What Is Fallibilism? Holding Beliefs You Might Be Wrong About

Fallibilism holds that no belief is immune to revision, that the justification supporting a claim never guarantees its truth, and that the appropriate response is to remain open to evidence rather than to treat any conclusion as final. It does not hold that nothing is known, which is scepticism and a different position entirely, since a fallibilist can consistently say that a claim is known while acknowledging it might be revised. It does not hold that all beliefs are equally uncertain, since evidence comes in degrees and a well-supported claim is not on a par with a guess. And it does not counsel indecision, since acting on the best available account is exactly what a fallibilist recommends. The position is about the status of a belief rather than about how firmly it should be held, which is the distinction most commonly missed by both its critics and its careless supporters.

Several considerations push towards it and they are difficult to answer:

• •The historical record, since well-established theories have been overturned repeatedly and there is no reason to think the present is different
• •The problem of induction, since no finite set of observations entails a general conclusion
• •Underdetermination, since more than one account is usually compatible with the available evidence
• •The fallibility of the faculties involved, since perception, memory and reasoning all demonstrably err
• •The absence of a reliable marker distinguishing certainty from confident error, since both feel the same from inside
• •The regress facing any claim to a foundation, since whatever grounds a belief can itself be questioned

What the position amounts to in practice is a set of habits rather than a general attitude of doubt. It means stating what evidence would change a belief, which is the single most useful discipline available and which separates a view held for reasons from one held for other motives. It means proportioning confidence to the support available rather than to the strength of feeling. It means taking criticism as information rather than as attack, which is the mechanism that makes collective enquiry work better than individual reasoning. It means treating a revision as a success of the method rather than an embarrassment. And it means distinguishing between claims at different levels of support rather than treating everything one believes as equally established. None of this requires announcing uncertainty constantly, which is a common misreading and is frequently a way of avoiding commitment rather than a form of intellectual honesty.

The position has a long history and several distinct sources. Ancient scepticism raised the underlying difficulties about justification and was mostly answered by attempts to find a secure foundation, which is the project that dominated early modern philosophy and that produced the search for beliefs immune to doubt. The nineteenth-century American pragmatists made the modern version explicit, treating enquiry as a self-correcting process in which no result is final, and coined the term. Popper developed a related view centred on the idea that theories can be refuted and never confirmed, and argued that science advances by the elimination of error rather than by accumulating proof. Later work in the history and sociology of science supplied the empirical support, documenting how thoroughly established frameworks have been replaced. The positions differ in important ways and share the conclusion that finality is not available and that arranging enquiry to catch its own errors is the practical response.

Two serious challenges are worth stating. The first concerns self-application, since the claim that every belief might be mistaken applies to itself, and critics argue this is either self-undermining or requires an exception that damages the position. Fallibilists generally accept the self-application and regard it as harmless, since holding a view that might be wrong is exactly what the view recommends. The second concerns mathematics and logic, where claims seem to be established in a way that empirical claims are not, and where a proof appears to deliver certainty rather than support. Responses vary, with some fallibilists restricting the thesis to empirical claims, others pointing out that proofs have contained errors that stood for years and that mathematical practice does involve correction, and others distinguishing the certainty of a conclusion given its premises from certainty about the premises themselves. The disagreement here is genuine and unsettled.

Holding that any belief might be revised is not scepticism, since a fallibilist can say a claim is known while acknowledging it might change. The historical record, the problem of induction and the absence of any internal marker separating certainty from confident error all push towards it. In practice it means saying what would change your mind and proportioning confidence to evidence, not announcing doubt constantly.