DBA 702

2Kuhns_the_Structure_of_Scientific_Revolutions_A..._3454.pdf

Home >Business & Finance homework help >Management homework help >DBA 702

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

(III) Higher-level, quasi-metaphysical commitments displayed by historical study.

(IV) Methodological commitments, commitments ‘which have held for scientists at all times’ (p. 42), such as the commitment to understand the world.

As well as holding over different durations and across different subsets of the relatively disunified body of science, these rules are obviously at very different levels of generality, some concrete and practical, others theoretical. However,

Though there obviously are rules to which all the practitioners of a scientific speciality adhere at a given time, those rules may not by themselves specify all that the practice of those specialists has in common. Normal science is a highly determined activity, but it need not be entirely determined by rules. That is why . . . I intro- duced shared paradigms rather than shared rules, assumptions, and points of view as the source of coherence for normal research traditions. Rules, I suggest, derive from paradigms, but para- digms can guide research even in the absence of rules. (p. 42)

This brings us onto the theme of Kuhn’s next section, section V, the priority of paradigms over rules.

According to Kuhn, historians can identify the paradigms of a mature scientific community pretty easily, by finding a set of recur- rent and repeated illustrations of theories ‘in their conceptual, observational, and instrumental applications’ (p. 43). What they thus find are the tools which members of the scientific community in question use to learn their trade, normally embodied in ‘text- books, lectures, and laboratory exercises’ (ibid.).13 But to determine the paradigms of a mature scientific community isn’t yet to identify the shared rules which members of that community follow. The rules followed are abstracted from paradigms, and are to be found by com- paring the community’s paradigms with one another, and with its current research reports. This search for rules, says Kuhn, is ‘both more difficult and less satisfying than the search for paradigms’ (ibid.). It is frustrating because scientists can agree in their identification of a paradigm, without agreeing in their interpretation of it. Agreeing in the identification of a paradigm is merely a matter of the scientists in a given field acknowledging that one of their number ‘has produced an apparently permanent solution to a group

Preston, John. Kuhn's 'the Structure of Scientific Revolutions' : A Reader's Guide, Bloomsbury Publishing Plc, 2008. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/indwes/detail.action?docID=1644308. Created from indwes on 2021-03-09 12:57:37.

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

of outstanding problems’ (p. 44). A paradigm, though, can guide research in the absence of a shared interpretation.

Kuhn acknowledges that this theme was prefigured by Polanyi, who had argued that science largely depends upon the scientist’s ‘tacit knowing’, knowing which cannot be articulated explicitly, and thus cannot be communicated via words, but which has to be acquired through practice.14 Polanyi contrasted the articulable contents of science, which one can encapsulate in books, with the ‘unspecifiable art of scientific research’ (Polanyi 1958, p. 53), which cannot:

[T]he actual foundations of our scientific beliefs cannot be asserted at all. When we accept a certain set of pre-suppositions and use them as our interpretative framework, we may be said to dwell in them as we do in our own body. Their uncritical accep- tance for the time being consists in a process of assimilation by which we identify ourselves with them. They are not asserted and cannot be asserted, for assertion can only be made within a frame- work with which we have identified ourselves for the time being; as they are themselves our ultimate framework, they are essen- tially inarticulable. (ibid., p. 60, emphasis added)

Polanyi’s approach was a necessary corrective to the then- dominant ones, giving Kuhn the idea that (although normal science consists in the articulation of paradigms) paradigms are not fully articulable. Kuhn was, in this respect, merely one of a number of mid-twentieth- century philosophers for whom science is as much a matter of skilled practice as explicit theory.15

Following the well-known discussion of ‘family resemblance’ concepts in Wittgenstein’s Philosophical Investigations (Wittgenstein 1958, §66ff.), Kuhn then urges that the various research problems and techniques current within a normal-scientific tradition need have nothing in common, but may be related only by resemblance. Because scientists do not need to know why the models they acquire through their scientific education have the status of paradigms, they do not need any full set of rules which tell them why. Paradigms, that is, ‘may be prior to, more binding, and more complete than any set of rules’ (p. 46). They guide research primarily by what Kuhn calls ‘direct modelling’ (p. 47). This is the reason why Kuhn regarded his term ‘paradigm’ as indispensable, and preferred it to terms urged on him by others, such as ‘basic assumption’, ‘intellectual framework’,

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

‘conceptual model’, etc. Paradigms, unlike such things, are largely inarticulable. The result is that although historians can identify paradigms, they can’t fully discover or say what paradigms are. Being exemplars, paradigms are the sort of things scientists learn by example, rather than by any other kind of linguistically explicit instruction.

This important aspect of Kuhn’s views, which commentators often miss or downplay, marks a difference with most preceding philosophies of science, including the logical positivist and logical empiricist tradition, Koyré’s intellectualism, and the work of the French conventionalist philosophers (conceived broadly, to include Pierre Duhem). Kuhn’s views do have something important in common with those of the conventionalists, in stressing certain ‘con- ventional’ components of science. But Kuhn doesn’t conceive the range of aspects of paradigms between which scientists really choose as widely as conventionalists do.16 This helps explain what we might call the invisibility of paradigms. As David Bohm, one of the first to register Kuhn’s focus on these ‘tacit’, or ‘subliminal’ aspects of science once put it, during normal science scientists ‘do not regard its basic character as open to question. Rather, they feel that they are not following a paradigm at all, but are instead simply investigat- ing the actual structure of the world’ (Bohm 1964, p. 378). What paradigm-scientists share is a practice, not a body of theory, rules, definitions, or any other sort of statements. (However, although scientists in a given scientific community share a paradigm, they don’t share the same understanding of it (pp. 44, 50).)

Here there is a further connection with the views of Polanyi, and of Wittgenstein, in the recognition that science has a basis that’s neither rational nor irrational, true nor false, but arational. In so far as science has a ‘foundation’, that is not some set or sets of premises, or even a single method, but rather certain forms of training. Scientists fall into communities not because they agree in their beliefs, but because they agree in what Wittgenstein called ‘forms of life’, i.e. their training, their education. Among the things inculcated in that training is a set of intuitions about similarity. Scientists, that is, are trained to see certain aspects of the phenomena they are study- ing as similar to others. Kuhn’s concept of a paradigm encompasses these similarity relations and, as we shall see, the radical nature of paradigm-changes is largely down to the fact that in such changes, one such set of intuitions about similarity are replaced by another.

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

In support of this ‘priority of paradigms’ over rules, Kuhn makes four points. The first is simply the extreme difficulty of discovering the rules that guide particular normal-scientific traditions. If rules did guide normal science, they should be relatively easy to recover from the actions and explanations given by the rule-followers in question, i.e. the scientists.

The second point is that because scientists always learn concepts, laws and theories as they’re applied, and because those applications accompany the theory in question into the textbooks from which future scientists learn their trade, the process of learning a theory depends on the study of applications. There’s no point in supposing that at any stage in the learning process scientists intuitively abstract a complete set of rules for themselves. Their ability to do successful research can be understood without recourse to ‘hypothetical rules of the game’ (p. 47). Although paradigms represent a kind of consensus, it isn’t a consensus about statements, such as axioms, or the definitions of theoretical terms (ET, p. xviii). Scientists aren’t taught such definitions, Kuhn claims, but ‘standard ways to solve selected problems’ in which theoretical terms figure (ET, p. xix, RSS, p. 298).

The third argument concerns the historical pattern of scientific development. According to Kuhn, rules have an important role in that development only during phases when the security of a para- digm is threatened:

The pre-paradigm period, in particular, is regularly marked by frequent and deep debates over legitimate methods, problems and standards of solution, though these serve rather to define schools than to produce agreement. [But] debates like these do not vanish once and for all with the appearance of a paradigm. Though almost non-existent during periods of normal science, they recur regularly just before and during scientific revolutions, the periods when paradigms are first under attack and then subject to change. . . . When scientists disagree about whether the funda- mental problems of their field have been solved, the search for rules gains a function that it does not ordinarily possess. While paradigms remain secure, however, they can function without agreement over rationalization or without any attempted ratio- nalization at all. (pp. 47–9)

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

And the fourth point is that seeing science in terms of paradigms instead of rules makes the diversity of scientific fields and spe- cialisms easier to understand. Where explicit rules exist, they’re usually shared by a broad scientific group, but paradigms, says Kuhn, need not be. Small scientific revolutions, which couldn’t be explained in terms of the replacement of explicit rules, can still be explained in terms of paradigm-shifts.

Because it is constituted by its paradigms, normal science as a whole, Kuhn stresses, isn’t a single monolithic and unified enterprise that stands or falls with any one paradigm. Often it seems ‘a rather ramshackle structure with little coherence among its various parts’ (p. 49, see also Kuhn in Crombie 1963, p. 387). This is an impor- tant comment, which suggests that Kuhn thinks the degree to which science is monolithic or unified changes over time. The idea that although scientists across a field share a paradigm, it may not amount to the same paradigm for them all is also pertinent (p. 50).

In the important postscript to his book, first published in 1970, Kuhn tells us quite a lot more about paradigms, and develops the concept of a scientific community. He worries, first, that the terms ‘paradigm’ and ‘scientific community’ are interdefined in SSR, and notes that if he were rewriting the book it would begin with discus- sion of the community structure of science.

Kuhn starts from what he calls ‘the intuitive notion of [scientific] community’ (p. 176), as ‘the practitioners of a scientific speciality’ (p. 177). Within this intuitive notion, scientific communities exist at many different levels. At the top, as it were, there is the community of all (natural) scientists. Next there are communities associated with each of the main natural-scientific professions (physicists, chemists, biologists, etc.). At the next level down there are ‘major subgroups’ such as solid-state physicists, radio astronomers, etc. Membership in communities at these three top levels is easy to estab- lish, but it becomes harder at the next level down. Nevertheless Kuhn is confident that such identification is possible, independently of paradigms, via use of citation data (data about which of their peers’ publications any given scientist refers to), for example. Communities at this fourth level, he says, will typically consist of around one hundred members, but individual scientists will often belong to several such groups. It is these communities, Kuhn says, which share paradigms.

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

At this point (p. 179, see also RSS, p. 295) Kuhn admits to another defect of SSR. There, scientific communities are simply identified with scientific subject-matters. But the way we now identify scientific subject-matters need not have been the way they were identified in the past. We may now count as ‘physicists’, for example, scientists who would not have been thought of thus by themselves or by their peers. An important insight of Kuhn’s here is that the struc- ture of the disciplines working on a given field of phenomena can itself change (RSS, p. 290). Scientific communities can therefore be identified with scientific subject-matters only as those subject- matters were conceived at the time in question.

Having identified such a community, though, one can then go on to ask what its members share. SSR’s answer, as we know, was ‘a paradigm’ or ‘a set of paradigms’. But in so far as this means an exemplar or set of exemplars, Kuhn now thinks it inappropriate. Instead, he suggests taking ‘paradigm’ here in the less important but more popular sense, in which it means ‘disciplinary matrix’. As regards this sense, Kuhn tells us that all scientists who work on the same disciplinary matrix will share:

(a) certain symbolic generalizations – theoretical assumptions and ‘laws’ that are deployed ‘without question’ (e.g. Newton’s three laws of motion);

(b) certain models and analogies, (e.g. seeing electrical circuits on the model of steady-state hydrodynamical systems);

(c) an idea of what are the good-making qualities of theories, the scientific virtues or ‘values’ (e.g. simplicity, accuracy, consis- tency, coherence);

(d) certain metaphysical principles – untestable assumptions which play a role in determining the direction of research (e.g. the cor- puscularian hypothesis);

(e) certain exemplars or concrete problem situations, which provide agreement on what constitutes the real problems in that field, and on what would constitute their solution.

These last components of disciplinary matrices, of course, are para- digms in the other sense, exemplars.

Study questions 1. Is a paradigm really a permanent achievement (p. 25), or merely

an apparently permanent one (p. 44)? If a new paradigm is also

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

a promissory note, what assurance do scientists have that it will be able to tackle the puzzles in which normal-scientific activity consists?

2. Should scientists actively seek out anomalies, and regard them as more threatening than Kuhn suggests they do? Is there any alter- native to the restricted vision that normal science imposes which would still preserve the potentially revolutionary character of normal science?

3. Should society exercise control over the puzzles or problems which scientists address?

4. Why is it so hard for the scientists involved to identify paradigms, but relatively easy for the historian of science to do so?

5. Would a wider conception of rules succeed in reinstating the idea that scientific activity is a matter of rule-following?

SECTION 4. ANOMALIES, CRISES, AND HOW SCIENTISTS REACT TO THEM

Sections VI and VII of SSR characterize the potential failures of, the

fragmentation of, and the exit from, normal science. They tackle the

problem: given that normal science does approximately fit the usual image of scientific work as piecemeal accumulation, how does it ever come to result in novelties, either of fact (discoveries) or of theory

(‘inventions’)? Kuhn’s answer is that normal science (work under a

paradigm) constitutes a uniquely powerful way of challenging and ulti-

mately overthrowing paradigms. The restriction of vision that com-

mitment to paradigms involves is a pre-condition for in-depth empirical

investigation, and for the recognition of anomalies. Section VIII

explores how scientists respond when anomalies turn, as they some-

times do, into crises. Kuhn here contrasts his view with Popper’s

falsificationism.

Periods of normal science are characterized by stability and the cumulative extension of knowledge: normal scientific research adds results to the growing stockpile of scientific knowledge. Such periods fit quite well the usual image of science, but they don’t include the elements of discovery or invention that the usual image of science also holds in such esteem. Normal science, Kuhn says, ‘does not aim at novelties of fact or theory and, when successful, finds none. New and unsuspected phenomena are, however, repeatedly

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

uncovered by scientific research, and radical new theories have again and again been invented by scientists. History even suggests that the scientific enterprise has developed a uniquely powerful technique for producing surprises of this sort’ (p. 52).

The discoveries and inventions in question are aspects of paradigm-changes, i.e. scientific revolutions. But how does this happen? How does normal science, research under a paradigm, come to constitute an effective way of inducing paradigm change? After all, normal science seems a deeply conservative and un-revo- lutionary activity. The general shape of Kuhn’s answer is that normal-scientific research, like any other human activity, doesn’t produce only what its practitioners aim at. Scientific revolutions are results of human action, but not of human design: they are unin- tended by-products of intentional activity.

Normal science is characterized by the kind of intensive research which Kuhn calls ‘puzzle-solving’, and which inevitably turns up problems. The normal scientist’s job is to make nature fit the existing paradigm, to solve these ‘puzzles’. To fail in this task is for the sci- entist, but not for their paradigm, to fail. When scientists do fail in this task, we have what Kuhn calls an anomaly, ‘a recognition that nature has somehow violated the paradigm-induced expectations that govern normal science’ (pp. 52–3).17 Anomalies are explored within normal science. Some will be resolved by finding ways to explain the recalcitrant phenomenon. But other anomalies won’t be resolved. Some are recognized for many years without inducing sci- entists to reconsider their theories. And some will prove intractable. It is these which will ultimately provide the impetus for discovery.

In section VI Kuhn distinguishes, in a loose and informal way, between two sorts of discovery: discoveries of fact and discoveries of theory (‘inventions’). His central thesis here is that discoveries are processes or episodes rather than events. As processes, they have a typical structure, starting with scientists becoming aware of an anomaly, continuing with their exploring the area surrounding the anomaly, and finishing when the paradigm theory has been adjusted in such a way that those phenomena are seen in a different way, what was previously anomalous now becoming expected. Because discov- eries of fact are processes, questions such as ‘When was oxygen discovered?’ are ill-formed. And because they are processes which involve the mobilization of paradigms, not just individual scientists, questions like ‘Who discovered oxygen?’ are also ill-formed. Such

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

questions, the ones the usual image of science prompts historians to address, simply have no answers. Discovery is a process which involves recognizing not just that something is but also what it is. But recognizing what something is involves conceptualization, and concept-formation is something that takes time, sometimes years.

Really fundamental discoveries can bring about, or at least involve, a change in paradigm. Often what is announced as a ‘dis- covery’ is in fact the genesis of a new theory. Kuhn discusses several examples of scientific discovery (the discovery of oxygen, of X-rays, and the development of the Leyden jar), in an attempt to generalize about its features. In considering how discoveries emerged, he says that ‘the perception of anomaly – of a phenomenon, that is, for which his paradigm had not readied the investigator – played an essential role in preparing the way for perception of novelty’ (p. 57). The new phenomenon violates deeply entrenched expectations, both theoretical and instrumental.

How it does so is, it has to be said, something of a mystery. Kuhn takes the categories of the existing theory to be in outright conflict with those of the new conceptualization, either in that they contra- dict each other, or in that belief in the old categories implicitly con- stitutes a commitment to the non-existence of items which fall under the new categories. Categories themselves, though, not being claims, cannot contradict one another, and it is unclear why we should think that any associated existence claims are exclusive in the way Kuhn thinks. In what way do scientists who come to believe in objects of a certain kind thereby deny that objects of other kinds exist? That this isn’t always the case is clear from Kuhn’s own example of Roentgen and his contemporaries, whose paradigms didn’t prohibit the exis- tence of X-rays (p. 58). Kuhn appears to think that it’s the scientists in question who make such exclusivity assumptions, rather than their theories (p. 59). But then the question is why they don’t just revise their assumption. The general question of when and why the- ories or paradigms are truly incompatible with one another threat- ens to raise metaphysical issues which Kuhn doesn’t really attempt to deal with.

Kuhn lists the typical characteristics of scientific discoveries as follows:

the previous awareness of anomaly, the gradual and simultane- ous emergence of both observational and conceptual recognition,

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

and the consequent change of paradigm categories and proce- dures often accompanied by resistance. (p. 62)

He then proceeds to argue, by reference to a psychological experi- ment by the Harvard psychologists Jerome Bruner and Leo Postman, that these characteristics are built into perception itself. By briefly presenting experimental subjects with both normal and ‘incongruous’ playing cards (black hearts, red spades, etc.), Bruner and Postman showed that

Perceptual expectancies, whether realistic or wishful, continue to operate so long as they are reinforced by the outcome of events. In short, expectancies continue to mold perceptual organization in a self-sustaining fashion so long as they are confirmed. (Bruner and Postman 1949, p. 208)

In a passage that makes an interesting complement to Popper’s normative account of falsification, they argued that

For as long as possible and by whatever means available, the organism will ward off the perception of the unexpected, those things which do not fit [its] prevailing set. [M]ost people come to depend on a certain constancy in their environment and, save under special conditions, attempt to ward off variations from this state of affairs. (ibid.)

‘Perceptual organization’, they concluded, ‘is powerfully determined by expectations built upon past commerce with the environment’ (ibid., p. 222).

For Kuhn, Bruner and Postman’s experiment then becomes a model for the process of scientific discovery. The norm, both in perception and in scientific research, is the presence of features which are to be expected. Novelties (such as the incongruous playing cards) are perceived only with difficulty, against resistance, within a setting provided by what is expected. At first, only what is anticipated is experienced. In fact, novelties aren’t perceived at all when they are first presented, and some subjects never come to see them. Most people, however, after prolonged exposure to the novel- ties, become aware of what Bruner and Postman called a ‘sense of wrongness’. As a consequence of this feeling, ‘conceptual categories

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

are adjusted until the initially anomalous has become the antici- pated’ (p. 64). Then, discovery can genuinely be said to have taken place.

But this, according to Kuhn, gives us the answer to our question, ‘Why does revolutionary scientific change arise out of normal science?’ The negative aspect of this answer is that normal science involves a degree of professionalization which restricts the vision of the scientist, as well as a constitutional resistance to paradigm change. On the more obviously positive side, normal science also leads to a wealth of detail, and a precise fit between theory and observation, that ‘could be achieved in no other way’ (p. 65):

The areas investigated by normal science are, of course, minis- cule; the enterprise now under discussion has drastically restricted vision. But those restrictions, born from confidence in a paradigm, turn out to be essential to the development of science. By focusing attention upon a small range of relatively esoteric problems, the paradigm forces scientists to investigate some part of nature in a detail and depth that would otherwise be unimaginable. And normal science possesses a built-in mecha- nism that ensures the relaxation of the restrictions that bound research whenever the paradigm from which they derive ceases to function effectively. (p. 24)

These features, rigidity and predictive accuracy, make normal science a taut and sensitive indicator of the presence of problems:

Without the special apparatus that is constructed mainly for antic- ipated functions, the results that lead ultimately to novelty could not occur. And even when the apparatus exists, novelty ordinarily emerges only for the [person] who, knowing with precision what [she or] he should expect, is able to recognize that something has gone wrong. Anomaly appears only against the background pro- vided by the paradigm. The more precise and far-reaching that paradigm is, the more sensitive an indicator it provides of anomaly and hence of an occasion for paradigm change. (p. 65)

Because normal science forms a settled background of precise expectations, novelties stand out clearly. Without that background, they would not do so, and scientists would be unable to tell problems

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

from achievements. What Kuhn once characterized as the ‘dogma- tism’ of mature science, the tendency for paradigm scientists tenaciously to defend their paradigm in the face of anomalies, ensures that the most serious of those anomalies spells the downfall of a paradigm only once that paradigm’s potential resources for dealing with it are exhausted.

Scientists, then, need paradigms (exemplars). Before they have par- adigms what they do constitutes something less than science, and when their paradigms are eventually undermined, their activity is less focused, more philosophical, and largely directed to finding a new par- adigm. Mature sciences, at least, cannot continue without paradigms.

Do these ideas justify the identification (e.g. by Feyerabend) of Kuhn as a paradigm-monist, one who thinks that each science, at any given time, can only have a single paradigm? This will depend which sense of ‘paradigm’ we have in mind. Kuhn is more prone to monism about disciplinary matrices, than to monism about exemplars. There are certain comments in SSR which reveal that he thinks that at any given time, each community involved in each scientific speciality pays homage to more than one exemplar. This is clearest at the beginning of section V, where he talks, for example, of the paradigms members of such communities study and practise, and ‘the para- digms of a mature scientific community’ (p. 43). So when, in the Preface, Kuhn admits there that there are circumstances, albeit rare ones, in which two paradigms can peacefully coexist (p. ix), this comment must presumably pertain to disciplinary matrices. It sug- gests that his view is that mature scientific fields are usually domi- nated by a single disciplinary matrix. (Whether scientific fields can be identified independently of disciplinary matrices, though, is a good question.)

But perhaps Kuhn’s work does contain materials for a certain kind of pluralist critique of science. Science, if Kuhn is right, works with an institutional mechanism which assumes that the natural world can be only one way. (That is what the presumption that rival para- digms exclude one another amounts to.) Kuhn, though, was groping towards a perspective that suggests how this might not be the case, how genuinely different but nevertheless equally good conceptual schemes might apply to the same phenomena (see section 6 of this Guide, below). If this is right, science has institutionalized a mecha- nism that prohibits a perfectly legitimate pluralism about conceptual schemes (or ‘disciplinary matrices’).

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

So much for novelties of fact. What about novelties of theory, which Kuhn calls ‘inventions’? These, he argues, are the more important factor in the production of paradigm-shifts. Discoveries of fact are never sufficient to produce a change in paradigm. They must be accompanied by theoretical innovation. But now we face a question parallel to the question about factual discoveries: how do theoretical innovations arise from normal science? The problem is exacerbated by the fact that normal science seems to be even less directed towards theoretical innovation than towards factual discovery!

Kuhn’s answer is very largely parallel to his answer to that previ- ous question. Awareness of anomaly again plays a leading role. The history of the natural sciences shows that before the transition to a new paradigm the old paradigm is always perceived to be in trouble in some way. The trouble usually takes the form of the failure of normal science puzzles to turn out as the paradigm says they ought. This can multiply until what we have is a breakdown in puzzle- solving activity, which then constitutes the core of what Kuhn calls ‘crisis’. External factors such as political or religious conditions, and the personal circumstances of particular scientists, etc. can deter- mine when a crisis occurs, when it comes to be recognized, and even perhaps the area of theory in which the breakdown happens, but they can’t precipitate crisis itself. That’s a matter internal to the paradigm.

Here we can contrast Kuhn both with his most illustrious fore- bear, and with certain more determinedly sociological thinkers. Koyré had already made a breach in the idea that the development of science was autonomous, but only to the extent of showing its rela- tions to other intellectual disciplines: philosophy (notably, meta- physics) and religion. By understanding the particular scientific ideas he studied in their purely intellectual context, and relating them so closely to philosophy, Koyré was widely perceived to have risked ‘disembodying’ them from their social and technological context. Other historicists and sociologists of knowledge went much further, wanting to show that the development of science depends decisively on precisely such ‘external’ influences. Kuhn, too, sought to go beyond Koyré in this respect, but only to a rather limited extent.

While he did pay attention to the social context of scientific activ- ity, Kuhn shouldn’t be thought of as having done more than

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

introduce discussion of social factors to Anglo-American philoso- phy of science; he was an ‘internalist’ rather than an ‘externalist’ (see Kuhn’s 1968 article ‘The History of Science’, in ET, Hacking 1979, p. 225, and RSS, pp. 287–8). By his own admission, he ignored the role of technological and social conditions, thinking that these would not affect SSR’s main theses (p. x). He insisted, rightly, that whether or not social contextual features are relevant to any area of inquiry must be an empirical and therefore open question. But he then went on to argue that the special social nature of scientific com- munities, their isolation and insulation from ‘external’ factors, was the reason why one generally didn’t have to take account of such factors in understanding the development of scientific ideas. Mature science, he felt, was ‘more fully, though by no means completely, insulated from its social milieu’ than any other discipline (ET, p. xv). So while he thought of his own work as ‘deeply sociological’ in a certain way (ET, p. xx), since it portrayed science as the product of groups, not of individuals, he also inveighed against certain sociolo- gists of science who presented themselves as ‘Kuhnians’, accusing them, for example, of seriously underestimating the role and significance of common values in science (ET, pp. xxi–xxii).

One of the signs of crisis is the re-emergence of some of the fea- tures which characterize the pre-paradigm period, such as the pro- liferation of versions. In extreme cases there are as many versions of the theory as there are leading scientists working on it. Foundational issues are explored in the hope that they will yield new fundamental insight. The solid core of agreement which characterizes normal science is eroded away. This makes it harder and harder to see what the theory is.

Kuhn mobilizes three examples of crisis (that which preceded the emergence of Copernican astronomy, that which preceded the oxygen theory of combustion, and the crisis in late-nineteenth- century physics) which he considers typical. In each of them, new theories emerged only after a crisis preceded by a distinct failure in normal-scientific puzzle-solving, and as a direct response to that crisis. But, ironically, the problems which precipitated the break- down had been around for some time, and had been considered already solved under the existing paradigm. The breakdown, there- fore, showed that even on its home ground, a paradigm cannot ensure that its solutions to problems are definitive. Paradigms can be embarrassed by having their own most illustrious successes collapse

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

on them. Kuhn suggests that the new solutions to the problems in question may well have been anticipated by the development of new theories during the preceding period of normal science. But his func- tional perspective leads him to insist (against Feyerabend, for example) that the development of such theories, being speculative, represents an extravagance at that time. A critic like Feyerabend would of course complain that if scientists didn’t come up with new theories, even during ‘normal’ science, crises might be even less prevalent than they are. But perhaps Kuhn might reply that although crises are a necessary condition for the ‘emergence’ of new theories (p. 77), meaning their advocacy, they aren’t, strictly speaking, a nec- essary condition for their existence. Paradigm scientists, that is, are free to invent new theories, but they are not free to advocate them until something has gone seriously wrong with their existing theory.

How, then, do scientists respond to crises? If one followed the Popperian model, and identified what Kuhn calls ‘anomalies’ with what Popper calls ‘falsifying instances’, one would be led to believe that they respond by giving up their theory. Kuhn instead claims that, as a matter of historical fact, scientists don’t respond to crisis by giving up their paradigm (even when that paradigm is a theory). They don’t compare their paradigms directly with nature, and they don’t treat ‘anomalies’ as falsifying instances, even though ‘in the vocabulary of philosophy of science that is what they are’ (p. 77).

Kuhn’s account of paradigm-transition explains why scientists don’t abandon their paradigm, even when it has encountered prob- lems. An accumulation of anomalies is a necessary but not sufficient condition for such a move. On Kuhn’s account, abandoning a para- digm must always involve taking up an alternative paradigm. Scientists only switch allegiance when there’s some more attractive paradigm to switch allegiance to. This is because science without par- adigms is inconceivable. For a scientist to reject a paradigm simply because of anomalies (counterinstances) is to cease to be a scientist, to leave the scientific community.

The way Kuhn has identified paradigms, after all, means that they include, at least, the scientist’s conceptual ‘toolkits’, their concepts, and without some such toolkit, a scientist is unemployable (see Kuhn in Crombie 1963, p. 387). Kuhn’s story also emphasizes that the rejection of a paradigm doesn’t arise just from comparing it with ‘the facts’. ‘The decision to reject one paradigm is always simultane- ously the decision to accept another, and the judgment leading to

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

that decision involves the comparison of both paradigms with nature and with each other’ (p. 77).

It’s tempting to compare Kuhn’s views with Popper’s here (as Kuhn himself does), and declare that Popper’s story fails to explain both why scientists with problematic theories don’t just drop them, and why scientists compare theories with one another, as well as with the results of observation. This, though, would be misleading, if not plain wrong. First, the theories Popper was talking about aren’t always the paradigms that Kuhn was talking about. One can only directly compare their views where the paradigms in question are theories. Even then, Kuhn’s interest in the actual sociological fate of those who relinquish their paradigm theory would not quite match up with Popper’s interest, which is in how the people in question ought to be treated. Second, Popper does, right from the start, allow a role for theory-comparison in theory-testing. He insists that a theory is tested for internal logical consistency, by way of empirical applications of its conclusions, and by ‘comparison with other theories, chiefly with the aim of determining whether the theory would constitute a scientific advance’ (Popper 1959, pp. 32–3). One might still complain, though, that Popper didn’t make much of this role.

It’s just as well that the presence of counterinstances doesn’t cause scientists to abandon their paradigm, since paradigms, Kuhn insists, are always confronted with counterinstances. In fact, normal science consists in the work scientists do as a response to these. Scientific the- ories which really do solve all their problems cease to be parts of pure science and become engineering tools. In a sort of Gestalt- switch relationship, one person’s ‘puzzles’ are another’s ‘counterin- stances’.18 Without these puzzles/counterinstances, there would be no normal science. But scientists spend most of their time applying their theories, not testing them by looking for negative evidence. That’s why Kuhn feels justified in treating the scientific work in ques- tion as the treatment of puzzles rather than as responses to coun- terinstances. Scientists do engage in theory-testing, as well as in other activities which fit the stereotype of scientific activity, such as con- structing speculative theories and trying speculative experiments, but they do so, according to Kuhn, principally during periods of extraordinary science, after one or more significant anomalies have already set in.

When considering how scientists respond to anomalies, then, it’s important to note that even ‘persistent and recognized anomaly does

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

not always induce crisis’ (p. 81). The reason is that anomalies don’t always remain anomalies: some of them are removed by subsequent puzzle-solving activity. If we now ask, ‘What makes an anomaly worthy of serious scrutiny?’, Kuhn tells us that there’s no general answer (and his account of when puzzles turn into crises is perhaps the sketchiest part of the book). The kinds of factors which enter the process are that the anomaly might be seen as calling into question the paradigm’s fundamental generalizations, or that it inhibits applications with a particular practical importance. When factors like these conspire together, the anomaly becomes more than just another puzzle within normal science. It becomes generally recog- nized as a real problem. If it resists the application of concerted activity on the part of the paradigm’s leading scientists, its resolu- tion becomes a point of honour. The activity that is then focused upon it causes minor deviations from the paradigm. Intense research goes into finding out what kinds of small alterations to the paradigm will yield a solution. The existence of minor deviations becomes exacerbated into the phenomenon of competing ‘schools’, and as a result, the paradigm itself is blurred. The solid agreement which con- stitutes the shared paradigm begins to erode, and attempts are made to articulate alternative theoretical structures. During this period, scientists will even go so far as to resort to philosophical analysis in order to find lines of solution. They will begin to examine the philo- sophical assumptions on which their existing paradigm is based, and doubts about these assumptions will be expressed. This is when the phenomenon of ‘thought-experiments’ comes into its own. It marks the onset of another period of non-normal science, a period of revolutionary scientific change, which is ‘a reconstruction of the field from new fundamentals’ (p. 85, emphasis added).

Kuhn tells us that ‘By concentrating scientific attention upon a narrow area of trouble and by preparing the scientific mind to rec- ognize experimental anomalies for what they are, crisis often prolif- erates new discoveries’ (p. 88). How does this happen? Sometimes the new paradigm is foreshadowed in the extraordinary research that has just gone into solving the anomaly. In other words, sometimes new paradigms are the result of solving some old problem. At other times, ‘no such structure is consciously seen in advance. Instead, the new paradigm, or a sufficient hint to permit later articulation, emerges all at once . . . in the mind of a [scientist] deeply immersed in crisis’ (pp. 89–90). This individual creativeness Kuhn treats as ineffable.

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

His sociological approach allows him to note that it’s usually a young scientist who has the new idea, since the young are less tightly bound into the old paradigm, but it doesn’t allow him to explore the ‘inscrutable’ (ibid.) nature of the psychological process itself.

In his Postscript (p. 181), Kuhn queries whether crises precede revolutions as invariably as SSR implies. But he argues that nothing essential hangs on the answer, that crises need only be the usual pre- ludes to revolution, and that crises may be generated, not by the work of the paradigm-scientists in question, but by the work of sci- entists in related areas.

Study questions 1. Does Kuhn successfully explain how conservative activities

can have revolutionary results, how ‘the productive scientist must be a traditionalist who enjoys playing intricate games by pre- established rules in order to be an innovator who discovers new rules and new pieces with which to play them’ (ET, p. 237, empha- sis added)?

2. Why should the historian’s inability to answer a question (such as ‘When was oxygen discovered?) mean that such a question is intrinsically problematic?

3. Are ‘anomalies’ merely in the eye of the beholder? Are Kuhn’s examples enough to convince us that scientific discoveries are always preceded by the awareness of anomaly?

4. If systems of categories differ, why should they be thought of as incompatible? Why should one such system be thought of as denying the existence of items falling under the categories of another such system?

5. If scientists didn’t have their vision restricted by their commit- ment to paradigms, would they be able to perceive anomalies? Is an informed mind better than an open mind in this respect? Is the restriction involved in commitment to a paradigm a necessary pre-condition for any in-depth investigation of nature?

6. Would scientists be right in responding to a crisis by giving up their paradigm? Would they be right to respond by giving up their theory? Does Kuhn’s view of theory-testing have advantages over Popper’s? Is science without paradigms possible? Is there reason to think such an activity would make significantly less epistemic progress than actually-existing science? In the absence of puzzles, would ‘normal science’ cease to exist?

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

SECTION 5: SCIENTIFIC REVOLUTIONS VERSUS CUMULATIVISM

SSR section IX raises questions about the nature of scientific revolu- tions, and their function in the development of science. Kuhn seeks to

justify his central political metaphor: the rest of his book is supposed

to demonstrate that the historical study of paradigm-change reveals

parallels between science and politics. In this section, Kuhn sets up a

conflict between the resulting view of science and the usual, cumula- tivist view, and suggests reasons for doubting the latter.

What of scientific revolutions? Kuhn has already told us that

The transition from a paradigm in crisis to a new one from which a new tradition of normal science can emerge is far from a cumu- lative process, one achieved by an articulation or extension of the old paradigm. Rather it is a reconstruction of the field from new fundamentals, a reconstruction that changes some of the field’s most elementary theoretical generalizations as well as many of its paradigm methods and applications. (pp. 84–5)

One of the central elements in such a transition, he argued in section VIII, is a change in the perceptions of the members of the scientific community. The scientists will have changed their perceptions of the field, its methods and its goals. Here Kuhn invokes the controversial ‘Gestalt-switch’ metaphor, which we shall return to in the next section.

Another metaphor that Kuhn is using is apparent in the title of his book itself, the political metaphor of revolution. He argues that the parallels between political revolutions and paradigm transitions are rich enough to support this metaphor. Some aspects of these par- allels are (pp. 92–3): the ways in which both kinds of revolutions come into being through an awareness of the inadequacy of the existing institutions, an awareness that they are failing to solve prob- lems; the idea that in both the sense of malfunction is a prerequisite to revolution; the fact that in each area one is faced with ‘a choice between incompatible modes of community life’ (p. 94, emphasis added); the idea that revolutions aim to change the existing institu- tions in ways which those institutions themselves prohibit; and the idea that in the interim transition period normal canons of debate are suspended in favour of decision-making methods involving

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

READING THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

techniques of persuasion rather than argument. As we shall see, this last point is one of the most important.

Scientific revolutions, then, are ‘those non-cumulative develop- mental episodes in which an older paradigm is replaced in whole or in part by an incompatible one’ (p. 92, emphases added). The same applies to the invention of new theories (p. 97). But why must the acceptance of a new paradigm or theory mean the rejection of the existing one?

One can imagine new paradigms and theories being introduced which don’t disturb the old ones, but according to Kuhn this rarely or never actually happens. This brings him into conflict with philosophers of science, such as the logical empiricists, who envis- aged science growing cumulatively through the development of new theories because they believed that old theories were reducible, by logical derivation, to new ones.

Kuhn associates cumulativism with a particular epistemological tradition or ‘paradigm’ which has dominated Western philosophy since the scientific revolution. (He tells us more about this in section X.) But, as we already know, he feels that there is increasing reason to doubt the latter. Some of these reasons derive from psychology, but Kuhn also suggests that cumulativism and its asso- ciated paradigm will be undermined if only we take the history of science seriously. He takes it that an unbiased look at that history will inevitably reveal the existence of a multitude of scientific revolutions.

This, it has to be said, is too superficial. Cumulativists are too many and probably too various to be clearly tied to a single philo- sophical ‘paradigm’. The logical positivists and logical empiricists may be open to the objection that their image of science was formed not by attending to actual science and its history, but rather by working under the a priori presumption that scientists must use methods which make sense when represented in formal logic. But cumulativism doesn’t arise only from ‘logical’, as opposed to histor- ical, approaches to science. It’s naive to think that an unbiased look at the history of science must result in non-cumulativism.

Cumulativist history has been written not only by ‘practitioner’ historians (i.e. scientists who write history) but also by professional historians and philosophers of science, including figures such as William Whewell, Émile Meyerson, Kuhn’s mentor James Bryant Conant and George Sarton.

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .

KUHN’S THE STRUCTURE OF SCIENTIFIC REVOLUTIONS

The prominent nineteenth-century scientist, historian and philosopher William Whewell, just to take one example, is a good example of a sophisticated cumulativist historian of science. He and Kuhn share approaches and views at least as important as those that divide them, including a thorough-going historicist insistence on seeing science as formed by processes which develop in time, a recog- nition that science proliferates and progresses by division and sub- division, and an assessment of metaphysics as essential to science. Whewell, like Kuhn, also divided the history of science into ‘epochs’ of different kinds, including ‘stationary periods’, and saw great indi- vidual scientists as synthesizing elements of a pre-existing context into generalized frameworks. But where Kuhn sees discontinuity in this process, Whewell tried explicitly to go beneath the apparently discontinuous surface of historical events in order to discern the rational developmental relations which drive them. The history of each science, according to Whewell, ‘which may . . . appear like a succession of revolutions, is, in reality, a series of developments’ (Whewell 1984, p. 8). This method of ‘rational reconstruction’ was later taken up by the logical positivists, logical empiricists and Imre Lakatos.

In addition, but crucially, there is an entire tradition of Darwinian historians and philosophers of science (including Ernst Mach and Pierre Duhem, but more recently Stephen Toulmin and David Hull) that not only provide an alternative to Kuhn’s revolutionism, but do so by drawing on a source (evolutionary theory) to which, as we shall see, Kuhn linked his own work.

Kuhn treats Whewell, Mach and Duhem as representatives of a single and determinedly philosophical tradition in the historiography of science (ET, pp. 106–7). But his accusation that cumulativists don’t take the history of science seriously fails to register the inter- pretive latitude available when doing history of science. Neither con- tinuity nor revolution is written on the face of science, and to suppose otherwise is to fail to take account of the fact (of which Kuhn was elsewhere well aware) that history is an interpretive (and therefore partly philosophical) discipline.

This is not to say that Kuhn’s critique of the particular version of cumulativism he discusses is unsuccessful. (Cumulativism can, of course, be naive. But so can revolutionism.) The version in question, which Kuhn finds in early logical positivism, and which is associ- ated with the ‘instrumentalist’ view of theories, certainly seems

C o p yr

ig h t ©

2 0 0 8 . B

lo o m

sb u ry

P u b lis

h in

g P

lc . A

ll ri g h ts

r e se

rv e d .