intermediary—stands out with unmistakable clearness in Say’s Traité. Among leading English authors, Lauderdale, Malthus, and Senior came more or less near to grasping this idea. But only Say made something like a success of it. It is nothing short of pathetic that, owing to a complete lack of understanding on the part of opponents and owing to complete ignorance of even the most elementary mathematical tools on the part of exponents, this promising start not only was left to hibernate for decades but also acquired a reputation for superficiality and sterility. On the other hand, we have the type of analysis of which the Ricardian detour is the outstanding example. It would of course be an exaggeration to say that Ricardo was entirely blind to the aspect of the economic process described above. He had glimpses of it now and then, and Professor Knight went perhaps too far if he accused Ricardo of not having seen the problem of distribution as a problem of valuation at all. 28 But it is true that Ricardo failed to see the explanatory principle offered by the valuation aspect. This failure is intimately related to a peculiarity of the Ricardian work that is essential for understanding him and proves better than does anything else that this work constitutes in fact a detour and falls out of the historical line of economists’ endeavors. For A.Smith, A.Marshall, and ourselves, the factors that explain the size and rate of change of the Social Product or National Dividend or Total Net Output are of primary significance. This was not Ricardo’s view. On the contrary, in his preface to the first edition of the Principles, he tells us: ‘To determine the laws which regulate this distribution [of total product between landowners, capitalists, and laborers, J.A.S.] is the principal problem in Political Economy.’ That is to say, he all but identifies economics with the theory of distribution, implying that he had little or nothing to say about—to use his language—‘the laws which regulate total output.’ This is a strange view to take, though it must be added at once that he did not always adhere to it, as his chapters on foreign trade and on machinery show. It enables us, however, to state the fundamental problem Ricardo wanted to solve in terms of an equation between four variables: net output equals rent plus profits plus wages (everything measured in Ricardian values, see below, ch. 6, sec. 2a). And it does still more for us. It rids us of one of these four variables. For, since we have nothing to say about total net output, we can accept its amount, whatever it is, as a datum. So we really start with an equation that contains only three variables. But one equation in three variables is still a hopeless business. Therefore, Ricardo (ch. 2) places himself on a margin of agricultural production where rent is zero. Observe carefully what this means for Ricardo’s analytic set-up. By numberless writers the West-Ricardo theory of rent has been discussed in isolation, and with nothing but the question in mind whether it was ‘right’ or ‘wrong.’ This question is completely pointless. The West-Ricardo theory 28 See F.H.Knight, ‘The Ricardian Theory of Production and Distribution,’ Canadian Journal of Economics and Political Science, vol. I, February 1935. But Knight supported his indictment effectively by a quotation from a letter of Ricardo’s to McCulloch to the effect that ‘the proportions in which the whole produce is divided between landlords, capitalists and labourers…are not essentially connected with the doctrine of value’ (ibid. p. 6n.). This is not true even from Ricardo’s own standpoint. [Substantially this same footnote appears above, sec. 3. J.A.S. had taken out this section (5) for revision (his notes indicated dissatisfaction with the looseness of the argument) just before his death.] History of economic analysis 542 of rent cannot be discussed in isolation, that is, without reference to the whole of the West-Ricardo system. It is within this system only that it acquires analytic meaning and, in fact, owing to Ricardo’s inability to deal with systems of simultaneous equations, imposes itself. Outside of the West-Ricardo system considered as a whole, it has very little meaning and is hardly worth bothering about. Let us go on. That theory of rent having fulfilled its only purpose, which is to get rid of another variable in our equation, we are left, on the margin of production, with one equation and two variables—still a hopeless business. But, so it occurred to Ricardo, wages are not really a variable either, at least not within that equation. He thought he knew, from external considerations, what they will be in the long run: here the old Quesnay theory comes in, reinforced by Malthus’ law of population—the wages will be roughly equal to what is necessary to enable ‘the labourers, one with another, to subsist and to perpetuate their race without either increase or diminution.’ And so we reach the blessed goal at last: profit, the only variable left, is determined too. Call this patchwork ingenious, if you so please, but do not deny that it is patchwork—and rather primitive patchwork at that. Marx’s schema is open to a similar objection. 29 He also eliminated rent from the fundamental problem, though in a different way. His equation of distribution, in terms of Marxist values, then reads: net output equals wages plus surplus. Again we can take the net output as a datum. And again the surplus is a residual, the determination of which depends upon the external considerations that determine wages. J.S.Mill’s system, on the contrary, absorbed enough of the Say conception—and in addition was sufficiently helped by Senior’s notion of abstinence—to be free from any such objection, and it offered all the elements of the complete model that Marshall was to build. But he retained so many Ricardian relics that there is some excuse for Jevons’ and the Austrians’ not seeing that they were developing his analysis and for believing instead that they had to destroy it. 6. THE ‘CLASSIC’ CONCEPTION OF ECONOMIC DEVELOPMENT I have tried to explain above (Part I, ch. 4, sec. 1d) the meaning and role of what I have called Vision—that first perception or impression of the phenomena to be investigated which factual and ‘theoretical’ analysis, in an endless relation of give and take, then work up into scientific propositions. But when we are concerned with nothing more ambitious than to formulate the way in which—on the plane of pure logic— economic quantities ‘hang together,’ that is, when we are concerned with the logic of static equilibrium or even with the essential features of a stationary process, the role of Vision is but a modest one—for we are really working up a few pretty obvious facts, perception of which comes easily to us. Things are very different when we turn to the task of analyzing economic life in its secular process of change. It is then much more difficult to visualize the really important factors and features of this process than it is to formulate their modi operandi 29 On Marx’s elimination of rent from the final problem, see above, ch. 1, sec. 4; on his equation of distribution in terms of Marxist values, see below, ch. 6, sec. 6g. General economics 543 once we have (or think we have) got hold of them. Vision (and all the errors that go with it) therefore plays a greater role in this type of venture than it does in the other. This may be illustrated by the Stagnationist Thesis of our own time, that is, the notion that the capitalist system has spent its powers; that the opportunities of private enterprise are giving out; that our economy is, amid convulsions, settling down to a state of Secular Stagnation or, as some prefer to call it, Maturity. No doubt facts and arguments have been collected in order to establish this notion, which has also been embodied in theoretical models. But it should be obvious that these facts and arguments rationalized a pre- existing vision or impression which they would have been powerless to create, if for no other reason, because the relevant observations extend over a period that is much too short and was much too much under the influence of clearly abnormal events to warrant any conclusions or predictions of that kind. Economists’ visions were in no better case a century or so ago. We shall consider three types of vision of the economic future of humanity that the writers of the period under survey tried to formulate and to establish. In other words, we shall consider three types of theories of economic development. The first type, associated primarily with the names of Malthus, West, Ricardo, and James Mill, fully justifies their being labeled ‘pessimists.’ Its wellknown features were: pressure of population, present already but still more to be expected; nature’s decreasing response to human effort to increase the supply of food; hence falling net returns to industry, more or less constant real wages, and ever-increasing (absolutely and relatively) rents of land. We are now not concerned with the manner in which these ‘classics’ gave analytic effect to this vision of theirs, that is, with the manner in which they formulated their ‘laws’ of population, decreasing returns in agriculture, and so on, and with the analytic use they made of them. To this we shall attend in the next chapter. Here we are concerned only with what they thought they saw, that is, with the vision that was at the back of their analysis—or, if you prefer, with their preconceptions. The most interesting thing to observe is the complete lack of imagination which that vision reveals. Those writers lived at the threshold of the most spectacular economic developments ever witnessed. Vast possibilities matured into realities under their very eyes. Nevertheless, they saw nothing but cramped economies, struggling with ever- decreasing success for their daily bread. They were convinced that technological improvement and increase in capital would in the end fail to counteract the fateful law of decreasing returns. James Mill, in his Elements, even offered a ‘proof’ for this. In other words, they were all stagnationists. Or, to use their own term, they all expected, for the future, the advent of a stationary state, which here no longer means an analytic tool but a future reality. Apparently, J.S.Mill was in better case. He dropped all ‘pessimism,’ and he was even intelligent enough to realize that there was no reason to look upon the future of the masses ‘as otherwise than hopeful.’ However, this was only because he believed—as other Malthusians such as Chalmers had believed before him—that mankind was learning the Malthusian lesson and was about to restrict propagation voluntarily so that the race between capital and population would be won by the former. In this he proved himself a better prophet than did others. But he had no idea what the capitalist engine of production was going to achieve. On the contrary, toward the end of his life (around 1870) he really became a stagnationist in the modern sense, believing that the private-enterprise economy had pretty much done what it was able to do and that a stationary state of the economic History of economic analysis 544 process was near at hand. But there is this difference between him and our own stagnationists. He did not, as A.Smith and Ricardo had done, view the stationary state with misgivings (Principles, Book IV, ch. 6), because he had eliminated the bogey of overpopulation. But neither did he share the modern stagnationist’s misgivings, because he did not fear the bogey of underconsumption. To him the stationary state looked rather comfortable—like a world without ‘bustle’ (his term) in which a philosopher like himself would not mind living and in which there would be moderate prosperity (or better) all round. 1 The question, whether the social structure of capitalism could persist in circumstances in which the main function of the capitalist entrepreneur was being lost, we may answer for him by saying that he visualized the advent of the stationary state as a very gradual process so that institutions and minds would have no difficulty in making the currently necessary adjustments. In agreement with all the English ‘classics’—perhaps we might say, with the spirit of his age—he greatly underrated the importance in economic de-velopment of the element of personal initiative and, correspondingly, he greatly overemphasized the importance of mere increase in physical producers’ goods. And in this again, he overemphasized the importance of saving. 2 Accepting the Turgot-Smith theory of the investment process, he took it for granted that the important thing was to have something to invest: the invesment itself did not present additional problems either as to promptness—it was normally sure to be immediate—or as to direction—it was sure to be guided by investment opportunities that were equally obvious to all and existed independently of the investing man. 3 Saving, then, was the powerful lever of economic development. And it never created obstructions; the saving act itself did not, since the sum saved was immediately spent on productive labor; the resulting expansion of productive capacity did not, since products of correctly planned production were always capable of being sold at cost-covering prices. 4 To use our own term, J.S.Mill’s schema of economic development, like Say’s, was essentially hitchless. Malthus’ and Sismondi’s schemata are examples of hitchbound ones, the hitches arising in both cases not so much from saving per se, but from the resulting increase in productive capacity. Ricardo’s also was hitchbound, but for another reason, namely, the reason enshrined in his interpretation of the law of diminishing returns. 1 This stationary state was a state of a peculiar kind and did not conform to the definition given above. It did not quite exclude technological progress or increase in capital. It was really stationary with respect to population only, it being assumed that this would cause everything to go on more quietly. 2 Of course we may define saving in such a way as to make that statement meaningless. But I mean by saving (or thrift) the distinctive phenomenon we all know (unless we are too familiar with the economic theory of the 1930’s) and therefore consider it as a factor in the process of the accumulation of physical capital goods. My statement then means that J.S.Mill, like all the authors that followed the Turgot-Smith line, was at fault in believing that thrift was the all-important (causal) factor in that process. 3 This mechanistic view, too, was an important element in the economic Weltbild of the ‘classics.’ They were entirely unaware of how great a part of capitalist reality they thereby suppressed in silence. General economics 545 4 The latter proposition is more delicate to handle than J.S.Mill and the economists of his time and line realized. But its opponents were below and not above the limited truth it asserts. Also, though in a particularly narrow way—through saving alone—Mill thus did recognize the most obvious of all truths about capitalist development, viz., that, by virtue of its logic, it tends to raise the standard of life of the masses. The second type of vision of the economic future—the ‘optimistic’ type—can be best illustrated by such names as Carey and List. Whatever we may think of the virtues of their technical analysis, at least they did not lack imagination. They felt intuitively that the dominant fact about capitalism was its power to create productive capacity, and they saw vast potentialities looming in the near future. With less imagination but with plenty of sound judgment, the majority of economists on the Continent refused to share the ‘pessimism’ of the Ricardians and of Malthus. At least, most of them watered it down. But beyond this, it was natural for those who more or less followed Say’s lead so far as technical theory was concerned to realize that neither facts nor analysis were bearing out the Ricardian vision. These were called ‘optimists’ and, partly but not wholly under Marxist influence, there grew up a tradition to despise them as shallow. This view is in fact historically associated with many writers—the Bastiat type—who fully deserve to be so called. But in itself this ‘optimism’ was the result of both a vision and a theory that were more correct than those of the ‘pessimists’: the degree of truth of a doctrine is by no means always positively correlated with the ability of its exponents. 5 The third type of vision of the economic future and of corresponding theories of economic development will be represented by Marx alone. Based upon a diagnosis of the social situation of the 1840’s and 1850’s that was ideologically vitiated in its roots, 6 hopelessly wrong in its prophecy of ever-increasing mass misery, inadequately substantiated both factually and analytically, Marx’s performance is yet the most powerful of all. In his general schema of thought, development was not what it was with all other economists of that period, an appendix to economic statics, but the central theme. And he concentrated his analytic powers on the task of showing how the economic process, changing itself by virtue of its own inherent logic, incessantly changes the social framework—the whole of society in fact. We have already dwelt on the grandeur of this conception; we shall briefly discuss its analytic aspects below. 5 In this and other respects, the case is similar to that of the doctrine of the essential harmony of class interests. On investigation, this doctrine turns out to be only partly tenable—but rather more so than does the doctrine of the essential antagonism of classes. But the latter has been preached with unsurpassable force and, moreover, renders the radical intellectual’s ideology. The former has never been put forcefully or even convincingly. And it does not suit the book of the radical intellectual. So he who holds it is likely to be sneered at as a sort of Caspar Milquetoast, and this is quite as effective as, or more so than, serious argument would be. In the present case, there is something else however. No matter what the reason is, it is a fact that pessimistic views about a thing always seem to the public mind to be more ‘profound’ than optimistic ones. 6 It was pointed out above that Marx took his picture of social reality from the radical ideology of his formative period. History of economic analysis 546 Only two points can be mentioned here. First, nobody—not even the most ardent of optimists with whom Marx had this point in common—had then a fuller conception of the size and power of the capitalist engine of the future. With a quaint touch of teleology, Marx said repeatedly that it is the ‘historical task’ or ‘privilege’ of capitalist society to create a productive apparatus that will be adequate for the requirements of a higher form of human civilization. However much our modern positivism may resent this way of putting it, the essential truth of what he meant to convey, in this respect, stands out clearly enough. Second, the motor of Marx’s economic development was indeed not quite the colorless Saving of J.S.Mill: he linked it—or investment—to technological change in a manner that is not to be found in the latter’s Principles. But all the same the motor is Saving, which with him is as promptly turned into investment as it is with Mill. This fact is hidden but not abolished by Marx’s use of the term Accumulation and by his violent diatribes against the ‘nursery tale’ (Kinderfibel) that physical capital is created by saving. There are good reasons as well as bad ones for Marx’s dislike of the latter term. In particular, capitalist fortunes do not typically arise from saving income dollars and piling them up neatly, but by the creation of sources of returns, the capitalized value of which then constitutes a ‘fortune.’ However, the implications of this he would not have liked any better than he did the picture of good and frugal boys who save till they find themselves rich. So it had to be Exploitation forever, so exclusively in fact as to endanger the explanatory value of his schema: for the social process as a whole, the essential point is in any case the capacity-creating use made of capitalists’ gains—no matter whether or not they arise from exploitation and are invested again for the purpose of further exploitation—and this essential point, so a history of analysis has to notice, is fundamentally the same with Marx and Mill, however different the phraseologies were in which they conveyed it. General economics 547 CHAPTER 6 [General Economics: Pure Theory] 1 [1. AXIOMATICS. SENIOR’S FOUR POSTULATES] TO SENIOR belongs the signal honor of having been the first to make the attempt to state, consciously and explicitly, the postulates that are necessary and sufficient in order to build up—it is misleading to say to ‘deduce’—that little analytic apparatus commonly known as economic theory, or, to put it differently, to provide for it an axiomatic basis. The merit of the attempt is but little decreased by the fact that his list of postulates was incomplete and otherwise defective and by the further fact that he invited attack by defining that ap-paratus so narrowly or else by equating this theory to ‘political economy.’ It is increased by the fact that the attempt occurs in the course of a general theoretical house cleaning and is part of a wider attempt at rigorous conceptualization. First he polished Wealth and (exchange) Value; then he stated his four Elementary Propositions—the postulates; finally he presented, under the inadequate heading of Distribution (Exchange or Value and Distribution would be more adequate), a set of additional concepts and of relations that together with the immediate development of the postulates, which settle most matters usually dealt with under the heading Production, are supposed to make up the theoretical organon. As a venture in pure theory, his performance is clearly superior to that of Ricardo. We shall now consider the postulates, using in doing so every opportunity that may arise of looking further afield. 1 [Section 1 of this chapter was written much earlier than the remainder of the chapter. The typescript was dated December 1943. It was obvious that J.A.S. intended to revise these pages on Senior and to make them the introductory section to this chapter. There were many notes clipped to the early typescript. There was no title for this section and no title for the chapter, but the remaining sections were relatively complete with titles for sections and subsections. This section is presented as written, although it lacks the proper introductory remarks and the revisions J.A.S. would have made.] [(a) The First Postulate.] The first reads as follows: ‘That every man desires to obtain additional Wealth with as little sacrifice as possible.’ 2 Implicitly at least, some such proposition underlies all theoretical reasoning and it would just as well fit into Ricardo’s or Malthus’ texts. Adam Smith and J.S. Mill took it for granted, Lauderdale came near to stating it explicitly. In the language of the next period—in Marshall’s for instance—it can be expressed by saying that every man desires to maximize the difference between the sum total of his satisfactions and the sum total of his sacrifices, both discounted to the present moment. But what is its nature and standing? Senior calls it ‘a matter of consciousness’ and distinguishes it from the three other propositions, which are ‘matters of observation.’ But it would not affect his meaning if we called it a matter of introspective observation. Moreover, there are, in his ‘development’ of this proposition (e.g. pp. 27–8) 3 various comments on the behavior of Dutchmen and Englishmen and Indians of Mexico that are obviously based upon an external observation of sorts. We may therefore provisionally speak of observation even in this case and proceed forthwith to state the following generalization about all four of them or any other proposition an economist may see fit to postulate. In so doing we exemplify—and in part justify—Say’s opinion on economics being an observational science (though he said experimental), which will thus be seen, appearances notwithstanding, to involve no disagreement with Senior. Nobody ever denied—or, by his practice, belied—the truth that economic theory, like any other theory, is founded upon observation. Senior, by taking little trouble with observations and concentrating upon inference from them, may have created a wrong impression and he may himself have held erroneous views about the relative importance of observing and inferring, but he did not in fact—though he did in phrase—treat economics as toto cælo ‘deductive.’ Now, the facts observed enter theory as hypotheses or assumptions or ‘restrictions,’ that is to say, as generalized statements induced or suggested by observation. 4 When we wish to stress our confidence in their validity, we often call them Laws—compare for instance Keynes’s ‘psychological law’ of the propensity to save. When we simply wish to stress our resolve not to challenge them in 2 [The postulates are discussed in Nassau William Senior, An Outline of the Science of Political Economy (1st ed. 1836; 6th ed. 1872; publ. in the Library of Economics, 1938).] 3 [Page references are to the edition published by the Library of Economics.] 4 It should be noticed that this is only one of several meanings of the word Hypothesis. We have met others. The same remark applies to Laws and Principles. General economics 549 the course of a particular argument, we call them Principles. But all these words really mean one and the same thing, and there is no point in philosophizing about them. This applies to frontier facts as well as to facts that belong to our field proper. The difference, as pointed out above, is only that in the first case we do not, in the second we do, feel fully responsible for the validity of our statements about them. Quite another question is whether or not we are to be satisfied with an observation of the Seniorian or, for that matter, the Ricardian or Millian kind. Three aspects of this question must be carefully distinguished if we are to understand ‘classic’ or any other theoretical procedure. There are first the two problems of observation by introspection and by common or everyday experience. Many economists of later times, especially the founders of the so-called Austrian School, have stoutly stood for both. Wieser in particular seems quite in accord with J.S.Mill in accepting common experience as a valid basis for theory to start from. Critics have sometimes gone so far as to rule both out entirely on the ground that introspection and common experience are nothing but cloaks for purely speculative assertions. This extreme form of the criticism is indeed open to the reply that some postulates—such as that businessmen on the whole prefer making money to losing it—are evidently not miles from the truth and that it is vexatious to insist upon elaborate research for the purpose of establishing them. But a less extreme form of the same criticism is not invalidated by such cases. There are others—savings habits for instance—in which introspection and common experience cannot be convincingly invoked; and even where they can, the relative importance and the modus operandi of the facts that enter a postulate may still have to be ascertained by more substantial methods. This opens up the second aspect of our question. Senior’s postulate embodies observation, but possibly inadequate observation. Does this justify us in rejecting everything that is between the covers of his book? Evidently not. The postulate divested of unnecessary utilitarian associations is plausible. All that could be objected to it on grounds such as that he overemphasized selfishness, overestimated the rational element in our behavior, and neglected historical differences in the intensity of the desire for ‘wealth’ at different times and in different places, is amply taken into account in his comments upon his proposition. If we feel misgivings nevertheless, all we have to do is to start appropriate research. Anything else is pure filibustering. So long as the strong prima facie plausibility of the postulate is not destroyed by the results of such research, and so long as particular problems are not specified for which plausibility is not enough and which were nevertheless attacked by those critic-economists, we may indeed feel that Senior’s analysis was primitive—we know that the whole of Senior’s, Ricardo’s, and Mill’s work is primitive, in inference no less than in observation—but we cannot deny its scientific character or call it wrong on principle. The third aspect of our question comes into view when we ask ourselves whether Senior’s first postulate could not be reformulated in a way that would get round the objections that have been or might be raised. But since the economists of that period, if guilty of Psychologism, were certainly much less guilty of it than were the economists of the next period, we had better defer discussion of this point until later. History of economic analysis 550 [(b) The Second Postulate: the Principle of Population.] Senior’s second postulate states the principle of population: ‘That the Population of the world, or, in other words, the number of persons inhabiting it, is limited only by moral or physical evil, or by fear of a deficiency of those articles of wealth which the habits of the individuals of each class of its inhabitants lead them to require’ (op. cit. p. 26). We take this opportunity to touch briefly upon the contribution of Malthus and the discussion that developed about it. Moreover, it will be convenient to add a few remarks about the history of the theory of population in the subsequent period so that we may drop it from our picture in Part IV. This decision is suggested by the fact that its interest for analytic economics greatly declined during the second half of the nineteenth century and that it then grew into a semi-independent science, which it is impossible to deal with in this book. [J.A.S. note: ‘but came back in our time.’] We have seen already that all the facts and arguments that Malthus presented in the first edition of his Essay (1798), down to the details of the analysis as well as of the applications, had been worked out before by so large a number of writers that we may speak of them as widely accepted at the beginning of the nineties. The case therefore differs essentially from the bulk of all those cases, still more frequent in economics than in other sciences, in which a proposition that we associate with an individual name has been anticipated by ‘forerunners.’ This does not amount to a charge of plagiarism or even to a denial of ‘subjective’ originality. But it does reduce Malthus’ contribution to effective co-ordination and restatement. The significance of its tremendous success at the time—with the profession and with political society—is underlined by the fact that, for about a century to come, theory of population was to mean arguments pro and con the Malthusian theory. Also, I have already alluded to the attempts that have been made to account for this success, and for Malthus’ performance itself, by the ideological mechanism. While giving my reason for refusing to accept this explanation, I have, however, admitted that there are two facts that do lend some support to it. The one of them is that the theory was immediately used as an argument against measures of social betterment. William Pitt availed himself of it. Malthus himself published a pamphlet, which is judged mildly if described by no worse term than silly, in which he argued, as Townsend had before him, that the proposal to encourage parishes to build cottages must on no account be entertained because building cottages would encourage early mar-riages (Letter to Samuel Whitbread…1807). In the public mind this sort of thing then assumed the form that the masses had themselves to thank for their economic situation and that nothing much could be done about it. The second fact is this: Malthus himself related that the argument developed in his mind in the course of discussions with his ‘social-minded’ father; and in the subtitle of the first edition of the Essay, he pointed significantly to the General economics 551 . third type of vision of the economic future and of corresponding theories of economic development will be represented by Marx alone. Based upon a diagnosis of the social situation of the 1840’s. similar to that of the doctrine of the essential harmony of class interests. On investigation, this doctrine turns out to be only partly tenable—but rather more so than does the doctrine of the essential. with the looseness of the argument) just before his death.] History of economic analysis 542 of rent cannot be discussed in isolation, that is, without reference to the whole of the West-Ricardo