The Forced Option
When the evidence is insufficient and the decision cannot be deferred, is it rational to let your will decide what to believe?
William James presented this argument in his 1896 lecture 'The Will to Believe,' responding directly to the Victorian skeptic W. K. Clifford. James argued that for a specific class of decisions, waiting for sufficient evidence is not a neutral act but a choice with its own consequences.
James, W. (1896). The Will to Believe and Other Essays in Popular Philosophy. Harvard University Press.
Live, forced, and momentous
James classified choices by three criteria. A choice is live if both options are genuine possibilities for the person making it. A dead option is one you cannot take seriously. A choice is forced if there is no third option: refusing to choose counts as one of the choices, not an escape from them. A choice is momentous if the stakes are high, the opportunity does not recur easily, and the decision significantly shapes what follows.
Religious belief, James argued, meets all three conditions for many people. You cannot simply wait indefinitely for more evidence, because life makes demands in the meantime. Not believing is not a neutral holding pattern. It is a choice with the same consequences as a deliberate decision not to believe.
Most choices, James conceded, are not like this. In science, you can suspend judgment, run more experiments, and wait. The costs of waiting are low and the benefits of accuracy are high. Clifford's principle, that evidence should govern belief, is appropriate there.
The will to believe
For forced, live, and momentous options, James argued that the will to believe is legitimate. When you cannot avoid committing, when both paths carry real costs, and when evidence alone does not settle the matter, your passional nature, your hopes, fears, and values, has a right to decide.
James was careful about what this licenses. It does not license believing whatever you want whenever you want. The option must be genuinely forced. If you can wait, you should. The principle is narrow but has a specific target: religious belief, which James thought often presents exactly this structure.
He also argued that some goods are only available through prior commitment. Trust in a relationship, confidence in a difficult task, and certain kinds of religious experience may require the belief before they can produce the evidence that would confirm it. A purely evidence-first approach to these goods might preclude them entirely.
Clifford's objection
W. K. Clifford stated the opposing view in its starkest form: "It is wrong always, everywhere, and for any one, to believe anything upon insufficient evidence."
Clifford was not merely making an epistemic point. He thought believing beyond evidence is a moral failure. Beliefs have consequences. Ships are sent to sea on the basis of beliefs about their seaworthiness. Public policy is shaped by beliefs about human nature. A person who forms beliefs carelessly contributes to a culture of credulity that causes real harm.
James thought Clifford's standard was itself an expression of a passionate commitment to a particular value, namely, avoiding error. But avoiding error is not the only epistemic value. Gaining truth matters too. A policy of never believing without sufficient evidence will successfully avoid many false beliefs and will also miss many true ones. James thought Clifford's rule made the avoidance of error lexically prior to the pursuit of truth, and that this prioritization was not obviously correct.
The debate between them is still the clearest framing of the question: what does rationality permit when evidence runs out?
Discussion questions
- Can you think of a case where refusing to believe something was itself a consequential choice?
- Is religious belief something you genuinely cannot stay neutral on, even if you wanted to?
- Is it possible to be genuinely neutral on a question that demands an answer?
Take it to the dinner table.
Get 3 thought experiments for memorable conversations, designed for dinner, with friends, at events, or anywhere small talk has gone on too long.