It’s a Very Long Way From Frankfurt to Virginia and Rochester: On the hierarchy/market distinction as a false dichotomy
by Bryce Weber (Perth)
for the British Columbia Political Studies Association Conference Simon Fraser University Burnaby, British Columbia May 5-6, 1995
Critical theory has always understood itself as an emancipatory project, and this project has been seen consistently by critical theorists themselves as involving a rejection of instrumental reason and strategic social and political behaviour as social norms. The pairing of critical theory with variants of rational choice theory would seem, then, to be anathema to critical theorists. And the idea that one has by necessity to choose between markets, which are if nothing else, the playgrounds of strategically rational social actors, and "hierarchy"_or arbirtary authority_is a particularly hoarey choice to a critical theorist. What I should like to argue in this paper is that the "choice" between these alternatives is based on a false dichotomy, and that the origins of this false dichotomy can be located in the inability_or unwillingness for whatever reasons_for rational choice theory and its variants, to move beyond a fundamentally methodologically solipsistic position. Rational choice based theories could avoid this dichotomy if they came to terms with the fact that people can, and necessarilly do, develop mutual understandings between and amongst themselves that are based on non-strategic motives, and which make perfectly rational sense without having to appeal to the idea that such strategically non- rational forms of social action are functions of some kind of biizarre_or at the least extravagant_act of altruism to do so. It is precisely this type of problem that Habermas's account of communicative action attempts to address. It argues that there is a reflexive component that functions as the presupposition of everyday language use, in those cases where meaning and/or shared social understanding is generated; and that the use of language is possible only to the degree that social actors are capable of understanding each other in non-strategic terms. Consequently, the resolution to the problem presented by the either/or of the market or tyranny can be found in an alternative suggested by the idea that people are capable of understanding what each other mean and so can discuss their mutual concerns reasonably. My position is that what is involved in these moments of communicative interaction is a moment of mutual recognition in which others are understood not as means to my ends, but as individuals who are the same as me when I'm not merely motivated by self interest. Herein lies the seeds of a notion of the "common good". It is evident that the better of the rational choice theorists have come to recognize the limitations of a simplistic rational choice account of the rationality of social and political action. This recognition has tended to focus on the inability of simple rational choice theory to account for how social institutions can exist at all. As Dryzek points out, these institutions could not exist if the only form of rational action were strategic.2 As I shall argue, James Johnson's3 and RŽjean Landry's4 approaches share the objective of attempting to get beyond the limitations of game theory, which they both depict as itself more adequate than simple rational choice theory. In both cases, this has to do with an increased acknowledgement by game theory, relative to less sophisticated derivatives of rational choice, of the social contexts and shared social understandings within which social action takes place. nevertheless, both Johnson and Landry attempt to avoid accounting for the inadequacies of a game theoretic approach through reference to the latter's refusal to come to terms with reflexive thought or forms of mutual recognition between social actors that go beyond the explanatory capacities of a competitive model of interaction. In maintaining this position, both Johnson and Landry may provide accounts of action with a degree of predictive power, and may provide accounts of how actors who understand themselves as acting in a strategic fashion will act. Since this appears to include most individuals currently involved in policy setting and implementation, this would be no small accomplishment in its own right. However, it also leaves the broader question aside; that is, how to solve the problem to which Dryzek refers as the "reflexive incoherence" of the rational choice based model of explanation. It is this question that lies at the heart of my concerns in this paper, not the adequacy of the models of explanation offered by Johnson and Landry on their own terms. It should be noted at the outset, however, that the problems that Dryzek points out are noted by Buchanan himself, that Johnson attempts to resolve these same problems through his attempt to appropriate the Habermasian notion of communicative interaction to the needs of game theory; and that Landry attempts to integrate elements of contract and transaction cost theories of interaction into an expanded, game theoretic explanatory framework to the same end. All are fundamentally driven forward by the inevitable inadequacy of the starting point of rational choice based theories. To put this in a nutshell, the utility maximizer of rational choice and game theory functions in the same way as the "Unhappy Consciousness" of Hegel's Phenomenology, not wanting out of the immanence of its isolation, in which it takes no little solace, but not being able to explain to itself or to anyone else why it feels compelled to reduce everything it encounters to the abstract formulae it dreams up in order to account for the reality it inevitably confronts as hostile. My thesis is that Johnson and Landry both see the need to move towards the social, but refuse to come to terms with what they ought to know is necessary in order to resolve the reflexive incoherence of the rational choice position: i.e., the abandonment of the model of utility maximization as the founding moment of the social institutions which the market, and similarly competitive forms of social interaction, rely upon for their continued existence. The title of the article by John Dryzek from which the title of this present talk is derived is "How Far is it from Virginia and Rochester to Frankfurt? Public Choice and Critical Theory."5 What Dryzek argues is that both rational choice theory and critical theory can, for different reasons, benefit from a dialogue.6 Dryzek's politics, (viz., his concern with new social movement politics) and his hope for the development of what he refers to as a "discursive democracy", place him more on the critical theory side of this divide. Another author who makes a similar claim, but who is at least methodologically more to the side of rational choice theory is James Johnson.7 In both cases, they claim to want to set a dialogue in motion between critical theory and rational choice theory. The argument they both present is that while critical theory and rational choice theory are not entirely subsumable to each other, their relative merits can be used to supplement the shortcomings of the other position. Now both Johnson and Dryzek point out the same weakness in rational choice-based theories, and that is their inability to account for the rationality of non-strategic action. This manifests itself most strikingly in their inability to account for the rationality of moral or ethical actions. At its extreme, this blind spot makes it difficult to explain how social institutions_which are admitted by the more thoughtful rational choice-based theorists as requiring a degree of cooperation_can operate successfully. What differentiates Dryzek's position on rational choice theory from Johnson's is that, although they both point out this incapacity of rational choice based theories, Dryzek treats the admission of this theoretical shortfall as an opening into which to inject critical theory content, whereas Johnson views the admission of this weakness as a merit which can be used to point in the direction in which rational choice theory needs to move. Both agree, however, that rational choice theory suffers from the problem described above. What Johnson recommends as a solution to this problem is a move towards game theory. This argument is similar to that made by RŽjean Landry, who argues that a third level of complexity can be added to the rational choice based paradigm; the recognition of the institutional setting within which social action takes place, understood as a set of largely indeterminant contractual arrangements between social and/or political actors. What I would like to argue in this paper, though, is that the move to game theory, and to the kind of theory of institutions which Landry8 does not provide a solution to the real problem faced by rational choice theory; that is, its inability to move beyond the position of a methodologically isolated individual. Or, to put this another way, of the inability of rational choice-based theories to come to terms with the fact that competitive social action_which it takes as its model of rational action_presupposes at both a logical and an empirical level, cooperative social interaction. Johnson's and Landry's acknowledgement that rational choice based theories ought to move towards a recognition of other social actors (as a function of seeking to be successful on self-interested terms, one might add), or towards the recognition of the institutionally located nature of social interaction are surely advances over the original, admittedly simplistic account of social action provided by unalloyed rational choice theory. However, this admission rather points once again to the weaknesses of rational choice based theories, since their accounts of social interaction and institutionally established paradigms of social interaction provide accounts of the latter which replicate at an epistemological level the "reflexive incoherence" which Dryzek points out, and which characterizes the admitted weaknesses of rational choice based theories as a whole. The reason that it is fruitful to follow out this sort of development_and Landry's article points this out quite well_is that what is called for to correct the limitations encountered rational choice theory is a move to the social which would violate the integrity of the rational choice paradigm itself. Without this, there can be no solution to the rational choice "problem." The mere acknowledgement of the social and of other social beings does not resolve the real problem of atomistic theory, for it is the accounts of what others' existences mean to the one who acknowledges them, rather than the mere acknowledgement of their existences, that is where the proof of the pudding is to be found. What Johnson and Landry both propose to resolve the problem of rational choice theory is a move to increasingly social levels of the explanation of the rationality of action. My argument is that this goal will necessarily proceed asymptotically: It will never get to the point it needs, since in both its strategic and games theory manifestations it continues to refuse to move beyond the fundamental limitations of its position; and it is these fundamental limits which are source of its problems to begin with. Society cannot be explained exhaustively in terms of strategic action_even when this action is contextualized_if this process of contextualization refuses to acknowledge that its possibility is predicated on the very thing which the logic of the rational choice position excludes from consideration as reasonable. In sum, rational choice theories, even in their most sophisticated forms, are flawed at a dialectical level in that they suppress the preconditions for their own sense in order to make sense of themselves.9 Both Johnson's and Landry's approaches are clever attempts to resolve the problems of rational choice theory without ever having to abandon the perspective which leads to the problem in the first place. In other words, their respective approaches are clever and strategic_the perfect scenario for a rational choice position_and so their attempts carry within themselves a kind of trope of the problem of rational choice-based theories as a whole.10 To even a casual observer the attempt by Dryzek and Johnson to develop a dialogue between rational choice and critical theory is somewhat strange. An earlier generation of critical theorists played the same role in Germany as the so-called "normativists" played in the similar North American debate of the 1960's, arguing against the positivist-derived positions of empirical social science in the Positivismusstreit in Germany.11 Rational choice theory may not be positivist in the strict sense of the term, but rational choice theory carries forward in the spirit of the positivist tradition, which is characterized by an aversion to reflexive thought, if Habermas is correct.12 On the other hand, Habermasian critical theory declares itself to be bent on a search for the social basis of the normative foundations of the original critical theory project.13 A conversation between rational choice and critical theory would thus appear to be a suspect proposition. Dryzek argues, however, that it can be worthwhile for both sides, inasmuch as critical theory has tended towards the abstract_an irony for a theoretical position which has traditionally prided itself for the acknowledgement of its inextricably socially located nature and as an applied form of thinking. Public choice theory, if nothing else, is applied. It asks, "Why do people make the choices what they do?" Yet as Dryzek argues, public choice theory suffers a shortfall in its approach which might be supplemented by critical theory in its present Habermasian form. Dryzek argues that the impasse suffered by rational choice theory manifests itself in a number of different ways. He says that at a conceptual level, the problem of rational choice-based theories manifests itself in ...an implicit and contradictory determinism of axioms for individual behaviour. Empirically, it is an inability to explain high baseline rates of co- operation and other instrumentally irrational behaviour in prisoners' dilemmas and other social situations (such as voting). Normatively, the problem is reflexive incoherence stemming from internal contradiction: normative public choice only makes sense so long as its own model of man is violated. Politically, the problem is hostility to democracy...14 Dryzek goes on to explain that for him, public choice theory is characterized by its assumption of a homo economicus as the paradigm social actor, even if this entails a moral theory, such as that of Rawls, whereas by critical theory he means the school of thought first developed by Horkheimer, Adorno and Marcuse, which focused on "instrumental rationality" as the source of domination in modern society and from which, in its present form, we needs must be emancipated. Habermas, he notes, attempts to move beyond what he sees as the limitations of his predecessors' position. Habermas has always maintained that instrumental rationality has its place in social life. The social pathologies of contemporary society, from Habermas's position, arise when instrumental reason comes to dominate more of social life than is appropriate. Some role is thus left by Habermas for "instrumental rationality" in social affairs, and this form of means/ends rationality is paradigmatic for public choice theory. What sets Habermas apart from rational choice-based theories, according to Dryzek, however, is his idea that rationality "means the competence to decide when it is appropriate to act instrumentally, or in conformity with social norms, or dramaturgically, as an expressive subject; and that the ability to judge these qualities in others. this communicative competence is, then, a matter of intersubjectivity, not just isolated subjectivity."15 Dryzek points out that the kinds of problems for democracy which come from the unconstrained pursuit of self-interest, as this is pointed out by the rational choice-based schools of public choice (Buchanan and Virginia Tech) and social choice (Riker and the University of Rochester) come down to a refusal to accept a cognitivist account of the possibility of actually discussing policy alternatives. From Dryzek's point of view, then, it seems that social and public choice theories place themselves in a dilemma: they understand democratic politics to be driven by self-interest, and see this as insurmountable; but they argue that this form of social action is inevitably destructive of the social fabric.16 Dryzek argues that the solution to this problem would be to make room for a cognitivist approach to problem solving in which individuals talk over their problems in a relatively non-strategic fashion. The non-cognitivism of public and social choice theory forecloses this possibility. From Dryzek's perspective, the cul-de-sac that rational choice gets itself into is avoidable if a communicative, rather than an instrumental or strategic, rationality is employed as the basis for the explanation of social action. Over and above the highly complex accounts of communicative rationality Habermas provides, the simplest way of explaining what "communicative action" means is to say that a conversation is "communicatively rational" if the parties to this discourse discuss things in order to understand the other person's position, independent of whether they agree with it or not. Now, this could never take place if the individuals involved in this type of discussion were speaking to each other with an instrumental intent, since in this situation the only point of listening to someone else is to find out enough about them to take advantage of the situation. Communicative rationality does not by any means guarantee agreement, but it does demand a willingness to hear out one's interlocutors. Public discussion informed by communicative rationality makes possible, as Dryzek says, the discussion of "generalizable interests" or "public goods".17 As Dryzek points out, the basis for the rational choice position is the prisoners' dilemma, and the scenarios appealed to by rational choice theorists posit non-cognitive accounts of this situation: the prisoners are simply not allowed to speak with each other, and it is assumed that in competitive social relations, they would not speak with each other either, at least not unless this were also part of the process of angling for advantage. Dryzek points out, however, that in social psychology experiments in which individuals were allowed to talk with each other, conflict over goal achievement by those involved was substantially reduced. Dryzek adds that the reason for these developments, from a critical theory perspective, "...is clear: a period of discussion enables the group's interactions to be governed by communicative rather than instrumental rationality."18 Applied to an institutional setting, Dryzek argues that this position can help rational choice out of its cul-de-sac. From his perspective, it is institutions which promote instrumental rationality and the unalloyed pursuit of private advantage which lead, as rational choice theorists themselves acknowledge, to social decadence. Dryzek's solution to this problem is to introduce new institutions which would facilitate communicatively rational forms of interaction; what Dryzek elsewhere works out as a theory of "discursive democracy", which places his work in the same ballpark as that of Cohen and Arato, for instance. Interestingly, Dryzek argues that hierarchies arise within the context of the attempted manipulation of subordinates, but this_i.e., hierarchy_is something that seems inevitable on the terms provided by rational choice- based theories, at least as these are described by Dryzek. He argues that the evidence of this can be found in the corrosion of the public's willingness at present to conceive of ways in which government might not be corrupt. He thinks that the concept, and practice, of communicative rationality offers an alternative to the instrumental self-interest taken as paradigmatic by rational choice. In short, the "reflexive incoherence of rational choice theory, as this is depicted by Dryzek, is similar to the Cretan paradox: "If a Cretan tells you all Cretans are liars, is he telling the truth?" If the only form of social action that is rational is self-interested action, then this claim can be comprehended rationally only to the degree that it is an attempt to manipulate those who hear it, so need not be believed. This is the "reflexive incoherence" of the rational choice position. Dryzek's position, in the end, is that the public will continue to have to make choices, and so some form of "public choice" theory will continue to be relevant. But if people make their choices about shared, public concerns on the basis of communicatively rational discussion, then the damage that public choice theory sees as unavoidable can be obviated. What public choice demonstrates, from Dryzek's point of view, is the "impossibility of political order founded on unconstrained rational egoism," even though it refuses to abandon this model of explaining human action. Johnson's take on the problem of attempting to maintain political order on the basis of unconstrained rational egoism is somewhat different from that of Dryzek. Dryzek's position is that public and social choice theory has built itself into a dilemma out of which it refuses to move, and that Habermasian critical theory offers a way out of this situation. Johnson's approach is two fold, in that he argues that rational choice-based theories are already building themselves out of the kinds of problems pointed out above through developments in game theory; and that unlike the Habermasian perspective, which deals with competitive, strategic action as something that is "parasitic" on prior, successful communicative action,19 it is communicative action which actually presupposes strategic action. The first of these points moves us towards the subsequent discussion of Landry's work; the second is intended by Johnson to refute Habermas's position by demonstrating that Habermas has an inadequately nuanced appreciation for the nature of strategic action. In effect, this sets up the ground work for arguing the case that rational choice-based theories have the capacity to work out, on their own terms, the problem of "reflexive incoherence" of social and public choice theory to which Dryzek points, which Buchanan himself acknowledges,20 and which Johnson accepts as a given to be worked with and beyond. I do not what to suggest in what follows that all of Johnson's criticisms of Habermas are without merit. A great many of them are, but they are numerous and many would require a more detailed treatment than what I shall attempt here. Like Dryzek, Johnson points out that rational choice theory needs to pay attention to the critical theory position since, as Johnson argues, critical theory points out the problem that Dryzek shows Buchanan locates which rational choice-based theories cannot answer; that is, ...the question of how and what basis social actors define the context within which their interactions transpire_for which rational choice theorists currently have no satisfactory solution...while Habermas's discussion of the problem is perhaps not persuasive, the problem [for rational choice theory] remains.21 Johnson takes up this "intuition" of critical theorists in order to prod rational choice theorists into a more critical understanding of their own position's key weakness, and he argues that he hopes to, "...show in a provisional way how game theory extends and deepens the critical theorists basic intuition that unembellished strategic rationality cannot adequately sustain social and political interaction.22 What game theorists attempt to do, according to Johnson, is to incorporate an account of communication into their own models of rational action.23 From Johnson's perspective, however, the lacuna in the game theoretic which the examination of communication is supposed to plug is created through the "widespread indeterminacy in the for of multiple equilibria and attendant coordination problems."24 The source of the problems Johnson examines, according to his account, do not come from an excess of competition, but rather are the result of excessive complication and a lack of comprehensive information. Johnson's examination of Habermas's approach to communicative rationality, and communicative action, draws into focus a number of ambiguities in Habermas's explanation of this which are well taken. But the pivotal difference between them lies in their respective understandings of the nature of strategic action, and the relation of strategic action to communicative action. According to Johnson, Habermas's account of strategic action is excessively narrow. He argues that Habermas's account of strategic action is more applicable to the analysis of "parametric action", that is, action in which social actors deal with others and the world around them as if they were actually dealing with a given, predefined "natural" world. This parametrically rational form of action would be appropriately analysed, according to Johnson, as the product of purely atomistic and irretrievably egoistic social actors. From Johnson's perspective, however, strategic rationality necessarily involves the recognition of other social beings, as social beings, and so strategic, rather than merely instrumental action is the focus of game theory. According to Johnson, game theorists presume that, ...social actors can (1)Êunderstand that their environment partly consist of other intentional agents, and (2) recognize those others as equally rational. these generalized capacities enable social actors to proceed strategically by attempting to accommodate the anticipated actions of relevant others when formulation their own plans.25 From Johnson's perspective, this means that game theory involves a recognition of other social selves, which might provide the grounds for a refutation of Habermas's position. But to this Johnson adds the argument that Habermas's understanding of the relationship of strategically to communicatively rational action is incorrect. Habermas's position is that communicative action is directed towards the development of understanding between social actors, whereas strategic action is teleologically oriented to success (and dramaturgically and normatively rational actions are oriented to social acceptance through norm conformity or to the recognition of one's needs).26 What differentiates communicative action, from Habermas's perspective, is that it is not teleological in the sense of the other three types of action, since its successful completion is not exhausted in achieving its goal, as he puts this in Theory of Communicative Action. Rather, he argues, only communicative rationality_or the communicatively rational component of any of the other three types of action_is capable of generating mutual understanding. And he argues further that the prior establishment of at least a vestigial degree of mutual understanding is necessary for the successful completion of any of the other types of action in question, to the degree that they are understood to take place within a social context. Habermas's critique of strategic accounts of the rationality of social action, then, is that they presuppose the prior existence (both logically, at the level of explanation and practically, at the level of social institutions and everyday social action) of successful communicative interaction. By way of example, the market, per se, is incapable of generating meaning, but consumerism requires meaning in order to successfully "market" products. If everything is reduced to the level of an exchange value, then nothing has meaning anymore and their will be no motivation on the part of consumers to purchase those goods which are marketed as providing meaning. Ironically, then, an ultimate "victory" of the market in its drive to render everything rationalizable as an exchangeable commodity according to the price system, would lead to its demise. There is a need, under these institutional conditions, to provide a space for the development of "meaning" independent of market rationality, and it is precisely this argument which "arts industry" representatives make in order to exempt certain forms of aesthetic and cultural production from market rationalization.27 From Habermas's perspective, strategic social action acts in a "parasitic" fashion because it exploits, but cannot replicate, previously successful communicatively established social understandings, and explanations of social action which rely on strategic models of the latter suffer the same inadequacy, according to his account. In a different treatment of this in TCA II, Habermas argues that "perlocutionary", or for him manipulative speech acts, presuppose a prior sense of trust and/or agreement which can only be established by non- perlocutionary speech acts. Or, to put this in less terminologically dense prose, someone has to trust you before you can trick them. Legitimation crises which derive from a shortfall of meaning in citizens' lives, such as the no-going one we have been experiencing_with its particularly acute symptoms which have manifested themselves in the United States recently_provide another example of the parasitic nature of manipulative political strategies which exploit and exhaust sources of meaning which neither the state nor the market can generate solely by themselves. Now, against this position of Habermas's which sees strategic action as parasitic, Johnson argues that Habermas himself "shows how the very capacity to engage in normatively governed action that ... practical discourse [capable of coordinating social action in a voluntary and unmanipulated fashion] presupposes can emerge only in intimate relation to the development of strategic competence."28 If Johnson can get away with this argument, what he will do is to subordinate communicative action to strategic action, and hence turn the tables on Habermas. In other words, if the "reflexive incoherence" of rational choice theory can be resolved by means of the move to communication game theory offers, but communicative action is then argued to be predicated upon strategic action, not the other way around, critical theory might provide a supplement to game theory, but it will not be in a position to criticize it as parasitic. Of course, Johnson's gambit will work here only to the degree that strategic rationality_which makes sustained political action within a relatively stable institutional settings impossible over the long term_turns out to be the source of communicatively rational interaction. And thus, strategic action will have to be made to appear to be more than what it really is in order to accomplish this. In other words, Johnson's approach will merely displace the rational choice problem into another setting, even if it is successful. The proof that Johnson offers to demonstrate his case is an argument by Habermas in which Habermas argues that the development of the capacity for strategic rationality is one stage of many, (and a relatively early one) in the development of moral consciousness. From Habermas's perspective, the development of this capacity is part of a learning process and is an advance over the child (or adult!) who acts as if the world were made up of objects, or of objectively given rules. But, from Habermas's perspective, the mastery of strategic competence _i.e., the ability to recognize the needs of others in order to manipulate them to get what one wants oneself_soon gives up to a higher level of moral reasoning. Besides, for Habermas, the mastery of levels of moral reasoning that represent stages of learning above and beyond the merely strategic provide the basis for criticizing strategically rational actions from a moral point of view. From Johnson's perspective, however, the need to develop a competence as a strategic actor as a stage in the development of moral consciousness is taken as, "crucial to the development of full moral consciousness. This provides critical theorist with a forceful reason to take seriously game-theoretic accounts of how strategic competence manifests itself in performance."29 Johnson seems to think that this allows him to refute the critique of strategic models of rationality as derivative of communicative accomplishments. Of course, the gambit here is to argue that communicative action is derivative, and hence in a sense "parasitic" upon instrumentally rational action, which would turn the tables on Habermas, but this attempt fails because it mistakes the temporal precedence of strategic action in the process of learning with the logical precedence of communicative action over strategically rational action.30 Johnson also argues that Habermas's inadequate account of strategically rational action leads him into the problem of not seeing that in order to be truly strategic, social actors must "presuppose a broader structure of communication in order to coordinate social and political action successfully."31 Of course, the point here is not that strategically rational actors recognize the communicative component of everyday life. The issue is whether they recognize this in order to be "successful" in instrumental terms, rather than in communicative terms. Johnson's approach attempts to bury this question rather than confront it. What Johnson then argues is that game theory recognizes that the limits to strategic action accounts of the rationality of action. The proof of this, he says, lies in game theory's own recognition of the high degrees of indeterminacy involved in social action, and then argues that it is this high degree of indeterminacy which defines the limits of the explanatory powers of strategic-rational accounts of social interaction. Indeterminacy, not a reflexive incoherence is used to account for the problems encountered by game theory. Johnson then introduces the notion of "cheap talk"_that is, open discussion with no costs to actors_as a possible way of accounting for the extra- strategic element of social and political coordination which he believes is what Habermas must really be attempting to account for by means of his theory of communicative action, according to Johnson. What game theory cannot account for, but then again Johnson argues neither can Habermas, is why "cheap talk"/communicative interaction has any binding force on its participants. In this much, Johnson argues that the critical theory approach does extend beyond that of game theory, for in acknowledging the communicative competence which Habermas argues makes communicative understandings possible, game theory must allow that an element of creativity is involved in speech, and this introduces a degree of indeterminacy into their new-found accounts of the non- strategic elements of social action that ruptures the desire of game theory models to provide comprehensive accounts of social action. Beyond this, as long as game theorists continue to explain the motivation for action as based on a model of incentives, and "cheap talk" offers none that is rationalizable on strategic terms, the idea that what someone else says is "credible" to the degree that it reveals the incentive for a speaker to have made this claim, the game theory position is further weakened. Johnson concludes by arguing that Habermas's claim that communicative competence involves the ability to defend one's arguments on the basis of the giving of good reasons, and to weigh the reasons others give for their opinions more or less fairly on these same terms, provides him with something that game theorists lack: i.e., "...a mechanism that might compellingly account for the binding force of language..."_so far so good_but Johnson them adds, "... the binding force of language in strategic interaction."32 Now Habermas's theory is intended precisely not to do this. For Habermas, communicatively developed mutual understandings have a binding effect precisely to the degree that they retain an element of non-strategic rationality. Johnson has attempted to employ Habermas's account of communicative action as a substitute for the "cheap talk" formulae of game theorists, but he has attempted to do so in a way that domesticates the critical capacity of communicative action. It is precisely the aspect of communicative interaction which can bind interlocutors which Johnson abandons in his subordination of communicative action to strategic action. When Johnson gets rid of the critical edge of communicative action, he gets rid of what he needs to save game theory from itself. What is instructive about Johnson's approach, however, is precisely its recognition of the methodological inadequacies of instrumental accounts of the rationality of social and political action, along with its inability to account for the existence of social institutions which admittedly make strategic action possible, but which themselves would be neither institutionally possible, nor coherently explicable, if they were expected to operate entirely on a strategic basis themselves. To reiterate Dryzek's expression here, at an epistemological level, this inadequacy of strategic action-based accounts of the rationality of action manifests itself in a "reflexive incoherence". In other words, the problem with such theories of the rationality of action is that if they are true, and all action is rational to the degree that it can be explained strategically, then this explanation of the rationality of action must itself be a strategic ploy, in which case, such a claim need not be taken seriously as anything other than an attempt to get something extrinsic to this claim itself. Johnson's approach, on one hand, attempts to subordinate the thing that is marginalized/negated by the application of this definition to social action (i.e., communicative interaction) by reducing the latter to a variety of cheap talk that somehow has a binding quality, but it can have a binding quality, in reality, only if it is both rational and qualitatively different (and hence irreducible) to instrumental rationality and its more complex and sophisticated variants, such a game theory. What game theory does, and what marks it as an advance on one level, particularly in Johnson's eyes, however, is that it acknowledges this shortcoming. Game theory recognizes others and recognizes social institutions, unlike parametric accounts of action. In fact, it makes sense for game theory to do so, for if one is truly going to get ones way and be capable of manipulating others to one's own advantage (however one conceives of this), then this level of success will involve flushing out other individuals personal motivations and figuring out what can be taken for granted as the accepted common sense in different social contexts: One learns to "play the person" rather than the game as such. In as much as game theory acknowledges the existence of others and social institutions, it marks and advance over the atomistic methodologically solipsistic fashion of more simple models of instrumental action. However, it continues to operate within an account of rationality which leaves the preconditions for its own sense unexplicated. What would make the preconditions more transparent? In short, what would be needed would be an intersubjective move. This is precisely what Habermas's theory of communicative action is intended to force its readers to undertake. The criteria of comprehensiveness, especially, and of credibility which are much prized by game theory, according to Johnson, would obviously take a beating in this process. But the point to be made here, is that the claims to comprehensiveness and "credibility" are attractive precisely because they work within an identitarian model of knowledge which continues to retain the formal structure_if not precise formulation_of what is to be accepted as valid evidence as did the old positivist model of the "correspondence" theory of truth. The "indeterminacy" to which Johnson points is a problem only so long as one thinks that indeterminacy needs must be reduced to zero if possible. But this, in turn, is tied to the notion of a set of alternatives that is defined by the polar opposites of the "market" on one hand, and "hierarchy", on the other. I shall not deal with the "market" here except to point out that it is conceived of as operating without ever involving communicative interaction, at least not at an immanent level. The idea of "hierarchy", on the other hand, is something that seems inevitable if the anonymous, non-communicative forms of interaction allowed (apparently) by the market are eliminated. Communicative action, which might be conceived of as being able to provide a binding force on its participants, offers an alternative to the anonymous actions of the market, but this aspect of communicative action cannot be made sense of at all if it is conceived of as a component of strategic action, as Johnson wants to argue. In the absence of a market which would anonymously coordinate action, and in the absence of communicative social interaction by means of which people could be convinced to take particular courses of action because of the giving of good reasons to do so, any form of coordinating social action other than the market would necessarily be seen as coercive. Thus, the hierarchy/market dichotomy arises. What alternative to "hierarchy" and the "market" can be offered? Clearly, from my perspective, if a communicative action approach is adopted, this dichotomy would turn out to be a largely false one, since alternatives to the market appear to be hierarchical only to those who operate within a methodologically solipsistic model of understanding, and to the degree that it accomplishes what it sets out to do, the move to communicative action accomplishes a move beyond methodological solipsism. However, it can be thought, as Johnson does, that game theory might provide a way out of this dichotomy, and RŽjean Landry's approach appears to bear this out as well. Landry is interested in rational choice-based explanations of policy making and implementation. He classifies rational choice models into "rational actor", "independent rational actor", and "institutional rational choice" models of explanation. The first is essentially the same as Johnson's notion of "parametric" instrumental action. It deals with individuals as isolated utility maximizers. In the second, a market or similar social setting is assumed. This is the point at which game theory begins to develop and the "prisoner's dilemma" model of interaction gains its relevance. But as Landry argues, The interdependent rational actor model provides a good theoretical tool in helping to think about complex strategic interaction that occur in most policy issues. To that extent, it offers a better tool than the individual rational actor mode. However, both type of models embody unsatisfactory assumptions regarding [sic.] institutional context of policy-making...In the game theory perspective, the institutional arrangements defining the decision making situations are considered as exogenous factors, i.e., they are given and stable. The purpose of institutional rational choice models is to tackle this problem of underspecification [i.e., of the lack of a concrete and realistic treatment of actual institutional contexts by rational choice and game theory] by incorporation contributions from transaction cost theory and contract theory.33 Johnson and Landry might well differ on the nature of the inadequacy of game theory, but both recognize, like Dryzek, that it cannot provide an adequate account of social institutions. What they also both recognize is that game theory needs to move beyond the restrictions of its present paradigm of explanation. Both also see the key to this in the placing of social agents in social contexts, and both claim that a supplement of some sort is needed to correct the problems they detect in the game-theoretic approach. Whereas Johnson thinks that this might be done via a modified communicative action theory reading of "cheap talk", hence taking care of problems of indeterminacy through a discourse analysis, Landry wants to explore the possibility of providing a more realistic account of social action via a more complex account the nature of the indeterminacies which arise in policy making contexts and for which game theory, in its present form, cannot account. The "contractual" variant of institutional rational choice analysis Landry examines argues that "the action of rational actors [can be] rendered compatible through contractual arrangements," wherein society comes to be seen as a nexus of contractual arrangements and the adequacy or inadequacy of various types of contractual arrangements can be seen as providing the opportunity for free-riding.34 From this perspective, government is needed to intervene in cases where private contractual arrangements fail.35 Landry's approach can provide a way of explaining when governmental intervention might best be undertaken or not and therefore promises to be a useful and potentially fascinating analytical tool. However, it does not solve the real problem game theory has in relation to institutions, and that is how social institutions can exist which are not part of the private contracting process while still not moving beyond the micro, utility maximizing account of the rationality of social action. The same can be said for the second of Landry's attempts to deal with the institutional shortfall of rational choice theory; in this case, "transaction theory." The key to this model is the idea that contractual arrangements are best explained not as means by which individuals reduce the costs of transactions, but rather as means by which individuals maximize their personal gains when they do transact with one another. What is really at stake in what Landry is doing, is the attempt to get beyond the zero-sum proclivities of more simplistic rational choice-based accounts of interaction. However, problems still arise, and are recognized as arising, by Landry, since contracts can never be complete.36 In other words, like Johnson, Landry recognizes that the acknowledgement of the social nature of interaction introduces a relatively high degree of "indeterminism" and a lack of "comprehensibility" into explanation. But unlike Johnson, who wants to account for how the social institutions necessary to deal with these levels of indeterminacy without violating the basic premises of the rational choice paradigm, Landry appears to take for granted the existence of the institutions of government which perform these functions. What the questions Johnson and Landry raise share in common, despite this not inconsiderable difference, is that in both cases they see the move from the isolated rational actor model of explanation to that of a model of explanation which places the social actor in a broader social context, as desirable and as marking an improvement. In both cases, the greater the degree of acknowledgement of the presence of other social actors and of shared social understandings and institutions, the more adequate the explanation of social and action can be said to be. But as Dryzek points out, none of this solves the "reflexive incoherence" of these models of explanation, and one has to ask just why it is that a recognition of the rationality of cooperative interaction is not allowed. Now, of course, since Hume, we have had the idea of "enlightened self- interest" to help us out of this problem: From this perspective, I help you now because I assume that you will help me later, and this will be of benefit to us both. But this is not cooperative interaction. This is merely clever and relatively far-sighted strategic interaction. Cooperative interaction is more akin to what Habermas describes as communicative interaction; or, at least, one could argue that communicative interaction is the prerequisite for subsequent cooperation. The ethic of cooperation, at its best, is one in which self-regard is not lacking, but in which competition is not the incentive for the coordination of one's action with that of someone else. The degree to which theorists of the rational choice school will go in order to generate an account of cooperation that does not involve the recognition of the other in the more full, hermeneutic sense I have in mind37 is the work of Robert Axelrod, who virtually invents cooperation out of innumerably iterated prisoner's dilemmas. What this accomplishes, in effect, is the conjuring up of an account of cooperative interaction on the basis of socially imbedded habit in which there is no need for either reflexive thought, nor for the recognition of the other.38 It is, then, a fair bit further between Frankfurt and Virginia or Rochester than Johnson, in particular, would like to think. In his case, his "Frankfurt" is a little bit like Disneyworld: Europe for the Americans, on American terms, and in Florida. The simulacrum of the non-existent original, then, is the French Eurodisney imitation of the American simulation of a Europe that occurs only in America; and Johnson's attempt to appropriate the notion of communicative action is apparently no more successful, when it comes to the "bottom line", than the above mentioned theme park has been. What policy application can this paper's argument be seen to have? If the Frankfurt School was right, and emancipation can only come about with the diminution of instrumental rationality_or as in Habermas's case, with the employment of instrumental rationality that is restricted to those spheres of life to which it is most appropriate_but the market is viable only to the degree that social interaction is rationalizable on instrumental terms, then the need to sustain the viability of rational choice based accounts of the rationality of social action is necessary ideologically, in order to sustain the appearance of the suzerainty/hegemony of the market and market rationality. To allow that the market itself is predicated upon prior social cooperation39, would be to allow concomitantly that the priorities of the market ought best not be the priorities that guide the fundamental concerns of the social policy agenda. As to the personal motivations which make the suppression of this acknowledgement attractive, I leave this question to the side.
1 I would like to thank Laurent Doubuzinskis for numerous discussions we have had concerning rational choice
and game theory approaches, and for his generosity in alerting me to relevant literature in this field. Clearly, the
opinions expressed here are my own.
2 John Dryzek, “How Far is it from Virginia to Rochester to Frankfurt? Public Choice as Critical Theory,” British
Journal of Political Science, volume 22, November 1992, pp.397-417; cf., John Dryzek Discursive Democracy
(Cambridge: Cambridge University Press, 1990).
3 See: James Johnson, “Habermas on Strategic and Communicative Action,” Political Theory, volume 19,
number 2, May, 1991, pp.181-201; James Johnson, “Is Talk Really Cheap?” American Political Science Review,
volume 87, number 1, March 1993, pp.74-86. My thanks to David Laycock for having brought this article to my
4 See: RŽjean Landry, “Rational Choice and Canadian Political Studies,” typescript provided to me by Laurent
5 Dryzek, BJPS. Actually, I owe the title of the paper at hand to a suggestion by Mike Howlett.
6 The allusions in the title are to the University of Rochester, Virginia Tech and Wilhelm Gšthe †niversitŠt and
the original Institute for Social Research in Frankfurt, the respective “homes” of rational choice theory and critical
7 Johnson, Political Theory; APSR.
8 And perhaps Laurent, if I understand his position properly.
9 To acknowledge this does not mean the abandonment of a rational choice position, but rather its Aufhebung.
10 My argument is that what is called for is an intersubjective move, but that the nature of this move renders it
unallowable from a rational choice perspective, because the criterion for theoretical adequacy, from a rational
choice perspective, remains moored at the level of an identitarian logic in which the move beyond the individual,
seen as a realm of evidential grounding, threatens to throw the appearance of certainty provided by a rational
choice based account of reality into jeopardy. While political arguments can be made against the rejection of
cooperation by rational choice based theories, (along with its embrace of a competitive model of social
interaction) the deeper reason that this style of theory has an appeal is that it supplies individuals who perceive
themselves_with some degree of justifiability_as isolated individuals with the assurance that even if what they
know may not be absolutely certain they can know with certainty what they do not know. The criteria of truth for
rational choice-based theories remains that of mathematics, whereas the measure of truth in social life can never
be this tidy. See Cornelius Castoriadis’s discussion on identity thinking and its roots in mathematics in, The
Imaginary Institution of Society, translated by Kathleen Blamey (Polity Press: Cambridge, 1987).
11 See: Adorno, Popper, et al., The Positivist Dispute in German Sociology.
12 JŸrgen Habermas, Knowledge and Human Interests, (Boston: Beacon Press, 1968).
13 JŸrgen Habermas, Theory of Communicative Action, two volumes, translated by Thomas McCarthy (Boston:
Beacon Press, 1984; 1987).
14 Dryzek, BJPS, p.398, my emphasis.
15 Dryzek, BJPS, p.401
16 Dryzek BJPS, p. 410
17 Dryzek BJPS, p.410. My point here would be that the kind of mutual recognition involved in the
intersubjective relations within this kind of discourse situation are the prerequisite for subsequent cooperative
interaction and form the basis for a substantive account of democracy that does not fall into Rousseau’s problem,
nor into Habermas’s excessively formal, neo-Kantian account of discourse.
18 Dryzek, BJPS, p.410.
19 Habermas, TCA II.
20 Dryzek points this out. BJPS
21 Johnson, Political Theory, p.194.
22 Johnson, APSR, p.74. My emphasis.
23 Johnson, APSR, p.75. Johnson would do well, then, to examine Habermas’s critical discussion of
Luhmann’s work on systems theory, which attempts to integrate elements of hermeneutic theories of meaning
into systems system.
24 Johnson, APSR, p.75.
25 Johnson, APSR.
26 Habermas, TCA I.
27 For instance, see the discussion of this in the Applebaum-HŽbert Report. See my, “The Applebaum-HŽbert
Report and the Internal Colonialization of Canadian Culture.” Canadian Journal of Social and Political
Theory/Canadian Conference for theÊArts Colloquium, Learned Societies, Vancouver, June 4, 1983.
28 Johnson, APSR
29 Johnson, APSR, p.77.
30 What Habermas can be criticized for in this connection, in my opinion, is his failure to tease out the
dialectical relation between communicative and strategic action, which lies at the heart of his critique’s challenge
to instrumental and strategic reason, although Johnson is correct to point out the ambiguities in Habermas’s
treatment of these issues. My contention, in this regard, is that if Habermas were to be more dialectical in the
explanation of his position, the types of criticisms mounted by Johnson would have less ground upon which to
31 Johnson, APSR, p.77.
32 Johnson, APSR, p.82.
33 RŽjean Landry, “Rational Choice and Canadian Policy Studies,” typescript, p.12.
34 Landry, pp.14-15.
35 Landry, p.17.
36 Landry, typescript, p.34.
37 See, for instance, Michael Theunissen, The Other: studies in the social ontology of Husserl, Heidegger,
Sartre, and Buber, translated by Christopher Macann with an introduction by Fred R. Dallmayr. (Cambridge,
Mass.: MIT Press, 1984).
38 Robert Axelrod, The Evolution of Cooperation, (New York: Basic Books, 1984).
39 I am not arguing that the market relies on collusion, but rather that at an aggregate level cooperative
interaction must exceed competitive interaction for the market to survive.