Rationality

views updated

RATIONALITY

Philosophers have, at least characteristically, aspired to possess "rationality" but have not thereby sought exactly the same thing. Portrayed vaguely, rationality is reasonableness, but not all philosophers take rationality as dependent on reasons; nor do all philosophers have a common understanding of reasons or of reasonableness. Some theorists consider rationality to obtain in cases that lack countervailing reasons against what has rationality; they thus countenance rationality as, in effect, a default status. In ordinary parlance, persons can have rationality; so, too, can beliefs, desires, intentions, and actions, among other things. The rationality appropriate to action is practical, whereas that characteristic of beliefs is, in the language of some philosophers, theoretical.

Many philosophers deem rationality as instrumental, as goal oriented. You have rationality, according to some of these philosophers, in virtue of doing your best, or at least doing what you appropriately think adequate, to achieve your goals. If ultimate goals are not themselves subject to assessments of rationality, then rationality is purely instrumental, in a manner associated with David Hume's position. Rationality, according to this view, is a minister without portfolio; it does not require any particular substantive goals of its own but consists rather in the proper pursuit of one's ultimate goals, whatever those goals happen to be. Many decision-theoretic and economic approaches to rationality are purely instrumentalist. If, however, ultimate goals are susceptible to rational assessment, as an Aristotelian tradition and a Kantian tradition maintain, then rationality is not purely instrumental. The latter two traditions regard certain rather specific (kinds of) goals, such as human well-being, as essential to rationality. Their substantialist approach to rationality lost considerable influence, however, with the rise of modern decision theory.

When relevant goals concern the acquisition of truth and the avoidance of falsehood, so-called epistemic rationality is at issue. Otherwise, some species of nonepistemic rationality is under consideration. One might individuate species of nonepistemic rationality by the kind of goal at hand; moral, prudential, political, economic, aesthetic, or some other. Some philosophers have invoked rationality "all things considered" to resolve conflicts arising from competing desires or species of rationality; even so, there are various approaches to rationality "all things considered" in circulation. The standards of rationality are not uniformly epistemic, then, but epistemic rationality can play a role even in what some call nonepistemic rationality. Regarding economic rationality, for instance, a person seeking such rationality will, at least under ordinary conditions, aspire to epistemically rational beliefs concerning what will achieve the relevant economic goals. Similar points apply to other species of nonepistemic rationality. A comprehensive account of rationality will characterize epistemic and nonepistemic rationality, as well as corresponding kinds of irrationality (e.g., weakness of will).

Taking rationality as deontological, some philosophers characterize rationality in terms of what is rationally obligatory and what is merely rationally permissible. If an action, for instance, is rationally obligatory, then one's failing to perform it will be irrational. Other philosophers opt for a nondeontological evaluative conception of rationality that concerns what is good (but not necessarily obligatory) from a certain evaluative standpoint. Some of the latter philosophers worry that, if beliefs and intentions are not voluntary, then they cannot be obligatory. Still other philosophers understand rationality in terms of what is praiseworthy, rather than blameworthy, from a certain evaluative standpoint. The familiar distinction between obligation, goodness, and praiseworthiness thus underlies three very general approaches to rationality.

Following Henry Sidgwick, William Frankena has distinguished four conceptions of rationality: (1) an egoistic conception implying that it is rational for one to be or do something if and only if this is conducive to one's own greatest happiness (e.g., one's own greatest pleasure or desire satisfaction); (2) a perfectionist conception entailing that it is rational for one to be or do something if and only if this is a means to or a part of one's moral or nonmoral perfection; (3) a utilitarian conception implying that it is rational for one to be or do something if and only if this is conducive to the greatest general good or welfare; and (4) an intuitionist conception implying that it is rational for one to be or do something if and only if this conforms to self-evident truths, intuited by reason, concerning what is appropriate. The history of philosophy represents, not only these conceptions of rationality, but also modified conceptions adding further necessary or sufficient conditions to one of (1)–(4).

Given an egoistic conception of rationality, one's being rational will allow for one's being immoral, if morality requires that one not give primacy to oneself over other people. Rationality and morality can then conflict. Such conflict is less obvious on a utilitarian conception of rationality. In fact, if morality is itself utilitarian in the way specified (as many philosophers hold), a utilitarian conception of rationality will disallow rational immorality. A perfectionist conception of rationality will preclude rational immorality only if the relevant perfection must be moral rather than nonmoral; achieving nonmoral perfection will, of course, not guarantee morality. As for an intuitionist conception of rationality, if the relevant self-evident truths do not concern what is morally appropriate, then rational immorality will be possible. An intuitionist conception will bar conflict between rationality and morality only if it requires conformity to all the self-evident truths about what is morally appropriate that are relevant to a situation or person. So, whether rationality and morality can conflict will depend, naturally enough, on the exact requirements of the conception of rationality at issue.

Richard Brandt has suggested that talk of what it would be rational to do functions to guide action by both recommending action and by making a normative claim that evaluates the available action relative to a standard. An important issue concerns what kind of strategy of using information to choose actions will enable one to achieve relevant goals as effectively as any other available strategy. Brandt has offered a distinctive constraint on such a strategy: A rational decision maker's preferences must be able to survive their being subjected to repeated vivid reflection on all relevant facts, including facts of logic. This constraint suggests what may be called (5) a relevant-information conception of rationality: Rationality is a matter of what would survive scrutiny by all relevant information.

A relevant-information conception of rationality depends, first, on a clear account of precisely when information is relevant and, second, on an account of why obviously irrational desires cannot survive scrutiny by all relevant information. Evidently, one could have a desire caused by obviously false beliefs arising just from wishful thinking, and this desire could survive a process of scrutiny by all relevant information where the underlying false beliefs are corrected. In any case, a relevant-information conception of rationality will preclude rational immorality only if it demands conformity to all relevant moral information.

The egoistic, perfectionist, utilitarian, and relevant-information conceptions of rationality are nonevidential in that they do not require one's having evidence that something is conducive to self-satisfaction, perfection, general welfare, or support from all relevant information. Many philosophers would thus fault those conceptions as insufficiently sensitive to the role of relevant evidence in rationality. If relevant evidence concerns epistemic rationality, we again see the apparent bearing of epistemic rationality on rationality in general. The latter bearing deserves more attention in contemporary work on nonepistemic rationality.

Philosophers currently divide over internalism and externalism about rationality. If rationality demands reasons of some sort or other, the dispute concerns two senses of talk of a person's having a reason to perform an action. An internalist construal of this talk implies that the person has some motive that will be advanced by the action. An externalist construal, in contrast, does not require that the person have a motive to be advanced by the action. Bernard Williams, among others, has suggested that any genuine reason for one's action must contribute to an explanation of one's action and that such a contribution to explanation must be a motivation for the action. He concludes that externalism about rationality is false, on the ground that external reasons do not contribute to explanation of action in the required manner. Externalism about rationality does allow that reasons fail to motivate, but this, according to externalists, is no defect whatever. Externalists distinguish between merely motivating reasons and justifying reasons, contending that only the latter are appropriate to rationality understood normatively; what is merely motivating in one's psychological set, in any case, need not be justifying. Perhaps, then, disputes between internalists and externalists will benefit from attention to the distinction between justifying and merely motivating reasons.

Modern decision theory assumes that, in satisfying certain consistency and completeness requirements, a person's preferences toward the possible outcomes of available actions will determine, at least in part, what actions are rational for that person by determining the personal utility of outcomes of those actions. In rational decision making under certainty one definitely knows the outcomes of available actions. In decision making under risk one can assign only various definite probabilities less than 1 to the outcomes of available actions. (Bayesians assume that the relevant probabilities are subjective in that they are determined by a decision maker's beliefs.) In decision making under uncertainty one lacks information about relevant states of the world and hence cannot assign even definite probabilities to the outcomes of available actions. Acknowledging that rationality is purely instrumental (and thus that even Adolf Hitler's Nazi objectives are not necessarily rationally flawed), Herbert Simon has faulted modern decision theory on the ground that humans rarely have available the facts, consistent preferences, and reasoning power required by standard decision theory. He contends that human rationality is "bounded" in that it does not require utility maximization or even consistency. Rather, it requires the application of a certain range of personal values (or preferences) to resolve fairly specific problems one faces, in a way that is satisfactory, rather than optimal, for one. Simon thus relies on actual human limitations to constrain his account of rationality.

Contemporary theorists divide over the significance of human psychological limitations for an account of rationality. The controversy turns on how idealized principles for rationality should be. This raises the important issue of what exactly makes some principles of rationality true and others false. If principles of rationality are not just stipulative definitions, this issue merits more attention from philosophers than it has received. Neglect of this metaphilosophical issue leaves the theory of rationality as a subject of ongoing philosophical controversy.

Bibliography

Audi, Robert. Action, Intention and Reason. Ithaca, NY: Cornell University Press, 1993.

Audi, Robert. The Architecture of Reason: The Structure and Substance of Rationality. New York: Oxford University Press, 2001.

Benn, S. I., and G. W. Mortimore, eds. Rationality and the Social Sciences. London: Routledge and Paul, 1976.

Brandt, R. A Theory of the Good and the Right. Oxford: Clarendon Press, 1979.

Brown, Harold. Rationality. London: Routledge, 1988.

Elster, J., ed. Rational Choice. New York: New York University Press, 1986.

Foley, Richard. The Theory of Epistemic Rationality. Cambridge, MA: Harvard University Press, 1987.

Frankena, W. "Concepts of Rational Action in the History of Ethics." Social Theory and Practice 9 (1983): 165–197.

Fumerton, Richard. "Audi on Rationality: Background Beliefs, Arational Enjoyment, and the Rationality of Altruism." Philosophy and Phenomenological Research 67 (2003): 188–193.

Gauthier, D. Morals by Agreement. Oxford: Clarendon Press, 1986.

Harman, Gilbert. Change in View: Principle of Reasoning. Cambridge, MA: MIT Press, 1986.

Hollis, M., and S. Lukes, eds. Rationality and Relativism. Cambridge, MA: MIT Press, 1982.

Jeffrey, Richard. The Logic of Decision. 2nd ed. Chicago: University of Chicago Press, 1983.

Katz, Jerrold. Realistic Rationalism. Boston: MIT Press, 1998.

Mele, A. Irrationality. New York: Oxford University Press, 1987.

Mele, A., and P. Rawlings, ed. The Oxford Handbook of Rationality. Oxford: Oxford University Press, 2004.

Moser, P., ed. Rationality in Action. Cambridge, U.K.: Cambridge University Press, 1990.

Nozick, R. The Nature of Rationality. Princeton, NJ: Princeton University Press, 1993.

Rescher, N. Rationality. Oxford: Clarendon Press, 1988.

Sidgwick, H. The Methods of Ethics. 7th ed. London, 1907.

Simon, Herbert. Reason in Human Affairs. Stanford, CA: Stanford University Press, 1981.