Social Research Glossary

Social Research Glossary

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home

Citation reference: Harvey, L., 2012-24, Social Research Glossary, Quality Research International, http://www.qualityresearchinternational.com/socialresearch/

This is a dynamic glossary and the author would welcome any e-mail suggestions for additions or amendments. Page updated 8 January, 2024 , © Lee Harvey 2012–2024.

_________________________________________________________________

Methodology of scientific research programmes

core definition

The methodology of scientific research programmes is a sophististication of falsificationism in response to the notion of scientific paradigms.

explanatory context

The methodology of scientific research programmes was developed by Imre Lakatos in response to Kuhn's paradigm thesis and its attack on falsificationism. Lakatos' model of the development of scientific knowledge develops and considerably sophisticates the falsificationist perspective.

Lakatos focuses on the internal history of science and sees science as a community of practitioners developing research programmes that are launched enthusiastically and worked on despite apparent falsification. Ignoring anomalies is not an irrational act by the programmatic researcher provided new developements are being made. However, once the momentum falters and the programme begins to stagnate practitioners will begin to develop other lines of enquiry or interests. It is the replacing of one programme by another with more empirical and theoretical content that constitutes the rational development of scientific knowledge.

A scientific research programme is characterised by a 'hard core' and a 'protective belt' of assumptions. The hard core is the 'unfalsifiable central tenets', the core assumptions upon which all work within the programme is based. The protective belt comprises the surrounding set of assumptions that are, ultimately, subject to negotiation and amendment in the light of work carried out in the programme, or as a result of discoveries made elsewhere that conflict with the operating principles of the programme.

The Lakatosian scientist works within a programme. The programme is based on a theory and the work is programmatic as it predicts phenonema and it is contingent upon the scientist to ascertain the accuracy of such predictions. A continual process of reformulation of the theory ensues, through changes in the protective belt. When the programme is in a progresive phase the reformulations lead to predictions of novel phenomena. As Lakatos points out, such reformulation is rarely ostensive but is easily rationally reconstructible.

Anomalies have only retrospective importance at the progressive stage. It is at the degenerating stage that anomalies achieve more prominence. Once the reformulation of a theory results in no new testable predictions, the programme starts to degenerate. Ad hoc adjustments are made to account for observed phenomena that conflict with the predictions of the theory. When observations, external to the programme, are cited against it, these too are usually accounted for by ad hoc adjustments or ignored altogether as irrelevant due to incompatible theoretical perspectives (all observation being theory dependent).

Nonetheless it is not observation and testing that is at the heart of Lakatos methodology. The progressive scientist is the single minded theoretician. Lakatos places a premium on the theoretical autonomy of science. Any mediation by 'fact' is supplementary.

See below for a detailed analysis of Lakatos' Methodology of Scientific Research Programmes

analytical review

Loyola University (undated):

Lakatos presented what many considered to be the Popperian side's response to Kuhn whose defense of dogmatism in normal science was already seen as "heretical" from a falsificationist point of view. Popper had claimed that his account of the history of science in terms of "conjectures and refutations" exhibited the actual historical "rationality" driving the growth of knowledge over time. For Kuhn the decision to switch paradigms is not determined by any such "rationality"; Lakatos explicitly attacked Kuhn as making scientific belief subject to non-rational methods of mass persuasion, as fickle as matters of taste and style. Thus he sets out to build a theory of the rationality of the growth of scientific belief over time which remained true to Popper's falsificationist views but admitted the hisotrical evidence that Kuhn had presented to show that scientists do not abadon theories when confronted by so-called "counterinstances."...

As a research programme progresses, scientists will attempt to refute or falisify the then accepted theory, in good falsificationist fashion. This is Lakatos's Popperian heritage. But when refuting evidence is encountered, according to the Lakatosian picture, the scientist will not consider the programme as "refuted." Instead he/she will begin to alter the assumptions of the "protective belt" in ways premitted or suggested by the positive heuristic, such that the "hard core" of the programme can be retained unscathed. As long as such moves enable scientists to predict more new phenomena (i.e. it is theoretically progressive), and at least some of those predictions get confirmed by observation (i.e., it is intermittantly empirically progressive), the programme is progressing and it is rational to pursue it....

But Lakatos's model offers no help in answering this crucial question. On the Lakatosian picture the individual scientist has to make a "subjective" judgment call on whether his programme is merely experiencing a temporary lull in its progressive shifts or if it has truly begun to degenerate. Thus such a scientist's "theory choice" is no more dictated by a scientific "rationality" than the decision to change paradigms is for the scientist on the Kuhnian image of science. But this was the very problem which Lakatos regarded as making Kuhn's analysis unacceptable and which he set out to remedy. For this reason, many philosophers have concluded that while Lakatos's image of science does indeed have much to teach us, as a model for scientific rationality -designed to avert the "rationality crisis"- Lakatos's methodology of scientific research programmes has failed. Unfortuantely Lakatos himself did not live long enough to develop his view further and/or answer these criticisms.

associated issues

Introduction

Lakatos' model of the development of scientific knowledge develops and considerably sophisticates the falsificationist perspective. He took account of Kuhn's objections to the Popperian edifice and offered an historically more 'accurate' model than Popper's. However, Lakatos embraces neither the paradigmatic mechanism nor Kuhn's ostensive irrationalism. The central element of Lakatos' approach is the scientific research programme whose history is rationally reconstructed. Lakatos focuses on the internal historiography of science, taking account of external factors only as they constitute rational elements in scientific decisions. These he then subsumes under the rubric of 'internal and rational'. (Unlike Kuhn who admitted psychologically irrational external factors in order to provide a reason for paradigm switches).

Lakatos posits a view of science as community based practitioners developing research programmes. Such programmes are launched enthusiastically and, in the initial stages, persevered with despite difficulties and evident falsifications. Anomalies at this stage are ignored. The programme becomes established and while a forward momentum persists, usually thanks to successful research results, the programme will continue to develop, and gradually anomalies will be overcome. Ignoring anomalies is not an irrational act by the programmatic researcher provided new developments are being made. However, once the momentum falters and the programme begins to stagnate practitioners will begin to move off, either developing entirely new interests or gravitate towards a programme that exhibits a progressive problemshift. It is the replacing of one programme by another with more empirical and theoretical content that constitutes the rational development of scientific knowledge.

Scientific research programmes

In Lakatos' scheme, scientists engaged in practice of any science are operating within a research programme. A scientific research programme is characterised by a 'hard core' and a 'protective belt' of assumptions and at any point a programme is apparently progressive stagnant or degenerating. The hard core is the 'unfalsifiable central tenets', the core assumptions upon which all work within the programme is based. The protective belt comprises the surrounding set of assumptions that are, ultimately, subject to negotiation and amendment in the light of work carried out in the programme, or as a result of discoveries made elsewhere that conflict with the operating principles of the programme. The research programme is, in itself, a constructive enterprise, seeking to discover novel phenomena and develop the theoretical framework through a sophistication of the protective belt

The methodological rules that outline the direction of a research programme are of two sorts. The negative heuristic informs the members of which paths of research to avoid and the positive heuristic informs them of the paths to pursue

Positive heuristic

Lakatos points out that a research programme will always be faced by anomalies as yet unexplained. However one should not presuppose that puzzle solving (the resolution of anomalies) is a random affair (as he implies Kuhn does) nor that the construction and manipulation of the protective belt is 'spontaneous', rather, Lakatos suggests, there is order in these activities.

'The order is usually decided in the theoretician's cabinet independently of the known anomalies. Few theoretical scientists engaged in a research programme pay undue attention to 'refutations'. They have a long-term research policy that anticipates these refutations. This research policy, or order of research, is set out—in more or less detail—in the positive heuristic of the research programme.' (Lakatos, 1970, p. 135).

Thus, argues Lakatos, the positive heuristic allows the researcher to make sense of the research sphere, principally by anticipating problematic areas and projecting likely resolutions. The positive heuristic saves the scientist from becoming confused by the anomalies, tackling the problem by setting out in a programmatic way an ever more complex and comprehensive model simulating reality. The scientist is thus concerned to build up the model following instructions that are laid down in the positive part of the programme. Effectively, then, the scientist ignores the available data in as much as counter examples are shelved. As Lakatos notes,

'If a scientist (or mathematician) has a positive heuristic, he refuses to be drawn into observation. He will "lie down on his couch and forget about the data"... Occasionally, of course, he will ask Nature a shrewd question: he will then be encouraged by Nature's yes, but not discouraged by its no.' (Lakatos, 1970, footnote 1 to page 135).

Newton's work again provides an example of the positive heuristic of research programmes. Lakatos claims that the subsequent developments in Newton's programme were all foreseeable at the time Newton developed his first naive model. It is essential, Lakatos maintains, to think of developments in research programmes as models (of increasing sophistication). A model is a set of initial conditions that one knows is bound to be replaced, the positive heuristic provides the key to foreseeing, more or less, how these initial conditions will be superseded.

'This shows once more how irrelevant 'refutations' of any specific variant are in a research programme: their existence is fully expected, the positive heuristic is there as the strategy both for predicting (producing) and digesting them.' (Lakatos, 1970, p. 136).

In short, then, Lakatos is arguing that scientists work out a programme and that falsifiability operates only when the momentum of the programme diminishes. The negative heuristic specifies the 'hard core' and is irrefutable by dint of a methodological decision of the research workers, thus providing secure and stable basis from which to work. The positive heuristic consists of a 'partially articulated set of suggestions and hints' on how to develop the research, that is, how to 'sophisticate the protective belt'. A programme is progressive while such sophistication continues.

Negative heuristic
Lakatos maintains that 'all scientific research programmes may be characterised by their hard core'. The negative heuristic involves the articulation of this hard core and the deflection of research interests away from it. The negative heuristic protects the 'hard core' by demanding ingenuity in the construction of auxiliary hypotheses that form a 'protective belt' around the core. All research activity aimed at the core must be directed to the belt of auxiliary hypotheses. This protective belt comes under scrutiny and is adjusted (or even replaced) to maintain the integrity of the core. The success of a research programme, then, depends on these changes in the protective belt resulting in 'progressive problemshifts'.

By way of example, Lakatos cites Newtonian theory that he regards as a classic example of a successful research programme.

'In Newton's programme the negative heuristic bids us to direct the modus tollens from Newton's three laws of dynamics to his law of gravitation. This 'core' is 'irrefutable' by the methodological decision of its protagonists: anomalies must lead to changes only in the 'protective belt' of auxiliary 'observational' hypotheses and initial conditions'. (Lakatos, 1970, p. 132)

Problemshifts
Lakatos sees science as a rational enterprise. The development of science, can, he asserts, be reconstructed objectively. The focus of attention of this objective rational reconstruction is to be 'problemshifts' in scientific theories.

While a programme continues to develop its positive heuristic successfully without compromising its hard core, and is similarly able to account for discoveries that are external to the programme, it is said to be a progressive programme. When it ceases to produce novel phenomena and has to content itself with amendments to its protective belt of assumptions in the light of external developments, it is stagnating and when such external developments fundamentally challenge the hard core without the programme answering through novel discoveries, then it is said to be a degenerating research programme.

Lakatos refers to problemshifts when examining the progress or degeneration of research programmes. A problemshift is a shift in theory due to anomalies that raise problems.

The concept of problemshift emerges from Lakatos' analysis of the debate between naive and sophisticated methodological falsificationists. Taking up Popper's banner, Lakatos argues that, like the conventionalists Popper agrees that theories and factual propositions can always be made compatible by the addition of auxiliary hypotheses to the theory (for example the development of phlogiston theory). However, Popper demands that one distinguish between scientific and pseudoscientific adjustments that are synonymous to rational and irrational developments of theory. Popper, somewhat vaguely maintains that auxiliary hypotheses that do not satisfy well-defined conditions (ad hoc hypotheses as he calls them) do not represent progress. Lakatos tries to articulate what constitutes admissible criteria for auxiliary hypotheses and thereby shifts the analysis away from a single theory to a series of theories. The serial nature of theories being the only way a sophisticated methodological falsificationist can determine the 'falsification' of a theory.

Lakatos sees a progressive problemshift as one where a problemshift is both theoretically and empirically progressive. Where this is not the case the problemshift is degenerative.

A series of theories will be theoretically progressive if each new theory has some excess empirical content over its predecessor. Excess empirical content means that the new theory predicts some novel, hitherto unexpected fact. Similarly an empirically progressive problemshift is characterised by the corroboration of excess empirical content, that is, each new theory leads to the actual discovery of a new fact.

Progress is measured by the degree to which a problemshift is progressive, that is, by the degree to which a series of theories leads to the discovery of novel facts. A theory in the series is 'falsified' when it is superceded by a theory with a higher corroborated content. If a research programme progressively explains more than competing research programmes then, for Lakatos, the competitor is 'superceded' and may be eliminated, or at least shelved.

The growth of science is thus characterised by this serial nature of theoretical development. Indeed Lakatos maintains that the most important series of theories for the history of science are those that are characterised by some continuity of personnel as well as ideas. Such a continuity evolves from a 'genuine research programme adumbrated at the start'. For Lakatos, a scientific research programme must be programmatic. It should map out future programmes of research and to do this it must be internally coherent and provide a context for the discovery of novel phenomena. This situation he sees as obtaining only in the natural and pure sciences and not in the social world. Freudianism and Marxism do not fulfill the latter criteria and sociology is not programmatic, according to Lakatos.

Anomalies
Lakatos' insistence on the programmatic nature of science leads him to differ with Popper. Unlike the naive falsificationist, whose motivation is the falsifying experiment, or the more sophisticated falsificationist whose bold conjecture is based upon a reflection upon an anomaly, Lakatos emphasises the construction of theory irrespective of anomalous examples.

Lakatos' scientist is single-minded in the pursuit of a programme, anomalies can be accosted later. The falsificationist maintains that learning from experience is learning from refuting instances, Lakatos, on the contrary, suggests that refuting instances (the crucial experiment of falsificationism) are only seen to have been crucial a long time afterwards.

For Popper, a crucial experiment is one in which an accepted basic statement is shown to be inconsistent with a theory. This is of no great consequence for Lakatos because the hard core of a programme is not a single statement but a universal unfalsifiable set of propositions and an inconsistent singular statement alone will not cause a scientist to reject the programme although it may prompt amendments to the protective belt.

Lakatos asserts that the history of science clearly shows the persistence of 'refuted' (in Popperian terms) research programmes. Some of these refuted programmes emerge as genuinely progressive and no programme would take off at all if the Popperian falsificationist principle were rigidly adhered to.

All new programmes, Lakatos suggests, are initially inadequately propounded. Even those programmes that are false may, through human ingenuity be defended 'progressively' for some considerable time and the crucial falsifying instance will only appear in retrospect when a new programme has emerged and replaced the old 'false' one. Anomalies become a driving force only as the positive heuristic of the research programme fades.

Convention plays a role in determining the hard core of a research programme. The adoption of universal theories, by convention, is, for Lakatos, a rational procedure and one that sustains the continuity of scientific growth. The scientist, having accepted a conventional 'universal theory' constructs a programme that seeks to supplement the irrefutable hard core so that it may explain and predict real phenomena. this positive heuristic 'defines problems, outlines the construction of a belt of auxiliary hypotheses, foresees anomalies and turns them victoriously into examples, all according to a preconceived plan. the scientist lists anomalies, but as long as his research programme sustains its momentum he may freely put them aside. It is primarily the positive heuristic of his programme, not the anomalies, which dictate the choice of his problems.' (Lakatos, 1975, pp. 9–10.).

Lakatos' model of the development of science
The Lakatosian scientist, therefore, works within a programme. The programme is based on a theory and the work is programmatic as the elaboration of the theory involves specification of applications and observations appropriate to it.

In short, it predicts phenomena and it is contingent upon the scientist to ascertain the accuracy of such predictions. The structure of that work is dependent upon an internally consistent set of predictions the consistency being assured by the immutability of the hard core of the theory. Scientific observation results in an assessment of the theoretical structure, and a revision of the protective belt of assumptions. The protective belt constitutes the negotiable element, it is these that are amended (progressively or otherwise) in order to achieve congruence between theory and observed 'reality'. The work of the scientist is thus to articulate the protective belt.

A continual process of reformulation of the theory ensues, through changes in the protective belt. When the programme is in a progressive phase the reformulations lead to predictions of novel phenomena. As Lakatos points out, such reformulation is rarely ostensive, but is easily rationally reconstructable.

Superficially, then, the researcher working in a programme appears to continuously test a series of theories with a common root, however, such 'testing' is not as the falsificationists would have it. The scientist working in the research programme does not abandon work when anomalies arise, rather, the anomalies are shelved and work proceeds by developing theories positively and not by rejecting them.

It may arise that subsequent external observation leads to the revival of the degenerating research programme but, until that happens (if it ever does), those scientists who persist with a degenerating research programme are conservatively clinging to a sinking vessel. The more likely event and certainly one not ruled out by Lakatos, is that scientists operate within more than one programme and are able to find clues in one that provide for development in another. This facilitates transition, provides a content for the synthesis of programmes (such as Maxwell's electro-magnetic theory) and undervalues revolutionary moments in the history of science.

Nonetheless it is not observation and testing that is at the heart of Lakatos methodology. The progressive scientist is the single minded theoretician. It is the positive element in the research programme that motivates the scientist.

This positive element is rationally organised and aimed at a more revealing exposition of the complexities of the real world. Such a goal provides the internal logic of scientific development and change. As Lakatos states

'According to my methodology the great scientific achievements are research programmes that can be evaluated in terms of progressive and degenerating problemshifts; and scientific revolutions consisting of one research programme superseding (overtaking in progress) another. This methodology offers a new rational reconstruction of science'. (Lakatos, 1975, p. 9)

Clearly, Lakatos is insisting on the theoretical primary of his model. It is important for Lakatos that this is so because he sees science as growing through theory.

'The methodology of scientific research programmes accounts for the relative autonomy of theoretical science: a historical fact whose rationality cannot be explained by earlier falsificationists. Which problems scientists working in powerful research programmes rationally choose, is determined by the positive heuristic of the programme rather than by psychologically worrying (or technologically urgent) anomalies. The anomalies are listed but shoved aside in the hope that they will turn, in due course, into corroborations of the programme. Only those scientists have to rivet their attention on anomalies who are either engaged in trial-and-error exercises or who work in a degenerating phase of a research programme when the positive heuristic ran out of steam. All this, of course, must sound repugnant to naive falsificationists who hold that once a theory is 'refuted' by experiment (by their rule book), it is irrational (and dishonest) to develop it further. One has to replace the old 'refuted' theory by a new, unrefuted one).' (Lakatos, 1970, p. 137–38)

Lakatos therefore places a premium on the theoretical autonomy of science. Any mediation by 'fact' is supplementary. But while the strong positive heuristic of a research programme is reflected by a heavy emphasis on theoretical development prior to experimentation (such that disagreements between theory and observation are resolved by the theoretician rethinking in terms of anomalies and non-perfect relations (i.e. in terms of the protective belt)) the very progressive nature of a research programme is exemplified by the theory's ability to be confirmed by observational test. This apparent paradox is removed, however, by a temporal component.

The testing is important in the formulation of the theory only once it is ready to be thus subjected to test. This readiness is (I) after the hard core is conventionally laid down, (II) once the protective belt has been articulated and (III) (contingent upon (I) and (II) ) once the legitimating context has been articulated and attained a credibility.

Problems of the Methodology of Scientific Research Programmes

One line of attack on Lakatos's model grows out of the inability of his scheme to determine, once and for all, when a research programme has degenerated into extinction. This internal deficiency was something Lakatos anticipated. He writes:

'It is very difficult to decide, especially since one must not demand progress at each single step, when a research programme has degenerated hopelessly or when one of two rival programmes has achieved a decisive advantage over the other. In this methodology, as in Duhem's conventionalism, there can be no instant—let alone mechanical—rationality. Neither the logicians proof of inconsistency, nor the experimental scientist's verdict of anomaly can defeat a research programme in one blow. One can be 'wise' only after the event.... There is never anything inevitable about the triumph of a programme. Also, there is never anything inevitable about its defeat'. (Lakatos 1975, p. 12)

It is possible, for example, for a programme to rise like a phoenix from the ashes and, Lakatos argues, the history of science has shown that this has been the case. However, given that

'It occasionally happens that when a research programme gets into a degenerating phase, a little revolution or a creative shift in its positive heuristic may push it forward again' (Lakatos, 1970, p. 137)

So. how can the scientific community dispose of a theory for good and all? How indeed does science progress in such circumstances? For just as it appears that empirical historical evidence supports the rejuvenation of some degenerating research programmes (particle theory of light, for example) history, equally clearly shows that some programmes are abandoned.

Kuhn, for example, objects to Lakatos' account on this point, and suggests that, to make it a viable model, Lakatos

'Must specify criteria which can be used at the time to distinguish a degenerative from a progressive research programme....otherwise, he has told us nothing at all.' (Kuhn, 1970, p. 239)

While Feyerabend, taking up much the same point, suggests that Lakatos'

'standards have practical face only if they are combined with a time limit (what looks like a degenerating problemshift may be the beginning of a much longer period of advance) but... if you are permitted to wait, why not wait a little longer? Thus the standards Lakatos wants to defend...[can be retained] as a verbal ornament, as a memoir to happier times when it was still thought possible to run a complex and often catastrophic business like science by following a few simple and 'rational' rules.' (Feyerabend, 1970a, p. 125)

However, Lakatos sees these objections as causing no great problem. Unlike, Kuhn, Lakatos does not see it as impossible for more than one programme to 'discover novel facts' at any one time, nor does he require that a scientist be psychologically committed to one programme to the exclusion of all others. Consequently it is possible for inputs from one programme to another to provide a new lease of life to an apparently degenerating programme (e.g., electromagnetic theory provided a new lease of life to the 'action at a distance' theory of magnetism).

'It is not, therefore, irrational for a scientist to persist in his application to a degenerating programme in the hope that it will revive, provided he publicly accepts the track record of competing programmes and does not fool himself or attempt to delude the scientific community. Ultimately, it is this community, through its effective veto of the publication of outmoded, ludicrous, old fashioned or inadequate theories that will ensure the effective burial of degenerative programmes, whether they are finally dead or not. The community of scientists will, in effect, undertake a policy of euthanasia at a public level. This is legitimate from a logical point of view, because the conflict between the programmes means that progress in one necessarily implies degeneration in the other. 'The progress of one programme is a vital factor in the degeneration of its rivals. If programme P1 constantly produces novel facts' these, by definition will be anomalies for the rival programme P2. If P2 accounts for those novel facts only in an ad hoc way, it is degenerating by definition. Thus the more P1 progresses,the more difficult it is for P2 to progress'. (Lakatos, 1975, p. 11. footnote 37.)

Furthermore, at a prescriptive level, Lakatos argues that new research programmes be given a 'breathing space'. While it may not be possible to dispense with a degenerating programme for good an all, it is also not possible to show that a new programme initially exhibits progressive problemshifts at both an empirical and theoretical level. He suggests that it may be 'a very long time' before a new programme is seen to produce 'genuinely novel' facts. Lakatos' prescription thus is that

'we must not discard a budding research programme simply because it has so far failed to overtake a powerful rival. We should not abandon it if, supposing its rival were not there, it would constitute a progressive problemshift'. (Lakatos, 1970, p. 157).

History and Methodology of Scientific Research Programmes
Lakatos is concerned with three interrelated but distinct enterprises: the history of science; the history of the theory of science; and a theory of the production of scientific knowledge. From his investigation of the history of the theory of science he has reconstructed the ongoing battle between rationalist explanations of knowledge and sceptical irrationalism. For Lakatos, the latter heralds nothing but chaos and cannot be conjunctive with the growth of science as he sees it. Consequently he sees science as a rational procedure that the history of science should reflect. The reconstruction of that history is therefore fundamentally an 'internal' production that emphasises the rationality of science.

The theory of the production of knowledge that Lakatos posits, his model of the methodology of scientific research programmes, shows how the development of science has progressed through the internal resolution of problems in a rational fashion. That which cannot legitimately be incorporated as 'internal' history (i.e. is apparently 'irrational'), is 'external' history in Lakatos' scheme. He argues that his methodology calls for supplementary 'empirical-external history'. No rationality theory will ever solve problems like why Mendelian genetics disappeared in Soviet Russia. Indeed, as humans are not entirely rational, a rationally reconstructed history will never be completely comprehensive. However, unlike falsificationists, for whom it does not matter whether the discovery of a fact preceded or followed a theory, the research programme methodologist differentiates between rational and irrational discovery. For example, Planck's discontent with his 1900 radiation formula was illogical for the falsificationists, for whom it constituted a bold falsifiable hypothesis. Planck's reaction to it was explicable only as a psychological reaction. For Lakatos, though, Planck's discontent can be explained internally as a rational condemnation of an ad hoc theory. Similarly, for Lakatos 'metaphysics' is a vital part of the rational reconstruction of science while it constitutes a 'useless' irrefutable irritant for the falsificationist. Lakatos concludes,

'Most historians have hitherto tended to regard the solution of some methodology of scientific research problems as being the monopoly of externalists.... [But] programmes turn many problems which had been external problems for other historiographers into internal ones. But occasionally the borderline is moved in the opposite direction. For instance there may have been an experiment which was accepted instantly—in the absence of a better theory—as a negative crucial experiment. For the falsificationist such acceptance is part of internal history; for me it is not rational and has to be explained in terms of external history. (Lakatos, 1975, p. 14.)

What Lakatos produces is an abstract from the history of science that shows how science has developed rationally.

In short, Lakatos' view of history as applied to his methodology of scientific research programmes is a form of utopian historicism. The utopian Weltanschauung is the progress of science, rationally construed. This constitutes the hard core of his programme and thus the rational reconstruction of the history of science is unashamedly constituted to illuminate this end. The 'objectivism' of this approach, as discussed above, is naive and illusory, and clearly at variance with Lakatos' nomothetic endeavours. He ignores any ideological critique, insisting on an internalist nomothetic approach by absorbing all impinging 'externalist' elements as 'internal' and 'rational', thus providing an ideal climate for the generation of myth.

Post hoc description and Methodology of Scientific Research Programmes
The concentration on the 'internal' history of science and the presupposition of the rational nature of the methodologist's of scientific research programmes reconstructions, exposes the Lakatosian model as itself an ideology of science rather than a theory of the production of scientific knowledge. There is a dialectical interrelationship between 'internal' history and rational reconstruction in Lakatos' scheme that insulates the methodology of scientific research programmes from either historiographic critique or disconfirmation via 'externalities'. The rational reconstruction of science in the Lakatos model is primarily a mythological system in which history is a resource used only when it does not conflict with normative prescriptions about what a rational scientist should do.

Lakatos refers to scientific tradition as supplier of wisdom in choosing between research programmes, however, once a tradition degenerates, historical evidence is ignored and philosophic reflection invoked through the positive heuristic. But the Lakatosian model is unable, except through post hoc reconstructions, to determine the 'power' of any metaphysical heuristic, its development, and the way it interacts with 'objective' scientific knowledge to produce that knowledge. Feyerabend points out that Lakatos expects the scientific community to mediate to dispense with degenerating programmes that have not been abandoned through a restriction of public exposition (journals not publishing articles on degenerative programmes, and no funds being provided for further work in such programmes). This process is clearly sociological and not a process contingent upon the 'rationality' of science. The metaphysical element in the positive heuristic of a programme is thus assailed socially not scientifically.

There is little in Lakatos, it would seem, but a dogmatic prescription for writing the history of science, couched in a sophisticated package, neatly disguised as a rationalist model of the development of science. Lakatos is indifferent to both the context and process of discovery. He is also unable to pursue a critical analysis of science, blinkered, as he is, by rationalist assumptions. Furthermore, as sophisticated as the research programme structure of his model is, it overplays the choice between competing research programmes and like all falsificationists, understates fundamental epistemological shifts in direction of the development of scientific knowledge.

Feyerabend argues that in no demonstrable way does the methodology of scientific research programmes produce rational reconstructions. Furthermore, he argues, so called rational reconstructions of the the history of science by proponents of the methodology are nothing but sociological case studies. Claims to rationality should be disregarded. All that such studies provide are documentations of power struggles in the sociology of knowledge. They are not rationalist histories and thus the methodology does not provide a model of the development of scientific knowledge that it claims to.

Feyerabend attacks the Lakatosian model because it merely assumes the rationality of its rational reconstructions. It can do no more than that. It is logically impossible to argue for that rationality because the whole enterprise is based upon the assumption of the excellence of the standards of science and that these standards are (by fiat) the valid basis for judgements about types of knowledge. The standards of science are taken for granted, they are nowhere assessed. the standards of science are assumed to be of a rational and uniform nature; for science (of the last two centuries) is regarded as the embodiment of 'true' knowledge, its superiority as a producer of 'objective' knowledge is incontrovertible for Lakatosians. Yet, Feyerabend suggests, there is no common scientific wisdom; there are no uniform standards of science. Science is impregnated by value judgements and conflicts of choice are quite evident in its history, such choices often being made for no rational reason.

Lakatos assumes that, while there is no universal criterion for deciding on the scientific character of theories, there is a scientific common wisdom that has assessed individual achievements. But, says Feyerabend, this assumption is unsupported by empirical evidence. The disciplinary and sub-disciplinary breakdown of modern science leads to the adoption of various attitudes towards a given theory.

'The basic value judgements of an experimentalist will differ from those of a theoretician (c.f. Rutherford, or Michelson, or Ehrenhaft on Einstein), a biologist will look at a theory differently from a cosmologist, the faithful Bohrian will regard modifications of the quantum theory with a different eye than will the faithful Einsteinian.' (Feyerabend, 1975, p. 316).

Feyerabend adds that revolutionary eras are characterised by a disunity of conceptualisations and not by uniform judgements and that choices between theories are rarely made for good reasons, thus Newton's theory of gravitation was accepted by some scientists who incorrectly believed it could be derived from Kepler's laws. Feyerabend concludes that

'Common scientific wisdom is not very common and certainly it is not very wise'. (Feyerabend, 1975, p. 317).

Assessment of Methodology of Scientific Research Programmes
Are there any elements in the Lakatosian scheme that would permit a less dogmatic model to be constructed or is the methodology of scientific research programmes simply a prescription for writing the internal history of science ? As a model, does it, or can it, confront questions as to how and why knowledge is produced?

Lakatos does draw attention to shortcomings in other accounts that have made claims to have uncovered the process of the production of scientific knowledge and makes only modest claims for his own work. His model simply says that scientists work on research programmes that grow and shift progressively until they are surpassed by other programmes (which may include the first as part of a synthesis) and that such a process is primarily of internal developments in science; new technology leading to new 'observations' and the disposal of anomalies, and reflective theoretical conjectures of a 'progressive' nature. This scheme operates in natural science, social sciences are excluded because of their irrationalism and failure to predict novel facts and construct a programme of research.

Lakatos's model suggests a smooth development of science, relying more on synthesis than revolution. Yet Lakatos has little to say about the social construction of research programmes. Why do scientists operate within certain programmes, how do such programmes get started, how is synthesis managed, and how, indeed, do scientists make the choice between one theory and another? Lakatos either consigns these problems to the sociology of science or he simply constructs normative idealisations retrospectively. Rationalism, for Lakatos, is the determinant of choice and direction. A scientist rationally chooses a programme that progressively accounts for 'observation'. Lakatos ignores the problematic nature of the rationality of science and the impossibility of any 'basic' observation against which theories may be judged. Therein resides his dogmatism, and this imposes limits on the usefulness of his model.

The one element in Lakatos' model that has potential for an interpretive development is the positive heuristic. The positive heuristic of a programme embodies both the context and process of discovery via its metaphysical element. As Lakatos says:

'One may formulate the 'positive heuristic' of a research programme as a 'metaphysical' principle. For instance one may formulate Newton's programme like this: 'the planets are essentially gravitating spinning-tops of roughly spherical shape'. This ideas was never rigidly maintained: the planets are not just gravitational, they have also, for example, electro-magnetic characteristics which may influence their motion.' (Lakatos, 1970, p. 136).

For Lakatos, the metaphysical principles of the positive heuristic are general flexible guidelines for the development of the protective belt and contrast fundamentally with the ever hardening central core of the theory. What Lakatos chooses to ignore is the process by which this metaphysic is developed, assimilated and propounded by both the scientist involved in the process of discovery and by science in the propagation and legitimation of ideas. The metaphysic of the positive heuristic conditions the scientific community to 'see' the world in a way compatible with the hard core of the theory and therefore permits the theory to be tested. Without the positive heuristic's metaphysical preparatory work the theory would never be 'ready to test'. Yet Lakatos is unconcerned with this area of analysis. He is preoccupied with rationalism and thus disinclined to shift analysis to the science-social milieu level.

Summary and conclusions
Lakatos's model of the growth of science is the outcome of his attempt to reconstruct sophisticated falsificationism so that it reflected the actualities of the history of science as he saw them. His extension of sophisticated falsificationism into an account of scientific progress lays the ground for the smooth transition from justificationist falsificationism to the developmental model of scientific research programmes. Lakatos proposed a model of the production of scientific knowledge that emphasises the internal history of science and exhorts the teleological supremacy of rationalism. The process of scientific advance is through successive and competing research programmes. The approach is useful in as much as it emphasises the heuristic nature of research and the 'guiding role' of metaphysics. It is however internalist, retrospective, normative and lacking in the explanatory power that Lakatos' nomothetic orientation presupposes. It sets aside epistemological shifts and the legitimation process of knowledge production. It adopts a rational reconstructivist approach to history, effectively a utopian historicism.

However, some sociological models of scientific knowledge production (such as Tiryakian's schools and Mullins' networks) utilise (or approximate to some degree) a Lakatosian framework, though usually in conjunction with Kuhnian paradigms. The potential of these models is thus constrained by the limitations of the methodology of scientific research programmes outlined in this review.

related areas