Navigation – Plan du site
Time in Economics, part I

The Original Time Approach of Georgescu-Roegen

L’approche originale du temps par Georgescu-Roegen
Roxana Bobulescu
p. 87-109


Dans cet article, nous analysons l’approche originale du temps de l’économiste Georgescu-Roegen. Notre principal argument est que pour Georgescu-Roegen le temps est un concept polychromatique. Tout en reconnaissant la distinction entre la nature logique et la nature historique du temps en économie, Georgescu-Roegen a défendu le temps historique à partir d’un système ouvert et dans une perspective dialectique. Il a montré que parmi toutes les approches du temps, celle du temps historique est la seule compatible avec les lois thermodynamiques, en particulier la loi de l’entropie. Cette approche du temps laisse également un espace pour tous les événements historiques et les activités humaines dans leur milieu de vie qui est la biosphère. Georgescu-Roegen a développé une approche originale du temps qui concilie sciences naturelles et sociales dans une perspective évolutionniste. Le temps est le courant de la conscience, un flux unidirectionnel qui entraîne des changements irrévocables.

Haut de page

Texte intégral

1Romanian-born American economist Nicholas Georgescu-Roegen (1906–1994) had a strong interest in epistemology. In Analytical Economics (1966), he called for economists and natural scientists to reconsider their respective fields with regard to the concept of time. While both physics and biology developed an evolutionary approach based upon the entropic and biologic transformations, economics still uses mechanical clock-time approaches. 

2Influenced by Schumpeter, Georgescu-Roegen distinguishes historical time and logical time, as did Joan Robinson and George Shackle before him. Unlike his predecessors, however, he does not totally dismiss logical time, though he favours the historical-time approach in an original way. This paper aims to describe this particular feature of Georgescu-Roegen’s defence of historical time, which sets that the past, the present and the future are qualitatively different. This time concept is concerned by social transformation.

3I will argue that taking into consideration historical time is the result of fully embracing an open-system methodology. From this viewpoint, reality is stratified and there are strong interdependences between the different strata. As a discipline, economics can be viewed as a sub-system of the biosphere, interacting not only with other sub-systems (such as the physical and the biological ones), but also with the institutional and historical context. The reality is complex, and scientists endeavour to grasp it by avoiding the reductionist fallacy. According to Georgescu-Roegen, the reductionist approach of neoclassical economics is the best example of such a methodological fallacy.

4In order to apply the open-system methodology to any object of study, time-consistency between the different fields or sub-systems involved is necessary. Scientists should be able to refer at the same time to physics, biology, economics and history; otherwise, the marriage of these disciplines is impossible. Furthermore, the historical time-frame is inherent to the evolutionary approach envisioned by Georgescu-Roegen. He distinguished irreversibility and irrevocability of a system. Irreversibility means that a system cannot follow the same path in a reverse order. Irrevocability means that a system cannot pass through the same state more than once.

5The common mainstream modelling position is to accept physical analogies while rejecting the historical approach. Modelling is based upon the deductive approach, and the positivist methodology underlying it rejects inductive temptations. From this perspective, the marriage between physics (Newtonian mechanics) and economics is sufficient. And both disciplines use logical time t or mechanic clock-time. This time concept is reversible; it reproduces the pendulum movements without any change intervening in the repeated process. Schumpeter called it “dynamic time”.

6Georgescu-Roegen challenges this view. He argues that integrating historical time into economics stems from integrating thermodynamic time T in economic analysis, and thus replacing logical time t. Thermodynamic time T is the entropy time. This Time T makes processes both irreversible and irrevocable.

7According to Georgescu-Roegen, “the root distinction (between t and T) does not lie in historical (evolutionary) sciences, but—as we have seen—in the heart of physics, between mechanics and thermodynamics” (Georgescu-Roegen, 1966, 71).

8The originality of Georgescu-Roegen’s approach is to reconcile two visions that were previously seen as contrasting (see Robinson, 1980, and Shackle, 1967)—the historical and the physical (logical, abstract) visions of time—by demonstrating that if we consider thermodynamic T instead of mechanical clock-time t, then we will be able to reconcile this physical time T with historical time. The mechanistic vision of the world with stable, timeless, reversible and measurable conditions is thus replaced by a vision of permanent change and qualitative transformation, by increasing entropy and exosomatic evolution. Time’s arrow is related to irreversibility and irrevocability of economic processes. It is the consequence of the entropy law: heat always moves from hotter to colder bodies. Time’s arrow is a property of entropy alone. Even in historical time there is a time’s arrow that goes in only one direction. This is the reason why historical time is consistent with thermodynamic time (or entropic time). 

9To preserve its scientific status, economics should not be bounded by mechanical time equilibrium, but instead should be open to new branches of physics, such as classical thermodynamics and statistical mechanics.

10Georgescu-Roegen allowed different concepts of time to coexist. In his view, philosophical time encompasses all time concepts. It is the stream of consciousness or a continuous succession of moments. It is unidirectional because it involves an “earlier” and a “later”.

11Despite his complex, plurichromatic time concept, Georgescu-Roegen uses only two symbols: T and t. In this paper, I retain these symbols but in between, I identify an intermediary time concept associated to t, which I call “t as duration” and which is the time of production and consumption. In order to clarify Georgescu-Roegen’s time concept, I suggest the following taxonomy:

Time concept






t as duration



T and historical time



12Georgescu-Roegen used t as duration in his models (fund–flow), aiming at demonstrating the time constraint of production. But he was unsatisfied with this partial analysis, which departs from the open-system approach and dialectic reasoning. This article will explain why Georgescu-Roegen preferred historical time to all other attempts to render economic analysis “in time”. In Section 1, we present the physical roots and lessons for time modelling, using a biological, evolutionary approach. In Section 2, we present the arguments in favour of historical time, from an open-system and dialectic perspective. Section 3 concludes.

1. The Physical Roots of Historical Time T

13The present study focuses on four main sources in Georgescu-Roegen’s bibliography. First, the major publications—such as Analytical Economics (1966) and “The Entropy Law and the Economic Process” (1971)—will provide the essence of the epistemology of and physical approach to time. These are completed by two more recent sources that specifically address the central idea behind this article: “Time and Change in economics” (1988) and “Time in Economics” (1994). These two groups of works take a critical approach to time; although the last two works dedicated to time are not well-known, they evidence Georgescu-Roegen’s strong interest in the adequate treatment of time in economics. Recurrent critics of the neoclassical “timeless” analysis reveal Georgescu-Roegen’s dissatisfaction with the treatment of time in economics, and his search for a more appropriate time concept.

14For Georgescu-Roegen, “time” is a polychromatic concept, used in various ways and subject to different interpretations. He held that if historical time is contrasted very simply with logical time, this is because logical time is inspired by the mechanical approach in physics. But this is not the only time approach in physics. Let us consider thermodynamics, as Georgescu-Roegen invites us to do: thermodynamic time T cannot be contrasted with historical time. Thermodynamic time is entropy time, characterised by change, irreversibility and irrevocability.

15Is this a reason for dismissing t (mechanical clock-time or logical time) from economic analysis? Not for Georgescu-Roegen, who uses this concept of time in a fund–flow model that he considers to be an “analytical simile” of production processes.

16The starting point of his critique can be found in his book Analytical Economics, in Section 10 (page 68), entitled “Time: An Ambiguous Term”. This is an eclectic approach to time, with an emphasis on natural sciences such as physics, biology and chemistry. The section starts with a philosophical definition of Time (differentiated by its upper-case “T”) as a “primitive stream of consciousness” (Weyl, 1950, 5). “In fact, temporal laws in any science require a distinction between earlier and later, which only consciousness can make” (Georgescu-Roegen, 1966, 69). For example, the entropy law gets its full meaning “as Time flows through the observer’s consciousness” (Georgescu-Roegen, 1966, 69).

17For Georgescu-Roegen, this Time is the universal time of consciousness, in which everything—facts, beings and knowledge—happens. In my view, this concept is the key to his open-system approach, because Time embraces all the different time concepts, as an umbrella term. Its definition underlies all other time concepts, be they from physics, history, economics or biology. Thus, Georgescu-Roegen wants to emphasise this by warning the reader not to invert the causal relation between Time and all other time concepts. It is not the physical law (the change in entropy) that defines Time; rather, Time derives from the stream of consciousness, regardless of the movements of the clock. The entropy changes occur according to the observation of an “earlier” and a “later” time. If we reverse the observation flow of consciousness, we invert the entropy law.

1.1 Lessons From Physics Regarding T and t

18The duality of time originates in physics. The thermodynamic law of physics stating that “heat always moves by itself from hotter to colder bodies” provoked a crisis in physics, because “mechanics cannot account for the unidirectional movement of heat, for according to mechanics all movements must be reversible” (Georgescu-Roegen, 1966, 70). The equations of mechanics are invariant with respect to t.

According to Georgescu-Roegen, the term “duality” is inappropriate to characterise the crisis in physics between mechanical, reversible time and thermodynamic, irreversible T. The two concepts cannot be confused, even if we use the same term “time” for both:

In fact, T represents Time, conceived as the stream of consciousness or, if you wish, as a continuous succession of moments, but t represents the measure of an interval (T’, T’’) by a mechanical clock. … T is an ordinal variable, but t is a cardinal one. The fact that a weak cardinal scale can be constructed for T on the bases of t = Meas (T’, T’’), does not mean that it is not necessary to distinguish between t and T, even though we must reject the duality of Time. (Georgescu-Roegen, 1966, 70-71)

19Mechanical phenomena (such as the swing of a pendulum, or the fall of a stone) are functions of t alone but are Timeless, because they occur the same way whenever they occur in Time. Consequently, there is no need to place the event in Time:

Because only in thermodynamics, of all branches of physics, laws are functions of T, there was no strong compulsion for physics to eliminate the ambiguous use of “time”. But it is hard to understand why other sciences, where the situation is not the same as in physics, have on the whole ignored the problem. All the greater is Schumpeter’s merit for stressing, in his later writings, the difference between historical and dynamic time, by which he understood T and t respectively. However, the root of the distinction does not lie in historical (evolutionary) sciences but—as we have seen—in the heart of physics, between mechanics and thermodynamics. (Georgescu-Roegen, 1966, 71)

20Time T in relation to thermodynamics is irreversible. According to the laws of thermodynamics, energy has two states: free (available) and bound (latent). The “heat death” of the universe happens when all energy is latent. The entropy law is known as the second law of thermodynamics, first formulated in 1824 by Sadi Carnot in his memoir on the efficiency of steam engines. Rudolf Clausius reformulated it in 1865, with the first law of thermodynamics stating that the energy of the universe remains constant; and the second law of thermodynamics that the entropy of the universe at all times moves towards a maximum. The entropy formula (Maxwell, 1921, 189) is as follows:

Entropy = Latent energy / Absolute temperature

21“It is strictly an evolutionary law with a clearly defined time’s arrow: entropy. Clausius seems to have thought of it in the same way, for he coined ‘entropy’ from a Greek word equivalent in meaning to ‘evolution’” (Georgescu-Roegen, 1966, 71). This corresponds to the classical formulation of the entropy law.

22In a more recent formulation, that of statistical mechanics, entropy has an additional meaning. It is a disorder index. Boltzmann’s formula of entropy is as follows:


with K = a physical constant and n = number of equivalent microstates

23The discipline of statistical mechanics has brought about a reorientation from rigid to stochastic laws that some consider as proof of the “free will” in nature. “Thermodynamics and biology have drawn continuously closer and … entropy now occupies a prominent place in the explanation of biological processes” (Georgescu-Roegen, 1966, 81).

24As entropy law has a Time’s arrow, it is an evolutionary law. At this point, biology meets thermodynamics due to time consistency with the law of evolution. Inspired by Lotka (1925) and Schrödinger (1944), Georgescu-Roegen explains how life is a permanent struggle against entropic degradation. Organisms compete for low-entropy matter and energy. However, life cannot be reduced to physical laws. Even if the entropy of the whole system must increase, life can affect this increase. Evolutionary laws in biology and thermodynamics mutually interact.

1.2 Time Scarcity, Change and Irreversibility

25In the light of the previous considerations derived from physics and biology, both entropy law and biological evolution—the root of life itself—point to the T concept. Georgescu-Roegen makes a solid statement of the prevalence of the Time concept in natural sciences. The only true temporal law is therefore the thermodynamic entropy law (i.e. the second law of thermodynamics). In this context, why should economics stick to its mechanics-based analogy? Economic processes are not mechanical phenomena. In this respect, Georgescu-Roegen calls for the abandonment of Juglar’s formula for business cycles. The idea is that the business-cycle formula implies an invariant parallelism between business activity and a mechanical clock, or a parallel relationship between some economic phenomena, by considering one of them as an economic clock for the others. Such parallelism between various aspects of economic activity is highly improbable:

However, the alternative idea that the march of the entire economic process can be described by a system of differential equations with clock-time as the independent variable—an idea underlying many macro-dynamic models—is in all probability vitiated ab ovo. (Georgescu-Roegen, 1966, 74)

26This is the case with growth macroeconomic models—such as Solow’s model (1957), which contains a disturbing anomaly of identifying t with the index of the changing technological condition over a given period of time in a given country. What is the meaning of a partial derivative of the aggregate production function with respect to time t, “none other than a play-thing with pencil and paper” (Georgescu-Roegen, 1994, 243)?

27Despite these “pencil-and-paper” modelling efforts, neoclassical economic theory cannot fully escape two related challenges—time scarcity and change. Concerning time scarcity, Georgescu-Roegen refers to Gossen and his theory of lifetime scarcity. Production and consumption processes take time. In the abstraction of the static processes, there is no waiting, no duration. There are no time boundaries between processes. Georgescu-Roegen proposes an alternative analytical representation of the production function in which he introduces duration and distinguishes between manufacturing and farming. Georgescu-Roegen defined three possible temporal arrangements of the elementary processes of production: (a) in series, elementary processes are sequential, without overlapping in time; (b) in parallel, they are carried out simultaneously, “started at the same time and repeated once completed” (Vittucci Marzetti, 2013, 212); and (c) in line, when they are carried out one after another so that they only partially overlap. The assembly line and the factory system strongly reduce the idleness of the funds because the funds can reach full specialization. In parallel processes, the periods of idleness of the funds cannot be eliminated (see Vittucci Marzetti, 2013). Farming has parallel processes and parallel time arrows, while the manufacturing process is a continuum. Thus, idleness of funds cannot be eliminated from agricultural mechanised processes.

28In showing the effect of duration on different fund and flow processes, Georgescu-Roegen uses t (Georgescu-Roegen, 1988). I call this the “t as duration”. He builds an “analytical simile” to show that production processes happen “in time” and are better explained by graphs than by the timeless traditional production function. Duration is the key variable in the fund–flow model:

time represented as duration, although fundamental in a process of production as in any human activity, has received hardly any attention simply because this process presented virtually no interest for an economic profession preoccupied with market phenomena. Yet that issue retained the attention of Adam Smith (in his classic example of the pin factory) and in even greater extent that of Karl Marx. Its revival in a form full of economic objects lessons inspired some fruitful extensions on the timing of economic activities, and on the economics of capital utilization. (Georgescu-Roegen, 1988, 43-44)

29Georgescu-Roegen also refers to Marx’s diagram of simple reproduction as suffering from the same timelessness default.

30Georgescu-Roegen argues that even statics needs some dynamics to explain change. Thus, the neoclassical static approach of production is not bound to timelessness. In the abstract continuum, the structure is always identical to itself, as production is instantaneous. Production and consumption overlap:

I do not wish to deny the analytical value of the static process when properly applied (which is not always the case). A proper use of that concept cannot ignore time. There is dynamics even in what we call a static condition for only by dynamics can we explain how a condition reproduces itself. Also what we may call economic dynamics does not differ essentially from statics. (Georgescu-Roegen, 1988, 40)

31The second challenge for economic theory is change. Reversibility in mechanics is reproduced in the neoclassical production function: “a process is said to be reversible if and only if it can follow the same course phase by phase in the reverse order” (Georgescu-Roegen, 1966, 83). The fund–flow model takes into consideration the entropy law transforming matter and energy into products and waste. Thus, the production process is irreversible (Georgescu-Roegen, 1988).

32The timeless analysis of static equilibrium in neoclassical economics has been criticised by neoclassical economists. Hicks distinguished the “economics of time” from “economics in time” (see Boland, 1978). “Economics of time” refers to t as a parameter, whilst “economics in time” departs from logical time by introducing change and thus irreversibility. Time “t as duration” used in Georgescu-Roegen’s fund-flow model is compatible with Hicks’s “economics in time” because it introduces change and irreversibility.

33In order to render economics “in time”, growth theorists developed dynamic models to explain the change and irreversibility of economies as a result of novelty and technological change, and thus increasing returns to scale. Boland suggested learning as an irreversible process in a Hayekian way (1978):

However, neoclassical economists could not remain for long indifferent to Change, that is, to the passage of time. With the strong mechanistic orientation of neoclassical economics, the interest in the role of time sprung in models that did not go beyond the elementary differentials with respect to time. The first signal came from a small monograph by Griffith E. Evans, a world-famous mathematician. But the novelty soon turned into a fashion especially after economic growth became the grand objective of our discipline. (Georgescu-Roegen, 1988, 40-41)

34Georgescu-Roegen discussed the integration of time in neoclassical growth models as the simple addition of an independent variable t representing time and virtually all possible changes (technological, of scale and factorial), of course without explaining change originating from social, institutional conditions. Time equals change rather like a magic potion in these models:

In the end, time came to represent the cause of change in a way that recalls Aeschylus’s idea that time brings about anything. In virtually all growth models t, an ordinary variable, is the ultimately independent variable. Many—e.g., Robert M. Solow and Allen—have even used that variable to represent technological change through the production function Q=F(L,K,t).

The wide-spread expression conceals from view the absurd idea that technological level is measurable both in the past and in the future. Also, in the so-called dynamic models of economic growth such as the salient ones of Sir Roy Harrod and Evsey D. Domar, time is the only effective element of change, all other variables representing initial conditions. (Georgescu-Roegen, 1988, 41)

35Taking a long-term view worsens the effect of timelessness in neoclassical economics. In Marshall’s work ([1890] 1961), production costs tend to decrease in the long run:

For short periods, the difficulties of adjusting the internal and external organization of a business to rapid changes in output are so great that the supply price must generally be taken to rise with an increase, and to fall with a diminution in the amount produced. But in long periods both the internal and the external economies of production on a large scale have time to develop themselves. (Marshall, [1890] 1961, 501-502).

36The long term increasing returns are problematic because they imply qualitative changes that are incompatible with the static approach of equilibrium. The long term supply curve requires reversibility. According to Schumpeter, Marshall was trying to make this curve more “realistic”, but by so doing he ruled away its reversibility. In Schumpeter’s words, the supply curves were “fragments of economic history through diagrams” (Schumpeter 1954). Marshall’s concept of a “law” of industrial organisation, i.e. the progressive specialization that increases productivity, was an evolutionary law, a biological analogy of the law of evolution.

37Long-term dynamics (with irreversibility) should replace long-term comparative statics (see Robinson, 1980)—and thus historical curves would also replace long-term static cost curves (see Schumpeter, 1954, who explains Marshall’s inconsistencies). This implies a time concept and a time’s arrow from the past to the future. But this is an evolutionary concept of time that does not fit into the mechanics-based analogy of neoclassical equilibrium. Marshall’s biological analogy of the individual firms as the “trees of the forest” would be a much better choice, in Georgescu-Roegen’s view. Increasing returns involve irreversible transformation and an evolutionary concept of time.

2. Historical Time From an Open-System Perspective

38While analytical similes can introduce time through duration and change (irreversibility) in economic models, they cannot fully explain the causes of change. We find here the core problem in economic theory, and the origin of the dissatisfaction that reoriented Georgescu-Roegen’s approach from modelling through analytical similes towards an open-system perspective.

39The main reason for discontent is that modelling effective causes to produce the same effects in given conditions again falls into the mechanical-analogy trap. In other words, history never repeats itself in the same way. For Georgescu-Roegen, economic and mechanical processes are incompatible:

By contrast, although economic crises have occurred one after another with some regularity, none was a repetition of the previous ones in the same sense as one apple falls just like all others. Each economic cycle was provoked by different factors and had different effects. In this difference lies Schumpeter’s idea of the difference between dynamic time, which only repeats, and historical time in which novelty emerge. (Georgescu-Roegen, 1988, 42)

40Different causes produce different effects in the economic cycle. Looking to individual processes in isolation induces a limit in economics. Only in physics can this isolation be effective through experiments. But biology and economics are concerned with evolutionary environments. Interdependence requires an open-system methodology, dealing with linkages, with relations inside wholes. Georgescu-Roegen stresses this point:

Perhaps in some domains it might be unscientific to experiment with wholes because, as Popper argues, we cannot thus impute effects to individual causes. That does not apply, however, to evolution which is inseparable from the Whole. Witness the fact that the only case in which we were able to formulate an evolutionary law is that of the whole universe. (Georgescu-Roegen, 1966, 88-89)

41Another important limit of using clock-time in economics is the presence of life itself, and the variety of time perceptions related to human consciousness. Humankind is the true agent of time. From philosophy, Georgescu-Roegen (1988) adopts Whitehead’s definition of time as “the flux of things” and not “the flux of a thing”. Thus, time is not monochromatic like clock-time or mechanical time. It is a “polychromatic matrix”, because things are many. According to Georgescu-Roegen, time in economics cannot be reduced to a monochromatic clock-time concept.

2.1 Historical Time is More than Irreversible; it is Irrevocable

42In the previous section, we explained how neoclassical growth economists introduced the historical factor through irreversibility. We also mentioned that Georgescu-Roegen does not reject physical analogies in economics. However, he does argue that the historical factor plays a very different role in economics and biology. Physical hysteresis can be an object of study, because the physicist can at any time find non-magnetised iron to prove that all magnets behave identically. Magnets do not have history. But economics and biology cannot find facts without history: ahistorical events do not exist in social sciences and biology. The historical factor cannot be reduced to a zero level. Turk echoes Georgescu-Roegen’s historical-time approach:

Historical time is not simply irreversible, in accord with the sense that time necessarily flows from past to future. Instead, it involves human activity and thus a stage, human history—especially economic history—on which economic activity and events transpire. (Turk, 2012, 477)

43Going back to Time in nature as a “primitive stream of consciousness” (see Section 1 of this paper), we understand that it moves only forwards. There is a temporal order established between two states of the universe S1 and S2, and this must be the same as that observed by a single or collective consciousness assumed to be contemporary to both states. It is thus assumed that Time is irrevocable through human consciousness. The irrevocability condition is formulated in classical thermodynamics. Georgescu-Roegen (1966) recalls this in order to call for the irrevocability of Time, and thus of economic processes.

44According to Georgescu-Roegen, there is a narrow sense of irreversibility: processes are irreversible but not irrevocable. Irrevocability is the condition of evolution: “The existence of evolutionary laws in nature depends upon whether there are irrevocable phenomena: the existence of only irreversible phenomena—in the narrow sense—does not suffice” (Georgescu-Roegen, 1966, 84).

45The originality of Georgescu-Roegen’s defence of historical time stems from his discussion of time in physics and of the irrevocability condition. He compares the time conceptions in statistical mechanics and in classical thermodynamics. In statistical mechanics, the constant-entropy-of-the-universe assumption implies that there are no evolution laws, that shuffling and sorting are possible, and that Time is reversible (i.e. there is no order). But the time arrow also affects statistical mechanics: “Historical time is a one-dimensional ordinal variable anchored by known chronological dates whether connected with mankind’s or with cosmological history … This time is not a cardinal variable” (Georgescu-Roegen, 1988, 42). In classical thermodynamics, the entropic degradation of the universe is irrevocable, because “the free energy once transformed into latent energy can never be recuperated” (Georgescu-Roegen, 1966, 84).

46Time’s arrow is an underlying concept of evolution and requires irrevocability. Irreversibility is a necessary but insufficient condition. According to Georgescu-Roegen, time’s arrow is necessary for defining reversibility as well—as, for example, in mechanical processes.

47Authors such as Boland define real-time processes as simply irreversible processes. In his 1978 paper on Hicks’s distinction of “economics of time” (time as just a parameter) and “economics in time” (time mattering because it is irreversible), Boland asserts that introducing historical time means introducing learning as an irreversible process in a Hayekian way (Boland, 1978). In a recent paper on time in economics, Boland revives the same argument: “time matters because it is irreversible” (Boland, 2005, 121). For Boland, modelling time is like copying the law of entropy:

While the Entropy Law would seem an obvious way to incorporate time’s arrow and thereby avoid the limitations of mechanical models, it is still an open question as to whether any social theory can be seen in a parallel fashion. (Boland, 2005, 122)

48According to Boland, neoclassical economics is not “timeless”. For Boland, the question is whether the treatment of time is adequate: “a model is not timeless if any of its parts is not a tautology” and “That a model or any explanation can be shown to be logically valid does not say that its truth status (as a compound statement) is timeless” (Boland, 1978, 249).

49Introducing knowledge is, according to Boland, a way to introduce learning processes—and thus irreversible processes—into economic models. But this is not sufficient, because irrevocability is an even stricter condition, stating that time’s arrow is one-way, and it is impossible to get back to a previous situation in any way. In Georgescu-Roegen’s view, it is an evolutionary law of nature:

evolution is the process that links birth to death: it is life in the broadest sense of the term. Witness the fact that even the whole universe must have a transient life between Creation and Heat Death if it is to be an evolving entity as pictured by the Classical Entropy Law. (Georgescu-Roegen, 1966, 87)

50If we use Hicks’s definition, standard economics is not timeless, but is “out of time”—and, according to Georgescu-Roegen, irreversibility is not sufficient to render it “in time”, as Boland asserts. History is the missing agent of change in standard economics, because it introduces a qualitative and crucial distinction between processes happening at distant points in Time. Today, production and consumption processes are different from yesterday. Their physical, social and biological context has evolved. To clarify this essential distinction, we offer here a long quotation from Georgescu-Roegen’s most recent paper (1994) on Time in economics:

time is plurychromatic, it enters in economics in multiple ways, both in production and in life enjoyment. But I have only alluded to that aspect of time that is gauged in its mysterious passing that has led to the peculiar image of history as an agent of change. The journey of a spaceship, say, to Mars certainly takes time, which, as Schumpeter distinguished it, is dynamical time because the phenomena it supports can be reproduced over and over again within very large circumstances. An adage says that history repeats itself, yes there has been one war after another, but the Punic War was surely different, essentially different, from the Gulf War. This is the historical time that Schumpeter (1954) opposed to dynamical time in connection with a still more important distinction—between “growth”, mere accretion and “development”. Standard economics stops its interest just as the problems turn to what happens in history and why, which means that the essential influence of time on the life of humans is not of the dominion of standard discipline. This is what must have escaped L.A. Boland (1978, 241) when he took to task Sir John Hicks (1976f) together with Shackle (1972) and myself (1971) for having maintained that standard economics is out of time. (Georgescu-Roegen, 1994, 256)

51Irrevocability and history are the necessary conditions for an evolutionary economic analysis; for Georgescu-Roegen, biology and economics are both evolutionary disciplines. And once we open the evolutionary Pandora’s box, the problem of boundaries between sub-systems and processes emerges as starkly relevant.

2.2 Open-system Methodology and the Evolutionary Perspective

52The analytical choice of boundaries and the separation of partial processes are delicate operations. How can we isolate processes that are simultaneous and interdependent? How can we isolate them in time without breaking essential causal chains? Georgescu-Roegen asserts the permeability—the dialectic penumbra—of the boundaries of economic processes. In this section, we study the different elements of Georgescu-Roegen’s evolutionary perspective, and explain why they require an open-system methodology.

53According to Georgescu-Roegen, the most appropriate definition of economics is Marshall’s “study of mankind in the ordinary business of life” (1966, 107). From an evolutionary perspective, humankind’s study is a difficult task, because there is no possible observation of other “humankinds” from which we could learn any Timeless laws. Humankind evolves biologically, socially and economically along Time’s arrow—and is characterised by social conflict. Following Lotka, Georgescu-Roegen explains how exosomatic instruments have been developed by human economic activity to give humankind an advantage over other species in using free energy. These instruments are very unequally distributed among individuals and communities, thus leading to conflicts. It is the exosomatic evolution of humankind that has transformed the production process into a social enterprise. It is not possible to reverse the endosomatic and exosomatic evolution. Development is evolution, and economic processes are irrevocable, evolutionary processes. In the concept of exosomatic evolution, we have interrelated economic, biological, physical and social factors. Economic facts are social facts, and economic value is social value. The exclusion of humankind from the discipline of economics is absurd. For Georgescu-Roegen, it is the exclusion of humankind from economic analysis that has contributed to creating a Timeless approach:

The ideal of a man-less science is gradually losing ground even in physics—ever since the discovery of quantum of action and Heisenberg’s indeterminacy. However, for a science of man to exclude altogether man from the picture is a patent incongruity. Nevertheless, standard economics takes special pride in operating with a man-less picture. As Pareto overtly claimed, once we have determined the means at the disposal of the individual and obtained “a photograph of his tastes … the individual may disappear”. The individual is thus reduced to a mere subscript of the ophelimity function Φ(X). The logic is perfect: man is not an economic agent simply because there is no economic process. There is only a jigsaw puzzle of fitting given means to given ends, which requires a computer not an agent. (Georgescu-Roegen, 1966, 104)

54As Time is defined by the “stream of consciousness”, its origin is anthropologic. Eliminating Time goes hand-in-hand with eliminating humankind from economic analysis. Timeless approaches invalidate evolutionary concepts such as increasing returns, because they are irreversible, and perhaps irrevocable as well. According to Georgescu-Roegen, the only authors who have transcended the static, Timeless framework are Marx, Schumpeter and Veblen. His criticism echoes the Methodenstreit rejection of a universally valid economic theory. Economic theory is historically and institutionally embedded—and thus, neoclassical theory, despite its aspiration to universal validity, actually reflects large urban communities in industrialised societies (Georgescu-Roegen, 1966, 109). By ignoring the evolutionary nature of all economic processes, standard economists are condemned to preach universal validity of their theory.

55Georgescu-Roegen stresses the importance of the cultural matrix in all human choices. Peasant communities may seem “irrational” from the standard-economics perspective when we disregard their cultural evolution. Georgescu-Roegen pleads for pluralism in economics, for a humanised science:

If we deny man’s faculty of empathy, then there really is no game we can play at all, whether in philosophy, literature, science, or family. Actually, we must come to recognize that the game is not the same in physical sciences as in sciences of man; that, contrary to what Pareto and numberless others preached, there is not only one method by which to know the truth. But perhaps one day we will all come to realize that man too is an instrument, the only one to study man’s propensities. That day there will be no more forgotten men, forgotten because today we allegedly do not know how to study them and report on what they think, feel and want. (Georgescu-Roegen, 1966, 29)

56Mechanistic epistemology is the root of misunderstanding the economic process. Georgescu-Roegen explains that clock-time in a closed system leads to a permanent circular flow of matter and energy that contradicts the laws of physics. The physical nature of the economic process is not circular, but unidirectional, continuously transforming low entropy into high entropy in the form of irrevocable waste. The Time of thermodynamics introduces this qualitative change into economic analysis. This conception of the economic process is similar to Schrödinger’s formula (or paradox) for the biological process of a living cell or organism, i.e. the unidirectional transformation of low (or, in Schrödinger’s terms “negative”) entropy into high entropy (or “positive entropy”) by a living organism (see Schrödinger, 1944, 71). However, analogies—isomorphism between economics and physics, or between economics and biology—are too reductionist. Economics cannot be reduced to physics or biology. The entropic process of the material environment is automatic, while the entropic economic process depends on human activity. The first is only “shuffling”, while the second is also “sorting”. Human action makes the difference between the pure physical and economic processes, in the form of institutions, of social and cultural determinants. And above all, the purpose of all economic processes is the “enjoyment of life”. The real source of economic value is the value of life. Purposive activity and enjoyment of life differentiate economics from physics. Economics is also more than biology—as all economic processes, through exosomatic instruments, are a continuation of the biological ones.

57Economic mechanisms cannot be reduced to physics or biology. As an open system, the economic system involves strata with causal mechanisms. According to the critical-realism approach, emerging properties may be rooted in the strata below, but they cannot be reduced to it. Georgescu-Roegen’s idea of dialectic penumbras or permeable boundaries of systems can be explained by Dow’s general systems theory, and also by Lawson’s critical realism and Mearman’s open-system approaches. Despite some disagreements on the definition of closed systems (Lawson, 1997; defining closeness using event regularities), these authors insist on the “openness of economics and to economics from other disciplines” (Mearman, 2003, 15). The critical realist world vision is an open system one, i.e. complexly structured, open, dynamic, characterized by emergence and novelty and “inclusive of totalities and causally efficacious absences, amongst other things.” (Lawson, 1997, 65). In this approach, “emergence may be defined as a relationship between two features or aspects such that one arises out of the other and yet, while perhaps being capable of reacting back on it, remains causally and taxonomically irreducible to it.” (Lawson, 1997, 63). He explains “real emergence clearly turns on the openness of being, a rejection of actualism (the doctrine that the real is reducible to the actual) and necessitates a conception of levels or strata (the physical, the biological, the social, etc).” (Lawson, 1997, 63).

58Historical time is a necessary condition of openness, ensuring time-consistency between the different strata or disciplines involved in the study of economic processes. Post-Keynesian economists such as Victoria Chick emphasise the role of history in an open-system approach: “History can make a theory which was right for its time become wrong as the economy evolves and the theory ceases to capture the salient relationship” (Chick, 2004, 11).

59By anchoring theory into the ocean of reality, Time fully sustains the open-system approach: “In an open-system approach, consistency is understood as a relation between theory and reality, rather than internal to theory, as in a closed-system approach” (Chick and Dow, 2005, 378).

60We interpret Georgescu-Roegen’s open-system approach as an exosomatic evolutionary analysis of economic processes. It involves a preference for historical time, which is also known as evolutionary time, entropy time or Time’s arrow.

61We will now turn to the methodological consequences of rendering economic analysis “in Time”.

2.3 Dialectical Reasoning and “Analytical Similes”

62According to Georgescu-Roegen, “historical time, the bed of real change, can be discussed fruitfully only by dialectics” (Georgescu-Roegen, 1994, 257). He refers explicitly to Marx’s dialectics in three steps: thesis, antithesis and synthesis. Being a faithful Schumpeterian, Georgescu-Roegen asserts that the three main qualities of an economist are: knowledge of economic history, knowledge of socioeconomic conditions, and analytical talent (Georgescu-Roegen, 1988, 44-45). He adopts the dialectic reasoning of Schumpeter’s epistemology. Historical time is a feature of Schumpeter’s evolutionary pattern that has acquired a legitimate fame among biological evolutionists, such as Levins and Lewontin (1985).

63An example of Schumpeter’s dialectics is the difference between small and great innovations; there is a dialectic penumbra separating the two concepts. Another example is the concept of entrepreneur, whose start and end points are blurred. At the core of these dialectic concepts, we find the idea of change. Sir John Hicks (1976) also argued that the frontiers of technological change are blurred, due to the dialectic concept of novelty. Marshall also expressed the dialectic nature of time when he explained that the short and long periods are separated by a dialectic penumbra: “It was a clear warning from one of the greatest economists that process is an intricate concept hard to represent arithmomorphically” (Georgescu-Roegen, 1988, 39). Lundberg and Hicks developed the “sequence analysis” in order to introduce some aspects of change into dynamic economic models. It is as if every economic process is divided into weeks (as an example of a finite period), and each Monday past decisions are adjusted in line with the evolution of data and expectations. This approach was viewed as a way of explaining economic fluctuations in time. It is essentially different from Marshall’s distinction between the short and the long run. Marshall explained how novelties enter the economic system and trigger evolution. Marshall’s argument is fully dialectic, while the Lundberg–Hicks model is arithmomorphic.

64Georgescu-Roegen argues that when it comes to change and evolution, dialectics must be preferred to arithmomorphism. By recalling the Aristotelian distinction between locomotion, quantitative changes and qualitative changes, Georgescu-Roegen (1988) argues that in economics, locomotion and quantitative changes can be described by arithmomorphic concepts. Qualitative change is an essential aspect of Time. But only a few qualities can be quantified at the price of disturbances to the linear arithmomorphic structure: “That the essential nature of qualitative change, of evolution (as understood in all sciences of man, at times even in astronomy) is dialectical has been pointed out, among other great thinkers, by Charles Darwin himself” (Georgescu-Roegen, 1988, 37, emphasis is mine).

65Georgescu-Roegen extends his argument against arithmomorphism even further by expressing a strong criticism of growth models. He mentions Solow (1964) and Allen (1967) who both used t as an independent, ordinary variable to represent technological change through the production function Q=F(L, K, t). For him, not only qualitative change but also growth-based input–output models require dialectics. Growth and change are evolutionary concepts. Biologists have for years studied the multiplication of cells. Economists cannot rely solely on paper-and-pencil models because “when we try to connect the symbols with facts we discover their basic but astounding implication that, say, either bulldozers are consumption goods or yogurt is capital equipment” (Georgescu-Roegen, 1988, 41-42). In a critical analysis of growth models, Georgescu-Roegen demonstrates that Leontief’s open input-output growth system (Leontief, 1951) no longer works if we separate consumption goods from production goods (as Marx did in his production model). Put plastically, the accumulation of consumption goods like yogurt cannot increase the number of bulldozers. Capital equipment is specific (see Georgescu-Roegen, 1976, 248).

66Georgescu-Roegen (1966) states that if economics is concerned with humankind and not only with “observable” quantities, then it must rely on dialectic reasoning. The correctness of dialectic reasoning can be tested through two methods: Socratic analysis and “analytical simile”.

67Analytical similes are authorised simplifications, reductionist reasoning, cautiously used because of the risk of fallacy. In analytical similes, we can use t. Similes reflect the Ernst Mach’s philosophy of sciences—the “economy of thought” in Georgescu-Roegen’s epistemology.

68Economics uses the mechanics-based analogy—for example, “the mechanics of utility and self-interest” that involve conservation (given means) and maximisation (optimal satisfaction). But economics cannot be reduced either to physics or to biology by a strict isomorphism (Georgescu-Roegen, 1966, 97).

69Georgescu-Roegen restricts the use of these analytical similes to the verification of dialectic reasoning that has the prevailing role: “To be sure, we can use analytical models in connection with qualitative changes, with some risks, however; but these models are only similes which may help us to detect possible errors in our dialectical reasoning” (Georgescu-Roegen, 1971, 337-338).

70This restrictive use of analytical similes implies a very discerning choice of boundaries or closures of economic processes under examination. The open-system methodology echoes these cautions by explaining that the choice of these closures may be inappropriate:

In an open-systems reality, methods which involve closure imply a compromise by the researcher. This should be recognised: there is a disjuncture between closed-systems methods and reality. This does not mean a lack of correspondence between the reality and methods—they cannot correspond—but that the methods introduce inappropriate closures. (Mearman, 2003, 20)

71Georgescu-Roegen distinguishes a blueprint from a simile—that is, homology (similarity in form or identity of structure) from analogy (similarity in function), in Bernard Cohen’s words 1994). In his paper “Economic Theory and Agrarian Economics” Georgescu-Roegen (1960) explains why an economic model is not a blueprint but an analytical simile that guides only an initiated scientist. A blueprint is a symbolic and operational representation of a mechanism. It is rather like an assembly guide for an apparatus, such as a radio. The uninitiated can easily follow the blueprint in order to build the radio. But this is not the case with economic models, which cannot simply be applied like blueprints. Georgescu-Roegen warns economists against the temptation to misuse pseudo-arithmomorphic concepts such as price index, cost of living, and aggregate production. Nonetheless, macroeconomics has made an increased use of these.

72In a recent article, Turk expresses a similar view on the dialectic methodology and the dialectic penumbras:

In a mathematical sense, historical contexts might be thought of as recursive, and their boundaries are always blurred. An evolutionary process may show the means by which one moved in time from A to B, but does not indicate how and why A emerged in the first place. (Turk, 2012, 489)

73Turk goes on to ask: “can activities conducted by humans, the subject matter of economics, be understood in relation to changes in real time, and yet be analysed through abstracted models that follow from a discrete set of general rules?” (Turk, 2012, 489).

74Macroeconomic models of growth—such as those of Michael Kalecki (1962) and Richard Goodwin (1991)—introduced instability, structural change, and cyclical movements for investments and innovations. But, according to Turk, all these models rely upon parametric, logical time t by excluding historical events and referring only to endogenous elements: “Moreover, the effort to create a model for the law of motion of capitalism presents an obvious contradiction with Robinson’s critique of time, in that any such law resonates with the Newtonian frame of reference she had rejected” (Turk, 2012, 490).

75In a very Machian “economy of thought” spirit, Georgescu-Roegen (1966, 104) argues that “the most valuable ladder of any science” is abstraction. But this is not satisfactory if it never gets down to reality:

However, the task of science is not to climb up the easiest ladder and remain there forever distilling and redistilling the same pure stuff. Standard economics, by opposing any suggestion that the economic process may consist of something more than a jigsaw puzzle with all its elements given, has identified itself with dogmatism. (Georgescu-Roegen, 1966, 104)

76Economic processes­­—like chemical, biological processes—do not have clear boundaries, but are surrounded by dialectic penumbras. The penumbra is even wider in the case of economics, which intertwines with social and political processes. Temporary closures and temporary boundaries have to be removed when it comes to understanding the causes of change, the inner engines of human evolution.

77The methodological conflict between dialectics and arithmomorphism originates in the introduction of historical time T. Many economists prefer avoiding the very uncomfortable dialectic and inductive method. Boland criticises Georgescu-Roegen’s dialectic approach and defends a positivist, arithmomorphic, deductive-logics approach:

Georgescu-Roegen’s allowance of dialectical concepts is nothing more than an admission of defeat. Since we cannot prove (using standard logic) that our knowledge of change is true—even when it is true—he abandons any pursuit of truth by allowing truth and falsity to coexist. This acceptance of contradictions can be attributed to his desire to maintain a belief in induction as the sole basis of “knowing” whenever knowledge is about observed facts. (Boland, 1978, 258)

78Georgescu-Roegen has a balanced position that is similar to that of the open-system economists. According to Mearman (2003, 3), “deductive logic exemplifies a group of ‘closed-systems methods’ because it entails the assumption that closed systems exist in reality; or they impose an assumption of closure on a reality.” Mearman explains the limits of deductive logics in open systems. If the mechanisms reside in strata, and if the strata have emergent properties, then the outcome of deduction is modified. Mearman’s argument, applied to two strata (philosophy and economics) can also be applied to physics, biology and economics. When we say that economics should take into consideration physical laws such as the entropy law, but that it should not copy physics because its domain is not reducible to physics, we mean that economics is rooted in—but irreducible to—physics. Entropy law affects economics, but economics cannot be reduced to entropy. Economics also inherits evolutionary properties from biology. All domains can be conceptualized as being on different strata of reality. The stratified reality is a hierarchical order of levels, where the lower level creates the conditions for the higher level. Each level has specific generative mechanisms. At the bottom, we find the physical mechanisms. Then follows the biological, psychological and at the highest level the social mechanisms (see Danemark, 2001). Economics is at the highest level.

79Now comes the emergent property effect. Economics has emergent properties; this means that these properties cannot be found in the lower strata (physics and biology)—for example, exosomatic evolution. Each higher stratum has developed characteristics that are not contained in the lower stratum, so we cannot deduce them from this. These characteristics, called also emergent properties, can be of two kinds: the first is a modified characteristic from a lower stratum (for example, exosomatic instruments is a modified biological characteristic in economics); and the second is a new characteristic, which is specific to the economic stratum (for example, increasing returns).

80The open-system approach requires the adoption of dialectic epistemology and historical time T, allowing all properties that are deduced (from lower strata) and emergent to be included in the economics stratum. The definition of Time itself is dialectic, as its boundaries are dialectic penumbras.

3. Conclusion

81Georgescu-Roegen’s plea for historical time is original. It is rooted in physics, and in biological evolution, but also embraces institutions and philosophy. For him, Time is the stream of consciousness; it is a unidirectional flow passing through the observer’s consciousness and driving irrevocable changes. As the universal time of consciousness, it encompasses all other forms of time.

82In Georgescu-Roegen’s view, historical time is the Time of thermodynamics, the entropic time, defined in an open-system perspective, which leaves space for all historical events and human activities. It is an anthropologic vision of time rooted in physics, and compatible with the biological evolutionary perspective of human species and the biosphere. This Time concept no longer allows any opposition between natural and social sciences.

83But the gain of such openness and compatibility between social and natural sciences has a high price. It calls for dialectic epistemology. This orientation is seen as an admission of defeat for many mainstream economists (see Boland, 2005 and popperian epistemology) who have tried hard to quantify, model and deduce economic Timeless laws by using logical time t. Parametric time t allowed the use of modelling techniques through the ceteris paribus assumption that assumes a partial closure of the economic system.

84However, Georgescu-Roegen’s plea for a historical-time and institutional approach does not completely dismiss the use of arithmomorphic models as analytical similes. He admits that analytical similes based on t can still be used in many cases as a deductive logic for checking the correctness of the dialectic reasoning. But the choice of these analytical similes and the closures they impose on an open system requires a very cautious, very discerning use (see Chick and Dow, 2005 on these partial closures in the open-system approach).

85The present paper has explored Georgescu-Roegen’s thoughts on time in economics. I presented his philosophical and physical definition of Time. Georgescu-Roegen explained the difference between mechanical time, clock-time, and thermodynamic, entropic Time. The entropy law is an evolutionary law with a Time’s arrow.

86Neoclassical economic theory was built upon a mechanics-based analogy involving clock-time. But even in this parametrical definition of time, mainstream economists fail to treat duration, change and irreversibility. Put differently, neoclassical economic models are not only Timeless, but also timeless (or “out of time”, if we use Hick’s terms). Georgescu-Roegen proposed an analytical representation of the production process that uses time as duration (the fund–flow model), and that integrates the unidirectional entropic transformation, producing not only output but also waste. To render economics “in time” and to model change, many growth theorists have added the independent variable t, which represents time and irreversible changes. Even if these dynamic models have managed to introduce irreversibility, they are still subject to criticism because they cannot reveal the real cause of change. Instead, the main cause of change is time itself. Georgescu-Roegen has argued that the physical analogy cannot apply to economic processes, and should be replaced by the evolutionary view. The example of increasing returns, an economic mechanism that is incompatible with long-term statics, is the illustration of the incompatibility between the concepts of mechanical, reversible time and of evolutionary, unidirectional, irreversible time.

87All the critiques that we presented (Section 1) pave the way to the adoption of Time—real, historical time, in which all changes are rooted. History is the agent of change. This Time is irreversible—and also irrevocable. Time as duration, time as irreversible and irrevocable, time for the enjoyment of life, and time as a scarce lifetime gift: these are some of the multiple facets of time. For Georgescu-Roegen, time is polychromatic. In his approach, different conceptions of time are reconciled if we look to Time through the eyes of a human. Today, even the discipline of physics, thanks to quantum theory, recognises the observer’s interaction with physical laws. Perhaps in economics, too, time has come to recognise that a science of humankind cannot be reduced to the mechanics of utility and self-interest.

I would like to thank the anonymous reviewers for their constructive comments. The usual disclaimer applies.

Haut de page


Allen, R.G.D. 1967. Macroeconomic Theory. A Mathemarical Treatment. New York: St. Martin’s.

Bernard Cohen, Ierome. 1994. Interactions: Some Contacts Between the Natural Sciences and the Social Sciences. Cambridge, MA: MIT Press.

Boland, Lawrence. 2005. Economics in Time vs Time in Economics: Building Models so that Time Matters. History of Economic Ideas, 13: 121-132.

Boland, Lawrence. 1978. Time in Economics vs. Economics in Time: the “Hayek problem”. Canadian Journal of Economics, 11: 240-262.

Chick, Victoria and Sheila Dow. 2005. The Meaning of Open Systems. Journal of Economic Methodology, 12(3): 363-381.

Chick, Victoria. 2004. On Open Systems. Brazilian Journal of Political Economy, 24(1/93): 3-16.

Danemark, Berth. 2001. Interdisciplinarity Research and Critical Realism—the Example of Disability Research. International Association for Critical Realism Conference. Denmark,

Georgescu-Roegen, Nicholas. 1994. Time in Economics. In Harald Hagemann and Omar F. Hamouda (eds), The Legacy of Hicks: His Contribution to Economic Analysis, London: Routledge, 241-256.

Georgescu-Roegen, Nicholas. 1988. Time and Change in Economics. In Eberhard K. Seifert (ed.), Okonomie und Zeit, Frankfurt am Main: Haag and Herchen, 29-52.

Georgescu-Roegen, Nicholas. 1976. Energy and Economic Myths: Institutional and Analytical Economic Essays. New York: Pergamon.

Georgescu-Roegen, Nicholas. 1971. The Entropy Law and the Economic Process. Cambridge, MA: Harvard University Press.

Georgescu-Roegen, Nicholas. 1966. Analytical Economics: Issues and Problems. Cambridge, MA: Harvard University Press.

Georgescu-Roegen, Nicholas. 1960. Economic Theory and Agrarian Economics. Oxford Economic Papers, 12: 1-40.

Goodwin, Richard M. 1991. Schumpeter, Keynes and the Theory of Economic Evolution. Journal of Evolutionary Economics, 1(1): 19-47.

Gossen, Hermann H. 1983. The Laws of Human Relations and the Rules of Human Action Derived Therefrom. Trans. by Rudolf C. Blitz. Cambridge, MA: The MIT Press.

Hicks, Sir John. 1976. Some Questions of Time in Economics. In Anthony M. Tang, Fred M. Westfield, and James S. Worley (eds), Evolution, Welfare and Time in Economics. Toronto: Heath, 135-151.

Kalecki, Michael. 1962. Observations on the Theory of Growth. Economic Journal, 72(285): 134-153.

Lawson, Tony. 1997. Economics and Reality. London: Routledge.

Leontief, Wassily. 1951. Studies in the Structure of the American Economy 1919-1939: An Empirical Application of Equilibrium Analysis. New York: Oxford University Press.

Levins, Richard and Richard Lewontin. 1985. The Dialectical Biologist. Cambridge, MA: Cambridge University Press.

Lotka, Alfred. 1925. Elements of Physical Biology. Baltimore: Williams & Wilkins (reprinted in 1956, New York: Dover).

Marshall, Alfred. [1890] 1961. Principles of Economics, 9th edition. London: Macmillan.

Maxwell, James Clerk. 1921. Theory of Heat, 10th edition. London: Dover Publications.

Mearman, Andrew. 2003. “Open systems” and Economic Methodology. Mimeo, 1-28,

Robinson, Joan. 1980. Time in Economics. Kyklos, 33(2): 219-229.

Schrödinger, Erwin. 1944. What is Life? Cambridge, UK: Cambridge University Press.

Schumpeter, Joseph Alois. 1954. History of Economic Analysis. New York: Oxford University Press.

Shacke, George L. S. 1972. Epistemics and Economics. Cambridge: Cambridge University Press.

Shackle, George L. S. 1967. Time in Economics. Amsterdam: North-Holland Publishing Company.

Solow, Robert. 1964. Capital Theory and the Rate of Return. Chicago: Rand McNally.

Solow, Robert. 1957. Technical Change and the Aggregate Production Function. The Review of Economic and Statistics, 39(3): 312-320.

Turk, Michael H. 2010. The Arrow of Time in Economics: From Robinson’s Critique to the New Historical Economics. European Journal of the History of Economic Thought, 17(3): 471-492.

Vittucci Marzetti, Giuseppe. 2013. The Fund-flow Approach: A Critical Survey. Journal of Economic Surveys, 27(2): 209-233.

Weyl, Hermann. 1950. Space, Time, Matter. New York: Dover Publications.

Haut de page

Pour citer cet article

Référence papier

Roxana Bobulescu, « The Original Time Approach of Georgescu-Roegen », Œconomia, 7-1 | 2017, 87-109.

Référence électronique

Roxana Bobulescu, « The Original Time Approach of Georgescu-Roegen », Œconomia [En ligne], 7-1 | 2017, mis en ligne le 01 mars 2017, consulté le 18 janvier 2018. URL : ; DOI : 10.4000/oeconomia.2580

Haut de page


Roxana Bobulescu

Grenoble Ecole de Management, Chair « Mindfulness, Well-being at Work and Economic Peace », 12 rue Pierre Sémard, BP 127, 38003 Grenoble, Cedex 01, France,

Articles du même auteur

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • OpenEdition Journals