falsificationism


falsificationism

the methodological position (particularly associated with Karl Popper, 1934) based on the notion that while an inductive universal generalization can never be finally verified, given the ever-present possibility of new and potentially refuting evidence, a single nonsupporting occurrence can refute a hypothesis (e.g. a single black swan refutes the general hypothesis that ‘all swans are white’). According to this view (and in contrast with LOGICAL POSITIVISM, see also EMPIRICISM), science can be defined in terms of the ‘falsifiability’ rather than the ‘verifiability’ of its theories and hypotheses, and the essential provisionality of scientific knowledge acknowledged. For Popper, the ‘falsifiability’ of a discipline's propositions is the decisive criterion ofdemarcation between science and non-science.

A virtue of this ‘realist’, rather than empiricist, position, is that it recognizes the importance of hypotheses and theories within science, and of changes in scientific knowledge, thus also captures something of the ‘critical spirit’ of science. Hence, this position is sometimes also referred to as critical rationalism.

Although it has attracted some support among social scientists, critics of falsificationism challenge its cogency on a number of counts:

  1. that ‘the facts’ which are put forward as the basis of the ‘independent’ test of theories and hypotheses are themselves ‘theory-laden’ – experiments, for example, are both constituted by and interpreted using theories;
  2. in practice, in science, and contrary to the position that can be termed naive falsificationism, it turns out that a single refutation is rarely decisive, the rejection and replacement of theories being a matter of a more overall judgement of the cogency and effectiveness of theories;
  3. the attempt (see Lakatos and Musgrave, 1970) to replace naive falsificationism with a sophisticated falsificationism, in which an overall judgement is made between progressive and degenerating scientific research programmes, fails to overcome the problems of falsificationism, for if no single observation is decisive, falsification loses its distinctive position; it no longer provides a clear cut rule of thumb in the day-to day procedures of science, or any clear overall demarcation between science and non-science.

For many commentators (e.g. see FEYERABEND, 1975), the procedures suggested for science by falsificationists simply fail to fit the past and present activities of science, and if used strictly would be likely to cripple it. See also COVERING-LAW MODEL AND DEDUCTIVE NOMOLOGICAL EXPLANATION, HYPOTHETICO-DEDUCTIVE EXPLANATION AND METHOD, SOCIOLOGY OF SCIENCE, SCIENTIFIC PARADIGM.