Rationality of Emotion

Standard Constraints on Rationality

The criteria of rationality commonly accepted for thought and action suggest four abstract constraints on rationality in general. These provide a starting point for any discussion of emotional rationality.

• 1.
  Norms of success. No entity can be assessed for rationality unless it is liable to success and failure. Rationality is not equivalent to success, nor can it ever guarantee success, but X can be said to be more rational than an alternative Y, insofar as X has the greater likelihood of success. Truth is the norm of success for a belief; hence, B1 is more rational than B2 if it is more likely to be true. Similarly, of two alternative actions, the more rational is the more likely to achieve a given goal.
• 2.
  Intentionality. The existence of a norm of success implies that whatever can be rational is susceptible of teleological explanation—that is, explanation in terms of some function or purpose. But the converse does not hold. Biological processes typically call for teleological explanations, but only those that are intentional can be said literally to be rational. Intentionality is informally characterized as “aboutness” and is widely thought to be an essential property of mental states. One could speak metaphorically of ants, plants, cells, or even genes as communicating and as choosing alternative strategies of survival or mating. But it would be eccentric to ascribe mentality literally to all biological organisms. Only intentional states can be rational.
• 3.
  Origins. Rationality is systematically related to future success, but ascriptions of rationality do not await the verdict of success. On the contrary, rationality hangs on provenance: It depends in part on the origin of the action or belief. If one belief derives from the consultation of astrological signs, [Page 628]while another is soundly inferred from scientific evidence, the latter is the more rational regardless of its truth.
• 4.
  Context dependency. If origin determines rationality, how do we identify the appropriate antecedents? In statistical reasoning, it may be rational to believe p relative to one set of facts and not-p relative to another, though both sets are equally correct and relevant. Beliefs and actions may be fairly judged both rational and irrational, depending on the extent of the background circumstances taken into account. This context dependency may be illustrated in terms of the tragic real-life case of Andrea Yates, who was induced by voices she heard to drown her five children. At her first trial, the insanity defense was disallowed, on the ground that her careful planning and execution of the drownings proved her rationality. But if one zooms out from its methodical implementation to the project of drowning one's five children, in obedience to the voice of God, to “save” them, one is bound to see that project itself as irrational. Yet while both Agamemnon and Abraham formed, in obedience to divine command, the project of killing their child, neither is usually thought to have been insane. Unlike truth, a verdict of rationality is never definitive. The framework of its assessment can in principle be extended or modified in an indefinite number of defensible ways.

Applying Standard Constraints to Emotional Rationality

To apply these constraints to emotion, we must first make any sense of the notion of emotional “success.” One approach to this is in terms of biological function: Emotions appraise typical life situations and prime the organism for appropriate response; any given emotion is successful if it fulfils that function. This evolutionary psychological approach has yielded many insights. But it doesn't quite get at a notion of success that would be relevant to the question of rationality, for not all functional processes or states are intentional. The intentionality requirement disqualifies even some of the states we loosely refer to as emotions: Moods, insofar as they lack intentionality, also lack conditions of success of the relevant sort. Any emotion that is clearly about something, by contrast, intrinsically defines what must be true of its target—the thing, person or situation at which it is directed—for the emotion to be appropriate. That feature is what many philosophers refer to as the emotion's formal object. Truth is the formal object of belief: “Because it's true” gives a trivial answer to the question “Why do you believe p?” Similarly, one can give gives a trivial answer to the question “Why are you E-ing?” for any given emotion E. But there is no global answer for all emotions: Each emotion has its own formal object. Some have obvious names: for fearing, the fearsome; being sad, loss; loving, lovable; being disgusted, the disgusting. Others call for awkward explanatory phrases. Just as the formal object of the sense of touch has no single name but relates to hardness, texture, and relative heat, so the formal object of anger has no single name but is awkwardly describable in terms of unjust harm or insult to oneself or others. Whether or not a formal object has a handy name doesn't seem to correlate with how easily we can tell whether it applies in a particular case. To find something fearsome, for example, is to perceive it as dangerous: an arguably objective, albeit probabilistic property. By contrast, shame is successful if its object is shameful, but the appropriateness of shameful, unlike dangerous, does not yield easily to objective confirmation.

...