Socrates seeks to define justice as one of the cardinal human virtues, and he understands the virtues as states of the soul. So his account of what justice is depends upon his account of the human soul.
According to the Republic, every human soul has three parts: reason, spirit, and appetite. (This is a claim about the embodied soul. In Book Ten, Socrates argues that the soul is immortal (608c-611a) and says that the disembodied soul might be simple (611a-612a), though he declines to insist on this (612a) and the Timaeus and Phaedrus apparently disagree on the question.) At first blush, the tripartition can suggest a division into beliefs, emotions, and desires. But Socrates explicitly ascribes beliefs, emotions, and desires to each part of the soul. In fact, it is not even clear that Plato would recognize psychological attitudes that are supposed to be representational without also being affective and conative, or conative and affective without also being representational. Consequently, ‘belief’ and ‘desire’ in translations or discussions of Plato (including this one) must be handled with care; they should not be understood along Humean lines as motivationally inert representations, on the one hand, and non-cognitive motivators, on the other.
The Republic offers two general reasons for the tripartition. First, Socrates argues that we cannot coherently explain certain cases of psychological conflict unless we suppose that there are at least two parts to the soul. The core of this argument is what we might call the principle of non-opposition: “the same thing will not be willing to do or undergo opposites in the same respect, in relation to the same thing, at the same time” (436b8–9). This is a perfectly general metaphysical principle, comparable to Aristotle’s principle of non-contradiction (Metaphysics G3 1005b19–20). Because of this principle, Socrates insists that one soul cannot be the subject of opposing attitudes unless one of three conditions is met. One soul can be the subject of opposing attitudes if the attitudes oppose each other in succession, even in rapidly alternating succession (as Hobbes explains mental conflict). One soul can also be the subject of opposing attitudes if the attitudes relate to different things, as a desire to drink champagne and a desire to drink a martini might conflict. Last, one soul can be the subject of opposing attitudes if the attitudes oppose in different respects.
At first glance, this third condition is unclear. The way Socrates handles putative counter-examples to the principle of non-opposition (at 436c-e) might suggest that when one thing experiences one opposite in one of its parts and another in another, it is not experiencing opposites in different respects. That would entail, apparently, that it is not one thing experiencing opposites at all, but merely a plurality. But Socrates later rewords the principle of non-opposition’s “same respect” condition as a “same part” condition (439b), which explicitly allows one thing to experience one opposite in one of its parts and another in another. The most natural way of relating these two articulations of the principle is to suppose that experiencing one opposite in one part and another in another is just one way to experience opposites in different respects. But however we relate the two articulations to each other, Socrates clearly concludes that one soul can experience simultaneously opposing attitudes in relation to the same thing, but only if different parts of it are the direct subjects of the opposing attitudes.
Socrates employs this general strategy four times. In Book Four, he twice considers conflicting attitudes about what to do. First, he imagines a desire to drink being opposed by a calculated consideration that it would be good not to drink (439a-d). (We might think, anachronistically, of someone about to undergo surgery.) This is supposed to establish a distinction between appetite and reason. Then he considers cases like that of Leontius, who became angry with himself for desiring to ogle corpses (439e-440b). These cases are supposed to establish a distinction between appetite and spirit. In Book Ten, Socrates appeals to the principle of non-opposition when considering the decent man who has recently lost a son and is conflicted about grieving (603e-604b) and when considering conflicting attitudes about how things appear to be (602c-603b). These show a broad division between reason and an inferior part of the soul; it is compatible with a further distinction between two inferior parts, spirit and appetite.
Socrates’ arguments from psychological conflict are well-tailored to explain akrasia (weakness of will). In the Protagoras, Socrates denies that anyone willingly does other than what she believes to be best, but in the Republic, the door is opened for a person to act on an appetitive attitude that conflicts with a rational attitude for what is best. How far the door is open to akrasia awaits further discussion below. For now, there are other more pressing questions about the Republic‘s explanation of psychological conflict.
First, what kinds of parts are reason, spirit, and appetite? Some scholars believe that they are merely conceptual parts, akin to subsets of a set. They would object to characterizing the parts assubjects of psychological attitudes. But the arguments from conflict treat reason, spirit, and appetite as distinct subjects of psychological states and events, and it seems best to take Socrates’ descriptions at face value unless there is compelling reason not to. At face value, Socrates offers a more robust conception of parts, wherein each part is like an independent agent.
Indeed, this notion of parts is robust enough to make one wonder why reason, spirit, and appetite are parts at all, as opposed to three independent subjects. But the Republic proceeds as though every embodied human being has just one soul that comprises three parts. No embodied soul is perfectly unified: even the virtuous person, who makes her soul into a unity as much as she can (443c-e), has three parts in her soul. (She must, as we shall see, in order to be just.) But every embodied soul enjoys an unearned unity: every human’s reason, spirit, and appetite constitute a single soul that is a unified source of that human’s life and is a unified locus of responsibility. (It is not as though a man is held responsible for what his reason does but not for what his appetite does.) There are questions about what exactly explains this unearned unity of the soul.
There are also questions about whether the arguments from conflict establish exactly three parts of the soul. Some worry that the discussion of Leontius does not warrant the recognition of a third part of the soul, and some worry that the appetitive part contains such a multitude of attitudes that it must be subject to further conflicts and further partitioning (and see 443e). Answering these questions requires us to characterize more precisely the kind of opposition that forces partitioning, in accordance with the principle of non-opposition, and to examine more carefully the broader features being attributed to the three parts of the soul.
Fortunately, the arguments from conflict do not work alone. Indeed, they cannot, as the principle of non-opposition merely establishes a constraint on successful psychological explanations. Appeals to this principle can show where some division must exist, but they do not by themselves characterize the parts so divided. So, already in Book Four’s arguments from conflict, Socrates invokes broader patterns of psychology and appeals to the parts to explain these patterns (cf. 435d-436b).
This appeal to reason, spirit, and appetite to explain broader patterns of human thought and action constitutes the Republic‘s second general strategy to support tripartition. It receives its fullest development in Books Eight and Nine, where Socrates uses his theory of the tripartite soul to explain a variety of psychological constitutions. In the most basic implementation of this strategy, Socrates distinguishes people ruled by reason, those ruled by spirit, and those ruled by appetite (580d-581e, esp. 581c): the first love wisdom and truth, the second love victory and honor, and the third profit and money. This simplistic division, it might be noted in passing, fixes the sides for an ongoing debate about whether it is best to be a philosopher, a politician, or an epicure (see, e.g., Aristotle, Nicomachean Ethics I 5 and X 6–8). But more important for our purposes here, this basic classification greatly illuminates the division of the soul.
First, we learn about the organizing aims of each of the psychological parts. In Book Four, reason is characterized by its ability to track what is good for each part and the soul as a whole (441e, 442c). In Book Nine, reason is characterized by its desire for wisdom. These are not bifurcated aims. Socrates argues that no one is satisfied merely with what he or she takes to be good for him- or herself but wants what is in fact good for him- or herself (505d). So reason naturally pursues not just what it takes to be good for the whole soul but also the wisdom that ensures that it would get this right. Nor is wisdom’s value merely instrumental to discovering what is good for one. If wisdom is a fundamental constituent of virtue and virtue is a fundamental constituent of what is good for a human being, then wisdom turns out to be a fundamental constituent of what is good for a human being. So it should not be surprising that the part of the soul that tracks and pursues what is good for the whole soul also loves wisdom. Spirit, by contrast, tracks social preeminence and honor. If ‘good’ is the organizing predicate for rational attitudes, ‘honorable’ or ‘fine’ (Greek kalon) is the organizing predicate for spirited attitudes. Finally, appetite seeks material satisfaction for bodily urges, and because money better than anything else provides this, people ruled by appetite often come to love money above all.
The basic division of the world into philosophers, honor-lovers, and money-lovers also illuminates what Socrates means by talking of being ruled by one part of the soul. If one part dominates in you, then aims of that part are your aims. If, for example, you are ruled by spirit, then your reason conceives of your good in terms of what is honorable. Reason has its own aim, to get what is in fact good for the whole soul, but in a soul perfectly ruled by spirit, where there are no genuine psychological conflicts between different parts, reason’s love for truth and wisdom must be limited to that which is also held to be honorable.
Still, Plato’s full psychological theory is much more complicated than the basic division of persons would suggest. First, there are different kinds of appetitive attitudes (558d-559c, 571a-572b): some are necessary for human beings; some are unnecessary but regulable (“lawful”), and some are unnecessary and entirely uncontrollable (“lawless”). So there are in fact five kinds of pure psychological constitutions: aristocratically constituted persons (those ruled by their rational attitudes), timocratically constituted persons (those ruled by their spirited attitudes), oligarchically constituted persons (ruled by necessary appetitive attitudes), democratically constituted persons (ruled by unnecessary appetitive attitudes), and tyrannically constituted persons (ruled by lawless appetitive attitudes).
The second complication is that some people are not perfectly ruled by one part of the soul, but are subject to continuing conflicts between, say, attitudes in favor of doing what is honorable and appetitive attitudes in favor of pursuing a shameful tryst. Socrates does not concentrate on these people, nor does he say how common they are. But he does acknowledge their existence (544c-d, cf. 445c). Moreover, the occurrence of akrasia would seem to require their existence. For if I am perfectly ruled by my spirit, then I take my good to be what is honorable, and how could I be akratic? My spirit and my reason are in line, so there will be no overpowering of rational preferences about what is best by spirit. You might suppose that my appetite could overcome my sense of what is honorable, but in that case, it would seem that I am not, after all, perfectly ruled by my spirit. Things might seem different with people ruled by their appetite. Certainly, if I were perfectly ruled by appetite, then I would be susceptible to akrasia of the impetuous sort, acting on appetitive desires without reflectively endorsing them as good. But impetuous akrasia is quite distinct from the standard akrasia in which I endorse φing as best for me and at just that moment intentionally ψ instead, and standard akrasia would seem to be impossible in any soul that is perfectly ruled by any one part of the soul. If you think that competing appetitive attitudes could give rise to a strict case of standard akrasia, you should recall how Socrates would have to explain these cases of psychological conflict in order to avoid multiplying his divisions in the soul.
The general strategy of the Republic‘s psychology—to explain human thought and action by reference to subpersonal homunculi—remains both appealing and problematic. Moreover, the dialogue is filled with pointed observations and fascinating speculations about human psychology. Some of them pull us up short, as, for example, the Freudian recognition of Oedipal desires that come out only in dreams (571c-d). The full theory is complex, and there remain numerous questions about many of its details.