The late Alfred North Whitehead is said to have once presided at a lecture by Bertrand Russell on the philosophical implications of what was then the new quantum mechanics, and to have closed the meeting by thanking Russell for his presentation and especially for “leaving the vast darkness of the subject unobscured.” The more than five-score contributors to the volume under review—a good number of them historians of science and technology, living in the United States and various European countries including Russia—deserve similar thanks. The book contains the proceedings of a week-long symposium, held at Oxford in July 1961, on “the intellectual, social and technical conditions for scientific discovery and technical invention from antiquity to the present.” Of the twenty-eight papers invited, about a third deal with the relations between scientific achievements and their socio-cultural environment, another third examine the internal development and the content of various scientific ideas and methods, while the remaining ones discuss questions bearing on the nature and the pursuit of the history of science as an intellectual discipline. However, though the learning most of the participants exhibit is impressive and often illuminating, only a few of them address themselves explicitly to the nominal theme of the symposium. There will doubtless be disagreement on whether the symposiasts have advanced our understanding of scientific change much beyond what was already known; but no one can seriously dispute that they succeeded in leaving the darkness of the subject unobscured.

Indeed, since its scope was not clearly limited, it is remarkable that the symposium succeeded in achieving anything else. The proceedings nowhere make plain just what is covered by the word “science.” However, the wide assortment of matters that are discussed suggests that perhaps the majority of the participants take the word to mean almost any continuing quest for reliable knowledge or for practical mastery of nature. The use of the word in this inclusive sense has obvious merits, but it also creates difficulties for a coherent discussion of scientific change. For in this sense, science has been pursued in diverse societies and within different intellectual traditions—and indeed by individuals whose personal motives for doing scientific work have generally been neither uniform nor unmixed. Moreover, the innumerable inquiries that make up the scientific enterprise were undertaken to resolve different sorts of specific problems—problems created by difficulties in prevailing beliefs, by the observation of new phenomena and the outcome of previous inquiries, or by the needs and opportunities of daily living. Accordingly, when science is viewed as a perennial and widespread search—whether for theoretical understanding or practical mastery—it is a truism to say that it undergoes various types of change. But it is also evident that the numerous enterprises labelled as “scientific” are often quite dissimilar in objectives, achievements, and intellectual methods. Merely because a variety of enterprises may be called “scientific” it should not be assumed that they are necessarily alike in respect to other significant features.

Unfortunately this assumption has often been made by historians of science and it has led them to make dubious claims. An example of this is the generally fascinating account by Joseph Needham of Cambridge of the remarkable advances in knowledge that were achieved in ancient and medieval China. In addition to presenting these achievements, his paper has two further aims: to explain why, between the second century B.C. and the fifteenth century A.D., China was much more efficient than the European West in applying knowledge of nature to useful purposes; and to challenge the widely held belief that science has been able to flourish only under the influence of Greek thought.

He attempts to show that the practical bent of Chinese science was intimately connected with the fact that water conservation and irrigation were of such great importance to the Chinese economy that their study had an official status. In addition, the orientation of Chinese philosophy, unlike European philosophical thought, was essentially secular, and was therefore favorable to the pursuit of useful knowledge. Needham maintains that despite their general distrust of speculative theory, the Chinese employed prototypes of wave and field theories in the study of optical, acoustic, and magnetic phenomena. The existence of vast quantities of practical knowledge in ancient and medieval China appears to be undeniable. But Needham’s claim that the Chinese developed anything even remotely resembling wave and field theories, or, more generally, that they achieved theoretical science in the Western sense, is seriously disputed by other sinologists. Indeed, Needham’s account strongly suggests that the knowledge possessed by the Chinese, while impressive in its range, was the product of a relatively crude trial-and-error empiricism, which is considerably different from the Western conception of science. Moreover, the close dependence of research upon the feudal bureaucracy of China, as well as the traditionalism of Chinese science (features that Needham himself stresses), argue against the supposition that relatively unhampered theoretical speculation—an indispensable ingredient in European science—could have flourished in China.


Even Needham’s claim that medieval China had a more efficient technology than did medieval Europe is rejected by Lynn White in an interesting paper about the accelerated technological progress in the Western Middle Ages. White also disagrees with another of Needham’s contentions, that Christian theology was unfavorable to the development of useful secular knowledge. On the assumption that animistic beliefs are an obstacle to a deliberate exploitation of nature, he advances the seemingly paradoxical hypothesis that the shift from animism to the cult of saints during the Western Middle Ages, by removing this obstacle, contributed to the progress of technology. However, the warrant for any of these claims, counterclaims, and assumptions is obscure. Key terms in them (e.g., “more efficient” or “unfavorable”) are so vague that it is uncertain just what can count and what cannot count as supporting evidence; and even if the meanings of the assertions were clarified, the further difficulty remains that the available evidence for them is at best fragmentary and inconclusive.

The same must be said for much of the discussion of the aetiology of scientific change in this symposium. Interest in this subject goes back to antiquity, and numerous explanations have been proposed to account for it. It is sometime argued that some single “factor” or condition accounts for it, sometimes a plurality of factors, sometimes ideational factors “internal” to a science, sometimes “external” or psychological and socio-economic factors. It is frequently unclear, however, whether these factors are supposed to explain the genesis of scientific ideas and discoveries, or whether they are intended to account for the acceptance of such innovations, whether by a single individual, a community of scientists, or society at large. Since relatively little is known about the conditions under which creative thought occurs, failure to recognize this distinction results in dubious if not inane historical analysis. Some of the so-called “single-factor” explanations of scientific change are notorious for confounding it; this is true, for example, of the view that scientific achievements are the products of genius, or of the equally familiar idea that the direction of scientific change is determined by economic needs. Moreover, even when the distinction is recognized, such explanations do not survive critical scrutiny.

In general, the contributions in this book do not subscribe to these single-factor views, but there are some exceptions. In a comparative study of ancient Babylonian and Greek astronomy, for example, the distinguished Swiss mathematician B.L. van der Waerden points out that while the techniques of analysis uniformly employed by the Babylonians were purely arithmetical, those used by the Greeks were predominantly geometrical and involved mechanical analogies. His explanation for this striking difference is that unlike the Babylonians “the Greeks were born geometers and engineers.” As it happens, however, the Greeks did not entirely lack outstanding arithmeticians—a circumstance which undoubtedly can also be credited to Greek genius, but which still raises the question why, granted their many-sided talents, their astronomical methods were geometrical rather than arithmetical. This question is not resolved by the postulation of specific faculties to explain dissimilarities in the scientific ideas of different ethnic groups. Although the Russian members of the symposium rejected Van der Waerden’s single-factor explanation, they advocated one of their own. Their contributions supply much valuable information on Russian science and technology during the past three centuries. But their papers also quote with approval Engels’ silly claim that electricity was understood only after its applicability had been discovered; and they try with boundless zeal to account for scientific development in Russia and elsewhere exclusively in terms of economic needs.

It is nonetheless clear from several other papers in the symposium that even the development of industrial technology is not completely determined by economic considerations. Thus, Sir Eric Ashby calls attention to the fact that in nineteenth-century England it was the market for trained industrial chemists that needed stimulation, rather then the actual training of such men; and one reason for this fact is the persistence of certain British traditions in education as well as in industry. In this connection it is often overlooked that until the latter part of the nineteenth century, industrial technology was largely the work of inventors who had little knowledge of theoretical science, and not the fruit of systematically applied basic science. Thus, developments in physical and chemical theory appeared for a long time to have no relevance to developments in industry. On the other hand, although industrial technology today is intimately dependent on applied science, it must not be assumed that the direction of applied research is completely controlled by economic needs. For as H. J. Habakkuk of Oxford points out, current applied science is in considerable measure the exploitation of technical ideas derived from fundamental theory, so that it is becoming “less responsive to market forces, and now forges ahead on its own steam.”


The contributions to the symposium that are concerned primarily with the internal conditions of scientific change cover a wide range of questions related to the evolution of modern science, including such matters as the interrelations between advances in theory and quantitative techniques, and the influence of various ideas on the often meandering growth of special branches of science. These papers are in general informative but uneven in quality. For example, an essay by Bentley Glass is a revealing account of the role and development of theoretical models in genetics—his analysis of how current genetic theory combines and reconciles preformationist and epigenetic notions is particularly illuminating. On the other hand, some of the contributions dealing with the influence of ideas illustrate the difficulties in establishing causal imputations in intellectual history.

C. C. Gillispie of Princeton claims that the development of an important branch of physics was influenced by ideas first broached in the social sciences. According to him, Quetelet gave a statistical interpretation of the Laplacian theory of probability theory in the use he made of probability in his proposals for a quantitative social science, thereby opening the way to a statistical (or probabilistic) description of physical nature as well. Subsequently, Maxwell employed the theory in a similar way in his seminal essay on statistical mecnanics; and Gillispie believes it is therefore likely that Maxwell’s work carries the impress of Quetelet’s ideas.

But unfortunately for this interesting conjecture, it is almost certain that Quetelet’s conception of probability was essentially the same as Laplace’s, and hence quite different from Maxwell’s views on probabilistic laws of nature. Moreover, Maxwell’s work on statistical mechanics falls into a tradition of theoretical analysis in physics that was well established when he wrote his paper. If there was indeed a “migration of probabilities from social studies to physics,” the available evidence does not confirm that it really took place.

For reasons that can only be surmised, the symposium includes no discussion of the conditions underlying change in the social sciences—although one paper is devoted to such problems in medicine, two in psychology, and one to a question in the logic of scientific research (but strangely classified as a problem in “the sociology of science.”) An essay by Owsei Temkin is an absorbing study of how the “ontological” notion of disease as a specific entity, and the “physiological” view that disease is simply a “state” of individual organisms, have alternated in the history of medicine. One paper on psychology by R. C. Oldfield provides a suggestive but much too brief analysis of changing conceptions of physiological mechanisms in human behavior. The second one by B. A. Farrell is not strictly on the aetiology of change, but is an attempt to compare so-called “objective” (or experimental) psychology with clinical (for the most part Freudian) psychology in respect to the character of their methods. Farrell thinks that the methods of objective psychology are rigorously scientific, even though its actual achievements have thus far been disappointing, while the methods and conclusions of psychoanalysis, which has transformed our view of human nature, have an uncertain validity. Nevertheless he denies that there is a sharp line separating methods which are genuinely scientific from those which are not—although he grants that magic, for example, is not a science. He claims that psychoanalysis is closer to science than magic; and he therefore believes that “it is inappropriate and misleading for the objective psychologist simply to maintain that psychoanalysis is an unscientific enterprise,” but that it is also quite inappropriate and misleading for the psychoanalyst to maintain the opposite.

Farrell’s argument is a puzzling one. In view of his own judgment that in its methods and theory psychoanalysis has only doubtful validity—on this question Farrell has written illuminatingly elsewhere—it is not clear why it should be misleading to assert that Freudian psychology does not qualify as a scientific enterprise. His case depends on the fact that the term “scientific method” is a vague one, but this does not really support it. There is no sharp demarcation between, say, the front and the back of a man’s head, or between animate and inanimate systems. But it does not follow that there are no differences between front and back or between living and non-living things, or that it may not be highly important for certain purposes to distinguish between them.

T.S. Kuhn’s “The Function of Dogma in Scientific Research” is undoubtedly the most vigorously written essay in the symposium and the only invited paper with pronounced philosophical overtones. Kuhn challenges the popular image of the scientist as an open-minded searcher after truth. According to Kuhn, the practitioners of a mature science are deeply committed to some “dogma” or “paradigm” (i.e, to a particular way of viewing and investigating nature). Scientific research is therefore normally directed to fitting various phenomena into the framework of ideas constituting the paradigm; and every effort is made to hold on to the paradigm or dogma, despite apparent incompatibilities between paradigm-based expectations and observation. Nevertheless, it may not be possible to take care of such anomalies by patching up the paradigm; and a new period of “normal” science, initiated by the adoption of a new paradigm, may come into being. The pattern of scientific development, as Kuhn formulates it, is therefore from paradigm to paradigm—a pattern in which commitment to a paradigm is a source of resistance to new ideas, but is also instrumental in making science the most revolutionary of all human activities. But except for his paradoxical language, most of what Kuhn says in the present paper is familiar. His defense of dogma in scientific research is little more than an implicit criticism of the widespread notion, often but mistakenly attributed to Francis Bacon, that brains unfurnished with ideas are of prime value in science, and that scientific inquiry consists in gathering “facts” without guidance from previously acquired convictions and hypotheses. He does not make clear, however, either in his present essay or in his other publications, just what is to be understood by a “paradigm”; and he exaggerates the discontinuities between successive “normal” periods of science, as well as the hold that paradigms generally have upon the entire community of scientists. In particular, the resistance to new ideas which he ascribes to scientists because they do not abandon a theory as soon as it runs into difficulties but try to patch it up instead, is not necessarily a sign of conservatism and commitment to some “dogma”; such efforts to save a theory are often instances of the sensible policy to reap whatever fruits even a partially satisfactory theory can yield until a better theory is devised.

But enough has been said to suggest that the Oxford symposium provides mixed and uneven fare, and to indicate that historians of science approach problems of scientific change in different ways and with dissimilar equipment. It is obvious that many of these problems can be adequately handled only by men with high competence in several domains of learning: in historical scholarship, in various branches of substantive science including social inquiry, and in the philosophical analysis of ideas and intellectual methods. But it is also evident from this symposium that those who attempt to handle such problems possess the needed competence in unequal measures. Indeed, although the history of science has recently acquired the status of a professional discipline, conceptions of the nature and aims of the subject differ widely, and there is similarly considerable variations in the kind of training that is regarded as desirable for a professional career in it. Such a plurality of conceptions is doubtless all to the good; and the questions still to be investigated concerning the development of science are certainly diverse enough not to require uniformity in the training of those who pursue them. On the other hand, large claims have been made, and continue to be made, for the value of the history of science as a subject of academic instruction. For example, the late George Sarton urged that it be taught regularly because it forms a natural bridge between the sciences and the humanities, and can therefore serve to heal the long-standing breach between “the two cultures.” And A. C. Crombie, the editor of the present volume, recommends the inclusion of the history and philosophy of science among academic studies on the ground that it can “give to the student—whether his main interest is historical, philosophical, or scientific—an informed and critical awareness of the scientific tradition that has grown up in our midst, and of its numerous links with many other aspects of our thinking and conditions of life. For scientists it has the more intimate purpose of providing an approach to the critical examination of what they are themselves doing.” These are admirable objectives, eminently worth being pursued with whole-hearted devotion. However, they are expressions of aims and hopes, not descriptions of achievements; and the problem of how to bell the cat still remains. In the light of the contributions most historians actually make to their subject, and the training most of them receive in preparing themselves for their profession, considerable skepticism is justified as to whether those high aims are likely to be soon transformed into realities.

This Issue

February 6, 1964