This is not to suggest, however, that a new paradigm ultimately succeeds through some mystical aesthetic. On the contrary, very few scientists abandon a scientific tradition for that reason alone. Those who do take such an attitude often prove to have judged wrongly. But for a paradigm to triumph, it must first gain a few supporters, who must develop it to the point where firm arguments can be produced and multiplied. Yet even those arguments, when they appear, are not individually decisive. Because scientists are reasonable people, one argument after another will ultimately persuade many of them. But there is no single argument that can or should persuade them all. What actually occurs is not so much the conversion of a single group as a gradually increasing shift in the distribution of professional belief.
At the outset, a new candidate for paradigm often has few supporters, and the motives of those supporters are frequently suspect. Nevertheless, if they are competent, they will improve the paradigm, explore its possibilities, and show what a community of scientists guided by it would be like. And as this proceeds, if the paradigm is destined to win the struggle, the number and force of persuasive arguments will increase. Accordingly, more scientists will be converted, and the work of exploring the new paradigm will continue. The number of experiments, instruments, papers, and books based on that paradigm will gradually grow. As still more people, convinced of the effectiveness of the new viewpoint, adopt the new way of practicing normal science, eventually only a few older holdouts will remain. And even they cannot be said to be wrong. Though the historian of science can always find in history irrational people—Priestley, for example—who held out as long as they could, it is impossible to know at what point a certain degree of resistance becomes illogical or unscientific. At most, the historian of science may wish to say that anyone who continues to hold out after the entire profession has been converted has ipso facto refused to be a scientist.
"Notes"
1) A concise summary of the main routes to a probabilistic theory of confirmation may be found in: Ernest Nagel, Principles of the Theory of Probability, Vol. I, No. 6, of International Encyclopedia of Unified Science, pp. 60-75.
2) K. R. Popper, The Logic of Scientific Discovery (New York, 1959), especially chapters i-iv.
3) For nonspecialist reactions to the concept of curved space, see: Philipp Frank, Einstein, His Life and Times, trans. and ed. G. Rosen and S. Kusaka (New York, 1947), pp. 142-46. Several attempts to preserve the gains of general relativity within Euclidean space are found in: C. Nordmann, Einstein and the Universe, trans. J. McCabe (New York, 1922), chap. ix.
4) T. S. Kuhn, The Copernican Revolution (Cambridge, Mass., 1957), chaps. iii, iv, and vii. That heliocentrism was more than a strictly astronomical issue is one of the book’s overall themes.
5) Max Jammer, Concepts of Space (Cambridge, Mass., 1954), pp. 118-24.
6) I. B. Cohen, Franklin and Newton: An Inquiry into Speculative Newtonian Experimental Science and Franklin’s Work in Electricity as an Example Thereof (Philadelphia, 1956), pp. 93-94.
7) Charles Darwin, On the Origin of Species... (authorized edition from 6th English ed.; New York, 1889), II, 295-96.
8) Max Planck, Scientific Autobiography and Other Papers, trans. F. Gaynor (New York, 1949), pp. 33-34.
9) On the role played by sun-worship in Kepler’s thought, see: E. A.
Burtt, The Metaphysical Foundations of Modern Physical Science (rev. ed.; New York, 1932), pp. 44-49.
10) On the role of reputation, consider the following: Lord Rayleigh, at a time when his reputation was firmly established, submitted to the British Association a paper on several paradoxes in electrodynamics. When the paper was first submitted, his name had accidentally been omitted, and it was initially rejected as the work of some “paradoxer.” Not long afterward, once the author’s name was inserted, the paper was gladly accepted with profuse apologies. [R. J.
Strutt, 4th Baron Rayleigh, John William Strutt, Third Baron Rayleigh (New York, 1924), p. 228].
11) On the problems raised by quantum theory, see F. Reiche, The Quantum Theory (London, 1922), chaps. ii, vi-ix. For the other examples in this paragraph, see the notes at the beginning of Section XII.
12) Kuhn, op. cit., pp. 219-25.
13) E. T. Whittaker, A History of the Theories of Aether and Electricity, I (2d ed.; London, 1951), p. 108.
14) On the development of the general theory of relativity, see: ibid., II (1953), 151-80. Einstein’s reaction when the theory exactly matched observations of the motion of Mercury’s perihelion appears in a letter quoted in: P. A. Schilpp (ed.), Albert Einstein, Philosopher-Scientist (Evanston, Ill., 1949), p. 101.
15) On Brahe’s system, which was geometrically entirely identical with Copernicus’s, see: J. L. E. Dreyer, A History of Astronomy from Thales to Kepler (2d ed.; New York, 1953), pp. 359-71. On the final version of the phlogiston theory and its success, see: J. R.
Partington and D. Micke, “Historical Studies of the Phlogiston Theory,” Annals of Science, IV (1937), 113-49.
16) On the problem posed by hydrogen, see J. R. Partington, A Short History of Chemistry (2d ed.; London, 1951), p. 134. On carbon monoxide, see H. Kopp, Geschichte der Chemie, III (Braunschweig, 1845), 294-96.
XIII. Progress through Revolutions
Progress through Revolutions
The preceding sections have provided, within the limits that this essay can address, my own schematic description of scientific development. Even so, it does not quite provide a conclusion. If this account captures the essential structure of the continuing development of science, then at the same time it raises a distinctive problem. Why does the scientific activity described above move steadily forward in a way different from, for example, the transformation of art, political theory, or philosophy? Why is progress a special condition reserved almost exclusively for the activities we call science? The most common answer to this question has been denied in the body of this essay. We must conclude by asking whether alternatives can be found.
It is necessary to note that part of this question is entirely semantic. In almost every case, the term “science” is used only for fields in which progress occurs in a definite way. This characteristic appears most clearly in the recurring debates over whether one or another of the modern social sciences is truly a science. These debates have their parallels in the pre-paradigm periods of the fields now unhesitatingly classified as sciences. Throughout, their ostensible subject is the definition of that troublesome term. Some argue, for example, that psychology is a science because it possesses such and such characteristics. Others reply that those characteristics are either unnecessary or insufficient to establish a field as a science. Such disputes often involve tremendous effort and arouse heated passion, leaving an outside observer bewildered as to the reason. Is the definition of “science” so very important? Can that definition tell someone whether or not he is a scientist? If so, why do natural scientists or artists not worry about defining the term? One cannot help thinking that the issue is more fundamental. Perhaps the questions actually being raised are these: Why does my field fail to advance in the same way as, say, physics? What changes in technique, method, or ideology would make it so progressive? But these are not questions that can be answered by agreement on a definition. Moreover, if precedents from the natural sciences apply, they will cease to be a source of concern not when a definition is found, but when the groups that now doubt their own status reach agreement about their past and present achievements. It is significant, for example, that economists debate less than scholars in many other fields of the social sciences whether their field is a science. Is that because economists know what science is?
Or is it because what they have reached agreement on is economics?
Though it is no longer merely a matter of semantics, this point of view is a transitional thesis that helps reveal the entangled relationship between science and the idea of progress. For centuries, both in antiquity and in early modern Europe, painting was certainly regarded as a field that underwent cumulative development. During that period, the painter’s aim was thought to lie in representation. Critics and historians such as Pliny and Vasari recorded their homage to a series of inventions, from perspective through chiaroscuro, that made more perfect representations of nature possible. 2) Yet it was also in that period, especially during the years of the Renaissance, that a small gap between science and art began to be felt. Leonardo da Vinci was one of many who moved freely among fields that would only later be clearly distinguished as separate categories.2) Moreover, even after such steady interchange had ceased, the term “art” continued to be applied not only to painting and sculpture but also to techniques and crafts that were likewise considered progressive. When painting and sculpture explicitly repudiated representation as their goal and began to learn again from primitive models, the gap we now take for granted deepened to its present depth. And even today, shifting fields once again, our difficulty in seeing the profound difference between science and technology must be related in part to the fact that progress is a conspicuous attribute of both.
But developing
Recognizing the tendency to regard any field as science only highlights our present problem; it does not solve it. What remains now is to understand why progress becomes such a marked characteristic of scientific activity carried on with the techniques and aims described in this essay. This question contains many questions within it, and each of them must be examined separately here. In all cases but the last, however, their solution will depend in part on a reversal of our standard view of the relation between scientific activity and the community of scientists that carries it out. We must learn to recognize as a cause what has commonly been regarded as an effect. If we can do so, the phrases “scientific progress” and even “scientific objectivity” will appear partly redundant. Indeed, one aspect of that redundancy has just been presented. Does a field make progress because it is science, or is it science because it makes progress?
Let us now ask why an activity like normal science should make progress
and begin the discussion by recalling several of the most striking characteristics of normal science. Usually, the members of a mature scientific community carry out research from a single paradigm, or from a set of closely related paradigms. It is extremely rare for different scientific communities to examine the same problems. In such exceptional cases, the groups will share several major paradigms. But from the standpoint of any single scientific community, whether scientists or non-scientists, the result of successful creative work must precisely be progress. We have seen how it records progress. Other creative fields also show progress of this same kind. A theologian expounding doctrine, or a philosopher discussing Kantian imperatives, contributes to progress, if only for the group that shares his premises.
No creative school recognizes a category of activity that is, on the one hand, a creative success and, on the other, not an addition to the total achievement of the group. If, as many do, we doubt that non-scientific fields make progress, it is because schools exist; the person who claims that there has been no progress in each of the differing schools is emphasizing not that Aristotelianism failed to progress, but that Aristotelians still remain.
But these doubts about progress arise in science as well. Throughout the pre-paradigm period, when many competing schools exist, evidence of progress is very hard to find except within a school. During the period described in Section II, when individuals practice science, the results of their research activity do not, as we know it, add to science. Also, during periods of revolution, when the fundamental doctrines of a field once again become matters of dispute, doubts are repeatedly expressed as to whether continued progress will be possible if one or another of the opposed paradigms is adopted.
Those who rejected Newtonianism argued that, by relying on innate forces inherent in matter, the theory would return science to the darkness of the Middle Ages.
Those who opposed Lavoisier’s chemistry saw the rejection of the concept of the chemical “element” in favor of laboratory elements as a rejection of chemical explanation by those seeking refuge in nominalism.
Though expressed more obliquely, a similar sentiment also seems to have underlain the opposition of Einstein, Bohm, and others to the dominant probabilistic interpretation of quantum mechanics. In short, it is only during periods of normal science that progress appears obvious and at the same time certain.
During those periods, however, the scientific community can view the fruits of its research in no other way.
Then, in the context of normal science, part of the answer to the problem of progress depends simply on the observer’s perspective. Scientific development is not of a different kind from development in many other fields, but the absence, during most periods, of competing schools that question one another’s goals and standards makes it easier to see the progress of a normal-scientific community. Yet this is only part of the answer, and by no means the most important part. For example, we saw earlier that once the acceptance of a common paradigm frees a scientific community from the need constantly to reexamine its first principles, the members of that community can concentrate wholly on the most subtle and esoteric aspects of the phenomena that interest it. Inevitably, this increases the group’s overall effectiveness and efficiency in solving new problems. Moreover, several characteristics of professional activity in science further enhance this very special efficiency.
Some of these characteristics arise from the unprecedented isolation of a mature scientific community
from laymen and from the demands of everyday life. Such insulation has never been complete—we are now discussing degrees of insulation. Nevertheless, there is no other professional community in which an individual’s creative work is announced so exclusively to the members of that profession and evaluated only by them. Even the most difficult poet or the most abstract theologian, though he may care less about applause, will be far more concerned than the scientist with public recognition of his creative work. And this difference proves to be necessary. Because the scientist works only for his colleagues, an audience that shares his own values and beliefs, he can take a single set of standards for granted. He need not worry about what other groups or schools will think, and therefore, after dealing with one problem, he can move on to the next more quickly than those working within a more heterogeneous group. More important than this, the insulation of the scientific community from society at large permits the individual scientist to focus his attention on problems that he has ample reason to believe can be solved by him. Unlike the engineer, many physicians, and most theologians, the scientist need not choose problems because their solution is urgently demanded, without regard to the means required to solve them.
From this point of view, the difference between natural scientists and many social scientists also proves highly suggestive. Whereas natural scientists almost never do so, social scientists often tend to justify their choice of research problems chiefly in terms of how socially important it is to secure a solution—for example, in problems such as the consequences of racial discrimination or the causes of the business cycle. Which group, then, might be expected to solve problems at a faster rate?
The effects of insulation from the larger society are greatly reinforced by another characteristic of the professional scientific community: the nature of an education that transmits its secrets. In music, painting, literature, and the like, one learns by encountering the works of other artists, especially those who came before.
Except for compendia or handbooks of original creations, textbooks play only a secondary role. In history, philosophy, and the social sciences, textbook literature has greater significance. Yet even in these fields, introductory university courses read original sources alongside them, some of which are the “classics” of the field and the rest contemporary research reports written by scholars for one another. As a result, the student in these fields remains continually aware of the exceedingly diverse problems that the members of his future group will attempt to solve over time.
More important, he is confronted with competing and incommensurable solutions to these problems—solutions that, in the end, he himself must evaluate.
Contrast this situation, at least,
with the situation in the modern natural sciences. The student in these fields of nature relies chiefly on textbooks until he begins independent research in the third or fourth year of graduate study. Many scientific curricula do not require even graduate students to read works not written for students. Even when research papers and specialized monographs are assigned as supplementary reading, such assignments are confined to the most advanced courses and are limited to materials that more or less supplement portions not found in the textbooks used. As one reaches the final stages of a scientist’s education, textbooks are systematically replaced by the original scientific literature that made the textbooks possible. Given their confidence in the paradigms that make this educational technique possible, few scientists would wish to change it.
Why, after all, should one read the research reports of Newton, Faraday, Einstein, or Schrodinger, when everything one needs to know about those works has been summarized in more concise, accurate, and systematic form in recent textbooks?
That this form of education is very
Rather than justify what has been carried on for so long, we cannot help but note that this method has, on the whole, been enormously effective. Of course, it is a narrow and rigid education—perhaps more so than in any field except orthodox theology. Yet for normal-scientific research, scientists are almost perfectly prepared. Moreover, it is also well suited to another task—the formation of significant crises through normal science. When crises arise, of course, the scientist is not so well prepared. Even if widespread crises may be reflected in less rigid educational practice, scientific training is not readily designed to produce people who will discover new approaches. But so long as someone—usually a young scholar or someone new to the field—comes forward with a new candidate for a paradigm, the losses caused by rigidity occur only at the individual level. By the time the generation affected by the change is reached, individual rigidity is compatible with a group capable of shifting away from a paradigm when circumstances demand it. This is especially so when that very rigidity sends the scientific community a sensitive signal that something has gone wrong.
Thus, in its normal state, the scientific community becomes an extremely efficient instrument for solving the problems or puzzles defined by its paradigm. Moreover, the result of solving those problems can only be progress. There is no difficulty here. But understanding this much merely brings into relief the second major part of the problem of scientific progress. Let us therefore turn now and ask about progress through extraordinary science. Why must progress also be an apparently universal accompaniment of scientific revolutions? Here again, the matter will become clear by asking what else the outcome of a scientific revolution could be. A revolution ends with the total victory of one of two opposing camps. Could the group say that the result of its victory was anything less than progress? To do so would be tantamount to admitting that they were wrong and their opponents were right. At least for them, the result of the revolution must be progress, and they are in a favorable position to ensure that future members of their community will view past history in the same way. Section XI described in detail the techniques by which that is achieved, and we have now returned to a characteristic closely connected with professional scientific activity. When a scientific community rejects a past paradigm, it simultaneously rejects most of the books and papers in which that paradigm is embodied as suitable subjects for professional research. Scientific education makes no use of anything corresponding to an art museum or a library of classics, and the result may appear as a dramatic distortion in the scientist’s perception of the past of his field. More than practitioners in other creative fields, the scientist comes to see the outcome as having led his field straight to its present advantageous position, and he sees that outcome as progress. So long as the scientist remains in the field, there can be no other alternative for him.
These remarks inevitably suggest that a member of a mature scientific community, like the typical figure in Orwell’s “1984,” is the victim of a history rewritten by the powers existing in that society. Moreover, this suggestion is not entirely unjustified. In scientific revolutions, there are losses as well as gains, and scientists tend to be particularly blind to the losses.3) On the other hand, however, no account of progress through revolutions can stop at this point. To do so would be to present a formulation that, if it did not obscure the nature of authority and the process by which choices between paradigms are determined, would not be entirely wrong: namely, that in the sciences might makes right. If authority, and especially nonprofessional authority, served as the arbiter of paradigm debates, the outcome of those debates might be a revolution, but it would not be a scientific revolution. The very meaning of science’s existence depends on granting the power to choose between paradigms to the members of a community of a particular kind. How special that community must be for science to survive and grow can be seen from humanity’s lack of understanding of scientific activity. Every civilization for which records remain possessed technology, art, religion, political systems, laws, and the like. And in ancient civilizations these domains were as developed as they are in our own. But only the civilization transmitted from Greece possessed anything more than the most rudimentary science. Most scientific knowledge has been the product of Europe during the last four centuries. Other regions and other eras failed to sustain the sort of special scientific community in which scientific productive activity appears.