The Live Skeptic's Argument

The live skeptic uses an argument just like the dinosaur one: if you believe P , hypothesis H inconsistent with P is a live hypothesis, and you are like Jo in being a well‐informed “mere mortal” with regard to H, then you do not know P . Let us make the argument moderately precise.

We start with a sufficient condition for a hypothesis H to be a “real, live” socioepistemic possibility in our intellectual community.

In our intellectual community, hypothesis H has been through a significant (not to say exhaustive) evaluation by experts over many years.

It is judged actually true or about as likely as any relevant possibility by a significant number of well‐informed, well‐respected, and highly intelligent experts in the field(s) H belongs to. Also assume that there are lots of these experts, and they are not crackpots.

Those experts reached that favorable opinion based on H 's merits in a familiar, epistemically responsible way (that is, the way they reached their opinion seems as epistemically responsible as any).

Those experts consider that there are several decent and independent sources of evidence for H , so it is not the case that the only reason people pay serious attention to H is the presence of one weird experiment or line of reasoning.

Many of those experts consider H a “real, live possibility” (i.e., this is what they would say if you asked them).

The second task is to articulate a set of sufficient conditions for being a well‐informed mere mortal with respect to a hypothesis H .

You know that the hypothesis H is inconsistent with P (where you believe P ).

You are at least somewhat familiar with H and the issues surrounding H , including the fact that H is live in the sense of (1)–(5). (Thus [7] entails [1]–[5].) In fact, you are as aware as just about anyone of the fact that H is live.

Your intelligence, understanding, and knowledge are not extraordinary for people in your intellectual community with regard to H or the issues surrounding H . (So although you may be an expert, you are certainly no better than the other experts—in particular, the ones who insist that H is or could very well be true.)

If you have any reasons or evidence you can marshal for casting doubt on H , and if they were carefully considered by the members of that community—in particular, by the well‐informed, well‐respected, and highly intelligent experts who were thoroughly familiar with the hypothesis—they would be nearly universally and confidently rejected as clearly insufficient to rule out the hypothesis (although they may have other merits).

Roughly put, “S is as aware as just about anyone that X ” in (7) means that X is true and S has all the usual good warrant for believing X , warrant that usually suffices for knowledge. I cannot say that it entails knowledge of X , however, for reasons that will become clear later. Briefly, the reason is that even though S has the usual “positive” warrant sufficient for knowledge of X , she also has some “negative” warrant that offsets or vetoes that positive warrant enough so that her belief in X no longer is, on balance, warranted or sufficient for knowledge.

The conjunction of (6) and (7) (which, as noted in [7], includes by entailment [1]–[5]) says, very roughly, “You are fully aware of what is going on with the live hypothesis, so it is a real threat to your knowledge, a threat that must be defused.” The conjunction of (8) and (9) says, again roughly, “You do not have what it takes to defuse the threat.”

Please note the strength of (9): it is not merely saying that you cannot convince experts that H is false. Condition (9) does not just mean that you lack reasons that would be accepted as establishing beyond serious dispute that H is false. It is saying that even the experts who reject H would say that your reasons for rejecting H are clearly inadequate by just about anyone's lights and not just by the lights of those who accept H . For instance, you might be a graduate student in paleontology who is aware of the rival hypotheses about the demise of the dinosaurs and who happens to believe the true meteor hypothesis. You go to see your PhD supervisor, and she asks you what you plan to say about the supervolcano hypothesis in your dissertation. You say that that theory is not very plausible, but you are happy to throw in a brief section showing why it is implausible. She agrees with you that the meteor hypothesis is correct, but she asks you what you plan to say against the supervolcano hypothesis. You give your spiel, and she tells you flat out that what you have said is clearly inadequate and you should either do much better with a critical section or drop it entirely and say in a footnote that you will be merely assuming the falsehood of the supervolcano theory. After all, professor so‐and‐so right down the hall is an advocate of the supervolcano theory, he is certainly no dope, he is not alone in his expert opinion, and you have said virtually nothing to put any pressure on his view.

The graduate student is a well‐informed mere mortal, “mere mortal” for short. A child is not, because she fails to satisfy either (6) or (7). Another kind of mere mortal is an expert in the field but whose specialization lies elsewhere. Professor Smith teaches various science classes. She is perfectly aware of the supervolcano and solar‐flare hypotheses but would not be able to say anything interesting against them. She has the true meteor belief, but like the graduate student, her belief is too lucky to amount to knowledge. That is the type of person I have in mind as a well‐informed mere mortal; my hope is that conditions (6)–(9) capture the important aspects of their epistemic position. Needless to say, there could be many such individuals.

The live skeptic now makes three key claims.

The modesty principle : if S is a mere mortal with respect to live hypothesis H (so conditions (1)–(9) hold), then H is not ruled out with respect to S. The live‐hypothesis principle : if S knows that P entails ¬ H and is as aware as just about anyone that H is live (so H really is live), then if H is not ruled out with respect to S, S does not know P . The liveness‐mortality premise : in the actual world or an extremely realistic, very close possible world, many members of our actual intellectual community of contemporary analytic philosophers, cognitive scientists, and their advanced students are mere mortals with respect to each of the following live hypotheses (so conditions 1–9 all hold in that world): Belief error theory: no one believes anything (endorsed to some significant degree at various times by, e.g., Patricia Churchland [1986], Paul Churchland [1989], Stephen Stich [1983], Daniel Dennett [1978], Paul Feyerabend [1963], Richard Rorty [1970], and Willard Quine [1960, 1985]). Pain error theory: pain is only in the brain (endorsed by Russell, Broad, Ryle, David Lewis, and David Armstrong; see Hyman 2003 for references). Color error theory: no ordinary physical objects are colored (endorsed by, e.g., Galileo, Larry Hardin, Paul Boghossian, David Velleman, Emmett Holmon, and Frank Jackson, as well as by scientists Semir Zeki, Stephen Palmer, Werner Backhaus, and Randolf Menzel; see Byrne and Hilbert 2003 for references). Trait error theory: no one has any character traits (endorsed by Gilbert Harman [1999, 2000[ and John Doris [1998, 2002]).

The live skeptic's validly inferred conclusion from these three claims is that in either this world or some very close possible worlds, many of us do not know that fire engines are red, we do not know that we sometimes have pains in our lower backs, we do not know that John Rawls was kind, and we do not even know that we believe any of those truths. This skeptic does not say that any of those four error theories are true; she can safely believe that they are (utterly, completely) false . The error theorist says that we do not know that fire engines are red because they are not red; the live skeptic says that we do not know that fire engines are red because even though our belief may well be true, it is not sufficiently warranted. Obviously, if the error theories are true, then the live skeptic's conclusion holds, assuming that knowledge requires truth. And for the very same reason, if the error theories are truth‐valueless, the live skeptic still wins. So no matter what the status of the live error theories is (true, false, without truth‐value), we mere mortals about those error theories do not know that anyone has any beliefs, anything is colored, and so on.

3. The Liveness‐Mortality Premise

Consider belief error theory, or eliminativism, as it is commonly called. 2 Eliminativism is currently a live possibility in our intellectual community (of contemporary analytic philosophers, cognitive scientists, and their advanced students). The hypothesis is a going concern among the experts. That is just a brute fact about the community of philosophers of mind and cognitive scientists. You might not like it, but that is just the way it is. The Churchlands, Feyerabend, Stich, Rorty, Quine, Dennett, and others have at various points in their exceedingly distinguished careers concluded upon careful reflection that it is very probably true or as serious a contender as any other hypothesis about folk psychology. They arrived at these views in the familiar and epistemically responsible way of looking hard at data and lines of reasoning and then thinking about them in highly intelligent, relatively unbiased and open‐minded ways. These people are not dopes. And it goes without saying that no magic bullet has been found for it, and people still take it virtually as seriously as ever. What this means is that the live status of a hypothesis is a sociological fact, not a matter for philosophical argument.

Okay, I lied: eliminativism is not live today and was not even in the 1980s or 1990s. It has too few supporters. But if Fodor, Dretske, Burge, and a few other realists had died in 1974, the reaction to the eliminativist‐leaning writings of Stich, Dennett, the Churchlands, and others had been more positive in the 1980s and 1990s, the “Who cares what common sense says?” attitude in contemporary metaphysics and other areas of philosophy was alive in the philosophy of mind … You get the idea. Surely, in some possible worlds practically next door to ours, eliminativism is a live hypothesis in both the cognitive science and philosophy‐of‐mind communities. In case you were not aware of it, color error theory is practically the dominant view among color scientists today; the same holds for error theory for pain location. And character trait error theory, although not actually live, could easily become live in the same way eliminativism could. My interest lies in the epistemic status of mere mortals in those actual or almost actual communities: what do they know about beliefs, colors, pain locations, or character traits? The live skeptical argument concludes that a society virtually just like ours but with sufficient philosophical confusion regarding eliminativism, for instance, could have a significant period of time (decades, even centuries) in which most philosophers and cognitive scientists knew nothing of the form “S believes P .” That is shocking enough. And yet, for all we know, our society will soon enter into philosophical depravity and wholeheartedly vote eliminativism as the theory of cognition most likely to be true.

The live skeptic need not think that eliminativism (or the other error theories) is remotely plausible. Even Fodor could be a live skeptic. In fact, the live skeptic herself could not only know that eliminativism is false but know that she has refuted it and know that she knows that it is false—even though she claims that in her own philosophical community eliminativism is live. She could hold herself to not be a mere mortal; and she could be right about that because she really does have a refutation of eliminativism. Exactly the same holds in the more familiar dinosaur case: a paleontologist genius could know full well that the supervolcano hypothesis is mistaken and the meteor theory true while holding that most of her students and colleagues do not know either of those facts because of their mere mortality with respect to the live supervolcano hypothesis. Throughout this chapter , I assume with the live skeptic that the error theories are false and nowhere in the vicinity of the truth. All the live skeptic is claiming, in endorsing the liveness‐mortality premise, is that there are easily possible philosophical communities in which the error theory in question is live in the sense of (1)–(5), and there are many ordinary individuals in those communities who satisfy (6)–(9)—individuals just like us philosophers in the actual world. There is no tension at all in endorsing that claim while holding that eliminativism is utterly implausible and has even been ruled out by oneself. Perhaps Kripke, Burge, and a few others have unpublished but rock‐solid proofs that eliminativism is false; even the eliminativists would admit defeat if they just had a chance to see them. It does not matter in the least because those proofs are irrelevant to the epistemic status of mere mortals in communities in which those proofs do not exist or are known by just a few philosophical hermits.

I assume that the only reason to balk at the liveness‐mortality premise lies in condition 9. One might think that even if eliminativism is live in the sense of (1)–(5), one could still easily rule it out by merely reflecting on the fact that one has beliefs. But (9) is just saying that virtually all experts would judge the evidence you can marshal for casting doubt on H to be clearly insufficient to rule it out. It does not say that the experts are right about that. Perhaps every one of us can rule out eliminativism, for instance, merely by saying to ourselves, “It sure seems to me that I believe that 2 + 2 = 4.” I will be considering and arguing against such ideas later in this chapter . But in the worlds in which eliminativism is live, philosophers sufficiently skeptical about the truth of common sense will not consider such a line of reasoning as sufficient to rule out eliminativism (so there are not too many John Searles there).

One might suspect that in the actual world, a large majority of philosophers of mind really do think that one can refute eliminativism merely by breezily reflecting on one's own beliefs. (Recall that we are assuming that eliminativism is false.) If so, then perhaps worlds that make (1)–(9) true—(9) in particular—are quite distant from actuality, contrary to the liveness‐mortality premise.

I think that the objector's claim about the actual world is mistaken, but it will help the live skeptic's cause if we get clearer on just how a world could make (1)–(9) true and yet remain very similar to actuality.

A quick answer is that just as color error theory is actually live today , as anyone familiar with the philosophy and science of color can attest, belief error theory could have the same respected status. Both error theories are radically and comparably opposed to common sense; so that hurdle can be met. Perhaps more convincingly, one additional feature, briefly mentioned earlier, easily makes (1)–(9) true at a world: more people being skeptical about the strength of the connection between common sense and truth. As matters stands today, in the philosophy of mind and epistemology communities, there is a strong confidence that statements that are conversationally appropriate in ordinary contexts are true. Epistemologists and philosophers of mind are typically loath to claim that large parts of common sense are radically mistaken. In metaphysics and the philosophy of language and logic, however, one does not find this faith in the inference from common sense to truth. (Think of theories of vagueness, truth [the semantic paradoxes], material constitution, and identity through time—not to mention the philosophy of physics.) Indeed, it is hard to find any contemporary metaphysician who does not hold claims that are about as outrageous, from the perspective of common sense, as eliminativism. In my experience, those who work only in epistemology or the philosophy of mind often find this attitude highly dubious or even hard to fathom, but there are good reasons why it is prominent in other areas of philosophy.

The live skeptic needs a weak claim, one operative in her rebuttal to the objection that worlds in which (1)–(9) are true are remote from actuality: the skeptical attitude we find in contemporary metaphysics, philosophy of language, and philosophy of logic regarding even the most well entrenched and nearly universal common sense could easily permeate the philosophy of mind, while eliminativism had more expert supporters and fewer detractors. 3 I am not claiming that such an attitude movement would be correct or fully justified; I am just saying that it could easily happen. In those worlds, which I claim are highly realistic in being very similar to the current time of the actual world, eliminativism is live, and many of us are mere mortals regarding it. So we have justified the live skeptic's claim that worlds satisfying (1)–(9) are next door to actuality.

4. The Nature of Live Skepticism

In later sections, I will address some objections to the live skeptic's argument. I cannot consider more than a small fraction of all the reasonable objections (see Frances 2005 a and 2005b for more thorough treatment), so your favorite might not get investigated here. And I do not treat the few I do consider as thoroughly as I should. But I hope to provide some initial reason for doubting that they succeed.

The objections will fall into two classes. First, there are reasons for thinking that we do not have to be able to rule out, in any significant sense, the error theories in order to know truths obviously inconsistent with them. For instance, relevant‐alternatives and contextualist theories are important here. These objections fault the live‐hypothesis principle, which says that we do need to rule out the error theories. Second, one can admit the ruling‐out requirement but argue that it is actually pretty easy to do the ruling out. Reliabilism and introspective accounts figure here, among others. These objections target the modesty principle, which says that mere mortals cannot manage the ruling out.

But before we put our critical hats on, we must understand the nature of live skepticism, for it is quite different from any traditional skeptical theory. In fact, it is so strikingly different, in several ways, that to a certain extent I lose the motivation to find fault with the live skeptic's argument. I will list three key marks of live skepticism.

First, the live skeptic is claiming that we lose our knowledge by means of a perfectly familiar mechanism. 4 Pretend that it had been common sense for hundreds of years that whales are fish; pretend that it is also true. Then scientists came along to challenge that bit of common sense. They admit that there are many fish and many whales, but they have some impressive arguments for the shocking claim that whales are not fish—they are mammals instead. If you take a quiz listing kinds of fish and you write “whale,” scientists will mark your answer as incorrect. If you say to your child, “Look at the whales, Julia; they are the biggest fish there are,” they will claim that what you say is fine in some respects but really false. For the purposes of this chapter , pretend that the scientists are wrong: whales are fish. So one familiar and even highly reliable method for finding out that something is a fish—find out that it is a whale—would clearly be question begging against the group of scientists. In this scenario, if you were a mere mortal regarding the whale‐fish controversy, then you would be familiar enough regarding the relevant issues that in order to know that Keiko (the whale) is a fish, you would need to have some way of neutralizing the live whales‐are‐not‐fish hypothesis.

Consider the advocate of the pain‐is‐only‐in‐the‐brain hypothesis and compare what she says with what the whales‐are‐not‐fish advocate says (keep in mind that we are pretending that whales are fish and assuming that toes sometimes throb in pain).

I realize that it is common sense that whales are fish.

I grant that there are loads of fish and loads of whales, and to a certain extent there is nothing wrong with classifying whales as fish.

But they are not fish, not really.

Whales are actually mammals, not fish, for various technical reasons.

There are zillions of fish to be found, but they are not to be found among the whales.

I realize that it is common sense that toes sometimes throb with pain.

I grant that there are loads of throbbing feelings and loads of toes, and to a certain extent there is nothing wrong with classifying some toes as throbbing.

But they are not throbbing, not really.

The throbbing feeling is actually in your brain, not your toe, for various technical reasons.

There are zillions of throbbing feelings to be found, but they are not to be found among the toes.

I take it as intuitive that in the envisioned whale‐fish scenario mere mortals would not know that Keiko the whale is a fish—even though they used the familiar, commonsensical, and highly reliable “If you find a whale, then you have found a fish” method to form the true belief that Keiko is a fish. By analogy, just because you (a mere mortal) have a true belief that Mary's toe is throbbing and you formed it in a common way—you saw her stub her toe and heard her curse and tell you that it was throbbing painfully—this method, as familiar, commonsensical, and reliable as it indeed is (in that world and our world), is not sufficient to make you know that Mary's toe is throbbing, not when you are a mere mortal with respect to the live pain‐is‐only‐in‐the‐brain hypothesis. Again, the reason for the lack of knowledge is exactly the same as in the whales‐are‐not‐fish case. The live skeptic is saying that our knowledge is sabotaged because of the operation of a perfectly familiar epistemic mechanism, the one governing the whales‐fish or dinosaur cases.

Now for the second interesting feature of live septicism. Suppose once again that live skepticism is true. Suppose that tomorrow the definitive refutations of color error theory and eliminativism are published and widely digested. It is reasonable to think that we then know that fire engines are red and that Moore believed skepticism false, for at that point the error theories have been ruled out on our behalf. In addition, it is plausible to hold that since my father knows nothing of eliminativism, color error theory, professional philosophy, or cognitive science, he currently knows that fire engines are red and that my mother believes that motorcycles are a menace. It is also reasonable and consistent with the new skeptic's position to think that my father knew those facts years ago before eliminativism (and, let us pretend, color error theory) was even on the radar screen in science or philosophy. Finally, we should hold that before we were mere mortals, say, when we were children, we had loads of knowledge inconsistent with the skeptical hypotheses even if the latter were already live.

This is just to say that the reach of the epistemic threat posed by the live skeptical hypotheses is not sufficient to affect those people for whom it is not live. In a way, this makes the new skepticism modest: by saying that it rules out knowledge only for mere mortals, we are restricting its scope severely. In other ways, the new skepticism is now unlike anything we have ever encountered.

Here is why. Skepticism has always been considered a permanent, blanket, and highly negative condition. If you are in it, then everyone else is in it too (that is “blanket”), forever (“permanent”); and this marks a significant epistemic deficiency on our part (“highly negative”). If skepticism is right, then it must reflect an everlasting, perhaps absolutely necessary, failure of our cognitive systems to achieve a certain result. The idea that some kind of radical, even if restricted, skepticism could be the result of a purely accidental, contingent, and temporary confluence of chance social events affecting a fully rational, cognitively well‐off individual seems absurd. And it is absurd, at least for classic versions of skepticism. But we have seen that skepticism can be thus accidentally generated. The skepticisms discussed in this chapter are relatively fleeting and are no indictment of our cognitive systems or evolutionary progress. Most surprisingly, the new skepticisms are the odd result of cognitive systems and procedures working well , not poorly: it is through the epistemically beneficial practices embodied in the premises of the new skeptical argument template that we have temporarily fallen into our regions of restricted skepticism. By being part of a community that pays due heed to expertly produced contrary evidence, we have temporarily robbed ourselves of large portions of knowledge, but we have done so as the result of following epistemic practices that almost always actually buy us knowledge. Once we understand how we got into the mess of not having knowledge of color, belief, or pain location, for instance, we see that it is not a bad thing. In fact, we are epistemically better off than we were before . Yesterday you knew; today you do not; but today you are smarter than you were yesterday regarding color, belief, and pain. People unacquainted with philosophy or color science or cognitive science may know more than we do, but this just shows that we need a new and improved measure of epistemic standing. Suppose I think that on balance, color error theory is very likely mistaken, and so, even in the full awareness of the live status of that theory, I continue to hold ordinary color beliefs. My mother's belief that fire engines are red is warranted enough for knowledge; mine is not because I remain a mere mortal; but in some sense I am in the superior epistemic situation with regard to the color of fire engines despite my lack of knowledge and lack of immortality vis‐à‐vis color error theory.

That may sound contradictory: if token belief A (my mother's) is better warranted than token belief B (mine), and A and B have the same content (e.g., that fire engines are red), then of course A is epistemically better than B . In the previous paragraph, I had the live skeptic deny this conditional, but she need not. Perhaps the collection of my attitudes regarding color is epistemically superior to hers, even though when it comes to the particular fire‐engine belief her position is superior to mine. The details on how best to describe the differences in the two believers are interesting, but in any case the end result is that a typical reaction to skepticism—the kind of epistemic failure attributed to us by the skeptic just could not be built into our cognitive systems—does not touch the live skeptical theses. Similarly for the objection that runs as follows: according to the skeptic, we are cut off from reality, unable to know it, and that is just implausible. On the contrary, by becoming a mere mortal and thereby falling into the live skeptical trap, I am in a better epistemic position than I was before; skepticism is an improvement , something to be bragged about, not ashamed of. Falling “victim” to the live skeptical snare does not cut me off from the reality of color, belief, pain, and knowledge compared with those nonphilosophical folk who in a real sense know more than I do. Instead, I am the one better in tune with the facts; my opinions dig deeper into the nature of reality than theirs do. This is not too surprising: by knowing more about the possibilities regarding color and belief, I have a better appreciation of the “whole color thing” or the “whole story regarding cognition.” The ordinary person may know that fire engines are red, while I do not, but her belief has a measure of accidentalness that mine will never have. If she had just been privy to the intelligent and sophisticated ruminations of some philosophers and scientists, then she would no longer know. It is just an accident that she does know. The nature of the accidentalness is not so great as to rob her of knowledge, or so the antiskeptic says and the live skeptic may admit, but coupled with my expanded knowledge of possibilities and evidence for and against those possibilities, it is enough to render her overall epistemic situation less secure than mine.

Furthermore, I have all the positive warrant that she does when it comes to our beliefs that fire engines are red. This point is so important that I will put it in a box, as if it were in a science textbook.

The live skeptic is not denying us any of the warrant we think we have; neither is she denying its quality.

The live skeptic's crucial claim is this: the overall warrant possessed by the beliefs targeted by the live skeptical hypotheses is anemic compared with that had by our other beliefs that amount to knowledge and are not targeted by any live skeptical hypotheses.

Engage in the useful fiction that warrant comes in units (we can make the same point without the fiction). She has 1,200 positive warrant units for her fire‐engine belief; I have just 800; one needs 1,000 for the true belief to be knowledge. I have had all the same fire‐engine and color experiences that she has had; I have each of her 1,200 units. My only problem is that I have 400 negative warrant units, coming from my foray into color science and philosophy. Perhaps we need to judge the epistemic standing of beliefs with additional measures, such as with positive warrants (I have all the good reasons, reliability facts, and so on that she has) and absolute‐value warrants (since the absolute value of my warrants is much greater than hers, I am much more familiar with the relevant issues regarding the content of the belief*).

In any case, I am not claiming that live skepticism does not apply to people who know nothing of the error theories. I am just allowing for that relatively likely possibility. If it does apply to them, then live skepticism is even stronger, covering a much wider range of beliefs, although it still does not affect the millions of beliefs not targeted by live contrary hypotheses.

Now for the third interesting feature of live skepticism. Suppose that live skepticism is true, so you do not know that the table is brown, nor do you know that you believe that. Still, no one said that you did not know perfectly well that the table looks brown and is for all practical purposes brown. Further, you still know that red is darker than yellow. In addition, I still know that I act like I believe P , that it seems to me that I believe P , that I feel like I believe P , that it seems as though Mary believes Q , and so on. I also know that for all practical purposes, Moore believed skepticism false. And I know that belief and truth are required for knowledge. If that is the case, then what have we lost, really, in losing our knowledge that fire engines are red? He knows that fire engines are red; we do not; but we are familiar with much more color theory than he is, and we still know that fire engines look red and are red for all practical purposes. Now who is in the better epistemic position vis‐à‐vis the redness of the fire engine?

5. Do We Really Need to Rule Out the Error Theories?

Now we move on to consider objections to the live skeptic's two principles. If we answer no to this section's title question, then we are rejecting the live‐hypothesis principle, which says that we mere mortals do need to rule out the error theories in order to know things obviously inconsistent with them.

The live‐hypothesis principle is akin to, but much more plausible than, any of the standard epistemic closure principles. Compare what might be called simple closure with the live hypothesis principle.

Simple closure : if S knows that P entails ¬ H , then if H is not ruled out with respect to S, S does not know P . 5 The live‐hypothesis principle : if S knows that P entails ¬ H and is as aware as just about anyone that H is live (so H really is live), then if H is not ruled out with respect to S, S does not know P .

The two principles differ in two crucial ways: only the live‐hypothesis principle demands that H be live and that the subject be fully aware of H 's live status. As we are about to see, by incorporating these conditions, the live‐hypothesis principle is immune to the plausible objections raised against simple closure.

It is sometimes said that one need not, in order to know P , rule out (or know the falsehood of*) counterpossibility H provided H is appropriately “irrelevant.” 6 Maybe so, but it is hard to see how this would apply if H were live in the manner described earlier, you were fully aware of the liveness, and you knew full well the inconsistency of H and P . If what I have argued previously is wrong, then some hypotheses do not need to be ruled out for you in order for you to know P even though all the following conditions obtain:

Most people including yourself believe P .

Everyone believes (correctly) the obvious fact that H is inconsistent with P .

You have actually put together P and H and know as well as anyone that H is inconsistent with P .

H is a real, live contender in your intellectual community.

You are aware that H is a real, live possibility actually endorsed by plenty of top experts.

Lots of people worry about their attitude toward P in light of what they think about H .

Even the experts who think that H is rubbish would pretty much laugh at any reasons you could give against H .

If just conditions i–iii hold, then perhaps I do not have to rule out H in order to know P . Such a situation may obtain when P is “I have hands” and H is some traditional skeptical hypothesis. We might (I do not know) get the same answer if just the first four conditions hold. Perhaps one could know P without ruling out H provided one believed P for the “right” reasons and was unaware that H was a live possibility endorsed by experts. But when all seven conditions hold, one cannot know P without being able to rule out H . If all seven conditions hold, then H needs to be ruled out because it is now a “relevant alternative” to what I believe; I am “epistemically responsible” to H . It seems to me that that is a reasonable constraint on relevant‐alternatives theories.

Contextualists hold that different uses of “S knows P ” can have different truth‐values even when they concern the same person, truth, and time. The idea is that depending on the context of utterance, different levels of epistemic standing for S's belief in P are necessary for the truth of a use of “S knows P ” in that context. In some contexts, for instance, when a lot is riding on whether P is true or S can be relied on, then S's warrant has to meet a high standard to make the knowledge attribution true. In other circumstances, when, for instance, little rides on the truth of P , S's warrant need meet only a low standard.

This theory schema can be filled out in many ways, depending on how one fills out “standards,” what one says about which contextual factors can raise or lower standards, and what one says about how much those factors can raise or lower standards. There are many examples that support contextualism to some extent. Independently of that matter, contextualism is also thought to offer a plausible response to skepticism by making room for the claim that in discussions of skepticism, the standards can, if the conversation proceeds just right, get raised absurdly high, so when the skeptic concludes with “So we do not know anything,” she speaks the truth—even though in other contexts of utterance, governed by low standards, we speak the truth when we say, “I know she hates my motorcycle.” But can the theory provide a decent response to live skepticism?

Take as our example eliminativism. It is natural to say that in theoretical contexts (e.g., discussions of connectionism or Fodor's theory of content), eliminativism definitely poses a threat that can be neutralized only by some strong epistemic factors. The live skeptic seems right about knowledge attributions in those contexts. But what about the completely ordinary conversational contexts in which mere mortals discuss their alleged knowledge of a friend's beliefs? Or the color of their socks? Or whether the doctor is aware of the new pain in their grandfather's hip? Here the error theories are far from anyone's mind. So the live eliminativist hypothesis does not have to be ruled out in order for the mere mortal's “I know she thinks motorcycles are dangerous” to be true. Or so the objection to live skepticism goes.

For the sake of argument in the remainder of this subsection, I will assume that contextualism of some sort is true. The issue here is the truth‐value of a specific kind of contextualism, one that implies that, for instance, “I know you think motorcycles are dangerous” is true when the mere mortal conversational participants are not thinking or talking about live eliminativism (that is the objection to live skepticism), even though when they discuss eliminativism or other relevant theoretical topics appropriately, then the very same knowledge sentence is false (that is the concession to live skepticism). I will include just one argument regarding this kind of contextualism.

It is plausible to think that the BIV hypothesis needs to be ruled out in order to save the truth of ordinary knowledge attributions only if it is being discussed or thought about appropriately. It is hard to see what else could make the BIV hypothesis a threat to one's belief that one has hands. However, that does not mean that occurrent discussion and thought are the only ways to make a hypothesis threatening. On the face of it, liveness conditions make a hypothesis just as much or even more of a threat than occurrent attention. We already saw this point when we were briefly discussing relevant‐alternatives theories. So we cannot just say, “Well, we have already seen that skeptical hypotheses can be truth‐conditionally irrelevant in some contexts; so that is what must be happening with the error theories too.” Neither can we get away with “Well, we know that if one is just minimally rational, then one has neutralized the BIV hypothesis, so the same must be true for the color error theory hypothesis.” The liveness conditions 1–5 and the mere mortality conditions 6 and 7 make the error theories truth‐conditionally relevant as they are “highlighted” or contextually salient.

6. Is It Not Actually Easy to Rule Live Error Theories Out?

So perhaps we do need to be able to rule out the live error theories. Can we mere mortals do it? If we can, then the modesty principle is false because it says that we cannot.

Maybe we can neutralize eliminativism without really doing anything. To see how, suppose that philosophers and cognitive scientists proclaim that there are no feelings at all. These eliminativists about feelings argue just as the eliminativists about belief argue:

No one has any beliefs/feelings. I realize that there are all sorts of cognitive/experiential processes or states in our heads and bodies, but none of them have what it takes to be a belief/feeling. The notion of a belief/feeling is a muddled folk notion that has been constructed in such a way that anything that is a belief/feeling must satisfy certain conditions. But in all probability nothing comes close to satisfying those conditions, which is why there are no beliefs/feelings.

I assume that eliminativism about feelings really is crazy. Even if all the philosophers of mind and cognitive scientists were screaming at my door, “No one has ever had a painful feeling! There are no feelings! No one has ever been in pain!” I would still know that I have had loads of painful feelings. My “access” to my painful feeling is so reliable or direct that I can epistemically neutralize, without even thinking about it, any hypothesis that says that I have no feelings. I might be wrong about the location of the feeling, but there is no way in hell that I am wrong that I have a painful feeling.

Now it might be thought that we have given the antiskeptic all she needs. Since I can know without any effort that the eliminativist about feeling is wrong, since I automatically have evidence (very broadly construed) that rules it out, similarly I can know without effort that the eliminativist about belief is wrong, since I automatically have evidence that rules it out. Perhaps this maneuver will not work for the other error theories, but it works for the most interesting one, eliminativism.

I do not think so. For one thing, I simply cannot imagine any remotely plausible reasons for thinking that no one is ever in pain. The no‐pain hypothesis just could not be live in a rational community. 7

More to the point, the notion of belief is just more theoretically loaded than that of feeling. For belief is not some subjective state like feeling. Perhaps a painful feeling just is something that seems painful, but the analogous claim for belief does not work. Sure, there is a feeling of subjective certainty toward a claim; to that extent, belief can involve something akin to sensation. I can know, no matter what the experts say, that I have a feeling of certainty that is directed toward a claim. But there is more to belief than that. Beliefs are not subjective feelings of certainty or anything else. They might include such feelings, but they are more than that. At the least, it is pretty unreasonable to reject this view about belief going beyond feeling. In fact, I do not know anyone who rejects this view. So even though I do have “subjective evidence” that neutralizes the hypothesis that I do not have any feelings of certainty or at least approval of a claim, I do not have subjective evidence that rules out eliminativism about belief. Similar points hold for pain locations, character traits, and colors. For those reasons, I do not think that the antiskeptic can reasonably hold that we all automatically, or upon easy reflection, possess warrant that is, on balance, sufficient to neutralize the error theories. We possess positive warrant that when left alone is sufficient for knowledge of facts about beliefs, but the liveness and mere mortality conditions generate enough negative warrant to sabotage that knowledge.

Maybe reliabilism will suffice to doom live skepticism. Suppose that reliability facts suffice to place a significant amount of positive epistemic warrant on our second‐order, first‐person beliefs. Perhaps they supply enough warrant that in ordinary circumstances, when we are not in a society fallen to serious consideration of eliminativism, true second‐order, first‐person beliefs are often good enough for knowledge. The live skeptic can agree with all of that. Perhaps they supply so much warrant that even in an eliminativist‐live society, people can have second‐order knowledge if they are quite divorced from and ignorant of both philosophy and cognitive science. Even so, eliminativism is actually live, and I am aware of it and its status. Suppose that I happen to reject the eliminativist hypothesis and so believe that I have beliefs. Still, I am no immortal regarding eliminativism. It seems that there is now a significant amount of negative epistemic warrant placed on my second‐order, first person beliefs (this is the negative warrant revealed by the eliminativist skeptical argument). And there is enough of it to render the positive reliability‐based warrant insufficient to render my second‐order beliefs warranted enough for knowledge. Only an exceedingly strong form of reliabilism can back up a neutralization of the eliminativist hypothesis. Everyone has admitted that factors like the recognition of expert counteropinion, liveness, independent contrary evidence, and whatnot can defeat a reliably formed belief—even highly reliable ones.

A relevant point here is that if the degree of reliability of the belief‐ producing process has significant epistemic weight (as the reliabilist reasonably holds), then the degree of reliability of the belief‐ sustaining process also has significant epistemic weight. This is no mystery. The belief‐sustaining process has to be reliable to retain knowledge, and in the cases described in this chapter —liveness, mere mortality—to continue to have beliefs that are known to be in contradiction with the recognizably live hypotheses is clearly highly unreliable. That is, the following belief‐sustaining procedure—which applies to the beliefs in question—is unreliable: continue to believe that P even though a contrary hypothesis Q has become live and one is a mere mortal with respect to Q . The reliabilist might be right that the reliability of the belief‐forming process is often enough to make a true belief knowledge; so when we are children those beliefs amount to knowledge. Our question has to do with the retention of that knowledge upon becoming aware of contrary expert evidence. The reliabilist who wants to use reliability factors to defeat the live skeptical arguments has to claim not only that the belief‐producing reliability confers a significant amount of warrant, but also that, mysteriously, belief‐sustaining reliability is irrelevant. Reliability of belief formation gives me 1,218 positive warrant units; standing fast with the belief that P in face of appreciated, well‐respected, expert evidence produces just 200 negative warrant units; I needed just 1,000 units to have my true belief turned into knowledge. Alternatively, belief‐forming reliability gives me positive warrant of sufficient quality for knowledge, and belief‐sustaining unreliability amounts to no veto. Should we swallow this extremism? I certainly do not feel any temptation to do so when it comes to philosophically uninteresting examples, so I do not see why we should when it comes to philosophically interesting examples.

7. Watering Down the Conclusion

Suppose that despite everything I have just argued, some antiskeptical solution for ordinary, everyday contexts is right even though perhaps in theoretical contexts the live skeptic wins. So in ordinary‐life contexts “I know my socks are blue” is true. I will make two comments in response.

I have given some pretty good arguments that in theoretical contexts—ones analytic philosophers and cognitive scientists often find themselves in—the new skeptical arguments go through. I find that skeptical result pretty amazing. If it is right, then you do not know much of anything right now about the color of objects around you, anyone's beliefs, anyone's character traits, or the pains in your knees. And you do not know any of that in many, many philosophical and even scientific contexts. In spite of the recent endorsement of skeptic‐friendly versions of contextualism, according to which we know very little if we are discussing a classic skeptical hypothesis and argument in the right way (thereby making those hypotheses mount real epistemic threats despite being nowhere near live), I still find it almost unbelievable that we do not know, right now , simple facts about the colors of our socks, our aches and pains, or what we believe.

Finally, suppose once again that some antiskeptical solution is right; so live skepticism is defeated. Then perhaps the antiskeptic may have won a battle but lost the war. Knowledge attributions are often true (that is the won battle), but the truth conditions of those attributions are so meager that we should feel a little ashamed (that is the lost war). The persistent live skeptic will say that even if the relevant ordinary knowledge attributions are true, skepticism wins because the truth conditions for those knowledge claims are impoverished. It should be clear that the persistent skeptic's position is not that although “S knows P ” is true, S's knowledge is not “high‐standards” knowledge. This skeptic is not complaining that our knowledge does not fulfill some super‐duper high‐octane condition that only a philosopher could love. The persistent live skeptic who objects that even if the relevant knowledge attributions are often true, the knowledge states would exist but be pathetic is not whining that knowledge is not what she always fantasized it should be. She has admitted that we can have all the positive warrant we thought we had; she can even admit that we often have super‐duper high‐octane knowledge of contingent matters of fact. She is targeting a special set of beliefs, ones targeted by the live error theories, and making one crucial claim: the overall warrant possessed by the beliefs targeted by the live skeptical hypotheses is anemic compared with that had by our other beliefs that amount to knowledge and are not targeted by any live skeptical hypotheses. There are seven reasons—(1) through (7)—for thinking that I need some powerful epistemic factors to defuse the expertly endorsed, highly respected contrary scientific error theories plus two reasons—(8) and (9)—to think that I have no such factors; and yet you are telling me that I can know lots of things inconsistent with those error theories? This cannot be a victory the antiskeptic should celebrate. If this realism about knowledge is accurate, then it is a gaunt realism, and the live skeptic has been closer to the truth than the realist.

I have introduced and evaluated live skepticism in more detail elsewhere (Frances 2005 a, 2005b).

For my purposes, “eliminativism” indicates just the view that no one believes anything (although it is convenient to conjoin it with the thesis that knowledge requires belief*). It is silent—neither pro nor con—on the status of other folk psychological concepts. Stich ( 1996 , 16–29) offers a very helpful summary of eliminativist arguments (eleven by my count). Since the details of these positions and arguments are both distracting and irrelevant to my purposes, I will not examine them here.

How large does the group have to be? Surely precision is out of place here. There is a group of well‐informed, mere mortal philosophers who take seriously the possibility that there are no chairs, but there are not enough of them to make it live, at least in my judgment. More to the point, that hypothesis does not even come close to meeting all five of the liveness conditions. There are deep issues involved in determining what kinds of hypotheses generate decent live skepticisms when plugged into the argument template of the previous section, none of which I can go into here. Suffice it to say that being a live hypothesis involves meeting a very demanding set of conditions. In part, I stick with hypotheses that have scientific backing because we defer to scientists more than we defer to philosophers, and this difference seems wise.

Here I am assuming that before we become mere mortals, say, as children, the liveness of the skeptical hypotheses does not sabotage our knowledge. I will say more on this point later in this section.

Of course, this principle is usually stated differently, but those differences will not matter. Minor objections to simple closure, e.g., that the subject has to “think through” the connection between the entailment and her belief in P , can be avoided by building additional conditions into the antecedent. Of course, these can be applied to the live‐hypothesis principle as well.

I have in mind contextualist and relevant‐alternatives theories, although what appears in the following discussion addresses contextualism primarily. These theories are discussed in Chapters 19 and 21 of this volume.

One can, of course, marshal good reasons for thinking that a certain kind of “phenomenal pain” does not exist, where a theoretically loaded reference‐fixing definition of “phenomenal pain” is in play, but that is another matter entirely.

Byrne, Alex , and David Hilbert . ( 2003 ). “ Color Realism and Color Science. ” Behavioral and Brain Sciences 26: 3–21.

Google Scholar

Churchland, Patricia S. ( 1986 ). Neurophilosophy: Toward a Unified Science of the Mind‐Brain . Cambridge, MA: MIT Press.

Google Preview

Churchland, Paul M. ( 1989 ). A Neurocomputational Perspective: The Nature of Mind and the Structure of Science . Cambridge, MA: MIT Press.

Dennett, Daniel . ( 1978 ). Brainstorms . Cambridge, MA: MIT Press.

Doris, John . ( 1998 ). “ Persons, Situations and Virtue Ethics. ” Noûs 32: 504–30.

———. ( 2002 ). Lack of Character . Cambridge: Cambridge University Press.

Feyerabend, Paul . ( 1963 ). “ Materialism and the Mind‐Body Problem. ” Review of Metaphysics 17: 49–66.

Frances, Bryan . ( 2005 a). Scepticism Comes Alive . Oxford: Oxford University Press. 10.1093/0199282137.001.0001

———. ( 2005 b). “ When a Skeptical Hypothesis Is Live. ” Noûs 39: 559–95.

Harman, Gilbert . ( 1999 ). “ Moral Philosophy Meets Social Psychology: Virtue Ethics and the Fundamental Attribution Error. ” Proceedings of the Aristotelian Society 99: 315–31. 10.1111/1467-9264.00062

———. ( 2000 ). “ The Nonexistence of Character Traits. ” Proceedings of the Aristotelian Society 100: 223–26. 10.1111/j.0066-7372.2003.00013.x

Hyman, John . ( 2003 ). “ Pains and Places. ” Philosophy 78: 5–24.

Quine, Willard.V.O. ( 1960 ). Word and Object . Cambridge, MA: MIT Press.

———. ( 1985 ). “ States of Mind. ” Journal of Philosophy 82: 5–8. 10.2307/2026508

Rorty, Richard . ( 1970 ). “ In Defence of Eliminative Materialism. ” Review of Metaphysics 24: 112–21.

Stich, Stephen . ( 1983 ). From Folk Psychology to Cognitive Science: The Case against Belief . Cambridge, MA: MIT Press.

———. ( 1996 ). Deconstructing the Mind . New York: Oxford University Press.

About Oxford Academic
Publish journals with us
University press partners
What we publish
New features
Open access
Institutional account management
Rights and permissions
Get help with access
Accessibility
Advertising
Media enquiries
Oxford University Press
Oxford Languages
University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

Copyright © 2024 Oxford University Press
Cookie settings
Cookie policy
Privacy policy
Legal notice

This Feature Is Available To Subscribers Only

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

I. Definition

Are you skeptical about holistic medicine? Climate change? The existence of extra-terrestrials? We’ve come to use the word “skepticism,” in our society, most often, to express doubt in new or “far out” ideas. Originally, in ancient Greece, skepticism was the philosophy of questioning all claims, religious, ethical, scientific, or otherwise. The point of skepticism was not so much to disbelieve claims, but to interrogate them; the word skepticism is derived from the Greek skepsis , meaning “inquiry.”

Philosophical skepticism can mean either:

questioning all claims, perhaps in order to better ascertain the truth, or . . .
the belief that there are no certain truths (including that statement itself) — no knowledge , only beliefs.

In practice, skeptics don’t always distinguish between these two attitudes, simply questioning claims, without worrying about whether absolute truth is possible.

More specific kinds of skepticism include religious skepticism, moral skepticism, legal skepticism, and scientific skepticism (see section five for details).

Skepticism is not the same as agnosticism, atheism , or faith in the current scientific model of nature, although it overlaps with all these attitudes. The true skeptic questions all beliefs including current scientific theories. However, skepticism is inherently associated with science because questioning and demanding proof is part of the scientific method.

II. The History of Skepticism

In the western tradition, skepticism is generally credited to the Greek philosopher Pyrrhus, from around 350 B.C.E–although the Sophists before him also promoted skeptical attitudes. It’s known that Pyrrhus traveled to India and communicated with “gymnosophists” a Greek term meaning “naked lovers of knowledge”; these were probably Hindu ascetics of some kind, many of whom lived literally naked and homeless, devoting their time to the development of mystical knowledge through meditation. Hinduism is not, in general, a highly skeptical belief system, with its faith in gods and transcendental reality, but it developed philosophical schools of thought that questioned the possibility of truth; and, Pyrrhus was particularly convinced by the Indians not to trust the evidence of the five senses.

It may not be inconsistent with Pyrrhus’s skepticism that his Hindu colleagues had spiritual beliefs; Pyrrhus and his students focused on the value of questioning rather than on disbelief. They argued most, in ancient Greece, with the Stoics, who had dogmatic beliefs about morality and the purpose of life. Otherwise, skepticism seems to be assumed by most other Greek philosophies (except Stoicism ); Socrates, Plato , Aristotle , and many others practiced skepticism, whether they considered themselves Pyrrhonists or not.

The medieval domination of Europe by dogmatic Catholicism put a relative end to public expressions of skepticism in the west for about 1,000 years; the Church was supposed to be an authoritative source of truth and to question its dictates could be heresy.

III. Controversies

Can anything be known? Nearly everyone agrees that inquiry is valuable in the quest for knowledge. But more radical forms of skepticism claim that there can be no certainty. Since figuring out truths is the central task of philosophy, many philosophers have struggled to determine what can and cannot be known, or to create a method for arriving at undoubtable truths. There are quite a few famous arguments and thought experiments with these goals:

Renee Descarte’s cogito ergo sum “I think therefore I am” was intended to answer the question, “What can I not doubt?” It is an appealing answer, however, various philosophies still question whether “I am.”
The dream argument (that you could always be dreaming)
The fake memory argument (you could have been created and put in this fake reality five minutes ago, with fake memories)
The phenomenological reduction : introduced by Edmund Husserl, this is the act of accepting as true only that one is perceiving or experiencing something, neither believing nor dis-believing the reality of those perceptions.

Thus, many philosophers agree that it’s impossible to know anything about external reality for sure. Yet, at the same time, most believe that, practically speaking, reliable knowledge comes from two sources—empirical observation and rational thought—especially logic. Most philosophers believe that logic, ideally, is not vulnerable to the brain-in-the vat argument. Living in a hallucinatory reality wouldn’t change the laws of logic—or math. And although all empirical observations could be hallucinatory, few philosophers consider that a good reason to especially doubt the natural laws derived by science; they have proven themselves reliable. But, in the final analysis, a good skeptic will always be ready to question them again.

Several perspectives that became well-known in our society since the 1950s question things that westerner philosophers have thought unquestionable. Many varieties of eastern mysticism maintain that if one meditates correctly, one will discover that the “I am” is merely an idea, not a reality. And Taoism , modern linguistic science, and post-modern philosophy all question the ability of logic to express truth, arguing that all symbols are inherently limited, incapable of representing reality truly. So, although philosophers and scientists continue to rely on logic and observation, radical, complete, skepticism remains a viable philosophy.

IV. Famous Quotes about Skepticism

Quotation #1.

“Extraordinary claims require extraordinary evidence.” ― Carl Sagan

Practically the mantra of modern scientific skepticism, Sagan’s pithy motto is taken to justify both the extremely stringent standards of proof in professional science, and the practice of requiring even more proof for truly extraordinary claims, such as ESP, even when a fair amount of evidence is reported. Sagan’s statement appeals to skeptics but it should be questioned (of course); is it really logical to require a higher standard of proof for any one claim as opposed to another? One could argue that Sagan’s statement implies unreasonable faith in ‘ordinary claims’!

Quotation #2

“She believed in nothing. Only her skepticism kept her from being an atheist.” ― Jean-Paul Sartre

Sartre’s quip cuts to the heart of the problem with regarding skepticism as an allegiance with science against beliefs in the supernatural. Radical skepticism, which could be interpreted as believing in nothing, dictates that one should question all beliefs, including those that deny the supernatural.

V. Types of Skepticism

Philosophical skepticism : We’ve talked about this one already; let’s distinguish its sub-types:
Pyrrhonic skepticism promotes questioning, as a method for better approaching truth, but does not deny the possibility of knowledge.
Academic skepticism denies the possibility of knowledge.
Global skepticism applies to all knowledge.
Local skepticism applies to a particular area of knowledge.
Moral skepticism is skepticism about moral truths
Religious skepticism is skepticism about religious beliefs
Metaphysical skepticism claims it’s impossible to know the ultimate nature of reality.
Scientific skepticism says that claims of truth about reality should be subjected to the scientific method and its requirements for proof.

VI. Skepticism versus Pessimism

Pessimism is the expectation that things will go badly. Skepticism is often used in everyday language to mean “pessimism”; a person can say, “I am skeptical about the outcome,” meaning that they question the likelihood of a positive outcome. This is confusing because skepticism and pessimism really have little in common. Pessimism is a belief in negative outcomes. While skepticism not a belief in anything and is neither positive nor negative, unless you feel that questioning is inherently negative.

VII. Skepticism in Popular Culture

Contact , the film

In this film based on a novel by Carl Sagan, Jodie Foster’s character, a scientist and atheist, finds herself in the odd position of believing in an extraordinary experience—contact with extra-terrestrials—without proof. In this scene, her critics use the same arguments against her, that she has always made against belief in God, which is why she glances over at her boyfriend the preacher. As a good scientist, she cedes the argument to the skeptics, but her final speech suggests a valuable criticism of skepticism—that there may be unprovable truths.

The problem of fake news and the reliability of internet sources

Basically, the solution to these problems is skepticism. The problem is that more and more people take whatever they read online without skepticism. Or are skeptical about only the things they already disagree with. The original Greek meaning of skepticism, inquiry , is the solution to this problem; knowledge depends on questioning. On the internet, this includes fact-checking and investigating the reliability of sources and alternative points of view.

a. Questioning whether evolution is true

b. Refusing to believe evidence for ESP

c. Believing that climate change is false

a. That “I am”

b. That “I think”

c. That “I am experiencing something”

d. Maybe nothing

a. Pyrrhonic skepticism

b. Academic skepticism

c. Metaphysical skepticism

d. Scientific skepticism

a. Scientific skepticism

b. Religious skepticism

c. Pyrrhonic skepticism

d. None of the above

Table of Contents
Random Entry
Chronological
Editorial Information
About the SEP
Editorial Board
How to Cite the SEP
Special Characters
Advanced Tools
Support the SEP
PDFs for SEP Friends
Make a Donation
SEPIA for Libraries
Entry Contents

Bibliography

Academic tools.

Friends PDF Preview
Author and Citation Info
Back to Top

Skeptical Theism

[ Editor’s Note: The following new entry by Timothy Perrine replaces the former entry on this topic by the previous author. ]

All of us—theist, atheist, and agnostic alike—experience suffering and evil in the world. There’s the annoyance of a stubbed toe, the disappointment of personal or professional setback, the endless frustration of debilitating chronic pain, and the soul-crushing experience of the suffering and death of those we care the most about (to name a few). It doesn’t require extensive education to worry if suffering and evil is evidence against the existence of God—or, at least, God understood classically, as a perfect being that is an all-powerful, all-knowing, all-good creator of the universe. This is the problem of evil, and when these worries are turned into arguments, they are known as arguments from evil.

But many of us—again, theist, atheist, and agnostic alike—are impressed by what God is supposed to be: a being with all perfections and knowledge, unencumbered by weaknesses or faults. Given our own ignorance, imperfections, and limitations, it is hard to fully imagine such a being and its potential plans for the world. Again, it doesn’t require extensive education to worry that it would be incredibly challenging for us—in such a comparatively benighted state—to predict, anticipate, or understand all of the plans of such a being.

Skeptical theism is a family of responses to arguments from evil. Skeptical theists think these two types of worries are related, that worries of the second type raise problems for worries of the first type. More specifically, skeptical theism embraces two claims. First, even if theism were true, we should be skeptical of our abilities to reasonably predict all of God’s plans for organizing the world, including those about the amount and nature of evil. Second, if this first claim is true, then it undermines or otherwise greatly mitigates arguments from evil. Now these two claims are vague and imprecise, and different skeptical theists spell them out in different ways depending on the argument from evil they are evaluating. Additionally, neither of these claims assume theism ; they can be coherently accepted by theists, atheists, and agnostics alike. Nonetheless, theists frequently appeal to skeptical theism in defense of theism. Thus, sometimes authors use the term ‘skeptical theism’ to refer to the conjunction of these kinds of conditionals with a commitment to theism; sometimes it is just these kinds of conditionals without a commitment to theism. For clarity here, we’ll use ‘skeptical theism’ and similar phrases to be neutral with regard to whether theism is true.

Skeptical theism has a number of historical precedents (see Rudavsky 2013). It was reintroduced into the contemporary literature in a 1984 paper by Stephen Wykstra. Since then, a variety of authors have developed and defended it including William Alston, Alvin Plantinga, Peter van Inwagen, Michael Bergmann, and Daniel Howard-Snyder. The literature on the topic is extensive and technical at points. Consequently, this entry must be selective and simplifying. The overall aim is to provide a basic understanding of the key issues surrounding skeptical theism, with extensive references for readers to dig into the meat of the debates.

1.1 Basic Outweighing Goods Argument

1.2 noseeum inferences, 1.3 cornea-based criticisms, 2.1 representative-based inductive inferences, 2.2 representative goods, 2.3 connecting goods, 3.1 individual case argument, 3.2 cumulative case argument, 4. skeptical theism and other responses, 5.1 objections to cornea principles, 5.2 common sense epistemology, 5.3 external world skepticism, 5.4 religious skepticism, 6.1 moral deliberation, 6.2 all-things-considered value, 6.3 all-things-considered obligations, 7.1 non-inferential arguments, 7.2 humean arguments, 7.3 divine hiddenness, 8. theory development, additional readings, other internet resources, related entries, 1. outweighing goods and noseeum inferences.

Skeptical theism is a cluster of responses to arguments from evil. One major argument from evil is what we’ll call the “Outweighing Goods” argument. This section explains a basic outweighing goods argument, a standard way of defending it using “noseeum inferences,” and a skeptical theistic critique of that defense.

A simple thought is that if God exists, there would be no evil. After all, God could prevent any evil there is; God would know about any evil there is; and God would want to prevent any evil there is. But this simple thought is too simple. For even if God exists, it need not be the case that God wants to prevents any evil there is. For God may have a justifying reason for permitting evil.

Frequently, in life, agents are justified in permitting evil if there is some reason that justifies their permission of it. A common and well-known kind of reason is that the permission of the evil is necessary for either bringing about some greater good or preventing some worse evil from occurring. (We’ll use the term ‘outweighing goods’ to refer inclusively to the greater goods or worse evils.) Indeed, the complexity of trading off permitting evils for outweighing goods is not just a part of consequentialist thinking but pervades most ordinary ethical and legal thinking. There may be other kinds of reasons that justify an agent in permitting evil. (For instance, perhaps the agent does not have the appropriate authority to prevent the evil or the actions necessary to prevent the evil would go against the agent’s character or personal projects.) But justifying reasons involving outweighing goods have dominated discussions of arguments from evil; so this entry will focus on them. Keeping these points in mind, one important argument from evil is this (compare Rowe 1979, 1984, 1986; Martin 1978):

This argument is deductive; thus, defenses of it focus primarily on the premises. N-O 1.1 is seen as a plausible claim articulating God’s obligations to the rest of us. N-O 1.2 is an existential claim, stating that some evil is, crudely put, not outweighed. A straightforward way of defending it is existential generalization: find specific evils such that God’s permission of them is not necessary for some outweighing good, and then generalize to N-O 1.2 .

Consider two situations given in the literature: a fawn, caught in a forest fire, that burned to death (cf. Rowe 1979), and a five-year-old girl who was raped and then strangled to death (cf. Russell 1989). Their sufferings are paradigms of evil. Now one might reflect on their suffering, and try hard to identify God’s reasons for permitting their suffering. An important sub-argument for N-O 1.2 follows:

This argument is not valid, but inductive in nature. In defense of N-O 1.21 , authors like Rowe argue that the suffering of the fawn and the girl seem unconnected from the concerns of traditional theodicies. Their suffering isn’t the result of their own free will, for instance; nor does their suffering contribute to their “soul-building” of creating a superior moral character. And while eternal union with God may be extremely good, it is hard to see how this suffering would be connected to it. Reflecting on their suffering, it is hard to identify outweighing goods that would justify God’s permission of that suffering.

This way of defending N-O 1.2 is sometimes called a “noseeum” inference (Wykstra 1996). From the claim that we “no see” an outweighing good, one infers that there is no such outweighing good and, thus, there is no God. (Wykstra’s inspiration for this term comes from the name of an insect – a ‘no-see-um’ – that is normally too small to be seen, but can still leave a nasty welt.) Alternatively, this way of defending N-O 1.2 is called an argument from “failure of theodicy” (Draper 2013), since it is the apparent failure of theodicies that make N-O 1.2 reasonable to believe.

Some skeptical theists critique noseeum arguments by using what is called a CORNEA-based criticism (Wykstra 1984, 1996). Like many skeptical theistic critiques, this critique has two components—a general epistemic principle and some claims in philosophy of religion. By combining the two components, skeptical theists aim to undermine some argument.

The term ‘CORNEA’ is an unprincipled acronym: Condition On ReasoNable Epistemic Access. A CORNEA principle is an epistemic principle stating conditions on when it is reasonable for a person to claim or believe p on the basis of something q . Wykstra himself formulated a range of principles that are not obviously equivalent. A simple CORNEA principle is:

Simple-CORNEA ( S-C ) A person is prima facie reasonable in believing p on the basis of q only if, given their cognitive abilities, it is reasonable for them to believe that if p were different, then q would likely be different as well.

To illustrate the plausibility of CORNEA principles, consider the following case. A doctor accidentally drops a needle on the floor. The doctor picks up the needle, visually inspects it, and announces that the needle is free of viruses and bacteria. Is the doctor prima facie reasonable in believing the needle is free of viruses and bacteria on the basis of his visual experience? Intuitively not. A principle like S-C is well-suited to explain why. It is not reasonable for the doctor to believe that, given his cognitive abilities, if the needle were contaminated with bacteria or viruses, then his visual experience would likely be different as well.

A CORNEA principle like S-C is a general epistemic principle that can be applied to any range of cases. Skeptical theists apply it to the inference from N-O 1.21 to N-O 1.22 . Suppose we are prima facie reasonable in believing that God’s permission of the fawn’s suffering and the girl’s suffering would not be necessary for some outweighing good on the basis of our failure to see an outweighing good connected to their suffering by reflecting. Then, by S-C , it must be reasonable for us to believe that if God’s permission of the fawn’s suffering and the girl’s suffering were necessary for some outweighing good, then we would likely not fail to identify an outweighing good connected to their suffering by reflecting.

But skeptical theists argue that this is not reasonable for us to believe. That is, they argue that even if God’s permission of the fawn’s suffering and the girl’s suffering were necessary for some outweighing good, then it is likely that we would fail to identify an outweighing good connected to their suffering simply by reflecting. Skeptical theists give analogies to support these claims. One extended analogy is a parent/child analogy (compare Wykstra 1984: 88; 1996: 142–5). A baby infant is in no position to evaluate its parent’s plans for its life. But our cognitive abilities are like a baby infant’s in comparison to God’s. So too, we are in no position to evaluate all of God’s plans. Similarly, parents with increased intelligence, understanding of goods, and abilities are more likely to plan for goods in their children’s future—a future which their children are not yet able to see. God, having maximum intelligence, understanding, and ability, is thus likely to make plans where many goods will be in the future in ways we cannot expect to understand. Another analogy involves the “physical depth” of the universe (cf. Russell and Wykstra 1988: 146–8; see also Alston 1991 [1996]: 109). If God created the universe, God created it with “physical depth,” where many of the features of the universe are difficult to learn about. So too, if God has plans for creation, God is likely to have created it with “moral depth,” where many of the goods that play a role in God’s plan are not “surface” goods easily identified but rather more obscure to us.

Thus, by S-C , believing N-O 1.22 on the basis of N-O 1.21 is reasonable only if it is reasonable to believe that, given our cognitive faculties, if God’s permission of their suffering were necessary for an outweighing good, we’d identify that good by reflecting. But that’s not reasonable for us to believe—even if there is an outweighing good justifying God’s permission of their suffering, then we likely wouldn’t see it by reflecting. So it’s not reasonable to believe N-O 1.22 on the basis of N-O 1.21 . Thus, according to this line of reasoning, the sub-argument for N-O 1.2 fails.

2. Outweighing Goods and Inductive Inferences

One of the key claims of the outweighing goods argument is N-O 1.2 : there is at least one evil e such that God’s permission of e is not necessary for some outweighing good. One way of defending this premise is to appeal to how things seem to us. But another way to defend N-O 1.2 is to eschew talk of how things seem to us and use more traditional inductive inferences.

Suppose I have a garden. I like spicy food, so every year I plant hot peppers. I plant a variety of hot peppers—differences in size, heat-level, place of origin, etc. And suppose that each year I notice that my hot peppers only begin flowering in late summer, early fall—never early spring. I might reason as follows:

No-Known-Pepper ( NKP ) No hot pepper I know of flowers in early spring.

None-at-all-Pepper ( NaaP ) Therefore, no hot pepper at all flowers in early spring.

The reasoning in this case is inductive. I have a sample size—hot peppers I know of—and they have a certain property—they don’t flower in early spring. I infer from this sample size to a larger group—all hot peppers—that they have the same property. Since my sample size is representative of the larger group, this inductive inference is a reasonable one.

Analogously, suppose we consider whether we know of any outweighing goods that would justify God’s permission of the girl’s suffering and the fawn’s suffering. We are familiar with a variety of outweighing goods in life. Suppose we think that none of the outweighing goods we know of would justify God’s permission of the girl’s suffering and the fawn’s suffering. We might reason inductively as follows (compare Rowe (1988, 1991):

No-Known-Goods ( NKG ) No outweighing goods we know of would justify God’s permission of the fawn’s suffering and the girl’s suffering

None-at-all-Goods ( NaaG ) Therefore, no outweighing goods at all would justify God’s permission of the fawn’s suffering and the girl’s suffering.

From NaaG it is a short-step to N-O 1.2 and then atheism. This way of defending N-O 1.2 avoids noseeum inferences and so might be thought to side-step the CORNEA-based criticisms of the previous section.

Skeptical theists have critiqued the inference from NKG to NaaG (see Alston 1991 [1996], 1996; Christlieb 1992: 48–54; Sennett 1993; Bergmann 2001). Once again, the critique can usefully be divided into two parts—an epistemological principle and some claims in philosophy of religion.

Skeptical theists have spent less time developing the epistemic principle used in this critique. But they assume a principle something like this: an inference from a sample of a group to the group vis-à-vis some property is reasonable only if one is in a position to reasonably believe that the sample is representative of the group vis-à-vis that property. (See Alston 1996: 326–7; Howard-Snyder 1996b: 297–9; Tucker 2014: 49ff.) for worries about how best to formulate such a principle.) Again, this principle can be intuitively supported by cases. If I plant in my garden only one kind of pepper—say, habaneros—then it would be unreasonable and irresponsible for me to generalize to all kinds of hot peppers.

Skeptical theists then apply something like that principle to the inference from NKG to NaaG . It is reasonable to infer NaaG from NKG only if it is reasonable for us to believe that the outweighing goods we know of are representative of all potential outweighing goods when it comes to the property of justifying God’s permission of evil.

But skeptical theists argue that it is not reasonable for us to believe that. Some skeptical theists reason by analogy. Sometimes, it is reasonable to believe our sample size is representative because it is sufficiently large and chosen at random. Other times, we have already “charted” the relevant group and we know what proportion our sample is from that group. But these are not the case here. The outweighing goods we know of aren’t picked at random. Nor have we “charted” the relevant group of all possible God-justifying outweighing goods. The standard techniques for generating representative samples don’t seem to apply to NKG .

Bergmann summarizes these ideas in a more careful statement as follows (2001: 279; cf. 2009: 376):

Notice that these theses do not claim that we have good reason for thinking they are not representative. They are more skeptical than that. Bergmann sees these claims as independently plausible, so that both theists and non-theists should endorse them, writing: “it just doesn’t seem unlikely that our understanding of the realm of value falls miserably short of capturing all that is true about that realm. One can recognize this even if one is not a theist.” (2001: 279).

Thus, given a relevant epistemic principle, the inference from NKG to NaaG is reasonable only if it is reasonable to believe that the outweighing goods we know of are representative of all outweighing goods when it comes to justifying God’s permission of evil. But, skeptical theists argue, that is not reasonable for us to believe. So the inference from NKG to NaaG is not reasonable. Thus, this representation-based inductive argument for N-O 1.2 also fails.

Skeptical theists object to the inference from NKG to NaaG . But some skeptical theists also object directly to NKG (see Alston 1996: 315-6; Plantinga 1998: 53; Bergmann 2009: 382). Given ST 3 , it may be that some outweighing goods we know of would justify God’s permissions of the fawn’s suffering and the girl’s suffering. It may simply be that the connection between the outweighing goods we are familiar with and God’s permission of suffering is not known to us. Again, skeptical theists might appeal to analogies to clarify their position. A novice chess player knows that a grandmaster’s objective is to win the game. But it doesn’t follow that the novice chess player understands how each move the grandmaster makes might contribute to the grandmaster’s objective. In fact, it might even seem to the novice player that various moves by the grandmaster don’t contribute to that objective at all.

3. Noseeum Arguments

Noseeum evils (or seemingly unjustified evils) are a type of evil—evils such that God’s permission of them do not seem necessary for an outweighing good. One might argue that those evils are good evidence for another type of evil that is inconsistent with the existence of God—evils such that God’s permission of them would not be necessary for an outweighing good. Alternatively, one might argue that noseeum evils are evidence against theism on their own—regardless of whether or not they are also evidence for evils of another kind. This section looks at two arguments of that kind and skeptical theistic critiques of them.

To understand the first argument, some background in confirmation theory will be useful. On one standard understanding, a claim \(E\) is evidence against a hypothesis \(H\) just in case the probability of \(H,\) given \(E,\) is less than the probability of \(H\) on its own. More formally: \(E\) is evidence against \(H\) just in case \(Pr(H\mid E) \lt Pr(H)\). Furthermore, the lower the difference between \(Pr(H\mid E)\) and \(Pr(H),\) the stronger \(E\) is evidence against \(H\). To determine if \(E\) is evidence against \(H,\) one considers two key conditional probabilities: first, the probability of \(E\) given \(H,\) and second, the probability of \(E\) given the denial of \(H\). More formally, \(Pr(E\mid H)\) and \(Pr(E\mid \neg H),\) respectively. Now if \(Pr(E\mid \neg H)\) is greater than \(Pr(E\mid H)\) then \(E\) will be evidence against \(H\). And the greater \(Pr(E\mid \neg H)\) is than \(Pr(E\neg H)\) the stronger \(E\) will be evidence against \(H\).

Next, reconsider:

No-Known-Goods ( NKG ) No outweighing goods we know of would justify God’s permission of the fawn’s suffering and the girl’s suffering.

(We’ll ignore the objection of 2.3 and assume NKG is true.) NKG is a noseeum claim, since it asserts that we “no see” any outweighing goods that would justify God’s permission of that specific suffering. To determine if NKG is evidence against theism, we need to consider two conditional probabilities: how likely is it given the denial of theism and how likely is it given theism. Rowe (1996: 267–70) argues that the probability of NKG given atheism is 1. For if there is no God, then God doesn’t permit anything. Thus, if there is no God, then nothing justifies God’s permission of anything, since God doesn’t permit anything. Thus, if there is no God, then no outweighing good we know of would justify God’s permission of anything—including the fawn’s suffering and the girl’s suffering.

If the probability of NKG given atheism is 1, then the lower probability it has given theism, the stronger the evidence NKG is against theism. And one might argue that the probability of NKG given theism is not high—at least below 0.5—because God would not let us suffer in silence. That is, if God permitted the suffering of the fawn and the girl for some outweighing good, then God would at least tell us what those goods are or—if that’s not possible for whatever reason—at least offer some kind of comfort for the suffering in some way (cf. Rowe 2006: 86–7). So given theism, we shouldn’t expect that no good we know of would justify God’s permission of these evils—God would tell us about those goods or at least comfort us.

Skeptical theists object to this argument that the probability of NKG , given theism, is not high (cf. Bergmann 2001). Presumably, God would not let us suffer in silence because such silence serves no outweighing good. But it is hard to see how one could establish that it serves no outweighing good, if ST 1 and ST 2 are true. After all, even if no good we know of would justify God letting us suffer in silence, given ST 1 and ST 2 , it would not be reasonable to infer that no such good would. So this argument fails to show that NKG is not high given theism.

An alternative argumentative strategy doesn’t focus on individual cases of evils that seem such that God’s permission of them is unnecessary for an outweighing good. Rather, it focuses on a larger class of such evils. One kind of argument that focuses on a larger class of these evils is a cumulative case argument like this:

But some skeptical theists object to N 1.1 on general methodological grounds (see Perrine and Wykstra 2014: 157–163). Evidence of a specific type might disconfirm a hypothesis by eliminating some specific version of that hypothesis that does not fit that type of evidence. But that evidence won’t disconfirm other specific versions of that hypothesis that do fit that type of evidence. Thus, evidence of the same type might initially disconfirm a hypothesis by eliminating specific versions of it; but that type of evidence won’t continue disconfirming that hypothesis, since the only remaining versions of the hypothesis accommodate that evidence. To give an example, the sun rises and sets every day. This type of evidence disconfirms heliocentrism by disconfirming a specific version of heliocentrism on which the earth revolves around the sun while always facing the sun. But this type of evidence won’t disconfirm other versions of heliocentrism on which the earth revolves around the sun, but regularly spins on an axis. Thus, the sun rising and setting each day might initially disconfirm heliocentrism—by eliminating some versions of heliocentrism—but that type of evidence won’t continue disconfirming heliocentrism. For the remaining versions of heliocentrism fit that type of evidence.

Analogously, some skeptical theists might allow that the existence of some seemingly unjustified evils disconfirms theism. Insofar as they do disconfirm theism, it is by disconfirming specific versions of theism on which God always explains, or indicates, the justifying reasons for the permission of evils. But the existence of seemingly unjustified evils won’t disconfirm versions of theism on which God’s justifying reasons for permitting suffering are frequently, even normally, beyond our ken. Thus, this type of evidence—the existence of seemingly unjustified evils—might disconfirm some versions of theism but won’t disconfirm other versions of theism. So, it won’t be the case that each seemingly unjustified evil is slight disconfirmation for theism in general and N 1.1 is false.

Skeptical theism is one family of responses to arguments from evil. But there are a number of other responses. Following Plantinga’s terminology (1974: 28), a theodicy purports to provide God’s justifying reasons for permitting suffering and evil. By contrast, a defense purports to provide potential reasons that would justify God’s permission of suffering and evil, even if those reasons may not be God’s own. By contrast, a skeptical theistic critique does not offer God’s actual or potential justifying reasons for permitting suffering and evil. Indeed, skeptical theists oftentimes find themselves agreeing with atheologians that they cannot identify connections between potential goods and evils that would justify God’s permission of specific evils.

Initially, it may seem that there is a tension, if not inconsistency, between skeptical theism on the one hand and defense and theodicies on the other (compare Stump 2010: 13–5; Coley 2015). However, there are many types of arguments from evil, utilizing different facts about evil. For the same argument from evil, utilizing specific facts about evil, it may be inconsistent to combine responses. That is, it may be inconsistent to respond with both (i) we know of reasons that do (or could) justify God’s permission of that type of evil and (ii) we don’t know of any reasons that do (or could) justify God’s permission of that type of evil. But some skeptical theists (e.g. Perrine and Wykstra 2014; DePoe 2014: 33) suggest that it may be reasonable to use different responses for different arguments from evil. Whether skeptical theistic critiques can be combined with other responses to various arguments from evil is a place in the literature that needs more attention.

Many authors who utilize skeptical theistic critiques also assume that God has obligations to us. That is, they tend to assume that God is permitted in letting us experience suffering and evil only if God has some kind of justifying reason. Many skeptical theists accept N-O 1.1 ; they assume the justifying reason must involve outweighing goods. Others might reject N-O 1.1 . Nonetheless, they think God must still have some kind of justifying reason; the reason just may not involve outweighing goods (compare van Inwagen 2006 (lecture 6), Mooney 2019). The claim that God has obligations to us is not normally a requirement of skeptical theistic critiques. Nonetheless, the particular way that skeptical theists critique arguments from evil indicates they assume that God does have obligations to us.

Recently, there have been other, more extreme responses to arguments from evil that deny that God has obligations to us. Various arguments have been offered for this view (see Adams (2013) and Murphy (2017) for some representative examples). But these responses deny that from the claim that God exist it necessarily follows that God would promote goods, prevent evils, or permit evils only if God has a justifying reason. These responses may be technically consistent with skeptical theistic responses, since the different responses will frequently object to distinct premises. Nonetheless, these responses do not sit well with the assumptions skeptical theists tend to bring to bear in evaluating arguments from evil.

5. Epistemological Objections

A number of authors have raised epistemological objections to skeptical theism. Some of these objections are to the epistemic principles skeptical theists use. Others are that the claims of skeptical theistic critiques may, themselves, produce skeptical problems.

CORNEA principles state a necessary condition on it being reasonable to claim or believe p on the basis of something q . The condition (roughly) is that it is also reasonable for one to believe that, given one’s cognitive faculties, if p were different, then q would likely be different as well. A standard worry for epistemic principles like this is that they will create an infinite regress (cf. Alston 1980, Fumerton 1995 (chapter 2), Leite 2008). And some have objected to CORNEA principles for exactly this reason (Swinburne 1998: 32–3). To see how the regress might go, suppose I’m reasonable in believing p on the basis of q . Then, by a CORNEA principle, I am also reasonable in believing that, given my cognitive faculties, if p were false, then q would likely be different. But suppose that belief is reasonable but based on something r . Well, applying a CORNEA principle again, it must be reasonable for me to believe that, were that belief different, then r would likely be different… ad infinitum .

A different objection is that CORNEA-principles are inconsistent with forming reasonable beliefs on the basis of inductive evidence (McBrayer 2009; Almeida 2014: 122ff). For instance, suppose the day care center informs me that each day, for the past three months, my child has been sleeping during nap time. On the basis of this inductive evidence, it is reasonable for me to infer that today my child will sleep during nap time. But if my child were not sleeping during nap time today, would my inductive evidence be different? Using standard possible-world semantics, it would not. For in the closest worlds in which my child is not sleeping, it is likely a fluke or random event. And in those worlds, I would still have my inductive evidence from the past. Thus, in this case, S-C implies that it is not reasonable to infer my child is sleeping today on the basis of my inductive evidence. But this example of inductive reasoning is not particularly distinctive. So S-C is inconsistent with forming reasonable beliefs on the basis of inductive evidence.

Several attempts have been made to clarify CORNEA-principles in ways that avoids these problems while still being strong enough for responding to arguments from evil (see Wykstra 1996, 2007; Wykstra and Perrine 2012; Perrine 2022). One clarification is that CORNEA principles are restricted to a specific kind of case: where the grounds or evidence for belief is “levering evidence,” which is (crudely put) evidence strong enough to reasonably “lever” one from agnosticism or disbelief about p to belief about p . The restriction of CORNEA-principles to this type of evidence or grounds is appropriate, since it is the kind of evidence that Rowe (1979) initially tried to provide. Restricting CORNEA-principles in this way may also block an infinite regress, since the CORNEA-principles are not fully general. A second clarification is that the condition in CORNEA-principles should not be interpreted as involving a subjunctive-conditional, but rather a conditional probability. That is, the condition is that it is reasonable to believe, given one’s cognitive faculties, that the conditional probability of one’s grounds or evidence, given one’s belief is false, is below .5. Though some of the details are technical, Wykstra and Perrine argue that a probabilistic understanding of CORNEA’s requirement handles cases of inductive evidence. But whether or not these clarifications ultimately help defuse objections is still a topic of debate. (For further discussion of CORNEA-principles, see Howard-Snyder (1992), Graham and Maitzen (2007), Stone (2011), Draper (2014a).)

Some authors have worried that skeptical theism is inconsistent with “Common Sense Epistemology” (see Dougherty 2008). More precisely, some might worry that the epistemological principles used in skeptical theistic critiques are inconsistent with epistemological principles that are part of Common Sense Epistemology. Such principles include Swinburne’s Principle of Credulity (2004: 303ff.) and Huemer’s Phenomenal Conservativism (2001: 99):

Principle of Credulity If it seems to a subject S that p , this is good grounds for S to believe p .

Phenomenal Conservativism If it seems to a subject S that p , then S is prima facie justified in believing p .

One might object to skeptical theism by arguing that these principles of Common Sense Epistemology are true, and thus their inconsistency with skeptical theism is a problem for skeptical theism.

In response, some authors argue that skeptical theistic critiques are consistent with various principles of Common Sense Epistemology. For even if one is prima facie justified in believing something, one’s prima facie justification might be defeated. For instance, I may be prima facie justified in believing my boss’ explanation of the new departmental policy just because he said as much; but that prima facie justification might be defeated when I recall the many times he has misled me about such policies. So too, skeptical theistic critiques might permit that individuals are prima facie justified in believing things like N-O 1.2 . Nonetheless, that prima facie justification is defeated by considering skeptical theistic critiques of arguments from evil. (For discussion, see Matheson 2011, 2014; McBrayer and Swenson 2012: 147; Coffman 2014: 83–4; Tweedt 2015; Rutledge 2017a; Hendricks 2018.)

Alternatively, other authors argue that the epistemological principles of skeptical theistic critiques are inconsistent with various principles of Common Sense Epistemology (see Wykstra 1984, Senor 2013, Perrine 2022). These authors argue that the inconsistency is not a problem for skeptical theistic critiques because those principles are false. For instance, Perrine argues that there are reasons independent of arguments from evil for thinking that these various principles of Common Sense Epistemology are false. For those principles are consistent with an agent having prima facie justification for believing something even if the belief was not formed in a responsible or reliable manner.

A frequent criticism of skeptical theism is that it leads to skepticism about the external world (see Russell and Wykstra 1988: 158–9; Weisberger 2007: 170–171; Wilks 2013; Law 2017; Russell 2017; 2018: 111ff). This criticism turns on the possibility of divine deception. More specifically, consider the following hypothesis:

Deceiving God ( DG ) God is deceiving us into believing the earth is extremely old, when in reality it is quite young.

If DG is true, then we are mistaken about the age of the earth. Now atheists can reject DG because it implies God exists; but clearly theists who are skeptical theists cannot reject DG for that reason. Theists, who are not skeptical theists, might reject DG by reasoning as follows: we can see no reason that would justify God in deceiving us in this way; so there is no such reason. But theists who are skeptical theists cannot rely on that line of reasoning. After all, as learned from their critiques of noseeum inferences, from the mere fact that it seems to us that God has no reason to (e.g.) deceive us about the age of the earth, it does not yet follow that we are prima face justified in believing God has no reason to deceive us about the age of the earth. So skeptical theists who are theists cannot use either of these line of reasoning to rule out DG . But this claim about the age of the earth is merely illustrative of any number of claims about the external world. Thus, skeptical theists who are theists cannot use either of these lines of reasoning to rule out any number of skeptical hypothesis about the external world that involve divine deception.

One type of response is to argue that, for skeptical theists to know about the external world, they do not have to first rule out hypotheses like DG . Otherwise put, there are other routes to knowledge about the external world that don’t involve inferences about potential divine deception (cf. Bergmann 2009: 391; 2012: 15). For instance, skeptical theists might embrace externalism about knowledge. According to that position, in order for one to know facts about the external world, one does not need to have access to what explains that one knows. Thus, one could know facts about the external world without producing an argument or first ruling out that one is not being deceived (cf. Bergmann 2008, Hendricks 2020). But there are other routes skeptical theists might explore. For instance, they might accept theories of perception on which we are antecedently justified in trusting our perceptual experiences, unless we are given evidence that our experiences are misleading in some way (cf. Boyce 2014). Such antecedent justification need not require first ruling out skeptical hypotheses like DG .

A similar objection is that skeptical theism leads to religious skepticism (Beaudoin 2000, 2005; Rowe 2006: 90–1; Wielenberg 2010; Hudson 2014). Skeptical theists might respond to the previous criticism by claiming that we have many ways of knowing about the external world that don’t, on the face of them, involve God telling us something and not deceiving us. But a staple of many religious traditions is that there are certain things we know about only because God has told us. For instance, consider a claim like:

Afterlife ( Af ) There is an eternal afterlife of union with God, full of bliss and free of suffering.

Many religious traditions hold that, insofar as we know or are reasonable in believing Af , it is only because God has told us, revealing it to us through a prophet, text, or direct religious experience.

Critics argue that theists who are skeptical theists are not reasonable in believing Af . The criticism here mirrors the one of the previous section. It is possible that God is deceiving us about these matters; but skeptical theists are unable to reasonably rule out the possibility of divine deception. Additionally, in the case of the external world, agents may have a variety of ways of coming to know about the external world. But that kind of response is less plausible here. Since it seems that for claims like Af , our only way of knowing them is to depend upon God—and depend upon God not deceiving us.

A different type of criticism focuses on knowledge of God’s commands. One starting point for this criticism is that many religious texts and traditions contain a wide range of commands that purport to be divine commands (Maitzen 2007). Nevertheless, most religious adherents do not think they are obligated to follow all of those purported commands. Perhaps the commands are issued by the wrong text or tradition; or perhaps the commands had been “superseded” or “updated” by God; or perhaps the commands seem to be issued by God but couldn’t be given the immoral content of the commands; or maybe the commands are just plain silly or arbitrary.

A natural way that a theist might sift through these purported commands is to consider what God would have reason to command. If the theist sees no reason why God would command (e.g.) not cooking goat in its mother’s milk, then the theist might conclude that such a command is not actually God’s. But this way of reasoning about divine commands looks incredibly similar to the noseeum argument for N-O 1.2 ! Thus, if theists who are skeptical theists are skeptical about whether that noseeum inference is reasonable, so too they should be skeptical that this inference about God’s commands is reasonable as well.

6. Moral Objections

A number of authors have argued that skeptical theism—especially when combined with theism—leads to various forms of moral problems, including moral skepticism. However, there are several distinct criticisms that are not always distinguished. I will distinguish three important objections involving morality.

One criticism is that theists who accept skeptical theism should deliberate about acting in problematic ways (see Almeida and Oppy 2003: 505–7; Piper 2007: 71ff; Rancourt 2013). More specifically, they should reason that they should never intervene in preventing seemingly unjustified evils, even if it would be easy for them to do so. For (as theists) they believe that for any seemingly unjustified evil, there is a justifying reason for it. They might fail to identify the justifying reason for some seemingly unjustified evil. But (as skeptical theists) they do not regard this failure as providing prima facie justification that the justifying reason does not exist. So they should reason that they should not intervene in preventing seemingly unjustified evils. But, the objection goes, it is clearly unreasonable —and immoral —to reason that we should never prevent a potential seemingly unjustified evil.

This basic objection can be extended. For instance, theists who are skeptical theists frequently believe that they have been commanded by God to act in various ways. Thus, in some concrete situation, a theist who is a skeptical theist may believe both that she should not prevent the suffering (because of her skeptical theism) and that she should prevent the suffering (because of her theism). The combination of her views leads to a kind of moral paralysis or aporia (see Piper 2007: 72–3). (Again, for the reasons mentioned in section IV.3, atheists and theists who are not skeptical theists have straightforward ways of resisting these objections.)

Skeptical theists have offered a number of responses. Some skeptical theists who are theists claim that God has commanded them to act in various ways. Therefore, when deliberating about how to act, it is reasonable to prevent evils since they have been commanded by God to prevent evils (compare Bergmann and Rea 2005: 244–5, Schnall 2007). This response requires knowledge or reasonable belief of God’s commands; but some critics of skeptical theism claim that skeptical theists who are theists do not have knowledge of God’s commands (see previous section).

Other skeptical theists argue that the objection misrepresents their position (Anderson 2012). Specifically, it neglects a distinction Wykstra drew (1984: 75–6) between different justifications for permitting evil (cf. Alston 1991 [1996: 101]). God may be justified in permitting an evil because the evil itself is necessary for some greater good. But it may be that God is justified in permitting an evil because the possibility of the evil—that is, not preventing it—is necessary for some good, such as developing moral character. A parenting analogy is apt here. A parent is able to prevent an embarrassingly low score on her child’s school work by forcing her child to do the homework. But by permitting the possibility of that low score, the parent gives her child an opportunity to achieve something of value—a more responsible character. In this case, the suffering—the embarrassingly low score—is not necessary for any further good; rather, the parent’s permission of it is necessary for a further good, namely, increased responsibility. Thus, some skeptical theists reject the claim that for any evil, there is a greater good that outweighs that evil. Rather, God may permit some evil may occur, not because that evil is necessary for some greater good, but because God’s permission of it is.

Some have objected that skeptical theism is committed to all-things-considered value skepticism (Jordan 2006, Hasker 2010, Ekstrom 2021: ch. 4). Crudely put, the all-things-considered value of something—a state of affairs, action, attitude, etc.—is the overall value or disvalue of it and all of its consequences. Some have argued that if skeptical theism is true, then it is frequently not reasonable to make all-things-considered value judgements about specific things, that is, specific states of affairs, actions, attitudes, etc. Given skeptical theism, we are frequently reasonable in judging that some specific thing is on its own valuable or disvaluable, good or bad. But, the objection goes, we are not reasonable in judging that things have all-things-considered value. Again, this objection might be extended to cases of obligations, since being unable to make reasonable all-things-considered value judgments may also undermine knowledge of our obligations (see next section).

In response, some skeptical theists concede their position undermines certain ways of arriving at all-things-considered value judgments (cf. Bergmann 2009: 379–80, 389; 2012: 25–6). Specifically, just as skeptical theistic claims undermine Rowe’s nosseum inferences, so too they will undermine nosseum inferences that certain things are all-things-considered valuable or disvaluable. But skeptical theists might also maintain that this all-things-considered value skepticism is unproblematic. One reason why it might be unproblematic is that there are reasons for accepting all-things-considered value skepticism independently of skeptical theism. For instance, no one knows all of the causal consequences of their action. Thus, no one knows all of the valuable or disvaluable consequences of their actions. Thus, no one knows the all-things-considered value of their actions. This line of reasoning supports all-things-considered value skepticism, but does not assume that skeptical theistic critiques are successful.

Many authors have argued that skeptical theism leads to skepticism about our obligations (see Jordan 2006: 409ff; Piper 2007: 68; Sehon 2010; Maitzen 2013: 448–451; Street 2014). Skeptical theists can be reasonable in believing that certain things are good and bad; and skeptical theists can be reasonable in believing that agents have certain pro tanto obligations (e.g. to bring about what is good, prevent what is bad, keep our promises, etc.). But oftentimes in ordinary life we take ourselves to reasonably believe—if not know—what we ought to do, what our all-things-considered obligations are. And some authors argue that skeptical theism leads to skepticism about our all-things-considered obligations.

The simplest form of this objection assumes maximizing act consequentialism. According to that view, an agent ought to act to bring about the outcomes with the greatest possible good. However, if skeptical theists cannot be reasonable in forming beliefs about the all-things-considered value of outcomes, then they cannot be reasonable in believing which outcomes have the greatest possible good. Thus, assuming maximizing act consequentialism, skeptical theists cannot be reasonable in believing some action is the one they ought to perform. But, arguably, skeptical theism leads to skepticism about our all-things-considered obligations assuming other moral theories as well. For instance, some robust forms of virtue ethics suggest that agents ought to perform the action that a perfectly virtuous person would perform; but presumably a perfectly virtuous person would not neglect salient outcomes of their actions. Similarly, some forms of Kantian ethics test moral principles by considering whether “universalizing” them in various ways results in some kind of contradiction; but presumably determining whether there is some kind of contradiction requires knowing the various outcomes of actions in those hypothetical scenarios.

Skeptical theists might respond by drawing a distinction between considerations relevant to God’s action and our own (compare Howard-Snyder 1996b: 292-3; Bergmann 2009: 393; Jeffrey 2019: section 3.1.3). For all we know, our knowledge of good and evil is not representative when it comes to considerations that determine what God ought to do; but it may be that our knowledge of good and evil is representative when it comes to considerations that determine what humans ought to do. While this response may be promising, skeptical theists haven’t developed it at great lengths, and some critics (notably Street 2014) have argued it is less promising than it might seem.

Howard-Synder (2009, 2014) offers a dilemma argument in defense of skeptical theism. On some moral theories, the facts that determine our all-things-considered obligations are frequently inaccessible to us. (Maximizing act consequentialism is one example.) Thus, given skeptical theism and those theories, we should frequently be in doubt about our all-things-considered obligations—but not in virtue of skeptical theism, but rather in virtue of those theories. By contrast, on other moral theories, the facts that determine our all-things-considered obligations are frequently accessible to us. Thus, given skeptical theism and those theories, we frequently should not be in doubt about our all-things-considered obligations—but not in virtue of skeptical theism, but rather in virtue of those theories. Either way, skeptical theism itself doesn’t lead to skepticism about our all-things-considered obligations.

7. Scope of Skeptical Theism

Many skeptical theistic critiques focus on “noseeum” inferences. These inferences might be used to infer that God’s permission of some evil lacks a justifying reason or even that there is no God. A natural thought is that skeptical theistic critiques, if successful at all, are only successful against arguments that make a “noseeum” inference. Other arguments from evil, using alternative lines of reasoning, may be immune from skeptical theistic critiques. This natural thought is not a criticism of skeptical theistic critiques per se . Rather, it suggests a certain limitation of skeptical theistic critiques. Such critiques have a limited scope in defending theism from arguments from evil.

A number of authors have argued that there are arguments from evil that are immune to skeptical theistic critiques. We next focus on three types of arguments—non-inferential arguments, Humean arguments, and Hiddenness Arguments.

Non-inferential arguments from evil begin with a standard observation. For a given proposition, sometimes we can be either inferentially justified in believing it—perhaps on the basis of some beliefs or argument—or non-inferentially justified in believing it. For instance, I am inferentially justified in believing my partner is home by inferring this from other things I am justified in believing like: the door is unlocked, the mail has been taken, her keys are on the ring, and her jacket is hanging on the wall, etc. But I can also be non-inferentially justified in believing my partner is home by seeing her in the home as well.

One way of defending N-O 1.2 — there is at least one evil e such that God’s permission of e is not necessary for some outweighing good—is inferentially, by giving an argument. Skeptical theists have critiqued various arguments for N-O 1.2 . But even if those critiques succeed, then at best what follows is that we do not (yet!) have inferential justification for believing N-O 1.2 . But it is conceivable that we have non-inferential justification for believing N-O 1.2 .

However, a non-inferential argument from evil still needs to show that we do have non-inferential justification for believing N-O 1.2 (or a similar kind of claim). While there might be a variety of ways of doing this (see Gellman 2013, 2017), a simple and straightforward way is to appeal to epistemic principles of Common Sense Epistemology like Phenomenal Conservativism (see Tucker 2014: 58f; Ekstrom 2021: 103ff). However, this way of defending a non-inferential argument from evil gets embroiled in the issues of section IV.2. Specifically, some skeptical theists will simply reject the epistemic principles; others may concede them but argue that skeptical theistic critiques provide defeaters. Regardless, whether or not non-inferential arguments from evil avoid skeptical theistic critiques—and are otherwise successful—is one of the more active areas of research on this topic in recent years.

Some of the most powerful arguments from evil are Humean arguments from evil (see Hume 1779 [2007], Draper 1989, Draper 1992: 313–317, Dougherty and Draper 2013, Morriston 2014). Humean arguments have several distinctive features. First, they appeal to facts about good and evil , such as the distribution of pleasure and pain we observe. Second, they are comparative: they compare theism with some alternative, inconsistent hypothesis like a “hypothesis of indifference” (roughly: it is not the case that life on earth as we observe it is the result of actions by either a benevolent or maleficent creator). Third, they are normally abductive, arguing that the inconsistent hypothesis does a better job of explaining certain facts than theism.

Humean arguments normally have a central premise like:

Distribution ( Dist ) The hypothesis of indifference does a much better job predicting the distribution of pleasure and pain we observe than theism.

On its own, Dist isn’t enough to make it reasonable to believe theism is false; after all, perhaps the hypothesis of indifference is worse off in other respects—it is very implausible or we have other excellent reasons for accepting theism (cf. Plantinga 1996: 247–250). So defenders of Humean arguments normally add additional premises to their argument to show that their rival to theism isn’t worse off in other respects as well (see Dougherty and Draper 2013: 69ff; Morriston 2014: 227). But the inclusion of a comparative premise like Dist is part of what makes Humean arguments distinctive.

Dist may be defended in various ways. One standard way appeals to our background knowledge of the biological role of pleasure and pain (Draper 1989). Simplifying, we antecedently know that pleasure—a good—and pain—an evil—play biological roles. For instance, pleasure will play a role in reinforcing adaptive behavior, whereas pain will play a role in deterring non-adaptive behavior. The hypothesis of indifference does not undermine predictions about the biological role of pleasure and pain. But theism will. For given theism it is expectable that pleasure and pain would play additional roles beyond biological ones such as moral or religious roles (e.g., the virtuous receive pleasure, the vicious pain). So theism, but not the hypothesis of indifference, will undermine the predictions about the biological role of pleasure and pain. Further, the argument goes, those predictions are accurate. Thus, the hypothesis of indifference does a better job of explaining Dist than theism.

Most skeptical theists object to claims like Dist by focusing on what theism predicts or explains. Some skeptical theists express general skepticism of our ability to effectively compare these hypotheses. For instance, Perrine (2019) argues that if theism is true, then skeptical theistic claims—like ST 1 – ST 3 from above—are also very likely true. Thus, if Dist is true, then so is:

Distribution+ ( Dist+ ) The hypothesis of indifference does a much better job predicting the distribution of pleasure and pain we observe than theism and skeptical theistic claims.

But it is doubtful that we are able to effectively predict the pleasure and pain we might observe given both theism and skeptical theistic claims. For skeptical theistic claims undermine our ability to create precise enough predictions about theism to effectively compare them with other hypotheses like the hypothesis of indifference.

Other skeptical theistic critiques of Dist focus on specific hypotheses that, it seems, we cannot reasonably rule out. For instance, consider the following claim (compare Van Inwagen 1991 [1996]; and 1996: 225–234):

Necessary Distribution ( ND ) A distribution of pleasure and pain for organisms similar to the one we observe is necessary for great goods associated with higher-level organisms and for avoiding disvaluable worlds that are massively irregular.

If ND is very probable, given theism, then it is not unlikely that we would observe a distribution of pleasure and pain similar to the one we observe. For if theism is true, God likely wants to bring about various greater goods. But if ND is true, God couldn’t bring about those greater goods without allowing for a distribution of pleasures and pains similar to the one we observe. Thus, if ND is very probable, given theism, then theism may predict the distribution of good and evil we observe just as well as the hypothesis of indifference. Thus, if it is reasonable to believe Dist it must also be reasonable to believe that ND is not very probable, given theism. But that hypothesis—that ND is not very probable, given theism—is not one we can reasonably rule out. To do that, we would have to have a greater understanding of goods, evils, their connections to organisms, laws of natures, irregularities, etc. than we in fact have. So it is unreasonable to believe Dist , since to reasonably believe it we would have to reasonably believe ND is not very probable given theism, and that’s not reasonable for us to believe.

Divine hiddenness arguments use facts about the apparent “hiddenness” or “absence” or “silence” of God to argue against the existence of God (cf. Schellenberg 1993, 2007, 2015). A simple argument from divine hiddenness is:

Non-resistant, non-belief refers to people who meet two conditions. First, they do not believe in God. Second, they are not resistant to belief in God, where not being resistant may include both having the conceptual and emotional abilities to have a relationship with God as well not having intentionally and culpably resisted such a relationship. H 1 can be defended in various ways, including: according to theism, being in a personal relationship with God is the greatest good for people. However, the argument goes, such a personal relationship requires belief. Thus, if God exists, and a person is not resistant to such a relationship—the person’s greatest good—then God would “make his presence known” to the person so they might reasonably believe that God exists and enter into such a relationship. In other words, God would not remain hidden from those who are receptive to a relationship with God. But H 2 claims there are such people—who are not resistant yet do not believe. Thus, there is no God.

Unsurprisingly, some authors claim this argument for H 1 is vulnerable to skeptical theistic critique (see Bergmann 2009: 382, McBrayer and Swenson 2012). So far as we can tell, there is no reason that would justify God in remaining hidden to people who are receptive to a relationship with God. But from the mere fact that we cannot identify such a reason, it does not follow that there is no such reason. Thus, attempts to establish H 1 by appealing to our knowledge of how God might plan to bring about certain goods—like a loving personal relation—may be vulnerable to skeptical theistic critique.

Theism is a cornerstone of many major religious traditions. But it is also a metaphysical view, being an ontological claim about what exists. It is also one potential way of filling in a “supernatural” worldview on which the world is not exhausted by “natural” things but also contains “supernatural” things. Many philosophers evaluate metaphysical claims in a manner similar to scientific claims, such as their ability to explain phenomenon, be well-confirmed, exhibit theoretical virtues (such as simplicity), etc.

However, there is a well-known lesson in scientific theorizing about how not to avoid a theoretical problem. Suppose we are evaluating a hypothesis H (say, Mars revolves around the Earth). But there is some data D that seems to strongly disconfirm H (say, Mars exhibits “retrograde motion”). One could expand or enrich H by adding further hypotheses to H that would predict the data D (say, Mars orbits the earth using epicycles on an orbit). The resulting expanded or enriched hypothesis may perfectly well predict the data D . But if the resulting expanded or enriched hypothesis is random, unprincipled, independently implausible, or otherwise ad hoc , it may not be plausible to believe the expanded hypothesis in virtue of these additions or expansions. The lesson here—crudely put—is that one should not respond to a problem for a hypothesis by just adding claims to that hypothesis. At the very least, one should also explain why those additional claims are also expectable given the hypothesis in question. In that way, one can show that the additional claims are not random, unprincipled, or ad hoc .

Among other things, theism is a metaphysical hypothesis. Many authors have argued that various facts about good and evil are strong evidence that disconfirms this hypothesis. In response, some authors have adduced skeptical theistic critiques, composed of further claims about God and human limitations. However, the lesson from above is that simply adding further claims is insufficient to defend a hypothesis from a theoretical problem. Thus, if these skeptical theistic claims are random, ad hoc , or unprincipled when added to theism, then even if the expanded theism adequately neutralizes the disconfirming evidence, the expanded theism may be too implausible to be reasonably believed.

Though not always framed in this way, skeptical theists have been sensitive to these points. Some skeptical theists argue that their skeptical theistic claims are independently plausible—that is, plausible independent of whether or not theism is true. Other skeptical theists argue that their skeptical theistic claims are quite plausible given theism . Skeptical theists offer various defenses, by appealing to considerations of human limitations or moral progress or ruminations about historical complexity or analogies involving expertise in various fields (see, inter alia , Alston 1996: 317ff; Howard-Snyder 1996b: 301ff; Durston 2000; Bergmann 2001: 284–6; Perrine 2019: 124–126; Hendricks 2020). Either way, these arguments can be used to try to show that skeptical theistic claims are not random, ad hoc , or unprincipled ways of “expanding” or adding to theism.

The issue of how to develop a theory was incipient in some of the exchanges between Rowe and Wykstra (see Rowe 1986: 246–7; Rowe 2006: 85ff; Wykstra 1984: 91ff; 1996: 142–5; Russell and Wykstra 1988: 159–160; Wykstra 1996: 140ff). For instance, Rowe concedes that certain claims are entirely expectable given theism, such as: there might be goods and evils beyond our ken but not beyond God’s. But other claims are not expectable given theism, such as: there is suffering that is necessary for goods beyond our ken and God would not tell us this. Rowe defends these claims by appeal to Wykstra’s parent-child analogy. A parent may allow a child to undergo a painful surgery for reasons that the child cannot yet understand; but a loving parent will surely not be absent while the child suffers (see also Dougherty 2012, Rutledge 2017b). Wykstra argued that such claims are expectable given theism by means of his analogy with scientific depth. More recently, other authors have tried to give other reasons why, even given theism, we should expect seemingly unjustified suffering without divine comfort or acknowledgement of that suffering (see DePoe 2014, 2017; Shields 2021)). These disputes can be understood as disputes over whether theism can be “expanded” by skeptical theistic considerations in a way that is both plausible and undermines arguments from evil, or whether the only plausible and principled ways of expanding theism do not undermine such arguments (compare Perrine and Wykstra 2014, 2017: 104–7).

The issue of theory development is important for two further reasons. First, many skeptical theists defend their claims and critiques by arguing that critics misunderstand them or that their positions are analogous to other cases that are perfectly unproblematic. However, as philosophers of science like Kuhn and Lakatos have suggested, many of our best scientific theories offer recommendations (implicitly or explicitly) for how to develop or expand hypotheses. Skeptical theists have spent less time making positive proposals for how to develop or expand a theistic worldview. Unless they do so, some critics may worry that skeptical theism makes theism into a “degenerating research program” (to use Lakatos’ charming phrase).

Second, many skeptical theists are not simply theists; they are adherents to some religious tradition. Religious traditions normally make many more claims than (mere) theism, and religious traditions normally reject other alternative religious traditions. Some philosophers of religion—such as Richard Swinburne—have continued the project of natural theology to not only defend theism but a specific religious tradition as well (Christianity, in Swinburne’s case). But many—including some skeptical theists—worry that such projects are undermined by skeptical theistic considerations. If so, skeptical theists who are theists need to explore alternative routes for explaining the justification of religious belief, such as reformed epistemology (see Plantinga and Wolterstroff 1983; Plantinga 2000). At the very least, more work could be done exploring positive models that incorporate justified belief in both skeptical theism and specific religious traditions.

Adams, Marilyn McCord, 2013. “Ignorance, Instrumentality, Compensation, and the Problem of Evil,” Sophia , 52: 7–26.
Almeida, M. J., 2014. “Skeptical Theism and Undercutting Defeat” in Dougherty and McBrayer (eds.) 2014, 115–131.
Almeida, M., and Graham Oppy, 2003. “Sceptical Theism and Evidential Arguments from Evil,” Australasian Journal of Philosophy , 81(4): 496–516.
Alston, William, 1980. “Level Confusions in Epistemology,” Midwest Studies in Philosophy , 5(1): 135–50.
–––, 1991 [1996]. “The Inductive Argument from Evil and the Human Cognitive Condition,” reprinted in Howard-Snyder (ed.) 1996, 97–125.
–––, 1996. “Some (Temporaily) Final Thoughts on Evidential Arguments from Evil,” in Howard-Snyder (ed.) 1996, 311–332.
Anderson, David James, 2012. “Skeptical Theism and Value Judgments,” International Journal for Philosophy of Religion , 72: 27–39.
Beaudoin, John, 2000. “Inscrutable Evil and Scepticism,” Heythrop Journal , 41(3): 297–302.
–––, 2005. “Skepticism and the Skeptical Theist,” Faith and Philosophy , 22(1): 42–56.
Bergmann, Michael, 2001. “Skeptical Theism and Rowe’s New Evidential Argument from Evil,” Noûs , 35(2): 278–296.
–––, 2008. “Externalist Responses to Skepticism,” in Oxford Handbook of Skepticism , John Greco (ed.), Oxford: Oxford University Press, 504–532.
–––, 2009. “Skeptical Theism and the Problem of Evil,” in Oxford Handbook of Philosophical Theology , Thomas P. Flint and Michael Rea (eds.), 374–399.
–––, 2012. “Commonsense Skeptical Theism,” in Reason, Metaphysics, and Mind , Kelly James Clark and Michael Rea (eds.), Oxford: Oxford University Press, 9–30.
Bergmann, Michael and Michael Rea, 2005. “In Defense of Sceptical Theism: A Reply to Almeida and Oppy,” Australasian Journal of Philosophy , 83(2): 241–251.
Boyce, Kenneth, 2014. “Some Considerations Concerning CORNEA, Global Skepticism, and Trust,” in Dougherty and McBrayer (eds.) 2014, 103–114.
Christlieb, Terry, 1992. “Which Theism Faces an Evidential Problem of Evil?” Faith and Philosophy , 9(1): 45–64.
Coffman, E. J., 2014. “Replies to Long and Tucker,” in Dougherty and McBrayer (eds.) 2014, 76–84.
Coley, Scott, 2015. “Skeptical Theism is Incompatible with Theodicy,” International Journal for Philosophy of Religion , 77: 53–63.
DePoe, John, 2014. “On the Epistemological Framework for Skeptical Theism,” in Dougherty and McBrayer (eds.) 2014, 32–44.
Dougherty, Trent, 2008. “Epistemological Considerations Concerning Skeptical Theism,” Faith and Philosophy , 25(2): 172–176.
–––, 2012. “Reconsidering the Parent Analogy: Unfinished Business for Skeptical Theists,” International Journal for Philosophy of Religion , 72: 17–25.
Dougherty, Trent and Justin McBrayer (eds.), 2014. Skeptical Theism: New Essays , Oxford: Oxford University Press.
Dougherty, Trent and Paul Draper, 2013. “Explanation and the Problem of Evil,” in McBrayer and Howard-Snyder (eds.) 2013.
Draper, Paul, 1989. “Pain and Pleasure: An Evidential Problem for Theists,” Noûs , 23: 331–350.
–––, 1992. “Probabilistic Arguments from Evil,” Religious Studies , 28(3): 303–317.
–––, 1996. “The Skeptical Theist,” in Howard-Snyder (ed.) 1996, 175–192.
–––, 2013. “The Limitations of Pure Skeptical Theism,” Res Philosophica , 90(1): 97–111.
–––, 2014a. “Confirmation Theory and the Core of CORNEA,” in Dougherty and McBrayer (eds.) 2014, 132–141.
–––, 2014b. “Meet the New Skeptical Theism, Same as the Old Skeptical Theism,” in Dougherty and McBrayer (eds.) 2014, 164–177.
Durston, Kirk, 2000. “The Consequential Complexity of History and Gratuitous Evil,” Religious Studies , 36(1): 65–80.
Ekstrom, Laura, 2021. God, Suffering and the Value of Free Will , Oxford: Oxford University Press.
Fumerton, Richard, 1995. Metaepistemology and Skepticism , Lanham, MD: Rowman & Littlefield.
Gellman, Jerome, 2013. “The Experience of Evil and Support for Atheism,” in McBrayer and Howard-Snyder (eds.) 2013, 98–112.
–––, 2017. “A Surviving Version of the Common Sense Problem of Evil: A Reply to Tweedt,” Faith and Philosophy , 34(1): 82–92.
Graham, Andrew and Stephen Maitzen, 2007. “CORNEA and Closure,” Faith and Philosophy , 24(1): 83–6.
Hasker, William, 2010. “All Too Skeptical Theism,” International Journal for Philosophy of Religion , 68(1–3): 15–29.
Hendricks, Perry, 2020. “Skeptical Theism Unscathed,” Pacific Philosophical Quarterly , 101(1): 43–73.
–––, 2018. “How to be a Skeptical Theist and a Commonsense Epistemologist,” Faith & Philosophy , 35(3): 345–55.
Howard-Snyder, Daniel, 1992. “Seeing Through CORNEA,” International Journal for Philosophy of Religion , 32(1): 25–49.
––– (ed.), 1996. The Evidential Argument from Evil , Bloomington: Indiana University Press.
–––, 1996b. “The Argument from Inscrutable Evil,” in Howard-Snyder (ed.) 1996, 286-310.
–––, 2009. “Epistemic Humility, Arguments from Evil, and Moral Skepticism,” in Jonathan Kvanvig (ed.), Oxford Studies in Philosophy of Religion , 2: 17–57.
–––, 2014. “Agnosticism, the Moral Skepticism Objection, and Commonsense Morality,” in Dougherty and McBrayer (eds.) 2014, 293–306
Hudson, Hud, 2014. “The Father of Lies?” in Jonathan Kvanvig (ed.), Oxford Studies in Philosophy of Religion , 5: 147–166.
Huemer, Michael, 2001. Skepticism and the Veil of Perception , Lanham, MD: Rowman & Littlefield.
Jeffrey, Anne, 2019. God and Morality , Cambridge: Cambridge University Press.
Jordan, Jeff, 2006. “Does Skeptical Theism Lead to Moral Skepticism?” Philosophy and Phenomenological Research , 72(2): 403–417.
Law, Stephen, 2017. “Skeptical Theism and Skepticism about the External World and Past,” Royal Institute of Philosophy Supplement , 81: 55–71.
Leite, Adam, 2008. “Believing One’s Reasons are Good,” Synthese , 161(3): 419–441.
Maitzen, Stephen, 2007. “Skeptical Theism and God’s Commands,” Sophia , 46: 237–243.
–––, 2013. “The Moral Skepticism Objection to Skeptical Theism,” in McBrayer and Howard-Snyder (eds.) 2013, 444–457.
Martin, Michael, 1978. “Is Evil Evidence Against the Existence of God?” Mind , 87(347): 429–432.
Matheson, Jonathan D., 2011. “Epistemological Considerations Concerning Skeptical Theism: A Response to Dougherty,” Faith and Philosophy , 28(3): 323–31.
–––, 2014. “Phenomenal Conservatism and Skeptical Theism,” in Dougherty and McBrayer (eds.) 2014, 3–20.
McBrayer, Justin, 2009. “CORNEA and Inductive Evidence,” Faith and Philosophy , 26(1): 77–86.
McBrayer, Justin, and Daniel Howard-Snyder (eds.), 2013. The Blackwell Companion to The Problem of Evil , Malden, MA: Wiley-Blackwell.
McBrayer, Justin and Philip Sweson, 2012. “Scepticism about the argument from Divine Hiddenness,” Religious Studies , 48: 129–50.
Mooney, Justin, 2019. “How to Solve the Problem of Evil,” Faith and Philosophy , 36(4): 442–462.
Morriston, Wes, 2014. “Skeptical Demonism: A Failed Response to a Humean Challenge,” in Dougherty and McBrayer (eds.), 2014, 221–235.
Murphy, Mark, 2017. God’s Own Ethics , Oxford: Oxford University Press.
Otte, Richard, 2002. “Rowe’s Probabilistic Argument from Evil,” Faith and Philosophy , 19(2): 147–171.
Perrine, Timothy and Stephen Wykstra, 2014. “Skeptical Theism, Abductive Atheology, and Theory Versioning,” in Dougherty and McBrayer (eds.) 2014, 142–163.
–––, 2017. “Skeptical Theism,” in The Cambridge Companion to the Problem of Evil , Chad Meister and Paul Moser (eds.), Cambridge: Cambridge University Press, 85–107.
Perrine, Timothy, 2019. “Skeptical Theism and Morriston’s Humean Argument from Evil,” Sophia , 58: 115–135.
–––, 2022. “On an Epistemic Cornerstone of Skeptical Theism: In Defense of CORNEA,” Sophia . 61: 533-555.
Piper, Mar, 2007. “Skeptical Theism and the Problem of Moral Aporia ,” International Journal for Philosophy of Religion , 62: 65–79.
Plantinga, Alvin, 1974. God, Freedom, and Evil , Grand Rapids, MI: Eerdmans Publishing.
–––, 1996. “On Being Evidentially Challenged,” in Howard-Snyder (ed.) 1996, 244–261.
–––, 1998. “Degenerate Evidence and Rowe’s New Evidential Argument from Evil,” Noûs , 32(4): 531–544.
–––, 2000. Warranted Christian Belief , Oxford: Oxford University Press.
Plantinga, Alvin and Nicholas Wolterstorff (eds.), 1983. Faith and Reason , Notre Dame: University of Notre Dame Press.
Rancourt, Benjamin, 2013. “Egoism or the problem of evil,” Religious Studies , 49: 313–325.
Rowe, William, 1979. “The Problem of Evil and Some Varieties of Atheism,” American Philosophical Quarterly , 16(4): 335–41.
–––, 1984. “Evil and the Theistic Hypothesis: A Response to Wykstra,” International Journal for Philosophy of Religion , 16(2): 95–100.
–––, 1986. “The Empirical Argument from Evil,” in Rationality, Religious Belief, and Moral Commitment , Robert Audi and William J. Wainwright (eds.), Ithaca: Cornell University Press, 227–247.
–––, 1988. “Evil and Theodicy,” Philosophical Topics , 16(2): 119–132.
–––, 1991. “Ruminations about Evil,” Philosophical Perspectives , 5: 69–88.
–––, 1996. “The Evidential Argument from Evil: A Second Look,” in Howard-Snyder (ed.) 1996, 262–285.
–––, 2006. “Friendly Atheism, Skeptical Theism, and the Problem of Evil,” International Journal for Philosophy of Religion , 59(2): 79–92.
Rudavsky, T. M, 2013. “A Brief History of Skeptical Responses to Evil,” in McBrayer and Howard-Snyder (eds.), 2013, 379–395.
Russell, Bruce, 1989. “The Persistent Problem of Evil,” Faith and Philosophy , 6(2): 121–39.
–––, 2017. “The Problem of Evil: Excessive Unnecessary Suffering,” in Ethics and the Problem of Evil , James Sterba (ed.), Bloomington: Indiana University Press, 90–107.
–––, 2018. “The Problem of Evil and Some Replies to Some Important Responses,” European Journal for Philosophy of Religion , 10(3): 105–131.
Russell, Bruce and Stephen Wykstra, 1988. “The ”Inductive“ Argument from Evil: A Dialogue,” Philosophical Topics , 16(2): 133–160.
Rutledge, Jonathan Curtis, 2017a. “Commonsense, Skeptical Theism, and Different Sorts of Closure of Inquiry Defeat,” Faith and Philosophy , 34(1): 17–32.
–––, 2017b. “The Parent Analogy: A Reassessment,” International Journal for Philosophy of Religion , 82: 5–14.
Schellenberg, J. L., 1993. Divine Hiddenness and Human Reason , Ithaca: Cornell University Press.
–––, 2007. The Wisdom to Doubt , Ithaca: Cornell University Press.
–––, 2015. The Hiddenness Argument , Oxford: Oxford University Press.
Schnall, Ira M., 2007. “Sceptical Theism and Moral Scepticism,” Religious Studies , 43(1): 49–69.
Sehon, Scott, 2010. “The Problem of Evil: Skeptical Theism Leads to Moral Paralysis,” International Journal for Philosophy of Religion , 67: 67–80.
Senor, Thomas D., 2013. “Skeptical Theism, CORNEA, and Common Sense Epistemology,” in McBrayer and Howard-Snyder (eds.) 2013, 426–443.
Shields, Jannai, 2021. “Representing the Parent Analogy,” European Journal for Philosophy of Religion , 13(4): 1–10.
Stone, Jim, 2011. “CORNEA, Scepticism, and Evil,” Australasian Journal of Philosophy , 89(1): 59–70.
Street, Sharon, 2014. “If Everything Happens For A Reason, Then We Don’t Know What Reasons Are,” in M. Bergmann & P. Kain (eds.), Challenges to moral and religious belief: Disagreement and evolution , Oxford: Oxford University Press.
Stump, Eleonore, 2010. Wandering in Darkness , Oxford: Oxford University Press.
Swinburne, Richard, 1998. Providence and the Problem of Evil , Oxford: Oxford University Press.
–––, 2004. The Existence of God , revised edition, Oxford: Oxford University Press.
Tucker, Chris, 2014. “Why Skeptical Theism Isn’t Skeptical Enough,” in Dougherty and McBrayer (eds.), 2014, 45–62.
Van Inwagen, Peter, 1991 [1996]. “The Problem of Evil, the Problem of Ari, and the Problem of Silence,” reprinted in Howard-Snyder (ed.), 1996, 151–174.
–––, 1996. “Reflections on the Chapters by Draper, Russell, and Gale,” in Howard-Snyder (ed.) 1996, 219–243.
–––, 2006. The Problem of Evil , Oxford: Oxford University Press.
Weisberger, Andrea, 2007. “The Argument from Evil,” in Cambridge Companion to Atheism , Michael Martin (ed.), Cambridge: Cambridge University Press, 166–181.
Wielenberg, Eric, 2010. “Sceptical Theism and Divine Lies,” Religious Studies , 46: 509–23.
Wilks, Ian, 2013. “The Global Skepticism Objection to Skeptical Theism,” in McBrayer and Howard-Snyder (eds.), 2013, 458–467.
Wykstra, Stephen, 1984. “The Humean Obstacle to Evidential Arguments from Suffering,” International Journal for Philosophy of Religion , 16(2): 73–93.
–––, 1996. “Rowe’s Noseeum Arguments from Evil,” in Howard-Snyder (ed.), 1996, 126–150.
–––, 2007. “CORNEA, Carnap, and Current Closure Befuddlement,” Faith and Philosophy , 24(1): 87–89.
Wykstra, Stephen and Timothy Perrine, 2012. “Foundations of Skeptical Theism,” Faith and Philosophy , 29(4): 375–399.
Aikin, Scott and Brian Ribeiro, 2019. “Skeptical Theism and the Creep Problem,” Logos & Episteme , 10(4): 349–362.
Bass, Robert, 2011. “Many Inscrutable Evils,” Ars Disputandi , 11: 118–132.
–––, 2014. “Inscrutable Evils: Still Numerous, Still Relevant,” International Journal of Philosophy and Theology , 75(4): 379–384.
Buchak, Lara, 2014. “Learning Not To be Naïve: A Comment on the Exchange Between Perrine/Wykstra and Draper,” in Dougherty and McBrayer (eds.) 2014, 178–190.
Draper, Paul, 2014a. “Confirmation Theory and the Core of CORNEA,” in Dougherty and McBrayer (eds.) 2014, 132–141.
–––, 2016. “When Skeptical Theism Fails, Skeptical Atheism Prevails,” In Jonathan Kvanvig, eds , Oxford Studies in Philosophy of Religion , 7: 63–80
Durston, Kirk, 2006. “The Complexity of History and Evil: A Reply to Trakakis,” Religious Studies , 42(1): 87–99.
Gilbertson, Eric, forthcoming. “Skeptical Theism, the Preface Paradox, and Non-Cumulative Inductive Evidence of Pointless Evil,” Philosophia .
Hendricks, Perry, 2021. “We Are Not In the Dark,” American Philosophical Quarterly , 58(2): 125–34.
Howard-Snyder, Daniel and Michael Bergmann, 2020. “Evil Does Not Make Atheism More Reasonable than Theism,” in Contemporary Debates in Philosophy of Religion , Michael Peterson and Raymond Van Arragon (eds.), Malden, MA: Wiley-Blackwell, 148–159.
Hudson, Hud, 2020. A Grotesque in the Garden , 2nd edition, Grand Rapids, MI: Eerdmans Press.
Jordan, Jeff, 2017. “The ”Loving Parent“ Analogy,” International Journal for Philosophy of Religion , 82: 15–28.
O’Connor, David, 2013. “Theistic Objections to Skeptical Theism,” in McBrayer and Howard-Snyder (eds.) 2013, 468–481.
Oliveira, Luis, 2021. “Sceptical Theism and the Paradox of Evil,” Australasian Journal of Philosophy , 98(2): 319–333.
Pegan, Philip, 2019. “Sharon Street’s Unsuccessful Argument Against Theism ,” International Journal for Philosophy of Religion , 86: 17–24.
Rea, Michael, 2013. “Skeptical Theism and the ”Too Much Skepticism“ Objection,” in McBrayer and Howard-Snyder (eds.) 2013, 482–506.
Russell, Bruce, 1996. “Defenseless,” in Howard-Snyder (ed.), 1996, 193–205.
Segal, Aaron, 2011. “Sceptical Theism and Divine Truths,” Religious Studies , 47: 85–95.
Snapper, Jef, 2011. “Paying the Cost of Skeptical Theism ,” International Journal for Philosophy of Religion , 69: 45–56.
Stone, Jim, 2003. “Evidential Atheism,” Philosophical Studies , 114: 253–277.
Timpe, Kevin, 2014. “Trust, Silence, and Liturgical Acts,” in Dougherty and McBrayer (eds.) 2014, 264–275.
Trakakis, Nick, 2003. “Evil and the Complexity of History: A Response to Durston,” Religious Studies , 39: 451–458.
Wielenberg, Erik, 2014. “Divine Deception,” in Dougherty and McBrayer (eds.) 2014, 236–249.
Wykstra, Stephen J, 2017. “Beyond the Impasse: Contemporary Moral Theory and the Crisis of Skeptical Theism,” in Ethics and the Problem of Evil , James Sterba (ed.), Bloomington: Indiana University Press, 108–140.

How to cite this entry . Preview the PDF version of this entry at the Friends of the SEP Society . Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entry at PhilPapers , with links to its database.

Skeptical Theism , entry in the Internet Encyclopedia of Philosophy .
Dougherty, Trent, “Skeptical Theism”, Stanford Encyclopedia of Philosophy (Winter 2022 Edition), Edward N. Zalta & Uri Nodelman (eds.), URL = < https://plato.stanford.edu/archives/win2022/entries/skeptical-theism/ >. [This was the previous entry on this topic in the Stanford Encyclopedia of Philosophy – see the version history .]

atheism and agnosticism | evil: problem of | goodness, perfect | hiddenness of God | providence, divine

Acknowledgments

For helpful feedback, the author thanks Michael Bergmann, Spencer Case, Paul Draper, Perry Hendricks, Michael Longenecker, and an anonymous reviewer.

Accessibility

Support SEP

Mirror sites.

View this site from another server:

Info about mirror sites

Library of Congress Catalog Data: ISSN 1095-5054

Help | Advanced Search

Computer Science > Machine Learning

Title: the platonic representation hypothesis.

Abstract: We argue that representations in AI models, particularly deep networks, are converging. First, we survey many examples of convergence in the literature: over time and across multiple domains, the ways by which different neural networks represent data are becoming more aligned. Next, we demonstrate convergence across data modalities: as vision models and language models get larger, they measure distance between datapoints in a more and more alike way. We hypothesize that this convergence is driving toward a shared statistical model of reality, akin to Plato's concept of an ideal reality. We term such a representation the platonic representation and discuss several possible selective pressures toward it. Finally, we discuss the implications of these trends, their limitations, and counterexamples to our analysis.

Submission history

Access paper:.

HTML (experimental)
Other Formats

References & Citations

Google Scholar
Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

COMMENTS

Skepticism
A skeptical hypothesis (with respect to a proposition p and a subject S) is a proposition SH such that if SH were true, then: (a) ... For example, suppose that I am justified, ceteris paribus, in believing that (pure) water is present if I am justified in believing that there is present, at standard temperature and pressure, a clear, odorless ...
Contemporary Skepticism
A standard example of a skeptical hypothesis is the so-called 'brain-in-a-vat' (BIV) hypothesis that one is being 'fed' one's experiences by computers. If this were true, then most of what one believes about the world would be false (or, at the very least, true in a different way from how one would expect), and thus one would lack ...
Philosophical skepticism
Philosophical skepticism (UK spelling: scepticism; from Greek σκέψις skepsis, "inquiry") is a family of philosophical views that question the possibility of knowledge. It differs from other forms of skepticism in that it even rejects very plausible knowledge claims that belong to basic common sense.Philosophical skeptics are often classified into two general categories: Those who deny ...
7.4 Skepticism
Offer and explain a skeptical hypothesis. Outline the general structure of argument for global skepticism. Philosophical skepticism is the view that some or all knowledge is impossible. A skeptic questions the possibility of knowledge—particularly justification—in some domain. ... We simply can never know either way whether, for example ...
Skepticism
Upon completion of this chapter, readers will be able to: Differentiate global skepticism and external-world skepticism by the scope of their doubts.; Explain the role of skeptical hypotheses in arguments for external-world skepticism.; Construct original skeptical hypotheses modeled on those discussed.; Analyze the Moorean and contextualist responses to skepticism, recognizing how Moorean ...
Skepticism
The first school of skeptical philosophy developed in the Academy, the school founded by Plato, in the 3rd century bce and was thus called "Academic" skepticism. Starting from the skeptical doctrines of Socrates, its leaders, Arcesilaus and Carneades, set forth a series of epistemological arguments to show that nothing could be known, challenging primarily what were then the two foremost ...
7.4: Skepticism
Offer and explain a skeptical hypothesis. Outline the general structure of argument for global skepticism. Philosophical skepticism is the view that some or all knowledge is impossible. A skeptic questions the possibility of knowledge—particularly justification—in some domain. ... We simply can never know either way whether, for example ...
Skepticism
Skepticism. First published Sat Dec 8, 2001; substantive revision Wed Aug 31, 2005. Much of epistemology has arisen either in defense of or in opposition to various forms of skepticism. Indeed, one could classify various theories of knowledge by their responses to skepticism. For example, rationalists could be viewed as skeptical about the ...
Moral Skepticism
There are two main responses to such skeptical hypothesis arguments (cf. May 2013). First, some anti-skeptics deny (1) and claim that skeptical hypotheses can be ruled out somehow. They might argue that moral nihilism is internally inconsistent or meaningless. If so, it can be ruled out by logic and semantics alone.
Sceptical Hypotheses and Transcendental Arguments
A different way to overcome a sceptical hypothesis is to use a transcendental argument. Transcendental arguments try to show that there is no way that the sceptical hypothesis could be true. That is, they try to show that a sceptical hypothesis can only be one of two things: it is either false, or it doesn't make any sense in the first place.
10 Live Skeptical Hypotheses
2. The Live Skeptic's Argument. The live skeptic uses an argument just like the dinosaur one: if you believe P, hypothesis H inconsistent with P is a live hypothesis, and you are like Jo in being a well‐informed "mere mortal" with regard to H, then you do not know P. Let us make the argument moderately precise.
Epistemic Closure Principles
The skeptic points out that if one knows an ordinary common sense proposition (such as that one has hands) to be true, and knows that this proposition entails the falsity of a skeptical hypothesis (such as that one is a handless brain in a vat, all of whose experiences are hallucinatory), one could know the falsity of the skeptical hypothesis ...
Skepticism: Explanation and Examples
Example #2. The problem of fake news and the reliability of internet sources. Basically, the solution to these problems is skepticism. The problem is that more and more people take whatever they read online without skepticism. Or are skeptical about only the things they already disagree with.
Skeptical hypotheses and moral skepticism
example, draws an analogy between perceptual skepticism and the kind of value skepticism that goes with a sense of life's absurdity. In the perceptual case, Nagel ... such skeptical hypothesis arguments by denying either the second or third premise. In the above argument, the crucial premise is the first one, namely, that I
PDF VARIETIES OF SKEPTICISM
1. Skepticism as an underdetermination problem. Skepticism about the external world is a philosophical problem, but there are importantly. different conceptions of what that problem is. One way of understanding such skepticism, which. I find fruitful, is to construe it as an underdetermination problem.
Skepticism and Content Externalism
Skepticism and Content Externalism. First published Wed May 23, 2018. A number of skeptical hypotheses or scenarios have been proposed which can be used as the basis for arguments to the effect that we lack knowledge of various propositions about objects in the external world, propositions that we normally take for granted and that we assume ...
Skeptical hypotheses and moral skepticism
Moral skeptics maintain that we do not have moral knowledge. Traditionally they haven't argued via skeptical hypotheses like those provided by perceptual skeptics about the external world, such as Descartes' deceiving demon. But some believe this can be done by appealing to hypotheses like moral nihilism. Moreover, some claim that skeptical ...
logic
We were told to formulate our own skeptical hypothesis and use it as part of a radical skeptical argument, and that the skeptical hypothesis needs to be . ... and that the skeptical hypothesis needs to be different from the usual examples (dreaming, virtual reality, BIV, etc). I've been thinking about it for a while but I couldn't really ...
PDF skeptical hypothesis
Neo is given very good evidence that some skeptical hypothesis is true. He rightly be-comes doubtful that his senses are or have been trustworthy. In fact, he becomes conﬁdent of a particular hypothesis: that AI's created the Matrix, etc. But this is just one of many types of hypotheses that might account for his experiences. These types ...
Examples of Skepticism in Different Fields
Exploring skepticism examples can make you a bit skeptical. See what it can look like in different situations, and be confident in what you've learned. ... John was skeptical when the television ad said the cleaner would take out all stains. I was still skeptical even after the word "natural" was printed on the food label.
How to Write a Strong Hypothesis
5. Phrase your hypothesis in three ways. To identify the variables, you can write a simple prediction in if…then form. The first part of the sentence states the independent variable and the second part states the dependent variable. If a first-year student starts attending more lectures, then their exam scores will improve.
Skeptical Theism
Skeptical theism is a cluster of responses to arguments from evil. One major argument from evil is what we'll call the "Outweighing Goods" argument. This section explains a basic outweighing goods argument, a standard way of defending it using "noseeum inferences," and a skeptical theistic critique of that defense.
Hypothesis Testing Explained (How I Wish It Was Explained to Me)
In this article, I won't delve into how sample size is computed (I will probably do it in a follow-up). For now, let's simply use the Statmodel's function for testing the difference between sample means as a black box: ### input (hypothesis + confusion matrix) control_mean = 10 control_std = 8 treatment_mean = 10.5 treatment_std = 9 confidence = .975 power = .80 ### compute sample size ...
[2405.07987] The Platonic Representation Hypothesis
The Platonic Representation Hypothesis. Minyoung Huh, Brian Cheung, Tongzhou Wang, Phillip Isola. We argue that representations in AI models, particularly deep networks, are converging. First, we survey many examples of convergence in the literature: over time and across multiple domains, the ways by which different neural networks represent ...

7.4 Skepticism

Global Skepticism

The Dream Argument

The Evil Demon Argument

Putnam’s Brain in a Vat

General Structure of Global Skeptical Arguments

Responses to Global Skepticism

Contextualism

Skepticism in Specific Domains

Think Like a Philosopher

4 Skepticism

Skepticism and Its Scope

External-World Skepticism

Zhuangzi’s butterfly dream.

The evil demon.

Brain in a vat.

Moore against Skepticism

The Contextualist Concession

Further Reading

Share This Book

Margin Size

7.4: Skepticism

Learning Objectives

Global Skepticism

The Dream Argument

The Evil Demon Argument

Putnam’s Brain in a Vat

General Structure of Global Skeptical Arguments

Responses to Global Skepticism

Contextualism

Skepticism in Specific Domains

Think Like A Philosopher

Sceptical Hypotheses and Transcendental Arguments

Putnam’s Brain In A Vat

Kripke’s Language Checking

10 Live Skeptical Hypotheses

1. Skeptical Threats, Real and Merely Academic

2. The Live Skeptic's Argument

3. The Liveness‐Mortality Premise

4. The Nature of Live Skepticism

5. Do We Really Need to Rule Out the Error Theories?

6. Is It Not Actually Easy to Rule Live Error Theories Out?

7. Watering Down the Conclusion

This Feature Is Available To Subscribers Only

I. Definition

II. The History of Skepticism

III. Controversies

IV. Famous Quotes about Skepticism

Quotation #2

V. Types of Skepticism

VI. Skepticism versus Pessimism

VII. Skepticism in Popular Culture

Bibliography

Skeptical Theism

1.1 Basic Outweighing Goods Argument

2. Outweighing Goods and Inductive Inferences

3. Noseeum Arguments

5. Epistemological Objections

6. Moral Objections

7. Scope of Skeptical Theism

Acknowledgments

Support SEP

Computer Science > Machine Learning

Submission history

References & Citations

BibTeX formatted citation

Bibliographic and Citation Tools

arXivLabs: experimental projects with community collaborators

COMMENTS