Edmund Gettier published two short counterexamples arguing that justified true belief (JTB) is not enough for knowledge. The cases made it look like someone can have good reasons and a true belief, yet still be right only by luck. This shifted late-20th-century epistemology toward explaining (and avoiding) “epistemic luck.”
Alvin Goldman argued that knowledge requires an appropriate causal connection between the fact and the believer’s belief. The idea was that Gettier-style luck can occur when the belief is not produced in the right way by what makes it true. Causal approaches became one major family of “post-Gettier” theories.
Fred Dretske argued that knowing a claim can require ruling out relevant alternatives—ways the claim could be false that matter in the situation. This approach aimed to explain why we can know ordinary things without eliminating every far-fetched skeptical possibility. It also set up later debates about whether knowledge is “closed” under implication (roughly: if you know p, and p implies q, can you know q?).
Peter Unger argued for an unusually strong skeptical view: that we do not (or cannot) know anything. By pressing skeptical arguments in a systematic way, the book helped make skepticism an urgent problem for late-20th-century analytic epistemology. Many later “anti-skeptical” theories were shaped by the need to answer this kind of challenge.
Gail Stine developed a relevant-alternatives treatment of skepticism while focusing on deductive closure (the idea that knowledge is preserved through known logical implication). She argued that whether a skeptical alternative is relevant can shift with context, and that tests of closure must avoid slipping between contexts. Her paper is an important bridge from relevant-alternatives views to later contextualist responses.
Saul Kripke introduced the “Pierre” case to show a paradox that can arise in reporting what someone believes using proper names. The puzzle suggested that ordinary belief reports can pull in conflicting directions even when the thinker seems rational. This mattered for epistemology because skeptical paradoxes and contextualist replies often depend on careful attention to how knowledge and belief attributions work in language.
Robert Nozick argued that knowledge involves beliefs that “track” the truth across nearby possibilities: if p were false, you wouldn’t believe p (and if p were true, you would). This aimed to block skeptical arguments by denying a strong closure principle that skeptics often rely on. Nozick’s approach became a major alternative to justification-centered theories in post-Gettier epistemology.
Barry Stroud argued that philosophical skepticism is significant even if we reject its conclusion, because it reveals deep tensions in how we think about knowledge of the external world. Rather than treating skepticism as an easy mistake, Stroud treated it as a serious philosophical problem needing diagnosis. This helped set the stage for late-20th-century work that tried to explain why skeptical arguments feel compelling.
Stewart Cohen defended the idea that we can be fallibilists (we can know without having perfect certainty) while still taking skeptical worries seriously. A key move was that what counts as knowing can vary with conversational context: when high error-possibilities become relevant, the standard for “knows” can rise. This became a central strategy for later contextualist replies to skeptical paradoxes.
Michael Williams argued that many skeptical arguments depend on hidden theoretical assumptions about how justification must work (for example, that all knowledge needs a certain kind of foundation). By challenging those assumptions, Williams aimed to show why radical skepticism is not forced on us by ordinary standards of inquiry. His work pushed skepticism debates toward examining the background “rules” of epistemic practice.
Keith DeRose defended attributor contextualism: the truth conditions of “S knows that p” can depend on the speaker’s context. His famous “bank cases” illustrate how knowledge attributions can shift when what is at stake changes, even if the person’s evidence stays the same. This paper helped make contextualism a leading late-20th-century response to skeptical pressure.
DeRose presented a detailed contextualist “solution” to skepticism: in ordinary contexts, skeptical alternatives (like being a brain in a vat) are not relevant, so many everyday knowledge claims are true. In philosophical contexts where skeptical scenarios are made salient, standards rise and the skeptic can correctly deny that we know. The account aims to explain both why skepticism feels powerful and why it does not overturn ordinary knowledge claims.
David Lewis argued that which error possibilities we must rule out depends on conversational rules (for example, a “rule of attention”: once an error possibility is raised, it becomes harder to ignore). This yields the idea that knowledge can be “elusive”: mentioning skeptical possibilities can make them relevant and undercut many knowledge claims in that context. Lewis’s paper became a major reference point for contextualist responses to skeptical paradoxes in the late 1990s.
Cohen argued that contextualism can explain how our reasons support ordinary knowledge claims while still allowing skeptical arguments to seem forceful in special contexts. The paper emphasized how the structure of reasons and relevance shifts with conversational setting, rather than treating skepticism as simply a logical mistake. By the end of the 1990s, contextualism had become a leading framework for explaining skeptical paradoxes without giving up everyday knowledge.
Late 20th-Century Epistemological Challenges: Gettier, Skeptical Paradoxes, and Contextualist Responses (1963–2000)