Falsification

Conceptual Overview and Discussion

A (scientific) theory is usually developed inductively. Observation is made, data collected, and a generalized theory that explains the data is created. Since the explanation covers only observations already made or data already gathered, it is impossible to substantiate the claim that the theory is universally true. A theory may be shown to be wrong in one of two ways. First, it is possible that some relevant observation or data will be found that is not explained by the theory or that contradicts it. Second, theories ultimately depend on assumptions. Predictions are logical entailments of the basic assumptions. If the predictions are found to be false, then the assumptions must also be false. It is possible that any theory may be false, and so all theories are falsifiable.

One peculiar aspect of falsification is that if a theory withstands attempts to falsify it, it is deemed to be somehow strengthened. Claims concerning its universality are considered less likely to be wrong than they were, and untested predictions made by the theory are accepted more readily. We have more confidence that the theory is “true.” It is not easy to explain why this is so. If we lack evidence to disprove a theory at a particular time, and we still lack evidence to disprove the theory at some later time, the situation has not changed: We lack evidence to disprove the theory. One approach to this situation is to claim that as additional data are gathered, knowledge is increased, and thus every unsuccessful challenge increases confidence in the theory. But again, it is hard to say precisely why a supposedly universal theory is stronger because it explains more data.

Popper argued that one goal of science should be to actively attempt to find evidence that would falsify a theory or its predictions. In his view, science should make, and attempt to refute, “risky” predictions by constructing “critical experiments” that would, if successful, conclusively demonstrate the falsity of the theory. If the theory robustly resists such attempts to falsify it, our confidence in the theory is justifiably increased.

...