While 2021 marks the 100th anniversary of the publication of John Maynard Keynes’s Treatise on Probability, it was not until roughly thirty-five years ago that serious scholarship in the history of economic thought began to examine the connections between his early work in the philosophy of uncertain knowledge and his later work as an economic theorist. When the Royal Economic Society, for instance, set about to honor Keynes’s contributions by publishing his Collected Writings, they did not think it worthwhile to include the correspondence concerning his work in probability, and this decision reflected the way that the profession weighed Keynes’s philosophical work as unimportant to understanding his economics at the time.Footnote 1 The cause(s) of the profession’s neglect of this part of Keynes’s work for so long is almost certainly overdetermined, but one of the major contributors was that the mainstream, neoclassical Keynesian economics that emerged in the postwar world did not contain expectations as an important component of its analysis. If “Keynesian economics” was not concerned with uncertainty, why would his early work on uncertain knowledge reveal anything to us about his economics?
The neglect of Keynes’s work in philosophy began to shift in the 1980s with two Cambridge dissertations by Anna Carabelli and Roderick O’Donnell, which were subsequently revised and published as books (Carabelli Reference Carabelli1988; O’Donnell Reference O’Donnell1989). By the 1980s, historians of economic thought had become increasingly interested in intellectual history as a model for their work, rather than simply looking to the frontiers of mainstream economics to define their work; thus, scholars began researching the full range of Keynes’s work, from his philosophical interests (Bateman Reference Bateman1987, Reference Bateman1988; Davis Reference Davis1994; Runde Reference Runde1994), to his work in economic policy making and investing (Clarke Reference Clarke1988). Suddenly, there was a different, fuller kind of interest in understanding the things that influenced Keynes’s work as an economist, and this change came from many directions.
When the editors of this Journal approached me to organize a symposium to celebrate the centenary of Probability, however, I sought not to recap the early years of scholarship but to look for new work on this aspect of Keynes’s thought. What has happened in the most recent decade of scholarship on Probability?
Arguably the most groundbreaking work in this area in the last ten years has been Jeff Biddle’s exploration of the way that Keynes’s book shaped empirical work in the United States in the 1920s. Biddle’s essay in this symposium documents how Probability influenced the way that several leading American statisticians understood “good method” in their empirical work, especially inductive work based on economic time series data. Biddle then goes on to discuss how Keynes’s ideas regarding “inferential methods based on probability theory” shaped the empirical work of several leading American agricultural economists.
Sylvie Rivot has been working on another frontier of scholarship focusing on Keynes’s work on probability. Traditionally, scholarship on the influence(s) of Probability on Keynes’s work as an economist has focused on the connections to his work as an economic theorist. Rivot, however, has chosen to explore how his work on probability affected his outlook on economic policy. Looking over the arc of his career, she finds in her contribution to the symposium that Keynes was concerned not only with people’s expectations but also with the confidence with which they hold them. This distinction reflects, of course, Keynes’s distinction between probabilities and the weight we place on them. Seen through this frame, Rivot provides the reader with a fresh view of Keynes’s thinking on the relative roles of fiscal and monetary policy.
The third contribution to the symposium draws on two recent contributions to our understanding of Keynes’s work: Robert Shiller’s (Reference Shiller2019) work on narrative economics and Cheryl Misak’s recent work (Reference Misak2020) on the influence of Frank Ramsey on Cambridge economics. My own essay in this symposium considers Keynes’s use of narrative beyond Shiller’s limited examination of The Economic Consequences of the Peace (Reference Keynes1920) and places this in the fuller context of Frank Ramsey’s pragmatism. This provides yet another lens through which to understand Keynes’s shift from the objective theory of probability in Probability to his use of an intersubjective theory of probability in his economic theory after 1932.
When the celebration of the fiftieth anniversary of Keynes’s General Theory of Employment, Interest, and Money took place in 1986, scholarly interest in the centrality of uncertainty and expectations to his magnum opus was a relatively new enterprise. When we celebrate the General Theory’s 100th anniversary in 2036, however, the foci will undoubtedly have changed. There is no doubt, for instance, that our understanding of Keynes’s work in probability theory will have continued to evolve and there is also no doubt that this work will influence our retrospective understanding of what Keynes meant to do in his work in economics. That’s how intellectual history works.
