Radical Skepticism — Ignorance (1975)

A Case for Global Scepticism

Peter Unger by Licentia Poetica

The Central Argument: No One Ever Knows Anything

Peter Unger’s Ignorance: A Case for Scepticism, published by Oxford University Press in 1975, represents one of the most sustained and uncompromising defenses of global skepticism in twentieth century philosophy. The book’s central thesis is disarmingly simple to state even though its implications are staggering, namely that no human being ever knows anything to be the case about the external world, and further that this conclusion extends beyond knowledge to justification itself, so that no one is ever reasonable or rational in holding any belief whatsoever. Unger arrived at this position not through the traditional route of invoking evil demons or brain in a vat scenarios, though such devices appear in his writing as illustrative aids, but through a careful semantic analysis of the very term knows and its relationship to certainty. His argument proceeds from the premise that knowledge entails certainty, that certainty is what linguists and philosophers call an absolute term, and that absolute certainty is a condition that human beings can never actually satisfy given the nature of belief formation and the ever present possibility of error. The conclusion Unger draws is what he termed universal ignorance, a thesis considerably more radical than the local or restricted skepticism many philosophers had previously entertained about particular domains such as perception or memory.

The Certainty Requirement and the Problem of Error Possibilities

At the heart of Unger’s argument lies a distinction he had already developed in earlier papers, most notably his 1971 Philosophical Review article A Defense of Skepticism, between what he called absolute terms and relative terms. Relative terms admit of degrees and comparison without any implication of a maximal or perfect instance, so that something can be tall or short without any requirement that a perfectly tall thing exist. Absolute terms, by contrast, denote a condition that is either fully satisfied or not satisfied at all, and Unger illustrated this with the adjective flat, arguing that a surface is properly called flat only if it has no bumps or irregularities whatsoever, however minute, and that anything short of this ideal is strictly speaking not flat but merely close to flat for practical purposes. Unger then classified certain as an absolute term of this same kind, arguing that to be certain that some proposition is true is to have absolutely no doubt whatsoever concerning it, a condition that admits no degrees despite our loose everyday willingness to say that we are more or less certain of things. Because knowledge, on Unger’s account, requires this same undiluted certainty, and because human cognitive life is pervasively marked by the possibility of error given the fallibility of perception, memory, testimony and reasoning, it follows that the stringent condition demanded by knowledge is one that virtually no belief can ever meet. Unger reinforced this argument with what commentators have called the evil scientist thought experiment, a precursor in some respects to later brain in a vat scenarios, designed to show that for any ordinary belief there remains some conceivable scenario, however remote, in which that belief would be false, and that the mere existence of such a possibility undermines the absolute certainty that knowledge is said to require.

Critique of Common-Sense and Contextualist Responses

Unger was acutely aware that his position appeared to fly in the face of common sense, and a substantial portion of Ignorance is devoted to anticipating and rebutting the various strategies by which philosophers had traditionally sought to reconcile epistemological theory with everyday certainty about knowledge claims. He was particularly critical of what might be called common-sense philosophy in the tradition of G E Moore, which sought to refute skeptical arguments simply by appealing to the overwhelming plausibility of ordinary claims such as knowing that one has hands, since Unger regarded this maneuver as question-begging rather than argumentatively substantive. He also engaged critically with early forms of what would later develop into contextualist theories of knowledge attribution, approaches that seek to explain away the apparent tension between skeptical arguments and ordinary knowledge claims by holding that the truth conditions of sentences containing the word knows shift according to conversational context, so that in low stakes everyday contexts we correctly say that people know things, while in the more demanding context of philosophical skeptical argument the standards for knowledge rise and such claims become false. Unger argued that this kind of contextual variability, whatever its merits in accounting for certain linguistic data, could not adequately explain away the force of the skeptical argument, since the argument targets the underlying absolute character of certainty itself rather than merely the pragmatic conditions under which we are willing to assert knowledge claims in ordinary speech. His skepticism about these accommodating strategies places Ignorance in direct opposition to a great deal of subsequent epistemological work that sought precisely to preserve ordinary knowledge attributions in the face of skeptical pressure.

Relation to Descartes, Hume, and Contemporary Skepticism

Unger situated his own argument in a lineage that stretches back to the classical skeptical tradition while also marking a clear departure from it. Descartes’s methodological doubt in the Meditations sought to identify a foundation of certainty secure against even the most radical doubt, using the possibility of a deceiving demon as a heuristic device to strip away all beliefs that could conceivably be false, before ultimately reconstructing a body of certain knowledge upon the cogito and the existence of a benevolent God. Unger’s evil scientist scenario functions similarly as a heuristic for exposing the fragility of our beliefs, but unlike Descartes, Unger offers no subsequent reconstruction of certainty, since his conclusion is that the demand for certainty itself can never be met and that the skeptical possibility can never be philosophically defeated. Hume’s skepticism, articulated most forcefully in his discussions of induction and causation, similarly emphasized the gap between the psychological compulsion to believe and any rational justification for that belief, and Unger’s insistence that we are never reasonable in our beliefs echoes Hume’s naturalistic diagnosis that belief outruns what reason can strictly license. Where Unger differs from both predecessors is in grounding his skepticism not primarily in epistemic scenarios of deception but in a semantic thesis about the logical grammar of absolute terms, making his argument in certain respects more akin to a linguistic or conceptual demonstration than a purely epistemological one. Contemporary skeptical arguments, including the widely discussed brain in a vat scenario popularized by Hilary Putnam and later closed under different premises by such figures as external world skeptics working in the Cartesian mold, share with Unger a reliance on underdetermination between the world as it appears and alternative scenarios equally compatible with our evidence, though most subsequent formulations abandon Unger’s distinctive absolute terms apparatus in favor of more straightforward appeals to evidential indistinguishability.

Unger’s Early Defense of Invariantism about Knows

A crucial feature of Unger’s position, one that has attracted sustained attention in later epistemology, is his implicit commitment to invariantism about the semantics of knows, meaning the view that the truth conditions for knowledge attributing sentences do not shift with conversational context but remain fixed regardless of the practical stakes or conversational standards in play. This sets Unger apart from contextualist theories of knowledge attribution that would later be developed in more systematic form by philosophers such as Keith DeRose and Stewart Cohen, who argued that the word knows functions somewhat like an indexical or context sensitive expression whose satisfaction conditions vary with features of the conversational context in which it is used. Unger’s invariantism, combined with his insistence that the fixed standard demanded by knows is the maximally stringent standard of absolute certainty, is precisely what generates his radical skeptical conclusion, since on his view there is no lower or more relaxed standard available in ordinary contexts that would allow everyday knowledge claims to come out true. This combination of invariantism with an extremely high fixed standard for knowledge is what many commentators have identified as the most philosophically distinctive and contestable feature of his overall position, since one could in principle accept his invariantism about the word knows while rejecting his characterization of certain as an absolute term, or conversely accept his semantic analysis of absolute terms while adopting a contextualist rather than invariantist account of knows.

Influence on Later Debates about Epistemic Standards and Contextualism

Ignorance exerted a significant influence on the subsequent development of epistemology, particularly on debates concerning the proper analysis of knowledge attributing sentences and the appropriate response to skeptical arguments. The book helped catalyze a wave of technical work throughout the 1980s and 1990s on contextualism, much of which can be read as an attempt to accommodate the intuitive pull of Unger’s skeptical argument without accepting its radical conclusion, by allowing that the skeptic’s demanding standards are appropriate in the special context of philosophical reflection while insisting that ordinary contexts operate with genuinely lower and satisfiable standards. David Lewis’s influential paper Elusive Knowledge, published in 1996, explicitly engaged with the puzzle that Unger’s work helped bring into sharp focus, namely how to explain the shifting plausibility of skeptical arguments across different conversational settings. Unger himself continued to refine his views on these questions in later work, including his 1984 book Philosophical Relativity, which extended his skepticism about determinate answers to many philosophical questions beyond epistemology narrowly construed. The distinction between relative and absolute gradable adjectives that Unger introduced has also had a lasting afterlife within linguistics and philosophy of language, where his classification of terms like flat and certain as absolute predicates anticipated later formal semantic treatments of gradable adjectives.

Legacy of the Book in Epistemology

More than four decades after its publication, Ignorance remains a touchstone in discussions of skepticism, cited regularly in handbook treatments of the topic and in the Stanford Encyclopedia of Philosophy’s entries on skepticism, contextualism, and the analysis of knowledge. Its legacy lies less in having persuaded the majority of epistemologists to accept universal ignorance as a live thesis, since most subsequent theorists have sought ways to resist or circumvent Unger’s conclusion, and more in having sharpened the terms in which the debate over knowledge, certainty, and error possibilities is conducted. Unger forced later philosophers to grapple seriously with the semantic question of what knows actually means and how its satisfaction conditions relate to certainty, a question that continues to generate technical literature in epistemology and semantics alike. The book stands as a landmark demonstration of how far a philosopher willing to follow an argument without regard for its palatability can push a discipline that had, in the decades prior, largely treated skepticism as a formal puzzle to be answered rather than a serious candidate for truth.

Further Readings

• Unger, Peter. Ignorance: A Case for Scepticism. Oxford University Press, 1975
• Unger, Peter. “A Defense of Skepticism.” The Philosophical Review, vol 80, no 2, 1971
• Stroud, Barry. The Significance of Philosophical Scepticism. Oxford University Press, 1984
• Pritchard, Duncan. Epistemic Luck. Oxford University Press, 2005
• DeRose, Keith. “Contextualism and Knowledge Attributions.” Philosophy and Phenomenological Research, vol 52, no 4, 1992
• Lewis, David. “Elusive Knowledge.” Australasian Journal of Philosophy, vol 74, no 4, 1996
• Cohen, Stewart. “Knowledge, Context, and Social Standards.” Synthese, vol 73, no 1, 1987
• Klein, Peter D. Certainty: A Refutation of Scepticism. University of Minnesota Press, 1981
• Olders, Diederik. “Unger’s Argument for Skepticism Revisited.” Philosophical journal article discussing the premises of Unger’s certainty argument
• Stanford Encyclopedia of Philosophy, entries on Skepticism, Contextualism in Epistemology, and The Analysis of Knowledge