"Emotion and Moral Judgment"Philosophy and Phenomenological Research, Vol. LXVI, No. 1 (pp. 104-124), by Linda Zagzebski (University of Oklahoma),
Reviewed by Kyle Swan (College of Charleston)
Moral cognitivists face the problem of moral motivation. How could it be that by merely taking the world to be a certain way we are inclined to do something? According to the Humean view of human psychology, since I can conceive the representation of the world apart from the motivational state, the cognitive state and the motivational state are distinct and independent states of mind. Therefore, if moral judgments are to be intrinsically motivating, they must be noncognitive. But moral noncognitivists face the problem of moral discourse. The structure of moral discourse suggests that moral judgments are genuine propositions that speakers assert. If moral judgments are genuine propositions, and are not just something else in disguise, then moral judgments must be cognitive.
Perhaps we ask too much of moral judgments. Linda Zagzebski, however, has offered an interesting and clear account of the structure of moral judgments, which aims to preserve their chief noncognitivist aspect and their chief cognitivist aspect. According to that account, moral judgments are intrinsically motivating and they are genuine propositions with truth conditions that are sometimes met. If Zagzebski is right about this, then the Humean test - separate conceivability implies independence - is irrelevant to at least some states of mind. She argues that this is true in the case of emotions. Emotions are both cognitive and affective and the cognitive aspect of emotions cannot exist independent of the affective aspect.
Zagzebski provides a set of examples of what she calls ground level moral judgments. They are ground level, or Level 1 judgments, in the sense they are made when someone is face to face with the object of the judgment. All such judgments employ terms that combine descriptive and affective aspects, so called thick concepts. Consider the judgment, "That is contemptible." A situation has the descriptive feature of someone acting with disregard for the well being of his or her child. My becoming aware of this situation, plus the requisite dispositional state, brings about in me the characteristic emotional state of feeling contempt for the person; I regard him or her as contemptible. According to Zagzebski, I cannot see the situation as contemptible without being in the emotional state, and cannot be in the relevant emotional state without seeing the situation as contemptible. I express this emotion of mine by issuing the judgment, "That is contemptible." According to Zagzebski, since the judgment expresses the emotion, it is intrinsically motivating. And since I am asserting that the intentional object of the emotion falls under the think affective concept, the judgment is propositional in form with truth conditions that may be met.
Zagzebski's account of moral judgment is internalist and cognitivist. I want to examine aspects of her internalism. It seems that people can make judgments of the form, "That is A", where A is some thick affective concept, yet not at all be motivated to act on the judgment. Psychopaths seem to do this. Internalists typically claim that whatever psychopaths are doing when they make such claims, they are not making moral judgments. Michael Smith, for example, suggests that in such cases inverted commas be affixed to the moral term in order to relay the idea that psychopaths must mean something like "That is what people say is A" when they use the moral term.
Zagzebski's strategy is similar, but she does not make use of inverted commas. When we say, "That is contemptible", and we do not feel contempt, Zagzebski refers to the judgment as a Level 2 judgment. The judgment has undergone a kind of "thinning." It is a step removed from the ground level judgment. As such, the judgment, "That is contemptible" is ambiguous between the Level 1 judgment where the speaker sees the object of the judgmentas contemptible and the Level 2 judgment where the speaker sees that it is contemptible. The difference is that "I cannot see something as [contemptible] without feeling [contempt], but I can see that it is [contemptible] without feeling [contempt]" (119). Level 2 judgments use thick affective concepts, like contemptible, but Zagzebski denies that the judgments express emotion. They merely express the proposition that the object of the judgment is contemptible. This distinction helps Zagzebski make sense of the many cases in which people make moral judgments but either struggle before acting on them, if indeed they do act on them, or fail to be at all motivated by them. In such cases, the person who made the judgment saw that the object of the judgment is A, but did not see the object as A.
The same is true of Zagzebski's Level 3 judgments, an additional step removed from ground level judgments. Thick affective concepts are no longer employed in Level 3 judgments. They are replaced with terms such as "right", "wrong", "should", and "ought". After being confronted with a situation that I see as falling under a thick affective concept, I make the judgment, "I ought to prevent this" or "It is right for me to prevent this." These are judgments about what to do in response to seeing the situation as falling under the thick affective concept. Like Level 2 judgments, these are not intrinsically motivating. I may judge that I ought to prevent something, but find myself struggling to act on that judgment or utterly without any motivation to do so. However, this should not be too much of a surprise if Zagzebski is correct and Level 3 judgments, like Level 2 judgments, are not expressions of emotion. According to Zagzebski, only ground level judgments, judgments employing thick affective concepts in a certain way, are expressions of emotion.
First, as a conceptual point, Zagzebski's idea about what it is to express a mental state seems a bit off and this affects the distinctions she wants to make among her three levels of moral judgment. The fact that "That is contemptible" can be used to express the emotion contempt towards the judgment's object is a feature of the semantics of the sentence. But the fact that in a given context a person who says, "That is contemptible" has expressed the emotion is merely of feature of the pragmatics of the sentence. What this means is that "expressed" should be understood in such a way that people can express a mental state, like an emotion, that they do not possess. To express an emotion is simply a speech act. The speaker simply takes advantage of the rules of language in such a way that he represents himself as having that conative state of mind. If this is the correct account of what it is to express a mental state, then Zagzebski is wrong to think that her ground level moral judgments are intrinsically motivating because they express emotion and also wrong to think that her Level 2 and Level 3 judgments do not express emotion.
First, if expressing an emotion is simply a speech act, then the Level 3 judgment, "It is wrong to act with disregard for the well being of one's children" does express a moral attitude and so an emotion. Perhaps it expresses the appropriateness of feeling guilty for having acted with disregard for the well being of one's children. Second, someone could utter the ground level moral judgment, "That is contemptible" without feeling contempt towards the object of the judgment. Nonetheless, the sentence expresses contempt for it. Again, given our linguistic conventions, it expresses that emotion. This does not necessarily mean that motivational judgment internalism is false. It just means that if it is true, then its truth does not depend on the fact that moral judgments express emotion. The truth or falsity of motivational judgment internalism, the thesis that moral judgments are intrinsically motivating, seems to have nothing to do with the fact that moral judgments express emotion.
My second point invokes some empirical data about the developmental psychology of young children.1 On the face of it, it seems odd to claim, as Zagzebski does, that judgments of the form "That is A", where A is some thick affective concept are more basic than judgments of the form "X is wrong", where X is some action or state of affairs. Research, which supports this intuition, has found that before their 4th birthdays children are fairly proficient at distinguishing between moral transgressions and conventional transgressions.2 Young children have the ability to make moral judgments. However, they do not seem to have the ability to make judgments about whether or not the intentional objects of their judgments fall under thick affective concepts. It is doubtful that children of that age understand what contempt is, let alone are able to make normative assessments of whether or not a given state of affairs calls for it. The concept is too complex. However, Zagzebski claims that making a ground level moral judgment, her most basic type of moral judgment, requires this ability. Since young children have the ability to make moral judgments, she must be wrong about this. The ability to understand and appropriately deploy thick affective concepts is not essential to making moral judgments.
I am inclined to agree with Zagzebski that the right account of moral judgment should make room for human emotions to play an essential role. However, if the conceptual point above is more or less correct, then allowing such a role for the emotions does not necessarily motivate motivational judgment internalism. Moreover, if the empirical point above is more or less correct, then the emotions that are thought to play an essential role in moral judgments had better not be too complex.
1. I thank Shaun Nichols for this point.
2. See S. Nichols, "Norms with Feeling: Towards a Psychological Account of Moral Judgment," Cognition 84, 221-236 and J. Smetana and J. Braeges, "The Development of Toddlers' Moral and Conventional Judgments," Merrill-Palmer Quarterly 36, 329-246.