Kuhn’s Structure of Sci Revolutions
Kuhn is widely credited with developing the notion of that sciences proceed by revolutionary changes in paradigm, as against the view that they proceed by accumulation. It is sometimes cited as evidence against the view that science is rational.
I. Introduction: A role for History
Kuhn’s approach is historiographic. He identifies a view of “normal science”:
If science is the constellation of facts, theories, and methods collected in current texts, then … scientific development becomes the piecemeal process by which these items have been added, singly and in combination, to the ever-growing stockpile that constitutes scientific technique and knowledge. (p 1)
II. The Route to Normal Science
While scientists may like to think of science as a principled attempt to refine and extend the scientific corpus, Kuhn shows that scientific work is greatly affected by its social nature. Thus:
… ‘[N]ormal science’ means research firmly based upon one or more past scientific achievements, achievements that some particular scientific community acknowledges for a time as supplying the foundation for its further practice. (The Route to Normal Science, p 10.)
[I]n the early stages of the development of any science different men confronting the same range of phenomena, but not usually all the same particular phenomena, describe and interpret them in different ways. What is surprising, and perhaps also unique in its degree to the fields we call science, is that such initial divergence should ever largely disappear.
For they do disappear to a very considerable extent, and then apparently once and for all. (p 17)
III. The Nature of Normal Science
In its established usage, a paradigm is an accepted model or pattern … . In a science, on the other hand, a paradigm is rarely an object for replication. Instead, like an accepted judicial decision in the common law, it is an object for further articulation and specification under new or more stringent conditions. (p 23)
Mopping-up operations are what engage scientist throughout their careers. They constitute what I am here calling normal science. [T]hat enterprise seems an attempt to force nature into the preformed an relatively inflexible box that the paradigm supplies. No part of the aim of normal science is to call forth new sorts of phenomena; indeed those that do not fit are often not seen at all. Nor do scientists normally aim to invent new theories, and they are often intolerant of those invented by others. …
Perhaps these are defects. (p 24)
[O]f empirical work undertaken to articulate the paradigm theory, resolving some of its residual ambiguities and permitting the solution to problems to which it had previously only drawn attention. (p 27) … Few of these elaborate efforts would have been conceived and none would have been carried out without a paradigm theory to define the problem and to guarantee the existence of a stable solution. (p 28)
IV. Normal Science as Puzzle-solving
Perhaps the most striking feature of the normal research problems we have just encountered is how little they aim to produce major novelties, conceptual or phenomenal. (p 35)
[O]ne of the things a scientific community acquires with a paradigm is a criterion for choosing problems that, once the paradigm is taken for granted, can be assumed to have solutions. To a great extent these are the only problems that the community will admit as scientific or encourage its members to undertake. … One of the reasons why normal science seems to progress so rapidly is that its practitioners concentrate on problems that only their own lack of ingenuity should keep them from solving. (p 37)
V. The Priority of Paradigms
Normal science can proceed without rules only so long as the relevant scientific community accepts without question the particuilar problem-solutions already achieved. Rules … therefore become important and the characteristic unconcern about them … vanish[es] whenever paradigms or models are felt to be insecure.(p 47)
VI. Anomaly and the Emergence of Scientific Discoveries
Discovery commences with the awareness of anomaly, i.e., with the recognition that nature has somehow violated the paradigm-induced expectations that govern normal science. (p 52/3)
To a greater or lesser extent (corresponding to the continuum from the shocking to the unanticipated result) [there are] characteristics common [to] all discoveries from which new sort 0f phenomena emerge. [These] include:
- the previous awareness of an anomaly,
- the gradual and simultaneous emergence of both observationa;l and conceptual recognition,
- and consequent change of paradigm, categories and procedures often accompanied by resistance. (p 62)
[R]ecognizing the process, we can at last begin to see why normal science, a pursuit not directed at novelties and tending at first to suppress them, should nevertheless be so effective in causing them to arise. (p 64)
VII. Crisis and the Emergence of Scientific Theories
All the discoveries [considered] were causes of or contributors to paradigm change. Furthermore, the changes … were all destructive as well as constructive. After the discovery had been assimilated [sic], scientists were able to account for a wider range of of natural phenomena or to account with greater precision for some of those previously known. But that gain was achieved only by discarding some previously standard beliefs or procedures and, simultaneously, by replacing the components of the previous paradigm with others. Shifts of this sort are, I have argued, associated with all discoveries made through normal science, excepting only the unsurprising ones that had been anticipated in all but their details. (p 66)
[T]he emergence of new theories is generally proceeded by a period of pronounced professional insecurity … generated by the persistent failure of the puzzles of normal science to come out as they should. Failure of existing rules is the prelude to a search for new ones. (p 67/8)
The novel theory seems a direct response to crisis. … in the absence of crisis … anticipations had been ignored. (p75)
Philosophers of science have repeatedly demonstrated that more than one theoretical construct can always be placed upon a given collection of data. History of science indicates that, particularly in the early developmental stages of a new paradigm, it is not very difficult to invent such alternatives. But that invention of alternatives is just whatscientists undertake except during the pre-paradigm stage of their science’s development and at very special stages occasions during its subsequent evolution. So long as the tools a paradigm supplies continue to prove capable of solving the problems it defines, science moves fastest and penetrates most deeply through confident employment of those tools. The reason is clear. As in manufacture so in science – retooling is an extravagance to be reserved for the occasion that demands it. The significance of crises is the indication they provide that an occasion for retooling has arrived. (p 76)
VIII. The Response to Crisis
Let us then assume that crises are a necessary precondition for the emergence of novel theories … . Though [scientists] may begin to lose faith and then to consider alternatives, they do not renounce the paradigm that has led them into the crisis. They do not, that is, treat anomalies as counter-instances, though in the vocabulary of the philosophy of science that is what they are. … No process yet disclosed by the historical study of scientific development at all resembles the methodological stereotype of falsification by direct comparison with nature … the act of judgement that leads scientists to reject a previously accepted theory is always based upon more than a comparison of that theory with the world. The decision to reject one paradigm is always simultaneously the decision to accept another, and the judgement leading to that decision involves the comparison of both paradigms with nature and with each other. (p 77)
To reject one paradigm without simultaneously substituting another is to reject science itself. (p 79)
[The objective of normal science] is to solve a puzzle for whose very existence the validity ofd the paradigm must be assumed. Failure … discredits only the scientist and not the theory. … [T]he man who reads a science text can easily take the applications to be the evidence for the theory, the reasons why it ought to be believed. (p 80)
All crisis begin with the blurring of a paradigm and the consequent loosening of the rules for normal research. (p 84)
IX. The Nature and Necessity of Scientific Revolutions
[S]cientific revolutions are here taken to be those non-cumulative developmental episodes in which an older paradigm is replaced by an incompatible new one. (p 91)
Science is likened to politics.
[T]he choice between competing paradigms proves to be the choice between incompatible modes of community life. … Each group uses its own paradigm to argue in that paradigm’s defense. (p 93)
To the extent … that two scientific schools disagree about what is a problem and what a solution, they will inevitably talk through each other when debating the relative merits of their respective paradigms. In the partially circular arguments that regularly result, each paradigm will be shown to satisfy more or less the criteria that it dictates for itself and to fall short of a few of those dictated by its opponent. (p 108/9)
Kuhn argues against what he conceives to be the logical positivist view that scientific theories are necessarily correct, and will only ever be refined.
X. Revolutions as Changes of World View
[P]aradigm changes do cause scientists to see the world of their research-engagement differently. In so far as their only recourse to that world is throught what they see and do, we may want to say that after a revolution scientists are responding to a different world. (p 110)
Scientists … often speak of the “scales falling from the eyes” or of the “lightning flash” that “inundates” a previously obscure puzzle, enabling its components to be seen in a new way that for the first time permits its solution. (p 121)
Far more clearly than the immediate experience from which they in part derive, operations and measurements are paradigm-determined. (p 125)
XI. The Invisibility of Revolutions
Textbooks … being pedagogic vehicles for the perpetuation of normal science … have to be re-written in the aftermath of each scientific revolution, and, once rewritten, they inevitably disguise not only the role but the very existence of of the revolutions that produced them. (p 136)
Partly by selection and partly by distortion , the scientists of earlier ages are implicitly represented as having worked upon the same set of fixed problems and in accordance with the same set of fixed canons that that the most recent revolution in scientific method has made seem scientific. (p 137)
XII. The Resolution of Revolutions
Any new interpretation of nature … emerges first in the minds of one or a few [whose] attention has been intensely concentrated upon the crisis-provoking problems … practice has committed them less deeply than most of their contemporaries to the world-view and rules determined by the old paradigm.(p 143)
In the sciences the testing situation never consists, as puzzle-solving does, simply in the comparison of a single paradigm with nature. Instead, testing occurs as part of the competition between two rival paradigms for the allegiance of the scientific community. (p 144)
[P]robabilistic theories disguise the verification situation as much as they illuminate it. (p 145)
I doubt that [falsifying experiences] exist. … [I]t is just the incompleteness and imperfection of the existing data-theory fit that, at any time, define many of the puzzles that characterize normal science. (p 145)
It makes a great deal of sense to ask which of two competing theories fits the facts better. (p 146)
[T]he proponents of competing paradigms will often disagree about the list of problems that any candidate for a paradigm must resolve. (p 147)
Within the new paradigm, old terms, concepts, and experiments fall into new relationships with each other. (p 148)
The source of resistance is the assurance that the older paradigm will ultimately solve all its problems, that nature can be shove into the box the paradigm provides. (p 150/1)
Probably the single most prevalent claim advanced by the proponents of a new paradigm is that they can solve the problems that have led the old one to a crisis. (p 152)
Claims … are particularly likely to succeed if the new paradigm displays a quantitative precision strikingly better than it older competitor. (p 152/3)
The claim to have solved the crisis-provoking problem is, however, rarely sufficient by itself. (p 153)
[Sometimes a new paradigm will have an apparently absurd consequence, unseen by its originators, which is subsequently verified.] Because of their shock value and because they have so obviously been “built into” the new theory from the start, arguments like these prove especially persuasive. (p 154)
In short, if a new candidate for a paradigm had to be judged from the start by hard-headed people who examined only problem-solving ability, the sciences would experience very few major revolutions. (p 156)
A decision between alternate ways of practicing science is called for, and in the circumstances that decision must be based less on past achievement than on future promise. (p 156/7)
XIII. Progress Through Revolutions
Normally, the members of a mature scientific community work from a single paradigm or from a closely related set. (p 159)
Throughout the pre-paradigm period when there is a multiplicity of competing schools, evidence of progress, except within schools is very hard to find. … the results … do not add up to science as we know it. (p 162)
[T]he abscence at most times of competing schools that question each other’s aims and standards makes the progress of a normal-scientific community far easier to see. … [O]nce the reception of a common paradigm has freed the scientific community from the need constantly to re-examine its first principles, the members of that community can concentrate exclusively upon the subtlest and most esoteric of the phenomena that concern it. Inevitably that increases both the effectiveness and the efficiency with which the group as a whole solves new problems. (p 162/3)
[F]or normal-scientific work, for puzzle-solving within the tradition that the text-boos define, the scientist is almost perfectly equipped. Furthermore, he is also well equipped for … the generation through normal science of significant crises. When they arise, the scientist is not, of course, equally well prepared. (p 165)
[B]ecause the group knows well which problems have already been solved, few scientists will easily be persuaded to adopt a viewpoint that again opens to question many problems that had previously been solved. (p 168)
Why should scientific communities be able to reach a firm consensus unattainable in other fields? (p 172)
At the heart of Kuhn’s thesis is the observation that scientists are rarely educated in a science in the round, but instead trained in a specific school or paradigm, to which they are then committed as if it were a religious orthodoxy. He proposes that it is the community of such normal scientists whose acceptance determines the development of mainstream sciences, selecting from and adapting the work of exceptional scientists, who appear as mavericks. There seems to be some truth in this. It seems to me that people who have been ill or unconventionally educated in the sciences, or who have come to science with some significant relevant intellectual baggage have often played a key role. People who have been ‘an empty slate’ and then received a conventional education, no matter how ‘young’, do not seem to me to challenge paradigms so fruitfully.
50 years on, what is the situation? I can vouch for the fact that, at least in the UK, schools have attempted to educate their students on the nature of ‘science proper’, the leading universities have taken this further, and that while leading scientists generally have good track records at ‘normal science’, in Kuhn’s sense, it is there exceptional insights that makes them leaders. But it is not so obvious that the up-coming generations of ‘young’ scientists have been so well educated. More seriously, it does seem that a generation of economists, in attempting to be more scientific, have aped normal science.
Kuhn presents normal science as a scholastic community, with its strengths and weaknesses, constrained by the scientific method. His contribution was to show that the development of science, in its ‘structure’, was only weakly influenced, if at all, by the use of the scientific method. (It would seem that any other corrective Oracle would do, hence the similarity to an orthodox religion.) But the veracity of science depends on its method.
Kuhn (1st Edition) raises the question of what it is that allows scientists to reach a consensus. It could be that it has more effective thought-police than other disciplines. For example, there are very high costs of entry into particle physics. Apparently in the 2nd Edition (not to hand) Kuhn emphasises the harmonising role of the scientific method, which ensures that two credible theories must be similar in some sense. This still leaves Kuhn’s question as to what it is about the world that makes it sufficiently regular for the scientific method (essentially, statistics) to lead to this harmony. But the ‘brute fact’ seems to be that in the traditional sciences, such as Physics, it does, whereas in economics it doesn’t.
Kuhn focusses on texts, so an appropriate measure of the ‘amount’ of supposed facts that are being changed would seem to be in how much of the mainstream textbooks need revising. In my experience, contra Kuhn, many scientists – as individuals – are seeking to have their names attached to the most complete and foundational re-writes, and actively seeking experiments and data to justify such profound changes. But as soon as they make any claim there is a social process, much as Kuhn describes, which seeks to minimise the impact of any innovations.
To me, the social process that Kuhn describes is common to most human activity, and not what makes science science. Rather, a key part of what makes science science is the respect that scientists have for each other’s search for truth, and their need to understand their subject in their own way. In its essence, good science is rather like a group of artists who encourage each other without being of the same ‘school’. Kuhn’s view is more like the history of art, which is a quite different thing from art.
Kuhn (p 168) seems to suppose that scientists only ever adopt the established viewpoint. From a social science perspective this may be correct: given the kind of social context that Kuhn envisages, scientists might be well advised to keep any doubts about the dominant paradigm to themselves, unless and until they have clear evidence, surprising their colleagues. But ideally, scientists seek to develop a range of theories, and to devise experiments to test them. They exchange ideas for theories and experiments, and experimental results, freely. This would be more scientific than Kuhn envisages, if less human. Perhaps it should be encouraged, particularly in economics. (It seems to me that much of Physics, for example, is already like this.)
Knowledge, Maps and Pragmatism
The Concise Oxford Dictionary defines a science as ‘a branch of knowledge conducted on objective principles …’, where the terms ‘knowledge’ and ‘objective’ can be taken, as Kuhn seems to, to imply that the things known are ‘facts’. But, for example, are the claims made by the laws of thermodynamics factual, or is it only a fact that their claims have not (yet) been falsified? Much of Kuhn’s account seems to assume the former. Thus he supposes that people believe the laws of thermodynamics and are committed to them. They believe that they have a perfect map of the territory. In any situation where the map can be used, it is used without thought or hesitation. As Kuhn says, this is more efficient than stopping to consider whether one has the right map for the job. But by contemporary standards, it clearly isn’t science. Kuhn is nonsense.
But even the best educated and most broadly experienced scientists are not only scientists, they are humans working under pressure. Much of Kuhn’s account would apply equally well to scientists behaving ‘pragmatically’ in the sense that they use ‘the established theory’ unless and until it is falsified and they have a better one. Perhaps scientists working in large organisations (as most necessarily do) are organisational people first, scientists second. Perhaps their outputs are organisational outputs first, science second.
Kuhn’s account is largely descriptive. It indicates that science is not what some peple thought it was, but does not in itself suggest whether this is a good or bad thing, or provide any guidance on possible changes. Indeed, it seems rather fatalistic.
The wikipedia page notes that Kuhn’s binary normal-revolutionary distinction is too crude. Accepting this, we have something more like Whitehead’s process model, Smut’s evolutionary account and Russell’s Theory of Knowledge. These offer more positive and potentially normative insights (some would say too much so).
One underpinning concept is that science has experiments, X, that give rise to data, D, that are to be explained by a theory, T. Kuhn mainly considers sciences like physics, where one can get a lot of data from controlled experiments. In such cases there are two logical criteria for a good theory:
- The likelihood, P(D|T,X) should not be exceeded by any other theory.
- The log likelihood should approximate its expected value.
Kuhn’s findings suggest that often theories have a part, F, that is assumed to be fact, and a part, V, that can be varied, so that one only considers theories T=(V,F) in (1) unless (2) cannot be satisfied, in which case one breaks down F into, for example, (F1,F0) , and considers variants of F1 and hence looks for theories (V,F0). More generally, one might subsequently find F0 to be unsatisfactory, and hence critique and look for variants. The notion of pragmatism here is something like:
To not challenge accepted facts until one has ‘hard’ data, and to then try to preserve as much of the supposed facts as possible.
In practice (2) rarely holds, because we lack complete theories, and so this seemingly obvious criterion is generally regarded as impractical. But – in principle – we could always modify an experiment to filter out any unexplainable ‘noise’, leaving just data that we think can be explained. Criterion (2) can then be used. If no theory holds then we can filter out more data until we have a theory of some aspect of the original data.
Without something like (2), at least informally, there is no guarantee that our theories are not widely out. It seems to me in the physics lab. (2) often holds, and this is an important part of why science is relatively trustworthy in what it actually says (as distinct from what people may think it says).
Even when there is a well-established theory, T, satisfying (1) and (2) above for all known experiments, a scientist might postulate an alternative, T’, for any reason. For example:
- They have observed a one-off event that – being one-off – does not statistically invalidate T but which suggests T’.
- They were brought up in a different tradition, and see the ‘facts’ in T=(V,F) as arbitrary and culturally-determined.
- They observe that some non-mainstream person or group behaves as if they do not believe T, and seem to be doing well.
- The recognize that T=(V,F) is – strictly – only a theory of a limited domain, but see F being applied analogously much more broadly, and are concerned about possible ill effects.
A scientist, then, can be part of a much broader social process than Kuhn considers. A scientist may not be just a scientist, indoctrinated in a particular school. Where Kuhn thinks of a scientific field as having a historical development like that of a single species, we might instead think of the evolution of sub-species within an ecology, who might develop apart for a while but then cross-breed.
Changes in science paradigms have led to nuclear bombs and computer chips, but have not subverted the things that ordinary life relied upon: the old technologies still worked. In this sense Kuhn’s paradigm shifts are of limited significance. The situation is different in finance. This has wished to mimic a science and has been subject to similar social processes. But the crisis of 2007/8 had a more profound impact, calling into question the previously dominant paradigm. It may be that behavioural economics is or will be the new paradigm, but it is not clear that this is adequate. We seem to be in limbo: the social process does not guarantee that (2) will ever be solved. In finance, belief in the paradigm influences the behaviour of the subject of study, hence a finance paradigm appears to be a much more complex thing than a science paradigm.
I am interested in notions of uncertainty, and particularly in the common idea that uncertainty is ‘nothing but’ (numeric) probability. I have two speculations:
- The nature of the process that Kuhn describes is such that even the exceptional scientist – as a scientist – would ever have cause to doubt the idea that uncertainty is just a number: they could always explain away anomalies by other means.
- Normal science could only work in sufficiently regular situations, and so one would only be faced with a problem with the ‘uncertainty is just a number’ heuristic when one has a crisis for the whole paradigm.
More radical uncertainty, then, is about changes to paradigms. As Kuhn says, this is currently ad-hoc and so outside normal science and perhaps even outside ‘established exceptional science’. Some might say ‘unscientific’. Mathematical, perhaps.
Often, one has what is assumed to be fact, F, together with the data, D, from experiments X, such that only one theory, T, satisfies (1). This happens, for example, when F leaves only one free variable. Where F leaves more degrees of freedom one may have various credible assumptions which, together with F, make a theory T unique. In practice when faced with a theory, e.g. of physics, it is hard to find out what the key assumptions were (if the original workers were even aware of them) and it is often hard to adequately characterise the experiments and experimental data, in order to consider alternatives. Thus if (when?) a flaw is found in the theory it is often very hard to consider alternatives.
We now have widespread use of distributed, collaborative, knowledge management technologies, such as wikis. These are already transforming the above situation. More progress (perhaps underpinned by XML) would be helpful, and might facilitate a more flexible approach to science.
My notes on science.