Scientific Method

One reason for taking this remark seriously is that Henderson was express­ing views that will have reinforced ideas about thermodynamics that Samuelson was also hearing from Wilson.

e. As is noted in chapter 22 this volume, Henderson was the first person whose work Samuelson cited in his Foundations.

Another figure in the Harvard complex was physicist Percy Bridgman (1862—1961), a teacher at Harvard since 1910, whose reputation rested on his experimental work on thermodynamics, for which he later (1946) received the Nobel Prize for physics. He attracted the attention of phi­losophers and students of other disciplines with The Logic of Modern Physics (1927), which put forward the method of “operational analysis.” Samuelson remembered having been introduced to operationalism by Henry Schultz when he was an undergraduate in Chicago.

Bridgman’s starting point in The Logic of Modern Physics was the “tru­ism” that “all our experimental knowledge and our understanding of nature is impossible apart from our own mental processes,” and that his essay was based on observing what physicists were currently thinking.²⁴ A reconsid­eration of the fundamentals of physics was needed because physicists’ ways of thinking had been challenged by recent developments, including both Einstein’s theory of relativity and quantum mechanics, the latter developing so fast that Bridgman had to concede some of his material was already out of date. Physicists were working with ideas that could not easily be related to everyday experience, which necessitated a revision of how we thought about concepts. That is, concepts could no longer be defined in terms of their prop­erties, but had to be defined in terms of operations: “In general we mean by any concept nothing more than a set of operations: the concept is synonymous with the corresponding set of operations.”^2,5 These operations might be physical (as with physical concepts such as length) or mental (as with mathematical concepts such as continuity).

This view had clear implications. It meant all knowledge was relative to the operations involved. It meant questions were meaningless if it was not possible to find operations by which answers could be obtained. Thus, notions of absolute time and space were meaningless because they had no connection with reality, for any operations to measure them would of necessity be relative to those operations.

I believe that many of the questions asked about social and philosophi­cal subjects will be found to be meaningless when examined from the point of view of operations. It would doubtless conduce greatly to clar­ity of thought if the operational mode of thinking were adopted in all fields of inquiry as well as the physical. Just as in the physical domain, so in other domains, one is making a significant statement about his subject in stating that a certain question is meaningless.²⁶

Bridgman went on to argue that thinking in terms of operations had con­sequences for science, for it would simplify thinking, rendering earlier speculations “unreadable.”²⁷ In developing his ideas, Bridgman paid much attention to “intuition,” which was tied to his ideas about what it meant to explain something, and “models.” Explanation involves reducing situations to ones that are consistent with our intuitions. At this time, neither the word “intuition” nor the word “model” was widely used in economics journals, but beginning in the 1940s, they came to be widely used by mathematical economists.

Bridgman’s philosophy, derived from his own observations of what appeared to be successful in physics, fits with Henderson’s craft-oriented view of the scientific method.²⁸ At the same time, Bridgman parted com­pany with Henderson in ways that would have been attractive to scholars such as Samuelson, skeptical about the latter’s leap into Paretian sociol­ogy. His focus on operations—he disliked the terms “operationalism” and “operationism,” as implying something more esoteric than his “simple ideas”—placed concepts firmly in the arena of scientific practices and cau­tioned scientists against claiming more than their methods could demon- strate.2⁹ It also left room for creativity, for operations could never be fully specified, allowing for nonrational decisions by scientists. That is, scien­tific knowledge was not something that could be understood apart from human psychology.

Bridgman’s operational methods were adopted by logical positivists and behavioral social scientists (notably Skinner), but were taken in directions dif­ferent from those in which Bridgman himself wished to go. Bridgman held that there was an irreducibly individual, subjective element in all knowl­edge, and he was critical of attempts, whether by positivists or the Vienna

Circle,^f to equate his operational method with their attempts to derive rules by which the objectivity of knowledge could be ensured: that was an impos­sible goal.³¹ Given this ambiguity, we need to be careful in imputing to Samuelson a particular interpretation of operationalism, for though he was to make the idea central to his work, there is little evidence of how much he read and precisely what he made of it—though he was to recommend that his MIT students read Bridgman—or of personal interactions with Bridgman that might have colored his reading of his work. In any case, his understand­ing of operational analysis will have been mediated by his discussions with Wilson, Schumpeter, Henderson, and colleagues in the Society of Fellows, all of whom had strong views on scientific method.^g