Gramsci, Wittgenstein, and Ricardo
Sraffa met Antonio Gramsci in 1919 while both were university students in Turin. Without ever becoming a registered member of the Socialist or the Italian Communist Party (founded in 1921), Sraffa developed a strong friendship and intellectual relationship with the Communist leader, with long discussions and collaboration on the journal founded by Gramsci, L’Ordine Nuovo. When Gramsci was imprisoned in 1926, Sraffa became his main point of reference in the outside world (after a short while, through Gramsci’s sister- in-law, Tatiana), getting him books and journals, searching for ways to obtain freedom without genuflecting to the Fascist regime, offering food for thought through Tatiana’s letters, and thus acting as a stimulus to the writing of the famous Quaderni del Carcere (Prison Notebooks, published posthumously, Gramsci 1975), which would exert a deep and wide influence on Western leftist thought in the post-Second World War period. Sraffa’s influence on Gramsci, and Gramsci’s influence on Sraffa (and, through him, on Wittgenstein), can still provoke heated debate; however, it would be beyond our scope to go into the issue here.[1]
Something more can be said about Sraffa’s relationship with Ludwig Wittgenstein. The two met in Cambridge in 1929, where the Austrian philosopher had just moved, under the auspices of Bertrand Russell, who a few years earlier had taken care of the publication of Wittgenstein’s celebrated book, Tractatus Logico-Philosophicus (Wittgenstein 1921 [1922]), considered by many as the high point of logical neopositivism. Wittgenstein’s long and frequent discussions with Sraffa exerted a decisive influence on the Austrian philosopher and played an important part in his transition from the logical neopositivism of the Tractatus to the new position developed in Philosophical Investigations, published posthumously (Wittgenstein 1953).
Wittgenstein himself in the Preface to Philosophical Investigations stated: ‘I am indebted to [the criticism] which a teacher of this university, Mr. P Sraffa, for many years unceasingly practised on my thoughts. I am indebted to this stimulus for the most consequential ideas of this book’ (ibid.: viii; italics in original). Wittgenstein had abandoned the idea, developed in the Tractatus, of language as a biunivocal representation of the world, with a connection between ‘propositions’ (constituting language) and ‘facts’ (constituting the world). In the Philosophical Investigations he developed a new theory of language and the relations between it and the world it should describe: a theory based on the idea that there is not just one type of language, but ‘there are countless kinds: countless different types of use of what we call “symbols”, “words”, “sentences”. And this multiplicity is not something fixed, given once for all; but new types of language, new language-games, as we may say, come into existence and others become obsolete and get forgotten’ Wittgenstein (1953: 21; italics in original). This is the so-called theory of language games: theoretical models focusing attention on particular aspects of real language, presenting them as the general language of a group of people.
This methodological position can be generalized, as in opposition to the idea that reality can be represented through a general, all-encompassing model—such as, in economics, general equilibrium theory—and preference for the construction of specific theoretical bricks connected into a more gen?eral world view by the requirement of conceptual consistency. In the case of economics, this means consistency with a representation of the economy as a circular flow of production and consumption, as in classical political economy, rather than as a one-way street from scarce resources to the satisfaction of economic agents’ desires, as in the marginalist/neoclassical approach.
Many contemporary economists are convinced that there is only one correct way of looking at the economy, namely the scarcity view. This is a frequent source of misunderstanding in debates between mainstream and heterodox economists. So much was quite clear to Sraffa himself, since in the early stages of his research career he too considered equilibrium between supply and demand as the only rule of the game, to the extent that in his 1925 article, after demonstrating that within Marshall’s approach both increasing and decreasing returns are logically unacceptable, he concluded that constant returns must be assumed (Sraffa 1925). According to philological reconstructions based on Sraffa’s papers (preserved at Trinity College, Cambridge), it is now believed that the turning point came around 1930, when Sraffa reached the conclusion that no assumption whatever on returns is necessary, since the level of production may be assumed as given when analysing the central issue in value theory, namely determining the set of prices of production that guarantee reproduction of the circular flow and their relationship with income distribution. According to Sraffa’s own reconstruction of classical political economy, this—the idea that equilibrium between supply and demand is not the fundamental pillar in classical value theory—is a crucial divide between the classical and the marginalist approach. Hence, a study of the classical conceptualization of the economy, as distinct from the marginalist conceptualization based on supply and demand equilibrium (i.e. the balancing of scarce resources with economic agents’ preferences), is a prerequisite for a theoretical reconstruction of the classical approach.
Thus, Sraffa’s meticulous work on his magnificent edition of Ricardo’s Works and Correspondence is not solely the fruit of a passionate interest in the history of economic thought; it was also part of his effort to bring economics back from the marginalist to the classical path—and, as we shall see, in such a way as to make the latter compatible with his friend Keynes’s own contributions. In fact, it was at Keynes’s suggestion that the Royal Society entrust Sraffa with the task of editing Ricardo’s writing in 1930, and it was Keynes who repeatedly defended Sraffa when he delayed completion of the work year after year, and who helped him in the years-long fruitful treasure hunt for lost Ricardian manuscripts (which led to many minor and two major discoveries, such as a chest of letters received by Ricardo from his correspondents, discovered in 1930, and a number of important letters from Ricardo to
James Mill and various manuscripts, including the essay on ‘Absolute Value and Exchangeable Value’ in 1943). But again, this is not the place to recount the story of this truly magnificent enterprise, the countless anecdotes about Sraffa’s meticulousness and hard work, and the eager expectation of economists such as Einaudi or Schumpeter for the completion of the work.[2]
Let us summarize, from our post-1960 (i.e. with our knowledge of Sraffa’s 1960 book) vantage point, Sraffa’s interpretation of Ricardo. Sraffa stresses the importance of the notion of the surplus, or in other words the difference between the set of commodities produced and the set of the means of production and subsistence necessary to carry on production. In the circular flow of production and consumption, when the division of labour prevails, each productive unit at the end of the production process obtains a product (or, in joint production, a set of products) and, in order to start the production process again, needs to obtain from other sectors (other units of production) its means of production and the means of subsistence for its workers. Thus, there arises a web of exchanges between productive units and between sectors. In a market economy, these exchanges also constitute the way in which the surplus is distributed between productive units and sectors; Sraffa assumes, with implicit reference to a capitalist economy, that this happens on the basis of the simplifying assumption of a unique rate of profits (which corresponds to the assumption of free competition, that is, free movement of capital between the different sectors of activity in the economy).
When a wage rate equal to the subsistence wage is also assumed, such as to allow workers to acquire the means of subsistence necessary for their and their family’s survival, and once rents are determined thanks to the ‘Ricardian’ theory of differential rent (which, in fact, Ricardo drew from Malthus, West, and Torrens), profits correspond to the part of the surplus not accruing to rents. If, as does Ricardo, we assume that all rents are devoted to luxury consumption and profits are entirely devoted to investment, the rate of profits (equal to the ratio between profits and capital advances) corresponds to the rate of accumulation (equal to the ratio of investments to capital advanced).
There is a problem, however: since profits are obtained as the difference between the value of the surplus (consisting of a set of heterogeneous commodities) and the amount of rents, and since capital advanced is also a set of heterogeneous commodities, we need to express in value terms such aggregates of commodities in order to determine the rate of profits and the rate of accumulation (hence, the rate of growth of the economy). This is no easy task, however, and—as Sraffa shows—Ricardo persevered in it up to the end of his life without, however, coming to the definitive solution, but only to approximate ones.
According to Sraffa’s interpretation, a preliminary, approximate solution (to be found in Ricardo’s 1815 Essay on Corn, but possibly devised one year earlier) consists in determination of the rate of profits as a ratio of two physical quantities of corn. This solution relies on a strong simplifying assumption, namely that there are only two commodities in the economy: corn (agricultural produce), utilized as a means of production in both sectors of the economy and as the only means of subsistence for the workers, and a manufactured commodity, which is only utilized as a means of production in its own sector. Thus, we may focus attention on the corn sector: once we have circumvented the problem of rent by focusing on no-rent land, subtracting from the amount of corn produced on this kind of land the amount of corn used as a means of production and for subsistence, we obtain profits as an amount of corn, while also capital advances are expressed as an amount of corn; the ratio between profits and capital advances, both magnitudes expressed in physical terms as amounts of corn, thus determines the rate of profits.
Malthus immediately objected to this solution that in no sector does any such situation occur, with a single commodity representing both the whole product and the whole of the means of production and subsistence. Thus, in the short time span of a couple of years, Ricardo turned to a different solution. In his Principles of Political Economy and Taxation (1817) he adopted a labour-contained theory of value, expressing both aggregate product and aggregate means of production and subsistence in terms of the labour directly or indirectly required for their production. Once again, this allowed him to determine the rate of profits as a ratio between two quantities, surplus produce and capital advances, both expressed in terms of a single magnitude, labour. Once again, Ricardo himself knew that this solution was not perfect, and in the very first chapter of the Principles indicated the limits of the labour theory of value, including the existence of different ratios between labour and capital advances in different sectors of the economy, different lengths of production period, and different durations of the various capital goods (this latter being the object of a specific critique by Torrens). As a result, the problem of value remained open, nor could the answer to what Marx (1894) called the problem of transformation—of labour values into prices of production—suggested in the posthumous Book 3 of Capital half a century later amount to a definitive solution. Finally, after years of searching, a solution was offered by Sraffa in his 1960 book, Production of Commodities by Means of Commodities, as we shall see in more detail below.
