Abstract
In their defence of Freud against the Popperian onslaught, Grant and Harari [1] argued that Popper's criterion of falsifiability did not disprove the claimed scientific status of psychoanalysis. However, they appear to have fallen into a commonmistake, in confusing Popper's notion of a demarcation criterion (between science and non-science) with his closely-related but separate theory on the progress of science.
Popper wanted a reliable means of distinguishing or demarcating between empirical science and non-science. In his view, the difference lay in the fact that a scientific theory can always be falsified by further experience whereas non-science, such as political ideas, religious beliefs and metaphysics, are never wrong. If a theory cannot generate predictions which can be proven wrong, it has no worthwhile empirical content and is not scientific in nature.
His theory of the progress of science derives from his view that there is a discrepancy in the values which attach to positive and negative test results. Essentially, when a test fails, it gives more information about the theory's status than if it succeeds. This is because a positive result tells us only that the theory was correct on that occasion, but, due to the problem of induction, the result can never state reliably that the theory is therefore a general truth. However, a negative result states emphatically that the theory is not general.
Science therefore progresses by people postulating bold theories, designing demanding tests for them and then doing their best to make their theory fail. A theory, which passes all the tests is held to be provisionally true. Of course, this means that there is no certainty in science but that is a minor issue.
These points must not be conflated because to do so leads to egregious errors, such as concluding that psychoanalysis was a valid scientific theory. There are other, equally serious errors in this paper, for example, believing that Freud refuted his own theories (he did not: he simply changed his mind), or believing that Popper ‘neglected the crucial role played by concepts and models in scientific theorising’. This, as I have shown [2], is simply not true of him, just as it is not true that he claimed there is ‘… but one scientific method, equally applicable to the natural sciences… social sciences… and all other endeavours which claim to be scientific’. His method of science was not the same as his demarcation criterion.
Psychoanalysis was simply bad science. It cannot be saved by bad philosophy.