"'Describing the parties’ choice as a rational choice subject to the reasonable constraints imposed by the original position allows Rawls to invoke the theory of rational choice and decision under conditions of uncertainty. In rational choice theory there are a number of potential “strategies” or rules of choice that are more or less reliably used depending on the circumstances. One rule of choice—called “maximin”—directs that we play it as safe as possible by choosing the alternative whose worst outcome leaves us better off than the worst outcome of all other alternatives. The aim is to “maximize the minimum” regret or loss to one’s position (measured in terms of welfare or, for Rawls, one’s share of primary social goods). To follow this strategy, Rawls says you should choose as if your enemy were to assign your social position in whatever kind of society you end up in. By contrast another strategy leads us to focus on the most advantaged position and says we should “maximize the maximum” potential gain—“maximax”—and choose the alternative whose best outcome leaves us better off than all other alternatives. Which, if either, of these strategies is more sensible to use depends on the circumstances and many other factors.
A third strategy advocated by orthodox Bayesian decision theory, says we should always choose to directly maximize expected utility. To do so under conditions of uncertainty of outcomes, the degree of uncertainty should be factored into one’s utility function, with probability estimates assigned to alternatives based on the limited knowledge that one has. Given these subjective estimates of probability incorporated into one’s utility function, one can always choose the alternative that maximizes expected utility. Since it simplifies matters to apply the same rule of choice to all decisions this is a highly attractive idea, so long as one can accept that it is normally safe to assume that that the maximization of expected utility leads over time to maximizing actual utility.
What about those extremely rare instances where there is absolutely no basis upon which to make probability estimates? Suppose you don’t even have a hunch regarding the greater likelihood of one alternative over another. According to orthodox Bayesian decision theory, the “principle of insufficient reason” should then be observed; it says that when there is no reason to assign a greater likelihood to one alternative rather than another, then an equal probability is to be assigned to each potential outcome. This makes sense on the assumption that if you have no more premonition of the likelihood of one option rather than another, they are for all you know equally likely to occur. By observing this rule of choice consistently over time, a rational chooser presumably should maximize expected individual utility, and hopefully actual utility as well.
What now is the appropriate decision rule to be used to choose principles of justice under conditions of complete uncertainty of probabilities in Rawls’s original position? Rawls argues that, given the enormous gravity of choice in the original position, plus the fact that the choice is not repeatable (there’s no opportunity to renegotiate or revise one’s decision), it is rational for the parties to follow the maximin strategy when choosing between the principles of justice and principles of average or aggregate utility (or any other principles that do not guarantee basic rights, liberties, opportunities, and a social minimum). Not surprisingly, following the maximin rule of choice results in choice of the principles of justice over the principles of utility (average or aggregate); for unlike utilitarianism, justice as fairness guarantees equal basic liberties, fair equal opportunities, and an adequate social minimum for all citizens.
Why does Rawls think maximin is the rational choice rule? Recall what is at stake in choice from the original position. The decision is not an ordinary choice. It is rather a unique and irrevocable choice where the parties decide the basic structure of their society, or the kind of social world they will live in and the background conditions against which they will develop and pursue their aims. It is a kind of superchoice—an inimitable choice of the background conditions for all one’s future choices. Rawls argues that because of the unique importance of the choice in the original position—including the gravity of the choice, the fact that it is not renegotiable or repeatable, and the fact that it determines all one’s future prospects—it is rational to follow the maximin rule and choose the principles of justice. For should even the worst transpire, the principles of justice guarantee an adequate share of primary goods enabling one to maintain one’s conscientious convictions and sincerest affections and pursue a wide range of permissible ends by protecting equal basic liberties and fair equal opportunities and guaranteeing an adequate social minimum of income and wealth. The principles of utility, by contrast, provide no guarantee of any of these benefits.
Rawls says that in general there are three conditions that must be met in order to make it rational to follow the maximin rule (TJ 154–55/134 rev.). First, there should be no basis or at most a very insecure basis upon which to make estimates of probabilities. Second, the choice singled out by observing the maximin rule is an acceptable alternative we can live with, so that one cares relatively little by comparison for what is to be gained above the minimum conditions secured by the maximin choice. When this condition is satisfied, then no matter what position one eventually ends up in, it is at least acceptable. The third condition for applying the maximin rule is that all the other alternatives have worse outcomes that we could not accept and live with. Of these three conditions Rawls later says that the first plays a minor role, and that it is the second and third conditions that are crucial to the maximin argument for justice as fairness (JF 99). This seems to suggest that, even if the veil of ignorance were not as thick and parties did have some degree of knowledge of the likelihood of ending up in one social position rather than another, still it would be more rational to choose the principles of justice over the principle of utility.
Rawls contends all three conditions for the maximin strategy are satisfied in the original position when choice is made between the principles of justice and the principle of utility (average and aggregate). Because all one’s values, commitments, and future prospects are at stake in the original position, and there is no hope of renegotiating the outcome, a rational person would agree to the principles of justice instead of the principle of utility. For the principles of justice imply that no matter what position you occupy in society, you will have the rights and resources needed to maintain your valued commitments and purposes, to effectively exercise your capacities for rational and moral deliberation and action, and to maintain your sense of self-respect as an equal citizen. With the principle of utility there is no such guarantee; everything is “up for grabs” (so to speak) and subject to loss if required by the greater sum of utilities. Conditions (2) and (3) for applying maximin are then satisfied in the comparison of justice as fairness with the principle of (average or aggregate) utility.
It is often claimed that Rawls’s parties are “risk-averse;” otherwise they would never follow the maximin rule but would take a chance on riskier but more rewarding outcomes provided by the principle of utility. Thus, John Harsanyi contends that it is more rational under conditions of complete uncertainty always to choose according to the principle of insufficient reason and assume an equal probability of occupying any position in society. When the equiprobability assumption is made, the parties in the original position would choose the principle of average utility instead of the principles of justice (Harsanyi 1975).
Rawls denies that the parties have a psychological disposition to risk-aversion. They have no knowledge of their attitudes towards risk. He argues however that it is rational to choose as if one were risk averse under the highly exceptional circumstances of the original position. His point is that, while there is nothing rational about a fixed disposition to risk aversion, it is nonetheless rational in some circumstances to choose conservatively to protect certain fundamental interests against loss or compromise. It does not make one a risk averse person, but instead it is normally rational to purchase auto liability, health, and home insurance against accident or calamity (assuming it is affordable). The original position is such a situation writ large. Even if one knew in the original position that the citizen one represents enjoys gambling and taking great risks, this would still not be a reason to gamble with their rights, liberties and starting position in society. For if the high risktaker were born into a traditional, repressive, or fundamentalist society, they might never have an opportunity for gambling and taking other risks they normally enjoy. It is rational then even for high risktakers to choose conservatively in the original position and guarantee their future opportunities to gamble or otherwise take risks.
Harsanyi and other orthodox Bayesians contend that maximin is an irrational decision rule, and they provide ample examples. To take Rawls’ own example, in a lottery where the loss and gain alternatives are either (0, n) or (1/n, 1) for all natural numbers n, maximin says choose the latter alternative (1/n, 1). This is clearly irrational for almost any number n except very small numbers. (TJ 136 rev.). But such examples do not suffice here; simply because maximin is under most circumstances irrational does not mean that it is never rational. For example, suppose n>1 and you must have 1/n to save you own life. Given the gravity of the circumstances, it would be rational to choose conservatively since you are guaranteed 1/n according to the maximin strategy, and there is no guarantee you will survive if you choose according to the principle of insufficient reason.
No doubt maximin is an irrational strategy under most circumstances of choice uncertainty, particularly under circumstances where we will have future opportunities to recoup our losses and choose again. But these are not the circumstances of the original position. Once the principles of justice are decided, they apply in perpetuity, and there is no opportunity to renegotiate or escape the situation. One who relies on the equiprobability assumption in choosing principles of justice in the original position is being foolishly reckless given the gravity of choice at stake. It is not being risk-averse, but rather entirely rational to refuse to gamble with one’s basic liberties, fair equal opportunities and adequate resources needed to pursue one’s most cherished ends and commitments, simply for the unknown chance of gaining the marginally greater social powers, income and wealth that might be available to some in a society governed entirely by the principle of utility.
Rawls exhibits the force of the maximin argument in discussing liberty of conscience. He says (TJ, sect. 33) that a person who is willing to jeopardize their right to hold and practice their conscientious religious, philosophical and moral convictions, all for the sake of gaining uncertain added benefits via the principle of utility, does not know what it means to have conscientious beliefs, or at least does not take such beliefs seriously (TJ 207–08/181–82 rev.). A rational person with convictions about what gives life meaning is not willing to negotiate with and gamble away the right to hold and express those convictions and the freedom to act on them. After all what could be the basis for negotiation, for what could matter more than the objects of one’s most sincere convictions and commitments? Some people (e.g. some nihilists) may not have any conscientious convictions (except the belief that nothing is worthwhile) and are simply willing to act on impulse or on whatever thoughts and desires they happen to have at the moment. But behind the veil of ignorance no one knows whether they are such a person, and it would be foolish to make this assumption. Knowing general facts about human propensities and sociability, the parties must take into account that people normally have conscientious convictions and values and commitments they are unwilling to compromise. (Besides, even the nihilist should want to protect the freedom to be a nihilist, to avoid ending up in an intolerant religious society.) Thus it remains irrational to jeopardize basic liberties by choosing the principle of utility instead of the principles of justice.
None of this is to say that maximin is normally a rational choice strategy. Rawls himself says it “is not, in general, a suitable guide for choices under uncertainty” (TJ 153). It is not even a rational strategy in the original position when the alternatives for choice guarantee basic liberties, equal opportunities, and a social minimum guaranteed by the principle of average utility''.
Stanford
No comments:
Post a Comment