The Poverty of Historicism

How can this objection be countered? There are, in the main, two positions which may be taken up by those who believe in a law of evolution. They may (a) deny our contention that the evolutionary process is unique; or (b) assert that in an evolutionary process, even if it is unique, we may discern a trend or tendency or direction, and diat we may formulate a hypothesis which states this trend, and test this hypothesis by future experience. The two positions (a) and (b) are not exclusive of each other.

It is important to point out that laws and trends are radically different things.¹⁵ There is little doubt that the habit of confusing trends with laws, together with the intuitive observation of trends (such as technical progress), inspired the central doctrines of evolutionism and historicism—the doctrines of the inexorable laws of biological evolution and of the irreversible laws of motion of society. And the same confusions and intuitions also inspired Comte’s doctrine of laws of succession—a doctrine which is still very influential.

Now we have seen that there are no laws that determine the succession of such a ‘dynamic’ series of events.¹⁹ On the other hand, there may be trends which are of this ‘dynamic’ character; for example, population increase. It may therefore be suspected that Mill had such trends in mind when he spoke of ‘laws of succession’. And this suspicion is confirmed by Mill himself when he describes his historical law of progress as a tendency. Discussing this ‘law’, he expresses his ‘belief.… that the general tendency is, and will continue to be saving occasional and temporary exceptions, one of improvement—a tendency towards a happier and better state. This … is … a theorem of the science’ (viz. of the social science). That Mill should seriously discuss the question whether ‘the phenomena of human society’ revolve ‘in an orbit’ or whether they move, progressively, in ‘a trajectory’²⁰ is in keeping with this fundamental confusion between laws and trends, as well as with the holistic idea that society can ‘move’ as a whole—say, like a planet.

My criticism of the doctrine of historical laws of succession is in one important respect still inconclusive. I have tried to show that the ‘directions’ or ‘tendencies’ which historicists discern in the succession of events called history are not laws but, if anything, trends. And I have pointed out why a trend, as opposed to a law, must not in general be used as a basis for scientific predictions.

Such a causal explanation will, of course, be scientifically acceptable only if the universal laws are well tested and corroborated, and if we have also some independent evidence in favour of the cause, i.e. of the initial conditions.

This, we may say, is the central mistake of historicism. Its ‘laws of development’ turn out to be absolute trends; trends which, like laws, do not depend on initial conditions, and which carry us irresistibly in a certain direction into the future. They are the basis of unconditional prophecies, as opposed to conditional scientific predictions.

In passing I should like to mention that neither the principle of methodological individualism, nor that of the zero method of constructing rational models, implies in my opinion the adoption of a psychological method. On the contrary, I believe that these principles can be combined with the view⁴¹ that the social sciences are comparatively independent of psychological assumptions, and that psychology can be treated, not as the basis of all social sciences, but as one social science among others.

Against my analysis of historical explanation ⁴⁵ it may be argued that history does make use of universal laws contrary to the emphatic declaration of so many historians that history has no interest whatever in such laws. To this we may answer that a singular event is the cause of another singular event—which is its effect—only relative to some universal laws.⁴⁶ But these laws may be so trivial, so much part of our common knowledge, that we need not mention them and rarely notice them. If we say that the cause of the death of Giordano Bruno was being burnt at the stake, we do not need to mention the universal law that all living things die when exposed to intense heat. But such a law was tacitly assumed in our causal explanation.

Among the theories which the political historian presupposes are, of course, certain theories of sociology—the sociology of power, for example. But the historian uses even these theories, as a rule, without being aware of them. He uses them in the main not as universal laws which help him to test his specific hypotheses, but as implicit in his terminology. In speaking of governments, nations, armies, he uses, usually unconsciously, the ‘models’ provided by scientific or pre-scientific sociological analysis (see the foregoing section).

I believe that this analysis clarifies some well-known controversies between certain students of the method of history.⁴⁷ One historicist group asserts that history, which does not merely enumerate facts but attempts to present them in some kind of causal connection, must be interested in the formulation of historical laws, since causality means, fundamentally, determination by law. Another group, which also includes historicists, argues that even ‘unique’ events, events which occur only once and have nothing ‘general’ about them, may be the cause of other events, and that it is this kind of causation that history is interested in. We can now see that both groups are partly right and partly wrong. Universal law and specific events are together necessary for any causal explanation, but outside the theoretical sciences, universal laws usually arouse little interest.

Tolstoy’s historicism is a reaction against a method of writing history which implicitly accepts the truth of the principle of leadership; a method which attributes much—too much, if Tolstoy is right, as he undoubtedly is—to the great man, the leader. Tolstoy tries to show, successfully I think, the small influence of the actions and decisions of Napoleon, Alexander, Kutuzov, and the other great leaders of 1812, in the face of what may be called the logic of events. Tolstoy points out, rightly, the neglected but very great importance of the decisions and actions of the countless unknown individuals who fought the battles, who burned Moscow, and who invented die partisan method of fighting. But he believes that he can see some kind of historical determination in these events—fate, historical laws, or a plan. In his version of historicism, he combines both methodological individualism and collectivism; that is to say, he represents a highly typical combination—typical of his time, and, I am afraid, of our own—of democratic-individualist and collectivist-nationalistic elements.

This example may remind us that there are some sound elements in historicism; it is a reaction against the naive method of interpreting political history merely as the story of great tyrants and great generals. Historicists rightly feel that there may be something better than this method. It is this feeling which makes their idea of ‘spirits’—of an age, of a nation, of an army— so seductive.

Such selective approaches fulfil functions in the study of history which are in some ways analogous to those of theories in science. It is therefore understandable that they have often been taken for theories. And indeed, those rare ideas inherent in these approaches which can be formulated in the form of testable hypotheses, whether singular or universal, may well be treated as scientific hypotheses. But as a rule, these historical ‘approaches’ or ‘points of view’ cannot be tested. They cannot be refuted, and apparent confirmations are therefore of no value, even if they are as numerous as the stars in the sky. We shall call such a selective point of view or focus of historical interest, if it cannot be formulated as a testable hypothesis, a historical interpretation.

The weakness of this ‘reduction’ of Mill’s seems obvious. Even if we should grant Mill’s premises and deductions, it still would not follow that the social or historical effect will be significant. Progress might be rendered negligible, say, by losses due to an unmanageable natural environment. Besides, the premises are based on only one side of ‘human nature’ without considering other sides such as forgetfulness or indolence. Thus where we observe the precise opposite of the progress described by Mill, there we can equally well ‘reduce’ these observations to ‘human nature’. (Is it not, indeed, one of the most popular devices of so-called historical theories to explain the decline and fall of empires by such traits as idleness and a propensity to over-eat?) In fact we can conceive of very few events which could not be plausibly explained by an appeal to certain propensities of ‘human nature’. But a method that can explain everything that might happen explains nothing.

If we wish to replace this surprisingly naive theory by a more tenable one, we have to take two steps. First, we have to attempt to find conditions of progress, and to this end we must apply the principle set out in section 28: we must try to imagine conditions under which progress would be arrested. This immediately leads to the realization that a psychological propensity alone cannot be sufficient to explain progress, since conditions may be found on which it may depend. Thus we must, next, replace the theory of psychological propensities by something better; I suggest, by an institutional (and technological) analysis of the conditions of progress.

In this connection, I may perhaps touch upon another of the doctrines presented in section 6 (Objectivity and Valuation). There it was argued that, since scientific research in social problems must itself influence social life, it is impossible for the social scientist who is aware of this influence to retain the proper scientific attitude of disinterested objectivity. But there is nothing peculiar to social science in this situation. A physicist or a physical engineer is in the same position. Without being a social scientist he can realize that the invention of a new aircraft or rocket may have a tremendous influence on society.

I have just sketched some of the institutional conditions on whose realization scientific and industrial progress depends. Now it is important to realize that most of these conditions cannot be called necessary, and that all of them taken together are not sufficient.

Modern historicists, however, seem to be unaware of the antiquity of their doctrine. They believe—and what else could their deification of modernism permit?—that their own brand of historicism is the latest and boldest achievement of the human mind, an achievement so staggeringly novel that only a few people are sufficiently advanced to grasp it. They believe, indeed, that it is they who have discovered the problem of change—one of the oldest problems of speculative metaphysics. Contrasting their ‘dynamic’ thinking with the ‘static’ thinking of all previous generations, they believe that their own advance has been made possible by the fact that we are now ‘living in a revolution’ which has so much accelerated the speed of our development that social change can be now directly experienced within a single lifetime. This story is, of course, sheer mythology. Important revolutions have occurred before our time, and since the days of Heraclitus change has been discovered over and over again.⁵⁸

¹ See F. A. von Hayek, ‘Scientism and the Study of Society’, Economica, N.S., vol. IX, especially p. 269. Professor Hayek uses the term ‘scientism’ as a name for ‘the slavish imitation of the method and language of science’. Here it is used, rather, as a name for the imitation of what certain people mistake for the method and language of science.

² I agree with Professor Raven when, in his Science, Religion, and the Future (1943), he calls this conflict ‘a storm in a Victorian tea-cup’; though the force of this remark is perhaps a little impaired by the attention he pays to the vapours still emerging from the cup—to the Great Systems of Evolutionist Philosophy, produced by Bergson, Whitehead, Smuts, and others.

³ Feeling somewhat intimidated by the tendency of evolutionists to suspect anyone of obscurantism who does not share their emotional attitude towards evolution as a ‘daring and revolutionary challenge to traditional thought’, I had better say here that I see in modern Darwinism the most successful explanation of the relevant facts. A good illustration of the emotional attitude of evolutionists is C. H. Waddington’s statement (Science and Ethics, 1942, p. 17) that ‘we must accept the direction of evolution as good simply because it is good’; a statement which also illustrates the fact that the following revealing comment by Professor Bernal upon the Darwinian controversy (ibid., p. 115) is still apposite: ‘It was not … that science had to fight an external enemy, the Church; it was that the Church … was within the scientists themselves.’

⁴ Even a statement such as ‘All vertebrates have one common pair of ancestors’ is not, in spite of the word ‘all’, a universal law of nature; for it refers to the vertebrates existing on earth, rather than to all organisms at any place and time which have that constitution which we consider as characteristic of vertebrates. See my Logic of Scientific Discovery, section 14 f.

⁵ See T. H. Huxley, Lay Sermons (1880), p. 214. Huxley’s belief in a law of evolution is very remarkable in view of his exceedingly critical attitude towards the idea of a law of (inevitable) progress. The explanation appears to be that he not only distinguished sharply between natural evolution and progress, but that he held (rightly, I believe) that these two had little to do with each other. Julian Huxley’s interesting analysis of what he calls ‘evolutionary progress’ (Evolution, 1942, pp. 559 ff.) seems to me to add little to this, although it is apparently designed to establish a link between evolution and progress. For he admits that evolution, though sometimes ‘progressive’, is more often not so. (For this, and for Huxley’s definition of ‘progress’, see note 26 on p. 117, below.) The fact, on the other hand, that every ‘progressive’ development may be considered as evolutionary, is hardly more than trivial. (That the succession of dominant types is progressive in his sense may merely mean that we habitually apply the term ‘dominant types’ to those of the most successful types which are the most ‘progressive’.)

⁶ See H. A. L Fisher’s History of Europe, vol. I, p. vii (italics mine). See also F. A. von Hayek, op. cit., Economica, vol. X, p. 58, who criticizes the attempt ‘to find laws where in the nature of the case they cannot be found, in the succession of the unique and singular historical phenomena’.

⁷ Plato describes the cycle of the Great Year in The Statesman; proceeding from the assumption that we live in the season of degeneration, he applies this doctrine in The Republic to the evolution of Greek cities, and in the Laws to the Persian Empire.

⁸ Professor Toynbee insists that his method is to investigate empirically the life cycle of 21 odd specimens of the biological species ‘civilization’. But even he does not seem to be influenced, in his adoption of this method, by any desire to counter Fisher’s argument (quoted above); at least, I do not find any indication of such a desire in his comments on this argument which he is content to dismiss as an expression of ‘the modern Western belief in the omnipotence of chance’; see A Study of History, vol. V, p. 414.1 do not think that this characterization does justice to Fisher, who says in the continuation of the passage quoted: ‘… The fact of progress is written plain and large on the page of history; but progress is not a law of nature. The ground gained by one generation may be lost by the next.’

⁹ In biology, the position is similar in so far as a multiplicity of evolutions (e.g. of different genera) may be taken as the basis of generalizations. But this com parison of evolutions has merely led to the description of types of evolutionary processes. The position is the same as in social history. We may find that certain types of events are repeated here or there, but no law describing either the course of all evolutionary processes (such as a law of evolutionary cycles) or the course of all evolution in general appears to result from such a comparison. See note 26 on p. 117, below.

¹⁰ Of nearly every theory it may be said that it agrees with many facts: this is one of the reasons why a theory can be said to be corroborated only if we are unable to find refuting facts, rather than if we are able to find supporting facts; see section 29, below, and my Logic of Scientific Discovery, especially ch. X. An example of the procedure criticized here is, I believe, Professor Toynbee’s allegedly empirical investigation into the life-cycle of what he calls the ‘species civilization’ (see note 8 on p. 101, above). He seems to overlook the fact that he classifies as civilizations only such entities as conform to his a priori belief in life-cycles. For example, Professor Toynbee contrasts (op. cit., vol. 1, pp. 147 to 149) his ‘civilizations’ with ‘primitive societies’ in order to establish his doctrine that these two cannot belong to the same ‘species’ although they may belong to the same ‘genus’. But the only basis of this classification is an a priori intuition into the nature of civilizations. This may be seen from his argument that the two are obviously as different as are elephants from rabbits—an intuitive argument whose weakness becomes clear if we consider the case of a St. Bernard dog and a Pekingese. But the whole question (whether or not the two belong to the same species) is inadmissible, for it is based on the scientistic method of treating collectives as if they were physical or biological bodies. Although this method has often been criticized (see, for example, F. A. von Hayek, Economica, vol. X, pp. 41 ff.) these criticisms have never received an adequate reply.

¹¹ Toynbee, op. cit., vol. 1, p. 176.

¹² This is so because of the law of inertia.—For an example of a typically ‘scientistic’ attempt to compute political ‘forces’ with the help of the Pythagorean theorem, see note 9 on pp. 57–8, above.

¹³ The confusion created by the talk about ‘motion’, ‘force’, ‘direction’, etc., may be gauged by considering that Henry Adams, the famous American historian, seriously hoped to determine the course of history by fixing the position of two points on its track—the one point located in the thirteenth century, the other in his own lifetime. He says himself of his project: ‘With the help of these two points … he hoped to project his lines forward and backward indefinitely … ‘, for, he argued, ‘any schoolboy could see that man as a force must be measured by motion, from a fixed point’ (The Education of Henry Adams, 1918, p. 434 f.). As a more recent example, I may quote Waddington’s remark (Science and Ethics, p. 17 f.) that ‘a social system’ is ‘something the existence of which essentially involves motion along an evolutionary path … ‘, and that (p. 18 f.) ‘the nature of science’s contribution to ethics … is the revelation of the nature, the character and direction of the evolutionary process in the world as a whole …’

¹⁴ See my Logic of Scientific Discovery, section 15, where reasons are given for considering existential statements to be metaphysical (in the sense of unscientific); see also note 28 on p. 118, below.

¹⁵ A law, however, may assert that under certain circumstances (initial condi tions) certain trends will be found; moreover, after a trend has been so explained, it is possible to formulate a law corresponding to the trend; see also note 29 on p. 119, below.

¹⁶ It may be worth mentioning that equilibrium economics is undoubtedly dynamic (in the ‘reasonable’ as opposed to the ‘Comtean’ sense of this term), even though time does not occur in its equation. For this theory does not assert that the equilibrium is anywhere realized; it merely asserts that every disturbance (and disturbances occur all the time) is followed by an adjustment—by a ‘movement’ towards equilibrium. In physics, statics is the theory of equilibria and not of movements towards equilibrium; a static system does not move.

¹⁷ Mill, Logic, Book VI, ch. X, section 3. For Mill’s theory of ‘progressive effects’ in general, see also Book III, ch. XV, section 2 f.

¹⁸ Mill seems to overlook the fact that only the very simplest arithmetical and geometrical sequences are such that ‘a few terms’ suffice for detecting their ‘principle’. It is easy to construct more complicated mathematical sequences in which thousands of terms would not suffice to discover their law of construction—even if it is known that there is such a law.

¹⁹ For the nearest approach to such laws, see section 28, especially note 29 on p. 119.

²⁰ See Mill, loc. cit. Mill distinguishes two senses of the word ‘progress’; in the wider sense, it is opposed to cyclic change but does not imply improvement. (He discusses ‘progressive change’ in this sense more fully, op. cit., Book III, ch. XV) In the narrower sense, it implies improvement. He teaches that the persist ence of progress in the wider sense is a question of method (I do not understand this point), and in the narrower sense a theorem of sociology.

²¹ In many historicist and evolutionist writings it is often impossible to discover where metaphor ends and serious theory begins. (See for example the notes on pp. 102 and 105 of the present section.) And we must even face the possibility that some historicists may deny that there is a difference between metaphor and theory. Consider, for example, the following quotation from the psycho-analyst Dr. Karin Stephen: ‘That the modern explanation which I have tried to put forward may still be no more than a metaphor I will concede … I do not think we need be ashamed … because scientific hypotheses are in fact all based on metaphor. What else is the wave theory of light …?’ (Cp. Wad-dington’s Science and Ethics, p. 80; see also p. 76 on gravity.) If the method of science were still that of essentialism, i.e. the method of asking ‘what is it?’ (cp. section 10 above), and if the wave theory of light were the essentialist statement that light is a wave motion, then this remark would be justified. But as things are, it is one of the main differences between psycho-analysis and the wave theory of light that while the former is still largely essentialistic and metaphorical, the latter is not.

²² This and the next quotation are from Mill, Logic, Book VI, ch. X, section 3.1 consider the term ‘empirical law’ (used by Mill as a name for a law of a low degree of generality) as very unfortunate because all scientific laws are empirical: they are all accepted or rejected on the basis of empirical evidence. (For Mill’s ‘empirical laws’, see also op.cit., Book III, ch. VI, and Book VI, ch. V, section 1.) Mill’s distinction has been accepted by C. Menger who opposes ‘exact laws’ to ‘empirical laws’; see The Collected Works, vol. II, pp. 38 ff., and 259 ff.

²³ See Mill, op. cit., Book VI, ch. X, section 4. See also Comte, Cours de philosophic positive, IV, p. 335.

²⁴ ill, op. cit., Book III, ch. XII, section 1. For the ‘derivation’ or ‘inverse deduction’ of what he calls ‘empirical laws’, see also lib. cit., ch. XVI, section 2.

²⁵ This paragraph, containing the analysis of a causal explanation of a specific event, is a near-quotation from my Logic of Scientific Discovery, section 12. At present, I feel inclined to suggest a definition of ‘cause’ on the basis of Tarski’s semantics (which I did not know when that book was written), along the following lines: The (singular) event A is called a cause of the (singular) event B if and only if from a set of true universal statements (laws of nature) a material implication follows whose implicans designates A and whose implicate designates B. Similarly, we could define the concept of a ‘scientifically accepted cause’. For the semantic concept of designation, see Carnap, Introduction to Semantics (1942). It appears that the above definition could be improved by using what Carnap calls ‘absolute concepts’.—For some historical remarks concerning the problem of cause, see note 7 to ch. 25 of my book, The Open Society and Its Enemies.

²⁶ For a discussion of evolutionary trends, see J. Huxley, Evolution (1942), ch. IX. Concerning Huxley’s theory of Evolutionary Progress (op. cit., ch. X) it appears to me that all that can reasonably be asserted is this: the general trend towards an increasing variety of forms, etc., leaves room for the statement that ‘progress’ (Huxley’s definition is discussed below) sometimes occurs, and sometimes not; that the evolutions of some forms is sometimes progressive, while that of most is not; and that there is no general reason why we should expect that forms will occur in the future which have made further progress. (Cp. Huxley’s contention—e.g. op. cit., p. 571—that, if man were wiped out, further progress is in the highest degree improbable. Although his arguments do not convince me, they carry an implication with which I am inclined to agree; namely, that biological progress is, as it were, something accidental.) Concerning Huxley’s definition of evolutionary progress as increasing all-round biological efficiency, i.e. as increasing control over and independence of the environment, I feel that he has indeed succeeded in expressing adequately the intentions of many who have used this term. Furthermore, the defining terms are not, I admit, anthropocentric; they contain no valuation. And yet, to call an increase in efficiency or in control ‘progress’ appears to me as expressing a valuation; it expresses the belief that efficiency or control is good, and that the spread of life and its further conquest of dead matter is desirable. But it is certainly possible to adopt very different values. I do not think therefore that Huxley’s claim that he has given an ‘objective definition’ of evolutionary progress, free from anthropomorphism and value judgments, is tenable. (See op. cit., p. 559; also p. 565, arguing against J. B. S. Haldane’s view that the idea of progress is anthropocentric.)

²⁷ That in Mill’s case it is this confusion which is mainly responsible for his belief in the existence of what I call ‘absolute trends’ can be seen by an analysis of his Logic, Book III, ch. XVI.

²⁸ There are some logical reasons for describing the belief in an absolute trend as unscientific or metaphysical (cp. note 14 on p. 106, above). Such a trend may be formulated by a non-specific or generalized existential statement (‘There exists such and such a trend’), which we cannot test since no observation of devi ation from the trend can disprove this statement; for we can always hope that, ‘in the long run’, deviations in the opposite direction will set matters right again.

²⁹ If we succeed in determining the complete or sufficient singular conditions c of a singular trend t, then we can formulate the universal law: ‘Whenever there are conditions of the kind c there will be a trend of the kind t’. The idea of such a law is unobjectionable from the logical point of view; but it is very different from Comte’s and Mill’s idea of a law of succession which, like an absolute trend, or a law of a mathematical sequence, characterizes the general run of events. Besides, how could we determine that our conditions are sufficient? Or what amounts to the same thing: how could we test a law of the form indicated above? (We must not forget that we are here discussing position (b) of section 27, which involves the claim that the trend can be tested.) In order to test such a law we have to try hard to produce conditions under which it does not hold; to this end we must try to show that conditions of the kind c are insufficient, and that even in their presence, a trend like t does not always occur. A method like this (it is sketched in section 32) would be unobjectionable. But it is inapplicable to the absolute trends of the historicist, since these are necessary and omnipresent concomitants of social life, and cannot be eliminated by any possible interference with social conditions. (We can see here again the ‘metaphysical’ character of the belief in trends which are not specific, such as general trends; the statements expressing such a belief cannot be tested; see also the foregoing note.)

³⁰ See V. Kraft, Die Grundformen der wissenschaftlichen Methoden (1925).

³¹ See my Logic of Scientific Discovery, on which the present section is based, espe cially the doctrine of tests by way of deduction (‘deductivism’) and of the redundancy of any further ‘induction’, since theories always retain their hypo thetical character (‘hypotheticism’), and the doctrine that scientific tests are genuine attempts to falsify theories (‘eliminationism’); see also the discussion of testability and falsifiability.

The opposition here pointed out, between deductivism and inductivism, corresponds in some respects to the classical distinction between rationalism and empiricism: Descartes was a deductivist, since he conceived all sciences as deductive systems, while the English empiricists, from Bacon on, all conceived the sciences as collecting observations from which generalizations are obtained by induction.

But Descartes believed that the principles, the premises of the deductive systems, must be secure and self-evident—’clear and distinct’. They are based upon the insight of reason. (They are synthetic and a priori valid, in Kantian language.) As opposed to this, I conceive them as tentative conjectures, or hypotheses.

These hypotheses, I contend, must be refutable in principle: it is here that I deviate from the two greatest modern deductivists, Henri Poincaré and Pierre Duhem.

Poincaré and Duhem both recognized the impossibility of conceiving the theories of physics as inductive generalizations. They realized that the observational measurements which form the alleged starting-point for the generalizations are, on the contrary, interpretations in the light of theories. And they rejected not only inductivism, but also the rationalistic belief in synthetic a priori valid principles or axioms. Poincare interpreted them as analytically true, as definitions; Duhem interpreted them as instruments (as did Cardinal Bellarmino and Bishop Berkeley), as means for the ordering of the experimental laws—the experimental laws which, he thought, were obtained by induction. Theories thus cannot contain either true or false information: they are nothing but instruments, since they can only be convenient or inconvenient, economical or uneconomical; supple and subtle, or else creaking and crude. (Thus, Duhem says, following Berkeley, there cannot be logical reasons why two or more theories which contradict one another should not all be accepted.) I fully agree with both these great authors in rejecting inductivism as well as the belief in the synthetic o priori validity of physical theories. But I cannot accept their view that it is impossible to submit theoretical systems to empirical tests. Some of them are testable, I think; that is, refutable in principle; and they are therefore synthetic (rather than analytic); empirical (rather than a priori); and informative (rather than purely instrumental). As to Duhem’s famous criticism of crucial experiments, he only shows that crucial experiments can never prove or establish a theory; but he nowhere shows that crucial experiments cannot refute a theory. Admittedly, Duhem is right when he says that we can test only huge and complex theoretical systems rather than isolated hypotheses; but if we test two such systems which differ in one hypothesis only, and if we can design experiments which refute the first system while leaving the second very well corroborated, then we may be on reasonably safe ground if we attribute the failure of the first system to that hypothesis in which it differs from the other.

³² For a surprising example of the way in which even botanical observations are guided by theory (and in which they may be even influenced by prejudice), see O. Frankel, ‘Cytology and Taxonomy of Hebe, etc.’, in Nature, vol. 147 (1941), p. 117.

³³ With this and the following paragraph, cp. F. A. von Hayek, ‘Scientism and the Study of Society’, parts I and II, Economica, vols. ix and x, where methodological collectivism is criticized and where methodological individualism is discussed in detail.

³⁴ For the two passages see Economica, vol. ix, p. 289 f. (italics mine).

³⁵ Cp. Erkenntnis, III, p. 426 f., and my Logik der Forschung, 1934, whose sub-title may be translated: ‘On the Epistemology of the Natural Sciences’.

³⁶ A somewhat similar argument can be found in C. Menger, Collected Works, vol. II (1883 and 1933), pp. 259–60.

³⁷ See the ‘null hypothesis’ discussed in J. Marschak, ‘Money Illusion and Demand Analysis’, in The Review of Economic Statistics, vol. XXV, p. 40.—The method described here seems partly to coincide with what has been called by Professor Hayek, following C. Menger, the ‘compositive’ method.

³⁸ Even here it may be said, perhaps, that the use of rational or ‘logical’ models in the social sciences, or of the ‘zero method’, has some vague parallel in the natural sciences, especially in thermodynamics and in biology (the construction of mechanical models, and of physiological models of processes and of organs). (Cp. also the use of variational methods.)

³⁹ See J. Marschak, op. cit.

⁴⁰ See E Sargant Florence, The Logic of Industrial Organisations (1933).

⁴¹ This view is more fully developed in ch. 14 of my Open Society.

⁴² These difficulties are discussed by Professor Hayek, op. cit., p. 290 f.

⁴³ See Econometrica, 1 (1933), p. 1 f.

⁴⁴ See Lionel Robbins, in Economica, vol. V, especially p. 351.

⁴⁵ My analysis may be contrasted with that of Morton G. White, ‘Historical Explanation’ (Mind, N.S., vol. 52, pp. 212 ff.), who bases his analysis on my theory of causal explanation as reproduced in an article by C. G. Hempel. Nevertheless he reaches a very different result. Neglecting the historian’s char acteristic interest in singular events, he suggests that an explanation is ‘historical’ if it is characterized by the use of sociological terms (and theories).

⁴⁶ This has been seen by Max Weber. His remarks on p. 179 of his Gesammelte Aufsdtze zur Wissenschaftslehre (1922) are the closest anticipation of which I know to the analysis offered here. But he is mistaken, I believe, when he suggests that the difference between theoretical and historical science lies in the degree of generality of the laws used.

⁴⁷ See, for example, Weber, op.cit., pp. 8 f., 44 f., 48, 215 ff., 233 ff.

⁴⁸ This anticipates the problems recently laboured but not answered by Professor Toynbee.

⁴⁹ For a criticism of the ‘doctrine … that all historical knowledge is relative’, see Hayek, in Economica, vol. X, pp. 55 ff.

⁵⁰ Comte, Cours de philosophic positive, IV, p. 335.

⁵¹ Mill, Logic, Book VI, ch. X, section 3; the next quotation is from section 6 where the theory is worked out in more detail.

⁵² Comte, op. cit., IV, p. 345.

⁵³ Mill, loc. cit., section 4.

⁵⁴ A fuller criticism of the so-called ‘Sociology of Knowledge’ will be found in ch. 23 of my Open Society and Its Enemies. The problem of scientific objectivity, and its dependence upon rational criticism and inter-subjective testability, is also discussed there in ch. 24, and, from a somewhat different point of view, in my Logic of Scientific Discovery.

⁵⁵ See note 44 on p. 83, above.

⁵⁶ See Waddington (The Scientific Attitude, 1941, pp. Ill and 112), who is prevented neither by his evolutionism nor by his scientific ethics from denying that this freedom has any ‘scientific value’. This passage is criticized in Hayek’s The Road to Serfdom, p. 143.

⁵⁷ The best immanent criticism of the teleological doctrine known to me (and one which adopts the religious point of view and especially the doctrine of creation) is contained in the last chapter of M. B. Foster’s The Political Philosophies of Plato and Hegel.

⁵⁸ See my book The Open Society and Its Enemies, especially ch. 2 f; also ch. 10,

IV

Criticism of the Pro-Naturalistic Doctrines