Science and Economics

It is not known when Samuelson had first met Knight, though a strong possibility is that it was on November 2, 1932, when Knight gave a talk for the National Student League with the title “The Case for Communism: From the Standpoint of an ex-Liberal.”²⁹ Delivered less than a week before the presidential election that brought Roosevelt to power, Knight offered a denunciation of the political system.

I am reminded of an incident told me the other day by a college book man. He went, with another book man, to call on the President of one of our largest and greatest universities. The President turned to book man No.2, with whom he was more acquainted, and said: “I am very glad to see you men. You meet different classes of people; you see the college and university people and you meet up with business men and all classes in hotel lobbies and smoking compartments. Tell me what the people of America are really talking about.” The book man responded without hesitation with a short monosyllable, which was not the word “sex,” but that is the way I have to report it, and that is the more important point I wish to make.³¹

Knight went on to argue that certain words were acceptable in public dis­course but others were not.

For instance, “cow-dung” is not a particularly “bad” word... but I must admit that I lack the courage merely to change the sex of the animal and use a different monosyllabic synonym for the substantive part of the expression.

People would not find one more shocking than the other, but they would pre­tend that they were. This was entirely right because social stability depends on public talk not meaning what it says. “Its first and most fundamental category must always be—‘B.S.’ But don't say it!”³² It was nihilistic, dismiss­ing the possibility of political processes involving serious thought; hence, the remark with which he opened his talk, that those who wanted change and wished to vote intelligently should vote Communist. A strong Communist vote might lead to the growth of a real conservative, aristocratic alterna­tive. This ability to entertain students makes clear why Samuelson would describe Knight as a “cracker-barrel Socrates,” but it is far from clear that he took much from Knight’s skepticism except insofar as it contributed to his conservatism. He later summarized Knight’s position as finding it difficult to choose between communism and fascism—a choice the despised Roosevelt, whom Samuelson had supported, rendered unnecessary.³³

Knight was to repent of this talk, given when he was particularly depressed because of personal problems and the political situation, and he tried to with­draw printed copies from circulation, but his skepticism about the state of politics remained. He gave a series of public lectures in June-July 1934, with the title “Intelligence and the Crisis in Western Culture,” in which many of the same themes recurred (though not the suggestion that people should vote Communist). Given his claim to have idolized Knight and his inter­est in both politics and economics, Samuelson would surely have attended. These lectures offered a historical perspective, contending that liberalism and democracy had worked in the nineteenth-century United States because of special circumstances. The vast resources made available by the movement of the frontier reduced the extent of competition between people: “With an essentially unlimited domain awaiting economic conquest, life ceased to be seriously competitive for individuals,” for success in life could involve exploit­ing nature rather than other people.³⁴ But the situation changed with the closing of the frontier, the result being a resurgence of the state and of politics.

This belief that there were higher ends than those typically stressed by economics was an important theme in Knight’s writing in the late 1920s and 1930s. Ten years before Samuelson’s encounter with economics, Knight had explored the relation between “ethics and the economic interpreta­tion.”³⁵ Knight challenged the view of some economists that human wants and desires should be taken as scientific facts—data that the scientist should take as a given. Wants were continually changing, but whereas many writers were focusing on the manipulation of consumers by business, Knight saw the development of wants as an important part of human nature. People aim not to satisfy their existing wants but to develop better wants.

Life is not fundamentally a striving for ends, for satisfactions, but rather for bases for further striving; desire is more fundamental to conduct than is achievement, or perhaps better, the true achievement is the refinement and elevation of the plane of desire, the cultivation of taste.³⁶

Where did this leave economics? Knight argued that there was no basis on which to distinguish between “economic” and “non-economic” wants. Biologically determined needs did not explain human motivation, for people sought not life but a good life. Neither were instincts a suitable basis for scientific analysis? This led Knight to the conclusion that “in so far as the ends are viewed as given, as data, then all activity is economic.”³⁷ However, rather than conclude that economics should dominate inquiry, he saw its significance as very limited. “Economics” he wrote,

treats of human conduct in so far as conduct is amenable to scientific treatment, in so far as it is controlled by definable conditions and can be reduced to law. But this, measured by the standard of mate­rial science, is not very far. There are no data for a science of conduct in a sense analogous to natural science. The data of conduct are pro­visional, shifting, and special to individual, unique situations in so high a degree that generalization is relatively fruitless.

He illustrated that with the example of the chess player who “acts as if the supreme end in life were to capture his opponent’s pieces,” even though he did not believe this.

The result was that a science of conduct was possible only if its subject matter became so abstract that it said little about actual behavior. Evaluation

l. The first of these was an implicit attack on Alfred Marshall and the second an attack on Thorstein Veblen, two very influential economists in this period. of motives was left to ethics, but this required standards by which to judge actions, and if this was to go beyond economics, it meant drawing on “some­thing more than scientific data.”³⁹ A scientific ethics was simply not possible.

In subsequent articles Knight developed this position. In “The Ethics of Competition,” he argued that it was not possible to pass judgment on eco­nomic policy without starting from some set of values, or ethical criteria— “social policy must be based on social ideals.”⁴⁰ He then went on to explore the standards of value implicit in what he called “the laissez-faire or individu­alistic social philosophy”—“the standards actually involved in making some familiar moral judgments in regard to the economic system.”⁴¹ This resulted in an argument that sounded very negative, for he was critical not only of Thorstein Veblen—the distinction between business and industry was with­out foundation and the idea that engineers should control the allocation of resources was “grotesque”—but also of liberal individualism.⁴²

The starting point for Knight's critique of individualism was the claim that economic activity performed several functions simultaneously.

Economic activity is at the same time a means of want-satisfaction, an agency for want- and character-formation, a field of creative self­expression, and a competitive sport.

He then proceeded to point to failings in each of these. The conditions neces­sary for perfect competition, the foundation on which the case for laissez-faire rested, were not met in real life. As the results produced by the economic system rested on tastes and purchasing power created by that system, it was impossible to impute to them ethical significance. “No one,” Knight claimed, “contends that a bottle of old wine is ethically worth as much as a barrel of flour, or a fantastic evening wrap for some potentate's mistress as much as a substantial dwelling house.”⁴⁴ The process of valuation was a “vicious” circle. Incomes went not to factors of production but to their owners, and the fact of ownership had no ethical validity. “The competitive system,” Knight concluded, “falls far short of our highest ideals.”⁴⁵ When he turned to production, he argued that there were values involved in the economic process itself; people value their social situation rather than con­sumption itself. There were conflicts between three ethical ideals: allocating goods in proportion to effort; efficient distribution of resources; and fairness. Furthermore, there could be no assumption that competition itself was mor­ally desirable.

In these arguments, Knight drew on many themes that would have been familiar from the Hobsonian welfare economics to which Samuelson had been introduced in Gideonse's social science course. Like Hobson and utilitarians before him, Knight presumed a consensus on values, for his conclusions rested on values that be believed were “part of our culture”— “the common-sense ideals of absolute ethics in modern Christendom.”⁴⁶ However, unlike Veblenian and Hobsonian critics of orthodox economics, Knight shied away from radical change, not because individualism was good but simply because “radical critics of competition as a general basis of the economic order generally underestimate egregiously the danger of doing vastly worse.”⁴⁷ He justified his rejection of radical change by claim­ing that the problem was to find the right mixture of policies: “there is no question of the exclusive use or entire abolition of any of the fundamen­tal methods of social organization, individualistic or socialistic.

These two articles were chosen by four graduate students—Friedman, Stigler, Wallis, and Homer Jones—to begin a volume of Knight's essays, published in 1935 to mark his fiftieth birthday, The Ethics of Competition and Other Essaysf^s Though this was published after Samuelson had left Chicago (Knight's birthday was in November), he would certainly have thought very carefully about the essays it contained.^m After these two pieces on ethics and welfare, a discussion of “Economic Psychology and the Value Problem” con­tinued in the same vein. The purpose of knowledge was to predict and to con­trol, and feelings could contribute to this only insofar as they could be linked to behavior: the only way to produce and manipulate feelings was “by means of established behavior sequences.”⁴⁹ However, though he thought that sci­ence had to work exclusively in terms of relationships between observable behaviors (as in Samuelson's later theory of revealed preference), Knight dis­missed the idea that this was all there was. “Much that the devotee of natural science methods,” he wrote, “dismisses contemptuously as ‘mere emotion' may turn out to have as strong a counterpart in ultimate reality as can be put forth by any human experience whatever.”⁵⁰ Drawing an analogy with

m. Even if Samuelson's claim to have read everything Knight wrote is not taken literally (though there seems to be no reason not to do so), these articles, assembled into a book by his friends, would certainly have been among those that he had read. mechanics, he pointed out that while physicists were uncomfortable with the notion of force, unobservable independently of the effects it produces, they used the notion freely. In discovering gravity, what Newton really discov­ered was that the same formula was applicable in apparently diverse cases.⁵¹ Feelings and motivations were important, in part because there was more to life than simply economics. After citing Ruskin's dictum “There is no wealth but life,” Knight argued that “in the connection and for the purpose of Ruskin's preaching, [it] is anything but nonsense. It is exactly what our over scientifically minded students of social problems need to be told, with all possible emphasis.”⁵²

A further chapter, “The Limitations of Scientific Method in Economics,” reinforced this message. Knight offered a very precise and all-encompassing view of scientific economics:

From a rational or scientific point of view, all practically real problems are problems in economics. The problem of life is to utilize resources “economically,” to make them go as far as possible in the production of desired results. The general theory of economics is therefore simply the rationale of life.⁵³

However, he then went on to say that this could not get us very far. It was the rationale of life, he claimed,

In so far as it has any rationale! The first question in regard to scien­tific economics is this question of how far life is rational, how far its problems reduce to the form of using given means to achieve given ends. Now this, we shall contend, is not very far; the scientific view of life is a limited and partial view; life is at bottom an exploration of the field of values, an attempt to discover values, rather than on the basis of them to produce and enjoy them to the greatest possible extent.⁵⁴

Knight worked into this argument about the limitations of the scientific method many more specific methodological observations. Economists fre­quently used the term “dynamics,” but did so in ignorance of the way it was used in mechanics. Statics was about equilibrium, raising the ques­tion of whether forces in operation would tend to produce an equilibrium. Economists had ignored this problem, leaving “a fatal gap in the science.” “The crying need of economic theory to-day,” he contended, “is for a study of the ‘laws of motion,' the kinetics of economic changes.”⁵⁵ This was a theme that Samuelson was later to take up.