Responses to Skepticism — I can evaluate responses to skepticism (contextualism, externalism) at an introductory level.

Historical Context & Motivation

The problem of skepticism — the philosophical challenge that we may not truly know anything at all — has been a persistent thorn in the side of epistemologists for over two millennia. Ancient Greek thinkers such as Pyrrho of Elis argued that for every claim we make about the world, an equally compelling counter-claim can be constructed, leading to a radical suspension of judgment. This tradition of radical doubt did not disappear with antiquity; it was revived and sharpened in the early modern period by René Descartes, who famously entertained the possibility that an evil demon might be deceiving him about everything, including the existence of the external world. Descartes ultimately used skepticism as a method to arrive at certainty, but the skeptical scenarios he described took on a philosophical life of their own.

In the twentieth century, analytic epistemologists reformulated skeptical arguments with greater logical precision. The central skeptical argument can be reconstructed as follows: if you cannot rule out the possibility that you are a brain in a vat — a disembodied brain being fed artificial experiences by a supercomputer — then you do not know that you have hands, that the sky is blue, or indeed anything about the external world. This argument exploits a principle called epistemic closure: the idea that if you know some proposition P, and you know that P entails Q, then you also know Q. The skeptic turns this principle against us, arguing that since we cannot know we are not brains in vats, we cannot know ordinary propositions either.

The question that motivates the rest of this lesson is: how should we respond to these powerful skeptical arguments? Do we accept that knowledge is impossible, or can we find principled reasons to resist the skeptic? Two of the most influential strategies developed in the late twentieth century are contextualism and externalism. Each offers a distinctive diagnosis of where the skeptical argument goes wrong, and each carries its own philosophical costs and benefits.

Core Principles & Definitions

Before evaluating specific responses to skepticism, it is essential to understand several foundational concepts that structure the debate. The traditional analysis of knowledge, often called the justified true belief (JTB) account, holds that a person S knows a proposition P if and only if P is true, S believes P, and S is justified in believing P. Skeptical arguments typically target the justification condition, arguing that our evidence for ordinary beliefs is insufficient once we take radical hypotheses seriously. Responses to skepticism differ in how they reconceive justification, the meaning of 'knows,' or the structural requirements for knowledge.

Visual Explanation — The Skeptical Argument and Its Responses

The diagram above captures the core logical structure of the debate. Notice that the skeptical argument is deductively valid — its conclusion follows from its premises. This means that any successful response must reject at least one premise or reinterpret a key term. Contextualists take the semantic route: they argue that the word 'knows' in the skeptic's conclusion means something different from the word 'knows' in ordinary conversation, so the skeptic's argument, while sound in a philosophy seminar, does not threaten everyday knowledge claims. Externalists take the epistemological route: they argue that the skeptic's demand — that you be able to introspectively rule out all alternatives — is too strong, because knowledge can be grounded in reliable processes or truth-tracking relations that the knower need not be aware of.

How Contextualism and Externalism Work

Contextualism: Shifting Standards

The core mechanism of epistemic contextualism rests on a thesis about language rather than about the nature of justification itself. Contextualists like Keith DeRose, Stewart Cohen, and David Lewis argue that 'knows' functions like indexical expressions such as 'tall,' 'flat,' or 'here.' When you say 'the table is flat,' the truth of your utterance depends on the standards operative in your conversational context. In a carpentry shop, the table might count as flat; in a physics laboratory measuring surface irregularities at the nanoscale, it would not. Similarly, when you say 'I know I have hands,' the truth of that claim depends on the epistemic standards operative in your context. In everyday conversation, where the relevant alternatives are mundane — perhaps your hands are hidden in gloves, or the lighting is poor — you easily meet the standards for knowledge. But in a philosophy seminar, where the brain-in-a-vat hypothesis is introduced as a live possibility, the standards rise, and the same evidence may no longer suffice.

This approach provides a diagnosis of why the skeptical argument feels compelling. When the skeptic mentions the brain-in-a-vat possibility, she raises the epistemic standards in the conversational context. Once those standards are raised, it becomes true — in that context — that you do not 'know' you have hands. But crucially, this does not retroactively undermine your earlier, everyday knowledge claims, which were made in a lower-standards context. The skeptic has not shown that ordinary knowledge is impossible; she has merely created a context in which the word 'knows' demands more than ordinary evidence can provide.

Externalism: Reliability and Tracking

While contextualism operates at the level of semantics, externalism operates at the level of epistemology itself. The externalist's central claim is that justification — and therefore knowledge — can depend on factors outside the believer's conscious awareness. The most prominent version, reliabilism (developed by Alvin Goldman), holds that a belief is justified if and only if it is produced by a cognitive process that reliably yields true beliefs. Perception, memory, and sound reasoning are reliable; wishful thinking, hasty generalization, and tea-leaf reading are not. On this view, you know you have hands because your perceptual system is, as a matter of empirical fact, a reliable producer of true beliefs about your immediate environment — even if you cannot prove this from the armchair.

A related externalist strategy, the tracking theory proposed by Robert Nozick, adds a counterfactual dimension. Nozick argues that S knows P if and only if: (1) P is true, (2) S believes P, (3) if P were false, S would not believe P (the sensitivity condition), and (4) if P were true, S would believe P (the adherence condition). Your belief that you have hands satisfies sensitivity: in the nearest possible worlds where you lack hands (say, after an accident), you would notice and would not believe you have hands. However, the belief 'I am not a brain in a vat' fails sensitivity: in the nearest possible world where you are a brain in a vat, your experiences would be identical and you would still believe you are not one. The crucial upshot is that Nozick denies the closure principle: you can know you have hands without knowing you are not a brain in a vat, because these are assessed by different counterfactual conditions.

Detailed Breakdown — Varieties of Response

It is worth noting an important near-cousin of contextualism: subject-sensitive invariantism (SSI), advanced by John Hawthorne and Jason Stanley. Like contextualists, proponents of SSI hold that the standards for knowledge vary. Unlike contextualists, however, they deny that this variation is a feature of the word 'knows.' Instead, SSI holds that whether a subject actually knows depends on features of the subject's practical situation — such as what is at stake. If nothing important hinges on whether you are right about the bank's hours, you may know they close at five on the basis of casual memory; but if you desperately need to deposit a check before the deadline, the same evidence may not suffice for knowledge because the stakes are higher. This view preserves an invariant semantics for 'knows' while still allowing knowledge attributions to be context-dependent at the level of the subject's situation.

Among the externalist options, it is also useful to distinguish between process reliabilism and tracking theories. Reliabilism is a theory of justification: it tells us when a belief is justified. Tracking theories are theories of knowledge proper: they offer conditions under which a true belief constitutes knowledge. Both are externalist because they appeal to factors (reliability of processes, counterfactual sensitivity) that may lie beyond the subject's awareness, but they differ in their precise commitments and their relationship to the closure principle. Reliabilism is generally compatible with closure; tracking theories are not.

Worked Example — Evaluating a Skeptical Scenario

Consider the following scenario, which is a classic in the epistemological literature and is often called the zebra case (inspired by Fred Dretske). You are at the zoo, standing in front of the zebra enclosure, and you say, 'I know those are zebras.' A skeptic approaches and asks, 'But do you know they are not cleverly disguised mules painted to look like zebras?' Let us work through this case from three perspectives.

Strengths and Limitations

Connection to Advanced Epistemological Theory

The responses to skepticism discussed in this lesson are not isolated curiosities; they connect to some of the most active research programs in contemporary epistemology. Contextualism has catalyzed a broader investigation into the intersection of philosophy of language and epistemology. If 'knows' is context-sensitive, are other epistemic terms — 'justified,' 'rational,' 'certain' — context-sensitive as well? This question has led to the field of epistemic expressivism and to experimental philosophy projects that empirically test how ordinary speakers use knowledge attributions. Meanwhile, externalism has deeply influenced debates about epistemic virtue theory, which asks whether knowledge is best understood as the product of intellectual virtues (like careful observation) rather than in terms of static conditions on beliefs.

One especially important development is Timothy Williamson's safety condition, which can be seen as a refinement of Nozick's tracking approach. Williamson argues that a belief is safe — and hence constitutes knowledge — if the subject could not easily have been wrong. Unlike sensitivity, safety does not require that you would stop believing P if P were false; it requires only that, in nearby possible worlds where you form the belief using the same method, your belief is true. This approach preserves closure (unlike Nozick) while retaining the externalist insight that knowledge depends on modal relations between belief and truth. For social science students, these debates matter because they inform broader questions about methodology: when does a survey respondent 'know' their own attitudes, and when are we justified in treating self-reports as reliable evidence?

Practice Problems

Lesson Summary