SOURCE: Aune, Bruce. “The Unity of Plato's Republic.” Ancient Philosophy 17, no. 2 (fall 1997): 291-308.

[In the following essay, Aune investigates charges of structural disunity between the two books of the Republic, maintaining that a close examination of the two parts reveals a style and method of inquiry in part II that are very similar to those of part I.]

How well does Republic i fit together with the books that follow? Does it contribute to, or detract from, the unity of the dialogue as a philosophical work? There is still disagreement about this matter.¹ Irwin 1995, 169 speaks of book 1 as the first of two ‘long introductions’ to the dialogue, the other being book 2; and he offers the supposition that, even though it seems to have more in common with the Laches and Charmides than with the rest of the Republic, it sketches some of the conclusions Plato meant to defend in books 2-10. By contrast, the late Gregory Vlastos, greatly impressed by similarities he saw between book 1 and a group of dialogues that contains the Laches and Charmides,² argued that the philosophical content of book 1 is seriously at odds with the rest of the Republic and features a method of inquiry in which Plato, when he wrote the later books, had actually lost confidence.³ Vlastos argued that the Socrates of book 1 is an early Socrates who pursues a philosophy so different from the philosophy of the Socrates of books 2-10 that ‘they could not have been depicted as cohabiting the same brain throughout unless it had been the brain of a schizophrenic’ (1991, 46 and 248-251). Different as their views are, I think neither Irwin nor Vlastos has the right perspective on book 1. A close reading shows it to be similar to the dialogues Irwin mentions only in superficial respects, to contain (in compressed form) the general theory that Socrates develops in the remainder of the dialogue, and to employ a method of inquiry that is not significantly different from that of the other books. I shall support these claims in the pages to follow.

The similarity that both Vlastos and Irwin see between book 1 and the dialogues they have in mind is owing to a method of inquiry and a characteristic outcome that they find in those dialogues. They call the method ‘the elenchus’.⁴ When Socrates employs it, he induces an interlocutor to offer a definition of (or thesis concerning) something and then, when it is given, proceeds to show that it is inconsistent with one or more propositions that the interlocutor, stimulated by Socrates' questions, suggestions, occasional assertions, and inferences, acknowledges to be true. The interlocutor, finding himself committed to the inconsistency, realizes that he does not know what he thought he knew. The typical outcome is ἀπορία: the interlocutor is ‘at a loss’ about what to believe in the circumstances. In dialogues such as the Laches and Charmides Socrates avows that he shares the ignorance of his interlocutors; he disavows all knowledge of the subject (always some virtue) being investigated. He expresses the same sort of ignorance, apparently, at the end of book 1.

The elenchus, as I have described it, is essentially a critical, destructive method, one by which an inconsistency is obtained and an interlocutor's ignorance is shown.⁵ Employing such a method is compatible, obviously, with employing other, more positive methods—and there is no question, I believe, that Socrates actually employs some other methods (at least in a limited way) in all his elenctic dialogues. One such method is epagôgê (ἐπαγωγή), as Aristotle (Meta. 1078b27-30) called it, which Socrates uses when he draws universal conclusions from particular instances—as he does, for example, in Euthyphro 13b when, having made certain observations about the arts of the horseman, huntsman, and shepherd, he concludes that attending to something always involves aiming at some good or benefit for the object of one's attention.⁶ Vlastos does not acknowledge epagôgê as a method distinct from the elenchus, but he emphasizes that Socrates actually draws some positive ‘moral’ conclusions in the Gorgias, a dialogue that he regards as an early elenctic one.⁷ Vlastos does not think these positive conclusions—e.g., that doing injustice is worse than suffering injustice and that being punished for an injustice is better than not being so punished—are obtained by a special method, however; he thinks they are based on a special assumption that Socrates is tacitly making, an assumption whose patent unacceptability, he believes, undermined Socrates' faith in the elenchus.⁸ I think there is no doubt that Socrates manages to introduce, either by inference or outright assertion, some positive doctrine in the dialogues Vlastos compares with book 1; but in dialogues that approach the ideal of a purely elenctic conversation most closely—the Laches and Charmides, which Irwin cites, are perhaps the best examples—this positive doctrine seems minimal. In these dialogues Socrates comes close to doing what, in the Apology, he says he does—namely, to ‘give aid to the god’ and show that people who seem to be wise about some important matter really are not so (Apol. [Apology] 23b, c).

Although I disagree with Vlastos' view of book 1, I agree that Socrates does employ the elenchus in it: His initial arguments with Cephalus and Polemarchus are obviously elenctic, and the large-scale features of his extended arguments with Thrasymachus accord with the description I gave of that form of argument. But these admissions do not require me to acknowledge that book 1 is merely an elenctic dialogue, or significantly similar to the ones Vlastos and Irwin have in mind. In fact, book 1 is significantly different from those dialogues. The detailed structure of the book 1 arguments—particularly the arguments with Thrasymachus, which occupy most of the book—have so much positive content that their elenctic character is almost incidental to them. Socrates effectively refutes Thrasymachus' total position, but he does so only by developing a theory of justice that is incompatible with it. In developing this theory he uses methods, moreover, that are constructive and not significantly different from the methods he uses later on. Since I am principally concerned to show that the philosophical content of book 1 is not only compatible with the doctrines developed in the later books but part and parcel of them, I shall begin with Socrates' arguments with Thrasymachus and leave his less constructive criticism of Cephalus and Polemarchus to my concluding section.

SOCRATES' ARGUMENTS WITH THRASYMACHUS

To appreciate the positive upshot of Socrates' discussion with Thrasymachus, it is helpful to recall the abstract structure of that discussion. It begins as a typical elenchus. After persuading Thrasymachus to provide a definition (or account) of what justice is, Socrates subjects him to a series of questions that result in an admission inconsistent with that definition (339d-343; the admission is noted at 340a). But Thrasymachus is not daunted by the inconsistency, and Socrates does not pursue the matter in a straightforward way. Instead of attempting to keep Thrasymachus focused on their initial subject of discussion (the nature of justice, δικαιοσύνη) he allows Thrasymachus to introduce a new thesis about justice, one concerning the attitude it is reasonable to take toward being just. In sharp contrast to what Socrates, Cephalus, and Polemarchus had been assuming, Thrasymachus declares that being unjust is really preferable to being just for someone whose eyes are open; if such a person has the power to be unjust on a large scale, he (or she) will be particularly well off (344a). Apparently setting aside the definition Thrasymachus initially proposed, Socrates proceeds to attack this new thesis, but his argument is complex and indirect. He refutes the new thesis and, incidentally, Thrasymachus's original definition of justice by developing an alternative account of justice (a theory of justice, as we should say today) which is inconsistent with the thesis and the definition.

What theory of justice does Socrates develop in his argument against Thrasymachus? The basic ideas of the theory are easily given, but their elaboration requires extended discussion. I begin with the basic ideas: Justice is something that can exist in a person and in a society (351c-352b). In both places it is a virtue (ἀρετή), something on account of which a function or job (ἔργον) is done well (353c). The jobs in question are jobs of ruling (ruling a self and ruling a polis—353d, 347c, d), and there are arts (τέχναι) in accordance with which these jobs are best conducted (342e, 345e). The self-governing job is best conducted in accordance with what might be called ‘the art of living’ (344e); the other job is best conducted in accordance with the art of government or politics. The results of good government in these cases are humanly valuable. A person who governs him- or herself well is just and happy; the society or polis that governs itself well is also just and happy. Both are εὐδαίμονεs: fortunate, or ‘well-off’ (352d).

Although Socrates develops these points in the course of his arguments with Thrasymachus, he does not develop them in the order in which I have listed them. His initial points emerge in his criticism of Thrasymachus' definition of justice as ‘the interest of the stronger’—a definition implying, when clarified by Thrasymachus, that justice is something defined in a given society by the laws and customs (νόμοι) in effect there, which serve the interests of the party in power (338e). Socrates begins his criticism by asking whether a ruler might make a legislative mistake and enact laws that do not serve his interest. Although Thrasymachus initially replies that such a mistake is possible, he soon reverses course and declares that a ruler in the precise or strict sense could not actually make a mistake of this kind: proper δημιουργοί cannot err, he says, in the performance of their τέχναι. This declaration accords with Socrates' premise that there is an art to ruling, but it misidentifies the aim of that art. As Socrates proceeds to argue, the proper aim of the ruler (the aim imposed by the ruler's art) is to care for the ruled and to serve their interests. The laws the ruler creates qua ruler—the laws that define justice in a properly conducted society (338e)—do not, Socrates concludes, represent the interest of the stronger there; they represent the interest of the weaker, the people who are ruled (339d-343). This conclusion, which relates justice (justice in a society) to the τέχνη of a proper or ideal ruler, a ruler in what Thrasymachus calls the ‘strict’ (ἀκριβη̑) sense, is a basic tenet of Socrates' theory.

Further points emerge when Thrasymachus stubbornly refuses to accept Socrates' conclusion as a refutation of his definition and propounds an additional thesis about justice, a thesis concerning the value of having this supposed virtue. Socrates proceeds to clarify his position by emphasizing that a δημιουργόs, in addition to practicing a given τέχνη, always practices further, higher-order τέχναι whose purposes may be accomplished by the successful practice of the lower-order one. As he explains, a physician who, in practicing medicine, is properly concerned to benefit his patients rather than himself is also apt to be practicing the higher-order art of wage-earning and may, by the successful performance of his healing art, secure the benefit he seeks as a wage-earner (346c-346e). This subordination of art to art (or τέχνη to τέχνη) can be extended further than Socrates suggests—and Plato, at a later stage of his argument, expects his readers to be aware of this fact, for the ultimate aim of human τέχναι is human happiness.⁹ Thus, if a man practices medicine to earn wages, he may do the latter to support himself and his family (there is an art to this, too); and he may practice this further art to earn the respect of his fellow-citizens (another art)¹⁰ and therefore promote his self-respect (an art), which is essential for being a good Athenian (also an art) and, ultimately, for living a good life (the highest art of all).

If Thrasymachus were right in what he says about the value of being just, justice in a person would not be an admirable trait: a just person would not be ‘wise or good’. Socrates attacks this conclusion in his next argument with Thrasymachus, an argument concerned with the relation between having a virtue (ἀρετή) and being accomplished in, or ‘wise and good at’, something—some τέχνη, in view of his examples (348b-350d). He argues that people who are wise and good at something (= have some virtue) do something, some job, well; and he implies that there are standards in accordance with which these jobs are best conducted: different people who are knowledgeable and good at these jobs do them in the same way (350a). Since the characteristic activity of a person who practices injustice is not relevantly like the characteristic activity of someone who is wise and good at something,¹¹ the unjust man is not virtuous—not wise and good at something. His opposite, however, is: the just person is like those who are wise and good at some task. Justice, then, is virtue and wisdom, Socrates concludes, and injustice is vice and ignorance (τήν δικαιοσύνην ἀρετὴν εαναι καὶ σοφίαν, τὴν δὲ ἀδικίαν κακίαν τε καὶ ἀμαθίαν, 350d).

In two further arguments directed against Thrasymachus' view about the value of being just, Socrates develops additional elements of his conception of justice. The first of these arguments is based on the fact that cities or groups of people can be just or unjust as well as individuals (351c-352b). Socrates begins by contending (by getting Thrasymachus to agree) that injustice in a group of people detracts from the group's ability to achieve a common purpose. He supports this premiss by observing that when injustice exists among the members of a group, they hate one another, fight, and refuse to cooperate; but that when justice exists among them, the effect is the opposite: there is friendship and a sense of common purpose. Claiming that injustice does not lose its power to create dissension even in the smallest group, he declares that injustice has a similar effect in individual people, making them at odds with themselves, not of one mind, and in state of internal civil war. People in such a state are incapable of achieving anything, he says, and are enemies of themselves, other people, and even the gods. Observing that justice has the opposite effect in the individual as well as in the group, he concludes that justice, wherever it occurs, is more profitable than injustice and that Thrasymachus' claim about the advantages of being unjust is therefore erroneous.

Socrates' next argument reintroduces the idea that a virtue is associated with a task or job, but it extends the notion of a task or job (ἔργον) in an important way. Socrates begins by getting Thrasymachus to acknowledge that natural things (animals, plants, their organs) as well as artifacts have functions, or do some characteristic work. This is true not only of horses but of eyes, ears, and pruning knives (352e). Thrasymachus also agrees with Socrates' proposed definition of a function: The function of a thing, X, is what only X can do, or what X can do best.¹² Thus, the function of eyes is to see; the function of ears is to hear; and the function of a pruning knife is to prune trees, shrubs, and vines. Socrates and Thrasymachus also agree that everything having a function has, at least potentially, a corresponding virtue (ἀρετή). Since (as Socrates emphasized earlier—346c) things can often do more than one job and, therefore, have more than one function and more than one virtue, the relevant notion of a virtue should be defined in a ‘with respect to’ way: The virtue of X with respect to the function F = that by which X performs F well.

Like other natural things, the human soul (ψυχή) has a function, Socrates says. To this he adds: The function of the soul is to plan, rule, take care of things—and also to live (353d). Socrates then says, and Thrasymachus agrees, that they have agreed that justice is the virtue of the soul. Since the virtue of a thing is that by which it performs its function well, it must be more profitable to have a just soul than to have an unjust one. The former does its work—of planning, ruling, living, and taking care of things—well, while the latter does not; and it cannot be more profitable to have a soul that does not do these things well. Such a soul would not ‘live well’! This completes Socrates' argument against Thrasymachus' claim about the value of being just. Thrasymachus must be wrong about this.

One thing patently problematic about this last argument is Socrates' premiss asserting that he and Thrasymachus ‘have agreed’ that justice is the virtue of the soul. Did they really agree to such a thing?¹³ Not in so many words. They did say, in the argument against Thrasymachus' claim that injustice is a virtue, that ‘justice is virtue and wisdom’ (meaning that being just is being good and wise at something—see 350d),¹⁴ but this is not the same as saying that justice is the virtue of the soul, for the latter implies that being just is being good at the soul's proper job. Yet they agreed in the preceding argument, where they compared unjust souls with unjust cities or groups, that injustice causes people to be at odds with themselves and in a state of internal civil war, so that they are incapable of accomplishing anything. They also agreed there that justice has the opposite effect—consisting, in the individual, in being of one mind and in a state of psychic harmony, so that one is capable of accomplishing one's goals. Obviously, anyone in this state has a soul that does its work (performs its function) well. Since it does this work because of its justice, and since the virtue of a thing with respect to a certain function is defined as that by which it performs the function well, it follows that justice is the virtue of the soul. Thrasymachus and Socrates can be said to have agreed tacitly on this assertion, if they have not done so explicitly.

An important qualification should be made at this point. According to the argument in which unjust individuals were compared with groups, injustice and therefore justice exist in groups as well as individuals. If this argument is tenable, it is a mistake to identify justice with the virtue of an individual soul: only personal justice, or justice in a person, can be understood this way; social justice requires a different definition. Such a definition is not difficult to give, however. If personal justice is the virtue of the individual soul, social justice is the virtue of a group or community: it is that by which (= that by virtue of which) the group does its job successfully. Socrates did not say, at least thus far, what the job (the function, ἔργον) of a group is, but his claims about the effects of injustice in a group make it clear that he regards the performance of that job as some kind of cooperative activity, one that serves the purpose for which the group exists. This conclusion has important implications for the general argument between Socrates and Thrasymachus.

As I have emphasized, Thrasymachus had two theses about justice, a definition of its nature and a claim about the value of possessing it. Socrates explicitly concluded that his arguments undermined the latter thesis, but it is obvious that they undermine the first thesis, the definition, as well. If personal justice is the virtue of the soul, and social justice is some kind of cooperative activity promoting the purpose of the group, justice cannot be identified with ‘the interest of the stronger’—that is, with something identified by the laws, the νόμοι, created by the stronger element in society, for those laws are often self-serving and sometimes even tyrannical. Although Socrates does not identify this implication of his criticism, we should not represent him as being unaware of it. His arguments against Thrasymachus' second thesis indirectly apply to the first one as well.

CONNECTIONS BETWEEN BOOK 1 AND SUBSEQUENT BOOKS

When Socrates completes his final argument against Thrasymachus, he is met with the bitter or sarcastic reply: ‘Let that [your victorious argument] be your banquet … at the feast of Bendis’.¹⁵ Socrates' response, surprisingly, is that he did not have a fine banquet: he behaved like a glutton, he says, snatching every dish that passed and tasting it before properly savoring its predecessor. Before finding out what justice is, he turned to the question whether it is a kind of vice and ignorance or a kind of wisdom and virtue. An argument then came up, he says, about injustice being more profitable than justice, and he could not refrain from abandoning the previous one and following up on the new one. As far as he is concerned, he says, the result of the discussion is that he ‘knows nothing’, for if he does not know what justice is, he hardly knows whether it is a kind of virtue or not, or whether a person who has it is happy or unhappy (354b-c).

This response is often taken to support the view that Socrates' argument in book 1 is comparable to his arguments in distinctively elenctic dialogues such as the Laches and Charmides. The argument has been critical, elenctic; and its outcome is not positive but ‘aporetic’: it ends in perplexity, ἀπορία.¹⁶ I have acknowledged that Socrates' argument bears some formal similarity to his arguments in distinctively elenctic dialogues, but I have insisted that the similarity is limited to the large-scale destructive features of his arguments, not to their positive theoretical content. In extracting admissions from Thrasymachus, or drawing consequences from other admissions,¹⁷ that are inconsistent with Thrasymachus' theses about justice, Socrates proceeds in a generally elenctic way, but he achieves his negative result (that Thrasymachus is wrong about justice and its value) only by developing a positive view of justice that requires it. This positive view is not supported elenctically, by negative criticism; it is supported by analogies (e.g., between injustice in groups and injustice in individuals) and by deduction from premises conceded to be reasonable (e.g., that things have functions and that virtues are related to functions in a special way). And although Thrasymachus cannot elude Socrates' criticism, he never concedes that he is wrong about justice, nor does he experience the ἀπορία that beset Socrates' interlocutors in more distinctively elenctic dialogues. Even Socrates himself is not genuinely perplexed by the subject of the conversation, as he appears to be in such dialogues. He says he ‘knows nothing’ about justice, not even ‘whether it is a kind of virtue’, but his expression of utter ignorance is implausible and unconvincing. He knows far more than he is admitting.

It is by no means obvious why Plato represents Socrates as claiming to know so little about something he has said so much about.¹⁸ It is not as if Socrates is prepared to abandon what he said. As a matter of fact, when, in book two, Glaucon and Adeimantus have completed what Glaucon calls their ‘renewal’ of Thrasymachus's argument and ask Socrates to respond, showing what effects justice and injustice have ‘because of themselves’ on individual souls (367e), Socrates replies that he thought what he said to Thrasymachus ‘showed (ἀποφαίνειν) that justice is better than injustice’ (368b). (How could he think he showed this—particularly in view of the arguments he used—if he ‘knows nothing’ about justice?) Also, when Socrates proposes, later in book 2, that they find out what sort of thing justice is in a polis and afterwards look for it in the individual—saying that if they could watch a city coming to be in theory, they would also see its justice coming to be (369a)—he does not raise any question about how they could identify a city's justice if they had no knowledge of what justice is. He does not raise this question, the context shows, because he knows how to identify a city's justice: the things he said about justice in his argument with Thrasymachus provide the necessary directions—and he follows them without bothering to announce that he is doing so.¹⁹

According to his earlier argument, personal justice is the virtue of an individual soul, an entity that is analogous, in moral respects, to a city or group. Both can be just, unjust, good, bad, enlightened, ignorant, and so on. In view of this analogy, social justice can be considered the virtue of a city or group. Now, the virtue of a thing is, Socrates argued, that by which it performs its function (does its job) well. Since a thing's job is, as he said, what it alone can do or what it can do best, to identify the function (ἔργον) of a polis we have only to consider what such a thing is uniquely capable of doing or can do best. Socrates does just this in book 2. Tacitly restricting the relevant doings here to those promoting the existence or welfare of social beings, he identifies the function (‘according to nature’) of a polis as that of meeting the needs of individual persons by utilizing the talents present in the group (see 369c, 428e). No one is self-sufficient, but each person has valuable talents. The group is uniquely capable of satisfying all genuine needs by using all available talents.

I think these last considerations—namely, Socrates' avowal about what he thought he had proved and his subsequent reliance on claims he had made in his arguments with Thrasymachus—make it quite certain that Socrates is not represented as having lost confidence in those arguments. Of course, those arguments did not purport to deal with justice in a detailed way: the view (or theory) of justice that emerges from them is highly compressed (or abbreviated), and the careful reader cries out for details. The tight connection between book 1 and the rest of the Republic is shown by the fact that these later books are largely concerned with working out the relevant details.

Here are some of the details (not minor details, of course!) that are worked out in later books. First, if justice can exist both in the individual soul and in a city, then an individual soul must be relevantly like a city. (This inference is supported by a premiss that Socrates used in one of his argument against Thrasymachus. In that argument [see 348b-350d] he claimed that if injustice is a virtue, as Thrasymachus' second thesis implies, then an unjust person must be relevantly like people who have other virtues; they must ‘have the quality’ of those they resemble.) But how can a soul be like, or have the quality of, a city? It is not enough to say, as he does at 352a, that a person ‘can be at odds with himself and not of one mind’; this does not imply that a soul is like a city. Socrates answers the question, however, in book 4, when he develops the view that the soul has three parts corresponding to the principal divisions (or parts) of the ideal polis, the polis created ‘according to nature’ (κατὰ φύσιν).²⁰

Second detail: If the justice in a city and in a soul is ‘that by which’ they do their job well, what is the intrinsic nature of these states? What, apart from their effects, are they like? Socrates' answer—the most detailed answer he actually gives—is that these states are harmonious arrangements among the parts of the city and the soul in which (see 434a) each part obtains its proper benefit and does its proper task (τε καὶ ἐαυτου̑ α̑ξιs τε καὶ πρα̑ξιs).²¹ To make sense of, and ideally to identify, these proper tasks and proper benefits, one has to have knowledge of the forms, Socrates thinks, and the relation of lower forms to the form of the good. Socrates discusses these matters in books 5, 6, and 7. He also shows, in book 9, how the opposing states of disharmony give rise to typical manifestations of injustice, such as stealing, lying, and slandering.

Third detail: If justice is the virtue of the soul or a community, what about the other cardinal virtues? Are not wisdom (σοφία), courage (ἀνδρεία), and moderation (σωφροσύνη) also virtues of the soul and the community? Socrates gives the answer to this question in book 4: what he says there implies that these other so-called cardinal virtues are parts or aspects of justice. If a soul is just, its parts must by his definition do their proper jobs; if they do, the soul will have these other virtues as well. Thus, if the rational part does its job, the soul will have the virtue of wisdom; if the spirited part does its job, the soul will have the virtue of courage; and if the appetitive and spirited parts recognize the leadership of the rational part and do the work it assigns them, the soul will have the virtue of moderation (see 442b-d).

Fourth detail: If a just person is really better off than an unjust one, exactly why is this? What does being unjust do to the individual soul? Socrates' specific answer to this last question emerges in book 9, where he pictures an unjust person as a man feeding a many-headed monster and a lion, both growing within him; but he also supports it by arguments about the nature and effects of justice on a person, which he develops in books 8 and 10 as well as 9.

Apart from details that are worked out in the ways I have mentioned, some of the key claims that Socrates made in book 1 are qualified in subsequent books. A crucial qualification concerns Socrates' definition of a thing's function or job. In book 1 he defined the function of a thing X as what only X can do or what X can do best, but this definition was merely provisional and not fully acceptable, all things considered. Socrates exposes its ultimate inadequacy when he discusses the role of women in the ideal polis.²² Although bearing children is something that only women can do, bearing children is certainly not the job or function that women should have in an ideal community. Children should, of course, be born in such a community, and women have to bear them; but bearing children is at best an occasional act, not a social role or permanent job. (Bearing children is not the same as carrying them inside you or caring for them once they are born: the former is not, in any case, a permanent job, and the latter is ideally a job for people, not necessarily women, who have a distinctive talent for it: 460b, c.) To identify a natural thing's proper function or job, Socrates concludes, one must understand its ideal role in nature—something one can do only by apprehending the form corresponding to it and the relation of that form to the other forms that, in accordance with the form of the good, collectively define the teleological structure of the κόσμοs.²³ The need to rectify the provisional notion of a thing's proper function that was introduced in book 1 and formed the basis of Socrates' positive argument there actually motivates Socrates' apparent digression about the role of women and children in the ideal polis. To perfect that argument—to qualify a key assumption and to add pertinent details—Socrates has to introduce the metaphysical and epistemological doctrines that appear in books 5 through 8.

THE OPENING CONVERSATIONS OF BOOK 1

Having pointed out what I take to be some undeniably tight connections between book 1 and the books that follow, I now want to say something about the opening conversations of book 1 in which Socrates appears to employ the elenchus in a largely negative way—as he did in such dialogues as the Laches or Charmides. The first point I want to make about these conversations is that although their upshot is primarily negative, they introduce themes that are developed later in the dialogue, and they prepare the way for the constructive arguments with Thrasymachus that dominate book 1. Their presence in book 1 is not therefore characteristic of the whole book and a satisfactory basis for assimilating it to purely (or largely) elenctic dialogues. The next, much more important point I want to make about these conversations is that the sort of refutation featured in them is in no way incompatible with the dialectical method that Plato (or the character, Socrates) endorses in book 7. Such refutation is, in fact, an important part of that method, and the specific elenctic arguments that Socrates employs in these initial conversations conform to the argumentative strategy of the entire dialogue.

To see the basis for my first point, note that although Polemarchus, interpreting Simonides, initially claimed no more than that it is just to help friends and harm enemies, Socrates interpreted him as advancing the doctrine that justice is a τέχνη—one of benefiting friends and harming enemies (332d). In refuting this doctrine, Socrates prepares the way for the alternative view he will propound in his argument with Thrasymachus—namely, that justice is not a τέχνη but an ἀρετή, something by which a job or function is well performed and something that may be brought about by the successful practice of a τέχνη (see my summary of Socrates' view on p. 294 above.) The idea that being just is (that is, =) having a certain τέχνη is easily confused with the idea that Socrates will defend;²⁴ so it is important for him to distinguish one idea from the other. In view of the line he will take against Thrasymachus, this is what he is, in effect, doing in his initial arguments with Polemarchus (332b-334b). If being just is having the τέχνη of helping friends and harming enemies, and being unjust is having the opposite τέχνη, then a just person would have to be an unjust one, since one who has the first art would also have the second. But this consequence is absurd; being just cannot be identified with having the τέχνη in point. Being just cannot be identified with having any τέχνη at all.²⁵

The arguments against Polemarchus' doctrine that I have just mentioned are almost entirely destructive, but Socrates develops an interesting positive claim when he argues that it is never just to harm anyone (335e). His argument involves the suggestive idea that harming a thing is damaging it in a way that renders it less excellent. This idea is very plausible, but Socrates draws the dubious conclusion that when humans are harmed, they are made less just—and it cannot be just, he says, to make another person less just, even if that person is an enemy. Socrates' conclusion is dubious because a person can evidently be harmed (and so made less excellent) in a purely physical way by losing a limb or being blinded—in which case being harmed would not entail being rendered unjust. Of course, if one accepts the view Socrates suggests in book 10 (621c-d) and defends in Phaedo 67c—that a living person is really an immortal soul temporarily lodged in a body—one will not accept this criticism, for an immaterial soul cannot be harmed in a physical way. Since Socrates declares, in one of his arguments against Thrasymachus, that justice in a person is the virtue of a soul (see 353e), he is committed to the idea that the soul must be made less just if, harmed, it is made less excellent. What appears to be a defective—perhaps even a sophistical—argument against Polemarchus (‘If a person is harmed, he or she is made less just’) is thus rendered plausible, at least in Socrates' eyes, by one of his arguments against Thrasymachus.

I turn now to my next point—namely, that the sort of elenctic refutation featured in Socrates' arguments with Cephalus and Polemarchus is in no way incompatible with the dialectical method that Socrates endorses in book 7. It will be helpful to approach this point by way of some general reflections on the Republic as a whole.

In spite of the positive doctrine with which the dialogue abounds, it is important to realize that what the Republic depicts is a quest rather than a finished result. To know how to fine-tune a polis or a human soul so that the result is ideally virtuous, one must have knowledge of the good, Socrates says—and this knowledge can be achieved, he implies, only after one has completed an ideal course of philosophical training. Speaking of the higher education of rulers, he says: ‘At the age of fifty, those who have survived the tests and been successful both in practical matters and in the sciences must be led to the goal and compelled to lift up the radiant light of their souls to what itself provides light for everything. And once they have seen the good itself, they must each in turn put the city, its citizens, and themselves in order, using it as their model’ (540a). But the good is not presented in the Republic; it is described there only by an analogy (a very abstract one) with the sun. Readers of the Republic are not, therefore, given the model by which to put their city and their souls in order. They have been given some directions, some helpful details, but to reach the desired goal they have a lot of philosophical work to do.

In book 7 Socrates tells us that this philosophical work is to be accomplished by means of dialectic. Those who practice this method ‘attempt to grasp with respect to each thing itself what the being of it is’ (533b), seeking in discourse a unified vision in which the form of the good is distinguished from everything else (534c). Yet to achieve this vision they must ‘survive all refutation’, coming through it (as from a battle) with their account ‘still intact’. Unlike a young man who plays at contradicting others for sport, older persons taking up dialectic at the proper time (at age 30, Socrates says) will take part in arguments, but they will do so only to discover the truth (533-540). Their elenctic arguments, like those of Socrates at the beginning of book 1, will thus accord with part of their epistemic strategy.

Socrates' fundamental concern with the right (or best) epistemic method is, in fact, probably represented in the first sentence of the Republic, where Socrates remarks that he went down to the Piraeus the day before to say a prayer to a goddess who is identified, later, as Bendis. As it happens, Bendis was the Thracian version of Artemis, a moon goddess associated with childbirth and the hunt. Since dialectic, as Socrates describes it, is clearly a hunt, the ultimate quarry of which is the good, and since Socrates is a self-styled midwife (Tht. [Theaetetus] 149a-151b), the reference to Bendis (in a supremely accomplished writer who did not do anything at random) is an indication, I suspect, that the dialectical process to be followed is particularly important in this dialogue.

This suggestion, which one might initially think is excessively fanciful, is reinforced by the presentation of alternative epistemic methods in the conversations of book 1. The basic aim of the dialogue (more basic than the queries about justice) is to discover the best way for a person to live—or, as Socrates describes it in his conversation with Thrasymachus, ‘to determine which whole way of life would make living worthwhile for each of us’ (344). But how can a matter like this be ascertained? What strategies are available? One obvious strategy—a strategy that would occur to a secular, twentieth-century thinker right away—is to find an intelligent person who has lived a long time with evident success (not unhappy, bitter, or defensive) and ask her/him what she/he thinks. This is exactly what Socrates does. He starts a conversation about life with Cephalus, an intelligent old man who appears to have had a successful life. Cephalus offers pretty good advice about the best way to live (it comes close to what Socrates recommends in book 9 for an ordinary person living in an imperfect society);²⁶ but Socrates induces him to give a definition (or interprets him as having given one) of being just, and he is quickly refuted. The implication? Experience and intelligent good sense, at least by themselves, provide an inadequate basis for philosophical understanding. This implication is well supported by the refutation of Cephalus, because Cephalus is a man of experience and intelligent good sense who nevertheless lacks philosophical understanding.

If experience and intelligent good sense do not provide a satisfactory means for answering philosophical questions, what does? How else can we determine the best way to live? Most ordinary people (past, present, and future) would answer this question by attending to inspired writings, which they regard as expressing the wisdom or will of some divinity. Socrates deals with this approach next, for Polemarchus tries to improve upon his father Cephalus' definition by appealing to the words of a poet, Simonides, whom many would concede to be wise and inspired. An obvious problem with an appeal of this kind—a problem often posed in Plato's time by reference to the pronouncements of the oracle of Delphi, which were misread in famous cases²⁷—is that of interpretation: How do we know if we have interpreted the poet's words correctly? How do we know if we have got the right message? Socrates raises this problem at once, saying ‘It is hard not to believe Simonides, for he is a wise and inspired man (σοφὸs γὰρ καὶ θει̑οs ἀνήρ); but what does he mean?’ Polemarchus offered an interpretation suggesting a soldier's²⁸ conception of admirable behavior (‘Help friends, harm enemies—that is, be loyal, courageous’), and this interpretation is reinforced by his admission that the actions and work in which the just person is most capable of benefiting friends and harming enemies are wars and alliances (332e). Yet Socrates had little trouble refuting the definition interpreted this way. The appeal to inspired writings, particularly when they are composed by poets, is a poor means of gaining philosophical knowledge.²⁹ (Socrates belabors this point in books 3 and 10.)

A final approach, alternative to the one Socrates will develop, is available: It is the sophisticated appeal to experience adopted by members of the sophistic movement. This approach to the subjects interesting Socrates (I have been calling them philosophical) is not easy to refute, and Socrates spends the rest of the Republic dealing with it. It is not until book 9 that Socrates completes his refutation of the restated version of Thrasymachus' views about the value of just behavior. However thorough Socrates' criticism may be, if one thinks of Thrasymachus as representing not just a view of justice that Plato is concerned to oppose at length but a representative method of solving philosophical problems, one will find Thrasymachus' irritable responses to Socrates' queries easier to understand. Thrasymachus is something of a moral positivist, who rests his view of justice on observation, not a priori argumentation. He observes that different societies have different systems of justice and that the only thing common to such systems (see 339) is that they express the interests of the parties that create them, the ‘stronger’ element in those societies. Since this sort of view can be confirmed, he thinks, by simple observation, it may strike him as silly for someone to criticize the view, or support some alternative, by rarefied ‘arguments’.

I have discussed these last matters to support my claim that, although Socrates' criticism of Cephalus' and of Polemarchus' theses about justice are, in fact, examples of elenctic refutation, they serve a larger purpose in the context of the Republic as a whole. Cephalus and Polemarchus are not just opinionated figures, like Euthyphro, whose ignorance deserves to be exposed; they represent epistemic strategies the refutation of which has philosophical significance for the entire dialogue. They also represent the primary human types whose motivation and epistemic biases are discussed by Socrates in books 8 and 9.³⁰ But I will not go into that matter here. My aim has been to show that the similarity of book 1 to Plato's elenctic dialogues is much more superficial than even distinguished scholars still suppose and that the principal topics pursued in the following books are required or suggested by the positive arguments set forth in book 1. I think I have already said enough to render this contention credible.

A final point needs to be made, however. In my first paragraph I claimed that the method of inquiry that Socrates employs in book 1 is really no different from the method he uses in the remainder of the dialogue. In this last section I argued that although Socrates conspicuously employs the elenchus in book 1, his use of that method not only accords with the larger aims of the entire dialogue but is an important component of the dialectical method that he prescribes in book 7. Interesting and important as these points may be, they do not really support the claim of my first paragraph—that there is a continuity of method in the entire dialogue. Socrates may not disavow elenchus in the later books, and the elenctic refutation of book 1 may accord with the plan of the dialogue, but it does not follow that the method, or methods, he actually uses in the later books are substantially the same as the method, or methods, he used in the first one. Prescribing a method for philosophcal investigation is not the same, after all, as actually using it.

Since the method, or methods, Socrates actually follows in the later books is now at issue, it is appropriate to remind the reader that Socrates' all-embracing argument in the remainder of the dialogue is actually elenctic: it is directed against the ‘renewal’ of Thrasymachus' position on justice that Glaucon and Adeimantus provide in book 2 (see 358b), and it is not fully completed until book 10, when the post-mortem consequences of a just life (which could be cut short by some calamity)³¹ are considered and the Myth of Er is presented. It is true that the renewed position does not represent the considered beliefs of Glaucon and Adeimantus, and in this respect it differs from the procedure of the most distinctively elenctic dialogues.³² In other respects, however, the large-scale structure of Socrates' argument against the renewed position conforms to the elenctic pattern. This pattern is, of course, dwarfed by the amount of positive theory that Socrates introduces in the process of obtaining his ultimate negative result. But as I have been arguing, this is also true, at least to a significant extent, of his argumentation in book 1.

The claim, urged principally by Vlastos, that Socrates employs a method in book 1 that he abandons in books 2 through 10 is based on the idea that the argumentation of book 1 is exclusively critical and elenctic, whereas the subsequent argumentation is not. But I have shown that both components of this idea are false: book 1 is not exclusively critical, and the following books are not entirely constructive. It seems to me that the theory-construction of book 1 is conducted by the same ‘methods’ that are used in the more extensive theory-construction of the following books: analogies are exploited, classifications are made and supported by everyday experience (as in identifying the three primary kinds of human beings), and hypotheses are adopted on the basis of their explanatory value (e.g., ‘Dissension in the soul causes unhappiness’, ‘Knowledge is focused on ideal unchanging objects’). Proving this last contention—showing that the constructive methods of book 1 and the subsequent books are, in fact, substantially the same—is a tall order, for it would require a logical analysis of the whole dialogue, something that cannot be undertaken in a journal article. But the burden of proof perhaps belongs to the other side. If the constructive methods in the later books are in fact different from the constructive methods of book 1, those who hold this view should endeavor to show that it is true. I have said enough here to support a skeptical verdict. I have no reason to think that a new constructive method is employed in the later books.³³

Notes

1. Some earlier disagreement is represented by the views of Taylor 1926, 264, Friedländer 1930, and Shorey 1933, 215; other references are given in Guthrie 1975, 437. Taylor argued that the Republic was planned as a whole and that book 1 serves its purpose as an introduction to the whole work ‘perfectly’; Friedländer denied that the Republic had been planned as a whole; he argued that book 1, which he called ‘the Thrasymachus’, was originally a separate dialogue written or planned as an integral part of ‘the early group of aporetic dialogues’; Shorey, allowing that the Republic might conceivably stand alone as a dramatic dialogue in Plato's ‘earlier’ manner, regarded hypotheses such as Friedländer's as mere ingenious conjectures.

2. The dialogues Vlastos had in mind are the dialogues belonging to what he regarded as Plato's ‘earlier’ period minus the Apology. The other dialogues in this group are Charmides, Crito, Euthyphro, Gorgias, Hippias Minor, Ion, Laches, Protagoras, and Republic i. See Vlastos 1991, 46-47. I myself have no considered views on the relative dating of Plato's dialogues. For a perspective on the order of Plato's dialogues different from that of Vlastos, see Kahn 1981; and for a perceptive review of recent works using stylometric methods to ascertain that order, see Young 1994.

3. Vlastos 1991, 250 argued that when he wrote the Republic, Plato had become aware that elenctic inquiries have purely negative results and cannot provide a satisfactory method for philosophical investigation. As a ‘tacit confession’ that the results he had so far achieved are insecure and must be reconsidered in the light of a new method he will now introduce, he attached book 1 to the others. Vlastos's view of Socrates' method in book 1 is evidently shared by Julia Annas, who writes, ‘In going from Book I to the rest of the work, Plato is passing from an unsuitable method of meeting Thrasymachus' challenge to a more suitable one’ (Annas 1981, 57). Reeve 1988, though holding that the Republic is a philosophically and artistically unified work, argues that book 1 is ‘a brilliant critique of Socrates, every aspect of which is designed to reveal a flaw in his theories’, and that the subsequent books develop ethical views that Socrates would have produced if he had been brought up in the right kind of polis (xi-xii, 22-23). White 1979, also holding that the Republic is a highly unified work, thinks that book 1 ‘shows the state of the discussion [concerning justice and the good life] as Plato saw himself taking it over from Socrates’ (3, 61). The position I take opposes all these views about the philosophical content of book 1.

4. My description of the elenchus differs in minor respects from the descriptions given by Vlastos 1983, 39 and Irwin 1995, 17.

5. Not everyone would agree that the Socratic elenchus is exclusively a critical, destructive method. Brickhouse and Smith 1994, 19 argue, e.g., that it can be understood as having a ‘constructive function’ because Socrates' repeated use of it in the examination of many different peoples' moral beliefs can provide inductive evidence for ‘the necessity of his own view for a coherent life’. But this argument does not show that the elenchus yields any positive conclusions itself. To get the positive conclusions Brickhouse and Smith have in mind, Socrates has to use (as Polansky 1985, 258 noted) an additional method that he does not actually use, namely induction, and even then only a negative conclusion would actually be warranted—the conclusion that no one who (as Brickhouse and Smith put it) ‘rejects Socrates' view of the value of injustice can maintain a coherent conception of how to live’ (p. 19). The premise that such people cannot maintain a coherent conception of how to live does not entail the positive conclusion that someone who accepts Socrates' view (or some view compatible with his) can maintain such a conception. (It is not a logical truth that fx(Rx r sCx) r fx(sRx r Cx).)

6. I think it is a mistake to regard ἐπαγωγή as induction, for Socrates does not regard the conclusion inferred by it as rendered merely probable to some degree. Vlastos 1991, 267ff. supports this claim, but he adds that the conclusion of such an inference is ‘built into the meaning’ of the predicate appearing in the premises. Evidently, he thinks the conclusion is analytic. I would not agree with such a strong claim.

7. See Vlastos 1983, 34, 47-57. In the same essay he also acknowledges that Socrates arrives at positive conclusions in Republic i; see p. 38 n.

8. The assumption is this: Anyone who ever has a false moral belief will always have at the same time true beliefs entailing the negation of that false belief. See Vlastos 1983, 52, 74. For critical comments on this assumption, see Kraut 1983, 67F and Polansky 1985, 254-257.

9. Plato later develops the ideas that things have jobs or functions (ἔργα), that the virtue of a thing is that by which it does its job (ἔργον) well, and that the virtue of the soul is (personal) justice. The job of the soul, though natural, is the object of an art (τέχνη), which the just soul must practice or be governed by. Since human beings ultimately pursue the goal of happiness (εὐδαιμονία), the other aims they pursue will be subordinate to this supreme τέχνη.

10. Socrates does not say that there are arts of supporting a family, promoting one's self-respect, and being a good Athenian; but he does say that there is an art of wage-earning (see 346c), and if that is an art, these other things are arts as well. After all, each confers a benefit; each can be done well or badly; each requires a particular know-how, and so on.

11. The person who practices injustice tries to ‘outdo’ those like him and those unlike him, whereas someone wise and good at something does not intentionally try to outdo other knowledgeable people or say something better than or different from what they say (350a, 350c).

12. At 353 he says a thing's work is ὃ ἂν e μόνον τι e κάλλιστα τω̑ν ἄλλων ἀπεργάζηται.

13. Annas 1981, 55 says no; in her view the premiss ‘has not been argued for’.

14. This is the suggestion of Adam 1969, 60.

15. 354b. I use the translation, here and elsewhere, of Grube and Reeve (1992).

16. This is Vlastos' claim (1991, 249-250).

17. Thrasymachus does not always acknowledge the difficulties that Socrates derives from his claims; sometimes he merely changes the subject. See 343a.

18. What is hard to understand in this connection is not why Socrates might disavow having a special, thorough kind of knowledge respecting justice—a kind, say, that would enable him to explain important things about it—but why he should speak as if he had not himself worked out and endorsed a general view of justice in his argument with Thrasymachus—a view that, as I emphasize, he does not disavow in his discussion with Glaucon and Adeimantus. For useful discussion about the paradoxical character of Socrates' frequent professions of ignorance (and related matters) see Brickhouse and Smith 1994, ch. 2.

19. At one time I thought that Socrates did not raise any question about how he and his interlocutors could recognize the justice of a city coming into being because he was tacitly assuming the doctrine of ἀνάμνησιs, which he had defended in the Meno and the Phaedo. But this doctrine is not actually mentioned in the Republic, and the explanation I give above makes it gratuitous to presume that Socrates is to be understood as relying on it in the passage in question.

20. The answer Socrates gives in book 4 makes it necessary for him to face the question, forced upon him by his argument in the Phaedo, how something composed of parts can possibly be immortal. He deals with this matter in book 10; see 611b-612.

21. A polis is just when all its classes of citizens do their proper job and receive their proper benefit; when this happens the function of the polis is realized: the needs of the citizens are met by means of the talents of all. Individual persons are just when the parts of their souls are harmonized in a way ensuring that they do their proper job (behave appropriately) and are satisfied by the benefits they properly possess (have the appropriate mental attitude). The parts of a properly harmonized soul fulfill their proper function in an individual person. When they do this, the person will not, e.g., succumb to rage or desire and behave in gluttonous or murderous ways.

22. Anyone doubtful that this example exposes the inadequacy of Socrates' definition can reflect on another: If I am the only person in the society who can walk on his or her hands, it hardly follows that walking on my hands is my function, or one of my functions, in the society.

23. In describing the practice of dialectic, the method to be followed by the philosopher-rulers of the Republic, Plato says (533b) that the practitioner must ‘attempt to grasp with respect to each thing itself what the being of it is’. See [Aune, Bruce. “The Unity of Plato's Republic.” Ancient Philosophy 17, no. 2 (fall 1997)] p. 304.

24. Reeve 1988, 8, 20, 23 appears to succumb to this confusion.

25. This conclusion is supported by two related arguments. The first, which is difficult to summarize briefly, involves the idea that any benefit a just person can confer on another can be conferred more effectively by someone with another, more specialized τέχνη; the other, focused more directly on the aim of helping friends, exploits the idea that one can help friends by unjust means—e.g., by stealing. Like the argument I cite in the text, these two arguments are subtle and require careful interpretation.

26. See 591e, where Socrates speaks of the value of moderation, justice, and moderate wealth (‘not too much money or too little’).

27. One such case was that of Croesus, the king of Lydia, who asked the oracle if he should go to war with the Persians. The oracle replied that if he goes to war, he will destroy a great army. Anticipating victory, Croesus proceeded to go to war; but the army he destroyed was his own. See Herodotus, Persian Wars i 91.

28. See n29 below. On the basis of the speech Against Eratosthenes, in which Lycias, Polemarchus' brother, describes the ruin of their family, there is no reason to believe that Polemarchus actually was or had been a military person; yet in thinking about what I called the ‘larger purposes’ served by Socrates' initial conversations with Cephalus and Polemarchus, one should not suppose that Plato was greatly concerned with historical fact. According to modern estimates, Cephalus had actually been dead from ten to thirty years at the time those conversations were depicted as occurring. See Lamb 1930, introduction.

29. Irwin 1995, 174 says that Plato does not reject Simonides' view in book 1, but he soon asks, ‘What needs to be added to the Simonidean view to make a satisfactory account of justice?’ When I say that Plato rejects the views of Cephalus, Polemarchus, and (by implication) Simonides, I do not mean to imply that he sees nothing in them. Neither is the true account he is looking for; neither will do as it stands.

30. At 581c Socrates identifies the three primary kinds of people as the wisdom-loving, the honor-loving, and the profit-loving. Obviously, Socrates is the first kind of person; Polymarchus, at least by name and his definition of being just, represents a victory-lover; and Cephalus, a merchant, is officially a profit-lover. Thrasymachus, whom Plato describes as breaking into the conversation like a wild beast about to spring, coming at the others as if he were to about to tear them to pieces, represents a fourth kind of person, the tyrannical man of books 8 and 9, who is a wolf among men (566a) but would be the best sort of man, a true wisdom-lover, were his nature not corrupted by a bad education and upbringing (491e). Taylor 1926, 266-267 noted the relation between Cephalus' occupation and his definition, and Friedländer 1930, 60 observed that Thrasymachus is representrative of Plato's tyrannical man.

31. Recall Glaucon's request (elaborated at 360e-362a) that Socrates show them that a just man who, believed to be unjust and consequently tortured, blinded, impaled, and, by implication, killed, is better off than the unjust man who, believed to be just, becomes wealthy and able to do as he wishes.

32. Vlastos 1991, 266 places great importance on this feature of distinctively elenctic dialogues.

33. For helpful comments on the manuscript, I thank Jyl Gentzler, Gareth Matthews, Ronald Polansky, and two anonymous referees.

Bibliography

Adam, James. 1902. The Republic of Plato. Cambridge: Cambridge University Press.

Annas, Julia. 1981. Introduction to Plato's Republic. Oxford: Clarendon Press.

Brickhouse, Thomas and Nicholas Smith. 1994. Plato's Socrates. New York: Oxford University Press.

Friedländer, Paul. 1964. Plato, vol. 2: The Dialogues, First Period. trans. H. Meyerhoff. New York: Pantheon Books.

Grube, G. M. A. and C. D. C. Reeve. 1992. Plato/Republic. Indianapolis: Hackett.

Irwin, Terence. 1995. Plato's Ethics. New York: Oxford University Press.

Kahn, Charles H. 1981. ‘Did Plato Write Socratic Dialogues?’ Classical Quarterly 31: 305-320.

Kraut, Richard 1983. ‘Comments on Gregory Vlastos, “The Socratic Elenchus”’ Oxford Studies in Ancient Philosophy 1: 59-70.

Lamb, W. M. R. trans. 1930. [Orations of] Lysias. Loeb Classical Library. Cambridge MA: Harvard University Press.

Polansky, Ronald M. 1985. ‘Professor Vlastos's Analysis of Socratic Elenchus’ Oxford Studies in Ancient Philosophy 3: 247-259.

Reeve, C. D. C. 1988. Philosopher-Kings. Princeton: Princeton University Press.

Shorey, Paul. 1933. What Plato Said. Chicago: University of Chicago Press.

Taylor, A. E. 1926. Plato, the Man and His Work. New York: Humanities Press.

Vlastos, Gregory. 1983. ‘The Socratic Elenchus’ and ‘Afterthoughts on the Socratic Elenchus’ Oxford Studies in Ancient Philosophy 1: 27-58 and 71-74.

Vlastos, Gregory. 1991. Socrates, Ironist and Moral Philosopher. Ithaca: Cornell University Press.

White, Nicholas P. 1979. A Companion to Plato's Republic. Indianapolis: Hackett.

Young, Charles M. 1994. ‘Plato and Computer Dating’ Oxford Studies in Ancient Philosophy 12: 227-250.