Rationalism in Ethics (Practical-Reason Approaches)

views updated


Practical-reasoning theory is a kind of metaethical viewalongside noncognitivism and other cognitivisms such as naturalism and rational intuitionismthat aims to understand ethics as rooted in practical reason.

Tradition divides the faculty of reason into two parts: theoretical and practical. Theoretical reason concerns what we should believe, practical reason what we should do. Beliefs aim to represent reality and are mistaken or in error when they do not. Theoretical reason's task, therefore, is to discover what is true of the independent order of fact to which belief is answerable. But what about practical reason? What could make it the case that an action is something a person ought to do?

Plainly, ethical convictions also aim at a kind of objectivity. If Jones thinks he should devote all his resources to conspicuous consumption but Smith thinks that Jones should donate some to help the poor, their convictions conflict. Only one, at most, can be true.

Practical-reasoning theories aim to explain the objective purport of ethical conviction, but in a way that respects a fundamental distinction between theoretical and practical reason. Like noncognitivism, these theories sharply distinguish between ethics and those theoretical disciplines that aim to represent some independent reality, whether the order of nature or some supersensible metaphysical realm. They therefore reject both naturalism and rational intuitionism. But they also deny noncognitivism, since they hold that ethical propositions can be true or false.

According to practical-reasoning theories, objectivity consists not in accurate representation of an independent order, but in demands that are universally imposed within an agent's own practical reasoning. What marks ethics off from science is its intrinsically practical character, its hold on us as agents. It is because there is such a thing as practical reason, a form rational agents' deliberations must take, that there is such a thing as ethics.

But what form does rational deliberation take? Uncontroversially, practical reasoning includes reasoning from ends to means. The interesting debates concern what else it involves, if anything, and how instrumental reasoning is itself to be understood. Humeans maintain that means-end reasoning exhausts practical reason and that instrumental reason can be reduced to the use of theoretical reason in discovering means to ends. They tend not to be practical-reasoning theorists, however, since they argue that ethics fundamentally concerns what engages human sympathy or moral sentiment rather than what it is rational for a person to do. By contrast, practical-reasoning theorists deny that practical reason can be reduced to theoretical reason. As Christine Korsgaard has argued, even instrumental practical reason directs an agent who has already used theoretical reason in determining that B is the only means to his end A to undertake B (or to give up A as an end). In this way instrumental practical reasoning parallels the structure of modus ponens in theoretical reasoning (the move from "p " and "if p, then q " to "q ").

Pursuing the analogy with theoretical reasoning (while insisting on irreducibility) further suggests that instrumental reasoning cannot exhaust practical reason. When we reason from our beliefsfor example, with modus ponens we reason from their contents, not from the fact of our believing them. We reason from p and if p, then q, not from the facts that we believe that p and that we believe that if p, then q. Similarly, when we adopt an end, we do not simply select it by sheer fiat. Rather, we choose it as something (we think) there is some reason to do. Thus, when we reason from our ends, we do not reason from the fact that they are our ends but from our commitments to them as things it makes sense to do. That is why instrumental rationality is so uncontroversial. As R. M. Hare argued, it is questionable at best that it follows from the facts that a person's end is to kill someone in the most grisly possible way and that using a cleaver is such a way that the person ought, or has some normative reason, to use a cleaver. What is uncontroversial is simply that the support of reasons transfers from end to means, other things being equal, and from not taking the (only available) means to renouncing the end, other things being equal. It follows only that a person ought to use a cleaver or give up my end.

On grounds such as these, practical-reasoning theorists tend to hold that instrumental rationality cannot exhaust practical reason. But how are we to deliberate about ends? What makes something a reason for adopting an end? Since they hold that reasons for action are necessarily connected to the agent's deliberative perspective, practical-reasoning theorists generally adopt what Korsgaard has called the internalism requirement, according to which a reason must be something the agent could, in principle, be moved by in deliberation and act on. This makes it a necessary condition of something's being a reason for an agent that she would be moved by it insofar as she deliberated rationally.

But what then is rational deliberation? Practical-reason theorists are loath to derive a deliberative ideal by independently specifying paradigm reasons for acting and holding that deliberation is rational when it responds appropriately to them. That would theorize practical reason too much on the model of theoretical reason. Rather, they maintain that rational deliberation must be understood formally, so that reason for acting is a status consideration inherit when it is such that it would move an agent who formed her will in accordance with that deliberative ideal.

The aspects that have been considered so far are relatively common among practical-reasoning theories, although not, perhaps, universal. Within these theories, however, there is a major division between neo-Hobbesians and neo-Kantians. Although nothing on the surface of practical-reasoning theory might suggest this result, it is notable that both camps attempt to vindicate the commonsense idea that moral obligations are supremely authoritative. Both argue that (at least some central) moral demands are demands of practical reason.


Recent versions of this view have their roots in ideas advanced by Kurt Baier in the late 1950s and attempt to address a significant problem faced by Baier's early view. Baier argued that reasons for acting must ultimately connect with the agent's interests. This does not reduce all practical reasoning to prudential reasoning, since other forms may advance agents' interests also. Specifically, Baier argued that morality may be viewed as a system of practical reasoning that is in the interest of everyone alike. Since it is mutually advantageous for everyone to regard moral obligations as supremely authoritative, Baier concluded that they actually do create overriding reasons for acting.

David Gauthier objected to Baier's theory that, while it is in the interest of each that all regard interest-trumping moral reasons as supreme, it is unclear how this can show that an individual agent should so regard them, since it will still most advance her interest to act prudentially when morality conflicts with self-interest. Why, then, might it not be true that instrumental and prudential reasoning exhaust practical reason, even if a person should hope to live in a world in which other people view things differently and (mistakenly) treat moral reasons as authoritative?

Gauthier is himself responsible for the major recent neo-Hobbesian practical-reasoning theory. Like Baier, Gauthier begins from the premise that practical reasoning must work to advance the agent's interests, although here his account is more nearly "internalist," since he understands a person's interests to consist in what she would herself prefer were she to be fully informed. Also like Baier, Gauthier argues that the fact that mutual advantage may require individuals to constrain their pursuit of self-interest can be used to show that practical reason counsels this constraint. However, it is not enough that it be true that everyone would do better if everyone so constrained his or her prudential reasoning. The crucial point for Gauthier is that individuals can do better if they constrain self-interest by a willingness to abide by mutually advantageous agreements.

Two agents who appear to each other to be unconstrained pursuers of self-interest simply cannot make agreements, however mutually advantageous the agreements might be, if these agreements would require the agents to act contrary to their own interests. In what have come to be known as prisoner's dilemma situations, therefore, mutually advantageous rational agreement between such persons is impossible. If each believes the other will rationally defect from the agreement on the condition that doing so is in her interest, then neither can rationally make the agreement.

Personal advantage therefore counsels presenting oneself to others as someone who is not an unconstrained maximizer of self-interest. Of course, it is possible, theoretically, for someone to do this while still deliberating as an unconstrained prudential reasoner. But it may not be practically possible, Gauthier argues, at least not for normal human beings. Human motivation may be sufficiently translucentthrough involuntary response, for exampleso that the least costly way of appearing to others as someone who can be relied upon to keep mutually advantageous, interest-constraining agreements is actually to be such a person. If that is so, then instrumental and prudential reason will not support themselves as principles to guide rational deliberation. On the contrary, they will recommend that agents deliberate in terms of an alternative conception of practical reason that counsels keeping mutually advantageous agreements, even when this is contrary to self-interest.

As a practical-reasoning theorist, Gauthier believes that reasons for acting cannot be understood except in relation to what should guide a rational agent in deliberation. And he believes that a rational agent is someone whose dispositions of choice and deliberation serve her best and most advance her interest. But just as indirect forms of ethical consequentialism, such as character- and rule-consequentialism, face the objection that they are unstable and threaten to collapse into either act-consequentialism or deontology, Gauthier's indirect consequentialist theory of rationality may face the same objection. What motivates the move away from unconstrained prudence, on the grounds that it cannot support itself in the agent's practical thinking, is a view about the role a principle of rational conduct must be able to play in the deliberations of an autonomous rational agent that may be more Kantian than Hobbesian in inspiration.


This contemporary tradition may be held to date from Thomas Nagel's The Possibility of Altruism (1970) and John Rawls's reinvigoration of Kantian moral and political philosophy in A Theory of Justice (1971) and "Kantian Constructivism in Moral Theory" (1980). Nagel's book was read as having both a modest and a more ambitious agenda. His more modest goal, suggested by his title, was to show how such "objective" (or, as he later termed them, "agent-neutral") considerations as "that acting would be relative someone's pain" can be genuine reasons for acting. A consideration can be rationally motivating, he argued, even if the agent lacks any relevant desire for acting on it other than one that is motivated by the awareness of that very consideration. A person may be moved, for instance, by considering long-term interests. And if motivation at a distance is possible with prudence, it can happen with altruism as well. Altruistic and other agent-neutral considerations can be rationally motivating.

Nagel's more ambitious agenda was to argue that practical reasoning is subject to a formal constraint that effectively requires that any genuine reason for acting be agent-neutral. Stressing the "motivational content" of genuine practical judgments, Nagel argued that avoiding a kind of solipsism is possible only if an agent is able to make the same practical judgment of himself from an impersonal standpoint as he does from an egocentric point of view. Since accepting practical judgments from one's own point of view normally motivates, Nagel maintained, making the same judgment of oneself from an impersonal standpoint should normally motivate also. But this will be so only if the reasons for acting that ground practical judgments are agent-neutral. So it is a necessary condition for avoiding practical solipsism that agents take considerations such as that something will advance their own ends or interests as reasons only if they regard them as instantiating more general, agent-neutral reasons, such as that acting will advance someone's ends or interests. Nagel later retreated from this strong claim in a direction that is arguably even more Kantian. Autonomous agency, he later argued, involves an agent's acting on reasons she can endorse from an objective standpoint, and such a set of reasons will include both agent-relative and agent-neutral ones.

Neo-Kantian practical-reasoning theories have been put forward by a number of philosophers, including Alan Gewirth, Stephen Darwall, and Christine Korsgaard. Korsgaard's sympathetic reconstruction of Immanuel Kant's own arguments in a series of papers has been especially influential. Common to all these neo-Kantian approaches has been the idea that the practical reasoning of an autonomous agent has a formal structure, with its own internal standards and constraints, and that these provide the fundamental truth and objectivity conditions for ethical thought and discourse. Thus, Gewirth maintains that fundamental moral principles are derivable from propositions to which a rational agent is committed from within the deliberative standpoint in acting. And Korsgaard argues that even instrumental theorists are committed to the "hypothetical imperative" as a practical norm. Since, however, we regard ourselves to be free as agents to adopt and renounce ends, practical reason cannot possibly be exhausted by any mere consistency constraint, such as the hypothetical imperative. It follows, the neo-Kantians argue, that practical reason requires norms to regulate the choice of ends no less than to guide the choice of means. In choosing ends for reasons we commit ourselves implicitly to principles of choice as valid for all. But such a commitment is not, they claim, a hypothesis about some independently existing order of normative fact to which we might have cognitive access. That, after all, is precisely the difference between theoretical and practical reason. So the standards to which deliberation is subject must ultimately be based on some formal principle of impartial endorsement that is internal to free practical reasoning itself. And this will be so, they conclude, only if practical reasoning is regulated by some such principle as the categorical imperative, which requires that one act only on principles that one can will to regulate the deliberation and choices of all. If moral demands are ultimately grounded in the categorical imperative also, it will follow that moral demands are demands of practical reason.

See also Baier, Kurt; Consequentialism; Decision Theory; Gewirth, Alan; Hare, Richard M.; Intuitionism; Kant, Immanuel; Metaethics; Nagel, Thomas; Naturalism; Noncognitivism; Practical Reason; Rationality; Rawls, John; Reason.


Baier, K. The Moral Point of View. Ithaca, NY: Cornell University Press, 1958.

Darwall, S. Impartial Reason. Ithaca, NY: Cornell University Press, 1983.

Falk, W. D. Ought, Reasons, and Morality. Ithaca, NY: Cornell University Press, 1986.

Gauthier, D. "Morality and Advantage." Philosophical Review 76 (1967): 460475.

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

Gewirth, A. Reason and Morality. Chicago: University of Chicago Press, 1978.

Korsgaard, C. "The Source of Normativity." In The Tanner Lectures on Human Values, edited by G. Peterson. Salt Lake City: University of Utah Press, 1994.

Nagel, T. The Possibility of Altruism. Oxford: Clarendon Press, 1970.

Nagel, T. The View from Nowhere. New York: Oxford University Press, 1986.

Rawls, J. "Kantian Constructivism in Moral Theory." Journal of Philosophy 77 (1980): 515572.

Rawls, J. A Theory of Justice. Cambridge, MA: Belknap Press of Harvard University Press, 1971.

Stephen Darwall (1996)