Professor Seligman has pointed out that Ricardo's contention that value is determined by the cost of production, and the contention of Jevons that value is determined by marginal utility, are not mutually exclusive, but, on the contrary, complementary to each other.[173] The present writer has long contended that the marginal utility theory and the Marxian labor-value theory are likewise not antagonistic but complementary.[174] This is not the place to enter into the elaborate discussion which this contention involves. Only a brief indication of the argument for the claim is here and now possible. First, as we have seen, Marx is very careful to insist that utility is essential to value, and that the utility must be a social utility. But social utility does not come of itself, from the skies or elsewhere. It is, so far as the vast majority of commodities is concerned, the product of labor. It is true that the value of a thing is never independent of its social utility; it is likewise true that this is determined by the social labor necessary to the reproduction of that utility. To regard the two theories as antagonistic, it seems to be necessary to say either (1) that the quantity of social labor necessary to produce certain commodities determines their value, utterly regardless of the amount of their social utility, or (2) that we estimate the social utility of commodities, estimate what we are willing to pay for them, utterly regardless of the labor necessary, on an average, for their reproduction. The latter contention would be absurd, and the former would involve the abandonment of the initial premises of the Marxian theory, contained in his definition of a commodity. In so far as the basis of social utility is the social labor necessary for its production, the labor-value theory of Marx may be said, I think, to include the marginal utility law, as one of the forms in which it operates. V