I. DON’T LOOK BACK?

HES Presidential lectures have a tendency to speak in grand generalizations concerning the state of the history of economics. I thought I might try to suggest a few diagnoses of its parlous state by instead performing a retrospective on one of the few modern orthodox economists who demonstrated a more-than-armchair fascination with its history, Paul Samuelson. More has been written on him than on almost any other modern contemporary (even Hayek!), and he enjoyed at least three major evaluative volumes devoted to singing his praises during his lifetimeFootnote ¹; moreover, soon after his death in 2009, there appeared a quite revealing set of retrospectives on Samuelson as an historian of economics in the History of Economic Ideas (2010, XVIII, 3). Two or more biographies are in the offing; and his archive at Duke has become a major pit stop for the band of brothers and sisters laboring in the twentieth-century vineyards. I had written a little bit on Samuelson earlier in my career, but then subsequently tended to ignore him; I have only recently learned he did not ignore me in the interim. (More on this later.) There already exists such a daunting mass of commentary that it seems quixotic to make the heap a little higher. But then, that’s never stopped me before.

An extended sojourn at his archives, courtesy of Duke CHOPE, has impressed upon me the extent to which Samuelson was a fascinating, even seductive, personality. Let me confess (with trepidation) before this audience that most economists are boring people. Not many can aspire to the sheer page-turning propulsion of Lord Skidelsky’s version of the life of Keynes. I am still unsure whether Samuelson fully measures up to that blockbuster standard, but only immersion in the archives can make you appreciate that, already in the mid-1940s, his characteristic voice was fully formed: funny, discursive, learned, self-confident, always ready with a good anecdote or apposite aperçu. On paper, at least, he was the kind of guy whom everyone in a local bar might have ambitions to emulate, or, at least, buy the next round for. I offer just one example from a 1945 letter to Henry Simons at Chicago:

This was just this sort of ‘hail fellow well met’ that was well-placed to cement the neoclassical orthodoxy in place in the US, something, thanks to Malcolm Rutherford and others, we now know to have occurred only after WWII. But, of course, he accomplished far more than that. Through his Foundations and his undergraduate textbook, he fashioned the persona of the modern orthodox economic scientist out of whole cloth, demonstrating by example what it was you would need to do to become accepted by the professional club that was rapidly becoming consolidated at MIT, Yale, Cowles at Chicago, RAND, and at the other lesser aspirant citadels of the new post-war order. You had to accept the neoclassical framework without cavil, build little models that solved well-posed puzzles in the space of thirty pages or so by finding equilibrium using maximization, and act like that practice would solve all the Big Concerns pressed upon economists by the keening populace. Within that framework, he was an astounding jack-of-all-trades, spinning off whole subdisciplines within economics with a judicious paper here or there, so that others might devote their whole lives to that one newly plowed furrow. One major holdout against this rush to judgment was the clutch of economists located at Samuelson’s old alma mater,the University of Chicago. Precisely because they seemed to conform to so many of his precepts, yet reject his version of science, this recusant tendency gave him grief throughout his life; it partially explains a grace note of regret that I believe suffuses his later work. Moreover, it gave rise to that deeply deceptive distinction between ‘freshwater’ and ‘saltwater’ economics, something that still bedevils modern comprehension of the fault lines within the economics profession, long after it ceased to refer to anything analytical.

I am not here tonight to psychoanalyze Samuelson; indeed, I generally try to skirt the biographical imperative that seems to fascinate so many of my colleagues. (Maybe I am congenitally dubious about the empathetic antennae of trained economists.) I can’t fathom what growing up in Gary, Indiana, might have done to him later as an economist, for instance. Rather, I am here concerned to explore Samuelson’s own attitudes towards the history of economic thought, and the ways in which he sought to bend history to fortify his triumphant orthodoxy. These days, it seems a simple epistemic Efficient Markets Hypothesis trumps the need to provide vindication through historical legitimation. As Robert Barro opined about the recent crisis, “As long as they keep paying us, we must be right.”

Things were different two generations ago. In Samuelson’s books, all useful intellectual history should be rendered subordinate to the imperatives of boundary maintenance surrounding what qualified as ‘sound economics’; over time, this created some problems for the Paul Samuelson who was genuinely curious about the past, and far more familiar with it than many contemporaries who called themselves historians. As time went on, his notions concerning the correct uses of history began to wreak some havoc with his internal image of Paul Samuelson’s place in the grand archive of the history of science, as he later conceived it. The irony of tonight’s story is that his personal and very real solicitous concern for the virtues of history culminated in the banishment of the history of economic thought from the triumphant economic orthodoxy, in the world he had made.

II. SAMUELSON’S THEORY OF HISTORY

The first thing one gleans from the Samuelson archives is that he had read a fair amount of intellectual history, far more than he lets on in his published work. Given his project, it comes as no surprise that much of his reading was of the history of the natural sciences. There are, for instance, three whole boxes of notes in the archive on thermodynamics and its history.Footnote ³ In a couple of places, he mentions the work of Clifford Truesdell, whose work used to be very much a narrow acquired taste, even among the rarified ranks of the community of historians of rational mechanics and the exact sciences. In other places, he mentions not only the ubiquitous Thomas Kuhn, but also Norman Hanson, Michael Polanyi, and Alexandre Koyré. A few choice citations of incidents from Newton’s life suggests he had read Richard Westfall’s biography of Isaac Newton; in another place, he discusses J.J. Thompson’s autobiography. He has an epistolary exchange with Gerald Holton over whether Lorentz legitimately could have scooped Einstein with regard to the theory of relativity.Footnote ⁴ And then there are the frequent citations to original works of economists of the previous two centuries, both famous and obscure, too numerous to recount.

Samuelson may have read those histories but, mostly, he didn’t like them. Musing on the experience, Samuelson often hinted at a frame tale where there abided a species of populist general history of ideas, and then there was something else, a more tough-minded and legitimate history of economic science.Footnote ⁵ It may have been that he learned this attitude from Schumpeter’s History of Economic Analysis; but there is also the possibility that he was channeling Hans Reichenbach’s distinction between the context of discovery and the context of justification, a key doctrine in the logical positivist movement dominant in Samuelson’s youth. In his impression of the wider history of ideas, only a very few economists had ascended to the status of general interest, and the ones who did, such as Smith, Marx, Veblen, or Keynes, served as stand-ins for a few Big Themes in cultural history. Such popular history of ideas neglected other figures whom Samuelson judged to have actually been far more consequential, like Wicksell, Walras, or Irving Fisher. Samuelson felt that a generalist intellectual history was, therefore, little better than a grab bag of just-so stories, entertainment for the rabble, or, as he declared to the sociologist Robert Merton Sr., “history is a myth not agreed upon.”Footnote ⁶ Elsewhere, he wrote, “one must never make the fatal mistake in the history of ideas of requiring that a notion be ‘true’. For that discipline, the slogan must be, ‘The customer is always right.”Footnote ⁷ It is not clear from the archives whether Samuelson ever understood the full welfare implications of his skepticism towards the marketplace of ideas—but that is the topic of a different talk. In his estimation, modern historians of ideas tended to be attracted to stories recounting the ways the target protagonist in question came to arrive at their discoveries/inventions, which included all the dead ends, false starts, detours, elisions, and travesties committed along the way. This, he felt, was utterly useless from the standpoint of the working economist:

Samuelson summarily dismissed this species of history of ideas; no touchy-feely hermeneutics for him. His vision of good, solid history of science counseled dispensing with context, Weltanschauung, alternative competing research programs, and ‘mistakes’ in favor of awarding gold stars to a few, select, individual mathematical exercises—which, if not actually already there in the text, he would himself cheerfully supply, foisting them upon the hapless author—and which seemed to lead, almost inevitably, to his own work. It resembled the most Hegelian of exercises, with Thought Thinking Itself inexorably to its foreordained conclusions, grinding away in an arid environment hermetically sealed off from the rest of human endeavor. For someone who never missed an opportunity to sneer at Mises’ a priorism, it was an amazing catechism of idealism. It goes without saying that Samuelson’s favored genre of scientific history turned out to be devoid of any social epistemology by construction, or, indeed, much of any epistemology whatsoever. Once, he even confessed that his own version of scientific history was, “in many ways … easier to write precisely because it need not involve the determination of social influencings.”Footnote ⁹ The preferred modality was not even to engage in close textual exegesis, but rather to deploy sweeping generalizations ex cathedra in pursuit of a pitiless reordering of the canon. It was paradigmatic of the Samuelsonian style to continually praise empiricism, but rarely offer any in his published articles; his work in history was hardly any different. Somehow, all that mattered in the end was whom it was Samuelson had deemed should be labeled as great, greater, greatest, or out of the running entirely.Footnote ¹⁰ It was almost as if he were setting himself up as a one-man judge and jury on American Idol, stringing along a sad parade of economists vying for a prize, telling them which tunes to sing, and then crushing their egos with his insouciant, cruel, and disparaging verdicts. Here, history was a moral tale that would of necessity impose agreement, if only because no one else managed to get a word in edgewise.

Perhaps the strangest thing of all about this species of ‘history of economic science’ was that Samuelson apparently remained convinced it would produce dicta that would dependably stand the test of the ages. Axel Leijonhufvud once paraphrased this attitude as imagining a disciplinary formation where contemporary answers kill questions so thoroughly that all sense of alternative possibilities disappears. Some historians responded that this verged on confusing solipsism with guaranteed validity. Samuelson had read enough outside economics to be aware that his own favored version of the history of science had lost support and credibility over the course of his lifetime, even among historians of the natural sciences: that is why, by the later 1980s, he resorted to a semi-ironic advocacy of “Whig history” (1987) or “Presentist history” (1991). There had been a brief interlude in the 1970s where a few other economists, such as Hans Brems, Samuel Hollander, and the younger Mark Blaug, joined him in pursuit of this species of the history of economic science,Footnote ¹¹ but the practice petered out fairly quickly. By the late 1980s, Samuelson realized he had lost the favor of the professional historians of economics: “However much [these papers] please their author, they do seem to gore the oxen who trod the fields of Dogmengeschichte” (1991, p. 3). Perhaps more to the point, the few orthodox economic theorists who possessed sufficient gravitas to address him as an equal did not react the way Samuelson had anticipated. A handful of mainstream economists who shared his passion for history, such as William Baumol or George Stigler, flatly rejected the cogency or relevance of his kind of history of economic thought.

There is a wonderful example of this in the archives. In a whole series of letters, Baumol challenges Samuelson’s interpretation of Marx, even sending him copies of certain selected passages and letters from the Marx–Engels correspondence. Samuelson does not give an inch, at which juncture, Baumol suggests the problem has to do with what counts as ‘history’: “In sum, my complaint is that you are interspersing your invariably first rate analysis with misleading and inaccurate doctrinal history. Marx may be as vulnerable to your attack as you believe, but not for views which he never held.”Footnote ¹²

Yet, somehow, such push-back did not prompt Samuelson to rethink his firm convictions concerning what constituted good history. Perhaps Carl Schmitt was right after all. Or, alternatively, perhaps this was rooted in another characteristic Samuelsonian practice, which was never to acknowledge the specific arguments, and even sometimes the identity, of his intellectual opponents. Once he had achieved early fame, Samuelson no longer seemed to be in the business of striving too hard to convince others of his ideas. As he once wrote, “One learns it is better to avoid an argument than to win one.”Footnote ¹³ If you were someone of insignificant status, and if you ventured to challenge him, he would ruthlessly cut you off without warning; if, conversely, you were someone who mattered in his books, then his public intellectual commitments would go all vague. As he told Baumol, “When Milton [Friedman] and I became identified with different views, I determined to soft-pedal my disagreements. He was an important influence in pushing economists rightward toward libertarian ethics. However, both in micro and in macro he was a Luddite.”Footnote ¹⁴ No niche reserved in the pantheon for Milton. Forget the fact he had successfully moved the entire economics profession in his direction from the 1970s onwards, an achievement of Samuelsonian proportions and ambition.

Giuseppe Bertola related an anecdote concerning a special lecture he heard Samuelson deliver to the first-year graduate micro class at MIT in 1985 about one of his very few intellectual donnybrooks in a long and successful career, the (now forgotten) ‘no reswitching theorem’ in the 1966 Quarterly Journal of Economics. Samuelson admitted to his tyro admirers embarking on their career that his theorem published there had been flawed, but then said: “So we were wrong at that time. Does it matter? No. We are here. You are here. The other people will be disappearing, funeral after funeral.”Footnote ¹⁵ There is that voice again: clear, direct, convivial.

In the end, Samuelson’s scientific history, which had started out with such good intentions, eventually amounted to little more than a rear-guard skirmish over who would get to write the obituaries. But, sad to say, no matter how powerful you get, you can never manage to write your own.

III. SOME SAMUELSONIAN LEGACIES

Just because Samuelson ultimately didn’t think very highly of historians and their obsessions does not mean he did not have profound impact upon the shape of the history of economic thought. Indeed, quite the contrary, as it seems many of the stories currently told to students in the sparse few remaining courses in the history of economics have frequently either originated with, or were endowed with staying power by, some Samuelsonian dictum or throwaway line. Conveniently, other historians have begun to sort this tangle out in specific cases, so that I can refer to them briefly here tonight, and not parse them in fine detail. Furthermore, I can avoid reconsideration of Samuelson’s hallmark Canonical Classical Model, since so many historians had objected to it during his lifetime that it would be otiose to recapitulate that chorus of repudiation here tonight.

Here are a few other unfortunate historical generalizations one can trace back to Samuelson:

[1] The whole fallacious drive to recast Adam Smith as a proto-neoclassical by recourse to the ‘invisible hand’ metaphor, now ubiquitous, was largely instigated by Samuelson.Footnote ¹⁶ There are a number of articles repeatedly pointing out that Adam Smith intended nothing resembling Pareto optimality or general equilibrium by his use of the metaphor; but Gavin Kennedy (Reference Hollinger and Capper2010) has demonstrated that it was Samuelson’s 1948 introductory textbook that first ushered us down that particular cul-de-sac.

This is corroborated by a graph of the occurrences of the term “invisible hand” from 1811 to 2008 in JSTOR’s Data for Research beta tool.Footnote ¹⁷ Here, each article that contains the term is counted once. If Samuelson was not the absolute original culprit, it appears he certainly fanned the flames.

[2] It was Samuelson and his early student, Lawrence Klein, who jointly launched the canard that Keynes was some alien species of idiot savant who had no cogent conception concerning what he was saying in the General Theory, a trope that came home to roost in the later rational expectations movement. Here was Samuelson’s rather backhanded obituary of Keynes: “In [the General Theory] the Keynesian system stands out indistinctly, as if the author were hardly aware of its existence or cognizant of its properties” (1946, p. 190). Also: “Keynes seems never to have had any genuine interest in pure economic theory.”Footnote ¹⁸ Lawrence Klein then chimed in shortly thereafter: Keynes “did not really understand what he had written, and chose the wrong thing to publicize as his innovation” (1947, p. 83). With friends like these, who needed the monetarists? The push-back by Axel Leijonhufvud and others was an attempt to correct the record, but seemingly with little success. As Leijonhufvud said in his lecture “The Uses of the Past,” “Either there were other possible futures inherent in the General Theoryor the book is dead.”Footnote ¹⁹

[3] Perhaps unexpectedly, it was Paul Samuelson who promoted the rumor that there was a ‘First’ and ‘Second’ Chicago School of Economics, linked by strong patterns of continuity. Of course, this did not preclude the denizens of Hyde Park from weaving their own individual genealogies out of whole cloth; but it was Samuelson, who was not party to the crucial post-1945 developments, who first sought to tell a story about the profound doctrinal importance of Frank Knight—as ‘founder’ (Samuelson Reference Samuelson1977, p. 915)—along with Jacob Viner, and Henry Schultz, to the more renowned Chicago School of Friedman, Director, Wallis, and Stigler.Footnote ²⁰ This insistence on parochial continuity was revealed dubious from the 1970s onwards, as Stigler and Friedman progressively repudiated Knight, Viner had renounced the notion of an earlier Chicago School (van Horn 2011), and Friedman and Director disparaged Schultz in various interviews. When economists such as David Laidler and Donald Patinkin sought to complicate the linear narrative, Samuelson rejected them out of hand.Footnote ²¹ Because he was leaning so hard on his own personal recollections, Samuelson apparently had no way of knowing the extent to which Friedrich Hayek had played a major role in reconstituting the post-war politics of the Chicago School, and the ways in which the Mont Pèlerin Society had sidelined Knight and, instead, served to connect key Chicago economists to the wider transnational political project, which resulted in a neoliberal political economy.Footnote ²² Indeed, he had been occupied so long in insisting that all good American economists tended to agree on basic economic theory that he missed much of the major rupture in post-war political economy.

[4] Orthodox neoclassical economists say the most extraordinary things about ‘utility,’ as every historian of economics knows. It exists; it doesn’t exist; it’s ordinal; it’s cardinal when we make wagers; we need it for our models to work; we don’t need it at all; it is a psychological entity that ‘causes’ behavior; it merely ‘rationalizes’ behavior in a post hocagnostic fashion; we must presume it imposes (our version of) ‘rationality’ as consistency; we can show people as being regularly irrational by appealing to it. One bit of terminology at the center of this cacophony is “revealed preference”; and while Samuelson doesn’t deserve all the blame, his strange inability to clarify its meaning over his lifetime has served to pump excess noise into the history of neoclassical economics. Here, I am just interested in how Paul Samuelson the historian tended to trip over Paul Samuelson the economist, when it came to defending his legacy.

Stanley Wong (Reference van Horn1973; 2006) performed the neat trick of showing that when one applied the principles of revealed preference to his own articles, Samuelson was himself revealed to be inconsistent concerning revealed preference. In his 1938 Economica article, Samuelson suggests revealed preference is an alternative to utility theory, which dispensed with all non-observational terms; after Hendrik Houthakker produced the strong axiom derivation of conventional demand theory in 1948, Samuelson then praised revealed preference theory as the culmination of orthodox utility theory, because they were identical. After Wong made the case for this reversal in the 1973 American Economic Review, Samuelson proceeded to insist he had never changed his mind, or underwent a change of allegiance. Examination of the Samuelson archive reveals he never actually responded to this intriguing contradiction, although one letter displays Wong complaining that Samuelson never followed up on denials tendered in person of being influenced by positivism, or questing for a non-orthodox formalization of demand.Footnote ²³ Wade Hands (Reference Freedman2011) has taken the inquiry one step further, showing that the most diverse and clashing contemporary research programs all claim they are formulating and operationalizing something they all call “revealed preference theory,” but, in practice, little substantive holds them all together other than vague family resemblance. Thus, Samuelson the historian promoted the widespread myth that neoclassical microeconomics in its modern manifestation is one, monolithic, consistent structure, reducible to a few empirical primitives, when patently, at least for Samuelson the historical figure and the minions who followed, it is not.

[5] Perhaps the audience can indulge me for briefly recounting a Samuelsonian historical red herring more closely related to my own work. As some of you know, I have argued that the origins of formal neoclassical theory can be found in the imitation of mid-nineteenth-century energy physics, and that this explains the otherwise miraculous supposed ‘simultaneous discovery’ by a usual roster of suspects in the 1870s. Samuelson began to take the position in the 1990s that neither he nor anyone else in his pantheon were ever beset with physics envy; and that any appearances to the contrary were due to the accident that some of the same mathematics had proven useful in both areas of research. This dismissal has been dutifully repeated by many down to the present.

I will not bore you tonight with excerpts from the letters from Edwin Bidwell Wilson directing Samuelson in 1938 to attend to specific pages in Willard Gibbs’ physics papers in his work on equilibrium and stability.Footnote ²⁴ Samuelson did not hide this dependence in his Foundations, but neither did he think very deeply about why the parallels were valid. Samuelson’s repeated appeals to physical theory had been noted by many historians of science well before anything I ever wrote, unbeknownst to me. Vernard Foley wrote to Samuelson in 1968: “The achievement of the thermodynamic synthesis in physics seems to have stimulated the construction of several systems of ‘social physics’, and even a few religious sects. I am approaching the problem from the point of view of the history of science.…”Footnote ²⁵ Samuelson, in response, admitted “the triumph of Newton gave impetus to a belief in self-regulating economic equilibrium,” but denied any more proximate influence of classical thermodynamics. Later on, the historian Robert Killingsworth raised the issue in the context of John von Neumann’s suggesting economics needed a more modern mathematics, not so rooted in older Principles of Least Action.Footnote ²⁶ When Duncan Foley sent him a paper on the thermodynamics of prices, he wrote back: “Herman Goring said, ‘When I hear the word culture, I reach for my gun.’ Likewise, when I hear the word ‘entropy’ in an economics ms., I reach for my tennis racket. This is a conditioned reflex not a praise-worthy attitude of an open-minded scientist.”Footnote ²⁷

Perhaps my audience will then realize how taken aback I was to find a folder in the archives marked “Philip Mirowski,” and contained within it a four-paged, typed, unpublished, undated text. I quote here an excerpt:

I will spare you the postit-by-postit replay, but, thoroughly predictably, Samuelson wins and Mirowski loses. I will not count just how many thunderbolts it takes for Jove to squash a bug, but will give you a flavor of the argument by quoting another passage:

It is one thing to observe it as spectator; it is another to feel what it is like to be subjected to a Samuelsonian Whig reading of something you wrote first-hand. First off, there is the remarkable Parnassian phenomenon of Paul reading himself in order to interpret himself, and you, to himself. Then, there is the utter disregard for what More Heat than Light actually asserts, which is that the first generation of neoclassicals imitated rational energetics, not classical thermodynamics, which was codified and stabilized only after their works. Therein I suggested that the role of conservation principles had been a problem for the neoclassical project, not that I or anyone else had actually claimed that potential utility was conserved. He then signals that, in his own mind, the pertinent mathematical key is rather the analogy between the ordinal entropy of classical thermodynamics and ordinal utility—something that would be what he derived from Gibbs—yet, this is a metaphorical comparison nowhere discussed in More Heat, but suitably dispatched in later papers by Duncan Foley.Footnote ²⁹ And finally, there is the imperious tone of “I don’t do analogies,” when numerous other historians had independently noticed this practice specifically in this regard. Were they all deluded?

Be that as it may, arguing with the deceased is a distressingly one-sided activity, guaranteed to disappoint. It was a letdown for Samuelson, and is a letdown for me. Indeed, that may be one of my take-home lessons for the evening. But I can’t help wondering: if it mattered so much to him, why didn’t he write?

IV. THE FRUSTRATIONS OF STRIVING FOR CONTROL OVER YOUR OWN HISTORY

Samuelson was certainly justified when it came to his wariness towards generalist intellectual history. For the sake of comparison, I consulted the top undergraduate reader in American intellectual history today, The American Intellectual Tradition,volume 2, covering the period 1865 to the present (Hollinger and Capper 2011). True to Samuelson’s expectations, none of the figures to whom he awarded his scientific gold stars make the table of contents. There are, nevertheless, four economists represented therein: Thorstein Veblen, Gunnar Myrdal, Walter Rostow, and Milton Friedman. Given the evidence we have cited from the archives, perhaps the last name would have proven the bitterest pill of all.

In Samuelson’s books, historians were good for nothing but disappointment and jibber-jabber. And yet, from a very early age, Samuelson seemed to be avidly concerned about how the verdict of history would treat his own career. To be sure, the Clark medal, the Bank of Sweden Prize, the best-selling textbook, the serial festschrifts, and the MIT heritage would dictate that his was a name to be reckoned with. For a while, it may have appeared that Samuelson had jump-started a special Whig history written mainly by economists who would tell his story the way he wanted it told. That was the real meaning of the oft-quoted finale of his AEA Presidential Address: “Not for us is the limelight and the applause.… In the long run, the economic scholar works for the only coin worth having—our own applause.” Footnote ³⁰ One can imagine the gales of applause that greeted this ending remark.

He thought that the world economics profession would faithfully keep his lamp lit. Nevertheless, as he grew older, Samuelson witnessed two trends that perturbed his generally upbeat nature. The first was the already mentioned failure of his Whig historiography to catch on among the economics profession. Instead, the profession progressively shed any and all history of economics; the marketplace of ideas had unexpectedly rejected the very possibility of the construction of the sort of pantheon that he had oriented his entire career to occupy. When this was pointed out to him, he responded with the counter-factual: “I shall say, touché rather than take defence [sic] in the observation that the field [of history of thought] might well have shrunk even worse under the older techniques.”Footnote ³¹ Historians were not amused. This, in turn, provoked some quasi-suppressed trepidation about how the despised scrum of subsequent historians would treat his legacy. As he aged, Samuelson began making harsher comments in correspondence about his contemporaries and the recently departed, as if to better situate and fortify his own position, but he would usually append some such disclaimer: “Please do not quote my heretical views in these matters. When LSE asked Viner what they should do with sour Edwin Cannan’s papers, he wisely advised: Burn them.”Footnote ³²

But, of course, he never once gave instructions to consign the Samuelson papers to the flames. Instead, they were lovingly preserved from the mid-1930s onwards, and deposited for all to see at Duke. Hope springs eternal.

This leads directly to the second, possibly more controversial, trend, which also factored into Samuelson’s trepidations about his legacy. When it came to steadfast loyalty, professional economists may have proven only marginally more trustworthy than the historians. As time went by, Samuelson witnessed many of his most cherished doctrines lose their purchase and legitimacy within the orthodox profession he had done so much to bring into being. One can start with the obvious: the Phillips curve had his fingerprints all over it, and (whatever the actual sequence of events) its demise was frequently cited as one of the major watersheds in the later repudiation of the neoclassical Keynesian synthesis. Although there are still a few recalcitrant holdouts, it seems this centerpiece of the Samuelsonian corpus is dead, not even revived by the experience of the Great Recession. Yet, there were more repudiations and renunciations, both great and small.

I have already mentioned the heritage of the theory of revealed preference, which wavers between modalities of superfluity and the endorsement of propositions that Samuelson had resisted much of his life. Samuelson’s hallmark dependence on the theory of public goods had been more or less demolished in practice by the public-choice school. As Arrow and Hahn wrote in General Competitive Analysis, “The correspondence principle isn’t.”Footnote ³³ Indeed, much of his core catechism that general equilibrium theory would underwrite and motivate the ‘well-behaved’ demand curves so central to many of his ‘applied’ models was thoroughly undermined by the Sonnenschein–Mantel–Debreu theorems of the 1970s.Footnote ³⁴ The ‘monopolistic competition’ movement long ago was ambushed in industrial organization, never to really recover. Samuelson notoriously disdained game theory, and much of the investiture of the neoclassical agent with strategic rationality. He had no patience with ‘behavioral economics,’ which would leave his welfare economics in shambles. Samuelson’s version of Heckscher–Ohlin has been repeatedly compromised in empirical exercises and in the experience of globalization. And, finally, there was the undeniable fact that Chicago had won the political war with MIT: his beloved neoclassical tradition had become affixed to the neoliberal political tradition in the minds of most people who never ventured beyond a principles course in college, because now, their principles of economics textbook renounced much of what Samuelson’s early editions had endorsed, however much they might mimic it in outward style.

Although repudiation of his trademark ideas may have stung, there was also the common phenomenon that disciplinary attention to any winner of the Bank of Sweden Prize often peaks right around the time of the bestowal of the award. This happened to Samuelson as well, as revealed in statistics of JSTOR citations compiled by Avner Offer and presented in Figure 2.Footnote ³⁵ Although citations are a poor proxy for engagement, nonetheless, it was undeniable that Samuelson was becoming progressively less and less the object of attention within the orthodox profession.

Figure 1. Number of articles mentioning “invisible hand” in JSTOR database, by year.

Figure 2. Number citations of Samuelson in JSTOR, normalized to ‘Arrows.’

Samuelson was too perceptive and clever not to notice that the orthodox profession was sloughing him off, doctrine by doctrine. While never directly reprimanding his students, late in life, he began to register his demurrers in various public outlets. But, most distressing, when the economic crisis hit in 2008, it dawned on him that he might be living through his worst nightmare: after entering economics because of the Great Depression in his youth, and having devoted his life to preventing such a debacle’s from ever happening again, here in his twilight years he glimpsed the fact that the disaster had recurred, and that the profession he had shepherded and nurtured in the protocols of science may have borne some responsibility for it.

Just so we do not interpret this as an isolated, momentary, lapse of judgment in the midst of crisis, Samuelson repeated this analysis in one of the last interviews he gave before his death:

He also went on record in early 2009 on the PBS News Hour Footnote ³⁸: “Fiendish Frankenstein monsters of financial engineering were created, a lot of them at MIT. Some of them by people like me. Nothing like this was present in 1929.”

Histories of theories of finance by Peter Bernstein and Donald MacKenzie regularly traced these financial innovations back to some early work of Paul Samuelson, as he acknowledged. In effect, in the wake of the crisis, he had observed yet another of his proud legacies transmuted into a poison pill; in this case, one that he himself believed had brought down the world economy. By contrast with his epigones, he was actually willing to admit that his beloved economics profession had borne some of the responsibility. The aftermath of the crisis had also demonstrated just how flimsy was the vestigial legacy of the Keynesian neoclassical synthesis within the orthodox profession. These admissions must have shaken his notorious self-assurance. The genre of Whig science had led to this: the smug sleep of scientific reason had spawned monsters of calamity and ruin.

V. TWILIGHT OF SCIENTIFIC HISTORY

Samuelson espoused a very ruthless instrumental view of doctrinal history. Only late in life did he begin to glimpse that a similar orientation, adopted by his epigones, might come to be repurposed and applied to his own legacy. Instead of every significant development in the history of economics having led up to his own accomplishments, they might be reconstructed as largely having passed him by. Of course, he was not personally responsible for the banishment of the history of economics from the orthodox curriculum; but in suggesting too avidly that intellectual triumphs occurred one funeral at a time, he could not have been unaware that his own impending funeral would mark the dissolution of much that had mattered to him. Innocent of most history, the younger generation could and would assert pretty much whatever they liked about the supposed economic consequences of Paul Samuelson, since now no one much cared one way or the other to make an issue of it, much less possessed sufficient historiographic gravitas or expertise to brook challenge to the thicket of idiosyncratic Whig interpretations. We now inhabit a strange world of DIY history: the past becomes whatever you personally wish it had been. Indeed, numerous contemporary economists have already been furiously rewriting the history of the current crisis, so as to pronounce absolution over the profession and escape any culpability for the onset and prolongation of the worst economic contraction since the 1930s. Certainly, there is no one of Samuelson’s stature now alive who would take upon themselves the onus of public admission that they should be held culpable, even if only indirectly. Instead, the world is full of economists who now claim they had it right all along.Footnote ³⁹ The absence of any serious intellectual chronicle of the crisis, as a consequence of the prior banishment of history of economic thought, has proven a boon to all manner of economist entrepreneurs in purveying their current wares.

Samuelson’s historiographic dictum had come to grief: history was just too dangerous to be left to the economists.