A selection of the best from recent issues of the Philosopher

Feyerabend states his purpose at the outset, which is to convince the reader that all methodologies are limited and that we should support the pluralism of theories and metaphysical views. From the major example of Galileo ( chapter 5 onwards ), Feyerabend expresses a radical extension of Popper that takes in Lakatos and ends in a very broad church, requiring very, very broad conjectures, although not a direct attack on induction like the one Popper made. He also, perhaps unanarchistically, supports the side of the church in the trials of Galileo, considering, as he does, that they acted correctly given the evidence. Feyerabend's point on page 152 that Popperian doctrine has observation at its start rather flies in the face of Popper's criticism of the use of probability as a basis for theory support, where he states that the hypothesis is the start. Feyerabend also leaves the dissenter without any rules ( page 162 ), whereas he or she is more likely to be operating within a different set.

In the rewrite of chapter 16, Feyerabend's discussion of incommensurability leads to a lengthy analysis of the Homeric world-view compared to the later one and he supports what he sees as the eclecticism of archaic man and debates whether incommensurabilities exist within quantum physics and finally drops the previous end of the chapter in favour of the difficulty in defining incommensurability.

For Appendix 2, Feyerabend relinquishes some of his support of Whorf's ideas that were in the previous edition and now accepts that concentrating on language is a limitation and puts forward the view that incommensurability is a problem for philosophers and not for scientists.

The summarizing questions of the latter chapter are the same but the answers have developed, but basically conclude that the general public should be assisted to realise that science is fallible. A new 'Postscript on Relativism', attacks the strong programme of the sociology of science whilst to some extent agreeing with recent Kuhn. 

Since it was first published in 1975, Against Method has followed Popper's The Logic of Scientific Discovery and Kuhn's The Structure of Scientific Revolutions into becoming a classic text in the debate about scientific methodology and scientific reasoning. In the way that Popper's contribution has been reduced by a number of his obituary writers, I suspect that Feyerabend's contribution may diminish in stature as well, but while continuing in the style of the others with some sweeping generalisations based on the science of antiquity, Feyerabend did manage to open up the field with his support of pluralisation and of the public being made aware of the limits of science. But it does seem that one can only gain sufficient publicity by pushing the bounds of sense past the extreme with a literary style that is, at times, opaque. 

Reviewed by Mick Phythian