Stove concisely explains both the aim of this book and its structure in the first part of a very short preface.
Stove starts chapter one by clarifying the sort of view that would uncontroversially constitute an irrationalist position regarding science. Stove then advances his reading of the philosophers he is criticising: "Popper, Kuhn, Lakatos, and Feyerabend, are all writers whose position inclines them to deny (A), or at least makes them more or less reluctant to admit it. (That the history of science is not "cumulative", is a point they all agree on)." The question he addresses in the chapter is, "How do these writers manage to be plausible, while being reluctant to admit so well-known a truth as (A)?" A general answer to this question is offered: "the constant tendency in these authors to conflate questions of fact with questions of logical value, or the history with the philosophy of science." Stove claims this tendency is "widely recognized", but waives both this general answer (and its supporters) in favour of seeking a more specific account.
Stove's first step in refining the general answer is observing what he calls mixed strategy writing in the authors he is examining. He uses this expression, since it is not always clear to him whether the writing expresses "equivocation" or "inconsistency". What is common to the examples Stove offers is that something well-known is mixed with something extraordinary, without the clash being resolved; the "irrationalism" is introduced simultaneously with orthodoxy, rendering it more plausible to the reader—disbelief is suspended.
A straight-forward example is provided by Thomas Kuhn's description of "paradigm shift", where he asserts the well-known fact that the world is the same after "paradigm shift" as before. Yet, at the same time, Kuhn also suggests that solutions to problems achieved under old paradigms are lost, redundant or "un-solutions" under new paradigms—denial of (A) above.
Examining Kuhn's use of the word solution more closely, Stove notes that Kuhn sometimes uses it in the ordinary way regarding practical knowledge, but at other times in a weaker sense, specific to Kuhn's theory, that a solution is relative to a paradigm, people, place and time. This equivocation on solution actually provides Stove with an answer of exactly the type he was looking for. All his authors, with many similar words, show similar equivocation. Stove lists knowledge, discovery, facts, verified, understanding, explanation and notes the list is far from complete. Idiosyncratic weak senses of these words are a characteristic of the writing of his subjects that explains clearly how a reader, presuming ordinary use of language, might believe them to be expressing something more orthodox than is, in fact, their intention.
At this point, Stove coins the expression neutralizing success words and provides an uncontroversial example from everyday language to illustrate it.
Stove also provides a quote from Paul Feyerabend explicitly directing his readers to "neutralize" his success words or not, according to their own preferences.
Stove notes that logical expressions can be sabotaged, just as success-words can be neutralized. He spends some time clarifying the relationship between these phenomena, since they are similar in intention but not, in fact, identical. Rather, they work together in the following way.
He also articulates the distinction in an informal (and wittily expressed) way, that sabotaging logical expressions is like derailing cognitive achievement en route, so that it can never arrive anywhere; while neutralizing success-words is more like blowing up any cognitive achievement at the destination, so it can never be recognized as having arrived.
Stove now presents a common method of sabotaging logical expressions in a generalizable form.
This simple pattern of expression makes historical rather than logical assertions (like an encyclopedia documenting debate, without making any truth-claims about what is said, only that it was said, see de dicto and de re).
Stove seems to restrain his witicisms in the course of presenting the evidence above. However, as he presents the last quote, he appears to experience his astonishment at such a statement as though again for the first time, expressing this via his characteristically barbed wit. Not only could Popper bring himself to make the last assertion, he is sufficiently comfortable with it to supply of course. Not only does Popper consider belief to be irrational, he considers this to be common knowledge!
This explains where many of Popper's ideas have come from—he shares Hume's scepticism about induction.
Stove considers this establishes what he set out to show in the chapter since, "Popper's philosophy of science is at any rate not more irrationalist than that of Feyerabend, Kuhn, or Lakatos, and at the same time, as a matter of well-known history, Popper's philosophy owes nothing to theirs, while Kuhn's philosophy owes much, and the philosophy of Lakatos and Feyerabend owes nearly everything, to Popper."
However, he explains that establishing both that these writers are irrationalists, and where their irrationalism comes from historically, still leaves the question of what it is they believe that leads them to accept this irrationalist conclusion. What implicit premise grounds their confidence in such an otherwise unattractive conclusion?
In chapter four, Stove presents Hume's argument for scepticism about the unobserved (A in diagram and table below), quoting from three primary sources — A Treatise of Human Nature, An Abstract [of A Treatise of Human Nature] and An Enquiry concerning Human Understanding. He supports his reading by quotes from the secondary literature, where his interpretation of Hume might otherwise be challenged. He concludes that deductivism (O in diagram and table below) is the "key premise of irrationalism". In Stove's words, "Nothing fatal to empiricist philosophy of science ... follows from the admission that arguments from the observed to the unobserved are not the best; unless this assumption was combined, as it was with Hume, with the fatal assumption that only the best will do [emphasis original]." He concludes the chapter with the following diagram and table.
|A||Scepticism about the unobserved||There is no reason to believe any contingent proposition about the unobserved.|
|B||Empiricism||Any reason to believe a contingent proposition about the unobserved is a proposition about the observed.|
|C||Inductive Scepticism||No proposition about the observed is a reason to believe a contingent proposition about the observed.|
|D||Impotence of the a priori||No necessary truth is a reason to believe any contingent proposition.|
|E||Accessibles necessary or observational||A proposition is directly accessible to knowledge or reasonable belief if and only if it is either a necessary truth or a proposition about the observed.|
|F||Reasons must be accessible||If P is a reason or part of a reason to believe Q then P is directly accessible to knowledge or reasonable belief.|
|G||Induction is invalid without Resemblance||Any inductive argument is invalid, and the validator of it is a Resemblance Thesis.|
|H||Resemblance is a contingent feature of the Universe||A Resemblance Thesis is a contingent proposition about the unobserved.|
|I||Resemblance is not provable a priori||A Resemblance Thesis is not deducible from necessary truths.|
|J||No contingents provable a priori||No contingent proposition is deducible from necessary truths.|
|K||Resemblance is not provable a posteriori||A Resemblance Thesis is not deducible from propositions about the observed.|
|L||Induction to Resemblance is circular if valid||A Resemblance Thesis is deducible from propositions about the observed only when to the latter is conjoined a Resemblance Thesis.|
|M||The validator of induction not necessary or observational||Any inductive argument is invalid, and the validator of it is neither a necessary truth nor a proposition about the observed.|
|M+||No validator of induction is necessary or observational||Any inductive argument is invalid, and any validator of it is neither a necessary truth nor a proposition about the observed.|
|N||Invalidity of induction incurable||Any inductive argument is invalid, and any validator of it is not a reason or part of a reason to believe its conclusion.|
|O||Deductivism||P is a reason to believe Q only if the argument from P to Q is valid, or there is a validator of it which is either a necessary truth or a proposition about the observed.|