Imre Lakatos: A Critical Appraisal

2. The Methodology of Scientific
Research Programmes

Book cover: The Invention of Science: A New History of the Scientific Revolution by David Wootton

I take Lakatos' description of science as a system of competing research programmes to be substantially correct. In other words, I do have misgivings and think that it needs to be revised in some of its details. Firstly, how did Lakatos see the nature and function of the research programme?

Lakatos had considered that the unit of appraisal in science was not the isolated theory, but a research programme. To know whether a theory constitutes a part of science, according to Lakatos, it is necessary to know its history. If it had been arrived at by content reducing ad hoc modifications, in the face of anomalies, of earlier theories, it is not 'scientific'. It is a series of theories—a research programme—then, that is deemed 'scientific' or 'pseudo-scientific'.

A research programme, Lakatos explained, is composed of a 'negative heuristic' and a 'positive heuristic'. The 'negative heuristic' specifies the 'hard core' of the programme; its metaphysical foundations or conceptual framework. This 'hard core' is deemed irrefutable by the methodological fiat of the programme's proponents. Every worthwhile research programme develops in an ocean of anomalies. It is the function of the 'negative heuristic' of the programme to prevent such anomalies from refuting the 'hard core' by directing the scientists' attention to the revision of the 'protective belt' of auxiliary hypotheses and initial conditions. Just how the 'protective belt' is to be modified is specified by a partially articulated plan; the 'positive heuristic'.

A research programme was regarded by Lakatos as 'progressive' if the successive modifications of its protective belt satisfy the following two conditions. Firstly, each successive modification must be 'theoretically progressive', or have 'excess empirical content' in the sense that the new theory, which consists of laws of nature, auxiliary hypotheses and initial conditions, must predict some hitherto unexpected, novel fact. Secondly, the modifications must be 'empirically progressive' in the sense that the predicted novel facts must be at least occasionally corroborated. Conversely, a programme that is not 'progressive' is deemed 'degenerating'. Lakatos considered that for a research programme to be 'scientific', it must be at least theoretically progressive. For one research programme to supersede a rival, he added, it must be progressive while its rival is degenerating. Furthermore, it must satisfactorily explain the previous predictive successes of its rival.

The main problems that I have with Lakatos' thesis, as I have outlined it here, are these. Firstly, it is historically inaccurate to claim that every proponent of a scientific research programme had deemed its hard core to be irrefutable by a methodological decision. Musgrave [1976: 457–67] has adequately argued this objection using the Newtonian mechanics programme as an example.

Book cover: Worldviews: An Introduction to the History and Philosophy of Science by Richard DeWitt

We must find a new explanation of the historical continuity of a hard core through successive phases of the programme of which it is a part. A promising explanation may be this. The metaphysical framework principles constituting the hard core would have reduced explanatory value if modified, in the face of anomalies, in the ways suggested by its proponents. Take Musgrave's example of the suggested modifications to the mathematical form of Newton's Universal Law of Gravitation. To have modified the law would have made it inexplicable, for its inverse square law form was understandable in geometric terms, just as the laws regarding light and radiant heat intensity were for the earlier Newtonians. The inverse square law was also later found applicable to magnetic and electrical attractions and repulsions; the later Newtonians similarly understanding the intensity of these phenomena in terms of the surface area of spheres with varying distances from a central point source. So, to have modified the inverse square law in the case of gravitational attraction would have rendered a partly understandable phenomenon wholly inexplicable.

My second misgiving with Lakatos' first thesis is that he regarded a research programme as a series of theories. But is it strictly correct to describe, say, Newton's development of his increasingly complex planetary models as formulations of new theories? It seems more in accord with accepted usage to say that Newton was articulating new versions of the same theory. I shall henceforth follow McMullin's [1976: 416, 420] suggestion and refer to a research programme as developing the one theory.

McMullin has made other important constructive criticisms of Lakatos' thesis; a second one deserving of mention here. Contrary to Lakatos, a 'theory' is not normally understood as including all of the background knowledge required to make specific predictions.[3] As McMullin [1976: 416, 421] notes, such 'auxiliary hypotheses' as theories of measurement are more properly regarded as 'theoretical assumptions over and above those that form an explicit part of the theory itself'. I shall adhere to McMullin's terminology here as well.

The fourth problem that I have is the difficulty in reconstructing the hard core and positive heuristic of actual research programmes. Lakatos [1978a: 119] himself admitted that this can only be done with hindsight, but even here we need reasonably clearly formulated rules to guard against arbitrary selection of materials for inclusion. Feyerabend [1979: 131–6] forcefully presses this point. I have no suggestions to offer on this problem. More work needs to be done in clarifying and strengthening Lakatos' first thesis. In spite of these shortcomings, I regard Lakatos' methodology as being far superior to earlier proposals. I now turn to his second thesis concerning theories of rationality.


  1. [3] For an example of where Lakatos states this view explicitly, see his [1978b: part 2, ch. 8, §6, 175].

Copyright © 2016

You will be interested in
Book cover: The Structure of Scientific Revolutions by Thomas S. Kuhn
Book cover: A Historical Introduction to the Philosophy of Science by John Losee
Book cover: Lakatos: An Introduction by Brendan Larvor
Book cover: Philosophy of Science: An Historical Anthology by Timothy McGrew, Marc Alspector-Kelly and Fritz Allhoff
Book cover: For and Against Method by Imre Lakatos and Paul Feyerabend
Book cover: Philosophy of Science: A Very Short Introduction by Samir Okasha

Share This

  • twitter
  • facebook
  • linkedin
  • googleplus
  • gmail
  • delicious
  • reddit
  • digg
  • newsvine
  • posterous
  • friendfeed
  • googlebookmarks
  • yahoobookmarks
  • yahoobuzz
  • orkut
  • stumbleupon
  • diigo
  • mixx
  • technorati
  • netvibes
  • myspace
  • slashdot
  • blogger
  • tumblr
  • email
Short URL: