Marx, Classical Political Economy and the Problem of Dynamics - Henryk Grossman 1941


Right from its origins Classical economic theory was a theory of abstract exchange-value: when it did concern itself with production it dealt merely with the value aspect, and passed over the labour process .[1] Since the rise of the theory of marginal utility and the mathematical school, the analysis of the concrete process of production has been increasingly excluded as an element of theory, and has been used only to establish its preconditions and overall framework. Analysis was directed almost exclusively at the relations between given market variables. It therefore took on a static character and was unable to explain dynamic changes in economic structure. Marx's economic theory marks a fundamental break in principle from both of these tendencies.

The capitalist mode of production is governed by the relation, exchange-value -increase in exchange-value, (M-M'). As a faithful expression of the bourgeois economy, the Classical doctrine was always simply a doctrine of abstract exchange-value.[2] Although Adam Smith did indeed begin his work on the 'Wealth of Nations' by stressing the division of labour as the source of wealth, taken to consist in an abundant supply of the products of labour -use values he nevertheless forgets use-values in the further course of his work ; they are not used in the further development of economic analysis .[3] Although there are presentations of material and structural relations they have an exclusively descriptive character. Smith's theory is one of abstract exchange-value. The social equilibrium between supply and demand, which yields the 'natural price', is solely an equilibrium of value.[4] The same applies to Ricardo. Chapter 20 of his Principles, where he elaborates the distinction between use-value and value, and the importance of 'wealth' as use-values, remains largely unrelated to the rest of the book. Ricardo's entire theoretical gaze is concentrated on the aspect of value (profit), and the use-value of commodities plays no role in his analysis.

The life of the working class depends on the mass of use-values which can be bought with a capital. However, the employer's sole interest is exchange-value, the expansion of exchange-value, i.e. profit, Ricardo expressed this in the now famous dictum, that for the employer who makes 2,000 profit on a capital of 20,000 - 10% -'it is utterly irrelevant whether his capital sets 100 or 1000 people into motion. . as long as in all instances profit does not fall below 2,000'. Whether a given capital employs 100 or 1000 workers depends on the particular structure of the economy. Marx points out that Ricardo is indifferent to this structure: he is only concerned with net revenue (pure profit), the value-surplus of price over costs, and not gross revenue, i.e. the mass of use-values necessary for the subsistence of the working population. For Ricardo these figure only as costs -which are to be pushed down as low as possible. Marx writes: 'By denying the importance of gross revenue, i.e. the volume of production and consumption apart from the value-surplus -and hence denying the importance of life itself, political economy's abstraction reaches the peak of infamy'.[5]

Ricardo's central interest was the theory of distribution: 'To discover the laws which determine distribution is the main problem of economic theory .' (Preface to the Principles). In a letter to Malthus he calls political economy a theory of those laws which govern the proportional division of a given wealth among the various social classes. He regarded the determination of the mathematical relation between the parts of a given totality as 'the only true object of the sciences' .[6] This point of departure renders Ricardo's method a prioristic and deductive: his theories can be deduced from a very small number of premisses. The Classical doctrine is more a system of logical deductions than a researching into and presenting of the real economic relations of the capitalist mode of production. In post-Classical economics this tendency to avoid the real labour-process becomes even more pronounced. In itself the principle of labour as the sole source of value contains a revolutionary element. It implies, as the Classical economists themselves explained, that in the prevailing social order the workers do not receive the full product of their labour, and that rent and profit represent deductions. The egalitarian Ricardians in England merely drew the conclusion implicit in the Classical labour theory of value when they declared that a society in which the workers would receive the full product of their labour would in fact be the only correct and 'natural' one .[7]

The reaction of the right-wing of the Ricardian school to this theoretical turn exercised by the left was to become even more conservative. They scented a threat to class harmony in Ricardo's labour theory of value,[8] and avoided any analysis of the production and labour-process in order to get round the ticklish question of the labour theory of value, and the dangerous consequences it held for distribution and the prevailing social order. Analysis was restricted to market phenomena, exchange: 'Exchange' stated Bastiat, 'is the whole of political economy. '[9] According to Leon Walras, the founder of the Lausanne school, political economy was 'the theory of value and the exchange of value: but he (Walras) forbade us from objectively studying production and distribution' .[10]

Out of fear of ending up in opposition to prevailing property interests, every effort was made to give economic theory the most abstract and formal shape possible, divorced from any qualitative-concrete content .[11] In short, they tried to erect a theory of distribution based on a theory of markets, in order to furnish the proof, by means of a theory of economic coordinates, that all factors of production are rewarded in proportion to the extent to which they are involved in producing the product, and that consequently the workers receive in the wage full recompense for the work done.[12]

A second line of development also began to become apparent at this time. Out of the same need to flee from reality this school urged economic theory onto another terrain, that of psychology. This started with ] .B. Say, who began with the use-value of commodities, but instead of seeing them as physical phenomena, conceived of them as psychological variables - the subjective utility of the object -and constructed a subjective theory of value on these 'services'. Beginning with Say, proceeding through Senior (1836) in England, Dupuits (1844) in France, and H. H. Gossen (1854) in Germany, the subjective theory of value led to the theory of marginal utility as a doctrine of general hedonism, in the course of which political economy's object of analysis shifted from the realm of things and social relations onto the terrain of subjective feelings. 'Boehm-Bawerk's subjective analysis of value contains the most precise and rational calculus of pleasure and pain that has ever been written',[13] as Boehm's special digression on the 'measurability of feelings' shows. The process of production is passed over,[14] and analysis is confined to market phenomena, the explanation of which is sought in human nature.

An even higher level of abstraction is represented by those attempts to make economics into a mathematically 'exact' science, and consequently to disregard any qualitative content in economic phenomena. Market phenomena are one-sidedly regarded as mere 'economic quantities', and where possible they are expressed in mathematical equations. This tendency in modern theory is perhaps formulated most clearly in the works of Joseph Schumpeter .[15] The process of production, as in fact all real relations in the economy, are excluded from analysis. In Schumpeter s view the essence of economic relations consists of the relation 'between economic quantities', which is in fact reduced to the relation of exchange: all other relations between economic variables are neglected as being immaterial.

Summarising this we can say the following: Although theoretical schools and tendencies have changed a great deal in the hundred years since Classical political economy, they possess a common feature which consists of the fact that the real labour process, and the social relations which are entered into in the course of it, have been expunged from their theoretical analysis.[16]

The Marxist critique is directed against the abstract-value mode of study of political economy, as was the critique made by the historical school. But whereas the latter sought to overcome the abstract 'absolute' character of theoretical deduction by means of the external and indiscriminate introduction of concrete historical or statistical material on production, consumption, trade, tax, the position of workers or peasants etc., and thus remained at the level of pure description, denying, in effect, the possibility of the knowledge of theoretical laws, Marx set himself the task of 'revealing the economic law of motion of modern society' .[17] This cannot be done, however, if one abstracts from the 'real world' and merely clings to its one aspect as 'economic quantities. Such a procedure is not political economy, but the 'metaphysics of political economy', which, the more it detaches itself from real objects by way of abstraction, 'the more it fancies itself to be getting nearer the point of penetrating to their core' .[18] Since reality does not consist merely of values, but is rather the unity of values and use-values, Marx's critique begins from the twofold character of economic phenomena, according to which the essential character of the bourgeois form of economy is a product of the specific connection of the valorisation process with the technical labour-process. Of course, subjectively, the sole interest of the businessman is with the value aspect, the valorisation process of his capital, profit. But he can only realise his desire for profit through the technical labour process, by manufacturing products, use-values. And it is the specific character of this labour-process, as the means of meeting the requirements of the valorisation process,[19] which gives the capitalist period its particular stamp. Marx criticises previous economic theory for only looking at individual, isolated sectors, instead of grasping the concrete totality of economic relations.

The monetary system of the Mercantilists merely analysed the circuit of capital within the sphere of circulation in its money-form. The Physiocrats (Quesnay) understood the problem at a deeper level: but they regarded the economic process as an eternal circuit of commodities, as the real production of commodities was not seen as the work of human beings, but Nature. Finally, the Classical economists (Smith, Ricardo) did indeed adopt the production process as the object of their analysis, but only as a valorisation process: this eventually placed them on the same path, skirting around production, and with the same formula, as Mercantilism[20] In contrast to his predecessors Marx stressed the crucial importance of the production process, understood not simply as a process of valorisation, but at the same time as a labour-process ; this does not mean, however, that the two other forms of the circuit of capital, as money and commodity, can be ignored. Capitalist reality is a unity of circuits: the process of circulation (both money and commodity), and the process of production (as the unity of the valorisation and labour-process). Only by being this unity of the labour- and valorisation-process does the production process, in Marx's view, constitute, 'The basis, the starting-point for the physiology of the bourgeois system-for the understanding of its internal organic coherence and life process.'[21] On the other hand, if the production process is merely understood as a valorisation process -as the Classicals understood it - it has all the characteristics of 'hoarding', becomes lost in abstraction, and is no longer capable of capturing the real economic process.[22]

Because Ricardo s categories of value are the expression-if one-sided-of concrete reality, namely the valorisation process, they are taken over by Marx in their basic principles and developed further. However, at the same time he modifies them by complementing their exclusively abstract value character with the material aspect, and elaborates their dual character. Marx's critique of Ricardo's categories of value, and the changes he made, closely resembles Marx's critique and transformation of Hegel's dialectic .[23] Both exhibit the same basic feature, being directed against the abstract and final character shared by Ricardo's categories of value and Hegel's dialectic, because each of these abstracts from the 'real characteristic form'. In his critique of Hegel's dialectic Marx compares, in characteristic fashion, the logic with which Hegel begins the Encyclopaedia, with money and value: it is the logic of 'money of the spirit' and the 'conceptual value of people and of nature, because it 'is utterly indifferent to all real forms' and has become 'abstract thought, abstracting from nature and real people' .[24] This is similar to the way in which money represents the 'least real' form of capital, and how capital has reached the 'pure fetishistic form' in interest-bearing capital, 'in which all the different forms . . . are obliterated, and it exists . . . as independent exchange-value'.[25]

This crucial philosophical position is also brought into play by Marx within political economy: the abstract study of value obscures the 'real forms', the qualitative content of the concrete labour process, which express the specific, differentiating features of the capitalist mode of production.

These can only be grasped by demonstrating the specific connection of the valorisation process to the technical labour-process in each particular period of history.[26] The 'value-form', whose final shape is the money-form, is completely without content.[27] The category of exchange-value leads an 'antedeluvian existence'.[28] One can find exchange-values in ancient Rome, in the Middle Ages and in capitalism; but different contents are hidden behind each of these forms of exchange-value.[29] Marx stresses that 'exchange-value' detached from the concrete relations under which it has arisen is an unreal abstraction, as exchange-value 'can never exist except as an abstract, one-sided relation to an already given concrete and living whole'. The use of the expression 'exchangevalue' presupposes, 'a population which produces under specific conditions'.[30] Of course, 'political economy is not technology .[31]] The point is not however to study the valorisation process separated from a particular labour-process, which makes up its basis and with which it constitutes a unitary whole. 'The concrete is concrete because it is the concentration of many determinations, hence unity of the diverse'. The task of science consists in the 'reproduction of the concrete via thought'.[32] just as the archaeologist reconstructs the entire skeleton, and even the supposed muscles and movements, of an animal from a few excavated bones, so Marx reads off the necessary tendencies of capital which are specific to an epoch from the structure of the labour-process in a particular period, and the type of tools used in it. This is possible as 'technology reveals the active behaviour of people to nature, the immediate production process of their lives, and hence their social relations'.[33] 'The hand-mill gives you society with the feudal lord: the steam-mill society with the industrial capitalist' .[34] Since social relations are closely bound up with the forces of production, changes in the tendencies of capital can be read from changes in these forces.

The best illustration of Marx's theoretical thought is provided by Chapters 14 and 15 of Volume I of Capital, the chapters on 'Manufacture' and 'Machinery and Large-Scale Industry'.[35] These chapters are by no means historical-descriptive depictions, whereby Marx seeks to provide a genetic exposition of the development of large-scale industry out of manufacture. Both chapters have an eminently theoretical character, which is proven by the fact that they are merely sub-sections of the part of Capital dealing with the 'Production of Relative Surplus-Value. What then characterises manufacture and 'machino-facture', large-scale industry as two different phases of capitalist production? Both have a capitalist character, both are based on wage-labour, and both are governed by the search for profit. However, since the technical labour-process is completely different in each, manufacture representing a 'productive mechanism whose organs are human beings',[36] in contrast to which modern large-scale industry is based on machines, this difference serves to distinguish the different phases of capitalism. The example of the derivation of these objective tendencies of capital from the analysis of the concrete labour-process and its instruments -machinery -is intended to illustrate the key distinction between Marx and other theoretical tendencies in the study of economic processes: later we shall analyse the additional consequences which arise from this method of study for the problem of crises and (economic) dynamics.

Whereas changes in the mode of production during manufacture begin with labour, in large-scale industry they proceed from the instruments of labour, machinery.[37] The process is as follows: machinery makes muscle-power dispensable, and thus facilitates the introduction of women and children into the production process on a massive scale. The price of labour-power is lowered and surplus-value increased, as the wages for the entire 'parcellised' family, for a greatly increased amount of work, are no higher now than they were previously for the individual bread-winner alone. The degree of exploitation of labour increases enormously.[38] In addition, the tendency towards the employment of young people and the simultaneous strengthening of the despotism of capital through the extensive employment of women and children works to break down the resistance previously put up by the male workers .[39] The material consumption of the machinery, which represents a large capital-value and which must be depreciated and have interest paid on it, does not only occur through use, but also through its non-use, as a result of the destructive effects of the natural elements. This explains the capitalists' tendency to make work continue day and night, a tendency reinforced by the fact that every new invention threatens to devalue the machinery; hence the capitalists' striving to minimise the danger of the 'moral' wear and tear of the machinery by reducing the period in which it produces its total value.[40] 'Hence too the economic paradox that the most powerful instrument for reducing labour-time suffers a dialectical inversion and becomes the most unfailing means for turning the whole lifetime of the worker and his family into labour-time at capital's disposal for its own valorisation.'[41]

A further incentive to the prolongation of labour-time is therefore the possibility of saving on the otherwise normal expansion of the scale of production through buildings. An expansion in the scale of production without these additional outlays implies an increase in the mass of surplus-value, with a simultaneous reduction in capital expenditure per unit of commodity produced, which further increases the mass of profit .[42]

Machinery leads to the tendency to intensify work, and in particular in those areas where the resistance of the working class has made the excessive prolongation of the working day impossible because of legal prohibitions. In the factory, 'the dependence of the worker on the continuous and uniform movement of the machine creates the strictest discipline' .[43] The increased speed of the machinery forces the worker to a greater degree of attentiveness and activity .[44]

This is the point at which the tendency towards a falling rate of valorisation and the creation of an industrial reserve army begin to play a role. At a higher level of capitalist development, and with the universal application of machinery, this machinery, the use of which has the task of enlarging relative surplus-value, and hence the mass of surplus-value, begins to operate in the opposite direction, i.e. towards a fall in the rate of valorisation. This is because the mass of surplus-value which can be obtained is dependent on two factors: the rate of surplus-value, and the 'number of workers simultaneously employed' .[45] In his hunt for an increase in relative surplus-value the capitalist is driven to the constant development of the productivity of labour through an increased use of machinery in relation to living labour; and he can only 'attain this result by diminishing the number of workers employed by a given amount of capital'.[46].A portion of the capital which was previously variable, and produced surplus-value, is progressively transformed into constant capital, which produces no surplus-value. The result is shown in the creation of a superfluous working population, and the tendency towards a reduction in the mass of surplus-value attainable in relation to the size of the capital employed. 'Hence there is an immanent contradiction in the application of machinery to the production of surplus-value, since, of the two factors of the surplus-value created by a given amount of capital, one, the rate of surplus-value, cannot be increased except by diminishing the other, the number of workers .'[47] Finally, Marx stresses the dynamic impulses which flow from the use of machinery. Whereas manufacture traditionally 'sought to retain the form of the division of labour which it found',[48] and was consequently unable to seize hold of society to its full extent, and change it in depth,[49] large-scale industry based on machinery is forced, by the fall in the rate of profit, continually to revolutionise the technology of the labour process, and therefore the structure of society.

[1] 'The crux of any theory of economic events is composed of theories of value and interest . . . and four fifths of theoretical economic literature consists of research into or controversies about these subjects'. Joseph Schumpeter, Eugen von Boehm-Bawerk in Neue oesterreichische Biographie Vienna 1935 Vol. I I p. 67.

[2]Marx consequently speaks of the 'accentuation of quantity and exchange value' by the Classical economist, in striking contrast to the writers of classical antiquity (Plato, Xenophon), 'who are exclusively concerned with quality and use-value'. (Capital I p. 486).

[3]Cf. Elster: Smiths Lehre and die Lehren der sogenannten 'Klassiker der Volkswirtschaftslehre' in Elster Voerterbuch der Volkswirtschaft Jena 1933 Vol. III p. 213. See too: G. H. Bousquet Essai sur l'evolution de la pensee economique Paris. 1927. p. 199 and Gunnar Myrdal, Das politische Element in der .nationaloekonomischen Doktrinbildung Berlin, 1932 p. 95.

[4]Elsters Smith Lehre.

[5]Marx, MEGA p. 514 et seq.

[6]'Political Economy you think is an inquiry into the nature and causes of wealth -I think it should be called an inquiry into the laws which determine the division of the produce of industry amongst the classes who concur in its formation. No law can be laid down respecting quantity but a tolerably correct one can be laid down respecting proportions'. Letters of David Ricardo to Malthus 1810-1823 ed. Bonar, Oxford, 1887. Letter of 9 October 1820.

[7]See the sharp formulation of the workers' rights to the full product of labour in Hodgskin's Labour defended against the Claim of Capital. By a Labourer. London 1825.

[8]See for example the work by Charles Knight, The Rights of Industry, Capital and Labour 1831, which attacks all opponents of the prevailing rights of property, including Hodgskin, and characterises them as 'enemies of the people', 'destroyers' and 'servants of despoliation'. Carey formulated this view the most clearly a little later: 'Ricardo's system is one of discord . . . . It has a tendency to create animosity between the classes . . . . His book is a handbook of demagogues who seek power by the confiscation of land, war and plunder. Carey The Past, the Present and the Future Philadelphia, 1848. p. 74-5.

[9]Bousquet op. cit. p. 226.

[10]ibid. p. 208. Walras's analysis is in fact confined to the exchange-relation. He deals with the entire 'production process' with one word. The production process is replaced by a symbol, the concept of 'coefficients of production', meant to mean those amounts of productive goods used in the manufacture of one unit of output. Each unit of production is then allotted a corresponding 'production coefficient', and in this formal manner, disposed of for Walras.

[11]August Walras makes this quite clear in a letter to his son (6 February 1859): 'One thing which I find especially pleasing in the plan for your work is the project which you have, and which I approve of totally, of keeping to the least offensive limits as far as property is concerned. This is very wise and very easy to observe. It is necessary to do political economy as one would do acoustics or mechanics' .

[12]J. B. Clark constantly tried to prove the principle that the formation of prices under free competition would allocate everyone exactly that which corresponds to their productive efforts. 'Natural law so far as it has its way, excludes all spoliation.' In a polemic against von Thuenen he assures that 'the natural law of wages gives a result . . . (that is) morally justifiable'. The Distribution of Wealth New York, 1931 p. 324.

[13]Cf. Myrdal op. cit. p. 152 -See too Boehm-Bawerk, Positive Theorie des Kapitals, Jena 1921, Abteilung 11/2 pp. 202-5.

[14]One could counter to this that Boehm-Bawerk deals with production in the Positive Theorie des Kapitals in the unknown sections on 'Produktionsunwege' (p. 15) and 'the capitalist production process' (p. 81). However, it would be deceptive to expect that Boehm really does deal with production. All that one discovers are general concepts which do not seem to capture the specific features of the capitalist period of production, but which are rather intended to apply, in their abstract universality, to all periods: thus for example, the statement that objects of use can be made in two ways: directly, such as picking wild fruits from a high tree: or indirectly, by first cutting a branch from another tree, and then knocking the fruits down. (p. 87) The creation of such an 'intermediary product' means the creation of a 'capital', and hence the carrying out of 'capitalist production', which for Boehm is identical with any form of indirect production. This confusion rests on a trivial confusion of the technical labour-process with the valorisation process, such that for Boehm any tool is 'capital': hence the Red Indian or Zulu who uses a boat for catching fish is a capitalist and carries on 'capitalist production'. (p. 86) According to Boehm's terminology capitalist production already existed at the most primitive levels of culture .

[15]Joseph Schumpeter, Das Wesen and der Hauptinhalt der theoretischen Nationaloekonomie Leipzig 1908 pp. 50 et seq.

[16]With the possible exception of the historical school in Germany led by Schmoller, which, however, because of its descriptive and eclectic character, and rejection of theory, can be passed over here.

[17]Capital I p. 92.

[18]Poverty of Philosophy, New York, 1%3 p. 106.

[19]'In the capitalist mode of production the labour-process only appears as a means for the valorisation process'. Capital II .

[20]In Marx's view the deep similarity between capitalist production and the Mercantilist system is particularly evident in crises. When all -values and prices are subject to enormous disturbances, suddenly there is a hunt for a stable metallic currency -hoarding of gold -, as the one secure thing in the midst of general insecurity, as the 'summum bonum"just as it is understood by the hoarder'. This hoarding of gold then acts to express that in a mode of production based on abstract exchange-value, 'the actual devaluation and worthlessness of all physical wealth' is the natural consequence, because alongside abstract exchange-value 'all other commodities -just because they are use-values -appear to be useless, mere baubles and toys'. (Contribution to a Critique of Political Economy p. 146). Although political economy imagines itself to be above Mercantilism, and assails it as a 'false theory', as illusion, it shares the same basic assumption as Mercantilism. As a consequence the Monetary system does not only remain. 'historically valid but retains its full validity within certain spheres of the modern economy'. (ibid, p. 159).

[21]Theories of Surplus-Value II, p166.

[22]Accordingly, for Marx, the only 'real' labour is the concrete labour which functions in the technical labour-process. (Contribution p. 36, 38): whereas abstract labour which produces exchange-value is simply the 'bourgeois form' of labour. (ibid. p. 48, 54). ' . . . the labour which posits exchange-value is a specific form of labour' (ibid. p. 36), and it is this exchange-value-positing labour which is responsible for all market catastrophes, devaluations, overproduction, stagnation. (Poverty of Philosophy p. 68).

[23]Capital I p. 103.

[24]MEGA p. 154.

[25]Theories of Surplus-Value III p. 464.

[26]Hegel had already criticised this tendency to mathematicisation, which only captures one side, the relations between quantities, in the concrete totality of reality, and neglects all the remaining qualitative aspects. 'Its purpose or principle is quantity. This is precisely the relationship that is non-essential, alien to the character of the notion. The process of knowledge goes on, therefore, on the surface, does not affect the concrete fact itself, does not touch its inner nature or notion, and is hence not a conceptual way of comprehending' (Phenomenology of Mind, London 1931 p. 102-3). He consequently emphasises that the task of political economy consists, not only in representing quantitative relations and movements, but also the qualitative side of their element in their 'realisation' (Verwirklichung) .

[27]Capital I p. 94.

[28]Grundrisse (Introduction) p. 101 .


[30]Ibid. p. 101.

[31]Ibid. p. 86.

[32]Capital I p. 493-494.

[33]Poverty of Philosophy p. 109. In a letter to Kautsky (26 June 1864) Engels criticises him for not having paid sufficient regard to the role of the labour-process. 'You should not separate technology from political economy to the extent that you have. . . . The tools of natives condition their society just as much as more recent tools condition capitalist society'. (Kautsky, Aus der Fruehzeit des Marxismus. Engels Briefwechsel mit Kautsky Prague 1935 p. 124.

[34]It is no accident that such a large part of the presentation in all the volumes of Capital is devoted to the technical labour-process. The chapter on machinery and large-scale industry on its own encompasses nearly 150 pages ; in addition a lot of space is given over to the presentation of the technical labour-process in connection with the valorisation process.

[35]Capital I p. 455 and 492.

[36]Ibid. p. 457.

[37]Ibid. p. 544.

[38]Ibid. p. 518-9. See pp. 489-90 on the insubordination of the workers in manufacture.

[39]Ibid. p. 526.

[40]Ibid. p. 528.

[41]Ibid. p. 532 .

[42]Ibid. p. 548.

[43]Ibid. p. 549-51.

[44]Ibid. p. 531 .

[45]Ibid. p. 531 .

[46]Ibid. p. 531 .

[47]Ibid. p. 531 .

[48]Ibid. p. 530.

[49]Ibid. p. 490 Cf. Jean Weller La Conception classique dun equilibre economique, Paris 1934. p. 11 and John M. Clark The Relation between Statics and Dynamics in Economic Essays in Honour of J.B.Clark New York, 1937 p. 51