'What happens after all is always that we exchange the whole of production for the whole production of the previous year. Besides, if production gradually increases, the exchange, at the same time as it improves future conditions, must entail a small loss every year.'[174] In other words, when the aggregate product is realised, accumulation is bound each year to create a surplus that cannot be sold. Sismondi, however, is afraid of drawing this final conclusion, and prefers a 'middle course', necessitating a somewhat obscure subterfuge: 'If this loss is not heavy, and evenly distributed, everyone will bear with it without complaining about his income. This is what constitutes the national economy, and the series of such small sacrifices increases capital and common wealth.'[174] If, on the other hand, there is ruthless accumulation, this surplus residue becomes a public calamity, and the result is a crisis. Thus a petty-bourgeois subterfuge becomes the solution of Sismondi: putting the dampers on accumulation. He constantly polemises against the classical school which advocates unrestricted development of the productive forces and expansion of production; and his whole work is a warning against the fatal consequences of giving full rein to the desire to accumulate. Sismondi's exposition proves that he was unable to grasp the reproductive process as a whole. Quite apart from his unsuccessful attempt to distinguish between the categories of capital and income from the point of view of society, his theory of reproduction suffers from the fundamental error he took over from Adam Smith: the idea that personal consumption absorbs the entire annual product, without leaving any part of the value for the renewal of society's constant capital, and also, that accumulation consists merely of the transformation of capitalised surplus value into variable capital. Yet, if later critics of Sismondi, e.g. the Russian Marxist Ilyin,[175] think that pointing out this fundamental error in the analysis of the aggregate product can justify a cavalier dismissal of Sismondi's entire theory of accumulation as inadequate, as 'nonsense', they merely demonstrate their own obtuseness in respect of Sismondi's real concern, his ultimate problem. The analysis of Marx at a later date, showing up the crude mistakes of Adam Smith for the first time, is the best proof that the problem of accumulation is far from solved just by attending to the equivalent of the constant capital in the aggregate product. This is proved even more strikingly in the actual development of Sismondi's theory: his views involved him in bitter controversy with the exponents and popularisers of the classical school, with Ricardo, Say and MacCulloch. The two parties to the conflict represent diametrically opposed points of view: Sismondi stands for the sheer impossibility, the others for the unrestricted possibility, of accumulation. Sismondi and his opponents