The paper tries to clarify certain problematical aspects that characterize the method at the basis of the neo-Walrasian version of the neoclassical theory by going back to the writings of J.R. Hicks, who has both been one of the most influential proponents of the neo-Walrasian approach (Hicks, 1939), and one of its first critics (Hicks, 1946, 1956, 1965, 1976, 1977).

In the first part of the paper (a first draft of which was presented at the 2016 Storep Conference held in Catania) we ascertain the intended relation between the objects determined by the neo-Walrasian theory on the one hand, and observations of economic reality on the other, in itself a task that is far from easy (Ciccone, 1999). We find that what Hicks’s writings seem to imply is that, just as in the traditional method of analysis, two notions of price ought to be distinguished within the neo-Walrasian method: an observable price, that the theory cannot generally determine, and a theoretical price, which it is the aim of the theory to show is the centre of gravitation of the former.

In the second part of the paper we reconstruct the main reasons why Hicks came to doubt that such an aim can actually be reached by the neo-Walrasian theory. First, we find new arguments to claim that the critique that Hicks so often presented in terms of the ‘slowness’ of the process of ‘price adjustment’ (Hicks, 1946, 1965, 1976) should in fact be interpreted as the sign of the awareness on Hicks’s part of the lack of persistence that characterizes the givens of the neo-Walrasian theory — such a lack of persistence being in its turn the effect of the treatment of capital on which the neo-Walrasian theory is built. Contrary to the reconstruction of this aspect of Hicks’s thought recently advanced in De Vroey (2006), we thus confirm the reconstruction already presented in Petri (1991). We next explore the possibility that a second critique by Hicks, that emerges at a certain point in his discussion of the way price-expectations are treated in neo-Walrasian theory, can be shown as going deeper than his first critique. For while according to the first critique the neo-Walrasian concept of equilibrium, though scarcely useful, is nonetheless theoretically conceivable, according to this second critique, which save for the exception of Leijonhufvud (1984) has gone practically unnoticed, that notion of equilibrium would simply be ‘nonsense’ (Hicks, 1977).