Is science the only way to know?

Many public voices hold that science is the only reliable route to knowledge: if a claim cannot be tested by experiment, it is either false or meaningless. Philosophers call this view scientism. It is different from science itself, which is a method for studying the natural world and is always open to revision. Scientism is a philosophical claim about all knowledge, and it has often been used to dismiss belief in God as superstition.

The paper examines two ideas that gave scientism its force. The first is verificationism, the view of A. J. Ayer, Rudolf Carnap and the logical positivists that a sentence means something only if it can be checked by experience or is true by definition. On this test, “God exists” was declared meaningless. The authors point out the problem: the verification principle can neither be checked by experience nor is it true by definition, so by its own rule it is meaningless. It would also rule out moral statements, claims about the past, and even general scientific laws, which can never be checked case by case. By the 1960s the theory was widely regarded as untenable.

The second is Karl Popper's falsificationism: a theory is scientific only if some observation could prove it wrong. Popper meant this to separate science from other fields, not to decide what is meaningful or rational, and he accepted that untestable metaphysical claims can have meaning. The paper adds the Duhem–Quine point that a failed test can be blamed on background assumptions, and Thomas Kuhn's observation that scientists rarely drop a theory after one failed prediction. It then sets out alternatives: Alvin Plantinga's case that belief in God can be rational without being inferred from other beliefs, William Alston's view that experience of God can justify belief much as sense perception does, Wittgenstein's idea that words gain meaning from their use within a way of life, and the critical realism of Ian Barbour and Nancey Murphy.

The authors conclude that knowledge is plural. Science remains the best tool for studying nature, though it rests on assumptions it cannot test, such as the uniformity of nature. History, law, ethics and religion each use methods suited to their questions, and questions of meaning, value and ultimate reality are not settled by measurement alone.