Customer Reviews

More Heat than Light: Economics as Social Physics, Physics as Nature's Economics (Historical Perspectives on Modern Economics)

5 star:		(4)
4 star:		(2)
3 star:		(0)
2 star:		(1)
1 star:		(1)

 

 

This is an amazing book! It exposes what is hidden in Samuelson and every other economics text, namely, that the theory of equilibria and utility were merely lifted from physics (Hamiltonian dynamics) without any support from economic data. What is more amazing is the complete discussion presented by the author that `utility' doesn't exist mathematically because the required differential form is nonintegrable. Economists and statisticians may not be able to undestand this point, but it should be familiar to researchers in dynamical systems theory. The inventors of marginal utility theory, Fisher, Walras, and Pareto, literally did not know enough mathematics to know what they were talking about. Equilibrium can't be reached in a Hamilton system (there is no friction to permit the approach to equilibrium), but this contradiction was no worse than all the others inherent in econo-pseudophysics.

This book should be read and understood by every economics student who reads Samuelson and asks: what has 'utility' to do with the data. A very good complementary book is Osborne's "The stock market from a physicist's viewpoint". Osborne introduced tha idea of lognormal stock prices (thereby paving the way for the Black-Scholes equation and derivatives trading). In his first two chapters, without the benefit of the historic perspective offered by Mirowski, Osborne explains why the supply-demand curves drawn by Samuelson are wrong and misleading, and then goes on to illustrate how one could obtain correct discrete plots real data. For example: if 25 tomatoes are available (supply) then what's the price? Clearly, there is no answer. Price does not exist as a function of supply, nor of demand (nonuniqueness). Osborne goes on to suggest that the three related assumptions of equilibrium, continous price changes and efficiency are not supported by the data, and observes that there are traders who make money out of the inefficiency of the market. Mirowski and Osborne are strongly recommended to anyone who wants to understand economics.

This is a work of prodigious scholarship and imaginative synthesis. Mirowski sifted through a tremendous amount of historical material, and approached it with great creativity. He makes an excellent prima facie case that even late 20th Century neoclassical economics is based on an early form of 19th Century thermodynamics -- the way it was before formulation of the Second Law (the one about entropy).

I came at this with more background in physics than in economics (which isn't saying much). I found the history of the principle of the conservation of energy (Ch. 2) fascinating in its own right. For its concise treatment of that topic, this book deserves to be better-known in the "physics-physics" (as distinguished from econophysics) community. As for Prof. McCauley's comment in an earlier Amazon review that Mirowski is confusing potential energy with the action as the appropriate analogue to utility, my impression was that this error isn't unique to Mirowski, but was made by at least some of the economists whose work he is critiquing (e.g. Irving Fisher).

I give this four stars, though, because of some genuine weak points.

First, Mirowski spills much ink faulting economists because they use a physics metaphor that's outdated, but relatively little on the question of empirical justification (or lack thereof) for using any physics metaphor at all. More discussion of this point would have been helpful.

Second, Mirowski's discussion of physics is at times very tentative, like a student who copies stuff into a term paper without understanding it fully, but hoping that he can sort of fake his way through. E.g., he refers to the definition of a curl of a function as a condition (when the condition he means is that the curl = 0), and twice to an exact differential as an "exact differential equation" when no equation is stated; he throws around frequent references to the Lagrangian without ever mentioning its familiar form as the difference between kinetic and potential energy (T-V); and although he includes some equations in his discussion of general relativity, he neither explains his notation nor seems to be sure of what the equations represent.

Finally, his writing style is often pompous and overly ornate. In the early part of the book he seems to have been possessed by the spirits of the 18th and 19th Century writers he's discussing. (I was amazed to learn that he's a Baby Boomer who was still in his 30s when he wrote this book.) He adopts a more entertaining and sarcastic tone when he gets to the neoclassical economists, especially in his take on P. Samuelson near the end of the book. But too often he sounds like a too-clever college student. It makes for an unfortunate contrast to the depth and originality of his argument.

This fascinating upgrage of the author's earlier _Against Mechanism_ gives a severe account of the state of mathematical economics as it has been since the marginalist revolution. It is a reminder that mathematical technique and basic modeling are two separate activities and that understanding what it is that one is attempting to make into a science is not so easy. It is probably true no deterministic mathematical science of any type known is possible in a medium involving human consciousness. Yet the obsession to treat these different domains of discourse as analogous to physics or amenable to predictive science via the apparatus of differential equations simply refuses to die. It is a peculiar history, that some very good mathematicians of the nineteenth century, who understood the physics, found bizarre at the start, before the bad habits of phantom thinking became institutionalized. Mirowski's expose of the whole game is priceless, and almost unnerving. Hm, ideology perhaps Very important book.

Since having reviewed this book in September 1999, I was inspired by it to resolve Mirowski's Thesis in a recent paper called 'The Futility of Utility' (Physica A,2000). My resolution shows that both Varian and Mirowski were partly wrong. Mirowski is right in one sense: when dynamics is taken into account in the theory of production then the generic case (nonintegrable dynamics) is that utility is a path-dependent functional, and so doesn't exist mathematically as a function of demand. In this case (as Osborne observed from empirical data) price as a function of demand does not exist (the 'cartoons' passing as graphs in Samuelson's textbook can't be constructed from real market data). Varian was wrong: the most trivial integrability conditions are violated in this case because utility as a function cannot be postulated but either exists or doesn't according to dynamics. On the other hand, Mirowski was wrong that an analog of kinetic energy, or even a conserved quantity, is required. Utility is not, as Mirowski believed, analogous to potential energy: it is analogous to what physicists call the action. When optimization-control dynamics (Hamiltonian dynamics in econometrics) is integrable, then the action is a function, not merely a path-dependent functional (see Liouville's integrability theorem, ca. 1880, also discussed in 1916 by Einstein in the context of why Bohr-Sommerfeld quantization fails).

Mirowski's incredible work deserves better than a 1000-word review by a layman. And yet the reviews of other economists have often been inept and misleading. (When a scholar allows a major mistake to see print, that's regarded as a pretty serious intellectual shortcoming. But there's a worse one. Accusing a scholar of making a major mistake he didn't actually commit. That's essentially what the well-known economist Hal Varian did in his review.)

As a history of the energy concept, MORE makes for facinating if demanding reading. An interest in economics is not at all required for that section of the book. Anyone with a little curiosity about physics and its history will be intrigued.

When it comes to criticizing neoclassical economics, Mirowski is in a class by himself. Witty but substantive zingers abound. Zingers that can't be dismissed as mere rhetorical potshots.

The grace and depth of MORE HEAT THAN LIGHT readily justify the effort required to finish it. Still, for laymen unwilling to tackle Mirowski, there is a pale substitute. Paul Ormerod's [spelling?] THE DEATH OF ECONOMICS. That's a breezy critique of neoclassical doctrine aimed at the general reader.

Whether you plan to ever look at it or not, buy Mirowski's book NOW. Forget rave reviews -- what he really deserves (to strike the proper, academic note)is your money.

This book is extremely imporatnt for economists - yet it is definitely mainly overlooked in mainstream economics and/or misinterpreted in some references what I saw. Actually no wonder why - it evidently shows that modern economics was based on incorrectly understood physical principles from its inception in 19 century, and in fact remains almost in the same mode now. The recent rise of econophysics just showed that from another angle. The aggregation of Mirowski's excellent histgorical findings and modern econophysical concepts is still waiting its hour. The only drawback I found in this book is some redundany, when the same thoughts are repeated several times, quite unexpectedly. It seems it could be reduced in two times without any loss of its value.

Mirowski's(M)book correctly points out that Samuelson's attempt to model economics "as if" it was the physics of Boltzmann and Gibbs fails to incorporate the 20th century physics of Einstein(the special theory of relativity(1904) and the general theory of relativity(1915)).However,if Samuelson had discovered the special nature of neoclassical economics,then he,and not John Maynard Keynes,would have been the greatest economist of the 20th century.Mirowski generally is correct that the economics profession has been too engrossed in the advanced Newtonian physics of the 1870-1900 time period.Of course,it is this type of physics that Samuelson was taught in his engineering physics courses when he was a student in the early 1930's(the same conclusion holds for this reviewer in the early 1970's).Unfortunately,Mirowski,instead of correctly pointing out that,despite Samuelson's great technical skills and ingenuity,such an approach could only yield special conclusions(Samuelson did point out that neoclassical theory is strictly limited to an analysis of points lying on the boundary of the static and dynamic production possibilities curves in his principles textbook),appears to come very close to claiming that Samuelson is a scientific fraud .Mirowski's claims are simply false. The second major problem with this book is in its assessment of Keynes's General Theory(GT;1936).Every statement about Keynes and the General Theory in this book is either an error of omission or an error of commission.Mirowski's knowledge of Keynes's mathematical modeling approach in chapters 20 and 21 of the GT is nonexistent.Mirowski's reliance on the error filled commentaries of the mathematically and economically illiterate,inept,and innumerant accountant,Hugh Townshend,whom Mirowski describes as"...a spectacularly perceptive critic..."(Mirowski,p.411)means that he has absolutely no idea of what Keynes is doing.On pp.261-262 of the GT,Keynes gives,not once but twice,the sufficient macroscopic optimality condition required for there to be no involuntary unemployment.It is that the marginal propensity to spend must equal 1.Unless this condition is met,no amount of wage and price flexibility ,even if instantaneous and simultaneous in all markets,will have any effect.Formally ,the mpc must equal 1.If the capital stock is not at an optimal level,then the mpc+mpi=1=mpc+mps condition is required,where mpc is the marginal propensity to spend on consumption goods,mpi is the marginal propensity to spend on investment goods,and mps is the marginal propensity to save.In chapters 20 and 21,Keynes derives this condition from the ground up,using a microfoundations of firms/industries operating under conditions of pure competition.This condition,which any competent mathematician can derive,is that w/p=mpl/(mpc+mpi),where mpl is the aggregated marginal product of labor.If mpc+mpi<1,involuntary unemployment will automatically exist.Keynes demonstrated that the capitalist system is ,in fact, a system of multiple,stable equilibria.Due to the ignorance of Mirowski, concerning Keynes's Einsteinian revolution in economics,his explicit unsupported attack on P. Samuelson,and his implicit attack ,again unsupported,on John Maynard Keynes,I can't recommend the purchase of this book unless it undergoes a complete revision,concentrating on the deficiencies of modeling economics on late 19th century Newtonian physics only.Mirowski could have written a 5-star book.Instead,he mixs his correct assessment of the misguided attempts of the economics profession to model economics as if it were physics with a series of unsupported attacks on Samuelson and Keynes.

I bought this book with high hopes. I found, to my disappointment, that it is baloney. Poorly written -reminscent of travel literature in which every noun is preceded by at least one adjective - and incompetent. If anything the author claims is true, he has not demonstrated it. It is just a meaningless polemic.