Nicholas Georgescu-Roegen was a Romanian mathematical economist whose application of thermodynamic concepts to economics led to the new academic sub-discipline of bioeconomics (ecological economics).

His extraordinary mathematics abilities led his teachers in Romania to encourage him to study abroad. In 1927 he went to the Institute of Statistics in Paris where his professor was Émile Borel. Borel had his Ph.D thesis published in special issue of an academic journal.

Georgescu-Roegen decided in 1931 to go to England to study with the great biostatistician Karl Pearson. His work with Pearson came to the attention of the Rockefeller Foundation, who offered him a research fellowship.

He was a visiting fellow at Harvard University in 1934, where he was introduced to the some of the best economic theorists by Joseph Schumpeter, who had himself recently emigrated, from Germany. Schumpeter organized weekly meetings of the thinkers, which was Georgescu-Roegen's education in economics. He never took an economics class.

He wrote several important papers in economics in the 1930's. After World War II he returned to Harvard as Research Associate and Lecturer. In the 1960's his fame began to spread. Harvard University Press published twelve of his most important papers in the 1966 volume, Analytical Economics: Issues and Problems. He wrote five introductory essays which first introduced his interest in the "Entropy Law."

Georgescu-Roegen's great work in 1971, The Entropy Law and the Economic Process, had an enormous impact on this "information philosopher." At the time, my work was on identifying negative entropy as the possible source for an "objective" value, based on remarks made by Arthur Stanley Eddington in his 1927 book, The Nature of the Physical World.

Georgescu's remarks depend heavily on a comparison of living thing with the hypothetical Maxwell's Demon that can manipulate the motion of molecules to (apparently) violate the second law of thermodynamocs.

7. Entropy and Purposive Activity. Among the various ideas surrounding the antinomy between physical causality and freedom is that of the inexorability of the physical laws. Properly understood, this idea is that man cannot defeat the physical laws in the sense of preventing their working. The law of gravitation, for instance, is at work even in the case of a flying aircraft. The Entropy Law of Classical thermodynamics is no exception to this rule. Heat is dissipated even when we refrigerate a warehouse, because more heat is “degraded” in the rest of the universe than that which is “upgraded” in the warehouse. The result is that bound heat-energy in the universe has increased, as the law requires. Refrigeration is an exception only to the crude law that heat cannot flow from the colder to the hotter body but not to the law proper which says that heat cannot do so by itself. However, the probabilistic formulation of the Entropy Law, based on the idea that heat is merely one manifestation of the irregular motion of particles, raised in some physicists’ minds doubts as to the inexorability of that law.

This view is related to a piquant fable of J. Clerk Maxwell’s. He imagined a minuscule demon posted near a microscopic swinging door in a wall separating two gases, A and B, of equal temperature. The demon is instructed to open and close the door “so as to allow only the swifter molecules to pass from A to B, and only the slower ones to pass from B to A.” Clearly, the demon can in this way make the gas in B hotter than in A. This means that it can unbind bound energy and, hence, defeat the Entropy Law of statistical thermodynamics.

Ever since Maxwell wrote it (1871), the fable has been the object of a controversy which, I submit, is empty. Taken on its face value, the fable reveals a conflict between the tenet that physical laws are inexorable and the statistical explanation of thermodynamic phenomena. In this perspective, Maxwell’s own point corresponds to eliminating the conflict by upholding the tenet and indicting the explanation. But one may equally well accept the statistical explanation and reject the tenet. This second alternative corresponds to the argument enthusiastically supported by all vjtalists that a living being—as proved by Maxwell’s demon— possesses the power of defeating the laws of matter. It is because of this last argument that the fable acquired a sweeping significance. However, like many other paradoxes, Maxwell’s is still an intellectual riddle. Like all paradoxes, Maxwell’s can only enlighten our thoughts but cannot become a basis for settling the very issue it raises.

The main line of the arguments aimed at disposing of the paradox descends from Boltzmann, who argued that “ if all differences of temperature would disappear, no intelligent being could emerge either.” The point has ever since been repeated in various forms by Einstein, Eddington, and many others. The issue was given a more explicit turn by L. Szilard. He argued that the demon cannot act without getting some information about the motions of the particles. This idea paved the way for equating entropy with deficiency of information and led to a series of exercises on the physical limitations of the demon. Their main point is that since a milieu in thermodynamic equilibrium is a black body it is impossible for the demon to see the particles. Should it be provided with some physical device for obtaining the needed information—say, a torch— it still could not unshuffle more energy than that consumed by the device. All these exercises, however, do not dispose of the paradox: they merely assume it away.62 Their very basis is that the Entropy Law prevents a physical device from performing more work than that warranted by the free energy it receives. Clearly, if this is the premise, the conclusion can only be the absurdity of the fable.

A more familiar argument, instead of providing the physical device, “exorcises” it, i.e., transforms it into an intelligent being in flesh and blood. It first observes that such a being must consume some free energy in order to survive, and then it asserts that if the being were able to unshuffle a greater amount it would contradict the Entropy Law This line of reasoning, therefore, is vitiated by the same circularity as that of the preceding paragraph. It has nevertheless the advantage of bringing to the forefront the most important implication of the fable. In the words of Helmholtz, it is the issue of whether the transformation of the disordered heat motion into free energy “ is also impossible for the delicate structures of the organic living tissues.” More exactly, if all is aimless motion (as statistical thermodynamics contends), we should expect the constituent particles of any organism to disintegrate promptly into a chaotic state just as the aimlessly running mice of G. N. Lewis’ metaphor supposedly do. Indeed, the probability that a living organism would not disintegrate promptly is fantastically small. According to the teachings of Boltzmann and of every advocate of the probabilistic approach, the event should never happen. Yet the “miracle” has happened over and over again on a fantastic scale. The miracle, therefore, needs an explanation. As Poincare put it, it is precisely because according to the laws of physics all things tend toward death “that life is an exception which it is necessary to explain.”

It is along this line of thought that Eddington argued that besides random there must be an opposite factor at work in nature: the antichance. “We have,” he said, “ swept away the anti-chance from the field of our current physical problems, but we have not got rid of it.” By this he may have meant that we have done away with strict causality and now we need the anti-chance to oppose mere chance in all those countless cases where the rule of chance is contradicted by enduring ordered structures. To be sure, similar suggestions for explaining the contradiction had been made long before by others—by Georg Hirth, for instance. By now, practically every thinker feels that “ something [a new principle] has to be added to the laws of physics and chemistry before the biological phenomena can be completely understood.”69 Suggestions such as Hirth’s and Eddington’s have only an indirect value which, moreover, calls for a great deal of conceding. But the alternative position— to maintain that there are no principles in nature other than those manifested in the test tube or on the photographic plate—amounts to a glorification of the fallacy of the puzzled zoo visitor mentioned a while ago. Certainly, as Wiener noted, “ it is simpler to repel the question posed by the Maxwell demon than to answer it.”

Yet Maxwell’s demon was not to remain without glory. The fable had a decisive influence upon the orientation of the biological sciences. To begin with, it compelled us all to recognize the categorical difference between shuffling and sorting. In thermodynamics we do not ask ourselves whence comes the energy for the shuffling of the universe, even though we know only too well that it takes some work to beat an egg or to shuffle cards. The shuffling in the universe—like that of the gas molecules surrounding the demon—goes on by itself: it is automatic. But not so with sorting: Maxwell invented a demon, not a mechanical device, for this job. “ Sorting is the prerogative of mind or instinct,” observed Eddington, and hardly anyone would disagree with him nowadays. Actually the more deeply biologists have penetrated into the biological transformations the more they have been struck by “ the amazing specificity with which elementary biological units pick out of the building materials available just the ‘right ones’ and annex them just at the right places.” Irrespective of their philosophical penchant, all recognize that such orderly processes, which are “much more complex and much more perfect than any automatic device known to technology at the present time,” occur only in life-bearing structures. This peculiar activity of living organisms is typified most transparently by Maxwell’s demon, which from its highly chaotic environment selects and directs the gas particles for some definite purpose.

Purpose is, of course, a concept alien to physics. But from what has been said in the preceding section, this point should not bother us. Physicists, in opposition to the positivist sociologists, have one after another admitted that purpose is a legitimate element of life activities, where the final cause is in its proper right, and that it leads to no logical contradiction if one accepts complementarity instead of monism. Eddington, as we have seen, goes even further. For although he argues that the “ nonrandom feature of the world might possibly be identified with purpose or design, [noncommittally with] anti-chance,” he does not suggest that anti-chance is absent from the physical world. “Being a sorting agent, [Maxwell’s demon] is the embodiment of anti-chance.” Norbert Wiener, too, sees no reason for supposing that Maxwell demons do not in fact exist hidden behind some complex structures, as it were. As the metastable properties of enzymes suggest, they may operate, not by separating fast and slow molecules, “ but by some other equivalent process.”76 It is not surprising therefore that thermodynamics and biology have drawn continuously closer and that entropy now occupies a prominent place in the explanation of biological processes.

Unfortunately, most students of life phenomena now shun the use of the concept of purpose. In all probability, this proclivity reflects the fear of being mocked as a vitalist more than anything else. As a result, only few students pay attention to the fact—a physico-chemical marvel in itself—that life-bearing structures are as a rule able to attain their individual purpose over unforeseen obstacles of all sorts or, as Bergson strikingly put it, to secure “ the constancy of the effect even when there is some wavering in the causes,” I should hasten to add that by emphasizing the legitimate place of purpose in life phenomena I do not intend to vindicate the ultravitalist position that living structures can defeat the laws of elementary matter. These laws are inexorable. However, this very argument uncovers the real issue of the vitalist controversy. Given that even a simple cell is a highly ordered structure, how is it possible for such a structure to avoid being thrown into disorder instantly by the inexorable Entropy Law? The answer of modern science has a definite economic flavor: a living organism is a steady-going concern which maintains its highly ordered structure by sucking low entropy from the environment so as to compensate for the entropic degradation to which it is continuously subject. Surprising though it may appear to common sense, life does not feed on mere matter and mere energy but—as Schrödinger aptly explained—on low entropy.

Sorting, however, is not a natural process. That is, no law of elementary matter states that there is any sorting going on by itself in nature; on the contrary, we know that shuffling is the universal law of elementary matter. On the other hand, no law prohibits sorting at a higher level of organization. Hence, the apparent contradiction between physical laws and the distinctive faculty of life-bearing structures.

Whether we study the internal biochemistry of a living organism or its outward behavior, we see that it continuously sorts. It is by this peculiar activity that living matter maintains its own level of entropy, although the individual organism ultimately succumbs to the Entropy Law. There is then nothing wrong in saying that life is characterized by the struggle against the entropic degradation of mere matter.81 But it would be a gross mistake to interpret this statement in the sense that life can prevent the degradation of the entire system, including the environment. The entropy of the whole system must increase, life or no life. Although the point in all its details is quite involved, its gist is relatively simple if we bear in mind a few things. The first is that the Entropy Law applies only to an isolated system as a whole. The second is that an isolated system in entropic equilibrium (in a chaotic state) is homogeneous in itself and also has no free energy relative to itself. ...

Whether a Maxwell demon, if introduced in such a world, could perform its task is still a moot question. But there is hardly any doubt that in a world whose entropy is still increasing a sorting demon can decrease the entropy of a subsystem. The fact that an exorcised demon, i.e., a living organism, can survive only in a world whose entropy increases has already been pointed out by more than one writer. I should add, however, that life, at least in the form it exists on this planet, is compatible only with a moderate entropy. In an environment of very low entropy, a living organism would not be able to resist the onslaught of the free energy hitting it from all parts. On the other hand, in an environment of very high entropy there would not be enough free energy going around for the sorting to be successful in the short run.

Let me observe that the case, however, is not completely closed by the above remarks. A perhaps even more difficult question confronts us now: is the increase of entropy greater if life is present than if it is not ? For if the presence of life matters, then life does have some effect upon physical laws. Our ordinary knowledge of the change in the material environment brought about by the biosphere seems to bear out the idea that life speeds up the entropic degradation of the whole system. And in fact, a simple laboratory experiment confirms that the entropic evolution of an isolated system is altered if life is introduced in it at a certain moment. All lifebearing structures work toward a purpose—to maintain their entropy intact. They achieve it by consuming the low entropy of the environment, and this fact alone should suffice by itself to justify the belief that life is capable of some physical manifestations that are not derivable from the purely physico-chemical laws of matter. There is, we remember, some freedom left to actuality by the Entropy Law of Classical thermodynamics. And as W. Ostwald, a Nobel laureate for chemistry, noted long ago,86 it is by virtue of this freedom that a living organism can realize its life purpose and, I should add, that man’s economic activity is possible. Another reason why I have dwelt in the preceding chapter on the main flaws ably couched in the reduction of thermodynamics to the law of mechanics should now be obvious: statistical thermodynamics completely denies the possibility of any purposive activity because it claims that everything is completely determined by the laws of mechanics. Accordingly, it would be nonsense to speak of purposive activity and to relate it to some “ vitalistic” principle not deducible from those laws. But without such a principle, I contend, we simply turn our backs to a wealth of highly important facts. Actually, if examined closely, many occasional remarks by physicists on the life process tend to show that they too share, however unawares, this “ vitalistic” belief.

The fact has a natural explanation. The scholarly mind cannot bear the vacuum left after the Classical concept of cause shared the fate of the mechanistic epistemology. The scholarly mind needs something to stimulate its imagination continuously or, as Planck said, to point in the direction of the most fruitful search. The domain of life-phenomena represents a very special case in this respect. For, as we have seen in this section, life is manifested by an entropic process that, without violating any natural law, cannot be completely derived from these laws—including those of thermodynamics! Between the physico-chemical domain and that of life there is, therefore, a deeper cleavage than even that between mechanics and thermodynamics. No form of cause that may fit other phenomena could do for the sciences of life. The final cause—that is, purpose—is not only in its right place in these sciences but it also constitutes an indispensable and extremely useful tool of analysis. A biologist or a social scientist has to be a “vitalist” and, as a result, to be in the habit of looking for a purpose. It is all right for an economist to rest satisfied with the explanation of a catastrophic crop by some efficient causes triggered by random events. However, the science served by him is ordinarily interested in problems involving human actions. And if an economist wishes to analyze the actions of those who tilled the soil and cast the seeds, or of all those who have been hit by the scarcity produced by the crop failure, he will not arrive at a penetrating understanding if he refuses to look for the purposes that move them. For the truth that cannot be oblitered by the current behavioristic landslide is that we all—the fans of behaviorism included—act from a purpose.

And one complete circle is now closed by recalling that all our important purposes—namely, to stay alive and to keep a place under the social sun— lead to entropic transformations of our neighboring universe. This means that the realization of our purposes sets us on a never-to-return journey.

(The Entropy Law and the Economic Process, pp. 187=-195)

Georgescu-Roegen's rambling thoughts show he does not understand the deep problems in statistical physics, but his work influenced a large community of scholars to apply principles of statistical physics to economics and ecology today, so they bear reading again.

Georgescu-Roegen died disappointed that his work had not been better understood and used to improve the well-being of mankind. This was partly because entropy is such a difficult subject for most readers and it remains a controversial subject among the scientists who founded thermodynamics, statistical mechanics, and the kinetic theory of gas particles.