ABSTRACT: I argue that Plato's psychological theories are motivated by concerns he had about moral theory. In particular, Plato rejects the modern account of rationality as the maximization of subjectively evaluated self-interest because, had he adopted such an account, his theory of justice would be subject to criticisms which he holds are fatal to the contractarian theory of justice. While formulating a theory to remain within ethical constraints sometimes violates the canons of scientific theorizing, Plato avoids this mistake.

The first serious account of justice Plato considers in the Republic is the contractarian account.(1) It holds that is always instrumentally rational for one to further her own interests and in that certain situations (exemplified by the prisoners dilemma) it is more rational to forego one's own interests (providing others do so also) than to behave in a straight-forwardly rational way. The rules allowing one to escape prisoner's dilemmas—the rules it is rational to accept providing all others accept them also—are simply the rules of morality. Hence it is rational to be moral.(2)

Plato agrees that rationality requires self-interested action. However, he distinguishes between perceived self-interest and actual self-interest and argues that any apparent conflict between rationality and morality is simply a conflict between one's perceived self-interest and the requirements of justice. Pursuing of one's actual self-interest never conflicts with the demands of morality. Since, for Plato, it is more rational to pursue one's real, than one's apparent, self-interest, rationality and morality do not conflict. It is rational to be moral.

Plato rejects the contractarian reconciliation of morality with individual rationality primarily because the thinks that the contractarian conception assumes that a person's motives for being just are necessarily based her self-interest, while our concept of the just person holds that to be truly just one must value justice for its own sake. The contractarian account is also unacceptable because it has no foorce in the case of the Lydia Shepherd.(3) Finally, Plato holds that we must reject the contractarian account because a better account is available to us, viz., his own account of justice. But to show this Plato must establish each of the following: 1. There really is a difference between perceived self-interest and actual self-interest, that there can be a difference between what one believes to be in one's interest and what really is in one's interest. 2. Provide an account of what one's actual self-interest is.(4) 3. Show that one's actual self-interest, like one's perceived self-interest, is the sort of thing one could come to know.(5) 4. Show that once one has knowledge of one's actual self-interest one will come to see that it really is of value, that it is worth pursuing and that we are the sort of entities which could be motived to act for the sake of our actual self-interest once we come to know what those interests are.(6) 5. Show that acting in accordance with one's actual self-interest is identical with (or, at the very least, compatible with) being a just person. That is, he must show us that the just person is also the rational person, one who bases her actions on reason. 6. Show how his account of the relationship between justice and rational self-interest illuminates our pre-critical concept of justice in an attractive way. That is, he must show that his account provides us with a truly valuable conception of justice.

In Book IV of the Republic, Plato provides us with an account of human psychology compatible with the idea that we might be motivated by the goal of acting in our actual, rather than our apparent, self-interest; he outlines a picture of human nature which allows for the possibility that moral motives could be our motives.(7) The first step in understanding Plato here is to see why he feels it is necessary to develop this account of human nature. Why doesn't he simply rely on the standard account of human nature, the account adopted by the contractarians with whom he is arguing?

The psychological theory assumed by Glaucon in stating the contractarian account of justice was best stated and most ably defended by Hume.(8) Hume held that the human mind is divided into two parts: reason and the passions. The passionate part provides us with goals, things we want for their own sake. The rational element serves to determine the best, or most efficient, means of attaining that which the passionate element has informed us we want. For Hume, reason does not play a direct role in providing us with motives; it does not determine what we want, only how to get that which we want. As Hume put it, reason "is, and ought only to be, the slave of the passions".(9) But this powerful and elegant picture will not serve Plato's purposes as can be seen by conjoining Plato's reconciliation of morality with rationality with Humean psychology. To see that this is so I will first sketch what the Platonic project would look like had Plato adopted a Humean psychology. I will then explain why he had to reject this Humean/Platonic account. My hope is that an understanding of Plato's rejection of the Humean story will illuminate the role Plato's account of the mind plays in his moral theory.

Had Plato accepted the Humean psychology, his account of the relationship between morality and rationality would have been much like the following. It is rational to maximize the satisfaction of one's desires for to do this is simply to maximize what is in one's self-interest. Human beings come to have certain wants, and reason tells them how to satisfy them. Yet working at satisfying the desires one happens to have will not necessarily maximize satisfaction, because our true natures are such that each of us has a basic and powerful passion for fulfillment. This desire, the desire for fulfillment, is one of which most people are unaware. Satisfying the wants we happen to have is not, as a matter of contingent psychological fact, a good, or efficient, means of satisfying the fundamental desire for fulfillment, which we all possess. It is the task of philosophy (particularly of the revisionist Platonism being outlined here) to bring to light the fundamental desire for fulfillment. Once we become conscious of this desire, we can turn our reason to the task of figuring out how best to satisfy it. Since the content of our basic desire—what I have called the desire for fulfillment—is the same as acting in the way that the just person acts, we all have a motive and a reason for being just. That is, once we see that our real self-interest lies in satisfying this desire we will see that the best, and as it turns out, only, way to satisfy this desire is to act justly. Plato would then tell us in what acting justly consisted. Thus, he would have provided us with a reason, that of satisfying our most fundamental desire, for acting justly.

According to the revisionist Platonic account, it is rational to be or become moral. It is rational to act justly because acting justly is the best available means of maximizing desire satisfaction. However, this account of the relationship between rationality and morality is unacceptable to Plato. Presumably, Plato thought of an account like this and rejected it for some reason. Why? This question is particularly pressing since the account just given is, compared with the account Plato actually gives us, such a simple and tidy account. I suggest the reason Plato rejected the revisionist Platonic account outlined above is the same as the reason he rejected the entire contractarian account of justice: the actors in the revisionist Platonic account are not acting justly because it is good to act justly. Rather, they are acting justly because doings so is a means of attaining that which they really want, viz., to satisfy their own desires. It is, of course, true that what they really want happens to require the same behavior that justice requires. However, Plato must reject this story because the revisionist Platonic actors are not motivated to act justly for its own sake. Rather, they are using justice as a means to some other end; they are not treating it as an end in itself.

The fact that the revisionist Platonic account given above is unsatisfactory reveals the tremendous difficulty of the task Plato has set for himself. He has to develop a theory which will motivate people to be moral, for any account of justice which conceives of this virtue as inert is useless. But he cannot motivate people to be moral just by appealing to any wants or desires they happen to have (whether or not they are aware of these desires), even if the desire appealed to is, in its content, equivalent to the desire to be or to become a moral person. Plato appears to have snookered himself. He has to develop an account of human motivation that does not run afoul of his principal criticism of the contractarian conception of justice, but even the best possible motive he could hope for—the desire which, in its content, is simply the desire to be just—runs afoul of this criticism. It is for this reason that Plato has to develop an alternative account of human motivation; one which differs radically from the Humean account.

While the contractarian divides the mind into two parts; the passions, which provide us with goals for action, and reason, which tells us how to attain those goals, Plato divides the mind into three parts; reason, spirit, and appetite. The difference here is not that Plato has added a third part to the two part Humean/contractarian account.(10) No element of the Platonic mind corresponds to either the reason or passions found in the Humean mind. No part of the Platonic mind by itself provides us with nothing but ends for our action; no element of the Platonic mind is concerned with nothing but determining how best to obtain goals provided by some other part of the mind. Rather, each part of the Platonic mind provides both its own ends and the means of attaining them. Each part of the Platonic mind may have either the virtue or the vice which is appropriate to it. Thus, an individual's reason may be either wise of foolish, her spirit either courageous or cowardly, her appetites either temperate or immoderate.

By denying that there is an element of the mind which, by itself, determines the content of our self-interest, Plato is able to have his cake and eat it too. Anyone who holds that one element of the mind both determines our self-interest and moves us to act is going to be prey to the charge that the just person is just simply because it is in her (amoral) interest to be just. (This holds even if one adds further elements to the mind in addition to the Humean passions.) But Plato's account of the mind opens the door to the possibility that people might be motivated to be just both for moral reasons and for reasons of self-interest. He has the chance to show that it is rational to be moral without it being the case that moral persons do good for the wrong (that is, solely self-interested) reasons. To see how Plato accomplishes this feat we need to remind ourselves of his account of the just individual. Plato's justice consists of a certain kind of harmony between the various elements of the soul, though today we usually speak of this virtue in terms of integrity or self-actualization. The easiest way to grasp Plato's picture of the just individual—that is, the individual with a harmonious soul—is to try to see what it is that she lacks. What has the unjust individual got that one with a harmonious soul does not? Plato's answer is that the person with a harmonious soul lacks psychic dissension. Some elaboration will clarify this point. Suppose it is my goal to satisfy my desires. One way in which I can fail to achieve this goal is by having desires that are incompatible or inappropriate. For example, may desires may be logically incompatible, in that I may desire both p and not-p. Alternatively, my desires may be causally incompatible; I may want to have breakfast in Halifax and lunch in Athens. Desires may be incompatible or inappropriate in other ways. I may find that sets of my desires are mutually unbecoming, aesthetically incompatible, or personally intolerable. Plato recommends that we eliminate the psychic tension which is simply the state of having incompatible desires.(11)

On Plato's view, the just person is one with a harmonious soul. That is to say, she lacks the psychic tension of having incompatible desires. This theory raises several points which deserve brief mention. First, it would be a mistake to suppose that Plato thought one could come to have a harmonious soul by the simple expedient of not having any desires at all. Indeed, this interpretation is so clearly absurd Plato does not even pause to consider it. Second, it is not clear whether it is the fact of conflict in one's soul or the perception of conflict this is the problem. Plato, as well as most of those who have been influenced by Freud, think it is the fact of conflict that is the evil. Yet it is not at all clear why this should be so. Third, it may be that some desires have a privileged position, such that if one satisfies them one will find that satisfying some others is either unnecessary or counter-productive. Plato holds that the desire for knowledge is of this type. As far as I can tell, he offers no argument for this claim, although, of course, the Platonic theory seems to require it.(12) Second-order desires also pose a problem. It does not follow from the fact that I want something that I want to want it, as any dieter can attest. Plato holds that our second-order desires (generated by the element of reason in our souls) are the ones to be preferred. But, although he gives us no argument for supposing that reason will provide us only with good, or proper second-order desires, perhaps a case may be made for thinking that reason will never provide me with a motive for seeking to develop a taste for torturing innocent barbarians.

Finally, Plato's theory demands that we think of justice (in a person) as being inversely proportional to the proportion of proper desires which go unsatisfied.(13) Why not say that it is proportional to the total amount of satisfaction or that it is inversely proportional to the total amount of unsatisfied desires? Nothing in Plato's writings suggests why we ought to favor his position over these alternatives. The most plausible position among these would appear to be that happiness is proportional to the total quantity of satisfaction, and not merely to the proportion of desires satisfied. This would include, as Plato surely wants to, the idea that it is not only existent desires that should count, but also desires one might come to have. This position and the complication posed by the inclusion of future desires are certainly problematic. Perhaps the best way to deal with them would be to point out that Socrates dissatisfied is happier than a pig satisfied. But Socrates satisfied would be even happier. Before leaving the concept of harmony, I want to say one favorable thing about Plato's idea that one good way to maximize our self-interest might be to rid ourselves of certain desires (especially those from the appetitive element) which are incompatible with other desires (usually those generated by the element of reason). We should all note that the reverse strategy is absurd. Imagine cultivating desires for both p and not-p on the grounds that, whatever happens, at least one of the desires will be satisfied. This seems about as sensible as the strategy of seeking to increase the number of true beliefs one has by believing (or at least claiming to believe) both p and not-p. Plato's analysis of harmony leads us, quite naturally, to a discussion of integrity. This is because Plato favors reaching harmony by having the reasoning element dominate the other two elements. This seems to be much like our conception of integrity. I take it that integrity consists in living up to one's commitments and living by one's principles. We may say that persons with integrity are those persons who do that which they have set for themselves to do. A person fails to have integrity when she succumbs to the seductive lure of lesser desires.

The connection between this conception of integrity and the conception of harmony discussed previously is obvious. The person with integrity is the person who can either resolve conflicts within herself in a fairly efficient way or prevent such conflicts from ever arising. She does this by relying on her principles, including the principle of keeping one's commitments to others, as the ultimate, and perhaps only, arbiter of what course of action to follow. This is also what Plato would say of the person with a harmonious soul; that reason provides that person with principle and, since reason is dominant in Plato's just person, these principles guide the person's actions.

At this point, one of the problems mentioned earlier in this paper becomes evident. We demand that the good person not only have principles, but that she have a certain kind of principles. Nothing we have said so far precludes the idea of a principled Nazi. Her problem may not be any lack of integrity or harmony in her soul. Rather, the problem is that she has the wrong principles.

Of course, Plato's just person does not have the sort of harmony that the principled Nazi has. Plato's "just" person has good principles. This brings us to my last two points. What argument has Plato got for thinking that the person with a harmonious soul will have the sort of principles (in terms of their content) that Plato thinks that person will have? I find no argument on this point (at least not in the Republic). Finally, it follows, that the best way to secure one's self-interest is to be what Plato calls the "just man". Hence, it is rational to be a "just man". What does not follow is that the "just man" is a good man; that being a "just man" is a morally worthy goal (though, of course, it does follow that being what Plato calls a "just man" is a prudentially worthy goal). Plato still has to show that (what he calls) a "just man" is in fact, a just person.

The first argument, which appears in Book IV, for this inference is that it is obvious that the Platonic "just man" is a good person.(14) A more compelling argument connects moral values to knowledge and the "just man" to knowledge occupies much of the rest of the Republic. Plato still needs the argument to convince those skeptics who refuse to grant that the "just man" is a good person and that consequently the Platonic conception of justice does connect justice with rationality in a manner superior to the connection offered by Glaucon in his defense of the contractarian position. Happily, Plato recognized this and saw that his account of the just person as one with a harmonious soul needed to be supplemented with an argument showing that the person with the harmonious soul necessarily would be a truly just individual. His account of why the person with a harmonious soul is necessarily a good person is provided by his theory connecting moral values to knowledge. The person with the harmonious soul will: (a) lack psychic dissonance, (b) only be governed by principles, and, most importantly, (c) will derive those principles from her knowledge of the Form of the Good. Plato brilliance lies not solely in his theory but also in anticipating problems which the theory might eventually confront. (15)


(1) Glaucon states the contractarian account of justice at Republic 358e to 367c. Plato also gives serious consideration to skepticism about justice. See Antony Flew, "Must Morality Pay or What Socrates Should Have Said to Thrasymachus" in Curtis L. Carter (editor), Skepticism and Moral Principles, New University Press, Evanston, 1973 especially section 3.

(2) Plato, of course, was unaware of the modern formulation of the prisoner's dilemma. But it is clear that he grasped the essential point that in certain situations what is individually rational leads to sub-optimal collective outcomes.

(3) Plato was the first to see clearly that contractarian accounts fails to have universal scope. There are two reasons why we are inclined to overestimate the importance Plato attaches to this objection. First, Plato manages to find such a graphic way of putting his objection and, second, due to Kant's continuing influence, we put much more stock in morality having universal application than Plato did. It does not apply to those who can get away with being immoral—presumably the very people it is most important to have accept the constraints of morality. This point is not widely appreciated because Glaucon introduces the contractarian account while he and Adeimantus are pressing the skeptical position raised by Thrasymachus. Hence, the case of the Lydian Shepherd is initially presented, not as an objection to the contractarian conception but as an explanation of why those who lack the power to do wrong with impunity would accept contractarian justice "as a mean between what is best—doing injustice without paying the penalty—and what is worst—suffering injustice without being able to avenge oneself". Republic 359a.

(4) That is to say, he must provide an ontology of value.

(5) Plato must provide an epistemology of value.For some, those for whom the Noble Lie is needed to move them to be moral, this is not the case.

(6) It is not sufficient to show that we can come to know what our actual self-interest is if, once we know it, this knowledge cannot motivate us to act.

(7) For a careful discussion of this part of the Republic see J.R.S. Wilson, "The Argument of Republic IV" in The Philosophical Quarterly, Volume 26, # 103, April 1976.

(8) Actually, the account of the soul assumed by Glaucon is closer to that put forward by Hobbes. Hume's account of the soul is more complex than Hobbes's or Glaucon's. In particular Hume allows that reason might tell an individual to cultivate new passions, something Hobbes does not consider. Furthermore, Hume's notion of the will and his principle of sympathy are, in the sense I will be using the term, rather "unHumean" additions to the core account of the soul.

(9) Treatise, Book II, Part III, Section III.

(10) Hume does hold that people have a will which is activated by both reason and by passion. He defines will as "an impression that we feel and are conscious of." But Hume's will does not correspond to any of the elements in Plato's account of the soul.

(11) The best discussion of the role of psychic tension in moral theory is Richard Brandt's A Theory of the Good and the Right.

(12) But see the Republic 582 a-d. J.S. Mill says similar things (Utilitarianism, chapter 2) in his argument that higher pleasures are superior to lower ones because those who have experienced both prefer the higher pleasures.

(13) For this point (and some that follow) I am indebted to Jan Narveson's "Morality, Integrity, and 'The Harmony of the Soul'" unpublished, undated, ms.

(14) The tripartite soul brings this out. Having the various virtues—wisdom, temperance, courage—are related to being just in that they are necessarily constantly conjoined. So, whoever is just is virtuous which, for Plato, is what it is to be good.

(15) I am grateful to Barry Curtis whose criticisms of my "Plato and the Social Contract" (Philosophy Research Archives, Volume XII, 1986-87) prompted me to write this paper.