Modifying Behavior: Making People Moral

Caroline West, in "Business as Usual? The Error Theory, Internalism, and the Function of Morality" (in A World Without Values: Essays on John Mackie's Moral Error Theory, (Richard Joyce and Simon Kirchin, eds.) presents a problem for anybody who wants to deny internalism.

And I want to deny internalism.

West first argues that a person who accepts the three propositions below cannot accept internalism. If these three claims are true, then internalism must be false.

(1) Cognitivism , the view that moral claims express beliefs that can be true or false.

(2) The Humean theory of motivation , the thesis that beliefs and desires are distinct existences, and that the motivational states are desires.

(3) The Humean theory of normative reason , which holds that reason alone mandates no revision to a subject's existing desire set, except instrumentally.

Well, I accept all three of these propositions as true; which means that I hold that internalism is false.

West then combines these three claims with an observation that I summarize as follows:

(4) The use of morality to alter behavior - to counteract limited sympathies, exploit role who can be persuaded that they have certain duties to others, cause people to consider the interests of others, and the like.

The idea is that for morality to work, we at least expect that a person’s behavior can be altered merely by convincing them of certain moral claims. Convince a serf that she is obligated to work the Earl’s lands, and the serf will then work the Earl’s lands. Convince a person that he has a duty to tell the truth in a court of law, and he will be motivated to tell the truth in a court of law.

Yet, this relationship between belief and behavior seems to require some form of internalism. It requires that there be some sense in which believing a moral claim rationally requires an agent to behave in a particular way.

If my interpretation of West's argument is correct, she must be saying that the problem exists if we also accept:

(5) Internalism is false. It is not the case that being persuaded of a moral claim rationally requires that an agent change his behavior.

The reported problem is that, once we add (5), we have to get rid of (4), or so it seems.

If, having been persuaded by the arguments against internalism, we all came to believe that there was no necessary rational connection between moral judgment, then we could not use morality in all of the ways that we presently do.

I would like to present a hypothesis for reconciling (5) with (4).

What I am going to argue is that moral statements are meant to alter behavior. However, the mechanism for doing so is not the motivational-rational implications of adopting a moral belief. The mechanism is causal, and does not involve reason or rationality at all. Specifically, moral claims contain an element of praise or condemnation, and that the purpose of this praise or condemnation is to alter desires directly. By altering desires, they alter behavior.

However, in doing this, I wish to maintain the truth of the first three propositions as well.

My first worry is that, by claiming that moral statements are statements of praise or condemnation, I will be taken as asserting some type of non-cognitivist emotivism. If this were true, then I would be contradicting item (1) on the list above, and my project would have already failed.

To confront this objection, I want to argue that moral claims both contain an element of praise or condemnation, and express beliefs that can be true or false – both at the same time.

Imagine being a teacher in a conversation with a student who is upset because he is not being allowed to take a make-up test. The student slaps his hand on the table top and shouts, “You promised!”

This statement expresses a belief that can be true or false. One of the ways in which you may answer the student is to say, “No, I primised that you could take a makeup test on Thursday. Today is Friday.” Thus, you can prove that the truth-apt component of the student’s statement is, in fact, false. The first proposition is preserved. Moral claims are not merely statements of praise or condemnation.

Yet, at the same time, the student is also expressing an attitude of condemnation. He is condemning you in specific for breaking a promise, and he is condemning promise-breakers in general (anybody who would break a promise).

It is the function of this praise or condemnation to modify behavior.

In order to explain this link, I want to speak more generally about the effects of biological reward and punishment on the brain and their effects on behavior. I will ultimately argue that the effects of praise functions as a reward and that condemnation functions as punishment – at least much of the time.

Please note, at this point I am talking about rewards and punishments in the biological sense – not the moral or legal sense. An animal presses a lever and gets a food pellet. The food pellet is, in the biological sense, a reward. There is no sense here that the animal deserved the pellet, or that it acquired the pellet as a result of good behavior. The pellet is a reward only in the sense that it is something that the animal is disposed to acquire, and that the animal is disposed to learn how to acquire.

The same is true of punishments. The same animal presses a bar and gets an electric shock. In calling this a punishment, I am not saying that the animal deserved to be shocked, or that the shock serves as some sort of retribution for pressing the bar. The shock is a punishment only in the sense that it is something that the animal is disposed to avoid, and is disposed to learn how to avoid.

There are two primary ways in which the food pellet can alter the animal’s behavior.

The animal can learn to press the button as a means to obtaining the food pellet. “If you want a food pellet, then you should press the button.” This method of altering behavior works entirely by altering beliefs. It is the belief that pressing the bar will release a food pellet, and a desire for the food pellet, and motivates the animal to press the bar.

However, rewards and punishments also have an effect on molding desires and aversions themselves.
What begins by being valued as a means to an end ultimately comes to be valued as an end itself. And that which an animal avoids as a means for avoiding some punishment (electrical shocks) he will come to avoid for its own sake, even when there are no more shocks.

I admit that I am glossing over this quite quickly. Research shows that the function of rewards and punishments alter attitudes to things other than the direct causal antecedents of that which brings the reward. For example, it also changes attitudes based on other associations. If the animal tends to collect his food in a blue room, the animal will begin to acquire a preference for being in a blue room.

A vital part of this hypothesis is that praise serves as a type of reward and condemnation serves as a type of punishment. Consequently, the praise and condemnation built into moral claims aim to work on the limbic system in the brain to alter attitudes. By altering those attitudes, they alter behavior.

It is also important to recognize that praise and condemnation have an impact far beyond those who are specifically praised or condemned. It effects those who witness the praise or condemnation, and even those who hear about the praise or condemnation. In fact, the praise or condemnation itself need not have happened – the praise or condemnation of a fictitious character can have an impact on molding the moral character of those who read or hear the stories. By altering the moral character of those in the audience, these stories alter behavior.

Once again, please note that this change in behavior does not come about through an application of reason. It comes about through an application of biological rewards (including praise) and punishments (including condemnation) on the limbic system of the brain. Consequently (and importantly) nothing in this violates provision (3) above - the provision that reason alone mandates no revision to a subject's existing desire set, except instrumentally.

I will offer it as a defense of this hypothesis that it accounts for a great many of our moral practices such as the ubiquitous use of rewards such as praise or punishments such as condemnation in making moral claims. It explains the use of rewards such as plaques, ribbons, and certificates – and the use of public award ceremonies. It accounts for the use of parables and stories to teach moral lessons. Those lessons, after all, aim not only to change beliefs, but to change desires as well.

Accepting that the purpose of morality is to change behavior only requires internalism if one assumes that the truth-bearing claim somehow is supposed to rationally require a change of behavior. If we place that work in a different mechanism instead – one that uses causation rather than rationality – then we can have a purpose of moral statements being a change in behavior without requiring that the change come from the belief itself.