RATIONALITY IS A MEANS TO AN END

Theories of knowledge and theories of rationality since antiquity centered on one major problem. It is the problem of the foundation of knowledge and of rationality. In Indian geography, you remember, the earth was made to rest on an elephant, the elephant on a tortoise, and so on; yet the suspicion remained that below it all was a fathomless abyss.

What is common to both new schools of thought in the post-Einstein era, the pro-science and the anti-science schools, is the idea that if science is totally unstable, then it is intellectually of little value. This is why the one school sees it as intellectually valuable and near-certain or highly probable, and the other as highly alterable and thus as intellectually valueless. The premiss common to both - that if science is in flux, then it is intellectually valueless - was contradicted by Popper, who considers the proposal of theories and the successful criticism of them both as the method of science and as the method of rational thinking - of intellectual development and intellectual achievement. Though he insists that the aim of science is truth, he has given up all hope or aspiration for certitude or high probability or any other stabilizing factor.

To avoid misunderstanding let us stress that the word ‘criticism’ is here used in its commonest and broadest sense in which opinions can be criticized - whether in every-day use or scientific use.

Popper has initially thought that all criticism is either empirical or logical. This led him to view moral proposals and metaphysical and theo­logical statements beyond the realm of rational debate. This was much better than the positivistic view of such statements as meaningless, but still not good enough, as Popper himself now stresses. His development came through his study of the ancient problem of rationality, and the contemporary irrationalist critique of classical rationalism. He claimed in his Open Society that indeed rationality has an irrational basis, but that it is most rational to minimize that basis. The minimum, he claims, is the Socratic assumption that we can learn from our mistakes. Once this idea was assumed, its very austerity ousted the idea that criticism in morality and metaphysics is impossible, and in the English edition of his Logic of Scientific Discovery of 1959 Popper makes this correction.

This is a nice case of how austerity of assumption may make one more broad-minded, contrary to what may perhaps be intuitively expected. Moreover, quite apart from whether we should be open-minded or not, we may find it very exciting to view science as a special case of Socratic dialogue, to wit the case of Socratic dialogue in which the criticism is empirical. But the theory of Socratic dialogue as presented in Popper’s Open Society (Chapter 24) is in need of modification, since in order to be effective, criticism must be selective and discriminating; and since in order to be properly selective and discriminating, one needs some criteria for selectivity and for discrimination.

Consider, again, scholasticism. Doubtlessly, it was critical towards its own minor assumptions and defended its major ones by any amount of auxiliary hypotheses. Here, then, Popper’s proposal to step up the degree of testability of the assumptions under consideration is justified by the desire to avoid stagnation. It is far from obvious, however, that in all cases Popper’s proposal is equally welcome. One may imagine cases, and even look for historical cases, where it was more profitable to consider a less testable but more promising ideas in terms of offering stimulus to thought or a unifying theme to diverse scientific problems, a kind of world-hypothesis or a metaphysics.

The modification here proposed solves other problems as well. Assuming that rationality is solving certain problems as best possible and subjecting these solutions to critical debate, one may characterize science as a proper special case of rationality, while using the following two criteria. First, the problems of science are related to explanations of facts, or more generally, to comprehension of the world. Second, the criticism of science is empirical. Popper himself has characterized science by the second criterion alone. Namely of empirical criticizability. He thought, I understand, that when we take care of the second criterion, that will also take care of the first: he thought, in other words, that good ex­planations are well open to empirical criticism, and vice versa. On this he is in error, but we can easily see that on the correction of Popper’s view of rationality and on holding to his view of science as a special case of rationality, this error gets eliminated rather naturally.

To show, again, that the modification is not vacuous, let us consider examples. Ambroise Pare refuted many medical superstitions of his day (such as the alleged healing effect of consuming bits of Egyptian mum­mies). It was indeed because many superstitions were refuted in the Renaissance that refutability came to be considered a vice rather than a virtue. Compare this with the - admittedly regrettable - irrefutability (or seeming irrefutability) of Einstein’s unified field theory or of Faraday’s adumbration of it!

Popper likes to liken scientific refutability with the ability of democ­racies to overthrow their governments with relative ease - which he even elevates to a definition of democracy.

The interesting task, now, is to follow through Popper’s idea of science again as a special case of rationality. It may turn out that this idea presents science as being much more in a state of flux than it really is, that the existing consolidating factors in science must be either erroneously overlooked by that theory or taken account of by adding to it extra assumptions, or by making some additional rules, perhaps, in order to render Socratic dialectics into the dialectics of science. If it turns out that one may explain the existing stability of science only by additional hypotheses, one may well prefer to condemn the stability than to add the auxiliary hypotheses. The question then will be: How shall we decide between these two, and can there be a rational method of arbitration? For my part, I think some of the stability may be explained by reference to the aim of science and our inability to execute them fast enough, some of the stability may be explained as the unwelcome intrusion of dogmatism and pomposity into science. The latter case - of intrusion - is very useful to study and experiment with, but it is not terribly interesting, except in that Popper’s theory does have practical reformist consequences. The idea that much of the stability is due to certain human sluggishness turns out to be more interesting, I think.

Assuming, for instance, that the faith in positive evidence is some residue from religion, we may try to explain, it rather than worship it. Different cases of positive evidence require different kinds of explanations, perhaps, or at least allow for different kinds of explanation. Einstein explained the positive evidence of Newton’s theory of gravity by his theory of gravity, plus claims such as that the field of the sun’s gravity is not too strong, etc. Doubtlessly, the positive evidence is thus explained partly as luck, partly as coming near enough to the truth. If Einstein’s theory turns out not to be the truth we may, indeed, feel the need for a different explanation of the positive evidence in question. One way or another, we can see that the more imaginative we are the more we may design new ways of viewing positive evidence. No doubt, the positive evidence concerning the success of different theories is even today radically different; the evidence for Rutherford’s theory of the atom, for instance, is often viewed as a kind of a fluke. How we explain positive evidence, then, much depends on our theory of the world in general.

The aim of science, to be sure, is the true explanation of phenomena - but only partly so. It does not attempt to cover all the phenomena at once, but it does attempt to provide the true world-view, the true meta­physical blueprint of the universe. Science, accordingly, not only proposes explanations to be examined, but also starts with metaphysical theories within which to incorporate scientific theories in manners which can be critically examined as well. The metaphysical frameworks may be criticized and rejected; they also provide some stabilizing elements as they are less easily criticizable than scientific theories and thus less ephemeral. They are less ephemeral because of our lack of ability to render them testable, at least not so quickly. Hence the stability is due to some defects! Similarly, we do not have to accept empirical evidence as authoritative. The aim of science is to have empirical information explained, but we may explain information as false, reinterprete infor­mation as merely approximation to the facts in order to render it con­sistent with our theories; and so on - as long as we are not too apologetic in doing so, that is to say as long as we present our views because they may be interesting, and as long as we remain open to criticism, there is nothing wrong in our stubborn refusal to toe the line. But this is easier said than done. In particular, we may always reinterpret corroborating evidence in a manner which will render it more as criticism than as corroboration, but the effort is not small! How much we are able to reinterpret and review depends on our imagination and ingenuity.

To take an example from a current situation, many dislike general relativity and they try to offer alternatives to that theory, which should, explain at least all the facts which that theory explains; as that theory is corroborated, their task is not so easy, and general relativity still lacks serious competition. To show how differently corroboration may look relative to other ends, take corroborations in technology, where positive evidence has to be acquired by a public organization prior to the imple­mentation of a technical innovation, such as civil jet-plane service. The positive evidence does not prove the organization’s decision to be correct; it does, however, prove it to be responsible. That in spite of all positive evidence in favour of a new technique, that technique may be disastrous is well known from quite a few dramatic cases. There is no doubt, also, that severe testing does eliminate some of these techniques before they even enter the market. Assuming, then, that the end of testing in technology is to eliminate some undesirable technical in­novations, the value of corroboration is of evidence of having done one’s fair share in avoiding catastrophe. How different is the corrobora­tion of some new aerodynamic formula in an airplane plant from the positive evidence in favour of general relativity is intuitively obvious; the explanation of this difference, I suggest, is in terms of the different cases of rationality involved, namely, the different purposes at hand.

Here I have touched upon the topic of the social stability of science as applied science or technology. There are other stabilizing factors in the social setting of science, such as the fact that school-teachers cannot be up-to-date in science but must follow certain requirements of school­boards. The requirements of school-boards, the requirements for in­dustrial standards of severity of tests, etc., are set by social conventions. And though social conventions are indeed alterable, they must be stable to some extent. But this is no reason to discourage fluidity in scientific research: our lack of imagination is a sufficiently strong factor in pre­venting high speed. Indeed, when we look at the geniuses of science, at the enormity of their ideas, at the fruitfulness and ease with which they can turn out new theories and new criticisms, we may well see that science would be very different if we all were such geniuses. Philosophy, said William Gilbert, is but for the few. This is no longer true, and Gilbert would have enjoyed seeing the thousands of scientists who live today. They are not all Gilberts, to be sure, and their work is less revolutionary than his. But even small thinkers can try to appreciate and to criticize and to imitate big thinkers, and there is no need to fear that if all scientists were Einsteins, science may become chaotic. Perhaps there would be chaos in the arts, too, if we had many Beethovens or Cezannes; but we do not preach stability in the arts. If and when we shall have too many Einsteins, or Beethovens, we may tackle some problems that this may give rise to. Meanwhile, we can well try to appreciate the revolutionary effect of some scientific geniuses. I have chosen to conclude with one very brief example, from the work of Faraday.

Michael Faraday, who interpreted Newtonian mechanics as an ap­proximation to a future gravitational field theory, admitted that he was rejecting the theory which had been as well corroborated, and as im­pressively as could be. He rejected it on metaphysical grounds and he argued very rationally about it. He invented a few ideas about gravitation­al fields and refuted them all. He accepted the refutations but not as refutations of his metaphysics. At least not at the time. He was a worship­per of facts, yet he was able to dismiss some information as false by his very fact-worshipping, because he declared that these reports were interpreted and he accepted as phenomenological only such reports as were neutral between his opponents’ views and his. One might object that such behaviour is dogmatic; Faraday must have worried about this objection, since he answered it: he said his opponents could not rein­terpret his evidence against them half as well as he could do this for them, and even he could not do this for them half as well as he could for himself. Finally, then, the advantage of his field-metaphysics was that it was the most imaginative and provocative available.

Perhaps nobody wishes to curb the imagination of men of science; I do not know. It is my impression that imagination, critical debate, interest in science, all these beautiful things are still viewed with suspicion, and for many people science and scholarship must be tedious and dull and pedestrian before they can appreciate it. Be that as it may, the theories of science which have gained traditional approval are certainly constrain­ing, and more often than not they obviously fail to do justice to the imagination and to criticism. Popper’s theory of science as imagination checked and stimulated by criticism is at least in intention so very wonderful. I have tried to examine how much of science it makes sense of without additional constraints and ad hoc hypotheses. If, contrary to my view, Popper’s program cannot be executed in its purity, we may try something entirely different, with the hope that we shall not do injustice to the value of creativity in free imagination and rigourous criticism.

Boston University

REFERENCES

* I am indebted to my wife Judith, to William W. Bartley, III, and to Robert S. Cohen, for their patient reading of many drafts, and making many corrections and suggestions.

¹ See my Towards an Historiography of Science, The Hague 1963.

² See my ‘The Nature of Scientific Problems and their Roots in Metaphysics’, in The Critical Approach: Essays in Honor of Karl Popper (ed. M. Bunge), New York 1964.

³ K. R. Popper, The Logic of Scientific Discovery, New York 1959.

⁴ I owe this to a private conversation with Burtt.

⁵ K. R. Popper, Conjectures and Refutations, Chapter 10, ‘Truth, Rationality, and the Growth of Scientific Knowledge’, § 5.

⁶ Alonzo Church, ‘Mathematics and Logic’, in Logic, Methodology, and Philosophy of Science (Proceedings of the 1960 International Congress) (ed. E. Nagel, P. Suppes, and A. Tarski), Stanford, Calif., 1962.

⁷ See Rudolf Carnap’s ‘Reply to Critics’, in The Philosophy of Rudolf Carnap (ed. P. A. Schilpp), Evanston 1964.