What Is Scepticism? Taking Doubt Seriously as a Method

Greek scepticism developed into distinct traditions with different conclusions. The Academic sceptics, working in Plato's school, argued that knowledge in the strict sense is unattainable and that the reasonable course is to act on what is most persuasive without claiming certainty. The Pyrrhonists went further and refused even that negative claim, since asserting that nothing can be known is itself a knowledge claim. Their practice was to set opposing arguments of equal apparent strength against each other, producing suspension of judgement, and their stated goal was not truth but tranquillity, on the reasoning that distress comes from holding firm beliefs about how things really are, and that letting them go produces peace. That is a striking position, since it treats philosophy as therapy rather than as inquiry. Sextus Empiricus compiled the arguments in works that survived and were rediscovered in the sixteenth century, which had an enormous effect on early modern philosophy and contributed directly to the questions Descartes set out to answer.

Several arguments recur and each targets a different route to justifying a belief:

• •The regress problem, since any justification can be asked for its own justification, leading to an infinite chain, a circle, or an arbitrary stopping point, none of which looks satisfactory
• •The problem of the criterion, which asks how one could identify a reliable method of distinguishing knowledge from error without already possessing knowledge
• •The unreliability of the senses, evidenced by illusions, dreams and disagreement between observers, which undermines the most direct source of belief
• •The problem of induction, which observes that no number of past regularities logically guarantees future ones, and that the assumption the future will resemble the past cannot be justified without circularity
• •The dream argument and its modern descendants, which observe that no internal feature reliably distinguishes waking from dreaming or reality from a sufficiently good simulation
• •Disagreement among apparently equally competent and informed people, which is taken to show that neither has a decisive reason to hold their view rather than the other's

The word has drifted a long way in ordinary use and several current senses should be kept apart. Scientific scepticism, meaning the practice of demanding evidence proportionate to a claim, is methodological and is compatible with confident belief once the evidence arrives, and it is close to what most people mean by the word now. Selective doubt, meaning the application of extreme standards to claims one dislikes and none at all to claims one prefers, is a rhetorical tactic rather than a position, and it has been used deliberately by industries seeking to delay regulation, which is documented in the histories of tobacco and of climate policy. Cynicism, meaning a general expectation that people act from bad motives, has no relation to the epistemological position despite being frequently conflated with it. Philosophical scepticism is distinguished from all of these by applying its doubts universally and consistently, including to itself, which is exactly what makes it hard to live by and hard to refute.

Philosophers have answered in several incompatible ways and no answer commands agreement. Foundationalism holds that the regress stops at beliefs that are self-justifying or evident without further support, and disputes concern what could qualify. Coherentism accepts the circle and argues that justification comes from the mutual support of a whole system of beliefs rather than from a foundation. Reliabilism relocates the question, holding that a belief is justified if produced by a reliable process, whether or not the believer can demonstrate that reliability. Pragmatism treats the demand for certainty as the error, arguing that beliefs are tools assessed by how well they work. Contextualism proposes that the standards for knowing vary with context, so a sceptical argument succeeds by raising standards artificially and ordinary knowledge claims remain true in ordinary contexts. Common sense responses insist that the certainty of ordinary knowledge exceeds the certainty of any sceptical premise, and that this is itself a reason to reject the argument rather than its conclusion.

The position is a systematic test of whether knowledge claims can be justified, and the Pyrrhonists refused even to assert that nothing is knowable, since that is itself a claim. The regress problem, the criterion problem and induction each block a different route to justification. Scientific scepticism, selective doubt used to delay regulation, and cynicism are three different things frequently confused with it and with each other.