June 1991

Book Review Column Mirowski's Screed: A Review of Philip Mirowski's More Heat than Light: as Social Physics, Physics as Nature's Economics.' Cambridge: Cambridge University Press, 1990, Pp. xii+ 450. $59.50 IBSN 0-521;.35042.. s.

Kevin D. Hoover consistently uncharitable readings of almost University of California, Davis everyone- Veblen, Georgescu-Roegen and a few obscure figures in the history of economic More Heat Than Light! - that sums it up thought excepted. quite accurately. Alas, it will take more than I four words to convince the reader of the Mirowski's argument is about metaphor soundness of such a stark judgement. and the role of metaphorical exchanges in the Mirowski's book is a sustained .attack on development of physics and economics. In the the foundations of modem neoclassical development of energy physics, which is central economics. He gives a succinct statement of his to this book, accounting notions and notions of thesis: economy of action are borrowed from The only way to fully comprehend economics. Mirowski cites the suggestive theory in economics is to situate it example of Joule, who may have gotten his wihin .. .. . the metaphorical simplex of inspiration for his research into the mechanical energy, motion, body and value, and to equivalent of heat from the carefully maintained regard it as part and parcel of the same accounts of his family's brewery: energy was like structures that undergird Western money - the diligent bookkeeper had to account physics. [pp. 141-2) for every last tuppence. Mirowski maintains If true, this thesis has severe consequences for that economics readily borrowed from physics our understanding of the history of economic· as well, and, by the time of the rise of thought and for the methodology of economics. neoclassicism, the exchange was pretty much Beyond economics, Mirowski aims to one way from energy physics to economics. undennine the nexus between social theory and Mirowski visualizes the metaphorical the natural sciences. A key slogan runs: complex involved in these exchanges as a "Physical metaphors used to describe social triangular pyramid, with "energy" at its apex processes are spuriously grounded in the natural and· "motion," "body," and "value," at the phenomena" (p. 318). vertices of its base. In chapter 2 he reviews the I hold no special brief for neoclassical development of energy physics from the late economics. Throughout his book, Mirowski 18th century through the early 20th century as scatters many telling criticisms of the details of elaborations of one or other of the metaphorical neoclassicism and of the inappropriate imitation relations of the three faces of this pyramid. of the physical sciences by economists. He offers Despite its stylistic infelicities, this is the most some attractive ideas for the development of the fascinating part of the book, even for someone discipline. Yet, reading this book gave me a who has heard the story before. slowly rising feeling of outrage. Taken as a Mirowski argues that the notion of energy whole, it is an outrageous book: neither the and its associated conservation law were not history nor the methodology are persuasive; the simultaneously discovered as standard histories scholarship is often slapdash; the tone is report. First, each of the putative discoverers, intemperate; and the style is often obnoxious. Mayer, Joule, Helmholtz and Colding, were Mirowski's hatred of neoclassical economi~ elaborating the metaphorical complex rather borders on the pathological: one sometimes than finding an "energy" that was out there wonders if his mother didn't run off with a independeritty of their own understandings - i.e., neoclassical economist, leaving little Phil bereft energy was not discovered. Second, the in the cradle. Mirowski strikes a flashy, bullying .. discoverers" were working on different faces of tone throughout the book, patronizing the the energy pyramid. This explains why, in reader, economists and physicists. He offers Mirowski's view, it did not occur to them that Junel991

CI:) t~ey were all doing the same thing The central problem for neoclassical . - m fact they were not. economics is that it failed to see that copying 'Boo(!l{tcJrL'W Mirowski then goes on to physics leads to absurdity. After the trace the development of the law of the protoenergetics stage, physics developed away conservation of energy and thermodynamics. By from substance accounts of energy toward field the end of the 19th century, it appeared that accounts. Conservation of energy can be physics was achieving a grand unification with expressed technica11y as the requirement that energy as the central concept. The energetics energy be represented by an irrotational movement associated with Ostwald began to conservative vector field. The characterization view energy as the key to everything. But then, of an economy as the simultaneous around the tum of the century, it an began to maximi:zation of utility functions subject to unravel. Planck and pthers attacked Ostwald as budget constraints is analogous to the field misunderstanding the basis of the existing formalisms of physics. Had economics energy physics; and that physics itself began to developed its metaphorical borrowing from disintegrate. Quantum mechanics and the physics along the same lines as physics itself ·theory of relativity fractured the hard-won unity developed, it would have been forced to impose of physics, and the law of the conservation of the conservation law in the form of an energy was transmuted into symmetry principles irrotational vector field. But then the analogue that were specifically tailored to different of the law of conservation of energy would state physical theories which no longer formed a that the sum of utility (analogo1,1s to potential unified whole. Not only had the conservation of energy) and expenditure (analogous to kinetic energy lost its overarching status, E = mc2 energy) would have to be conserved quantity. suggested that energy might not be conserved, But that is an economic absurdity; utility and but could be converted into matter and vice expenditure do not have the same dimensions. versa. Further elaborations of relativity theory Mirowski faults for not suggested that the homely thought behind the exploring the complete implications of the law of the conservation of energy - nothing energy metaphor; and, because those comes from nothing - might be wrong: "It now implications are unsavory in the extreme, for appears possible that the universe is a free adopting the energy metaphor at all. lunch" (Mirowski, p. 392, quoting Guth 1983, p. Mirowski notes various physicists and 215). mathematicians who raised questions about the All economics, at least from the advent of appropriateness of the energy metaphor for mercantilism, was, in Mirowski's view, involved economics, and, particularly, about the in the elaboration of the same metaphorical counterpart to the law of the conservation of complex of energy/motion/body/value, But the energy. These took the form: are utility fields two-way st~t of metaphorical exchange integrable? He maintains that, after the turn of between physics and economics became the century, when these questions were not essentially a one-way street with the marginal satisfactorily answered, economists - in large revolution of the 1870s, and a divided highway measure because of their mathematical once physics itself lost its unity in the 20th incompetence - simply ignored the question of century. integrability for nearly thirty years. The sores of According to Mirowski (p.3), "..... the an inappropriate metaphor continued to fester. progenitors of neoclassical economic theory In the 1930s, an influx of engineers and boldly copied the reigning physical theories of . mathematicians raised mathematical the 1870s." He dubs this physics competence among economists to a new height. "protoenergetics." It is the energy physics that Integrability was rediscovered. Now, however, it developed out of rational mechanics before the was seen as an economically insigificant second law of thermodynamics introduced the technical point. Mirowski believes that this notion of the irreversibility of thermodynamic attitude was part of an elaborate shell game in processes. Mirowksi's claim is extremely strong: which the physics of protoenergetics continued '' ..... those neoclassicals did not imitate physics to drive the development of neoclassical in a desultory or superficial manner; no. they economic theory, while economists denied the copied their models mostly term for term and centrality of the physics metaphor for their own symbol for symbol, and said so'' (p.3). Of this discipline. For this charade · for· Mirowski claim. more anon. repeatedly questions the motives of . t.he

1 AO June 1991

economists involved Paul theory. "Ex nihilo fit nihil fit" was already Samuelson is held chiefly to blame. known to metaphysics when "economics" II referred to housekeeping and was beneath a No one can doubt that economists in the philosopher's contempt. 19th century, as well as before and after, looked The direct and conscious parallels drawn to physics as an inspiration for scientific between physics and neoclassical economics that economics, they borrowed its mathematics and Mirowski cities are invariably surrounded with found economic analogies for some of its .caveats noting that no analogy with physics will concepts, such as energy. Physics was a be exact in every detail. Jevons's clearest direct resplendent jewel in the crown of the modem borrowing from physics in his Theory of intellect. Other disciplines, not just economics, Political Economy is the discussion of the law of sought to stand in the reflected glory of its the lever. In context, however, it is evident that scintillating light Mirowski maintains, his point has to do with what sorts of however, that physics and economics were mathematics can be used in certain classes of bound together more tightly than this suggests: problems, and not with precise analogies both were elaborations of the single between levers and economical systems. metaphorical complex represented in the energy Similarly, Fisher is clear that his use of pyramid. While it is clear in retrospect that the hydrostatic analogies in The Purchasing Power energy pyramid has considerable taxonomic of Money (p. 108) and elsewhere are merely purchase in the history of physics, Mirowski suggestive, and not exact. Mirowski reproduces provides no evidence that it had any heuristic a table from Fisher's doctoral dissertation in power, that it . in any way guided the which he draws explicit analogies between development of either physics or economics. economics and physics. But Mirowski finds it Emulation of physics was part of the Zeitgeist of necessary to construct his own supplement this the 19th century: it was simply in the air; people table drawing his own further analogies in order talked about energy the way they now talk about to convict Fisher of a complete and precise quantum mechanics or relativity, usually borrowing of the physics metaphor. Nor is this without rigor and often (as Mirowski himself harmless filling in of obvious lacunae: Mirowski notes) without understanding or perspicacity. · (p. 230) must refer to his own additions to Morowski denies that the linkage was this loose: ·convict Fisher's system of involving absurdities...... the research program at each vertex At the same time, Mirowski is not at all happy [of his energy pyramid] is essentially the that Fisher attempts to insert economic same metaphor. Here is the sense in considerations into his table that are not which we are no longer dealing with one-for-one with the elements of physics. prosaic notions of intellectual It is a good thing that modem physics cross-disciplinary influences, Zeitgeist. supports the notion that something may come or epistemes. The research program from nothing, for Mirowski repeatedly draws situated at each vertex derives legitimacy substantive conclusions from what is not there. for its radically unjustifiable principles His treatment of Fisher's table is a typical from the homeomorphisms with the example of this rhetorical tactic, which we might structures of explanation at the other name the "evidential free lunch." A few further vertexes. [p. 116] · examples follow. Significantly, Mirowski does not provide a In Mirowski's view, Fisher's thesis stands single instance of the "word-for-word, convicted of contravening the logic of symbol-for-symbol" borrowing of physics that metaphorical reasoning and failing to perceive he promises. There is much equivocation. the fundamental conflict between the physics of Apparently, any time any economist uses the the conservative vector field and neoclassical words "energy" or "conservation" Mirowski economics. It happens that J. Willard Gibbs, the reports them as buying into the fine details of the imminent thermodynamicist, was one of protoenergetics program. Similarly, any time a Fisher's advisers. Unable to produce evidence of physicist uses a word like "value" or a phrase Gibbs criticizing Fisher, Mirowski like "nothing comes from nothing" Mirowsld ..conjectures" (to use his own word, p. 242) sees an appeal to economics. But all these words Gibbs's objection: have now, and had then, meanings that were not Gibbs . . . . . undoubtedly asked Fisher closely rooted in any economic or physical why Fisher's indifference lines should ~ be able to be integrated into treatise" (p. 247). The fact that Laurent - -. utility surfaces. Far, from supports the Laussane school and does not 'Boo(~VltW being a minor technical recapitulate his questions suggests to me either complaint, Gibbs probably tried to that he regarded his questions as having been make Fisher aware that the absence of adequately answered in the end or that he integrability would necessarily mean regarded the issue as being of secondary that there could exist no such. quantity importance or that he doubted his own standing as total utility, and path independence in the debate. In any case, absent Mirowski's ofequilibrium would be compromised. own interpretations of which Laurent was no What he apparently never understood doubt innocent, the evidence of Laurent against was that Gibbs wanted to know why neoclassical economics is weak. Fisher did not explore integrability as III the next logical step towards a dynamic A central problem in Mirowski's view of theory of optimization..... Fisher, history and method is that metaphors are all uncomprehending, instead went on to important, yet there is no explanation of their say that he did not need integrability for mode of influence. Indeed, it is not at all clear his theory, and indeed, he did not need what a metaphor is for Mirowski: almost every utility, period. This statement only term in every context is described as served to demonstrate that he was out metaphorical, so one wonders if Mirowski of his depth. We can date the collective recognizes words as ever having a primary neoclassical neurosis with regard to the non-metaphorical usage at all. He may not; for physics metaphor from this point. [p. he frequently expresses radical skepticism about 243) the ..thereness" of energy, of value, of motion, Fisher is convicted for wrongly responding to of body, of just about everything. The mildest objections for which there is no tangible forms of "realism" are suspect. He speaks of evidence that Gibbs made. And this is the critical metaphorical resonances. On the one hand, he juncture in the history of neoclassical accuses economists of detailed copying; on the economics! I submit that there is a very good other hand, he asserts that they (consciously or reason to believe that Mirowski's conjectured unconsciously) strive to cover up their tracks. exchange never took palce: Gibbs was Fisher's Fisher again provides a good example: conjectured exchange never took palce: Gibbs His chosen tactic was to avoid was Fisher's adviser, and he signed off on his discussion of the conservation of energy thesis, which he surely would not have done had at all costs, even if it meant some he imagined it to be fundamentally flawed. Or misrepresentation of the model does Mirowski have further conjectural appropriated from physics. [p. 230) "evidence" that Gibbs was unusuaJJy negligent Here Mirowski takes Fisher's primary task to be in fulfilling his academic obligations? the appropriation and elaboration of the physics A second example comes from Mirowski's metaphor. An alternative and more reasonable recounting of the inquiries of the interpretation was that Fisher's primary task mathematician, Hermann Laurent. Laurent, at was to make sense of economic problems. If the different times, wrote letters fuU of searching physics metaphor implied the absurdity that questions about the mathematics, particularly money and utility were directly commensurably about integrability, of Walras's and of Pareto's (not that there is any evidence that Fisher systems. Mirowski presents the economists as, understood this implication), then so much by turns, dunder-headed and evasive. He worse for the metaphor. concludes that they failed to communicate Mirowski subscribes to the metaphorical because .. Laurent understood the physics, and imperative: to use a metaphor is to commit Pareto [and, by a weJJ-supported inference from oneself to the complete mapping of that Mirowski's explicit comments, Walras] did not" metaphor onto the subject at hand. But surely, if (p. 24 7). Mirowski then wonders at Laurent's my love is like a summer's breeze, I still have no Petit traite d'economie politique mathematique reason to think that an anemometer would help for supporting the Lausanne school. ..His to gauge the intensity of her devotion. questions •about integrability'..... were never Mirowski recognizes this but asserts that adequately answered by the protagonists, and so scientific metaphors are different from poetic it appears he just passed them by in his own metaphors and "have different criteria of

142 Junel991

r::c::>efficacy and success" (p. 278). But without allowing false trading) (p. 252). The . why? Mirowski's only halfway Arrow-Debreu model is similarly free of taint. 'IJoo{'R,_tVltW persuasive reason is that " ..... one What an admission! Walras saw the necessity of of the most attractive aspects of analogical . altering his model, not because of the energy reasoning is the prefabricated nature of an metaphor (Mirowski assures us that he was too interlocked set of explanatory structures and incompetent to pursue that very far) but because constructs, allowing quickened evaluation of of internally generated problems with his earlier logical coherence" (p. 272). But if the chocie account. comes down either to adapting to the Mirowski's view of neoclassical economics requirements of the economic problem or to is narrow and blinkered: it appears to be further prosecuting a metaphor beyond the coextens.ive with utility-based general equilibrium point at which it is apt, what advantages does price theory. But few self-described neoclassicals such prefabrication provide? Aside from the worry much about general equilibrium. Marshall, advantages of prefabrication there simply is no Friedman. and most practitioners of applied argument in favor of the metaphorical microeconomics, including those who invoke imperative. There is only bald assertion. Poetic general equilibrium most heavily (estimators of metaphors, ip Mirowski's view, need not be demand systems and computable general (indeed, should not be) fully prosecuted; but to equilibrium modelers) do not expect a perfect deny that scientific metaphors must " ..... is to match between neoclassical price theory and the deny the possibility of scientific metaphor" (p. economy. Instead. they find the model 279). This from an avowed enemy of scientism in suggestive of important aspects of the economy; econom1cs. they are aware of many of the humbler criticisms Recall that Mirowslc.i is suspicious of aJI that Mirowski, along with many others, have forms of realism, so the imperative of the made (e.g .• absence of an auctioneer, unstable metaphor replaces the imperative of the preferences, failure to characterize process); and economic problem. Explorations of they are not slaves to the metaphor. lexicographic preferences are condemned, for In this, they are hardly different from the example; because they reflect ignorance of the physicists. As Mirowski tells it 19th-century root metaphor of neoclassical economics, no physicists dreamed the Laplacian dream of a matter how much they may suit the economic complete dynamic, deterministic model of behavior of people (p. 366). "Contrary to the everything: a giant Hamiltonian equation in ideology of neoclassicism, we are not so which one need only specify the initial indifferently free to choose [the aspects of the conditions and the future and past of the physics metaphor we like or dislike]" (p. 272). universe would unfold before one's eyes. But The metaphorical imperative helps to structure Pioncare· demonstrated that Hamiltonian inquiry" ..... which might otherwise be even dynamics had severely limited applicability. more rife with rampant individualism ..... than is Nonetheless, Hamiltonian dynamics are still already the case" (p. 279). The intellectual central to many areas of physics. Mirowski connection between nominalistic metaphysics observes: "Thus, if the Poincare theorem was the and intolerance and totalitarian impulses could rude awakening from the Laplacian Dream, not be more neatly illustrated. most of the dreamers merely rolled over and The most frequently employed rhetorical went back to sleep" (p. 73). Physics has seen no paradigm in Mirowski's accounts of economists reason to abandon Hamiltonian dynamcis or is, "When did you stop beating your wife?" energy or its conservation laws (pp. 90-91 ). Walras, for example, is, on the one hand, The physicists have very good reasons for savaged for his mathematical incapacities and maintaining 19th century mechanics in spite of his inability to complete the analogy between its failure as a model of everything. Similarly, physics and his economics. On the other hand, neoclassical economics has its uses as well as its when Mirowski completes the analogy on his flaws. Marshall saw this clearly. Mirowski, behalf, Walras is convicted of economic however, is no friend of Marshall: ..There was absurdity. Interestingly, Mirowski admits that no solution, so Marshall papered the whole whatever the problems of the first three editions thing over with a florid pattern of Victorian of Walras's Elements, the fourth edition is free of common sense., (p. 302). Mirowski is hardly the taint because it restricts itself to virtual trades Poincare of economics; and .economists would (i.e., the auctioneer coordinates everything do well to stick to Marshallian common sense June 199 1 r=r=--\ (florid or plain). spared. For example, of the Cambridge - IV (England) post-Keynesians, he writes (p. 342): -n,..,.t. fPetJiew D"''"\:'\! Mirowski's tone is uniformly If they had been acquainted with a little patronizmg. The reader (presumed to be an of the history of physics, they would economist) is patroniz.ed: the phy$.ics in chapter 2 have seen that their mandate was to may be too hard or the reader may not be patient explore all of the ways in which a enough or perhaps not civilized enough to have substance theory of value was interests beyond economics. The neoclassical inconsistent with a field theory of value. economists discussed in the book are patronized: In general, if economists are not presented as too they are all"coy," "disingenuous," ..incompetent," stupid to understand the physics, they are and "uncomprehending" with respect to physics presented as sleepwalkers in the thrall of the and its metaphorical imperatives. Even the energy metaphor, not quite understanding how physicists are patronized: it is truly amazing that it disfigures their theories. the benighted founders of energy physics There is a glaring omission in Mirowski's advanced their subject at all. And there can be discussion of the relationship between no doubt that Mirowski understands economics and physics: . conservation laws better than modern physicists Von Neumann has no entry in Mirowski's (cf. p. 90). But Mirowski is inconsistent about index. Indeed, he is mentioned only in the physics: on the one hand, the scientists do not process of asserting that his growth model is always get it right; on the other hand, we are genuinely neoclassical in comparison with repeatedly told that if we, or economists of the Sraffa's formally similar model. This · is a past, had only known physics, we or they would stunning omission; for von Neumann was an have seen through neoclassical economics. economist and a physicist - and worthy of a If physicists are sometimes benighted, Nobel prize in both fields. If competence in neoclassical economists are damned; and physics was the central stumbling block, it is Lucifer himself is called Paul Samuelson. Or, queer that von Neumann did not immediately perhaps, he is only a malevolent wizard: .... : .. the grasp the difficulties of the energy metaphor. conjuration of scientific legitimacy by means of v vague innuendo abounds in Samuelson's Mirowski's intemperate tone. is reinforced oeuvre" (p. 384). There is more than a little of by his epideictk (I learned this form him) style. the pot calling the kettle black here. Samuelson Length and repetition replace argument and says that his forays into thermodynamics, which evidence. Two-bit words abound. One needs a he does not think bear a close relation to his good dictionary to read Mirowski. My Concise economics, are part of a search for a civilizing Oxford English Dictionary was insufficient. intellectual breadth; and they are "fun." Many times I bad to repair to the big OED, and Mirowski comments: "I think almost everyone once I came up short even there. This may have would agree this is ingenuous in the extreme: improved my vocabulary, stiU "quotidian," . People generally are not given Nobel prizes for .. tyro," 0 tergiversations," "ukase" and •fun' " (p. 385). (Perhaps Mirowski's own work .. farrago" are not words that improve with is a sore burden. Still, if he would consider the frequent use. It was also annoying to find a fair biographies of Nobel prize winners in physics number of Cooperisms (so named in honor of (Feynman, for instance) he might find that fun Mark Twain's rule in "Fenimore Cooper's plays a bigger part than this suggests.) He then Literary Offenses," .. 1.2. Say what he is goes on to accuse Samuelson of promoting proposing to say, not merely come near it. 13. as a smokescreen to Use the right word, not its second cousin"): cloak the true nature of neoclassical economics '1risson of · excitement," ..self-reflexive " and from those who might expose the incoherence of "from whence" are redundant; "Hobson's the physics metaphor with philosophical or choice" is not a hard choice, but no choice at alt; other evaluative discussion. a power series expansion is not a "Taylor Earlier Mirowski suggests that Pareto, expansion"; meteors fall, they do not rise; while Walras and Fisher wished to browbeat and there may be some difficulty in homogenizing hoodwink other economists with their them, there should be no problem in mathematics (pp. 249-50). Browbeating with "pasteurizing chalk and cheese"; many, not few, mathematics and physics is something Mirowski heard the "siren song," it is just that, since knows something about. No school of thought is hearing it resulting in their being smashed on the

14..1 June 1991

/=+= , rocks, they did not live to tell of it - we ·must ·presume that Joan IDAA(, "'t'View D

Note I am grateful to my colleague, Julie Nelson, for useful comments on the first draft of this essay.

References Fisher, Irving. (1985) The Purchasing Power of Morsey: Its Determination tUld Relation to Credit Interest and Crues. Fairfield, NJ: Augustus Kelley. Jevons, W. Stanley. (1965) The Theory of Political Economy, fifth edition. New York: Augustus Kelley.