Originally I simply intended to elaborate and to put into publishable English my talk in Hayek’s seminar (first given in German in Brussels in the house of my friend Alfred Braunthal),170 showing more closely how “historicism” inspired both Marxism and fascism. I saw the finished paper clearly before me: a fairly long paper, but of course easily publishable in one piece.
My main trouble was to write it in acceptable English. I had written a few things before, but they were linguistically very bad. My German style in Logik der Forschung had been reasonably light—for German readers; but I discovered that English standards of writing were utterly different, and far higher than German standards. For example, no German reader minds polysyllables. In English, one has to learn to be repelled by them. But if one is still fighting to avoid the simplest mistakes, such higher aims are far more distant, however much one may approve of them.
The Poverty of Historicism is, I think, one of my stodgiest pieces of writing. Besides, after I had written the ten sections which form the first chapter, my whole plan broke down: section 10, on essentialism, turned out to puzzle my friends so much that I began to elaborate it, and out of this elaboration and a few remarks I made on the totalitarian tendencies of Plato’s Republic—remarks which were also thought obscure by my friends (especially Henry Dan Broadhead and Margaret Dalziel)—there grew, or exploded, without any plan and against all plans, a truly unintended consequence, The Open Society. After it had begun to take shape I cut it out of The Poverty and reduced The Poverty to what was more or less its originally intended content.
There was also a minor factor which contributed to The Open Society: I was incensed by the obscurantism of some examination questions about “the one and the many” in Greek philosophy, and I wanted to bring into the open the political tendencies linked with these metaphysical ideas.
After The Open Society had broken away from The Poverty, I next finished the first three chapters of the latter. The fourth chapter, which until then had existed only in a sketchy form (without any discussion of what I later called “situational logic”), was completed, I think, only after the first draft of the Plato volume of The Open Society had been written.
It was no doubt due partly to internal developments in my thought that these works proceeded in this somewhat confused way, but partly also, I suppose, to the Hitler-Stalin pact and the actual outbreak of the war, and to the strange course of the war. Like everybody else, I feared that after the fall of France, Hitler would invade England. I was relieved when he invaded Russia instead, but afraid that Russia would collapse. Yet, as Churchill says in his book on the First World War, wars are not won but lost; and the Second World War was lost by Hitler’s tanks in Russia and by Japan’s bombers at Pearl Harbor.
The Poverty and The Open Society were my war effort. I thought that freedom might become a central problem again, especially under the renewed influence of Marxism and the idea of large-scale “planning” (or “dirigism”); and so these books were meant as a defence of freedom against totalitarian and authoritarian ideas, and as a warning against the dangers of historicist superstitions. Both books, and especially The Open Society (no doubt the more important one), may be described as books on the philosophy of politics.
Both grew out of the theory of knowledge of Logik der Forschung and out of my conviction that our often unconscious views on the theory of knowledge and its central problems (“What can we know?”, “How certain is our knowledge?”) are decisive for our attitude towards ourselves and towards politics.171
In Logik der Forschung I tried to show that our knowledge grows through trial and error-elimination, and that the main difference between its prescientific and its scientific growth is that on the scientific level we consciously search for our errors: the conscious adoption of the critical method becomes the main instrument of growth. It seems that already at that time I was well aware that the critical method—or the critical approach—consists, generally, in the search for difficulties or contradictions and their tentative resolution, and that this approach could be carried far beyond science, for which critical tests are characteristic. For I wrote: “In the present work I have relegated the critical—or, if you will, the ‘dialectical’ —method of resolving contradictions to second place, since I have been concerned with the attempts to develop the practical methodological aspects of my views. In an as yet unpublished work I have tried to take the critical path.…”172 (The allusion was to Die beiden Grundprobleme.)
In The Open Society I stressed that the critical method, though it will use tests wherever possible, and preferably practical tests, can be generalized into what I described as the critical or rational attitude.173 I argued that one of the best senses of “reason” and “reasonableness” was openness to criticism—readiness to be criticized, and eagerness to criticize oneself; and I tried to argue that this critical attitude of reasonableness should be extended as far as possible.174 I suggested that the demand that we extend the critical attitude as far as possible might be called “critical rationalism”, a suggestion which was later endorsed by Adrienne Koch,175 and by Hans Albert.176
Implicit in this attitude is the realization that we shall always have to live in an imperfect society. This is so not only because even very good people are very imperfect; nor is it because, obviously, we often make mistakes because we do not know enough. Even more important than either of these reasons is the fact that there always exist irresolvable clashes of values: there are many moral problems which are insoluble because moral principles may conflict.
There can be no human society without conflict: such a society would be a society not of friends but of ants. Even if it were attainable, there are human values of the greatest importance which would be destroyed by its attainment, and which therefore should prevent us from attempting to bring it about. On the other hand, we certainly ought to bring about a reduction of conflict. So already we have here an example of a clash of values or principles. This example also shows that clashes of values and principles may be valuable, and indeed essential for an open society.
One of the main arguments of The Open Society is directed against moral relativism. The fact that moral values or principles may clash does not invalidate them. Moral values or principles may be discovered, and even invented. They may be relevant to a certain situation, and irrelevant to other situations. They may be accessible to some people and inaccessible to others. But all this is quite distinct from relativism; that is, from the doctrine that any set of values can be defended.177
In this, my intellectual autobiography, a number of the other philosophical ideas of The Open Society (some of them pertaining to the history of philosophy, others to the philosophy of history) ought really to be mentioned—more, indeed, than can be discussed here. Among them is what was the first fairly extensive exposition of my anti-essentialist position and, I suspect, the first statement of an anti-essentialism which is not nominalistic or observationalistic. In connection with this exposition, The Open Society contains some criticisms of Wittgenstein’s Tractatus; criticisms which have been almost completely neglected by Wittgenstein’s commentators.
In a similar context I also wrote on the logical paradoxes and formulated some new paradoxes. I also discussed their relation to the paradox of democracy (a discussion which has given rise to a fairly extensive literature) and to the more general paradoxes of sovereignty.
A voluminous literature, which in my opinion has contributed little to the problem, has sprung from a mistaken criticism of my ideas on historical explanation. In section 12 of Logik der Forschung I discussed what I called “causal explanation”,178 or deductive explanation, a discussion which had been anticipated, without my being aware of it, by J. S. Mill, though perhaps a bit vaguely (because of his lack of distinction between an initial condition and a universal law).179 When I first read “The Poverty of Historicism” in Brussels a former pupil of mine, Dr Karl Hilferding,180 made an interesting contribution to the discussion, to which the philosophers Carl Hempel and Paul Oppenheim also contributed: Hilferding pointed out the relation that some of my remarks on historical explanation had to section 12 of Logik der Forschung. (These remarks eventually became pages 143–46 of the book edition [1957(g)] of The Poverty. Hilferding’s discussion, based on Logik der Forschung, brought out some of the points now on pages 122–24 and 133 of [1957(g)];181 points connected partly with the logical relation between explanation and prediction, and partly with the triviality of the universal laws much used in historical explanations: these laws are usually uninteresting simply because they are in the context unproblematic.)
I did not, however, regard this particular analysis as especially important for historical explanation, and what I did regard as important needed some further years in which to mature. It was the problem of rationality (or the “rationality principle” or the “zero method” or the “logic of the situation”).182 But for years the unimportant thesis—in a misinterpreted form—has, under the name “the deductive model”, helped to generate a voluminous literature.
The much more important aspect of the problem, the method of situational analysis, which I first added to The Poverty183 in 1938, and later explained a little more fully in Chapter 14 of The Open Society,184 was developed from what I had previously called the “zero method”. The main point here was an attempt to generalize the method of economic theory (marginal utility theory) so as to become applicable to the other theoretical social sciences. In my later formulations, this method consists of constructing a model of the social situation, including especially the institutional situation, in which an agent is acting, in such a manner as to explain the rationality (the zero-character) of his action. Such models, then, are the testable hypotheses of the social sciences; and those models that are “singular”, more especially, are the (in principle testable) singular hypotheses of history.
In this connection I may perhaps also refer to the theory of the abstract society, which was first added in the American edition of The Open Society.185
For myself The Open Society marks a turning point, for it made me write history (somewhat speculative history) which, to some extent, gave me an excuse to write about methods of historical research.186 I had done some unpublished research in the history of philosophy before, but this was my first published contribution. I think it has, to say the least, raised a number of new historical problems—in fact, a wasps’ nest of them.
The first volume of The Open Society, which I called The Spell of Plato, originated, as already mentioned, from an extension of section 10 of The Poverty. In the first draft of this extension there were a few paragraphs on Plato’s totalitarianism, on its connection with his historicist theory of decline or degeneration, and on Aristotle. These were based on my earlier reading of the Republic, the Statesman, Gorgias, and some books of the Laws, and on Theodor Gomperz’s Greek Thinkers, a book much beloved since my days in secondary school. The adverse reactions of my New Zealand friends to these paragraphs produced in the end The Spell of Plato, and with it The Open Society. For it turned me back to the study of the sources, because I wanted to give full evidence for my views. I reread Plato most intensively; I read Diels, Grote (whose view, I found, was essentially the same as mine), and many other commentators and historians of the period. (Full references will be found in The Open Society.) What I read was determined largely by what books I could get in New Zealand: during the war there was no possibility of getting books from overseas for my purposes. For some reason or other I could not get, for example, the Loeb edition of the Republic (Shorey’s translation), though the second volume, I found after the war, had been published in 1935. This was a great pity, since it is by far the best translation, as I was to discover later. The translations which were available were so unsatisfactory that, with the help of Adam’s marvellous edition, I began to do translations myself, in spite of my very scanty Greek, which I tried to improve with the help of a school grammar which I had brought from Austria. Nothing would have come of this but for the great amount of time I spent on these translations: I had found before that I had to rewrite again and again translations from Latin, and even from German, if I wanted to make an interesting idea clear, in reasonably forceful English. I have been accused of bias in my translations; and indeed they are biased. But there are no unbiased translations of Plato and, I suggest, there can be none. Shorey’s is one of the few which has no liberal bias, because he accepted Plato’s politics, in the same sense, approximately, in which I rejected them.
I sent The Poverty to Mind, but it was rejected; and immediately after completing The Open Society in February, 1943 (it had been rewritten many times), I sent it to America for publication. The book had been written in trying circumstances; libraries were severely limited, and I had had to adjust myself to whatever books were available. I had a desperately heavy teaching load, and the University authorities not only were unhelpful, but tried actively to make difficulties for me. I was told that I should be well advised not to publish anything while in New Zealand, and that any time spent on research was a theft from the working time as a lecturer for which I was being paid.187 The situation was such that without the moral support of my friends in New Zealand I could hardly have survived. Under these circumstances the reaction of those friends in the United States to whom I sent the manuscript was a terrible blow. They did not react at all for many months; and later, instead of submitting the manuscript to a publisher, they solicited an opinion from a famous authority, who decided that the book, because of its irreverence towards Aristotle (not Plato), was not fit to be submitted to a publisher.
After almost a year, when I was at my wit’s end and in terribly low spirits, I obtained, by chance, the English address of my friend Ernst Gombrich, with whom I had lost contact during the war. Together with Hayek, who most generously offered his help (I had not dared to trouble him since I had seen him only a few times in my life), he found a publisher. Both wrote most encouragingly about the book. The relief was immense. I felt that these two had saved my life, and I still feel so.