Published online by Cambridge University Press: 01 January 2020
Game theorists assume that rational defensibility is a necessary condition for moral, social, or political justification. By itself, this is a fairly uncontroversial claim; most moral or political philosophers would agree. And yet game theorists tend to be advocates of the free market. External critics of game theory usually claim this is because game theorists assume that individuals are atomistic and self-interested. Game theorists themselves deny this, however, for what strike us as good reasons. In principle, game theory has no necessary ties to right-wing distribution schemes. Why, then, is game theory almost exclusively the province of conservative philosophers, political scientists, and economists? The problem, we believe, lies in the theory of rational choice standardly employed by game theory. Even if we accept, for the purposes of argument, game theory's account of the justification of moral dispositions — that a disposition is morally justified if and only if, in its absence, it would be game theoretically rational to acquire it — we need not be led to right-wing solutions. If we expand the kinds of choices facing individuals to include choices about what we will call ‘institutional roles,’ then we can explain the game theoretic rationality of the kinds of emotions and behavior exemplified by duty, loyalty, and love.
1 Gauthier, David ‘Deterrence, Maximization, and Rationality,’ in his Moral Dealing: Contract, Ethics, and Reason (Ithaca: Cornell University Press 1990) 298–321CrossRefGoogle Scholar
2 To see how it might be game theoretically rational to adopt a disposition to retaliate upon provocation (that is, a disposition for vengefulness), see Kavka, Gregory Moral Paradoxes of Nuclear Deterrence (Cambridge: Cambridge University Press 1986)Google Scholar.
3 Gauthier, David Morals By Agreement (Oxford: Clarendon Press 1986)Google Scholar
4 Narveson, Jan ‘Have We a Right to Non-Discrimination?’ in Poff, Deborah and Waluchow, Wilfrid eds., Business Ethics in Canada, 2nd ed. (Scarborough, ON: Prentice Hall 1991) 279-94.Google Scholar
5 Danielson, Peter Artificial Morality (London: Routledge 1992)Google Scholar
6 See, for example, Held, Virginia ‘Non-Contractual Society,’ in Hanen, Marsha and Nielsen, Kai eds., Science, Morality and Feminist Theory (Calgary: University of Calgary Press 1987), 111-37Google Scholar, as well as her Feminist Morality: Transforming Culture, Society, and Politics (Chicago: University of Chicago Press 1993), especially ch 10.
7 Pateman, Carole ‘Women and Consent,’ Political Theory 8 (1980), 153CrossRefGoogle Scholar; emphasis hers. Reprinted in Pateman, Carole The Disorder of Women: Democracy, Feminism and Political Theory (Stanford: Stanford University Press 1989), 74Google Scholar.
8 See, for example, Morals By Agreement, 11 and 253.
9 Even if relations were ontologically prior to their relata (i.e., individuals), we believe it still would remain important to address moral justification to individuals (even if not only to individuals).
10 Just exactly what ‘well-orderedness’ involves is itself an issue in game theory. (Even transitivity is not universally accepted.) But nothing in this paper hinges on the resolution of these disputes.
11 See, for example, Jaggar, Alison M. Feminist Politics and Human Nature (Totowa, NJ: Rowman and Littlefield 1983)Google Scholar, or some of the essays in Beyond Self-Interest, Mansbridge, Jane J. ed. (Chicago: University of Chicago Press 1990)Google Scholar.
12 Rawls, John A Theory of Justice (Cambridge, MA: Harvard University Press 1971), 127; see also 13Google Scholar.
13 This ‘standard’ reasoning has been questioned within game theory itself, and models for preferences that co-refer are now being developed. See, for example, Vallentyne, Peter ‘Contractarianism and the Assumption of Mutual Unconcern,’ in Vallentyne, Peter ed., Contractarianism and Rational Choice (Cambridge: Cambridge University Press 1991), 71-5Google Scholar.
14 See, for example, Morals By Agreement, especially chapter 9.
15 David Gauthier made this reply when asked by one of us at a talk whether a rational individual would ever choose to be a parent.
16 See, for example, George, Rolf ‘Who Should Bear the Cost of Children?’ Public Affairs Quarterly 1 (1987) 1–42Google Scholar.
17 Our claim here is that motherhood itself is not a ‘motherhood issue’ - that the institutional role of ‘mother’ warrants rational moral investigation.
18 See, for example, Gauthier, David ‘Reason and Maximization,’ Canadian Journal of Philosophy 4 (1975) 411-33CrossRefGoogle Scholar; reprinted in Moral Dealing: Contract, Ethics, and Reason, 209-33.
19 Schotter, Andrew The Economic Theory of Social Institutions (Cambridge: Cambridge University Press 1981 ), 11CrossRefGoogle Scholar. Schotter's technical definition of a social institution is: ‘A regularity R in the behavior of members of a population P when they are agents in a recurrent situation r is an institution if and only if it is true and common knowledge in P that (1) everyone conforms to R; (2) everyone expects everyone else to conform to R; and (3) either everyone prefers to conform to R on the condition that others do, if r is a coordination problem, in which case uniform conformity to R is a coordination equilibrium; or (4) if anyone ever deviates from R it is known that some or all of the others will also deviate and the payoffs associated with the recurrent play of r using these deviating strategies are worse for all agents than the payoff associated with R’ (11). This definition is a modification of Lewis's, David definition of a social convention in Convention: A Philosophical Study (Cambridge, MA: Harvard University Press 1969)Google Scholar.
20 Schotter likens the development of social institutions to the evolution of species. He writes: ‘Economic and social systems evolve the way species do. To ensure their survival and growth, they must solve a whole set of problems that arise as the system evolves. Each problem creates the need for some adaptive feature, that is, a social institution …. For instance, the problem of multilateral exchange in neoclassical economies (economies satisfying all of the ‘proper’ neoclassical assumptions) is solved by the creation of competitive markets’ (1-2). We do not wish to push the species analogy, however. Social systems, unlike species, are teleological, purposive and normative; they can be changed, or at least directed, by human effort in ways that evolution cannot. Our task as moral and political philosophers is prescriptive, not descriptive. We use game theory as a tool to indicate which social institutions can be justified, not just to explain the evolution of whatever social institutions happen to exist in a particular society.
21 Once again, we are not claiming this is the only reason for forming families.