I thought it might be helpful if I were to write up a few notes on how I see the material in Unit III all fitting together. (In some cases the logical order of topics deviates somewhat from the order in which the readings were assigned but that was deliberate - I like to make sure the core problem solving material is covered well before the exam.)

Let me begin with recapitulating what happened in the first two units of this class. In Unit I we got the basics of Rational Choice Theory (RCT), primarily the method of maximizing expected utility, along with a few other strategies, such as "play it safe" or "best option" that can be used in special cases.

In Unit II we learned some of the ways in which people systematically fail to make quantitative probability and utility assignments in a manner consistent with RCT. And we looked in some detail at a proposed alternative theory of decision making (Prospect Theory) which tries to describe how people actually evaluate probabilities and utilities. At the very least, Unit II suggests that RCT needs to be modified if it is to describe a lot of actual decision making behavior.

In Unit III we go back to RCT as a starting point. (Unit III does not reprise any of the material in Unit II.) We now introduce two new questions about the adequacy of Rational Choice Theory. The first concerns the role of moral values in decision making. The example that we dealt with at length is the decision about whether to vote.

What seems to be the case here is that people's perception that they have a duty to vote often plays a crucial role in their decision making. But for people who feel they have such a obligation, the "duty" factor does not appear to just get added into a cost/benefit analysis or entered into a calculation of expected utility. Rather it seems to act as a different kind of variable, one that places restrictions on or even overrides the kinds of factors that we normally invoke in expected utility decision making. So we conclude that RCT also needs to be supplemented with an account of moral reasoning if it is to be a good model for decision making.

It would take us too far afield in this course to try to present a detailed account of moral deliberation, but we do look briefly at Kohlberg's theory of the development of moral reasoning in children and adolescents. This example of moral theory is nice because it illustrates the fact that there can be a lot of variation in what people believe their duty to be, at least in Kohlberg's preliminary stages. But it also shows that at the post conventional levels there is also a convergence in basic values that holds across cultures. So, for example, all of Kohlberg's mature subjects recognize that the life of Heinz's wife is more important than the druggist's property rights or Heinz's convenience. In any event, the excursion into Kohlberg's higher levels provides some examples of the moral factors that might need to be added to the more ordinary costs and benefits that are central to RCT.

So to summarize the first main point of Unit III: RCT is incomplete as a description of human decision making because it does not provide a natural framework for the inclusion of moral factors. The second point strikes much closer to the heart of the RCT approach because it suggests an inadequacy in taking RCT to be a normative ideal even in cases where moral considerations are not relevant. Here I am referring to the situations illustrated by the Prisoners' Dilemma (PD) and the Tragedy of the Commons (ToC).

Suppose we view the decisions to be made in PD and ToC as ones which should be dictated entirely by narrow self-interest. (Let us begin by postulating that the examples we are dealing with do not and should not involve moral considerations. Later we will relax this assumption.) The point of the PD is that even if all I care about is my own time spent in jail, I will not achieve a good result unless I find some way of cooperating with the other prisoner. If each of us acts as an isolated decision maker using RCT the result for each of us will be sub-optimal. Ditto for the decision about how many cows to graze on the commons.

The significance of these examples is not that we should be nice and think about the other prisoner or the other farmers just because it's the moral thing to do. (Of course, it is nice to be thoughtful, but that is not the primary lesson to be drawn from the examples.) The point is that you cannot even be efficiently selfish without taking into consideration what the other person is going to be doing if they also also act according to RCT! Neither of you is going to be effective at getting your own selfish desires satisfied unless you can develop some sort of method for cooperating!

What we have here is a sort of amoral argument for morality. If one wants to maximize one's self-interests in a PD or ToC situation, one should adopt a cooperative strategy that on the surface makes it look like you care about the the welfare of the other party. (Of course, it may be easier to be cooperative if you really do care about the other guy, but there are also selfish reasons for doing so!) The problem for RCT now becomes to describe the relationship between the standard RCT analysis where the decision matrix considers only the straightforward situation for an isolated individual and some sort of super-RCT analysis in which individual choices are coordinated in a way so as to arrive at good group decisions. The approach which attempts to do just that is called Game Theory, but we will not pursue it any further in this course.

So where are we so far? Well, the voting case reminds us that people do invoke moral considerations. (This example is invoked in part because there has been so much research done on it.) The PD and ToC examples show us one sort of situation that show us why cooperation has lots of practical benefits. What remains to be done in this unit is to sketch a picture of a kind of biological bridge that links the purely pragmatic justification of cooperation with the kinds of higher level moral reasoning discussed by Kohlberg and invoked in the voting case.

In the final part of Unit III we look briefly at the ways in which evolutionary biologists talk about altruism and other forms of social behavior in animals and relate these ideas to what we have discovered about the role of moral considerations and the need for cooperative strategies in human decision making.

One goal of these intro Honors courses is to illustrate how approaches from various disciplines can be used to illuminate important issues. In this unit we have a mixture of material drawn from economics, political science, cognitive psychology, evolutionary theory and moral philosophy. This obviously poses a somewhat different sort of challenge to students than the typical discipline based course. But if you get an overview of what is happening I hope you won't get too confused. At least you know that if you don't much like one reading assignment the next one is sure to be completely different!