Making Natural Knowledge

Constructivism and the History of Science, with a new Preface

Jan Golinski

• Publisher: University of Chicago Press
• Year: 2008
• Print ISBN: 9780226302317, 9780226302294
• eBook ISBN: 9780226302324
• Topic: Biological Sciences
• Subtopic: Science History
• Index: Biological Sciences

Arguably the best available introduction to constructivism, a research paradigm that has dominated the history of science for the past forty years, Making Natural Knowledge reflects on the importance of this theory, tells the history of its rise to prominence, and traces its most important tensions.

Viewing scientific knowledge as a product of human culture, Jan Golinski challenges the traditional trajectory of the history of science as steady and autonomous progress. In exploring topics such as the social identity of the scientist, the significance of places where science is practiced, and the roles played by language, instruments, and images, Making Natural Knowledge sheds new light on the relations between science and other cultural domains.

"A standard introduction to historically minded scholars interested in the constructivist programme. In fact, it has been called the 'constructivist's bible' in many a conference corridor."—Matthew Eddy, British Journal for the History of Science

1

An Outline of Constructivism

The only condition on which there could be a history of nature is that the events of nature are actions on the part of some thinking being or beings, and that by studying these actions we could discover what were the thoughts which they expressed and think these thoughts for ourselves. This is a condition which probably no one will claim is fulfilled. Consequently the processes of nature are not historical processes and our knowledge of nature, though it may resemble history in certain superficial ways, e.g. by being chronological, is not historical knowledge.

R. G. Collingwood, The Idea of History (1946/1961: 302)

In a break analogous to the one effected by astronomy and physics when they excluded the metaphysical question of the why in favor of the positive (or positivist) inquiry into the how, the human sciences substitute for inquiries into the truth of beliefs (in the existence of God or of the external world, or in the validity of mathematical or logical principles) a historical examination of the genesis of those beliefs.

Pierre Bourdieu, “The Peculiar History of Scientific Reason” (1991: 4)

FROM KUHN TO THE SOCIOLOGY OF SCIENTIFIC KNOWLEDGE

Apparently much against its author’s wishes, Thomas Kuhn’s The Structure of Scientific Revolutions (1962/1970), has come to be seen as the harbinger of the constructivist movement. It therefore seems worth discussing Kuhn at some length to discern what he contributed to the development of this approach to science studies. This will inevitably be a selective reading: Quite different pictures of Kuhn have been given elsewhere, for example, in work on the philosophy of science, where he appears in the company of Karl Popper and Imre Lakatos in discussions of rational criteria for theory choice. The Kuhn I shall describe addresses a rather different set of issues, including practical reasoning in science, the role of authority in pedagogy, the nature of controversies, and the definition of scientific communities. These and other matters were drawn out in commentaries on Kuhn by the proponents of the Strong Programme in the 1970s and early 1980s. David Bloor and Barry Barnes, working at the University of Edinburgh, showed how Kuhn’s work could yield valuable resources for building a constructivist sociology of scientific knowledge.

I shall follow Bloor and Barnes in their appropriation of certain aspects of Kuhn, but I also want to open up other, more directly historical, implications of his work. While his philosophical leanings are evident, Kuhn was not a sociologist, and his primary disciplinary affiliation was to history. It is paradoxical, therefore, that his direct influence among historians has been at best limited. Outsiders might think of Kuhn as the preeminent historian of science of recent decades, but few inside the field have followed the lead of his schematic model of historical change. Kuhn earned widespread recognition for producing a work of broad chronological sweep and clear philosophical significance, in line with a tradition that stretches back to Priestley and Whewell. But it is also striking, as Steve Fuller has noted (1992: 272), that he has turned out to be the last contributor to that genre of “didactic macrohistory of science.” We might ponder what this suggests about the implications of his work for historiography. It appears that some components of Kuhn’s approach proved radically subversive of the narrative themes traditionally used in macrohistories of science.

As is well known, Kuhn’s work was based on a distinction between “normal” and “revolutionary” science. For each scientific discipline, he suggested, a common pattern of maturation would be found, albeit one that occurred in different disciplines at different times. A confused initial period, when research was lacking in coherent organization or aim, was succeeded by the achievement of a “paradigm,” which Kuhn defined as a “universally recognized scientific achievement that for a time provides model problems and solutions to a community of practitioners” (1962/1970: viii). Having achieved its paradigm, a mature discipline would enter the phase of normal science, in which research was directed toward developing the paradigm and applying it to solve appropriate problems. Normal science could, however, break down in a “crisis,” when the accumulation of anomalies (problems that resisted solution by the accepted methods) would obstruct attempts to continue applying the paradigm. Then a revolution would occur: A new paradigm would come in to succeed the old one, in a process that involved both a shift in the psychological framework within which scientists operated and a transformation in the social organization of their community. After the revolutionary upheaval, normal science would resume, governed by the new paradigm. As examples of revolutions, Kuhn referred to the historical transformations associated with the names of Copernicus, Newton, Lavoisier, and Einstein.

The varying interpretations of Kuhn’s work are, to some degree, consequent upon the ambiguities of his crucial term, “paradigm.” One commentator discerned no fewer than twenty-one different meanings of the word in Kuhn’s text (Masterman 1970). This may exaggerate things slightly, but there is no doubt that ambiguities exist. It is not clear, however, what is the most productive way to resolve them. In the second edition of his book, Kuhn himself distinguished two senses of “paradigm.” First was “the entire constellation of beliefs, values, techniques, and so on shared by members of a given community” (Kuhn called this the “sociological” sense). And, second, “the concrete puzzle solutions which, employed as models or examples, can replace explicit rules as a basis for the solution of the remaining puzzles of normal science” (this sense, Kuhn said, was “philosophically . . . the deeper of the two” [1962/1970: 175]).

Constructivist commentators, however, have interpreted the term in a way that merges elements of both of these definitions. They have accepted the basic form of the second definition but combined it with the sociological application suggested for the first. Barnes and Bloor, and more recently Joseph Rouse (1987: chap. 2), have argued that the notion of a paradigm as a concrete exemplar – a model problem solution – suggests a pragmatic alternative to the traditional philosophical view that science is governed by a logical structure of theory, a worldview or Weltanschauung. If paradigms are seen primarily as models, then science appears as an enterprise of practical reasoning governed by accepted conventions rather than by logical deduction from some theoretical structure. This understanding of paradigm, it has been suggested, is both philosophically deeper and sociologically more productive.

There is plenty of warrant for the pragmatic notion of paradigm in Kuhn’s text. When he introduced the idea that scientific education operated primarily through conveying concrete exemplars, he noted that such an exemplar “cannot be fully reduced to logically atomic components which might function in its stead” (1962/1970: 11). Rather than a set of stipulations from which deduction might proceed, a paradigm comprises a pattern or model: “like an accepted judicial decision in the common law, it is an object for further articulation and specification under new or more stringent conditions” (23). Analogical reasoning plays a large part in the application of a paradigm, because “a paradigm developed for one set of phenomena is ambiguous in its application to other closely related ones” (29). Kuhn gave two examples of the extension of a paradigm by analogy: the work to apply the caloric theory of heat to a range of chemical and physical phenomena in the eighteenth and early nineteenth centuries; and the monumental enterprise of rational mechanics in the same period, which was devoted to extending the range of Newton’s laws of motion to cover the movements of terrestrial and celestial bodies. In each case, as paradigm models were extended to new phenomena, significant experimental and theoretical work was required to make them apply to the new situation. These applications cannot be said to have been specified in advance in the original paradigm; they were, rather, new and original extensions of it.

This view of the way science proceeds is in line with the position that Bloor and Barnes described as “finitism.” Bloor defined finitism as “the thesis that the established meaning of a word does not determine its future applications. . . . Meaning is created by acts of use” (1983: 25).

Barnes added, “Finitism denies that inherent properties or meanings attach to concepts and determine their future correct applications; and consequently it denies that truth and falsity are inherent properties of statements” (Barnes 1982: 30–31). Kuhn himself, drawing on the same philosophical resources as Bloor and Barnes, had made the comparison between the application of a paradigm and the attachment of meanings to words. It was Wittgenstein who had first argued that a word derives its meaning, not from an exhaustive definition that can specify all of its possible references, but by a process of extending its usage by analogy from instance to instance. We do not need to know the essence of a “game,” for example, or be able to give a complete definition of the characteristics a game must have, in order to learn to apply the term to a family of activities. As we learn, by trial and error, the range of its conventionally acceptable applications, we learn its meaning. Kuhn noted:

Something of the same sort may well hold for the various research problems and techniques that arise within a single normal-scientific tradition. What these have in common is not that they satisfy some explicit or even some fully discoverable set of rules and assumptions that gives the tradition its character and its hold upon the scientific mind. Instead, they may relate by resemblance and modeling to one or another part of the scientific corpus which the community in question already recognizes as among its established achievements. . . . Paradigms may be prior to, more binding, and more complete than any set of rules for research that could be unequivocally abstracted from them. (1962/1970: 45–46)

The picture we have, then, is one in which exemplary achievements yield a family of techniques, which, in the course of the paradigm’s extension, prove appropriate for solving certain problems. A paradigm is not specifiable as a list of theoretical propositions or methodological stipulations; it is not developed by logical deduction from premises. Rather, the exemplar is learned as a model problem solution and is applied by analogy to what are judged to be similar phenomena. To the extent that the problems presented by new phenomena are solved, the paradigm continues to be adhered to, expanding and modifying its range as time goes on.

As exemplary problem solutions, paradigms are learned as ways of seeing and doing. Quite a lot of the process of scientific education, in Kuhn’s view, consists of imparting unarticulated skills and interpretive dispositions. The perceptual and motor abilities that apprentice scientists have to learn cannot be fully spelled out as a set of rules. Anyone with school experience of learning dissection techniques, for example, will recall how little was taught by the words in a textbook, and how much one depended on the guidance of the teacher, on tinkering, and on comparing one’s findings with those of others. Those who have pursued their scientific education further will be familiar with the sense of accomplishment that comes from mastering a particular experimental skill, and will probably agree that there is no other way to achieve it than learning by doing. Kuhn adopted the phrase “tacit knowledge” from the philosopher and physical chemist Michael Polanyi to characterize a large part of what the scientist learns in the course of being trained in a particular paradigm. Polanyi had argued that scientific practice requires the learning of substantial unspoken skills: “When we accept a certain set of presuppositions and use them as our interpretative framework, we may be said to dwell in them as we do in our own body. . . . [A]s they are themselves our ultimate framework, they are essentially inarticulable” (Polanyi 1958: 60). Education in the sciences appears less like a process of logical programming than an apprenticeship in a traditional craft. Hence, Jerome Ravetz’s (1971) vision of science as “craftsman’s work” in a book that took Polanyi’s perspective as its point of departure.

The appeal of this view for Barnes and Bloor was that it made science look more like other aspects of culture. As a kind of craft, science appears more open to the naturalistic approach to understanding its construction; the epistemological barriers around it are lowered. It is then plausible to regard scientific beliefs as being inculcated by relations of authority within educational institutions and sustained by conventions in specific communities. Echoing the statement of Polanyi (1958: 53) that “to learn by example is to submit to authority,” Barnes wrote, “paradigms, the core of the culture of science, are transmitted and sustained just as is culture generally: scientists accept them and become committed to them as a result of training and socialization, and the commitment is maintained by a developed system of social control” (1985b: 89).

Such a view opens the way to a wide range of possible empirical studies of scientific education, of the workings of institutions, and of the ways in which the authority of science is maintained in the wider society. We shall be considering some of these studies, both sociological and historical, in subsequent chapters. In analyzing these topics, constructivist research has broken with what had been the prevailing approach to understanding the social relations of science. Traditionally, science had been seen as maintained by certain institutions that might be necessary for it to flourish but did not affect the content of what was believed. The constructivist outlook suggests, however, that science is shaped by social relations at its very core – in the details of what is accepted as knowledge and how it is pursued. Kuhn can be read as endorsing one position or the other, depending on the reader’s orientation. For the proponents of the Strong Programme, the more significant reading was the more radical one, in which science was seen as social through and through. This reading built upon Kuhn’s comments to the effect that paradigms are integral to the definition of scientific communities.

Kuhn used the researchers on electricity (known as “electricians”) in the mid-eighteenth century as an example of a scientific community defined by common acceptance of a paradigm. Prior to Benjamin Franklin’s work in the 1740s, he noted, there were numerous views about the nature of electricity, but most were held by only a single experimenter and derived from a limited subset of all the known experiments. Franklin’s research, focused on the Leyden Jar (a device that could store a remarkably large quantity of electric charge), led to his articulation of a single fluid theory of electricity. The idea was that neutral bodies contained a certain proportion of electrical fluid to normal ponderable matter, and that they could be charged positively or negatively by adding or removing fluid. The theory provided reasonable explanations of conduction and neutralization phenomena; it yielded a plausible account of how the Leyden Jar could be charged; and it could even be extended to cover most (though not all) attractions and repulsions between charged bodies. According to Kuhn, the “Franklinian paradigm . . . suggested which experiments would be worth performing and which . . . would not” (1962/1970: 18). The theory, which offered an exemplary solution of an outstanding problem, ended the confusing debate between rival interpretations and gave the electricians a coherent plan for further research.

Kuhn’s claim that Franklin’s theory deserves to be designated a “paradigm” has been questioned by some subsequent historical research. But I am not concerned with that issue as much as with the way Kuhn identified the achievement of a paradigm with consolidation of the social community of researchers. He noted that “the emergence of a paradigm affects the structure of the group that practices the field.” Following Franklin’s work, “the united group of electricians” achieved a “more rigid definition” (1962/1970: 18–19). Those who were not willing to adopt Franklin’s model were condemned to work in isolation. It seems that group solidarity was the result of consensus acceptance of the paradigm, a coherent social group coalescing around recognition of a specific scientific accomplishment. Kuhn did not talk, at this point, about what might be called the “external” sociological dimensions of the group. He did not describe its place in scientific or educational institutions or its members’ positions in society. The implication is that these factors were of secondary importance to the social cohesion that flowed from achievement of the paradigm.

In the “Postscript” to the second edition of his book, however, Kuhn seemed to qualify this point. He noted that there was a circularity inherent in his assumptions that, “A paradigm is what members of the scientific community share, and, conversely a scientific community consists of men who share a paradigm” (1962/1970: 176). The circularity was a source of problems that could be avoided, he suggested, if the investigation were to begin with “a discussion of the community structure of science.” An objective study of scientific institutions could provide a grid upon which the social influence of various paradigms could be mapped. Accordingly, Kuhn referred to research done by sociologists working in the tradition of Robert K. Merton, who had charted some of the institutional features of modern science. They had drawn attention to such factors as uniform educational experiences, a high degree of unanimity of professional judgment, and publications and organizations serving specific disciplines, as the external conditions that sustained the social cohesion of the scientific community.

From the standpoint of the Strong Programme, however, to attempt to identify social structures that were independent of scientists’ commitment to particular forms of practice was to betray one of the fundamental insights suggested by the notion of a paradigm. The circularity of defining a social group and a form of practice in terms of one another need not be a vicious one. Rather, it comes close to what Barnes and Bloor took Wittgenstein to have meant by a “language-game” or “form of life.” For them, it was precisely Kuhn’s willingness to explore the kinds of communities that form around an exemplary model of practice that gave his work its appeal. These groups would be expected to be narrower than scientific disciplines, and certainly narrower than the population of all professional scientists studied by the Mertonian sociologists. Kuhn was suggesting that it is essential to study smaller-scale groups whose identity is bound up with allegiance to a specific mode of practice. While it might be possible to describe the social locations of these groups in institutional or disciplinary terms, their identity was not to be defined in such terms. Rather, their defining feature was their consolidation around a particular way of doing science. Only with this kind of analysis could one hope to show how social relations penetrate to the core of scientific practice.

Kuhn himself had noted that defining communities in terms of external institutional criteria could not help with isolating the interdisciplinary groups that come together to study particular phenomena. He gave the example of the “Phage Group” – the consortium of biochemists, microbiologists, and geneticists that pursued the study of bacterial viruses in the 1940s and 1950s (1962/1970: 177). Barnes talked about the research “sub-cultures,” which Kuhn’s work had illuminated. Kuhn had shown, he claimed:

[J]ust how profound and pervasive is the significance of the sub-culture in science, and the communal activity of the organised groups of practitioners who sustain it. The culture is far more than the setting for scientific research; it is the research itself. It is not just problems, techniques and existing findings which are culturally specific; so, too, are the modes of perceiving and conceptualising reality, the forms of inference and analogy, and the standards and precedents for judgment and evaluation which are actually employed in the course of research. (Barnes 1982: 10)

The place in Kuhn’s analysis where this came out most clearly was in his treatment of the controversies that occur in transitions between paradigms. During scientific revolutions, he claimed, the proponents of competing paradigms engage in debates that are unsatisfying and ultimately likely to prove inconclusive. It is rarely possible for upholders of the new paradigm to provide definitive proof of its superiority to those who continue to defend the old one. This is because the paradigms are themselves the sources of outstanding problems, techniques for solving them, and standards for assessing the solutions. The perceptual skills of scientists are refined to pick out the data that are meaningful for their particular paradigm. Furthermore, because commitment to a paradigm involves acquiring skills in the use of certain instruments, it may be said that adherents to different paradigms are living in different perceptual worlds. We should not expect it to be possible for them to agree on a single set of data against which both paradigms can be assessed. Instead, new paradigms yield new data which are simply outside the range that the previous paradigm would have considered meaningful.

Since, Kuhn said, “there are no standards higher than the assent of the relevant community,” there are no neutral standards, external to the paradigms, against which they can both be measured. This is a situation of “incommensurability.” “When paradigms enter, as they must, into a debate about paradigm choice, their role is necessarily circular. Each group uses its own paradigm to argue in that paradigm’s defense” (Kuhn 1962/1970: 94). In the...

Table of contents

• Cover
• Copyright
• Title Page
• Contents
• Illustrations
• New Preface (2005)
• Preface
• Introduction: Challenges to the Classical View of Science
• 1. An Outline of Constructivism
• 2. Identity and Discipline
• 3. The Place of Production
• 4. Speaking for Nature
• 5. Interventions and Representations
• 6. Culture and Construction
• Coda: The Obligations of Narrative
• Bibliography
• Index