* * *

This article is part of the UnderstandingSociety research gateway. Please visit the site, where you will find other useful articles, blogs, and an international social network site on the philosophy of social science, Marxism, and globalization.

* * *

 

Comments and discussion of the theory of practical rationality

Daniel Little

1991

 

Topics:

Comments and discussion of the theory of practical rationality 1

Theories of practical rationality 3

Altruism and commitment 3

Fairness 3

Decision rules 4

Possible alternative decision rules 4

Classification of criticisms of SER 7

Criticisms of narrow economic rationality 10

Principles of inference 15

The role of theories of rationality in social science 15

Fairness, Cooperation, and Rational Self-interest 18

Rationality Cooperation and Fairness 19

Economic rationality. 20

Problems with this conception. 20

Possible remedies. 22

Implications for the moral economy debate. 24

References 25

Altruism and fairness 27

Summary 30

Time preferences 31

Social science theories of rationality 32

Explanation without relativism 33

Where does the debate arise? 34

I. What is the thin theory of rationality? 35

Practical vs. belief rationality 35

A series of positions: 36

The role of theories of rationality in social science 36

References 39

 


Theories of practical rationality

Altruism and commitment

So far the qualification we have considered to the "collective action theorem" have accepted the premise of economic rationality; it has been argued on a variety of grounds that even economically rational individuals will engage more frequently in collective action than the CAT predicts.  Let us turn now to a more fundamental objection to the CAP: arguments to the effect that the concept of rationality as maximizing of private interests is itself defective.

 

            First, this conception of rationality is difficult to reconcile with the idea of moral decision making.  If rationality essentially involves maximizing self-interest, then acting out of moral concerns--moral rules, altruistic concern for others, or personal or political commitments--would appear to be irrational.  And in fact economists and public policy analysts tend to subsume moral motivations under the category of "non-rational influences" (e.g., Hardin (1982)).  

            Let us begin by making it clear that the objections which follow are not romanticizing efforts to displace the concept of rationality from our concept of human activity, or to elevate sentiment or passion to a dominant position in our understanding of action.  Rather, the critics we will consider argue that human beings are (imperfectly) rational; but that rationality does not consist simply in maximizing one's private interests.

 

Fairness

             A number of philosophers have attempted to incorporate the idea of fairness into the concept of rational decision making.  See for example the extensive literature on utilitarianism and fairness (Regan (1980), Griffin (1985), and Harsanyi (1985)). A reason for my performing an act is that I benefit from widespread performance of this sort of act, and I recognize that fairness requires that I pay my share of the cost of these public benefits.  John Rawls's A Theory of Justice represents an extended argument to the effect that there are principles of justice which ought to regulate the just society, and these principles derive from the principle of fairness.  Rawls's construction is at some distance from our primary concern, since he is concerned with global features of justice, while we are concerned with individual rationality.  But the kernel of Rawls's construction is relevant here: if individual rationality involves evaluating alternative lines of action in terms not only of the costs and benefits of each alternative, but also in terms of the fairness of each alternative, then we have arrived at a structured concept of rationality.  And it is a concept which involves the imposition of side constraints on the decision making process.

             A second, and more general, approach, is one which attempts to modify the conception of rationality so as to introduce a mechanism for moral deliberation.  Is it possible to describe a conception of practical reason which incorporates maximizing rationality and moral considerations?  The goal would be to arrive at a conception of rationality which contains economic prudence as a special case; provides a mechanism of deliberation within which moral constraints and reasons have scope;  and is sufficiently precise to permit some degree of reasoning about rational decision making.

             Let us now take stock from this excursion into recent moral philosophy.  What consequences, if any, follow from these considerations for the concept of rationality?  We saw in our discussion of Sen's work that there are good analytical and empirical reasons for judging that much actual human behavior is not explicable on the basis of a simple utility-maximizing scheme.  This finding might lead us to suppose that human beings are typically not rational; or it might lead us to question the concept of rationality associated with SER.  Sen suggests the latter course, and proposes that we attempt to construct a more structured concept of practical reason which permits us to take account of moral, political, and personal commitments as well as concern for welfare.  Moreover, Sen shows that the former cannot be subsumed under the simple concepts of welfare-maximizing or preference rank-ordering.

             Next, we saw in Nagel's work a series of arguments to the effect that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests.  This line of thought directly addresses the egoism assumption of SER.  It does not, however, do quite enough for us; for it does not give us a way of incorporating the idea of moral principles (or other normative requirements) into the decision making process.

             Finally, we saw that other recent moral philosophers have outlined the sort of structured decision making process necessary in order to take account of the role of principle in decision-making: the decision maker can combine a set of side constraints on action (normative commitments, in Sen's terms), as well as a set of goals (personal interest, social goals, the welfare of others, etc.).  And the decisions he arrives at will be a complex function of constraints and goal-maximizing actions.

             How do these findings relate to our central concerns?  First, they offer both a set of criticisms of SER (SER is an inadequate conception of practical reason because it makes it impossible to take account of certain features of the decision making process which are intuitively crucial), and an alternative model of the decision making process which promises to be a more adequate analysis of the concept of human rationality.  Moreover, this richer conception of practical reason promises to offer a new set of solutions to different classes of collective action problems: if individuals are altruistic to some degree (that is, responsive to the interests of others), and if they are principled (that is, moved by considerations of fairness, reciprocity, or justice), then they will be practically motivated to act differently when confronted with occasions for collective action than the CAT predicts.

             A second, and more general, approach, is one which attempts to modify the conception of rationality so as to introduce a mechanism for moral deliberation.  Is it possible to describe a conception of practical reason which incorporates maximizing rationality and moral considerations?  The goal would be to arrive at a conception of rationality which contains economic prudence as a special case; provides a mechanism of deliberation within which moral constraints and reasons have scope;  and is sufficiently precise to permit some degree of reasoning about rational decision making.

             A second, and more general, approach, is one which attempts to modify the conception of rationality so as to introduce a mechanism for moral deliberation.  Is it possible to describe a conception of practical reason which incorporates maximizing rationality and moral considerations?  The goal would be to arrive at a conception of rationality which contains economic prudence as a special case; provides a mechanism of deliberation within which moral constraints and reasons have scope;  and is sufficiently precise to permit some degree of reasoning about rational decision making.

           

Decision rules

Possible alternative decision rules

Generally the position to be argued here is that practical reason--the application of reasons to problems of choice--has a more complex structure than the maximizing conception allows; that it includes a capacity for the human actor to take account of moral reasons for acting and refraining; and that in consequence several classes of public goods problems may be seen to have solutions which are unexpected from the point of view of simple economic rationality.

            This discussion may begin with an obvious fact:  Individual human action, in a wide variety of social contexts from traditional societies to modern political organizations, is influenced by moral considerations.  Members of an organization or group may be induced to contribute to collective projects out of an unwillingness to take unfair advantage of other members' contributions (i.e., motives of fairness and reciprocity, and an unwillingness to be a free rider).  Or they may be induced to contribute to a collective scheme for aiding the indigent out of simple recognition of the reality of the needs of other persons which the scheme is intended to address (i.e., motives of altruism).  It is not necessary to provide extensive empirical justification for these assertions; they are a commonplace of human social life.

            The question is, how are we to describe action prompted by these sorts of motives?  One possible approach is to lump these sorts of behaviors together under the heading of "nonrational" or "extrarational" influences on conduct‑‑on a par with habitual conduct, conduct stimulated by subliminal persuasion, etc.  (This appears to be Hardin's approach (1982:103 ff.).)  On this approach, moral motivations are non‑rational impulses which push and pull action away from the optimal solutions which correctly represent rational self‑interest.  Thus to the extent that one is influenced by moral factors, one is less than wholly rational.

            Second, we might attempt to describe this sort of conduct as a rational calculus of a more complicated sort than that usually discussed in economic explanation, and refer to psychic benefits to which "moral behavior" gives rise.  The agent takes some amount of pleasure or satisfaction in acting altruistically, or he suffers some discomfort in acting unfairly; and when these pleasures and pains are taken into account, the action taken may be seen to maximize the individual's utility calculus.

            This approach looks suspiciously ad hoc, in that it introduces a hypothetical effect ("psychic benefit") in order to balance the rational calculus equation.  More importantly, though, this approach runs contrary to the best current thinking about the process of moral reasoning.

            Both these approaches appear to depend upon an emotivist theory of morality and moral motivation.  They assimilate moral action to a causal model of action in which a variety of non‑rational motive causes distortion of the process of rational deliberation.  Thus to be moral is to act contrary to what reason dictates.  But emotivism is not a particularly persuasive view of moral psychology.  And it would appear that one can do better than this by introducing a more complex model of practical reasoning which accords moral reasons some rational force.

            Let us briefly reconstruct a conception of "practical reason" which reflects recent thinking by moral philosophers and moral psychologists about the process of moral reasoning.[1]  The goal would be to arrive at a decision rule which contains economic prudence as a special case; provides a mechanism of deliberation within which moral constraints and reasons have scope; and is sufficiently precise to permit some degree of reasoning about rational decision making.

            Some philosophers and rational choice theorists have argued that the postulate of economic rationality may be modified with the addition of a motive supporting conditional fairness or reciprocity (Elster, Sen, Harsanyi, Margolis).  The decision rule on this approach might look something like this:  Maximize private interests unless the proposed action involves a public goods problems; in that case, estimate the likelihood of enough cooperative behavior to secure the collective good; if success is likely, cooperate; otherwise defect.

 

IF the action does not affect a public good

            THEN maximize private welfare.

            OTHERWISE estimate the likelihood of success of cooperative/                               collective action;

                        IF success is sufficiently probable,

                                    THEN cooperate;

                                    OTHERWISE defect.

 

This rule does not embody strict moral principle or altruism, but rather conditional reciprocity.  It represents a limited willingness to contribute to public goods, but only on the assumption that there is a realistic prospect that collective action will succeed. 

            The welfare-maximizing rule was flawed by being predisposed against cooperative action.  This rule is flawed in the opposite direction, however.  It doesn't correctly describe voters' behavior or collective action either, since it would predict virtually universal contribution to public goods.  The problem is that cooperative behavior preempts private striving on this rule whenever cooperation is sufficiently likely in others.  What the rule appears to need, therefore, is a weighting function which permits the decision maker to balance private costs and benefits against the gains of cooperative behavior and the prima facie presumption in favor of fairness or reciprocity.  Let's see if we can incorporate this feature into the model.

 

IF the action does not affect a public good

            THEN maximize private welfare.

 

            OTHERWISE estimate the likelihood of success of cooperative/                               collective action;

                        IF success is sufficiently probable,

                                    THEN assign an index V to the value of the collective good and cooperation in this case and compare with the cost of cooperation C.

                                                IF V > C

                                                            THEN cooperate;

                                                            OTHERWISE defect.

                                    OTHERWISE defect.

 

This rule begs an important question: how to compare private costs and the value of cooperation and the collective good.  It is consequently a less rigorously applicable rule than the simpler welfare-maximizing rule.  At the same time, it would appear to correspond more faithfully to the decision-making procedure which real persons undergo when deliberating about collective projects.  It is also noteworthy that this rule appears to collapse onto a Harsanyi-like dual-utility maximizing rule.  The individual assigns utilities to private goods and values to public goods, and possesses a weighting function which defines the rate at which tradeoffs are made between the two goods.  The decision rule may then be stated quite simply:  Act so as to maximize the weighted sum of private and public welfare.

            This rule is specifically constructed to give a more satisfactory account of individual decision making in circumstances of collective action.  It does not on its face take account of other moral considerations: altruism, fairness, and moral commitment.  Another promising insight is the idea that a full theory of rationality must also involve different sorts of considerations altogether.  An example of an attempt to embody moral principle in the decision rule is Nozick's notion of a "side-constraint theory."  Such a theory would presumably have a welfare-maximizing component, so that the theory would contain economic rationality as a special case (i.e., in circumstances where moral principles and requirements of fairness have no application).  But the rule would have an overriding set of moral constraints which limit choice of actions according to their consistency with a list of moral rules.

            This concept of rationality has a fundamentally different logical structure than that of SER.  It is not fundamentally a maximizing function, but rather a complex of side‑constraint rationality and maximizing rationality.  This more elaborate structure permits us to bring the idea of moral limitation and moral obligation into the domain of rationality.

            This sort of rule might be modeled along these lines:

 

            LIST alternative actions in order of the (private/general) welfare they generate.

            ELIMINATE actions which violate specified moral rules.

            CHOOSE the highest ranking alternative which survives the filter. 

 

This rule becomes either "morally constrained egoism" or "morally constrained altruism" depending on how welfare is understood; either as self-interest, or as interests of all affected persons.

 

 

Classification of criticisms of SER

 

             Criticisms of the concept of economic rationality have occurred along two independent dimensions.  First, there is a psychological criticism having to do with the assumption of self-interestedness.  The usual concept of economic rationality assumes that each individual is out to maximize the satisfaction of his own private interests; thus economic man is an egoist.  Against this assumption, various critics have argued that human beings have a variety of altruistic motives; they are capable of acting out of regard for the interests of others as well as out of self-interest.  Thus human beings sometimes accept sacrifices to private interest for the sake of family, friends, anonymous strangers, or the social group.  And it is sometimes held that this form of behavior is ruled out by the assumption of economic rationality.

             One especially important example of this type of view is that of the utilitarian decision maker who impartially chooses strategies according to the sum contribution which each alternative makes to the welfare of all.

             Let us refer to this as the "content objection": it maintains that the theory of economic rationality assumes that the content of rational planning is defined by self-interest, whereas most human beings sometimes pursue the interests of others as well.  This objection is not quite on target, however.  For the theory of economic rationality is not in fact forced to assume egoism (though it frequently does).  Rather, it assumes only that each individual has an ordered set of goals or interests which he would like to further; and then he engages in activity which is designed to effectively maximize the attainment of those goals.  Consequently, it is consistent with the concept of economic rationality to attribute altruistic goals to the agent (though it is necessary to provide an ordering to those goals).  And so long as the agent pursues those goals by attempting to maximize or optimize their attainment, he may be said to conform to the requirements of economic rationality.

             

             When we explicitly add the assumption that individuals attach some weight to the welfare or interests of others, we are immediately able to solve some of the paradoxes of collective rationality.  In particular, the 2-person prisoner's dilemma looks more tractable on the assumption that both players are making their decisions on the basis of both players' payoffs, both will arrive at the "cooperate" strategy.  This is true absolutely for the impartial decision maker who weights both players' welfare equally.  And it is frequently true even for players who are partial to their own interests in the sense that they discount payoffs to the opponent, so long as the conflict between interests of the two parties is not too severe.  Thus in the game matrix above, strict impartiality leads to a choice of "cooperate"; but so does a principle of choice in which one values his own welfare at a rate twice that of his opponent.

             However, the addition of the assumption of altruistic consideration of the welfare of others does not solve all types of collective action problems.  In particular, the n-person PD persists even in the light of this additional assumption.  This follows from two assumptions.  First, an individual's contribution adds only a negligible amount to the total collective good.  And second, the individual's cost is not negligible.  On these assumptions, the maximizing altruist will reason that his contribution will be wasted, and will refrain from contributing.

             Strict and limited utilitarianism, then, represent a family of views which overcome the egoism feature of standard economic rationality, but which preserve the idea of rational choice as a welfare-maximizing process.  (This discussion relates closely to recent work in utilitarian ethics.  Particularly important is the debate over act and rule utilitarianism.  See Sen and Williams (1982) for relevant contributions.)

             

             The second type of criticism of standard economic rationality is more fundamental.  This view holds that the conception of rationality as maximizing welfare is itself fundamentally flawed--regardless of the issue of egoism and altruism.  Or rather, the view holds that maximizing decision making is one part of a full theory of rationality, but that a full theory of rationality must also involve different sorts of considerations altogether.  The chief candidate for addition to the rational decision making procedure is the idea of following a moral rule or principle.  On this account (which is manifestly Kantian in tone), the rational decision maker will be mindful of two fundamentally different kinds of considerations.  For a range of possible actions A<1>, ... A<i> the decision maker will first ask whether any of these actions is required by moral principle, and whether any are ruled out by moral principle.  Any actions which are ruled out by moral principle are dropped from the decision making process.  If a unique action is required by moral principle it will be adopted.  If no decision has been reached, the decision maker now moves to the optimizing stage: he assesses the alternative actions in terms of their contributions to his table of preferences and selects that action which optimize his preferences.

             This concept of rationality has a fundamentally different logical structure than that of SER.  It is not fundamentally a maximizing function, but rather a complex of side-constraint rationality and maximizing rationality.  This more elaborate structure permits us to bring the idea of moral limitation and moral obligation into the domain of rationality.

             Moreover, this additional assumption permits the solution of a wider class of collective action problems.  If the requirement of fairness or reciprocity is a moral requirement, then "free-rider" obstacles to collective action will be diminished.  The prospective contributor will consider contributing and taking a free ride.  He will rule out the free ride on grounds of fairness.  And he will choose out of the restricted set, and decide to contribute.

             We might note parenthetically that the Kantian moral philosopher goes one step further, by trying to show not only that it is rational to act out of regard for moral principle even in cases where doing so leads to sub-optimal outcomes in terms of preferences; but further, that a particular range of moral principles are rationally derivable or justifiable.  (See Nagel, Rawls, Kant.)  We might call this the concept of practical reason.

             These considerations suggest a possible classification of models of rational decision making:

 

 

TYPE OF INTEREST REFERRED TO

 

TYPE OF DECISION PROCESS

 

 

 

Maximizing

side-constraint

Egoism

SER

reciprocal egoism

altruism

utilitarianism

Kantian ethics

             

             Two important criticisms of SER emerge from this discussion.  First, SER has historically been inclined towards an assumption of egoistic self-interest; and in fact prisoner's dilemma situations, game theoretical reasoning, and the like depend upon this assumption.  However, it is possible to introduce the assumption that decision makers take the interests or welfare of others into account (with or without a discount function) and preserve the main outlines of SER.  More fundamentally, we found that telling objections have been put forward the maximizing structure of SER; philosophers and economists alike have argued that a full theory of rationality needs to give a place for the operation of moral principle in the decision making process.  Further, it was shown that introducing altruistic consideration of others solves some collective action problems but not all; but that when we add the requirements of moral principle, we find that a wider class of collective action problems are solved.

             If these points are to have empirical significance (as opposed to purely philosophical plausibility), it is necessary to give some account of the mechanisms by which moral principles and altruistic motives come to be motivationally effective.  And it is necessary to give some concrete evidence to the effect that actual human decision making reflects the motivational diversity and rational structure which these criticisms predict.  We will turn to a discussion of the former below when we consider the causal role of moral systems; briefly, the account will emphasize socialization processes through which individuals come to internalize the norms of the social order around them.  There are some suggestions in the biological literature, however, that altruism in primates has an evolutionary basis as well; that is, that altruism is not merely the result of socialization, but is rather in some sense built into the human decision making system (much as various features of linguistic competence are evidently built into the neurophysiology of cognition).  See Margolis (1982) for further development of this possibility.

             How do these factors affect Popkin's arguments?  To the extent that Popkin's analytical work--and the predictions which he derives concerning village institutions--depends upon a theory of human motivation and decision making which is importantly mistaken, it will be evident that the theoretical inferences have little application to human society.  If villagers are altruistic in the narrow sense of being affected by their perceptions of the needs and welfare of others; and if villagers' decision making processes are importantly constrained by simple moral constraints like fairness, reciprocity, and the like; then the CAT does not validly apply to village society because village society does not satisfy its premises.

             We must be careful not to draw an overly strong conclusion, however; for no one maintain that human beings are indifferent to private welfare.  Indeed, generally speaking it would seem reasonable to assume that each decision maker places high priority on personal and familial welfare; human beings generally do not behave like impartial utilitarians.  This finding suggests that human behavior is the resultant of several different forms of motive: self-interest and altruism; and several different types of decision making processes: maximizing and side-constraint testing.  And to the degree that self-interest and maximizing behavior are prominent in a particular type of circumstance, to that degree the CAT will be empirically significant.  In village society one presumably will find a mix of these four possibilities; consequently it is invalid to directly apply the CAT.  Rather, it is necessary to offer empirical arguments and analysis to show in what circumstances the premises of the CAT are most fully satisfied, and in which as a consequence we would most confidently predict collective action problems.  But we have now seen a variety of features of community life which plausibly work to enhance both the impulse to altruism and the impulse towards fairness and reciprocity; and these considerations in turn suggest that some communities in some circumstances may manage to effectively overcome collective action problems.

 

Criticisms of narrow economic rationality

 

            Much of the argument in this chapter has been based on the supposition that rational-choice analysis sheds substantial light on a variety of processes of social change.  A number of criticisms have been advanced in recent years, however, to the narrow conception of economic rationality.  These criticisms have direct bearing on the issue of the adequacy of the rational choice framework of analysis of social phenomena.  These criticisms must be addressed, and their significance evaluated, before we can reasonably conclude that the rational-choice model has significant application in the explanation of concrete social phenomena.  Now we will consider the adequacy of the assumption of narrow economic rationality itself and the body of public choice analysis which depends upon it.  We will find that this theory is inadequate as a comprehensive account of rational deliberation; it is therefore not a sound base for the theory of rational-intentional action; and social science is therefore to some extent misguided when it founds explanations of aggregate patterns on the simple public choice model.  It should be emphasized that this position does not deprecate the importance of rationality in explaining human activity, or elevate emotion or impulse to a dominant position in our understanding of action.  Rather, the critics we will consider argue that human beings are (imperfectly) rational, but that rationality does not consist simply in maximizing one's private interests.

            We may begin with A. K. Sen's critique of narrow economic rationality.  Sen criticizes the assumption of pure self‑interest which is contained in the standard conception (Sen 1982:84‑90).  "The purely economic man is indeed close to being a social moron" (Sen 1982:99).  Against the assumption of self‑interested maximizing decision‑making, Sen argues for a proposal for a more structured concept of practical reason: one which permits the decision maker to take account of commitments.  This concept covers a variety of non‑welfare features of reasoning, but moral principle (fairness and reciprocity) and altruistic concern for the welfare of others are central among these.  "Commitment is, of course, closely connected with one's morals.  But moral this question is in a very broad sense, covering a variety of influences from religious to political" (Sen 1982:93).  Sen believes that the role of commitment is centrally important in the analysis of individuals' behavior with regard to public goods.  For example, he suggests that the voters' paradox may be explained by assuming that "voters are not trying to maximize expected utility, but . . . to record one's true preference" (Sen 1982:97).  And he draws connections between the role of commitment and work motivation.  "To run an organization entirely on incentives to personal gain is pretty much a hopeless task" (Sen 1982:98).  He argues, therefore, that in order to understand different areas of rational behavior it is necessary to consider both utility‑maximizing decision making and rational conduct influenced by commitment; and it is an empirical question whether one factor or the other is predominant in a particular range of behavior (Sen 1982:104).

            Sen suggests the outline of an alternative decision-making rule.  On this account the agent is postulated to embody two sorts of preference rankings: a ranking of first-order preferences and a meta‑ranking of preference rankings in terms of the relation "ranking x is more moral than ranking y" (Sen 1982:100‑101). 

            We might ask briefly to what aspect of the theory of rationality these points are directed--the formal theory or the empirical theory.  Does Sen mean to assert that the narrow conception of economic rationality is flawed on logical grounds, or is his point rather that this conception does unacceptable violence to the actual processes of human rational decision making?  It would appear that Sen believes both claims, but the logic of his position flows from the second point to the first.  That is, he appears to reason, first, that narrow economic rationality is not a good account of actual human decision-making; and second, that this fact is a reason for modifying the formal theory of rationality as well.

            Thus Sen holds that an adequate theory of rationality requires more structure than a simple welfare‑maximizing model would allow; in particular, it needs to take account of moral principles and commitment.  Sen's proposals for a more structured conception of practical reason, however, are the least convincing part of his account.  The idea of a meta‑ranking of preference rankings is unintuitive, to say the least; it should be possible to give a better account of the relation between means‑end rationality (utility maximizing) and rational subordination of action to principle (moral reasoning).  Therefore let us now turn briefly to important contributions by recent moral philosophers.

            Sen's arguments show that there are good analytic and empirical reasons for judging that much actual human behavior is not explicable on the basis of a simple utility‑maximizing scheme.  This finding might lead us to suppose that human beings are typically not rational, or it might lead us to question the concept of rationality associated with the standard conception.  Sen suggests the latter course, and proposes that we attempt to construct a more structured concept of practical reason which permits us to take account of moral, political, and personal commitments as well as concern for welfare.  Moreover, Sen shows that the former cannot be subsumed under the simple concepts of welfare‑maximizing or preference rank‑ordering.  (Sen's main contributions are contained in "Rational Fools" and "The Impossibility of a Paretian Liberal" in Sen, ed. 1982.)  Thus Sen holds that an adequate theory of rationality requires more structure than a simple welfare‑maximizing model would allow; in particular, it needs to take account of moral principles and commitment.  Sen's proposals for a more structured conception of practical reason, however, are the least convincing part of his account.  The idea of a meta‑ranking of preference rankings is unintuitive, to say the least; it should be possible to give a better account of the relation between means‑end rationality (utility maximizing) and rational subordination of action to principle (moral reasoning).  Therefore let us now turn briefly to important contributions by recent moral philosophers.

            Various moral philosophers have argued that practical rationality is more comprehensive than narrow economic rationality.  Thus Thomas Nagel provides a series of arguments to the effect that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests (Nagel 1971).  This line of thought directly addresses the egoism assumption of the standard conception.  It does not, however, do quite enough for us; for it does not give us a way of incorporating the idea of moral principles (or other normative requirements) into the decision making process.  But other recent moral philosophers have outlined the sort of structured decision making process necessary in order to take account of the role of principle in decision making: the decision maker can combine a set of side constraints on action (normative commitments, in Sen's terms), as well as a set of goals (personal interest, social goals, the welfare of others, etc.).

            Thomas Nagel's The Possibility of Altruism provides one possible way of developing a more comprehensive theory of practical reason.  Nagel's work is an extended essay on the logical character and requirements of practical reason.  In particular, Nagel argues that once we have correctly analyzed the requirements of reason, we will discover that simple egoism is irrational.  "Just as there are rational requirements on thought, there are rational requirements on action, and altruism is one of them" (Nagel 1970:3).  This argument thus represents a direct challenge to the assumptions of standard conception of rationality, and a defense of the view that practical rationality has a substantially more complex structure than simple utility‑maximizing models allow.

            Nagel begins with a conception of rationality which is substantially different from the standard conception of rationality: being rational is acting for good reasons.  "We may explain what a man does by referring to his reasons. . . .  A consideration can operate as a motivating reason only if it has, or is thought to have, the status of a reason in the system of normative principles by which individuals govern their conduct.  Such normative principles therefore specify significant features of the motivational structure" (Nagel 1970:15).  And Nagel believes that once we work out the implications of this conception of "acting on the basis of reasons", we will discover that certain consequences follow for the actor.  In particular, it follows that the actor will act out of regard for the interests of others; he will act altruistically.  "The general thesis to be defended concerning altruism is that one has a direct reason to promote the interests of others‑‑a reason which does not depend on intermediate factors such as one's own interests or one's antecedent sentiments of sympathy and benevolence" (Nagel 1970:16).

            It should be noted that Nagel's concept of altruism is one which holds that the rational agent will give practical weight to the interests of others as well as to his own interests.  "By altruism I mean not abject self‑sacrifice, but merely a willingness to act in consideration of the interests of other persons, without the need of ulterior motives" (Nagel 1970:79).  For our purposes Nagel's arguments suggest one important conclusion: the standard model of economic rationality cannot serve as a full theory of practical reason.  Instead, it is necessary to conceive of the rational decision maker as being subject to a variety of moral constraints which modulate behavior and which sometimes restrict self‑seeking actions.

            Thus Nagel intends to establish that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests (Nagel 1971).  This line of thought directly addresses the egoism assumption of the standard conception.  It does not, however, do quite enough for us; for it does not give us a way of incorporating the idea of moral principles (or other normative requirements) into the decision making process.

            Finally, a number of philosophers have attempted to incorporate the idea of fairness into the concept of rational decision making.  A reason for my performing an act is that I benefit from widespread performance of this sort of act, and I recognize that fairness requires that I pay my share of the cost of these public benefits.  John Rawls's A Theory of Justice represents an extended argument to the effect that there are principles of justice which ought to regulate the just society, and these principles derive from the principle of fairness.  Rawls's construction is at some distance from our primary concern, since he is concerned with global features of justice, while we are concerned with individual rationality.  But the kernel of Rawls's construction is relevant here: if individual rationality involves evaluating alternative lines of action in terms not only of the costs and benefits of each alternative, but also in terms of the fairness of each alternative, then we have arrived at a structured concept of rationality.  And it is a concept which involves the imposition of side constraints on the decision making process.  A more structured decision making process is necessary in order to take account of the role of principle in decision making: the decision maker can combine a set of side constraints on action, as well as a set of goals.  And the decisions he arrives at will be a complex function of constraints and goal‑maximizing actions.

            How do these findings relate to our central concerns?  First, they suggest that the narrow conception of economic rationality is an inadequate conception of practical reason because it makes it impossible to take account of certain features of the decision making process which are intuitively crucial.  Further, these considerations suggest an alternative model of the decision making process which promises to be a more adequate analysis of the concept of human rationality.  Moreover, this richer conception of practical reason promises to offer a new set of solutions to different classes of collective action problems:  If individuals are altruistic to some degree (that is, responsive to the interests of others), and if they are principled (that is, moved by considerations of fairness, reciprocity, or justice), then they will be practically motivated to act differently when confronted with occasions for collective action than the collective action theorem predicts.

            These findings have direct import for the applicability of rational-choice models in social explanations.  For example, recall the discussion in chapter 2 of freeriding and public goods problems.  Once we consider a more complex theory of practical deliberation, formal arguments predicting the emergence of public goods problems in real social groups will be found to be misleading.  On a more complex, and more empirically adequate, account of practical reason, altruism, cooperation, and reciprocity are rational choices; therefore we would expect a social group consisting of rational individuals to show marks of cooperation and altruism. 

            We must be careful not to draw an overly strong conclusion, however; for no one maintain that human beings are indifferent to private welfare.  Indeed, generally speaking it would seem reasonable to assume that each decision-maker places high priority on personal and familial welfare; human beings generally do not behave like impartial utilitarians.  This finding suggests that human behavior is the resultant of several different forms of motive: self‑interest and altruism; and several different types of decision making processes: maximizing and side-constraint testing.  And to the degree that self‑interest and maximizing behavior are prominent in a particular type of circumstance, to that degree the collective action theorem will be empirically significant.  In village society one presumably will find a mix of these four possibilities; consequently it is invalid to directly apply the collective action theorem.  Rather, it is necessary to offer empirical arguments and analysis to show in what circumstances the premises of the collective action theorem are most fully satisfied, and in which as a consequence we would most confidently predict collective action problems.  But in chapter 2 we saw a variety of features of community life which plausibly work to enhance both the impulse to altruism and the impulse towards fairness and reciprocity; and these considerations in turn suggest that some communities in some circumstances may manage to effectively overcome collective action problems. 

            These findings concerning the adequacy of the concept of economic rationality have immediate consequences for Popkin's case against the moral economists.  For Sen and Nagel give reasons for holding that moral commitments and values are a constitutive part of human rationality.  In the context of village society we must ask whether there are moral norms which might affect individual action; and the answer to this question appears to be unavoidably affirmative.  Scott's arguments to this effect are particularly strong (1976:13‑56); but almost any study of traditional society would do just as well.  And the arguments advanced by Sen and Nagel suggest a way of incorporating adherence to moral standards (e.g., reciprocity, fairness, or unwillingness to be a freerider) into the process of rational decision making.  Thus it is consistent to hold that rational peasants may modulate their conduct out of regard for various forms of traditional morality.  These arguments therefore cast doubt on Popkin's assumption that the rational peasant is exclusively motivated by considerations of familial or personal welfare, and they undermine Popkin's skepticism about the motivational efficacy of moral principles and village assumptions about fairness.  If villagers are altruistic in the narrow sense of being affected by their perceptions of the needs and welfare of others; and if villagers' decision making processes are importantly constrained by simple moral constraints like fairness, reciprocity, and the like; then the collective action theorem does not validly apply to village society because village society does not satisfy its premises.

            Two important criticisms of the standard conception of economic rationality emerge from this discussion.  First, the standard conception has historically been inclined towards an assumption of egoistic self‑interest; and in fact prisoners' dilemma situations, game theoretical reasoning, and the like depend upon this assumption.  However, it is possible to introduce the assumption that decision makers take the interests or welfare of others into account (with or without a discount function) and preserve the main outlines of narrow economic rationality.  More fundamentally, we found that telling objections have been put forward the maximizing structure of narrow economic rationality; philosophers and economists alike have argued that a full theory of rationality needs to give a place for the operation of moral principle in the decision making process.

 

Principles of inference

 

What, then, are the principles of inference that constitute rationality?  The following are illustrative.

 

Principles of deductive inference:

 

if you believe a set of sentences Si, and P logically follows from Si, then accept P as well (or else give up belief in some of Si).

 

Principles of inductive inference:

 

Inductive generalization: if you have observed n individuals and all have the property P, infer that all have the property P.  [qualifications on sample size and sample technique.]

 

All hypotheses must be empirically testable in some fashion.

 

Testing of hypotheses through deductive consequences

 

Inference to the best explanation

 

Evaluate causal hypotheses using Mill's methods.

 

Choose the simplest of a group of empirically equivalent hypotheses

 

Occam's razor: don't postulate new entities unless forced to by empirical findings.

 

 

The role of theories of rationality in social science

 

            Explanations in social science often refer to the rational actions of individuals.  Before considering the logic of such explanations in detail, let us ask a preliminary question:  What is the intended import of various theories of rationality?  Several possibilities suggest themselves.  First, the theory of rationality may be a normative discipline which is intended to establish universal rules of reasoning in support of belief and action.  (Call this the "pure theory" of rationality.)  On the side of belief this discipline would include deductive and inductive logic, probability theory, and (perhaps) applied philosophy of science.[2]  These disciplines are normative in that they attempt to establish criteria of sound inference; to be rational is to reason in accordance with these criteria.  On the side of action the normative theory of rationality would include an account of the rules of reasoning which "ought" to allow a decision maker to consider his beliefs, goals, and (perhaps) values, and arrive at the optimal action in the circumstances.  Partial efforts in this direction include the disciplines of decision theory and game theory; one might also include various theories of applied ethics as potential candidates in this part of the theory of reason.[3]  Overall, the goal on this approach is to arrive at a decision procedure for action which generates the "right" decisions on the basis of a given range of information about goals and beliefs.

            It should be noted that on this account the theory of rationality is sufficiently idealized that it bears little relation to actual human decision making.  For symbolic logic, decision theory, and game theory abstract from limits on computational power, information access, memory, and the like.  These abstractions entail that ordinary human reasoners cannot fully embody these models of rational calculation.  For example, even two-person games for which game theory shows that solutions exist are too complex for anyone but a mathematical economist with ample computational resources to solve.  This finding shows that human reasoners do not solve strategic decision problems using two-person game theory.[4]

            A rather different conception of the import of a theory of rationality is one according to which the goal is to provide an empirically grounded theory of deliberative human action.  (Call this the "cognitive theory" of rationality.)  On this account, the theory of rationality is a branch of cognitive psychology.  One might pose this question:  How do human beings arrive at beliefs and make decisions under optimal circumstances (i.e., ample time for decision making, low level of emotional influences on the process of deliberation, and adequate informational resources)?  On this account we would conceive of the theory of rationality much as Chomsky conceives of the theory of grammar: as an idealized representation of psychologically real processes and protocols.[5]  On this approach it is an open question whether there is one form of deliberative decision making or many.  It is logically possible, for example, that different cultures have developed different structures of reasoning which are transmitted through education from one generation to the next.

            If after appropriate empirical research we were to conclude that there are important features in common among all cultures in terms of deliberative reasoning (as Chomsky argues there are features in common among all natural languages), then the theory of rationality would have important application in social science.  One would be able to construct explanations of various social patterns on the basis of analysis of the decision-making procedures of individuals within the environments of choice which their societies pose for them.

            What are the relations between these different sorts of theories?   Consider first the relation between the pure theory and the cognitive theory.  One possible view is that the formal discipline is in fact a highly idealized representation of what the empirical discipline would uncover after full investigation.  Formal logic correctly identifies the rules of deductive inference, and human beings (through favorable evolution, possibly) embody these rules in their ordinary processes of reasoning.  Likewise, decision theory and game theory establish the correct principles of reasoning in circumstances of uncertainty and competition, and human beings imperfectly embody these rules.  The political economy approach to social science appears to presuppose something like this.[6]

             The relation between the cognitive theory and the means-end theory might go along these lines.  The means-end theory presupposes that agents embody some cognitive process of deliberation or other, but it tacitly presupposes that this process is a relatively simple one.  However, whatever the findings of the empirical discipline of deliberative reasoning, the theory of rational action will simply incorporate the resulting theory into its attempts to explain behavior as the calculated effort to achieve goals on the basis of beliefs.

            Finally, what is the relation between the interpretive theory and other theories of rationality?  Its proponents view the interpretive theory as the most comprehensive account of intelligible human action among these alternatives, and an account within which other views of rationality constitute special cases.  For much human action--in our culture and others'--may be understood as calculated to maximize the attainment of one's ends, given one's beliefs.  Therefore means-end explanations are often the correct way of understanding the agent's behavior.  But other forms of action do not conform to this model; therefore interpretation theory requires that we construct other models of action within the context of which to understand these types of action.

            Where does the theory of economic rationality lie in this taxonomy?  First, it falls within the formal theory of rationality in its guise as formal decision theory.  Economists and decision theorists are interested in working out the logic of the simple assumption of maximization of utility, and the complications which risk, uncertainty, and strategic interactions introduce into this model.  Second, it plays a role in the empirical theory of deliberation in that some have presented narrow economic rationality as a model of the cognitive process of deliberation.  The argument here is that narrow economic rationality correctly describes a wide range of human behavior, and the best explanation for this fact is that human beings embody a process of calculation which approximates the formal theory of economic rationality.  That is, human beings really act so as to maximize private interests.  The theory of narrow economic rationality also has a central role in the means-end theory of rational action; it is the procedure which common sense recommends for understanding action in terms of goals and beliefs.

            The main assumptions of the rational-choice paradigm are these: human beings are deliberative; they are capable of forming true beliefs about their natural and social environment; and they are concerned about individual and family welfare.  Many criticisms of this model derive from the suspicion that rational-choice analysis makes unreasonable assumptions about individual behavior as excessively calculating and excessively materialistic.

 

            We might ask briefly how public choice theory fits into the taxonomy of theories of rationality provided earlier.  The goal of public choice theory is to analyze the collective or aggregate consequences for a social group as a whole of the rational choices made by individuals within structured circumstances of choice.  At present this field depends upon the narrow conception of economic rationality; but what changes would be forced by incorporating a fuller and more adequate conception of decision making?  Much of the power of public choice theory derives from the simplicity of the assumption of economic rationality; it is quite possible that a substantially more complex theory of rationality would make it impossible to derive formal results in the aggregate.  Taking an extreme case, it would appear unpromising to require a description of the full range of possible motives of individual action and attempt to derive aggregate consequences.  So some degree of abstraction is required in the description of individual decision making.  At the same time, whatever results are derivable from the richer assumption would presumably be more directly applicable to the behavior of existing social groups.

 

Fairness, Cooperation, and Rational Self-interest

 

Problems of rationality and morality

economic rationality

 

Standard conception: maximize rational self-interest.

 

Problems of collective rationality:

public goods problems

Hardin

Popkin

freerider problems

prisoners' dilemma

Taylor

Axelrod

 

Does the standard conception fit with actual individual behavior?

Sen: Rational Fool, Paretian Liberal

Simon: substantive rationality

Taylor: community

 

Practical reason

Morality and altruism

Nagel

fairness

Rawls

cooperation

Regan

 

Is it possible to describe a conception of practical reason which

incorporates maximizing rationality and moral considerations?

Harsanyi and Margolis: two preference structures.

Sen: metaranking of preference structures.

 

The goal would be to arrive at a conception of rationality which:

 

contains economic prudence as a special case;

 

provides a mechanism of deliberation within which moral constraints and reasons have scope;

 

is sufficiently precise to permit some degree of reasoning about rational decision making.

 

Cases to consider.

Hardin

Popkin

Axelrod

Taylor

 

Rationality Cooperation and Fairness

 

            I am currently working on a project in the philosophy of social science which is concerned with the use of assumptions about economic rationality to explain patterns of village life in premodern agrarian societies.  There are two important schools of thought in this area.  The first (the "moral economy" school) holds that peasant societies are knit together by powerful communal institutions, and that individual peasant families organize their daily lives (e.g., labor contributions to collective projects, decisions about cultivation, contributions to collective welfare funds, etc.) through recognition of the requirements of communal life.  On this approach a shared morality, communal institutions, and universally recognized obligations of reciprocity are the guiding features of village life. 

            The other school (the "rational peasant" school) holds that peasants are rational decision makers; they are chiefly concerned with family welfare over the medium run; and as a result peasant societies show the mark of classical collective action problems, prisoners' dilemmas, etc.  In particular, this school holds that collective goods are consistently underachieved because of individual noncontribution, and that communal institutions are either undermined by noncompliance or exploited by powerful individuals for their own benefit.   Crucial to this dispute is the status of the assumption of economic rationality.  However, there is significant controversy about this assumption along precisely those dimensions which are most significant for the moral economy debate: the adequacy of the theory of economic rationality in circumstances where cooperation and fairness are at issue, where moral values arise, and where public goods are concerned.  Philosophers, economists, and game theorists have raised serious objections to the theory of economic rationality on the ground that it cannot give a satisfactory account of behavior motivated by moral principle, by altruism, or by fairness and reciprocity.

            In the following I'd like to introduce some of these issues for discussion.  My goal here isn't to give you a finished position; I've put this together in a few afternoons.  Rather, I'd like to bring together the main outlines of some of the thinking in this area in order to see if we can make some progress on the issue in conversation.

 

Economic rationality.

 

            The standard conception of economic rationality is one according to which the agent acts so as to maximize private interests, preference satisfaction, utility, or welfare.  Recall Rawls's premise in the original position:  Individuals are privately rational and mutually disinterested.  This is the standard assumption of game theory, economics, and decision theory.  In order to make use of this conception in a particular decision problem, the analyst identifies a measure of utility for the players, and then tries to work out an optimal strategy for each player to achieve the highest score.

             

Problems with this conception. 

 

(a)        First, this conception of rationality is difficult to reconcile with the idea of moral decision making.  If rationality essentially involves maximizing self-interest, then acting out of moral concerns--moral rules, altruistic concern for others, or personal or political commitments—would appear to be irrational.  And in fact economists and public policy analysts tend to subsume moral motivations under the category of "non-rational influences" (e.g., Hardin (1982)). 

            Consider a case.  Suppose a prosperous farmer Smith attends a repossession sale of the property of failed farmer Jones, but refrains from bidding on an attractively priced tractor which he needs and will have to purchase later at a higher price.  His action fails to maximize his own private welfare; so let's canvass possible motives which might have led to this choice.  First, the farmer may feel personal loyalty or friendship for the bankrupt Jones, and chooses not to benefit from Jones's misfortune.  Next, Smith may be acting according to a principle of solidarity:  If members of the community refuse to participate in farm sales, banks and creditors will have less incentive to foreclose.  Or Smith may be a Kantian; he may reason that he could not will the universalization of his action.

Finally, Smith may be flatly irrational; e.g., he may have miscalculated, or he may be under the influence of magical thinking, leading to the conviction that involvement in Jones's misfortune now will lead to his own failure in the future.  The point of the example is this:  The standard conception of economic rationality cannot embrace any of these motives as rational.  But a priori, we need a conception of practical rationality which can give appropriate place to moral motives, and which can distinguish between irrational beliefs and miscalculations, on the one hand, and legitimate moral reasons, on the other.

 

(b)       Second, this conception of economic rationality has difficulty dealing with collective action and interactive behavior.  This model predicts that a group of rational persons will have a great deal of difficulty coordinating individual action to achieve collective benefits.  Here two large categories of problems have arisen: public goods problems and cooperation problems.  Mancur Olson showed that groups composed of rational individuals would find it virtually impossible to achieve collective action in pursuit of universally recognized collective interests.  And the classic prisoners' dilemma shows that rational individuals are unable to achieve a cooperative solution to a transparent situation of choice.

            Olson's collective(c)action result can be illustrated as follows.  Imagine a group of rational individuals who share a common interest in an unpolluted lake.  Each individual assigns a value to his share of this public good of $2000.  The primary threat to the lake is garbage disposal; each household can either bury its garbage for free, or purchase a garbage sanitizer for $1000.  Finally, the effect of each household's untreated garbage on the cleanliness of the lake is small, and no household can be denied access to the lake, whatever its garbage disposal system.  Now consider the reasoning of a representative decision maker when confronted with the opportunity of purchasing the garbage sanitizer.  He reasons that either most households will sanitize or they will not.  His decision will not determine which of these alternatives materializes.  His payoff for any of the four possibilities is equal to the difference between his share of the public good and his private cost in installing the sanitizer.  His payoffs may be listed as follows:

                                                            OTHERS

                                                            sanitize                        don't sanitize

 

            sanitize                                    1000                            (c)1000

SELF

            don't sanitize                2000                            0

 

Whatever others do, therefore, SELF is better off not sanitizing.  This setting is symmetrical for all players, however; so the conclusion is that no one sanitizes, the lake is polluted, and no amount of the public good is purchased.  The general result is this:  Rational, mutually disinterested persons will not act so as to secure collective goods.

 

(c)        Finally, however, this conception of rationality faces significant empirical problems.  Actual human behavior seems widely different from what this theory of rationality would predict in a wide variety of cases.  In an extremely wide range of examples human beings act individually and collectively in ways which appear to violate the requirements of economic rationality.  This finding could be taken to suggest "either• that human beings are characteristically irrational (or only partially rational); or it could suggest that this conception of rationality is suspect.

            A few examples will illustrate this point.  First, Olson-like public goods arguments show that it is individually irrational to make the private expenditure of going to the polls--even though the individual's share of the collective good of a successful election exceeds the private cost.  This is because the individual will receive the public benefit whether or not he has contributed.  Rational choice theorists take the fact of 50% turnouts in presidential elections as evidence for their view: half the electorate behaves as this theory of rationality requires.  The point may be reversed with good effect, however: fully half does not behave "rationally". 

            Other public goods cases show much the same result--e.g., subscription drives for public radio and television, or political fundraising efforts.  These examples seem to show that significant numbers of educated, intelligent American citizens act contrary to the dictates of private rationality. 

            Finally, consider strikes, rebellion, and revolution.  Various authors (e.g., Allen Buchanan) have applied Olson's analysis to collective behavior of classes.  When considered as a problem of rational choice, the decision to participate in a strike or rebellion appears to be yet another example of a public goods problem.  There is a collective interest to be gained through successful collective action; but no individual has a sufficient private interest in contribution.  So the strike or rebellion should fail.  However, there are ample examples of successful collective class action—the Solidarity movement in the shipyards of Gdansk, general strikes in England and Germany early this century, and revolutions in China, Russia, and Cuba.

            A. K. Sen supports this general line of argument in "Rational Fools" when he holds that no social organization could maintain itself only through the working of private economic rationality.  "Rational economic man is very close to a social moron."  Rather, economic incentives must be accompanied by moral principle, political and personal commitments, and altruistic motivations.

These examples have a common point:  Social groups typically find it easier to achieve cooperative action and collective goods than the model of economic rationality would predict.  At the very least this would seem to show that the theory of economic rationality does not provide a good basis for explaining cooperative behavior and collective action; social science would be well(c)served with a conception of practical rationality which gives appropriate scope to the sorts of factors which "do• explain the occurrence of cooperative behavior and collective action.

 

Possible remedies.

 

(a)        Some recent work in applied game theory has accepted the concept of

economic rationality, but has tried to show that more careful analysis of

certain kinds of "games" shows that cooperation is the individually rational

strategy.  Robert Axelrod provides an important example of this approach in

"The Evolution of Cooperation•.  He considers the problem of repeated

prisoners' dilemmas (which models a large class of problems of collective

action) and asks what the best strategy is from the point of view of each

individual's longterm private interests.  On any given play of the

prisoners' dilemma "defect" is the best private strategy.  In repeated play

Axelrod shows that "Tit for Tat" is the best strategy over repeated

encounters of repeated prisoners' dilemmas with a variety of competing

strategies.  (According to Tit for Tat, the player opens with "cooperate"

and then reciprocates his opponent's previous play.  If the opponent defects

on play N, then the player defects on play N+1.)  This is a rather nice

result, since it shows that cooperative behavior can emerge spontaneously

through individually rational behavior in the right circumstances.  Michael

Taylor arrives at similar results through analysis of premodern social

groups and the mechanisms through which such groups secure collective goods.

 

(b)       A second, and more general, approach, is one which attempts to modify

the conception of rationality so as to introduce a mechanism for moral

deliberation.  Is it possible to describe a conception of practical reason

which incorporates maximizing rationality and moral considerations?  The

goal would be to arrive at a conception of rationality which contains

economic prudence as a special case; provides a mechanism of deliberation

within which moral constraints and reasons have scope;  and is sufficiently

precise to permit some degree of reasoning about rational decision making.

 

(1)       Some philosophers and rational choice theorists have argued that the

postulate of economic rationality may be modified with the addition of a

motive supporting conditional fairness or reciprocity (Elster, Sen,

Harsanyi, Margolis).  The decision rule on this approach might look

something like this:  Maximize private interests unless the proposed action

involves a public goods problems; in that case, estimate the likelihood of

enough cooperative behavior to secure the collective good; if success is

likely, cooperate; otherwise defect.

 

IF the action does not affect a public good

            THEN maximize private welfare.

            OTHERWISE estimate the likelihood of success of cooperative/

                                    collective action;

            IF success is sufficiently probable,

                                    THEN cooperate;

                                    OTHERWISE defect.

 

This rule does not embody strict moral principle or altruism, but rather

conditional reciprocity.  It represents a limited willingness to contribute

to public goods, but only on the assumption that there is a realistic

prospect that collective action will succeed. 

            The welfare(c)maximizing rule was flawed by being predisposed against

cooperative action.  This rule is flawed in the opposite direction, however.

It doesn't correctly describe voters' behavior or collective action either,

since it would predict virtually universal contribution to public goods.

The problem is that cooperative behavior preempts private striving on this

rule whenever cooperation is sufficiently likely in others.  What the rule

appears to need, therefore, is a weighting function which permits the

decision maker to balance private costs and benefits against the gains of

cooperative behavior and the prima facie presumption in favor of fairness or

reciprocity.  Let's see if we can incorporate this feature into the model.

 

IF the action does not affect a public good

            THEN maximize private welfare.

 

            OTHERWISE estimate the likelihood of success of cooperative/

                                    collective action;

                        IF success is sufficiently probable,

THEN assign an index V to the value of the collective good

and cooperation in this case and compare with the cost

of cooperation C.

                                                IF V > C

                                                            THEN cooperate;

                                                            OTHERWISE defect.

                                    OTHERWISE defect.

 

This rule begs an important question: how to compare private costs and the

value of cooperation and the collective good.  It is consequently a less

rigorously applicable rule than the simpler welfare(c)maximizing rule.  At the

same time, it would appear to correspond more faithfully to the decision-making procedure which real persons undergo when deliberating about collective projects.

 

(2)       Finally, we might briefly consider efforts to construct a conception of

practical reason which begins from a different premise altogether: not

welfare(c)maximizing, but rather the concept of the person.  Here I'm

thinking, obviously, of Nagel's "Possibility of Altruism•.  For Nagel argues

that the first requirement of practical reason is that we recognize the

reality of other persons.  To recognize another person's reality is to

recognize the reality of his or her interests.  And to recognize the reality

of an interest is to recognize that I have a prima facie reason to try to

satisfy that interest.  Therefore I have a prima facie reason to act

altruistically; or in other words, practical reason embodies a rational

impulse towards altruistic action.  (It might be noted that this account

would appear to be entirely consistent with a welfare(c)maximizing decisionÜf______Ü

rule, with the nonstandard assumption that the welfares to be considered

include the interests of other persons.  Thus Nagel's position appears to

have a utilitarian implication.)

            Another example of this sort of effort would be an attempt to embody

moral rules directly within the decision rule.  (Nozick seems to have such a

view in "Anarchy, State, and Utopia• when he speaks of "side(c)constraint

theories.")  Such a theory would presumably have a welfare(c)maximizing

component, so that the theory would contain economic rationality as a

special case (i.e., in circumstances where moral principles and requirements

of fairness have no application).  But the rule would have an overriding set

of moral constraints which limit choice of actions according to their

consistency with a list of moral rules.

            This sort of rule might be modeled along these lines:

 

LIST alternative actions in order of the (private/general) welfare they generate.

ELIMINATE actions which violate specified moral rules.

CHOOSE the highest ranking alternative which survives the filter.

 

This rule becomes either "morally constrained egoism" or "morally

constrained altruism" depending on how welfare is understood; either as

self-interest, or as interests of all affected persons.

 

Implications for the moral economy debate.

 

            Let us briefly take stock of these issues for the theoretical debate I

briefly alluded to above.  The debate is often cast as one between

rationality and community, as if the two were polar cases.  Many of the

considerations identified above, however, tend to cast doubt on that way of

characterizing the dispute.  First, the Axelrod(c)Taylor efforts to show that

cooperation and public goods can emerge even from strict economic

rationality have direct relevance to the circumstances of the traditional

village.  Axelrod's account, for example, shows that cooperation can develop

if relations between players are sufficiently stable to give each player

reason to expect continuing interactions with the others, and sufficiently

familiar to permit the players to reidentify each other and keep track of

other players' history of cooperation.  The traditional village satisfies

both these conditions; relations are face(c)to(c)face over an extended period of

time, and villagers have notoriously long memories for failures to

cooperate.  Likewise, Taylor identifies a number of informal mechanisms

through which village societies can change the payoffs to players in such a

way as to elicit contributions to collective goods.  Thus non(c)contributors

may be shamed or ridiculed, making it rational to contribute.  These

considerations would suggest that the village "moral economy" is backed by

social arrangements which make it rational for villagers to cooperate, even

on a narrow conception of rationality.

            Second, to the extent that criticisms of narrow economic rationality

lead us to a broader conception of practical reason(c)(c)one which incorporates

motives favoring reciprocity, fairness, or altruism(c)(c)patterns of village

behavior which appear to be communal or traditional may be recast as the

natural consequence of rational individual behavior.  When a villager makes

a voluntary contribution of labor services to a public water control-

project, for example, we now have an alternative to the traditionalist's

explanation ("the villager's action was blindly subordinated to traditional

norms"): namely, the villager recognizes the public good to be gained by

collective water control projects and judges that fairness requires that he

make a contribution.  On this account, the villager's behavior is both

rational and cooperative, and village society is both to some degree

communal and based on rational individual decision(c)making constrained by

existing village practices and institutions.

 

References

 

            Axelrod, Robert.  "The Evolution of Cooperation•.  New York: Basic, 1984.

            Barry, Brian.  "Sociologists, Economists and Democracy•.  Chicago:

University of Chicago Press, 1978.

            Buchanan, Allen.  "Revolutionary Motivation and Rationality."

"Philosophy & Public Affairs• 9:1 (1979).

            Elster, Jon.  "Ulysses and the Sirens•.  Cambridge: Cambridge University

Press, 1979.

            Hardin, Russell.  "Collective Action•.  Baltimore: The Johns Hopkins

University Press, 1982.

            Harsanyi, John C.  "Essays on Ethics, Social Behavior, and Scientific

Explanation•.  Dordrecht: D. Reidel, 1976.

            Margolis, Howard.  "Selfishness, Rationality, & Altruism A Theory of

Social Choice•.  Chicago: University of Chicago, 1982.

            Nagel, Thomas.  "The Possibility of Altruism•.  Oxford: Oxford University

Press, 1971.

            Olson, Mancur.  "The Logic of Collective Action: Public Goods and the

Theory of Groups•.  Cambridge: Harvard University Press, 1965.

            Popkin, Samuel.  "The Rational Peasant•.  Berkeley: University of

California, 1979.

            Rapoport, Anatol.  "Two-Person Game Theory•.  Ann Arbor: University of

Michigan, 1966.

            Rapoport, Anatol and Chammah, A. M.  "Prisoners' Dilemma: A Study in

Conflict and Cooperation•.  Ann Arbor: University of Michigan, 1965.

            Rawls, John.  "A Theory of Justice•.  Cambridge: Harvard University

Press, 1971.

            Regan, Donald.  "Utilitarianism and Co-operation•.  Oxford: Oxford

University Press, 1980.

            Schelling, Thomas.  "Micromotives and Macrobehavior•.  New York: Norton,

1978.

            Scott, James.  "The Moral Economy of the Peasant•.  New Haven: Yale

University Press, 1976.

            Scott, James.  "Weapons of the Weak• "Everyday Forms of Peasant

Resistance•.  New Haven: Yale University Press, 1986.

            Sen, Amartya.  "Choice, Welfare and Measurement•.  Cambridge: MIT Press,

1982.

            Taylor, Michael.  "Anarchy and Cooperation•.  London: Wiley, 1976.

            Taylor, Michael.  "Community, Anarchy and Liberty•.  Cambridge: Cambridge

University Press, 1982.

 

 


Altruism and fairness

            A central criticism of the narrow conception of economic rationality concerns the implicit egoism that is generally associated with this framework.  It should be emphasized that this position does not deprecate the importance of rationality in explaining human activity, or elevate emotion or impulse to a dominant position in our understanding of action.  Rather, the critics we will consider argue that human beings are (imperfectly) rational, but that rationality does not consist simply in maximizing one's private interests.

            I begin with A. K. Sen's critique of narrow economic rationality.  Sen criticizes the assumption of pure self‑interest which is contained in the standard conception (Sen 1982:84‑90).  “The purely economic man is indeed close to being a social moron” (Sen 1982:99).  Against the assumption of self‑interested maximizing decision‑making, Sen argues for a proposal for a more structured concept of practical reason: one which permits the decision maker to take account of commitments.  This concept covers a variety of non‑welfare features of reasoning, but moral principle (fairness and reciprocity) and altruistic concern for the welfare of others are central among these.  “Commitment is, of course, closely connected with one's morals.  But moral this question is in a very broad sense, covering a variety of influences from religious to political” (Sen 1982:93).  Sen believes that the role of commitment is centrally important in the analysis of individuals' behavior with regard to public goods.  For example, he suggests that the voters' paradox may be explained by assuming that “voters are not trying to maximize expected utility, but . . . to record one's true preference” (Sen 1982:97).  And he draws connections between the role of commitment and work motivation.  “To run an organization entirely on incentives to personal gain is pretty much a hopeless task” (Sen 1982:98).  He argues, therefore, that in order to understand different areas of rational behavior it is necessary to consider both utility‑maximizing decision making and rational conduct influenced by commitment; and it is an empirical question whether one factor or the other is predominant in a particular range of behavior (Sen 1982:104).  Sen suggests the outline of an alternative decision-making rule.  On this account the agent is postulated to embody two sorts of preference rankings: a ranking of first-order preferences and a meta‑ranking of preference rankings in terms of the relation “ranking x is more moral than ranking y” (Sen 1982:100‑101). 

            Sen's arguments show that there are good analytic and empirical reasons for judging that much actual human behavior is not explicable on the basis of a simple utility‑maximizing scheme.  This finding might lead us to suppose that human beings are typically not rational, or it might lead us to question the concept of rationality associated with the standard conception.  Sen suggests the latter course, and proposes that we attempt to construct a more structured concept of practical reason which permits us to take account of moral, political, and personal commitments as well as concern for welfare.

            Various moral philosophers have argued that practical rationality is more comprehensive than narrow economic rationality.  Thus Thomas Nagel provides a series of arguments to the effect that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests (Nagel 1971). Nagel's The Possibility of Altruism provides one possible way of developing a more comprehensive theory of practical reason.  Nagel's work is an extended essay on the logical character and requirements of practical reason.  In particular, Nagel argues that once we have correctly analyzed the requirements of reason, we will discover that simple egoism is irrational.  “Just as there are rational requirements on thought, there are rational requirements on action, and altruism is one of them” (Nagel 1970:3).  This argument thus represents a direct challenge to the assumptions of standard conception of rationality, and a defense of the view that practical rationality has a substantially more complex structure than simple utility‑maximizing models allow.

            Nagel begins with a conception of rationality which is substantially different from the standard conception of rationality: being rational is acting for good reasons.  “We may explain what a man does by referring to his reasons. . . .  A consideration can operate as a motivating reason only if it has, or is thought to have, the status of a reason in the system of normative principles by which individuals govern their conduct.  Such normative principles therefore specify significant features of the motivational structure” (Nagel 1970:15).  And Nagel believes that once we work out the implications of this conception of “acting on the basis of reasons”, we will discover that certain consequences follow for the actor.  In particular, it follows that the actor will act out of regard for the interests of others; he will act altruistically.  “The general thesis to be defended concerning altruism is that one has a direct reason to promote the interests of others—a reason which does not depend on intermediate factors such as one's own interests or one's antecedent sentiments of sympathy and benevolence” (Nagel 1970:16).

            It should be noted that Nagel's concept of altruism is one which holds that the rational agent will give practical weight to the interests of others as well as to his own interests.  “By altruism I mean not abject self‑sacrifice, but merely a willingness to act in consideration of the interests of other persons, without the need of ulterior motives” (Nagel 1970:79).  For our purposes Nagel's arguments suggest one important conclusion: the standard model of economic rationality cannot serve as a full theory of practical reason.  Instead, it is necessary to conceive of the rational decision maker as being subject to a variety of moral constraints which modulate behavior and which sometimes restrict self‑seeking actions.

            Thus Nagel intends to establish that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests (Nagel 1971).  This line of thought directly addresses the egoism assumption of the standard conception.  It does not, however, do quite enough for us; for it does not give us a way of incorporating the idea of moral principles (or other normative requirements) into the decision making process.

            Finally, a number of philosophers have attempted to incorporate the idea of fairness into the concept of rational decision making.  A reason for my performing an act is that I benefit from widespread performance of this sort of act, and I recognize that fairness requires that I pay my share of the cost of these public benefits.  John Rawls's A Theory of Justice represents an extended argument to the effect that there are principles of justice which ought to regulate the just society, and these principles derive from the principle of fairness.  Rawls's construction is at some distance from our primary concern, since he is concerned with global features of justice, while we are concerned with individual rationality.  But the kernel of Rawls's construction is relevant here: if individual rationality involves evaluating alternative lines of action in terms not only of the costs and benefits of each alternative, but also in terms of the fairness of each alternative, then we have arrived at a structured concept of rationality.  And it is a concept which involves the imposition of side constraints on the decision making process.  A more structured decision making process is necessary in order to take account of the role of principle in decision making: the decision maker can combine a set of side constraints on action, as well as a set of goals.  And the decisions he arrives at will be a complex function of constraints and goal‑maximizing actions.

            How do these findings relate to our central concerns?  First, they suggest that the narrow conception of economic rationality is an inadequate conception of practical reason because it makes it impossible to take account of certain features of the decision making process which are intuitively crucial.  Further, these considerations suggest an alternative model of the decision making process which promises to be a more adequate analysis of the concept of human rationality.  Moreover, this richer conception of practical reason promises to offer a new set of solutions to different classes of collective action problems:  If individuals are altruistic to some degree (that is, responsive to the interests of others), and if they are principled (that is, moved by considerations of fairness, reciprocity, or justice), then they will be practically motivated to act differently when confronted with occasions for collective action than the collective action theorem predicts.

            These findings have direct import for the applicability of rational-choice models in social explanations.  For example, recall the discussion in chapter 2 of freeriding and public goods problems.  Once we consider a more complex theory of practical deliberation, formal arguments predicting the emergence of public goods problems in real social groups will be found to be misleading.  On a more complex, and more empirically adequate, account of practical reason, altruism, cooperation, and reciprocity are rational choices; therefore we would expect a social group consisting of rational individuals to show marks of cooperation and altruism. 

            We must be careful not to draw an overly strong conclusion, however; for no one maintain that human beings are indifferent to private welfare.  Indeed, generally speaking it would seem reasonable to assume that each decision-maker places high priority on personal and familial welfare; human beings generally do not behave like impartial utilitarians.  This finding suggests that human behavior is the resultant of several different forms of motive: self‑interest and altruism; and several different types of decision making processes: maximizing and side-constraint testing.  And to the degree that self‑interest and maximizing behavior are prominent in a particular type of circumstance, to that degree the collective action theorem will be empirically significant.  In village society one presumably will find a mix of these four possibilities; consequently it is invalid to directly apply the collective action theorem.  Rather, it is necessary to offer empirical arguments and analysis to show in what circumstances the premises of the collective action theorem are most fully satisfied, and in which as a consequence we would most confidently predict collective action problems.  But in chapter 2 we saw a variety of features of community life which plausibly work to enhance both the impulse to altruism and the impulse towards fairness and reciprocity; and these considerations in turn suggest that some communities in some circumstances may manage to effectively overcome collective action problems. 

            These findings concerning the adequacy of the concept of economic rationality have immediate consequences for Popkin's case against the moral economists.  For Sen and Nagel give reasons for holding that moral commitments and values are a constitutive part of human rationality.  In the context of village society we must ask whether there are moral norms which might affect individual action; and the answer to this question appears to be unavoidably affirmative.  Scott's arguments to this effect are particularly strong (1976:13‑56); but almost any study of traditional society would do just as well.  And the arguments advanced by Sen and Nagel suggest a way of incorporating adherence to moral standards (e.g., reciprocity, fairness, or unwillingness to be a freerider) into the process of rational decision making.  Thus it is consistent to hold that rational peasants may modulate their conduct out of regard for various forms of traditional morality.  These arguments therefore cast doubt on Popkin's assumption that the rational peasant is exclusively motivated by considerations of familial or personal welfare, and they undermine Popkin's skepticism about the motivational efficacy of moral principles and village assumptions about fairness.  If villagers are altruistic in the narrow sense of being affected by their perceptions of the needs and welfare of others; and if villagers' decision making processes are importantly constrained by simple moral constraints like fairness, reciprocity, and the like; then the collective action theorem does not validly apply to village society because village society does not satisfy its premises.

            Two important criticisms of the standard conception of economic rationality emerge from this discussion.  First, the standard conception has historically been inclined towards an assumption of egoistic self‑interest; and in fact prisoners' dilemma situations, game theoretical reasoning, and the like depend upon this assumption.  However, it is possible to introduce the assumption that decision makers take the interests or welfare of others into account (with or without a discount function) and preserve the main outlines of narrow economic rationality.  More fundamentally, we found that telling objections have been put forward the maximizing structure of narrow economic rationality; philosophers and economists alike have argued that a full theory of rationality needs to give a place for the operation of moral principle in the decision making process.

 

Summary

            Let us now take stock from this excursion into recent moral philosophy.  What consequences, if any, follow from these considerations for the concept of rationality?  We saw in our discussion of Sen's work that there are good analytical and empirical reasons for judging that much actual human behavior is not explicable on the basis of a simple utility‑maximizing scheme.  This finding might lead us to suppose that human beings are typically not rational; or it might lead us to question the concept of rationality associated with SER.  Sen suggests the latter course, and proposes that we attempt to construct a more structured concept of practical reason which permits us to take account of moral, political, and personal commitments as well as concern for welfare.  Moreover, Sen shows that the former cannot be subsumed under the simple concepts of welfare‑maximizing or preference rank‑ordering.

            Next, we saw in Nagel's work a series of arguments to the effect that rationality requires altruism: recognition of the reality of the interests of others and a direct willingness to act out of regard for those interests.  This line of thought directly addresses the egoism assumption of SER.  It does not, however, do quite enough for us; for it does not give us a way of incorporating the idea of moral principles (or other normative requirements) into the decision making process.

            Finally, we saw that other recent moral philosophers have outlined the sort of structured decision making process necessary in order to take account of the role of principle in decision-making: the decision maker can combine a set of side constraints on action (normative commitments, in Sen's terms), as well as a set of goals (personal interest, social goals, the welfare of others, etc.).  And the decisions he arrives at will be a complex function of constraints and goal‑maximizing actions.

            How do these findings relate to our central concerns?  First, they offer both a set of criticisms of SER (SER is an inadequate conception of practical reason because it makes it impossible to take account of certain features of the decision making process which are intuitively crucial), and an alternative model of the decision making process which promises to be a more adequate analysis of the concept of human rationality.  Moreover, this richer conception of practical reason promises to offer a new set of solutions to different classes of collective action problems: if individuals are altruistic to some degree (that is, responsive to the interests of others), and if they are principled (that is, moved by considerations of fairness, reciprocity, or justice), then they will be practically motivated to act differently when confronted with occasions for collective action than the CAT predicts.       A second, and more general, approach, is one which attempts to modify the conception of rationality so as to introduce a mechanism for moral deliberation.  Is it possible to describe a conception of practical reason which incorporates maximizing rationality and moral considerations?  The goal would be to arrive at a conception of rationality which contains economic prudence as a special case; provides a mechanism of deliberation within which moral constraints and reasons have scope;  and is sufficiently precise to permit some degree of reasoning about rational decision making.

 

Time preferences

 

            This conclusion allows us to infer that agents do in fact discount future utilities.  Why should we expect there to be a positive interest rate?  Perhaps this circumstance merely reflects people's (irrational) propensity to discount future utilities.  Upon inspection, however, it is apparent that interest rates do not merely reflect intertemporal preferences on the part of agents; they also reflect processes of real growth.  By investing a quantity of resources productively today it is possible to create a larger sum of resources tomorrow.  And this growth rate is a feature of current technology--not a function of preferences.  Even if there were no credit market (and therefore no positive interest rate), it would still be true that it is possible to transform a quantity of resources today into a larger quantity tomorrow.  If tomorrow's utility is not discounted, then there is no upper bound on the proportion of present income the rational agent should put aside into productive ventures.  Given that utility is an increasing function of resources, the utility I get from $100 worth of consumption today is less than that I would get from $110; there are production possibilities that will permit me to turn today's $100 into resources worth $110; if, by assumption, I do not discount future utilities, then the utility maximizer should always choose to make the productive investment.

            Suppose that the real growth rate is g: a sum of resources m invested productively today will result in a value of (1+g)*m resources in the next period.  What is the relation between the three rates i (interest), r (utility discount), and g (real growth rate)?  In a competitive market the interest rate will be less than the growth rate, reflecting the entrepreneurial gains available to producers.  (If i = g, then no entrepreneurial investment would occur.)  Imagine that there are two classes of agents: entrepreneurs and savers/consumers.  Given that g is positive, there will be demand for savings (as entrepreneurs seek out resources for productive investment).  This will induce a positive interest rate less than g.  If r is zero (that is, if savers do not discount future utilities), then any positive interest rate will elicit savings--driving i to zero as well.  If 0<r then there will be an equilibrium rate of interest such that r < i < g.

            We found above that actuarial discounting is not exponential.  Interest compounding and discounting, by contrast, are exponential.  And for plausible values of the parameters the two curves cross.  What implications does this finding have for savings behavior?  It implies that saving will cease for those periods in which the effective utility discount is greater than the interest rate.

            Consider first a problem having to do with the structure of preferences over time: how should the agent take account of future benefits in deliberating about actions today?  Many of our choices involve a distribution of benefits to ourselves over the present and future.  The act of saving is a choice in which the agent defers consumption today for the sake of future consumption; implicitly, then, the agent is making a comparison between the utility of today's consumption versus the utility today of the future utility produced by the savings.  Suppose that the act of saving today leads to a net loss of one unit of utility, while tomorrow's enjoyment of the savings produces a positive benefit of 1.1 units.  Is there a rational base for deciding what to do in such a case?  Elster raises two problems about time preferences: first, whether it is rational at all for an agent's choices to embody pure time preference (that is, to give extra weight to present utilities over future utilities), and second, whether there are consistency conditions that can be imposed on time preference. 

 

            It appears, however, that this account understates the role of uncertainty.  Constructing a lifeplan involves adopting complex strategies in which various circumstances must occur simultaneously if the plan is to come to fruition.  And the more remote in time the moment of enjoyment, the more likely it is that one or another of these unforeseen obstacles will arise--giving me a new reason for discounting the future.

            The problem of inconsistent time preferences arises in the context of planning the expenditure of utility-producing resources over time, using a formula that leads the agent to change plans repeatedly from one time period to the next.  Suppose the agent has a sum of money M to distribute over n time periods, and that the utility function is the same in each period.  An agent with no time preferences would arrive at the following distribution: m1=M/n, m2=M/n, ... mn=M/n.  The agent with exponential time preferences and discount rate r will select this distribution:

 

            Consider a simple numerical example.  Imagine that the agent is deliberating about how to use $1000 over a period of five years which can either be consumed or saved for future consumption with accumulated interest at 10%.  (Utility is based on the Cobb-Douglas function described below reflecting the diminishing marginal utility of income.)

period  1                      2                      3                      4                      5

future value:

income $1000              $1100              1210                1331                1464

utility               3.16                 3.31                 3.48                 3.64                 3.83

 

Present discounted value:

                        1000                1000                1000                1000                1000

                        3.16                 2.87                 2.74                 2.61                 2.49

 

The present value of each sum is $1000, since each sum is discounted by the rate of interest and the number of periods that will elapse.  What is the present utility of $1464 five years hence, however?  If we say that it is the utility of a present consumption of $1000, then we have discounted future utilities, but have done so at a slower rate than we discount income.  If we discount at the interest rate, then the present utility of the future income is 2.49--lower than the present utility of $1000.

 

 

Social science theories of rationality

 

            A number of social scientists and philosophers deny that the concept of individual rationality can be used to explain behavior in a wide range of societies and cultures.  Instead, it is held that the concept of individual rationality is specific to modern western cultures; rationality is parochial and culturally relative.  What is involved in this position?  There are several versions:

 

*          Modes of behavior and motivational structures are highly culture-relative and variable.  The model of means-end rationality is merely one culturally specific mode of behavior, characteristic of western cultures.

*          People aren't in general rational, deliberative, or calculating.  The model of means-end rationality is culturally specific to modern western culture.

*          People are deliberative and are in a general way sensitive to means-end considerations.  But there are wide variations in the ways in which means-end deliberation is embodied in different cultures and historical periods.  So the thin theory does not help us understand people's behavior.  It does describe their behavior; but their choices are highly sensitive to the culturally specific implementation.

            We may refer to the first view as the radical relativist view; while the second is the substantivist-relativist view.  On the latter approach, the dispute is not properly over the role of the assumption of rationality and deliberativeness, but rather the thickness of the description of rationality and social institutions in the context of explanation.  The account of rational deliberation may be more or less detailed--taking into account only utilities or preference rankings, taking into account normative constraints, and taking into account culturally specific worldviews and beliefs.  And the account of the environment of choice may be more or less concrete.

            The substantivist's argument against the economists and the rational choice model may now be construed in this way: the thin theory of rationality is too thin to properly describe rational deliberation in most social circumstances.  (This appears to be James Scott's view in The Moral Economy of the Peasant.)  And the thin theory of the environment of choice--a structure of costs and benefits of various actions--is overly abstract; it is necessary to provide a more detailed description of the political, social, and economic structures and institutions within which individuals deliberate.  Institutions impose a set of opportunities, powers, and constraints on actors in the context of which they deliberate.

 

Explanation without relativism

            What explanatory models must be given up if we accept the radical relativist view?  First, most obviously, all forms of economic analysis--classical, neoclassical, Marxist, or general equilibrium theory--depend on the assumption of individual rationality.  If we conclude that Chinese peasants are not rational calculators sensitive to cost and benefit of alternative decisions, then we must foreswear explanations based on predicted equilibria--equalization of wage rates, correlations among price phenomena in linked markets, etc.

            Second, theories of organization and bureaucracy depend on the assumption that the behavior of individuals within organizations can be shaped through incentives and disincentives--again, an assumption that cannot be sustained unless we postulate that individuals make calculating decisions based on cost and benefit.

            Third, explanations of political behavior based on material interests, individual self-interest, or class interests depend on the assumption that individuals choose actions in an effort to further their interests--once again, the postulate of individual rationality.  Without this postulate, great swaths of analysis of peasant politics, both Marxist and non-Marxist, fall by the wayside.  Thus both public choice theory and Marxist political theory depends on the assumption of individual rationality.

            The discipline of human geography likewise depends on the assumption that individuals make deliberative choices about their place of residence; G. William Skinner's analysis of market hierarchies collapses if we reject the assumption of individual rationality.

            Giving up the assumption of individual rationality thus comes perilously close to giving up on the use of theoretical models in social science altogether, since the vast majority of such models presuppose indidivual rationality.  This is an extremely high cost; for what sort of social analysis can be performed without the assumption of individual rationality?  There are at least two possible alternatives.  First, we may do as Geertz seems to recommend: construct our interpretations of social settings as if they were sui generis.  We might start from scratch and attempt to discover the basis of motivation and behavior anew in every distinct culture (or even sub-culture).

            And second, we might aspire to a different sort of generalization about human agency--e.g. a general theory of social psychology, a generalization about the ubiquity of symbolic behavior, or a Freudian theory of unconscious motivation.  This approach, however, is at least as vulnerable to the charge of ethnocentrism as the rational-choice approach; so I will consider it no further.

 

Where does the debate arise?

            In what context does the debate over rationality arise?  First and foremost it arises in cases where the social scientist or historian is faced with the problem of explaining a pattern of individual or group action in a particular social setting.  Here the basic question is this:  What assumptions is it legitimate to make about the motives and deliberations of individuals in order to explain their behavior?  A second context involves the effort to explain the characteristics of a set of social arrangements--e.g. the placement of villages in the Chinese countryside, the cell structure of a revolutionary organization, or the slowness of a bureaucracy to adapt itself to new problems.  Here the question is:  What features of individual activity, within a specific social and natural environment, lead them to shape their institutions and arrangements in the particular ways that are to be observed?

            Are there important social-science research tasks for which assumptions of individual rationality are not appropriate or demanded?  Certainly there are; much of the terrain covered by interpretation theory, cultural anthropology, and ethnography involves the non-rational significance of individual and social activity.

            But note that there are many anthropologists who do attempt to explain cultural arrangements as the result of economic circumstance--e.g. Pasternak or the book I am now reviewing.


I. What is the thin theory of rationality?

 

            The claim made for rational-choice theory:

 

*          If we postulate the thin theory of rationality and work out (at an acceptable level of description) (i) the beliefs and goals of participants and (ii) the social and natural environment of choice, then it will be possible to (a) explain patterns of individual behavior in this context, and (b) provide aggregative explanations of collective behavior in this context.

 

II. Qualifications: norms and values, class interests, more or less materialist description of goals, critique of economic rationality.

 

III. Arguments against the rational choice approach.

 

IV. Replies.

 

Practical vs. belief rationality

 

            The notion of rationality involves several distinct ideas--what we may call belief rationality and practical rationality.  The first concerns the standards by which a person comes to accept beliefs about the world, while the second has to do with the way a person decides what action to perform.  The claim of cultural relativism of rationality has been applied to both; but in this paper I am concerned with practical rationality, not belief rationality.  On this account, the investigator should take beliefs as fixed and explain behavior as a prudent effort to achieve goals given beliefs.  It is possible, for example, that a given culture may attribute causal powers to local gods; in which case it is prudent to appease the gods through appropriate ritual performances.


            This topic raises several questions.  First, is the concept of individual rationality applicable across cultures, or is it specific to modern western European-based cultures?  And second, is it possible or legitimate to attempt to explain social phenomena in non-western cultures on the basis of rational choice theory?

 

A series of positions:

 

1. Economic rationality based on egoistic self-interest and thin description of institutions.

2. Individual rationality as prudence, based on a broader conception of interests; thin institutions.

3. Individual rationality with broad interests and thick interests.

4. Deliberative goal-directed agency within a system of norms, values, and worldviews.

5. Culturally unique agency.

 

Too abstract a description of agency

Too little attention to norms and values

Ethnocentric notion of rationality.

            The problem of individual rationality in area studies is a difficult one.  On the one hand, some RC advocates offer an overly narrow conception of rationality: egoistic, economic, and sparse.  On the other hand the critics go too far in rejecting the concept of individual rationality.

 

1. Interests and beliefs cannot be specified in a culture-independent way.

2. The model of calculating rationality is itself culturally specific.  There are radical differences in agency across culture.  Tradition-driven behavior is a subvariant of this position.

3. The RC model does not permit a proper treatment of norms and values in agency.

4. RC models depend on overly schematic specifications of agency and structure.

            a. These models implicitly project anachronistic institutional arrangements onto non-western societies--e.g. exchange relations or market institutions.

 

The role of theories of rationality in social science

 

            Explanations in social science often refer to the rational actions of individuals.  Before considering the logic of such explanations in detail, let us ask a preliminary question:  What is the intended import of various theories of rationality?  Several possibilities suggest themselves.  First, the theory of rationality may be a normative discipline which is intended to establish universal rules of reasoning in support of belief and action.  (Call this the "pure theory" of rationality.)  On the side of belief this discipline would include deductive and inductive logic, probability theory, and (perhaps) applied philosophy of science.[7]  These disciplines are normative in that they attempt to establish criteria of sound inference; to be rational is to reason in accordance with these criteria.  On the side of action the normative theory of rationality would include an account of the rules of reasoning which "ought" to allow a decision maker to consider his beliefs, goals, and (perhaps) values, and arrive at the optimal action in the circumstances.  Partial efforts in this direction include the disciplines of decision theory and game theory; one might also include various theories of applied ethics as potential candidates in this part of the theory of reason.[8]  Overall, the goal on this approach is to arrive at a decision procedure for action which generates the "right" decisions on the basis of a given range of information about goals and beliefs.

            It should be noted that on this account the theory of rationality is sufficiently idealized that it bears little relation to actual human decision making.  For symbolic logic, decision theory, and game theory abstract from limits on computational power, information access, memory, and the like.  These abstractions entail that ordinary human reasoners cannot fully embody these models of rational calculation.  For example, even two-person games for which game theory shows that solutions exist are too complex for anyone but a mathematical economist with ample computational resources to solve.  This finding shows that human reasoners do not solve strategic decision problems using two-person game theory.[9]

            A rather different conception of the import of a theory of rationality is one according to which the goal is to provide an empirically grounded theory of deliberative human action.  (Call this the "cognitive theory" of rationality.)  On this account, the theory of rationality is a branch of cognitive psychology.  One might pose this question:  How do human beings arrive at beliefs and make decisions under optimal circumstances (i.e., ample time for decision making, low level of emotional influences on the process of deliberation, and adequate informational resources)?  On this account we would conceive of the theory of rationality much as Chomsky conceives of the theory of grammar: as an idealized representation of psychologically real processes and protocols.[10]  On this approach it is an open question whether there is one form of deliberative decision making or many.  It is logically possible, for example, that different cultures have developed different structures of reasoning which are transmitted through education from one generation to the next.

            If after appropriate empirical research we were to conclude that there are important features in common among all cultures in terms of deliberative reasoning (as Chomsky argues there are features in common among all natural languages), then the theory of rationality would have important application in social science.  One would be able to construct explanations of various social patterns on the basis of analysis of the decision-making procedures of individuals within the environments of choice which their societies pose for them.

            What are the relations between these different sorts of theories?   Consider first the relation between the pure theory and the cognitive theory.  One possible view is that the formal discipline is in fact a highly idealized representation of what the empirical discipline would uncover after full investigation.  Formal logic correctly identifies the rules of deductive inference, and human beings (through favorable evolution, possibly) embody these rules in their ordinary processes of reasoning.  Likewise, decision theory and game theory establish the correct principles of reasoning in circumstances of uncertainty and competition, and human beings imperfectly embody these rules.  The political economy approach to social science appears to presuppose something like this.[11]

             The relation between the cognitive theory and the means-end theory might go along these lines.  The means-end theory presupposes that agents embody some cognitive process of deliberation or other, but it tacitly presupposes that this process is a relatively simple one.  However, whatever the findings of the empirical discipline of deliberative reasoning, the theory of rational action will simply incorporate the resulting theory into its attempts to explain behavior as the calculated effort to achieve goals on the basis of beliefs.

            Finally, what is the relation between the interpretive theory and other theories of rationality?  Its proponents view the interpretive theory as the most comprehensive account of intelligible human action among these alternatives, and an account within which other views of rationality constitute special cases.  For much human action--in our culture and others'--may be understood as calculated to maximize the attainment of one's ends, given one's beliefs.  Therefore means-end explanations are often the correct way of understanding the agent's behavior.  But other forms of action do not conform to this model; therefore interpretation theory requires that we construct other models of action within the context of which to understand these types of action.

            Where does the theory of economic rationality lie in this taxonomy?  First, it falls within the formal theory of rationality in its guise as formal decision theory.  Economists and decision theorists are interested in working out the logic of the simple assumption of maximization of utility, and the complications which risk, uncertainty, and strategic interactions introduce into this model.  Second, it plays a role in the empirical theory of deliberation in that some have presented narrow economic rationality as a model of the cognitive process of deliberation.  The argument here is that narrow economic rationality correctly describes a wide range of human behavior, and the best explanation for this fact is that human beings embody a process of calculation which approximates the formal theory of economic rationality.  That is, human beings really act so as to maximize private interests.  The theory of narrow economic rationality also has a central role in the means-end theory of rational action; it is the procedure which common sense recommends for understanding action in terms of goals and beliefs.

            The main assumptions of the rational-choice paradigm are these: human beings are deliberative; they are capable of forming true beliefs about their natural and social environment; and they are concerned about individual and family welfare.  Many criticisms of this model derive from the suspicion that rational-choice analysis makes unreasonable assumptions about individual behavior as excessively calculating and excessively materialistic.

 

            We might ask briefly how public choice theory fits into the taxonomy of theories of rationality provided earlier.  The goal of public choice theory is to analyze the collective or aggregate consequences for a social group as a whole of the rational choices made by individuals within structured circumstances of choice.  At present this field depends upon the narrow conception of economic rationality; but what changes would be forced by incorporating a fuller and more adequate conception of decision making?  Much of the power of public choice theory derives from the simplicity of the assumption of economic rationality; it is quite possible that a substantially more complex theory of rationality would make it impossible to derive formal results in the aggregate.  Taking an extreme case, it would appear unpromising to require a description of the full range of possible motives of individual action and attempt to derive aggregate consequences.  So some degree of abstraction is required in the description of individual decision making.  At the same time, whatever results are derivable from the richer assumption would presumably be more directly applicable to the behavior of existing social groups.

 

References

 

Akerlof, George. 1970. The Market for ‘Lemons’: Quality, Uncertainty and the Market Mechanism. Quarterly Journal of Economics 84:488-500.

 

Alexander, Jeffrey C., and Steven Seidman, eds. 1990. Culture and society: contemporary debates. Cambridge [England] ; New York: Cambridge University Press.

 

Anderson, Elizabeth. 2000. Beyond Homo Economicus. Philosophy & Public Affairs 29 (2).

 

Arrow, Kenneth Joseph. 1963. Social choice and individual values. 2d ed, Cowles Foundation for Research in Economics at Yale University. Monograph 12. New York,: Wiley.

 

Arrow, K.J., and F.H. Hahn. 1971. General Competitive Analysis: San Francisco.

 

Axelrod, Robert M. 1984. The evolution of cooperation. New York: Basic Books.

 

----------. 1997. The complexity of cooperation: agent-based models of competition and collaboration, Princeton studies in complexity. Princeton, N.J.: Princeton University Press.

 

Baden, John, and Douglas S. Noonan, eds. 1998. Managing the commons. 2nd ed. Bloomington, Indiana: Indiana University Press.

 

Baier, Kurt. 1985. Maximization and Fairness. Ethics 96 (1):119-129.

 

Baron, Jonathan. 1985. Rationality and Intelligence. Cambridge: Cambridge University Press.

 

Barry, Brian M., and Russell Hardin, eds. 1982. Rational man and irrational society?: an introduction and sourcebook. Beverly Hills: Sage Publications.

 

Bates, Robert H. 1998. Analytic narratives. Princeton, N.J.: Princeton University Press.

 

Becker, Gary. 1976. The Economic Approach to Human Behavior. Chicago: University of Chicago.

 

----------. 1986. An economic analysis of the family. Dublin, Ireland: Economic and Social Research Institute.

 

Ben-Ner, Avner, and Louis G. Putterman, eds. 1998. Economics, values, and organization. Cambridge ; New York, NY: Cambridge University Press.

 

Block, Peter. 1993. Stewardship: choosing service over self-interest. 1st ed. San Francisco: Berrett-Koehler Publishers.

 

Bonner, John. 1986. Introduction to the Theory of Social Choice. Baltimore: The Johns Hopkins University Press.

 

Brams, Steven J. 1994. Theory of Moves. Cambridge UK: Cambridge University Press.

 

Brams, Steven J., and Alan D. Taylor. 1996. Fair Division: From cake-cutting to dispute resolution. Cambridge UK: Cambridge University Press.

 

Bratman, Michael. 1987. Intention, plans, and practical reason. Cambridge, Mass.: Harvard University Press.

 

Brinton, Mary C., and Victor Nee, eds. 1998. New institutionalism in sociology. New York: Russell Sage Foundation.

 

Brocheux, Pierre. 1983. Moral Economy or Political Economy? The Peasants are Always Rational. Journal of Asian Studies 42 (4):791-803.

 

Buchanan, Allen. 1979. Revolutionary Motivation and Rationality. Philosophy and Public Affairs 9 (1):59-82.

 

Buchanan, James M., and Gordon Tullock. 1962. The calculus of consent, logical foundations of constitutional democracy. Ann Arbor,: University of Michigan Press.

 

Campbell, Donald E. 1990. Intergenerational social choice without the Pareto principle. Journal of Economic Theory 50 (2):414(10).

 

Campbell, Richmond, and Lanning Sowden, eds. 1985. Paradoxes of Rationality and Cooperation: Prisoner’s Dilemma and Newcomb’s Problem. Vancouver: University of British Columbia Press.

 

Cherniak, Christopher. 1986. Minimal Rationality. Cambridge: MIT Press.

 

Chong, Dennis. 1995. Rational Choice Theory’s Mysterious Rivals. Critical Review 9 (1–2):37-58.

 

Coleman, James. 1986. Individual Interests and Collective Action: Studies in Rationality and Social Change. Cambridge: Cambridge University Press.

 

Darwall, Stephen L. 2002. Welfare and rational care, Princeton monographs in philosophy. Princeton, N.J.: Princeton University Press.

 

Dessler, David. 1989. What’s at Stake in the Agent-Structure Debate? International Organization 43 (3):441-473.

 

Eatwell, John, Murray Milgate, and Peter K. Newman, eds. 1989. Game theory: the New Palgrave. 1st American ed. New York: W. W. Norton.

 

Elster, Jon. 1979. Ulysses and the Sirens: studies in rationality and irrationality. Cambridge [Eng.] ; New York: Cambridge University Press.

 

----------. 1982. Belief, Bias and Ideology. In Rationality and Relativism, edited by M. Hollis and S. Lukes.

 

----------. 1982. Sour Grapes--Utilitarianism and the genesis of wants. In Utilitarianism and beyond, edited by A. K. Sen and B. Williams.

 

----------. 1983. Sour Grapes: Studies in the Subversion of Rationality. Cambridge: Cambridge University Press.

 

----------. 1985. Rationality, Morality, and Collective Action. Ethics 96 (1):136-155.

 

----------, ed. 1986. Rational choice, Readings in social and political theory. Washington Square, N.Y.: New York University Press.

 

----------. 1986. The Multiple self, Studies in rationality and social change. Cambridge [Cambridgeshire] ; New York: Cambridge University Press.

 

----------. 1989. Solomonic Judgements: Studies in the Limitations of Rationality. Cambridge: Cambridge University Press.

 

----------. 1993. Political psychology. Cambridge [England] ; New York, NY, USA: Cambridge University Press.

 

----------. 1999. Alchemies of the mind: rationality and the emotions. Cambridge, U.K. ; New York: Cambridge University Press.

 

Elster, Jon, and John E. Roemer, eds. 1991. Interpersonal comparisons of well-being, Studies in rationality and social change. Cambridge [England] ; New York: Cambridge University Press.

 

Elster, Jon, and Rune Slagstad, eds. 1988. Constitutionalism and democracy, Studies in rationality and social change. Cambridge ; New York; Paris: Cambridge University Press ; Maison des sciences de l’homme.

 

Emirbayer, Mustafa, and Ann Mische. 1998. What Is Agency? American Journal of Sociology 103 (4):962-1023.

 

Feeny, David. 1983. The Moral or the Rational Peasant? Competing Hypotheses of Collective Action. Journal of Asian Studies 62 (4):769-789.

 

Ferejohn, John, and Debra Satz. 1995. Unification, Universalism, and Rational Choice Theory. Critical Review 9 (1-2):71-84.

 

Flew, Antony. 1975. Thinking about thinking: or, Do I sincerely want to be right? [Glasgow]: Fontana/Collins.

 

Friedman, Jeffrey, ed. 1995. Critical Review; Special Issue: Rational Choice Theory and Politics. Vol. 9:1-2.

 

Frohlich, Norman. 1974. Self-Interest or Altruism, What Difference? Journal of Conflict Resolution:55-73.

 

Frohlich, Norman, and et al. 1975. Individual Contributions to Collective Goods: Alternative Models. Journal of Conflict Resolution 19:310-329.

 

Frohlich, Norman, Joe Oppenheimer, and Oran R. Young. 1971. Political Leadership and Collective Goods. Princeton: Princeton University Press.

 

Frohlich, Norman, and Joe A. Oppenheimer. 1974. The Carrot and the Stick: Optimal Program Mixes for Entrepreneurial Political Leaders. Public Choice 19:43-61.

 

----------. 1978. Modern Political Economy. Englewood Cliffs, NJ: Prentice-Hall.

 

Gardenfors, Peter, and Nils-Eric Sahlin, eds. 1988. Decision, probability, and utility: selected readings. Cambridge ; New York: Cambridge University Press.

 

Gauthier, David P. 1970. Morality and rational self-interest. Englewood Cliffs, N.J.: Prentice-Hall.

 

Gert, Bernard. 1970. The moral rules; a new rational foundation for morality. [1st ] ed. New York: Harper & Row.

 

Godelier, Maurice. 1972. Rationality and irrationality in economics. London: Nlb.

 

Goldstein, William M., and Robin M. Hogarth, eds. 1997. Research on judgment and decision making: currents, connections, and controversies, Cambridge series on judgment and decision making. Cambridge, U.K. ; New York: Cambridge University Press.

 

Green, Donald P., and Ian Shapiro. 1994. Pathologies of rational choice theory: a critique of applications in political science. New Haven: Yale University Press.

 

Hardin, Garrett. 1968. The Tragedy of the Commons. Science 162:1243-48.

 

Hardin, Russell. 1982. Collective Action. Baltimore: The Johns Hopkins University Press.

 

----------. 1995. One for All: The Logic of Group Conflict. Princeton: Princeton University Press.

 

Harsanyi, John C. 1977. Rational behavior and bargaining equilibrium in games and social situations. Cambridge: Cambridge University Press.

 

----------. 1982. Morality and the Theory of Rational Behavior. In Utilitarianism and Beyond, edited by A. Sen and B. Williams.

 

Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge: Harvard University Press.

 

----------. 1977. The Passions and the Interests: Political Arguments for Capitalism before its Triumph. Princeton: Princeton University Press.

 

----------. 1986. Rival Views of Market Society and Other Recent Essays. Cambridge: Harvard University Press.

 

----------. 1991. The Rhetoric of Reaction: Perversity, Futility, Jeopardy. Cambridge: Harvard University Press.

 

Hollis, Martin, and Edward J. Nell. 1975. Rational economic man; a philosophical critique of neo-classical economics. [London]: Cambridge University Press.

 

Horton, Robin. 1970. African Thought and Western Science. In Rationality, edited by B. Wilson.

 

Jarvie, I. C. 1984. Rationality and Relativism: In Search of a Philosophy and History of Anthropology. London: Routledge.

 

Kahneman, D., P. Slovic, and A. Tversky. 1982. Judgment under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

 

Keyes, Charles et al. 1983. Peasant Strategies in Asian Societies: Moral and Rational Economic Approaches--A Symposium. Journal of Asian Studies XLII (4):753–868.

 

 Keyes, Charles F. 1983. Economic Action and Buddhist Morality in a Thai Village. Journal of Asian Studies XLII (4):851-868.

 

Kieve, Ronald A. 1986. From Necessary Illusion to Rational Choice? A Critique of Neo-Marxist Rational-Choice Theory. Theory and Society 15:557-582.

 

Kiser, E. 1998. The Debate on Historical Sociology: Rational Choice Theory and Its Critics. American Journal of Sociology 104 (3):785-816.

 

Klitgaard, Robert E. 1988. Controlling corruption. Berkeley: University of California Press.

 

Korner, S., ed. 1974. Practical Reason. New Haven: Yale University Press.

 

Levi, Margaret. 1988. Of Rule and Revenue. Berkeley: University of California.

 

----------. 1997. Consent, Dissent, and Patriotism. Cambridge: Cambridge University Press.

 

Lichbach, Mark Irving, and Alan S. Zuckerman, eds. 1997. Comparative politics: rationality, culture, and structure, Cambridge studies in comparative politics. Cambridge, U.K. ; New York, NY, USA: Cambridge University Press.

 

Little, Daniel. 1988. Collective Action and the Traditional Village. Journal of Agricultural Ethics 1 (1):41-58.

 

----------. 1991. Rational-Choice Models and Asian Studies. Journal of Asian Studies 50 (1):35-52.

 

Luce, R. D., and Howard Raiffa. 1958. Games and Decisions. New York: Wiley.

 

Lukes, Steven. 1970. Some Problems about Rationality. In Rationality, edited by B. Wilson.

 

----------. 1982. Relativism in its Place. In Rationality and Relativism, edited by M. Hollis and S. Lukes: in Hollis and Lukes, eds. (1982).

 

MacIntyre, Alasdair C. 1988. Whose justice? Which rationality? Notre Dame, Ind.: University of Notre Dame Press.

 

Mansbridge, Jane J., ed. 1990. Beyond self-interest. Chicago: University of Chicago Press.

 

Margolis, Howard. 1982. Selfishness, altruism, and rationality: a theory of social choice. Cambridge [Cambridgeshire] ; New York: Cambridge University Press.

 

McAdam, Doug, Sidney G. Tarrow, and Charles Tilly. 2001. Dynamics of contention, Cambridge studies in contentious politics. New York: Cambridge University Press.

 

Moser, Paul K., ed. 1990. Rationality in action: contemporary approaches. Cambridge [England] ; New York: Cambridge University Press.

 

Mueller, Dennis C. 1976. Public Choice: A Survey. Journal of Economic Literature 14 (2):395-433.

 

----------, ed. 1989. Public choice II. Cambridge [England] ; New York: Cambridge University Press.

 

----------. 1993. Intergenerational Justice and the Social Discount Rate. In The public choice approach to politics, edited by D. C. Mueller. Aldershot, U.K.: Elgar.

 

Newall, A., and Herbert Simon. 1972. Human Problem Solving. Englewood Cliffs, N.J.: Prentice-Hall.

 

Newton-Smith, William. 1982. Relativism and the Possibility of Interpretation. In Rationality and Relativism, edited by M. Hollis and S. Lukes.

 

Nisbett, R., and L. Ross. 1980. Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs, N.J.: Prentice-Hall.

 

Nozick, Robert. 1993. The nature of rationality. Princeton, N.J.: Princeton University Press.

 

Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge: Harvard University Press.

 

Ostrom, Elinor. 1990. Governing the commons: the evolution of institutions for collective action. Cambridge [England] ; New York: Cambridge University Press.

 

Oye, Kenneth A., ed. 1986. Cooperation under anarchy. Princeton, N.J.: Princeton University Press.

 

Popkin, Samuel L. 1979. The rational peasant: the political economy of rural society in Vietnam. Berkeley: University of California Press.

 

Przeworski, Adam. 1985. Marxism and Rational Choice. Politics & Society 14 (4):379–409.

 

----------. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America, Studies in Rationality and Social Change. Cambridge: Cambridge University Press.

 

Rapoport, Anatol. 1966. Two-person game theory; the essential ideas, Ann Arbor science library, AAS515. Ann Arbor,: University of Michigan Press.

 

Rapoport, Anatol, and A. M. Chammah. 1965. Prisoners’ Dilemma: A Study in Conflict and Cooperation. Ann Arbor: University of Michigan.

 

Reed, Michael I. 1997. In Praise of Duality and Dualism: Rethinking Agency and Structure in Organizational Analysis. Organizational Studies 18 (1):21-42.

 

Roemer, John. 1986. ‘Rational Choice’ Marxism: Some Issues of Method and Substance. In Analytical Marxism, edited by J. Roemer.

 

Russell, Clifford S., ed. 1979. Collective Decision Making. Baltimore: Johns Hopkins Press.

 

Sandler, Todd, and John T. Tschirhart. 1980. The Economic Theory of Clubs: An Evaluative Survey. Journal of Economic Literature 18 (4):1481-521.

 

Schelling, Thomas C. 1960. The strategy of conflict. Cambridge,: Harvard University Press.

 

----------. 1978. Micromotives and Macrobehavior. New York: Norton.

 

----------. 1984. Choice and consequence. Cambridge, Mass.: Harvard University Press.

 

----------. 1995. Intergenerational discounting. Energy Policy 23 (4-5):395(7).

 

Schotter, Andrew, and Gerhard Schwodiauer (?). 1980. Economics and Game Theory: A Survey. Journal of Economic Literature 18 (2):479-527.

 

Sen, Amartya. 1970. Collective Choice and Social Welfare: San Francisco: Holden-Day.

 

----------. 1973. Behaviour and the Concept of Preference. Economica 40:241-59.

 

----------. 1982. Rational Fools. In Choice, Welfare and Measurement, edited by A. Sen.

 

Shilling, Chris. 1997. The Undersocialised Conception of the Embodied Agent in Modern Sociology. Sociology 31 (4):737-754.

 

Shubik, Martin. 1982. Game Theory in the Social Sciences: Concepts and Solutions. Cambridge: MIT Press.

 

Simon, Herbert A. 1957. Models of Man: Social and Rational. New York: Wiley, 1957.

 

----------, ed. 1979. Models of Thought. New Haven: Yale University.

 

----------. 1983. Reason in Human Affairs. Stanford: Stanford University.

 

Tarrow, Sidney. 1996. The People’s Two Rhythms: Charles Tilly and the Study of Contentious Politics. Comparative Studies in Society and History 38 (3):586-600.

 

Taylor, Michael. 1976. Anarchy and Cooperation. London: Wiley.

 

----------. 1982. Community, Anarchy and Liberty. Cambridge: Cambridge University Press.

 

----------. 1987. The possibility of cooperation. Cambridge: Cambridge University Press.

 

----------. 1988. Rationality and Revolutionary Collective Action. In Rationality and Revolution, edited by M. Taylor.

 

----------, ed. 1988. Rationality and revolution, Studies in Marxism and social theory. Cambridge [Cambridgeshire] ; New York; Paris: Cambridge University Press ; Editions de la Maison des Sciences d’Homme.

 

Tong, James. 1988. Rational outlaws: rebels and bandits in the Ming Dynasty, 1368–1644. In Rationality and Revolution, edited by M. Taylor.

 

----------. 1991. Disorder Under Heaven: Collective Violence in the Ming Dynasty. Stanford, California: Stanford University Press.

 

Tsebelis, George. 1990. Nested Games: Rational Choice in Comparative Politics. Berkeley: University of California Press.

 

Wason, P. C., Stephen E. Newstead, and Jonathan St B. T. Evans, eds. 1995. Perspectives on thinking and reasoning: essays in honour of Peter Wason. Hove: Erlbaum Associates.

 

Wilson, Bryan R., ed. 1970. Rationality. Oxford: Blackwell.

 

Winch, Peter. 1970. Understanding a Primitive Society. In Rationality, edited by B. Wilson.

 



    [1] See Korner 1972 for a collection of useful papers on the subject.

 

 

cking (19??) provides a useful survey of the main elements of this part of the theory of rationality.

    [3] Central contributions include Jeffrey (1965), Luce and Raiffa (1958), Schelling (1963), and Rapoport (1966).

    [4] Elster (1979) provides discussion of some of the relations between real human reasoners and formal decision theory and game theory.

    [5] Chomsky's account may be found in Aspects of the Theory of Syntax (1965).  This sort of approach gives rise to an idealized theory in that it abstracts from computational errors, memory limitations, etc., which are found in practical decision making or speech, but it is empirically grounded in that it is intended to represent the actual processes through which the cognitive task is performed.  Rawls takes something like this approach in his analysis of our moral competence (1971).

    [6] One might argue that this correlation is unlikely.  On the side of inductive reasoning, for example, much recent empirical research shows that actual inductive reasoners commit systemic fallacies in inferring from a set of facts to a generalization.  Likewise, the processes of strategic deliberation recommended by two-person game theory are sufficiently mathematical and complex to make it unlikely that the human cognitive system embodies these procedures in any approximate fashion.

cking (19??) provides a useful survey of the main elements of this part of the theory of rationality.

    [8] Central contributions include Jeffrey (1965), Luce and Raiffa (1958), Schelling (1963), and Rapoport (1966).

    [9] Elster (1979) provides discussion of some of the relations between real human reasoners and formal decision theory and game theory.

    [10] Chomsky's account may be found in Aspects of the Theory of Syntax (1965).  This sort of approach gives rise to an idealized theory in that it abstracts from computational errors, memory limitations, etc., which are found in practical decision making or speech, but it is empirically grounded in that it is intended to represent the actual processes through which the cognitive task is performed.  Rawls takes something like this approach in his analysis of our moral competence (1971).

    [11] One might argue that this correlation is unlikely.  On the side of inductive reasoning, for example, much recent empirical research shows that actual inductive reasoners commit systemic fallacies in inferring from a set of facts to a generalization.  Likewise, the processes of strategic deliberation recommended by two-person game theory are sufficiently mathematical and complex to make it unlikely that the human cognitive system embodies these procedures in any approximate fashion.

Go to UnderstandingSociety gateway