This paper seeks an answer to the following question: By what process did the utility concept in economics evolve from its Benthamite to its modern form? Jeremy Bentham applied his principle of utility to ethical, legal, and constitutional issues, and only after more than a century of adaptation did it become the dominant explanation for consumer choice. The paper identifies fitness for mathematical analysis as the underlying principle of selection for this evolutionary success, exemplified in particular by the mutations of utility between Bentham and W. S. Jevons, and between Jevons and Alfred Marshall.

In Chapter 2, Book IV, of The Wealth of Nations, Adam Smith assesses the appropriateness of the policy of tariff retaliation or reciprocity. His deliberations are prefaced by a discussion of two rather innocuous cases that turn out to have very little to do with the main topic. In the first case it is advantageous “to lay some burden upon foreign, for the encouragement of domestic industry” (WN, p. 463) when that particular industry is necessary for the defence of the country. Smith's second case refers to the situation when some domestic tax is being imposed on a good produced at home. In this situation Smith believed it reasonable that an equal tax should be imposed on the same good when it was imported (WN, p. 465).

Despite its recent re-emergence to analytical importance, the phenomenon of increasing returns remains outside the central core of neoclassical economics. The history of this idea (or set of ideas) might have been quite different if Adam Smith's explanation of the origins of trade had not been replaced by that of David Ricardo. To Adam Smith, mutually beneficial exchange emerges because of specialization, which, in its turn, implies the presence of increasing returns to the size of the exchange nexus. Even in a world of equals, trade offers mutuality of gain. There is no need for participants in the economic nexus to differ one from another. In the Ricardian logic, by contrast, trade presumably emerges because productive resources differ in their capacities to create economic value, at least among separate “goods.” Specialization is a “natural” feature of resource endowments—a feature that is exploited by trade. Comparative advantage ensures the mutuality of gain. But, in this explanation, there is no direct linkage between the size of the exchange network and the degree of specialization that is viable. There is no need to introduce increasing returns. Comparative advantage may be present even if there are constant returns to scale, both for the economy and for its separate productive sectors.

Smith explicitly spells out the term “invisible hand” in only three texts, each in a different work. What is embarrassing to neoclassical welfare theory is that these explicit references have little to do with what came to be understood by the metaphor. In particular, the explicit reference to the invisible hand in The Wealth of Nations (WN) is not about the efficiency of the market in allocating resources. Rather, it is about, ironically, how impediments in the international capital market generate great benefits.

There was no Jevonian revolution in international trade theory. The new ideas of marginal analysis entering British economics in the late nineteenth century seem to have left the field untouched. According to Schumpeter (1954, p. 609), Marshall “did not do more than polish and develop Mill's meaning … Edgeworth's famous restatement added many interesting details but also did not go beyond Mill in fundamentals.”