The Australian philosopher David Stove (1927-1994) published the first edition of this book in 1982 with the title Popper and After: Four Modern Irrationalists. The book was reprinted in 1998 with the title Anything Goes: Origins of the Cult of Scientific Irrationalism. The 2007 edition has a different title: Scientific Irrationalism: Origins of a Postmodern Cult. The book is a criticism of Popper, Kuhn, Lakatos, and Feyerabend's philosophy of science. The four authors are charged of irrationalism; they defend, according to Stove, the "extremely implausible position" (21) that science is not cumulative and that there has been no growth of knowledge.

The book is organized in two sections. In the first section Stove claims that the four authors make their position credible by means of literary tricks. An example is what Stove calls "sabotaging logical expressions" (51): frequently the four authors embed a logical expression (i.e. an assertion about logical relations between statements) in an epistemic context, thus depriving the expression of its logical implications. The reader is not conscious of this trick, and wrongly thinks that the sabotaged expression still keeps its logical consequences.

A schematic example … is this: instead of saying "P entails Q", which is of course a logical statement, to say "P entails Q according to most logicians" (55)

While "P entails Q" is an assertion of logic, which implies definite relations between P and Q (e.g. it is not the case that P is true and Q is false, if an interpretation M satisfies "not Q" then M satisfies "not P") the statement "P entails Q according to most logicians" is not an assertion of logic – it says nothing at all about logical relations between P and Q – but it is an assertion about history of science or history of logic. The reader is unaware of this fact, and supposes that a statement of logic is made, and that a definite logical property concerning P and Q is asserted. "P entails Q according to most logicians" suggests that P entails Q, but this is not the case. The result is a "ghost-logical statement" (56) which is immune to logical criticism.

In the second section Stove searches for the historical origins of the irrationalism espoused by the four authors. Kuhn, Lakatos, and Feyerabend derive their irrationalism from Popper, who is profoundly influenced by Hume's argument against induction. At the core of Hume's argument, says Stove, is deductivism, that is the thesis that P is a justification for Q only if P entails Q. Deductivism does not accept non-deductive justifications, such as the justifications provided by inductive logic. For example, deductivism claims that the statement "I hold 999 of the 1000 tickets of a fair lottery" (P) is not a reason to believe "I will win the lottery" (Q), because P and Q are logically independent statements. Deductivism is not only false, according to Stove, but also frivolous: a deductivist has no means of analysing the logical relations between rival scientific theories because rival scientific theories are logically independent (179). Therefore, a deductivist sometimes manifests a tendency to erroneously ascribe logical relations to propositions which are independent; of course, such philosopher must exhibit an unusual carelessness about logic. The charge against Popper and the others is very serious: according to Stove, they have voluntarily made mistakes in elementary deductive logic.

I am sympathetic with Stove's reaction against Feyerabend's irrationalism, but this is the only substantial point in which I agree with him. The questions posed by the four authors are genuine problems, not mere literary tricks. In 1981 Laudan reformulated the doubts about the accumulation of scientific knowledge in a straightforward way: every past scientific theory was proved false, thus we have reasons to believe that our present scientific theories are false too – the so called pessimistic meta-induction. Is thus science really cumulative? If our past and present scientific theories are false, in what sense we can say that there has been a growth of knowledge? A philosopher who points out this kind of difficulties is not an irrationalist and an enemy of science; he is a bona fide philosopher.

I have found Stove's arguments unconvincingly and flawed, but others have found his work interesting and worth of mention. One of the most important journals dedicated to philosophy of science has published a favourable review, in which we can read that the "points about the way in which Popper, Lakatos, Kuhn and Feyerabend corrupt the language are convincingly and entertainingly developed in Part I" (Philosophy of Science, 50, p. 351). Surely the book is entertaining, but I don't search for entertainment in a work about philosophy of science; I search for solid philosophical arguments, but they are lacking in Stove's book.

© 2007 Mauro Murzi

Mauro Murzi, Rome, Italy

Comment on this review

Categories: Philosophical