Epistemological Problems of Economics

One can completely agree with Max Weber when he declares:

Wherever the causal explanation of a “cultural phenomenon”—an “historical individual”—comes into question, knowledge of laws of causation cannot be the end, but only the means of investigation, It facilitates and makes possible for us the imputation of the culturally significant components of the phenomena, in their individuality, to their concrete causes. As far and only as far as it accomplishes this is it valuable for the cognition of concatenations in individual cases.58

Weber is wrong, however, when he adds

The more “general,” i.e., the more abstract, the laws, the less they accomplish for the requirements of the causal imputation of individual phenomena and thereby, indirectly, for the understanding of the meaning of cultural events . . . From the point of view of exact natural science, “laws” are all the more important and valuable the more general they are; from the point of view of the cognition of historical phenomena in their concrete setting, the most general laws are also always the least valuable because they are the most empty of content. For the more comprehensive is the validity of a generic concept—i.e., its scope—the more it leads us away from the fullness of reality; because, in order to contain the most common element possible of many phenomena, to be as abstract as possible, it must consequently be devoid of content.59

Although Weber even goes so far as to speak of “all so-called ‘economic laws’ without exception” in the arguments by which he arrives at these conclusions, he could, nevertheless, only have had in mind the well-known attempts to discover laws of historical development. If one recalls Hegel’s famous proposition: “World history . . . depicts the development of the spirit’s consciousness of its freedom, and the material realization brought about by this consciousness,”60  or one of Breysig’s propositions, then Weber’s statements at once become understandable. Applied to the propositions of sociology, they appear inconceivable.

Whoever undertakes to write the history of the last decade will not be able to ignore the problem of reparations.61  At the center of this problem, however, stands that of the transfer of the funds involved. Its essence is the question whether or not the stability of the gold value of German money can be affected by the payment of sums for reparations, and particularly by their transfer to foreign countries. This question can be examined only by the methods of economic theory. Any other way of examining it would simply be nonsensical. It is worthy of note that not just some of those who have participated in this discussion, but all without exception, from first to last resort to the universally valid propositions of economic theory. Even one who starts from the balance-of-payments theory, which science has decisively rejected, adheres to a doctrine that makes the same logical claim to universal validity as the theory that modem science acknowledges as correct. Without recourse to such propositions, a discussion of the consequences that must follow on certain assumptions could never be carried on. In the absence of a universally valid theory, the historian will be unable to make any statements connected with the transfer of funds, no matter whether the payments are actually made according to the Dawes Plan or whether they cease for some reason not yet given. Let us assume that the payments are made and that the gold value of the mark does not change. Without recourse to the principle of the theory of purchasing-power parity, one could still not infer from this that Germany’s payment had not affected its currency. It could be that another causal chain, acting at the same time, did not permit the effect on currency anticipated by the balance-of-payments theory to become visible. And if this were so, the historian would either completely overlook this second causal chain or would not be able to understand its effect.

History cannot be imagined without theory. The naive belief that, unprejudiced by any theory, one can derive history directly from the sources is quite untenable. Rickert has argued in an irrefutable way that the task of history does not consist in the duplication of reality, but in its reconstitution and simplification by means of concepts.62  If one renounces the construction and use of theories concerning the connections among phenomena, on no account does one arrive at a solution of the problems that is free of theory and therefore in closer conformity with reality. We cannot think without making use of the category of causality. All thinking, even that of the historian, postulates this principle. The only question is whether one wants to have recourse to causal explanations that have been elaborated and critically examined by scientific thought or to uncritical, popular, prescientific “dogmas.” No explanations reveal themselves directly from the facts. Even if one wanted to draw conclusions uncritically—post hoc, ergo propter hoc—one would be completely at a loss in view of the confusing plethora and diversity of phenomena. It is precisely the “multifarious causal complexity” of processes of which Muhs speaks,63  i.e., the concurrence in them of a multiplicity of causal factors, that makes theory necessary.

For ages historians have made use of theories provided by nonscientific thought and laying claim to universal validity. Consider to what an extent such a theory is contained in the simple sentence, “The defeated king found himself forced to conclude peace under unfavorable conditions.” What is involved here are simple and scarcely disputed theories, which, by their very character, are nonscientific, but this does not change the fact that they are still theories, i.e., statements understood as universally valid. In addition, the historian employs theories taken from all the other sciences, and it goes without saying that one is justified in demanding, in such cases, that the theories used conform to the present state of science, i.e., they must, in our view, be correct theories. The old Chinese historian could trace extraordinarily dry weather back to moral lapses on the part of the emperor and report that after the monarch’s expiation rain fell again. The ancient historian could ascribe the early death of the king’s son to the jealousy of the gods. Today, in the present state of meteorology and pathology, we look for a different explanation. Even though the sources were to inform us unequivocally that Numa Pompilius was acquainted with Camena Egeria, we would be unable to believe it and would disregard them. The intercourse of witches with the devil has been established as proved according to the rules of legal evidence; yet, on the strength of our theory, we deny this possibility, all documents to the contrary notwithstanding.64  The historian must regard all other sciences as auxiliary to his own and must be thoroughly familiar with as much of them as is required by the particular tasks he has set for himself. Whoever treats of the history of the Julian-Claudian dynasty will scarcely be able to do without a knowledge of the theory of heredity and psychiatry. Whoever writes a history of bridge-building will need a thorough knowledge of bridge-building; whoever writes a history of strategy will need a thorough knowledge of strategy.

Now the proponents of historicism, of course, admit all this as far as all other sciences are concerned, but they deny it with reference to sociology. Here the matter seems to them to be different. No substantial reason for this difference is to be discovered, but, psychologically, the resistance of many historians is easily understood. As far as the other sciences are relevant to history, the alternative is either that the historian needs to acquire a moderate degree of knowledge, which does not exceed the amount possessed as a matter of course by every educated person, or that special fields of historical knowledge not closely connected with the sphere proper to history become autonomous disciplines. One does not have to be a meteorologist to know that no matter how serious the failings of the monarch, they cannot influence the weather. And even one who understands only very little of the theory of heredity will know what weight to attach to the divine extraction that historical sources attribute to many dynasties. Making the history of medicine and similar disciplines autonomous affects but slightly the sphere proper to history. The claims of sociology, however, even if only as a result of the failure to recognize the boundaries between sociological and historical investigations, are felt by many historians as an infringement on their very own domain.

Each and every proposition of history implicitly contains theorems of sociology. No statement concerning the effect of political measures is conceivable that could forgo recourse to universally valid propositions about human action. Whether the topic under discussion is the “social question,” mercantilist policy, imperialism, power politics, or wars and revolutions, we again and again encounter in the historian’s discussions statements that are inferences from universally valid propositions of sociology. just as Monsieur Jourdain was astonished to learn that what he had always been speaking was prose, so historians too show surprise when one points out to them that they make use of the theorems of sociology from first to last.

It is regrettable, however, that these theorems, which they unhesitatingly employ, occasionally belong to prescientific thought. One who disregards the results of modern sociology does not therefore work “free of theory.” He employs the naive, obsolete theory of an epoch of scientific thought long since superseded or else the still more naive theory of prescientific thought. The effect this has on economic history is nothing short of grotesque. Economic history did not become possible until classical economics had produced a scientific apparatus for political and economic thought. Previous attempts—for example, those dealing with the history of trade—were nothing but a compilation of memoranda. Nowadays the economic historian seeks to emancipate himself from theory altogether. He disdains to approach his task with the logical tools of a developed scientific theory and prefers to content himself with the small measure of theoretical knowledge that today reaches everyone through the newspapers and daily conversation. The presuppositionlessness of which these historians boast consists, in reality, in the uncritical repetition of eclectic, contradictory, and logically untenable popular misconceptions, which have been a hundred times refuted by modern sciences.65  Thus, the diligent work performed by entire generations of scholars has remained unproductive. The Historical School failed precisely in the province of social and economic history, which it claimed as its proper domain.

Now the champions of history “devoid of theory” maintain, of course, that their concepts and theorems must be derived from the historical data, inasmuch as there are no universally valid, supertemporal laws of human action. As we have seen, the thesis that there can also be irrational action and that rational action is generally only the result of a long historical development rests on a gross misunderstanding. Historicism, however, goes still further. It dismisses the doctrine of the supertemporality of reason as a prejudice of the Enlightenment. The logical structure of human reason, we are informed, has changed in the course of the ages, in the same way as, for example, technical knowledge and skills.66

We shall not enter here into what is to be said in principle, from the standpoint of sociology, against this postulate of historicism.67  In any case, such reasoning would prove unacceptable to the proponents of historicism, who deny the possibility of any supertemporal theory in contradistinction to historical experience. Therefore, we must confine ourselves to what even historicism must acknowledge as an immanent critique of its thesis. The first point to be established, however, is that none of the sources of historical information accessible to us contains anything that could shake the assumption of the immutability of reason. Never has even an attempt been made to state concretely in what respects the logical structure of reason could have changed in the course of the ages. The champions of historicism would be greatly embarrassed if one were to require of them that they illustrate their thesis by pointing out an example.

In this respect, the failure of ethnology has been no less conspicuous than that of history. Wilhelm Jerusalem to be sure, has emphatically stated: “Kant’s firm belief in the timeless, completely immutable logical structure of our reason . . . has not only not been confirmed by the findings of modern ethnology, but has been proved completely incorrect.”68  But even Jerusalem has not undertaken in a single instance to show us in what way the logic of primitive peoples is structurally different from our logic. A general appeal to the writings of ethnologists is not sufficient here. Ethnology shows only that the conclusions arrived at by the reasoning of primitive peoples are different from those which we arrive at and that the range of things primitive peoples are accustomed to think about is different from the circle of our intellectual interests. When primitive man assumes magical and mystical connections where we assume connections of a different kind, or where we find no connection at all, or when he sees no connection where we do see one, this shows only that the content of his reasoning differs from that of our own, but not that his reasoning is of a different logical structure from ours.

In support of his statement, Jerusalem refers repeatedly to the works of Lévy-Bruhl. However, nothing that Lévy-Bruhl sets forth in his admirable Writings on this topic says anything more than that members of primitive races have no understanding of the problems with which, in the civilized countries, a narrow circle of intellectually distinguished men concern themselves. “An African,” says Lévy-Bruhl, borrowing from Bentley’s narrative,

never thinks a matter out if he can help it. . . . They never recognized any similarity between their own trading and the coast factory. They considered that when the white man wanted cloth, he opened a bale and got it. Whence the bales came and why and how—that they never thought of.

The primitive man has a habit of mind which makes him

stop short at his earliest perception of things and never reason if he can in any way avoid it.69

Lévy-Bruhl and Bentley seem to have confined their association to the members of primitive races. Had they also looked about in Europe—and, one might add, among European economists and politicians—they would certainly not have considered the practice of never thinking matters out and never reasoning as peculiarities of primitive peoples alone. As Lévy-Bruhl says, citing a report by Mangin, the Mossi on the Niger river are lacking in reflection. For that reason they are also wanting in ideas.

Conversation with them turns only upon women, food, and (in the rainy season) the crops.70

What other subjects did many contemporaries of Newton, Kant, and Lévy-Bruhl prefer?

It must be pointed out, moreover, that from the data he compiled, Lévy-Bruhl never draws the conclusions that Jerusalem wants to infer from them. For example, expressly summing up his observations about the causal reasoning of primitive races, Lévy-Bruhl remarks:

The primitive mind, like our own, is anxious to find the reasons for what happens, but it does not seek these in the same direction as we do. It moves in a world where innumerable occult powers are everywhere present, and always in action or ready to act.71

And, on the basis of searching investigations, Cassirer arrives at the conclusion:

When one compares the empirical-scientific and the mythical conceptions of the world, it becomes immediately obvious that the contrast between them is not based on their employing totally different categories in the study and explanation of reality. It is not in the nature, the quality of these categories, that myth and empirical-scientific cognition differ, but in their modality. The methods of connecting things that both employ in order to give the perceptibly diverse the form of unity so as to fit the manifold into a framework demonstrates a thoroughgoing analogy and correspondence. They exhibit the same most general “forms” of perception and reasoning which constitute the unity of consciousness as such and which, therefore, constitute the unity of mythical consciousness in the same way as that of pure cognitive consciousness.72

What the proponents of historicism fail to see is that even propositions like: “The theorems of classical economics possessed relative truth for the age in which they were constructed” can be enunciated only if one has already adopted a supertemporal, universally valid theory. Without such a theory the historian could not consider his task anything more than the compilation and publication of source materials. Thus, it has been no fortuitous coincidence, but inner necessity, that the age in which historicism has held sway has been characterized by a progressive decline in historical research and historical writing. With a few laudable exceptions, for history the upshot of historicism has been, on the one hand, the publication of sources, and, on the other hand, dilettantist constructions, such as those of Chamberlain and Spengler.

If history is not to be a meaningless absurdity, then every statement that it makes about a causal relationship must be thought through to its conclusion and examined for its compatibility with the entire structure of our knowledge. However, this cannot be done without sociological theory.

One must agree completely with Max Weber when he says that for the causal explanation of cultural phenomena “knowledge of laws of causation cannot be the end, but only the means of investigation.” Sociology is an auxiliary—though, to be sure, an indispensable auxiliary—of history. Sociological—and especially economic—theory stands in the same relationship to politics. Every science is an end in itself only for him who thirsts after the knowledge of it.

• 58Weber, Wissenschaftslehre, p. 190.
• 59Ibid., pp. 178 ff.
• 60G.W.F. Hegel, <em>Vorlesungen über die Philosophie der Weltgeschichte</em>, ed. by Lasson (Leipzig, 1917), Vol. 1 (<em>Philosophische Bibliothek</em>, Vol. 171a), p. 148.
• 61In judging this example it should be noted that it has been carried over unchanged from the first publication of this article, which appeared in 1929.
• 6262Cf. Rickert, <em>Kulturwissenschaft und Naturwissenschaft</em>, pp. 28 ff. Cf. further, Sombart, “Zur Methode der exakten und historischen Nationalökonomie,” <em>Schmollers Jahrbuch</em>, LII, 647.
• 63Cf. Muhs, op. cit., p. 808.
• 64“Historiquement, le diable est beaucoup plus solidement prouvé que Pisistrate: nous n’avons pas un seul mot d’un contemporain qui dise avoir vu Pisistrate; des milliers des ‘temoins oculaires’ déclarent avoir vu le diable, il y a peu de faits historiques établis sur un pareil nombre de témoignages indépendants. Pourtant nous n’hesitons plus à rejeter le diable et à admettre Pisistrate. C’est que l’existence du diable serait inconciliable avec les lois de toutes les sciences constituées.” Langlois-Seignobos, Introduction aux études historiques (3rd ed.; Paris, 1905), pp. 177 f.
• 65Cf. Celestin Charles Alfred Bouglé, Qu’est-ce que la sociologie? (5th ed.; Paris, 1925), pp. 54 ff.
• 66Cf. Karl Mannheim, “Historismus,” Archiv für Sozialwissenschaft, LII, 9.
• 67Cf. Husserl, Logische Untersuchungen, I, 136 ff.
• 68Franz W. Jerusalem, “Die soziologische Bedingtheit des Denkens und der Denkformen,” Versuche zu einer Soziologie des Wissens, ed. by Max Scheler (Munich and Leipzig, 1924), p. 183.
• 69Cf. Lévy-Bruhl, Primitive Mentality, trans. by Lilian Clare (New York, 1923), pp. 27 f.
• 70Ibid., p. 27.
• 71Ibid., p. 437.
• 72Ernst Cassirer, Philosophie der symbolischen Formen (Berlin, 1925), II, 78.