/ On the Backward Passage from Neo-Classical Utility Maximizers to Value-Based Social Coalitions

1 Universalism, Difference, and Glocalism: A Call to Action

In The Passage West, Giacomo Marramao joins a penetrating analysis of the transformations currently shaping the contemporary socio-political landscape to a powerful argument about the new shape that contemporary thought must take in order to cope successfully with the changed face of modernity. Marramao’s thesis is that “we are confronted with a thrilling and perilous transit to modernity which is intended to produce radical changes [...] not only of the ‘other’ civilizations but of Western civilization itself ” (x). Universalism and difference are the sign posts marking this “transit”: they are the true Scylla and Charybdis we must go through while seeking a passage West that will inaugurate a new era.

Universalism is the effective label Marramao adopts for the prevalent reduction of social subjects to atomic, unconnected, rational individuals only keen on maximizing their utilities. It is the deeply anti-social view of human communities (“There is no such things as society,” as Margaret Thatcher famously declared) that is embedded in the current reduction of any form of politics to the “logic of the market,” whether through a subordination of the former to the latter or through a complete elimination of political processes of mediation. The problem with universalism lies in its inevitable contradiction with local non-economic aggregations. The emergence of local symbolic identities based on values throws the universalist paradigm into question, because values, contrary to interests, are not commensurable (44-45). Values cannot be reduced to a common currency.

With the term “difference,” on the other hand, Marramao denotes the reverse attitude to universalism: the existence of communities who come together on the basis of non-economic value-based principles that supersede and indeed pre-cede the rational preferences of its constituent subjects. The difficulty with local identities is the converse of universalism’s problem. Groups that aggregate around values (and not around interests) cannot be simply tolerated because tolerance presupposes a central authority whose claim to truth and power cannot be questioned, and which, as a consequence, implies “a radical devaluation of the ‘truthfulness’ of the [tolerated] position” (183). In the absence of a central authority, tolerance moves to “respect”. But this only worsens the situation, as respect gives way to new intolerance rising from “‘armor-plated’ differences that relate to one another as monads with neither doors nor windows” (183).

The distinguishing feature of the new epoch of modernity we are going through (“modernita’-mondo” or world-modernity, as Marramao calls it) is that the contradictions affecting universalism and difference have now entered a positive feedback loop. Universal globalization produces differences. These new differences challenge universalism, which extends itself and produces more differences, and so on. The not-so-hidden reason behind this development is what Marramao calls “the end of the Westphalian order.” It is the breakup of the Western system of allocation of power that emerged with the Peace of Westphalia in the 17th century and marked the inauguration of the nation-state as the only sovereign political subject in European politics (13). Globalization has effectively ended the Westphalian order by breaking down the “isomorphism between ‘a people,’ ‘territory’ and ‘sovereignty’ that had hereto sustained it” (35). This all- too-recent break up, characteristic of world-modernity, has generated the contemporary paradoxical phenomenon of glocalism. Marramao argues that we are living in a permanent short-circuit between the global production of locality and the localization of the global. The short circuit between universalism and difference epitomized by glocalism distinguishes this globalization from all the phases that preceded it.

The “passage West” the book calls for is defined and at the same time prompted by globalisation’s new phase and by the short circuit of glocalism it produces. The “passage West,” in other words, is both a condition and a task: glocalism’s positive feedback loop between universalism and difference demands our active intervention, whose general outline Marramao concisely and effectively summarizes in the English preface to his book he wrote in 2012. The tasks ahead, he states, consists in the following:

The reconstruction of a cosmopolitan perspective (which is now caught in the grip of technological standardisation and by diasporic identity) [that] must now pass through a radical redefinition of the universal: we need a universal dimension setting out from the criterion of difference. (xii)

I would suggest that one of the fundamental moments towards the “radical redefinition of the universal” Marramao wishes for entails a reconstruction of the dominant cultural formation that underwrites the contemporary form universalism has taken. I am referring to the discipline of economics, whose epistemological and ontological assumptions provide the philosophical foundations of “universalism” in Marramao’s sense. In my opinion, a necessary (albeit not sufficient) condition for a “radical redefinition of the universal” is an equivalently radical reframing of economic thought that must start with an overhaul of its current technical kernel, game theory, toward a rediscovery of its original political vocation and a future opening to the play of differences.

2 From Economics to Political Economy

Game theory is the highly formalized discipline that studies decision making in social contexts of conflict or cooperation. It was born out of the intense debate that occurred between the second half of the 19th and the first third of the 20th century over the possibility of a mathematical science of economic behavior. After an intense period of development that started in the 1950s, game theory started to invade the core of economic thought from the late 1970s, a position it never relinquished. Nowadays, it provides the most sophisticated models of decision-making carried out by intelligent rational subjects who interact in conflicting and cooperative context. From this point of view, game theory could be considered as the epitome of the universalist position Giacomo Marramao describes in The Passage West. The theory considers human interactions as games in which rational agents always try to choose the strategy that will maximize their utilities (normally, but not necessarily, measured by monetary rewards).[1] Game theory is universalist precisely because it holds a simple, yet universal model that it applies to every single human being: individuals are all, according to the theory, rational utility maximizers and nothing but utility maximizers. Indeed, the extension of game theory’s basic approach to non-economic fields has tried to show how, for instance, the emergence of the seemingly non-egoistic behaviors that could be identified as the core of ethics could be explained as a particularly useful rational strategy carried out by single individuals.[2]

Yet, game theory’s incorporation of rational decision theory and, therefore, its role as universalism’s epitome is, to a large extent, a historical contingency. During game theory’s incubation period (the first third of the 20th century) economic theory was in the final phase of a profound shift. The study of human economic behavior—originally named “political economy” and best exemplified by the classic works of Adam Smith, David Ricardo, and Karl Marx—focused on the study of the production, formation, and distribution of wealth by social groups. Political economy took into consideration and was particularly interested in the economic interactions among social groups and in the possibility of coalitions and/or antagonisms among them. Toward the end of the 19th century, political economy was replaced by the science of economics solidly based on mathematical foundations and rooted in the neo-Classic paradigm, which prescribed that the subject of economics is the study of the large scale effects of independently acting individuals who always choose to maximize their own utility. The goal of economics became the study of the laws that determine the coordinated behavior of all the subjects involved. It became the study of the laws describing the behaviors that bring the economic system to equilibrium (or, equivalently, that may push it away from it). Since prices, in neo-classic economic theory, are determined by individual (person-to-person) transactions, the existence of a general equilibrium represents the crowning achievement of the neo-Classic economic school. The existence of a general equilibrium demonstrates that the overall behavior of a complex system can be explained by a large but solvable system of equations describing the behavior of single individuals. A simple—indeed, from many points of view a simplistic— existence proof was provided by Léon Walras when he showed that the number of equations required for the formal description of an economic system contains an identical number of unknowns. Following Walras’ work, throughout Europe but especially in Vienna, an intense discussion erupted on the possible mathematization of the discipline of economics. Karl Menger, Ludwig von Mises, Friedrich Hayek, Karl Polanyi, and Oskar Morgenstern were at the center of the debate and game theory—which received its first canonical exposition in Morgenstern and von Neumann’s 1944 work—is one of its direct outcomes.

Interestingly, Morgenstern was very critical of the neo-Classic paradigm and in particular of the concept of “general economic equilibrium” that Walras had introduced and which would become one of the staples of economic theory. Morgenstern was keen to stress the implicit contradiction between the existence of such equilibrium and the neo-Classic explicit assumption of economic agents’ perfect knowledge. He pointed out that the theory assumes that economic subjects must be able to rank their own preferences and determine the effects of their own transactions on prices at the same time as they determine the effects of every other economic agents’ present and future transactions on the same prices. Such agents would not just be rational, “they would be semi-Gods,” he concluded (1935, 341–342). The theory of games and strategic behavior he set out to develop with John von Neumann a decade later was explicitly devised as an antidote to such improbable assumptions, and as an explicit rejection of the perfectly rational utility-maximizing subject that lies at the center of neo-Classic “universalist” economics.

In its original 1944 formulation, game theory sees economic interactions as essentially marked by political behavior. Strategic interactions in actual societies are never modeled as a sum of individual-to-individual transactions among a multitude of equally rational and similarly undifferentiated universal subjects. On the contrary, economic interactions are always conducted by “coalitions” of players who aggregate on the basis of extra-economic (political) principles, and where the distribution of power is inevitably unequal.[3] Already in his initial 1928 venture into game theory, and more explicitly in the 1944 work with Morgenstern, von Neumann emphasized that a collection of three or more subjects represents a set of virtual coalitions only a few of which can be actualized in a concrete strategic confrontation. In a 3-player game, for instance, there are 3 virtual setups: 1 against (2,3), 2 against (1,3), and 3 against (1,2). It is social praxis that determines the transition from virtual to actual coalitions, where social praxis is understood as an agonistic practice which is not, nor could ever be, ruled by formal principles. We could perhaps express von Neumann and Morgenstern’s approach with Marramao’s words by saying that the subjects of strategic (and, a fortiori, economic) interactions are the results of a political process of (possibly) value-based differentiation. Far from replacing politics with the market—as contemporary universalism is wont to do—von Neumann and Morgenstern, in an ideal continuation of a genuinely Machiavellian theory of politics, attempt to develop an instrument that can help us better understand the unavoidable social conflicts that underwrite social and political praxis. An important side effect of their attempt is the reduction of mathematical economic theory to an important but partial component of a much broader political economy that economics relentlessly tries to replace.

The brief sketch of the conflict between neo-Classic economic orthodoxy and game theory covered only the years of its conception and its early formulation in the late 1940s. Its successive development witnessed an almost complete reversal of Morgenstern and von Neumann’s original insights. In the early 1950s, John Nash suggested an alternative concept of solution for strategic games, now called, in his honor, “Nash-equilibrium.” Nash developed his alternative concept on the basis of a very different articulation of elementary strategic interactions. We have seen that von Neumann and Morgenstern modeled all real economic interactions as strategic confrontations between coalitions of players. This approach implies, as I pointed out, that economic struggles go hand in hand with social and economic cooperation: players must cooperate in order to form coalitions that will later compete according to the formal rules of the theory.

John Nash’s approach is exactly the opposite. His fundamental distinction is between cooperative and non-cooperative games and posits the first class as more fundamental than the second one. As he said in his original PhD dissertation, later published in the Annals of Mathematics, “our theory is based on the absence of coalitions in that it is assumed that each participant acts independently, without collaboration or communication with any of the others” (1951, 286). Nash proceeded to provide a concept of solution for non-cooperative games—the previously mentioned “Nash-equilibrium”—according to which a game is in equilibrium when no player has anything to gain by switching to an alternative strategy. In other words, a non-cooperative game is in equilibrium when all players are simultaneously choosing their optimal strategies against each other. Finally Nash proved that every finite non-cooperative game has at least one equilibrium. Conceptually speaking, the most important part of Nash’s original contribution is contained in the last two paragraphs of his short 1950 paper. Nash switches back to cooperative games, in the sense of von Neumann and Morgenstern, and declares explicitly, without providing many details, that he has developed a “‘dynamical’ approach to the study of cooperative games based upon reduction to non-cooperative form” (1951, 195; my emphasis). These few lines, minimally expanded in a later article (1953), were later dubbed the “Nash program” in game theory: the attempt, that is, to reframe the whole theory upon the concept of existing equilibria in non-cooperative games. The Nash program was pursued tentatively in the 1950s and gained increasing momentum in the following decades. By the 1980s it was effectively completed and Nash’s concept equilibrium in cooperative games had become the cornerstone of the whole theoretical apparatus, to the point that even the original results by von Neumann, which precede Nash’s theory by more than 20 years, are routinely presented in terms of Nash-equilibria.[4]

This is not the place to dwell on the technical merits and demerits of von Neumann’s versus Nash’s concepts of solution.[5] What is apparent from even these cursory notes, though, is that the Nash program is a complete reorientation of game theory that rejoins it to the neo-Classic universalist view. As Robert Aumann stated in a 1985 paper published when the Nash program was effectively completed (in no small part thanks to his efforts), the “Nash equilibrium is the embodiment of the idea that economic agents are rational; that they simultaneously act to maximize their utility. If there is any idea that can be considered to be the driving force of economic theory, this is it. Thus in a sense, Nash equilibrium embodies the most important and fundamental idea of economics, that people act in accordance with their incentives” (2000, 19). Aumann explicitly reconnects game theory to the Walrasian ideal against which it had been developed.[6]

However, the ever increasing technical sophistication of “post-Nash program” game-theoretic results and its related production of more and more abstract concepts of “solution,” joined to the emergence of conceptual paradoxes of various kinds, has turned several early enthusiasts into stern critics of its overall ambitions. I am mainly referring to the paradoxes of “common knowledge,” which show that, according to game theory, all the participants in strategic interactions must assume (a) the knowledge that every one is and will behave rationally (i.e. as a utility maximizer), (b) a mutual knowledge of the results of one own’s and others’ actions, and (c) that each player has a specific preference (Lewis, 1969). Technical results by Robert Aumann (1976) showed that these conditions do not generally entail an infinite regress, as it would appear at first sight. Nonetheless, they have paradoxical consequence. For instance, they exclude a priori the very possibility of market speculation: it follows from Lewis’s original intuition and from Aumann’s results that rational agents would never speculate against each other. Given the obvious counterfactual falsity of this claim, it follows that either game theory would have to admit the existence of limited forms of rationality that are, from the point of view of neo-Classic theory, just examples of irrational behavior.[7] From a general conceptual standpoint, these paradoxes constitute almost a reductio ad absurdum of neo-Classic rationality, since they show that a theory built upon the axiomatic assumption of perfect rationality (which is embedded, as we saw, in the concept of Nash-equilibrium) leads the theory to reject the existence of a perfectly rational agent. The theory of rational strategic interactions concludes that some of the best examples of economic behavior (such as the behavior of economic agents in the market) are not rational at all and should not exist.

Jon Elster is perhaps the best example of a game theory enthusiast turned stern critic. Elster was convinced that a convincing and updated reformulation of Marxism—the foremost example of “political economy,” perhaps—had to be based on an individualistic micro-economic foundation. Rational choice theory and game theory were the two essential components of such a refoundation that would assure “solid microfoundations for any study of social structure and social change” (1982, 477). Thirty years later, Elster’s judgment of game theory could not be more different. The formal models that the theory of rational choice and game theory have produced, he stated in 2009, could be defended, in principle, on the basis of their predictive, explicative, or normative value; or because of their aesthetic or mathematical value. My thesis, he concludes, “is that much work in economics and political science is deprived of any empirical, aesthetic, or mathematical interest and therefore has no value whatsoever.”[8]

I think Elster’s conclusion is perfectly legitimate only if we limit its scope to the status game theory reached after the successful implementation of the Nash program. On the contrary, a return to the original, non neo-Classic foundation of game theory would prompt a recentering of the economic subject. It would move it away from the undifferentiated utility maximizer inhabiting neo-Classic universalism. Such a recentering could, in turn, be seen as a response to Marramao’s call to action that was rightfully prompted by the pernicious feedback loop of glocalism. The rejection of universalism from the very core of economics and its welcome reabsorption into a broader political economy would allow the “irrational” symbolic aggregation of differentiated communities to enter strategic competitions confined to limited forms of rationality. It would open the field of economics—long the bastion of the most unforgiving forms of universalism with its consequent reduction of political action to market dynamics—to the play of differences. The rediscovery of the forgotten origins of universalist orthodoxy might even constitute a possible path toward “the passage West” that Marramao envisions.

Notes

    1. For game theory, a “game” is simply a sequence of interactions between two or more agents defined by strict rules. A game always has a starting state and a well- defined series of end states (outcomes) with specific payoffs to the players. Parlor games such as chess, poker, etc. are examples of games that can be analyzed with game-theoretic tools, even though the theory considers many more classes of games.return to text

    2. See Binmore (1994-2005) for an ambitious extension of game theory to ethics and social justice. The extension of game theory to biology (evolutionary game theory) became a widely pursued research field after the publication of Maynard- Smith (1982) and Axelrod (1984). See Leonard (2010) and Mirowski (2002) for a history of the formative years of game theory. Aumann and Hart (1992-2002) is a comprehensive, multi-volume systematic account, while Binmore (2007) is a manageable introduction. Some of the elements of game theory’s history I am sketching below are discussed in more detail in Franchi (2013).return to text

    3. Technically speaking, von Neumann and Morgenstern model a non zero-sum N- player game (which is the closest approximation of real economic interactions) as a 2-player zero-sum game between N+1 players aggregated in 2 coalitions. See von Neumann and Morgenstern (1944, 220 ff.) for the classic exposition and Lucas (1992) for a recent well-informed discussion; Franchi (2013) gives an informal presentation of the classic approach (which is now called the “Stable set solution”).return to text

    4. Von Neumann’s most important technical contribution is the proof of existence of a winning strategy for any 2-player zero-sum game (the so called minimax strategy). It is the main result of von Neumann’s 1928 paper and the foundation of the 1944 work with Morgenstern. Von Neumann’s original proof exploited Brouwer’s fixed-point theorem. Nowadays, the proof is usually given in terms of Nash equilibria by showing that any 2-person zero-sum game has such an equilibrium (see, for instance, Osborne [2002]).return to text

    5. See Lucas (1992) for a balanced discussion of “Stable sets” and Aumann (2000) for a very critical assessment. The concept of Nash equilibrium ran into its own share of technical problems, which generated a plethora of alternative and less and less intuitive equilibrium concepts. See Binmore (2007) for an accessible discussion of the overall consequences for game theory of the proliferation of concepts of equilibrium (or, as they are also called, of “solution concepts”).return to text

    6. The link between Walrasian theory and Robert’s Aumann implementation of the Nash program is stressed by Mirowski (2002). It should be noted that von Neumann was very dismissive of Nash when the latter approached him at Princeton in the early 1950s. In an interview with Robert Leonard (2010, p. 503), Martin Shubik reports that von Neumann “hated Nash’s concept and considered it antithetical to game theory” whereas Nash held that his own conceptions of game theory based on equilibria instead of minimax strategies was “more individualistic and more American”.return to text

    7. See Geanakoplos (1992b) and Geanakoplos (1992a) for an extended discussion of the problem of common knowledge in game theory.return to text

    8. See Elster (2009, 9), the first volume of his lectures at the Collège de France. See also the second volume (2010), which is devoted explicitly to the theme of “irrationality”, and especially to a discussion of the overall economic and political implications of behavioral economics, the discipline that has contributed more than any other to cast profound doubts on the rationality of social agents.return to text

    Works Referenced

    • Aumann, Robert. “Agreeing to Disagree”. In Annals of Statistics. (4) (1976): 1236–1239.
    • ———. “What is Game Theory Trying to Accomplish?” In Collected Papers. Vol. 1. Cambridge, MIT Press, 2000: 5–46.
    • Aumann, Robert and Sergiu Hart. Handbook of Game Theory with Economic Applications. 3 vols. Amsterdam: Elsevier, 1992-2002.
    • Axelrod, Robert. The Evolution of Cooperation. New York: Basic Books, 1984. Binmore, Ken. Game Theory and the Social Contract. Cambridge: MIT Press, 1994-2005.
    • ———. Game Theory, A Very Short Introduction. Oxford: Oxford University Press, 2007.
    • Elster, Jon. “Marxism, Functionalism, and Game Theory. The Case for Methodological Individualism”. In Theory and Society. (11.4) (1982): 453–482.
    • ———. Le désintéressement: Traité critique de l’homme économique. Vol. 1. Paris: Seuil, 2009.
    • ———. L’irrationalité: Traité critique de l’homme économique. Vol. 2. Paris: Seuil, 2010. Franchi, Stefano. “Teoria dei giochi e complessità”. In Decisioni e scelte incontesti complessi. Ed. by Sergio Barile, Valerio Eletti, and Maurizio
    • Matteuzzi. Milano: Wolters Kluwers, 2013: 343–387.
    • Geanakoplos, John. “Common Kowledge”. In Handbook of Game Theory. Vol. 2. Ed. by Robert Aumann and Sergiu Hart. Amsterdam: Elsevier, 1992, 1437–1496.
    • ——— “Common Knowledge”. In The Journal of Economic Perspectives. (6) (1992): 53–82.
    • Leonard, Robert. Von Neumann, Morgenstern, and the Creation of Game Theory. Cambridge: Cambridge University Press, 2010.
    • Lewis, David. Conventions: A Philosophical Study. Cambridge: Harvard University Press, 1969.
    • Lucas, William F. “Von Neumann-Morgenstern Stable Sets”. In Handbook of Game Theory. Vol. 1. Ed. by Robert Aumann and Sergiu Hart. Amsterdam: Elsevier, 1992: 543–590.
    • Maynard-Smith, John. Evolution and the Theory of Games. Cambridge: Cambridge University Press, 1982.
    • Mirowski, Philip. Machine Dreams. Economics Becomes a Cyborg Science. Cambridge: Cambridge University Press, 2002.
    • Morgenstern, Oskar. “Vollkommene Voraussicht und wirtschaftliches
    • Gleichgewicht”. In Zeitschrift für Nationalökonomie. (6.3) (1935): 337–357.
    • Nash, John. “Non-Cooperative Games”. In Annals of Mathematics. (54.2) (1951): 286–295.
    • ———. “Two-Person Cooperative Games”. In Econometrica. (21.1) (1953): 128–140.
    • Osborne, Martin J. An Introduction to Game Theory. Oxford: Oxford University Press, 2002.
    • von Neumann, John. “Zür Theorie der Gesellschaftsspiele”. In Mathematische Annalen. (100) (1928): 295–328.
    • von Neumann, John and Oskar Morgenstern. The Theory of Games and Economic Behavior. Princeton: Princeton University Press, 1944.