Abstract How to do economics was one of Menger’s primary interests, his point of view being distinctly Aristotelian, differing in this aspect greatly from his immediate successors, such as Wieser and Böhm-Bawerk, and also from later Austrian economists. Menger’s Aristotelian realism (or classical realism) ran against the nominalist trend of his own and subsequent time, but it may have a more sympathetic hearing now, with remarkable parallels between Aristotelian essentialism and the thesis of theory-laden facts, associated with Rorty, Feyerabend and others. The most important aspect of his Aristotelian realism was his belief that economic theory had to be abstracted from the phenomena by a rational grasp of economic phenomenal forms, Menger explicitly stating that he was not dealing with deductions from a priori axioms. Instead, he was eager to promote what he called exact science. The pursuit of exact science is simply a certain way of treating any subject matter whatever it may be, a certain direction of cognitive endeavour. An exact law provides a theoretical understanding of only one aspect of actual phenomena and neither can be nor need be verified by full empirical actuality. One can still find significance in our time in Menger’s ideas on how to do economics. The question of how to do economics must surely rank as Carl Menger’s major preoccupation. The irony of Menger is that his observations on this subject, published in his Untersuchungen of 1883, had, with one notable exception, a minimal influence on later economists of the Austrian school, a school of thought he is generally acknowledged to have founded. The most probable reason for this is that the honour of having founded the school really belonged to Friedrich von Wieser and Eugen von Böhm-Bawerk and that neither of these cared very much for what Menger had to say in the Untersuchungen. Wieser came about as close to saying so as he reasonably could in Menger’s obituary of 1922 and in a contribution on Menger to an Austrian biography a year later.1 On matters methodological he revealed himself to have views which in our day we might associate with McCloskey’s aversion for hard and fast procedural rules. Wieser’s admiration for Menger was based purely on the idea of marginal utility. He described the misgivings he and Böhm-Bawerk (they were classmates both at school and at university) had had as students about value theory and how, as among the first readers of Menger’s 1871 Grundsätze, they had found the ‘Archimedean point’ that provided the firm ground on which they could proceed.2 Wieser proceeded to apply the idea exhaustively, among other things, to costs, and Böhm-Bawerk to an essentially Ricardian problem.3 But Menger himself had not placed his emphasis on marginal utility. He appears never to have written specifically about it again. Where he did mention the growing literature on the subject, we have it from Howey, he appears not to have connected it with himself.4 Moreover, as Wieser also pointed out, Menger took no part at all in the subsequent working out of the ‘modern theoretical system’ by members of the Austrian school.5 Menger not only wanted to defend theorizing but he also believed that the economic theory of his day was badly in need of reform6 and that his mission in such reform was to point out the importance of research into what he called ‘exact’ laws. In the event, the reform carried out by the people who are counted as his followers was based on only one specific case of an exact law, albeit an important one, and with time the connection with Menger’s original conception faded away. Of course, when marginal utility spread through other quarters in economics, it soon came about, in Hutchison’s felicitous turn of phrase, that the adjective proved to be more important than the noun. Von Mises, as far as one can tell, was the only prominent member of the Austrian school who seriously took to heart what Menger had to say on how to do economics. His ‘praxeology’ almost certainly was a reinterpretation, in a different philosophical system, of Menger’s exact laws. Since Mises was the fountainhead of the modern Austrian school, it is fair to say that something of Menger still lives on in the ‘subjectivism’ professed by the school. But the connection is rather remote. Menger’s original Aristotelian position was put by Mises initially into a neo-Kantian and ultimately into a plain rationalist form. These are rather unfamiliar positions for many economists today, and it is therefore hardly surprising that the subjectivism of the modern Austrians does not display any notable uniformity. It may therefore be of some interest to those who, like Menger, feel that economic theory is again in need of reform to have a broad outline of Menger’s ideas on how to do economics and a brief assessment of whether such antiquated ideas may be of any practical significance today. It is this that the present paper attempts to provide. Section 2 deals with the basics and Section 3 with the philosophical realism inherent in Menger’s position. Sections 4, 5 and 6 deal with different aspects of his idea of an exact economic theory. Section 7 leads up to the role Menger envisaged for exact theory, and Section 8 makes some suggestions about how one could still find significance in our time in Menger’s ideas on how to do economics. 2 Menger’s deep and abiding interest in the philosophy and method of economics apparently occupied him even before the publication of the Grundsätze.7 His point of view was distinctly Aristotelian.8 According to Kauder, this was common among Austrian intellectuals in the second half of the nineteenth century.9 In the wor1d at large, however, Aristotelianism was as unfashionable in Menger’s day as it is today, if not more so. If Menger was aware of the unfashionability of his views—and it seems reasonable to suppose that he must have been—he concealed this completely from his readers. In fact, he spoke of what amounts to Aristotelian essentialism as though no reasonable person could reject it as an intellectual basis. Menger outlined his standpoint very briefly in the preface to the Grundsätze. He had tried, he said, to resolve the complicated phenomena of economic life into the simplest elements still open to certain observation; then, focusing attention on these, to investigate how complicated economic phenomena develop according to law out of their elements.10 This of course is not a very usual way of stating an Aristotelian position, though it was not foreign even to Aristotle himself. It is interesting that the very first footnote reference in the Grundsätze is to Aristotle’s Politics I.3, and that in I.1 (a page or two earlier in a book of normal size) there appears the sentence: ‘As in other departments of science, so in politics, the compound should always be resolved into the simple elements or least parts of the whole.’11 Such analysis led to an understanding of what in a few places Menger referred to as the eigenthümliche Beschaffenheit (characteristic constitution) of something, but more commonly he used the word Wesen. This word is often translated as essence, especially in an Aristotelian context. And since the etymology of both words seems to indicate an original meaning of being, it is no doubt a correct translation. But Wesen in German usage does not have the connotation of something abstruse that essence has in English usage, so that the word nature perhaps conveys the meaning more accurately. Thus, the title of the first section of the first chapter of the Grundsätze, viz., ‘Ueber das Wesen der Güter’, simply means ‘On the nature of goods’. When Menger wanted to reduce the nature of something to its elements (as he always set out to do), he spoke of Erscheinungsformen (literally, phenomenal forms, i.e. the different forms in which things and events appear) and, when these were recurrent, of Typen (types) to which he added the word strenge (strict) when he wanted to indicate that they were the simplest elements, the simplest, as he sometimes suggested, that could be thought.12 The study of strict types in economics—the morphology of economic phenomena, as he eventually called it13—and of typical relations between phenomenal forms was of fundamental importance to Menger. It systematizes what is in any case done in everyday life. Without a knowledge of types, one could not have a mental grasp of the world of experience, and without a knowledge of the typical relations between phenomenal forms, one would forgo not only a deeper understanding of reality, but also the possibility of a control over things.14 All this is consistent with an Aristotelian outlook, but it could also be called an unusually elaborate articulation of a common-sense approach to a subject. As examples of the phenomenal forms of economics, Menger mentioned purchase, money, supply, demand, price, capital and interest.15 An author who singles out one of these for his attention and begins by discussing the nature of his subject matter and then proceeds to reduce it to its elements, i.e. to analyse it, would not be doing anything out of the ordinary. An author who really understands, and moreover takes seriously the fact-hypothesis dualism often invoked in economics nowadays, would of course avoid any discussion of the nature or essence of things. But a thoroughgoing positivism is probably less common even today than a basic essentialism. Menger, it must be stressed, did not claim to have put forward an unusual viewpoint; in fact, he more or less claimed the opposite. Nevertheless, to keen minds such as Menger’s or Aristotle’s, common sense (at least as it appears, as has sometimes been suggested, to speakers of Indo-European languages) may have implications that are not so familiar. Menger certainly adopted positions which differ markedly not only from those of the majority of present-day mainstream economists but in many cases also from those of later members of the Austrian school. The following sections will try to provide a few glimpses of these not-so-familiar positions. 3 Most important, perhaps, one must appreciate that Menger’s was a realist position, where the realism in question is contrasted with nominalism and may perhaps be called classical realism to distinguish it from the many other meanings of the word. To delve into the distinctions between Platonism, Aristotelianism, rationalism, etc. in the one camp and positivism, falsificationism, instrumentalism and whatever in the other is a pretty bizarre business. But to grasp the distinction between the broad categories of classical realism and nominalism, while it is no easier for a novice, is at least more worthwhile because it sheds some light on many of the problems of the subject of economics. Let us broach the matter with a mental experiment which we may call the ontology test. A number of economists gather and are faced with the problem of choosing between two theories, A and B, which appear to explain some occurrence equally well. Neither explanation is of a kind that suggests an immediate further step, such as ‘Let’s go and have a look’ or ‘Let’s ask Anna, she was there’, but rather, both involve relations between abstract-sounding entities. One of the group, after looking the matter over, immediately maintains, in the spirit of Occam’s razor, that B is the better explanation because it is shorter, i.e. invokes fewer entities or needs fewer assumptions. Whereupon another in the group asks: ‘But now, quite apart from the number of assumptions needed, which, if either, is the correct explanation; which is the true one.’ Now, if you, looking on, think that this is a silly question, then the chances are that you are a nominalist; if you think that it is a perfectly sensible question, then you are likely to be a classical realist; and if you are quite convinced that you do not know whether it is a sensible question, then you may call yourself ontologically modest. The distinction hinges on whether the so-called universal (like a word that may be applied to more than one thing or event) is in some sense real, has some non-physical existence, or whether there are only particular things, stones and plants and so on, to which we attach names (nomina) like labels. If the latter, then it would hardly do to attach only proper names since we would not be able to say anything beyond calling out names, and so, if the latter is indeed the case, we invented the universal as a mental expedient to get along. If we could decide this question, then we would also know whether an economist, or anyone else, could, merely by thinking or reflecting on all he has seen, by a mere rational grasp, come to conclusions of scientific or practical value—and that would be good to know. Platonic ideas are an example of a reality reached by a rational grasp, a reality which (according to this view) helps us both to understand and to achieve our purposes in the seething welter of the physical world. A child of nature could, simply by thinking, arrive at the properties of the circle and thus understand more about cross sections of tree trunks, ripples in a pool and so on; and, should he also invent the wheel, he would know how to achieve a smoother ride. A modern Platonist might point out that the theorems of mathematics could in principle have been worked out by people confined to their rooms, staring at a wall, and yet the results of their reflections eventually helped to put a man on the moon. Plato was preoccupied with mathematical and moral concepts. But Aristotle, his student, had a more empirical and linguistic bent. He was struck by the fact that, though we may be concerned with an individual thing, a ‘this’ as he put it, we can only understand it as a ‘this such’. When we ask: ‘What is this?’, the answer can only be: ‘It is a such.’ In other words, an individual thing or event can only be understood by its attributes, by the universals that may be predicated of it. The universal, according to Aristotle, is not, as Plato had said, in a non-material realm, but in the phenomena, from where we abstract it by a rational grasp. The nature, the essence, the being of a thing or event is understood by its so-being, its definition. In various places, Menger used distinctly Aristotelian expressions. For instance, when distinguishing between cognition and understanding (‘Erkenntniss’ and ‘Verständniss’) in the Untersuchungen, he said that we understand phenomena when we (re)cognize the grounds of their being and their so-being (‘den Grund ihres Seins und ihres So-Seins’) and the so-being of an economic phenomenon is its eigenthümliche Beschaffenheit (characteristic constitution) or nature.16 But much more troublesome to a modern reader is the Aristotelian influence on Menger’s references to causality (other than those in the early parts of the Grundsätze). Aristotle had stated that essence may often be defined in terms of causes and this is also evident in such of Aristotle’s remarks as that ‘to know a thing’s nature is to know the reason why it is’.17 But the current English word cause is a rather misleading translation of whatever the original Greek word was, because only one of the four types of causes listed by Aristotle corresponds at all closely to present-day concepts of causation (e.g. the Aristotelian causes of a house include the purpose for which it is built and the materials out of which it is made).18 The meaning is rendered better by the German word Bedingung, which Menger used most commonly when investigating the nature of something. This word has a most usual meaning of a condition or a stipulation, but may also be used, as Menger did, in the sense of a rational requirement. Thus Menger, who had no word for utility, would say that one of the Bedingungen, one of the rational requirements, for a price is that there should be individuals who attach significance to the capacity of a thing to satisfy a want. But, confusingly to a modem reader, he might also refer to this as a causal law. However, the most important aspect of Menger’s Aristotelian realism is that he believed that economic theory had to be abstracted from the phenomena by a rational grasp of the economic phenomenal forms. He stated explicitly that he was not dealing with concepts or deductions from a priori axioms.19 Of course, one could be mistaken about a theory, just as one could be mistaken about any observation. But there was no question of theory choice. The one true theory was already out there in the phenomena. 4 In the preface to the Grundsätze, Menger claimed that the method of research he had used (i.e. reduction to elements, etc.) was exactly the same as that which had been used to good effect in the physical and natural sciences.20 He frequently drew parallels, especially in his later writings, between chemistry and physics on the one hand and economic theory on the other. However, Menger may easily be misunderstood on this point. The common method he had in mind extended no further than the reduction to elements or types and to the typical relations between types. He was to clarify his ideas on this, at least in print, only much later. As he eventually explained it, he saw two criteria for the division between the various sciences. There was division by subject matter, i.e. by kinds of phenomenal form or simply by the nature of the things studied. By this criterion there was of course a difference between economics and the natural sciences. But there was also division according to the direction of research, or more literally, the direction of cognitive endeavour (Richtung des Erkenntnisstrebens), and the way of seeing (Betrachtungsweise).21 By this criterion there could be various types of science, but the cognitive direction and method (reduction to types and typical relations) that Menger was eager to promote led to what he called exact science and in this sense there could be and were exact natural sciences and an exact science of economics. Menger tended to attach more importance to the second criterion. In one place in the Untersuchungen, for instance, he remarked that the circumstance that research in the field of human action proceeds from an assumption of a definite intention (Willensrichtung) on the part of the human agent was a characteristic peculiar to the social sciences and did not establish an essential difference between the exact natural and social sciences. The exact natural sciences also had characteristics peculiar to themselves.22 To arrive at Menger’s conception of exact science, one has to set aside all consideration of a specific subject matter. The pursuit of exact science is simply a certain way of treating any subject matter as whatever it may be, a certain direction of cognitive endeavour. Menger is at odds on this point with the great majority of the later Austrian school. His explicit statement that there is no ‘unbridgeable gulf’ between any of the sciences23 is directly contrary to Wieser’s contention, which has become something of a tradition in the Austrian school, namely, that the economist may approach his subject matter, as it were, from the inside while the physical scientist can view his only from the outside.24 Menger’s almost obsessive preoccupation with exact science and its universal character must be seen in the context of his defence of theory and, in particular, his opposition to the idea that economics can only be done historically or, as we might say now, hermeneutically. However, Menger did oppose the slavish imitation of the physical sciences. Immediately after pointing out in the Grundsätze that his method of research was the same as that which had led to the great successes of the natural sciences, he said that every method acquired peculiarities according to the sphere of science in which it is applied and insisted that for this reason there could be no question of using the natural-science approach in economics. The attempts which, according to him, had been made to carry the peculiarities of the method of the natural sciences over to economics had led, he said, to an empty playing with superficial analogies between the phenomena of economic life and of nature.25 He did acknowledge that one should not lose sight of the ultimate goal of establishing the connection between all the sciences and the unity of their highest principles, but he thought that this goal could not be attained until the individual sciences had been thoroughly explored.26 Superficially, therefore, Menger sounded much like his followers. But his position did give him the advantage of being able to pinpoint where the difficulty lay and did not require him to invoke the idea of a human mystique that could not be treated like other subjects.27 5 In a paper he read in 1971, Erich Streissler posed the question: ‘To what extent was the Austrian school marginalist?’28 He confined himself to the question whether the Grundsätze provided a basis for the application of calculus and came to the conclusion that ‘marginalism was not the essence of their endeavour.’ ‘Marginalism’, he said, ‘is introduced in the middle of Menger’s Grundsätze, but it is for this very reason not central to ... this very logical construction.’ He drew attention to Menger’s preoccupation with information transmission, ‘interindividual differences in information’, uncertainty, market imperfections, divergences between demand and supply prices, price conflict, goods with more than one price at the same time, bargaining processes, changing qualities of goods, input-output relations which were not fixed and so on. For Menger, he said, ‘everything immediately ramified in some five to ten dimensions’ and summed up: ‘Menger had incorporated into his founding volume practically all the ideas which make the application of the marginal calculus difficult and hazy.’29 Streissler was no doubt correct that Menger’s economics does not lend itself to a marginalist approach. But the conclusions he drew from this accorded Menger less than his due: ‘He wrote a completely subjective theory; and that meant that he eschewed deriving concrete results. He was content to show all the manifold dimensions of causation in the economic field. In his view the final outcome of all these forces at work could not be fully described. And that is the basic failure of his theory: he ended in doubt and not in positive theorems.’ And a little further on: ‘Thus, when we wish to continue his suggestions nowadays, we have to give ourselves a push and jump a little beyond Menger’s paralysing scepticism.’30 Through the eyes of an equilibrium theorist, it must seem that Menger had profound but unsystematic insights, that he delighted in floundering in a morass of perplexity, that his ‘very logical construction’ enriched the subject matter, but also left it more perplexing and intractable than it appeared to be before the analysis was begun. But the point, of course, is that Menger was not looking through the eyes of an equilibrium theorist and it is not hard to imagine how, through his Aristotelian eyes, marginalism would also have looked most unpromising. If economics, like every other science, had its own proper set of phenomenal forms, then marginalism would be an innovation in economic theory in which the specifically economic forms fade from consideration while the economic theorist concentrates on the forms peculiar to the sciences of mechanics and mathematics—and to Menger’s way of thinking, such an innovation would have been downright foolish. Menger’s attitude to mathematics should be seen in a similar light. The question has often been discussed, and it has usually been accepted that Menger disliked the use of mathematics in economics. Opinions have differed on the degree of his mathematical competence, though one should perhaps accept as authoritative that of Menger’s own son, a mathematician, who did not rate his father very highly in this respect.31 Hutchison refers to an extract from a letter Menger wrote to Walras (published by Jaffe) in which he asks rhetorically how one could attain a knowledge of the nature of value, rent, profits, etc. by means of mathematics.32 On the view put forward here, Menger would not have objected to mathematics as a Hilfswissenschaft (auxiliary science, a favourite term with Menger) but rather to its use when it displaced what he regarded as the properly economic phenomenal forms. To him, mathematics was a rigorous exploration (an exact science) of certain selected phenomenal forms just as economic theory was an exploration of its own phenomenal forms. Streissler, however, raised an important point when he said (in the quotation above) that Menger thought that his theory could not handle the total picture. In fact, Menger stressed this constantly, but would have objected strongly to the suggestion that it was a failure or a matter of doubt and a paralysing skepticism. But before we can turn to this rather difficult question, we have to consider some further distinctions that Menger insisted on. 6 On Menger’s own admission, the Untersuchungen was a polemical work.33 It was directed against the historical school of economists which dominated academic economics in Germany during his lifetime, and especially against Schmoller, its leading figure at the time. Menger charged the historical economists with not being able to distinguish between a historical and a theoretical understanding of economic phenomena. The book begins with a discussion of two main directions of research or cognitive endeavour. In one, attention is focused on concrete phenomena in space and time and on relations between them, while in the other, it is focused on recurrent phenomenal forms and on typical relations between them. The former concentrates on the individual, the latter on the general aspects of phenomena, the mark of the ‘individual’ being a definite spatial and temporal setting. When the two main directions of research are brought to bear on economic phenomena, Menger argued, quite distinct disciplines arise. Concentration on individual aspects leads to historical and statistical studies. Concentration on general aspects leads to theoretical economics which includes but is not co-extensive with the exact science of economics. In addition, Menger recognized practical sciences of economics (he gave the example of finance) which were not concerned with cognition at all but with making practical suggestions for achieving certain aims and which drew on other disciplines for this purpose. He conceded a useful role to all these subdivisions of economics, provided that their separate identities were respected, but by implication he put economic theory into a pivotal position because the other sides of economics had to use it as an auxiliary science.34 Menger had formed the impression that the German historical economists were going about their task as though they were combining the historical, statistical, theoretical and practical sides of economics into one comprehensive science—and in a curious use of a word, he called this ‘one-sided’. The main purpose of the Untersuchungen was to show that such a science would be a monstrosity. In this context he introduced a subdivision of the theoretical direction of research which apparently had not occurred to him when he wrote the Grundsätze. He now spoke of the realistic-empirical and exact directions of theoretical research.35 All theory, as before, was concerned with the general, more specifically with laws, or, in the phrase which Menger never tired of using, with the regularities in the coexistence and sequence of phenomena and of phenomenal forms (in the realistic-empirical and the exact directions of theoretical research respectively).36 But the particular, with which the general was contrasted, was now referred to as full empirical actuality37—an expression which was used in the sense of actuality as it is experienced. The realistic-empirical direction of theoretical research extracted from phenomena in their full empirical actuality certain empirical (as opposed to exact) laws about the coexistence and sequence of phenomena and it is therefore nothing other than what is normally called induction. Menger himself occasionally spoke of induction in this context, such as when he praised Aristotle for having already denied the strictly scientific character of induction.38 Menger conceded a useful role to empirical laws in economics, but they were no more than rough-and-ready rules of thumb to which there were countless exceptions and for which there was no guarantee of their continued usefulness.39 In contrast, the exact laws derived by the exact direction of theoretical research did not have these shortcomings. Just how the exact direction of research could arrive at regularities in the coexistence and sequence of phenomena, if not by induction, is not clear. It is the kind of question which all who have professed an essentialism seem to have found difficult, not least Aristotle himself.40 The principle which Menger enunciated in this regard is hardly enlightening and it plays no further role in his subsequent discussion.41 But an entirely clear picture does not emerge there either. He indicated, for instance, that exact science was not to be understood as conceptual analysis nor exact laws as deductions from a priori axioms.42 If they were taken as such, he thought, the exact direction of research would have a methodologically subordinate position in so far as exact laws would have to be corrected by empirical laws whenever a contradiction arose between these types of laws.43 As Menger saw it, this was far from the case and yet much of what he said about the exact direction of research did suggest conceptual analysis. He said, for instance, that exact laws must hold without exception because, according to the laws of our thought, they simply cannot be thought to have exceptions.44 Again he said that exact research does not arrive at exact laws of the actual phenomena of economic life but at laws of economizing (Wirtschaftlichkeit).45 In a book review in 1887, Menger upbraided one of the authors for dealing only with empirical laws. Mentioning laws of economizing again, he described these as laws of rational economic means-ends relations (Gesetze der rationalen ökonomischen Zweckbeziehungen).46 Little wonder, then, that von Mises was later to revive Menger’s exact laws in the form of a priori concepts and propositions. Since the history of Western philosophy is strewn with the flotsam and jetsam of past attempts to deal with the universal, it is perhaps not too surprising that Menger’s exact laws should evince some ambiguity. Moreover, if one does not probe too deeply, it is not too difficult to see what Menger had in mind when he distinguished between exact and empirical laws. The kind of analysis he had presented in the Grundsätze is somehow different from a statement of a connection that is empirical in the original meaning of the word—such as when someone claims that garlic is good for gout. The relation between value and the significance people attach to the capacity of a good to satisfy a want is somehow necessary and intelligible, whereas the claimed connection between garlic and gout is not. There is nothing in the layman’s understanding of what garlic is and what gout is that makes it intelligible to him what, if any, the connection between them is. If there is a connection, it is just a dumb fact, whereas value theory speaks for itself. Econometrics aside, what does ordinary price theory tell us but (in Mengerian terms) the rational requirements for prices and price changes, the exact laws linking the phenomenal forms of a price to other phenomenal forms. That, at any rate, appears to be how Menger would have seen it, and he was convinced that there was a clear distinction between exact and empirical laws. Having expounded the distinction, Menger proceeded to use it as a stick with which to beat the histoncal economists. He charged them with being confused about (what he called) the most elementary principles of scientific method.47 They had rejected exact economics, as Menger understood it, on false grounds in so far as they had insisted on judging it by a criterion appropriate only for empirical laws, namely, that it should be in accordance with full empirical actuality. Empirical laws, being gleaned from full empirical actuality, must continue to be confirmed by it. They always have exceptions, but if they have many exceptions, they become worthless and simply fall away. Not so, exact laws. Tested in this way, they will always seem inadequate. An exact law provides a theoretical understanding of only one aspect of actual phenomena and neither can be nor need be verified by full empirical actuality. One need not test the principle of economizing against what people actually do. Menger likened such a test to an attempt by a mathematician to verify geometry by measuring actual objects.48 A few years earlier, Walras had also alluded to economics pursued more geometrico, in the manner of geometry.49 7 In Appendix 6 of the Untersuchungen, Menger remarked that actual economic phenomena paradoxically are in a large measure of an uneconomic nature.50 The principle of economizing may help us to make practical recommendations. It may also help us to understand better what many people are trying to do. But it cannot enable us to predict ‘full empirical actuality’ because we cannot presume that people actually do economize. Since Menger treated the questions of free will, of error and of imperfect knowledge in an analogous way, we have an indication of what he would have thought of present efforts to model economic action under conditions of uncertainty. We also have an indication, finally, of Menger’s conception of the role of exact economic theory. In the preface to the Grundsätze, Menger warned his readers not to heed the opinion of those who, from a consideration of human free will, denied the regularity of economic phenomena (‘Gesetztmässigkeit’, thus regularity in the sense of conformity to law or principle). On that basis, he thought, the possibility of economics as an exact science was negated altogether and that of course he was intent to deny.51 He conceded that free will was a good reason for not expecting the full regularity of human action itself, and in a later work he said that he did not wish to deny free will as a practical category.52 But he considered the objection to be irrelevant. Free will did not preclude an exact science of economics. On this point, he certainly sounded rather different from present-day Austrian economists. Unfortunately, the argument Menger put forward in 1871 in support of his contention was rather obscure. Among other things, he said that economic theory is related to economic activity as chemistry is related to the activities of the practising chemist.53 He made his meaning clearer in the Untersuchungen when he dealt with the historical school’s objections to the central role of self-interest in economic theory. He outlined the historical school’s strictures on self-interest (understood more or less as selfishness): Mankind acts upon countless, to some extent incompatible, motives and this precludes the strict regularity of human action. Political economy in the Smith tradition (according to the historical school) makes self-interest into a fundamental axiom. Only by thus falsely regarding man as acting upon only one motive can the factor of free will (Willkür) be excluded and economic laws without a temporal and spatial context be conceived.54 Menger agreed that the multiplicity of motives precludes the strict regularity of human action. He then added, as it appears in the English translation:55 But there is another factor, equally important, that does the same thing. I mean error, a factor which surely can be separated still less from human action than custom, public spirit, feeling for justice, and love of one’s fellow man can be separated from the economy ... Our historians are too lenient towards their scholarly opponents. The presupposition of a strict regularity in economic phenomena ... includes not only the dogma of ever-constant self-interest, but also the dogma of the ‘infallibility’ and ‘omniscience’ of men in economic matters. Commenting on this passage, Hutchison remarked:56 in emphasizing this assumption of correct knowledge, and the exclusion of ignorance and error, Menger was taking the first step towards the opening up of the analysis of expectations ... But in doing so he was admitting that his ‘exact’ laws had a very limited applicability to actual economic behaviour and its prediction. And again: Menger himself seemed to admit the seriously limiting nature of the abstraction involved here when he implied that the historical critics had been too forbearing regarding the limitations of economic theory. Hutchison of course is quite correct that Menger’s exact laws had severe limitations. But he is surely wrong in suggesting that Menger was admitting anything, that he was conceding a point to the historical school. The rather heavy-handed sarcasm which Menger reserved for members of the historical school is quite unmistakable in the pages following the passage quoted above. It makes it quite plain that he was conceding nothing to them. Menger went on to say what astonishing folly the greatest minds of all nations had evinced for thousands of years in striving after strict social theories and under what lamentable misconceptions mankind would still labour if the historical school of German economists had not opened its eyes. Moreover, since the same misconceptions were found in other fields of science, also chemistry, physics, mechanics and mathematics were untenable, worthless and in need of reform, deplorable confusions of human thinking. And no one had suspected it until the historical school had opened our eyes to it, partly with the instinct of genius, not fully aware of the epoch-making upheavals in the sphere of exact research. Truly, our historical economists can take pride in these their achievements!57 After carrying on in this vein for more than two pages, Menger even announced to his readers that he was returning to the serious side of the matter. Nor was this the only place where Menger discussed the issue. He mentioned free will, error and/or ignorance in various parts of the text as well as in Appendix 5 and 6 and almost always in the context that they preclude a strict regularity in the phenomena of human action. The word he used for ignorance (Unkenntnis) has a sense purely of a lack of knowledge, in which sense one may be ignorant of the future. Error he sometimes also understood in the sense of uncertainty. Paragraph 4 of the first chapter of the Grundsätze, headed ‘Time - Error’, deals with time and uncertainty. The heading was apparently changed to read ‘Uncertainty’ in what was to be the revised edition of 1923.58 What is significant is that Menger was not in the least perturbed by what many economists regard as an embarrassing untidiness in the subject. Human action in its full empirical actuality, i.e. what people manifestly do before one’s eyes, was simply not explicable by exact laws and yet Menger left no doubt that he regarded research into exact laws as the economic theorist’s highest calling.59 It is hard to understand this attitude unless one takes into account that Menger’s is a voice from another era. One has to be able to see in Menger’s ‘full empirical actuality’ something of the ‘perceptible flux’ of antiquity. Things and events not only have rational requirements which define their essential natures but they also have accidental features. There was a time when the attitude Menger brought to bear on economics appears to have been quite common, and it has lived on in sundry ways until quite recent times: The economic man is not the whole man. An empirical measure is one guided by mere experience and not by scientific knowledge.60 However, over the past four hundred years this attitude seems largely to have been displaced. One may speculate that the development of celestial mechanics out of astronomy has had something to do with this. Observation of celestial bodies, at least over the distances involved, presented no flux, no intractable welter of experience, but only a law-like movement. One may speculate further that the conviction that all else may be reduced to such law-like behaviour arose out of the awe which these scientific achievements inspired. At any rate, the great influence of this conviction, especially in modern economics, can hardly be denied. The old attitude, it appears, has been revived in the signal/noise metaphor current among physical scientists as a façon de parler: One pictures oneself intently listening to a radio, trying to pick out signals from a background of atmospherics. In terms of this metaphor, Menger was listening for signals amidst the noise. Exact science, to extend the metaphor, attunes the ear to recognize signals, but the various sciences specialize in discriminating different kinds of signals. We understand things through theories, said Menger, in so far as each concrete case comes to our attention as an instance of a general uniformity, and in this way we reach an understanding that goes beyond immediate experience (whatever Menger thought that was).61 This, then, is the role Menger saw for theory. It provides a link between the matter of the moment and a whole network of things already known; it widens our perspective; it literally makes us see more. Passing an office block under construction, an engineer sees a system of stresses, an economist sees an investment and a child sees big cranes and yellow safety helmets. However, though theories enlarge our vision in specialized directions, we cannot expect them to explain everything. Menger’s rationale for economic theory was less ambitious than that apparently envisaged for many more recent theories, but at least it had a ring of something familiar about it. Why, for instance, are many aspirant accountants required to take a course or two in economics? How would we answer such a question? Whether or not all things are presumed to have determinants makes a considerable difference. With the belief in universal law-like behaviour, a time must eventually come, as unfalsified hypotheses accumulate or models provide ever closer approximations to economic life, when theory must show the determinants not only of error and ignorance, not to mention frivolity, stupidity and so on, but also of such formidable things as creativity. Since this appears to be a daunting task, the fullness of human conduct becomes something of an embarrassment best ignored. To Menger, however, the fullness of human conduct was not in the least an embarrassment. He spoke of economic life as he was acquainted with it. It was the obvious and natural thing to do and not a matter of a ‘paralysing skepticism’. After all, he was not making assumptions and he was not constructing models. He probably did not even consider the possibility of that procedure sufficiently to be skeptical about it.62 8 Broadly viewed, the conditions in the discipline of economics of our time are remarkably similar to those which perturbed Menger. Much effort is expended on statistical work, more particularly on the search for connections of the dumb-fact garlic-is-good-for-gout sort. There is also much activity in which the purely economic considerations, or what used to be regarded as such, appear to be fast fading from sight while the fun, one suspects, is derived from the delights (and the status) of quite another discipline. Merger of course did not object to the pursuit as such of at least the former of these sorts of activity, because something of great value might well be found. What perturbed him most was the odium of the trivial, of the unscientific, which fell upon the intellectual activity he considered important. One may argue about whether there is a parallel here too. Would things have been different if Menger’s followers had taken up not only subjective value theory but also the matters closer to his heart? Probably not. The earlier parts of this century were so much dominated by various nominalist doctrines that Menger’s ideas on how to do economics would have been laughed out of court. But times have changed and there is now a far greater spirit of anything goes in these matters. In these more fluid times, it may well be worthwhile to try out Menger’s ideas on how to do economics, but to take one’s cue from a Menger without essences. To him this would probably have seemed like doing Hamlet without the Prince of Denmark. But, in fact, most of what he had to say on the role of exact economic theory—as well as on the mistakes and confusions of his contemporaries—would still be there if the classical realist ontology were simply removed. In the years since the gradual fall from grace of positivism, at least among philosophers, many new and interesting ideas have emerged. Among them is a set of ideas and insights which are quite diverse in detail and yet linked by a common thread. The principal creators of these developments were Wittgenstein and language games, Michael Polanyi and the tacit component of knowledge, Kuhn and paradigms and normal science, Quine and ontological relativity and Feyerabend (the very embodiment of the spirit of anything goes) and theory-dependent observations.63 There were others, most recently perhaps Richard Rorty. What all these have in common, to a greater or lesser extent and in some cases only by implication, is the idea of a theory-laden fact, i.e. the idea that our observation, what we see manifestly happening before our eyes, is influenced by our presuppositions, including especially beliefs to which we are firmly committed. We do not come to know things as starkly unique. We can know things (as Aristotle had said) only by their attributes which, taken one by one, are also the attributes of other things. What we see has deep roots in our previous experience of these other things, in our view of what the world is like or, more generally, in our presuppositions which may be based on our personal experience but more likely are acquired (like the first language we learnt) in discourse with the people with whom we associate. What we see must be intelligible to us when our presuppositions are taken into account. Hence our observations may be said to be theory laden. However, what has this to do with Menger? Well, let us suppose that we could still speak to Menger or to a Menger clone who thinks just like him. We would ask him how he knew that his exact laws were true; how it is that a rational or intellectual grasp must eventually lead to the truth. We may produce another person who claims that he has been led by a rational grasp to exact laws which are incompatible with Menger’s. We would then ask them how they could ever tell which of them, if either, has really reached the truth. After carrying on like this for a week, a month, perhaps some years, Menger may eventually be reduced to saying: ‘I am tired of talking about truth. But I still maintain that my exact laws make events more intelligible not only to me but to the business people I have spoken to. Furthermore, I am sure that these people would regard the exact laws as the truth.’ By then, his whole enterprise would for most purposes be quite compatible with the thesis of theory-laden facts. In other words, but for the classical realist conviction—and of course it is a vast difference—there seem to be some remarkable parallels between Aristotelian essentialism and the thesis of theory-laden facts. The structure that makes for coherence is no longer in the phenomena but in the facts as they are perceived. In reading the philosophers mentioned above, one gets the impression (or at least the present writer does) that, after centuries of problems and doubts about the concept of truth, there are people who have returned to an ancient way of thinking, simply having dropped the embarrassing baggage that used to go with it. It is a more modest venture to examine one’s presuppositions and those of people one may talk to or whose books one may read. For these purposes, it does not matter whether the presuppositions are true in some ultimate sense. Much less does one have to commit oneself either to classical realism or to nominalism, to a view on what kind of entities there are. One may be ontologically modest. It does no harm to be agnostic on these matters. It would therefore still be possible to follow something like Menger’s programme of exact economics, but a version without essentialism which would have the aid of a wealth of insights and ideas in the literature on (what for the sake of brevity has here been subsumed under) theory-ladenness. Such a programme would still be able to clarify, if not enlarge, our vision without the pretence of being able to show how everything is determined. In particular, it could investigate the question of whether there are common paradigms in terms of which economic and social problems come to our attention; and if so, whether they really are incommensurable with the paradigm of classical mechanics as, mutatis mutandis, Menger claimed they were. It could, for instance, take up Shackle’s challenge: How to find a scheme of thought about the basic nature of human affairs, which will include decision in the meaning we give to this word in our unself-conscious, intuitive, instinctive attitude to life, where without examination or heart-searching we take it for granted that a responsibility lies upon us for our acts; that these acts are in a profound sense creative, inceptive, the source of historical novelty ...64 Of course, it would not be, and Menger never intended it to be, the whole of economics. It would be that part of it only which might have the potential of clarifying and of lessening confusion and talking at cross purposes. Obviously, there would be those who would regard this as a task fit only for a Shakespearean court jester, and not for an economic scientist. A more serious charge against Menger’s whole scheme is that the interest in everyday terms is misplaced since most of the great advances in physical science were couched in terms largely unintelligible to the layman. This charge is not so easily answered. However, it must be remembered that these advances in physical science came after a centuries-long process of distilling from the one time jumble of changes in those elements and their interconnections which were to be regarded as physical and mechanical. Perhaps Menger regarded this process as one of discovering exact laws. In this regard, a remark by a leading mathematician and physical scientist of yesteryear is most interesting. Discussing the role philosophical questions had played in the innovations in physics of his time, Sir Arthur Eddington ascribed the discomfort this had engendered among most physicists to ‘a tendency to treat the mathematical development of a theory as the only part which deserves serious attention’. He continued: ‘But in physics everything depends on the insight with which the ideas are handled before they reach the mathematical stage.’65 Footnotes 1 Wieser (1929, pp. 111–12 and 123–25). 2 Wieser (1929, pp. 113–19). 3 See, for instance, Lachmann’s comments in his review of Hicks’s ‘Capital and Time’ (Lachmann, 1973, pp. 196 and 203–7). 4 Howey (1960, p. 142). 5 Wieser (1929, p. 125). 6 See, for instance, Menger ( 1934, pp. xliii–xliv;  1933, p. 81;  1935, p. 23;  1935, p. 211). 7 Hutchison (1973, p. 31). He was citing Emil Kauder, who drew the conclusion from Menger’s papers in Tokyo. 8 See Kauder (1965, pp. 95–100) and Hutchison (1973, pp. 18ff). 9 Kauder (1965, p. 95). 10 Menger ( 1934, p. xiv), repeated in Menger ( 1933, p. 41). 11 Translation by Benjamin Jowett. 12 Menger ( 1933, pp. 4, 12, 325–27, 41, etc.). 13 Menger ( 1935, pp. 197–99). 14 Menger ( 1933, pp. 5 and 33). 15 Ibid., p. 4. 16 Menger ( 1933, p. 14). 17 Posterior Analytics II. 2, translation by G. Mure. In one place, Menger ( 1933, p. 87) states the maxim: Scire est per causas scire. 18 See also Alter (1982, pp. 152–53). 19 Menger ( 1933, footnote on pp. 6 and 7). See also Winch (1972, p. 331). 20 Menger ( 1934, p. xiv). 21 Menger ( 1935, pp. 189–92). 22 Menger ( 1933, footnote on p. 260). 23 Menger ( 1935, p. 34). 24 Wieser (1929, p. 17ff). He was criticizing Schumpeter, who also insisted in the unity of the sciences, but from a nominalist standpoint, in his ‘Wesen und Hauptinhalt’. 25 Menger ( 1934, p. xiv). 26 Ibid., p. xlvi. 27 Mittermaier (1986) goes into this question at greater length. 28 Streissler (1973). 29 Ibid. The quotations, in order, are from pp. 160–61, 166, 169 and 172. 30 Ibid., pp. 174–75. 31 Menger (1973, p. 45). 32 Hutchison (1953, p. 148), where the extract quoted is from Jaffe (1935, p. 200). See also Hayek in the introduction to the LSE reprint of the Grundsätze (Menger,  1934, p. ix), Kauder (1965, p. 901), Jaffe (1976, p. 521), Lachmann (1977, p. 481), Streissler (1973, p. 174) and Winch (1972, p. 330). 33 Menger ( 1933, p. xx). 34 Ibid., for instance, pp. 18 and 3. See also Winch (1972, p. 330). 35 Ibid., p. 25 and chapters 4 and 5. In the preface to Grundsätze, he had said that the reduction to types, etc. is properly called the empirical method. 36 ‘Regelmässigkeiten in der Coexistenz und Aufeinanderfolge.’ Ibid., first mentioned on p. 4 and then throughout the book. 37 ‘Volle empirische Wirklichkeit.’ Ibid., pp. vii and 341 and thereafter. He also spoke of concrete phenomena. 38 Ibid., p. 35. 39 Ibid., pp. 34–38 and 46–48. 40 At the end of the Posterior Analytics (Book II, Ch. 19), Aristotle discusses induction, intuition and primary premises in an uncharacteristically vague manner. 41 Menger ( 1933, p. 40). The principle states that whatever has been observed once would occur again under exactly the same conditions. 42 Ibid., footnote on pp. 6 and 7. 43 Ibid., p. 531. 44 Ibid., p. 42. See also pp. 25 and 401. 45 Ibid., pp. 59 and 265. 46 Menger ( 1935, p. 105). See also Menger ( 1935, pp. 190, 192 and 199), where he speaks of inner connection and inner causation (innerer Zusammenhang, innere Verursachung). 47 Menger ( 1933, p. 59). 48 Ibid., p. 54. 49 Walras (1874, p. 71). After explaining how he understood the method of geometry, Walras continued: ‘Following the same procedure, the pure theory of economics ought to take over from experience certain type concepts, like those of exchange, supply, demand, market ... From these real-type concepts the pure science of economics should then abstract and define ideal-type concepts in terms of which it carries on its reasoning. The return to reality should not take place until the science is completed and then only with a view to practical applications.’ Menger may have had misgivings about the fact that phenomenal forms were described as concepts, but he could not have objected much to the way Walras put this. That their legacies are so different is probably due to the circumstance that Walras’s theory was taken over by Pareto, who had strong nominalist convictions, whereas Manger’s value theory was established by men who appear not to have had strong philosophical views at all. 50 Menger ( 1933, p. 265). See also the third paragraph of Section 6 above. 51 Menger ( 1934, p. xlvi). 52 Menger ( 1933, p. 259). 53 Menger ( 1934, p. xlvii). 54 Menger ( 1933, pp. 73–74). 55 Menger ( 1960, p. 84). 56 Hutchison (1973, pp. 22–23). 57 Menger ( 1933, pp. 75–77). 58 See Borch (1973, p. 61). 59 He even mentioned the limitation (Menger,  1935, pp. 27–29) in a two- to three-page summary of the Untersuchungen. 60 This is the third of four senses of the word empirical given in the Oxford English Dictionary. The first two are similar to it but applied only to medicine. The fourth is the sense common in economic discussion. 61 Menger ( 1933, p. 33). In this place he also showed what he meant by prediction: We may achieve control over something when we are in a position to set what we know from our theories as its rational requirements. 62 Shackle has on a number of occasions vividly outlined the peculiar conditions marginalist theory has to assume. See, for instance, Shackle (1972, pp. 592–95 and 599–602). 63 The works in which the questions of interest here were first brought up are: Wittgenstein (1953), Polanyi (1958), Kuhn ( 1970), Quine (1960, 1969) and Feyerabend (1981). 64 Shackle (1966, p. 73). 65 Eddington (1939, p. 55). Bibliography Alter, M. 1982. Carl Menger and Homo Oeconomics: some thoughts on Austrian theory and methodology, Journal of Economic Issues , vol. 16, 149– 60 Google Scholar CrossRef Search ADS Borch, K. 1973. The place of uncertainty in the theories of the Austrian school, pp. 61– 74 in Hicks, J. R. and Weber, W. (eds.), Carl Menger and the Austrian School of Economics , Oxford, Clarendon Eddington, Sir A. 1939. The Philosophy of Physical Science , Cambridge, Cambridge University Press Feyerabend, P. K. 1981. An attempt at a realistic interpretation of experience, pp. 17– 36 in Realism, Rationalism and Scientific Method, Philosophical Papers , vol. 1, Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Hicks, J. R. and Weber, W. (eds.) 1973. Carl Menger and the Austrian School of Economics , Oxford, Clarendon Howey, R. S. 1960. The Rise of the Marginal Utility School 1870–1889 , Lawrence, University of Kansas Press Hutchison, T. W. 1953. A Review of Economic Doctrines 1870–1929 , Oxford, Clarendon Hutchison, T. W. 1973. Some themes from Investigations into Method, in Hicks, J. R. and Weber, W. (eds.), Carl Menger and the Austrian School of Economics , Oxford, Clarendon Jaffe, W. 1935. Unpublished papers and letters of Leon Walras, Journal of Political Economy , vol. 43, 87– 207 Google Scholar CrossRef Search ADS Jaffe, W. 1976. Menger, Jevons and Walras de-homogenized, Economic Inquiry , vol. 14, 511– 24 Google Scholar CrossRef Search ADS Kauder, E. 1965. A History of Marginal Utility Theory , Princeton, NJ, Princeton University Press Google Scholar CrossRef Search ADS Kuhn, T. S.  1970. The Structure of Scientific Revolutions , 2nd ed., enlarged, Chicago, University of Chicago Press Lachmann, L. M. 1973. Sir John Hicks as a neo-Austrian, South African Journal of Economics , vol. 41, 195– 207 Lachmann, L. M. 1977. Capital, Expectations and the Market Process: Essays on the Theory of the Market Economy , Kansas City, Sheed Andrews & McMeel Menger, C.  1934. Grundsätze der Volkswirthschaftslehre, The Collected Works of Carl Menger , vol. 1, LSE Reprints Menger, C.  1933. Untersuchungen über die Methode der Socialwissenschaften und der Politischen Oekonomie insbesondere, The Collected Works of Carl Menger , vol. 2, LSE Reprints Menger, C.  1960. Problems of Economics and Sociology , translation of Menger 1883 by F. J. Nock, Champaign, University of Illinois Press Menger, C.  1935. Die lrrthümer des Historismus in der deutschen Nationalökonomie, The Collected Works of Carl Menger , vol. 3, pp. 1– 97, LSE Reprints Menger, C.  1935. Zur Kritik der Politischen Oekonomie, The Collected Works of Carl Menger , vol. 3. pp. 99– 131, LSE Reprints Menger, C.  1935. Grundzüge einer Klassifikation der Wirtschaftswissenschaften, The Collected Works of Carl Menger , vol. 3, pp. 185– 218, LSE Reprints Menger, K. 1973. Austrian marginalism and mathematical economics, pp. 38– 60 in Hicks, J. R. and Weber, W. (eds.), Carl Menger and the Austrian School of Economics , Oxford, Clarendon Google Scholar CrossRef Search ADS Mittermaier, K. 1986. Mechanomorphism, pp. 236– 51 in Kirzner, I. (ed.), Subjectivism, Intelligibility and Economic Understanding. Essays in Honor of Ludwig M. Lachmann on His Eightieth Birthday , New York, New York University Press Google Scholar CrossRef Search ADS Polanyi, M. 1958. Personal Knowledge: Towards a Post-Critical Philosophy , London, Routledge Quine, W. V. 1960. Word and Object , Cambridge, MA, MIT Press Quine, W. V. 1969. Ontological Relativity and Other Essays , New York, Columbia University Press Shackle, G. L. S. 1966. The Nature of Economic Thought , Cambridge, Cambridge University Press Shackle, G. L. S. 1972. Marginalism: the harvest, History of Political Economy , vol. 4, 587– 602 Google Scholar CrossRef Search ADS Streissler, E. 1973. To what extent was the Austrian school marginalist?, in Black, R. et al. (eds.), The Marginal Revolution in Economics , Durham, NC, Duke University Press Google Scholar CrossRef Search ADS Walras, L. 1874. Elements of Pure Economics , translated by W. Jaffe, London, Allen & Unwin Wieser, F. 1929. Gesammelte Abhandlungen , edited by F. A. Hayek, Tübingen, Mohr Winch, D. 1972. Marginalism and the boundaries of economic science, History of Political Economy , vol. 4, 325– 43 Google Scholar CrossRef Search ADS Wittgenstein, L. 1953. Philosophical Investigations , translated by G. Anscombe, New York, Macmillan © The Author(s) 2017. Published by Oxford University Press on behalf of the Cambridge Political Economy Society. All rights reserved.
Cambridge Journal of Economics – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud