In 1927, when the thirty-year-old Sraffa arrived in Cambridge, Anglo-Saxon economics was dominated by Marshall’s thought. Morever, both in Europe and America, one or another form of marginalist economics held undisputed sway—with its subjectivist theory of value and its anti-socialist implications. Sraffa himself had recently become notorious in the Anglo-Saxon world, having published in a 1926 number of the Economic Journal an article criticizing Marshall’s theory of value. But the first reactions to this work already clearly showed the need to go deeper: that is to say, to combine the critique of marginalist theory with a reconstruction of the alternative approach of the classical economists and Marx. A research undertaking of more than thirty years first took shape during this period. It was to result in Production of Commodities by means of Commodities (1960) and a few years earlier (1951–5) in Sraffa’s edition of the works of Ricardo, the greatest of the classical economists.

These two projects went hand in hand, complementing each other. In Production of Commodities, Sraffa not only laid the basis for a critique of marginalist value-theory, but thoroughly resolved an analytical problem with which the classical economists had grappled in vain, and which had been tellingly used against them by their adversaries. This is the problem of determining the prices of production and the relationship between these prices and the intervening variables: wages and the rate of profit. With his edition of Ricardo’s works, Sraffa re-presents the classical economic approach based on the concept of surplus, rescuing it from oblivion and from the misleading interpretations made current by the marginalist school.

When Sraffa began a series of lectures on the theory of value in October 1928, Ricardo had a significance for economists that is very different from the one we attribute to him today. For Marshall, he was the immature and inexact forerunner of modern theory; while for Jevons, he had been responsible for a pernicious deflection of economics from the road to science. At any event, Ricardo aroused very little interest. At best, his theory of rent was evoked as prefiguring the principle of diminishing marginal productivity; or some mention was made of his theory of money, and his grounding of international trade on the principle of comparative costs. Ricardo’s labour theory of value was normally disregarded by theorists, only to be mentioned by economic historians as an object of antiquarian interest; and his idea of economic process as being based on the concept of surplus had been buried altogether.

In this cultural atmosphere of 1928, the Royal Economic Society entrusted the task of preparing an edition of Ricardo’s works to an economic historian mainly interested in monetary problems: Professor Gregory of the University of London. However, it did not attach very much weight to the project; and since Gregory was himself not fully committed to the undertaking, he willingly agreed to pass it over to Sraffa. (The good offices of Keynes were of crucial importance, as they had been in inviting Sraffa to Cambridge in the first place.) The Italian economist set to work with great enthusiasm and exasperating tenacity. The appearance of the work was several times reported as imminent; but Sraffa continued his veritable ‘hunt’ for manuscripts, receiving help from Keynes both in this search and in warding off publishers’ complaints about the delay. In July 1943, Ricardo’s very important letters to James Mill were rediscovered, together with a number of other manuscripts, including the major essay ‘On absolute value and exchange value’, on which Ricardo worked during the last weeks of his life. Maurice Dobb began to collaborate with Sraffa during the last phase of the project. And then, between 1951 and 1955, it was completed with the publication of the ten-volume Works and Correspondence of David Ricardo—a critical edition which, in the unanimous view of both ideological friends and opponents, remains a model of textual rigour.

Through the published texts, the apparatus of notes and, above all, his own introduction in Volume One to The Principles of Political Economy and Taxation, Sraffa disentangles Ricardo’s thought from marginalist misrepresentation, restoring his key place in economic theory (and with it, the place of the entire trend of classical economics). Drastically oversimplifying Sraffa’s exposition, we may distinguish two successive phases in Ricardo’s analysis. The first culminates in his 1815 ‘Essay on the low price of corn’, where he employs what is today called a ‘single commodity model’. The use of a given quantity of corn as means of production (that is, as seed and as subsistence-wages for workers engaged in the production process) makes it possible to obtain a greater quantity of corn; once the initial supply of means of production has been reconstituted, there thus remains a surplus which goes to the property-owning classes (profit to the capitalists, and rent to the landowners). If the fertility of the land varies, competition among capitalists will result in their paying a level of rent for the best lands equal to the difference between the surplus obtainable on them and the surplus obtainable on the worst land under cultivation. When a rise in population brings ever less fertile land into use, the surplus on the worst land will grow smaller and smaller, as will the rate of profit; whereas rents will continually rise and real wages remain fixed at subsistence level. Accumulation, which is dependent on profits, will therefore slow down. And so, in order to favour the capitalists (and hence economic progress) in their struggle with the landowners, it is necessary to promote imports of corn. Which is the origin of Ricardo’s commitment to free trade.

In the second phase, which is that of the Principles (1817), Ricardo abandons the ‘single commodity model’ under the pressure of Malthus’s criticisms. He then adopts a labour theory of value (according to which the value of any commodity is determined by the amount of labour directly or indirectly necessary for its own production), using it to measure the surplus and the anticipated capital, and thus to derive the rate of profit as a relationship between two heterogeneous aggregates of commodities. In this respect, the theory of value plays a role that is subordinate and functional to the theory of distribution; and it allows the latter to bring out the class conflict of interests among workers, capitalists and landowners, with regard to the distribution of the surplus.

However, another aspect of the theory of value still has to be discussed: namely, that of the ‘invariable measure of value’, and the relationship between absolute value and exchange value. Ricardo already broached the question in his Principles, and went on to deal with it more directly in his last writings. But in his introduction to the Principles, Sraffa makes a contribution which not only interprets Ricardo’s views but is itself of the utmost theoretical importance. Indeed, it has perhaps not yet received all the attention it deserves—including for its bearing on the problem of the relationship between classical economics and Marx. Sraffa shows that Ricardo merged in the ‘invariable measure of value’ two attributes which ought to be kept distinct: 1. that of having an invariable value (with respect to the means of production themselves), when the distribution of income between wages and profits changes at a fixed level of technology; and 2. that of having an invariable value with respect to changes in technology. In Production of Commodities by means of Commodities, Sraffa demonstrates that the first problem can be solved once the two are distinguished. As to the second, it is clear that measuring in terms of labour-contained does here preserve a certain meaning; but it is also clear that the problem is in danger of assuming metaphysical or subjectivist dimensions (labour as a ‘sacrifice and chore’).