Feyerabend discussed what I have called his « heuristic meta-attitude » toward the diverse rational methods, rules, criteria, and structures in a number of places, under the name of the « way of the scientist » to distinguish it from the « way of the philosopher ». The way of the scientist is based on provisional, local, temporary, approximative, and tentative rules. The criteria and prescriptions have no universal application but are intended as possible procedures and considerations, rules of thumb, that concrete research has to fill out and decide on:
« I neither want to replace rules, nor do I want to show their worthlessness; I rather want to increase the inventory of rules, and I want to suggest a different use for all of them…Usually it is assumed that rules determine the structure of research in advance, they guarantee its objectivity, they guarantee that we are dealing with rational action. By contrast I regard each piece of research both as a potential instance of application for a rule and as a a test case of the rule: we may permit the rule to guide our research, i.e. to exclude some actions and to mould others, but we may also permit our research to suspend the rule, or to regard it as inapplicable even though all the known conditions demand its application… We are guided, rather, by the vague hope that working without the rule, or on the basis of a contrary rule we shall eventually find a new form of rationality that will provide a rational justification for the whole procedure ».
The « way of the scientist » is based on a discussion of of actual scientists: Mach, Einstein, Planck, Bohr, Pauli, von Weizsäcker. This is the sort of research that he favours. We see that Feyerabend is not « anti-science », nor does he think that a scientific approach has no distinctive characteristics. Yet he refuses to encapsulate them in a fixed, universal, and prescriptive model.
Feyerabend carefully distinguishes forms of rationalism based on a-historical principles or on abstract structures (the way of the philosopher) from his own historical approach (the way of the scientist). So it would be a mistake to affirm without suitable qualification that Feyerabend is anti-rationalist. As we see in the quote above, Feyerabend is in favour of rationality and of rational justification.
It is also a mistake to lump him in with « postmodernist philosophers » without discussing their actual ideas, and not the travesties proposed by the science warriors.