INTRODUCTION

I am grateful for Professor Grünbaum's criticism of the ‘antifalsificationist’ features of my methodology of scientific research programmes, and I am glad to have the opportunity to reply. I have to start by trying to clarify a basic misunderstanding. My paper opened with the question: ‘Exactly how and what do we learn about scientific theories from experiment?’ Later I made what Grünbaum called my ‘provocative claim’ that ‘we cannot learn from experience the falsehood of any [scientific] theory’. Now if one interprets ‘theory’ as ‘proposition (fallibly) mirroring a fact’, then because of the (epistemologically unbridgeable) gap between fact and proposition my claim is far from being provocative: it is an orthodox common-place. It says that if crucial experiments are to provide experimental disproof, there can be no crucial experiments. If I have a provocative claim, that claim is a stronger one; namely that no experimental result, in isolation, can ever defeat a ‘theory’, whether in my sense (that further work on it is irrational) or in Grünbaum's sense (that the experiment should change our rational belief into rational disbelief). That is, there are no ‘crucial experiments’ even in either of these two weaker senses.

THERE HAVE BEEN NO CRUCIAL EXPERIMENTS IN SCIENCE

I explained and elaborated my negative position concerning ‘crucial experiments’ in several papers between 1968 and 1971, and I tried to put it in a nutshell (admittedly with the aid of dozens of back references) at our Penn symposium. In my view, in science we do not learn simply from conjectures and refutations.