PLATO AND ATHENIAN JUSTICE

Plato’s interest in justice is pronounced and familiar. So too are his criticisms of Athenian democracy. This article suggests that Plato’s conceptualization of justice constituted a direct and conscious confrontation with the highly democratic mode of justice pursued in Athens’ popular courts. Yet Plato did not resist all Athenian judicial norms. His approach recalls Athenian homicide trials, which operated quite differently from the ordinary kind. Plato’s signal contribution to the history of political thought may be characterized as having taken the conception of justice associated with homicide to be paradigmatic, with remarkably enduring effects. In the study of Plato, two points seem so obvious as hardly to need restating. One is the special place of justice in his writings. As Eric Havelock observed, though Plato devoted several dialogues to single virtues, only justice received the honour of a treatise in ten books: the Republic, or ‘On Justice’ as its first editors subtitled it. Yet justice is also prominent elsewhere, as in the Euthyphro, where it eventually supersedes holiness as the principle regulating man’s relations with the gods, or the Theaetetus, an inquiry into knowledge trained specifically on the question of what is just. Indeed, as Jay Kennedy has recently shown, justice was often literally central: the cluster ‘philosophy, justice and god’ recurs at the exact centre of many Platonic texts. If justice was in some sense Plato’s lodestar, then Athenian democracy was the port from which he set sail, and this is the other obvious feature of his work. Athens is the target of explicit criticism in the Protagoras and the Laws, but all Plato’s writings are shot through with scepticism about the kind of democratic norms most Athenians took for granted. The trial of Sokrates in 399 provides the obvious occasion for articulating this scepticism in the Euthyphro, Apology, Crito and Phaedo, but the perils of majority rule are also HISTORY OF POLITICAL THOUGHT. Vol. XXXVI. No. 4. Winter 2015 1 Junior Fellow, Harvard Society of Fellows, 78 Mount Auburn Street, Cambridge MA 02138, USA. Email: cammack@fas.harvard.edu 2 I thank Clifford Ando, Paul Cammack, Jill Frank, Bryan Garsten, Eugene Garver, David Grewal, Verity Harte, Matthew Landauer, Jane Mansbridge, George Scialabba, Richard Tuck and John Tully for comments and questions on this article, not all of which I was able to address here. Translations are my own unless otherwise stated. 3 E.A. Havelock, The Greek Concept of Justice (Cambridge MA, 1978), p. 308. Cf. J. Annas, An Introduction to Plato’s Republic (Oxford, 1981); D. Allen, World of Prometheus (Princeton, 2000), pp. 243–81. 4 J. Kennedy, ‘Plato’s Forms, Pythagorean Mathematics, and Stichometry’, Apeiron, 43 (2010), pp. 10–13. 5 Prt. 319b–320c; Laws 693d. Cf. J. Ober, Political Dissent in Democratic Athens (Princeton, 1998), pp. 157–247; E.M. Wood, Citizens to Lords (London, 2008); D. Allen, Why Plato Wrote (Malden, 2010). C op yr ig ht ( c) Im pr in t A ca de m ic 2 01 3 F or p er so na l u se o nl y -n ot fo r re pr od uc tio n in view elsewhere, as in the Gorgias, Theaetetus and Republic. Emile Faguet characterized Plato’s position in the strongest possible terms. ‘Socrates’s death inspired all Plato’s hatreds. And his hatreds inspired all his ideas . . . The foundation of his politics is nothing other than a horror of the Athenians.’ Others detect more ambivalence, even admiration for certain practices. Yet an impression of alienation from Athenian convention remains. These points are familiar, yet they are seldom drawn together. Plato’s engagement with justice is not normally read as an intervention in Athenian politics. Indeed, for most of the last century Plato was regarded as uninterested in practical affairs, though the opposite position was once well respected and has recently been revived in a novel form by Danielle Allen. Typically, Plato’s political engagement is assumed to be co-extensive with the institutional proposals found in the Republic and the Laws (possibly to be supplemented by the evidence of the dubiously authored ‘Seventh Letter’) — which is to say not very extensive at all, since those proposals are widely (and surely rightly) regarded as flights of fancy designed to serve particular philosophical ends rather than serious recommendations for reform. Yet Plato’s institutional proposals may be the wrong place to look for signs of his political activism. The crucial evidence arguably lies elsewhere, in his sustained attention to democratic judicial activity. Assemblies and councils appear regularly in his work, but these appearances are swamped by references to courts ( ), which in Athens were staffed by hundreds of ordinary citizens with full discretion over verdicts and no provision for appeal. These references far exceed what one might expect in relation to Sokrates’ trial. Courtrooms, juries, forensic oratory, criminal charges, penal612 D. CAMMACK 6 E. Faguet, Pour Qu’on Lise Platon (Paris, 1905), p. 16. 7 S. Monoson, Plato’s Democratic Entanglements (Princeton, 2000). Cf. J. Annas, ‘Politics in Plato’s Republic: His and Ours’, Apeiron, 33 (2000), p. 322. 8 Allen, Why Plato Wrote. Cf. Robertson, ‘Plato as a Critic of Athens’, Classical Weekly, 29 (19) (1936), pp. 145–51; W. Agard, What Democracy Meant to the Greeks (Chapel Hill, 1942), p. 207; W. Jaeger, Paideia (Oxford, 1944); J. Frank, ‘Wages of War: On Judgment in Plato’s Republic’, Political Theory, 35 (2007), pp. 443–67. 9 M.I. Finley, ‘Plato and Practical Politics’, in M.I. Finley, Aspects of Antiquity (New York, 1969), pp. 73–88; Annas, ‘Politics in Plato’s Republic’, pp. 303–10, 323; Wood, Citizens to Lords, p. 69; P. Cartledge, Ancient Greek Political Thought in Practice (Cambridge, 2009), p. 21; M. Schofield, ‘Approaching the Republic’ and ‘Plato and Practical Politics’, in The Cambridge Companion to Greek and Roman Political Thought, ed. C. Rowe and M. Schofield (Cambridge, 2000), pp. 190–232 and 293–302. 10 Cf. G. Morrow, ‘Plato and the Rule of Law’, The Philosophical Review, 50 (1941), pp. 108–10. 11 M.H. Hansen, Athenian Democracy in the Age of Demosthenes (Norman, 1999), pp. 178–224; A. Lanni, Law and Justice in the Courts of Classical Athens (Cambridge, 2006), pp. 31–40. C op yr ig ht ( c) Im pr in t A ca de m ic 2 01 3 F or p er so na l u se o nl y -n ot fo r re pr od uc tio n PLATO AND ATHENIAN JUSTICE 613 ties and verdicts feature throughout Plato’s work, whether the topic is ostensibly judicial or not. Some of these references are easy to miss. ‘Are we to fix the limits of truth by the clock?’ asks Sokrates in the Theaetetus, which readers may not recognize as a nod to judicial practice unless they know that speakers in Athens’ courts — and only in its courts — were subject to strict time limits. Similarly, the line ‘Are we to count names like votes and determine their correctness that way?’ in the Cratylus may be read as voicing scepticism about majoritarianism in general, unless it is known that counting votes (as opposed to estimating a majority from raised hands) was a distinctively judicial practice. Other items are less obscure. The Gorgias initially associates rhetoric with ‘the courts, council, assembly and other places’, then with ‘the courts and other places’, and finally focuses solely on the courts. In the Cleitophon, Sokrates identifies politics fully with judging, and in the Laws, a polis is said to be no polis at all without law-courts properly established, while those who are excluded from judging are said to be justified in feeling they have no share in the polis at all. In the ancient Greek context, this emphasis is less strange than it may seem. Aristotle divided politics into two parts, deciding what is advantageous (to sympheron, a task for assemblies) and deciding what is just (to dikaion, a task for courts); and while today politics is primarily associated with the former, that is, prospective policy decisions, the ancient Greeks seem to have associated it far more with the latter, that is, retrospective dispute-resolution. Moreover, judicial activity was especially significant in democratic Athens. Aristotle believed it was specifically through its power in the courts that the Athenian had gained control of the entire political system. The author 12 E.g. Tht. 148b–c, 173c, 174c; Phil. 20e, 51a, 52e, 60c, 64d, 67d; Smp. 189c, 215b, 219c; Phdr. 266d–e; Hipparch. 225c; Lch. 196b; Lys. 222e; Prt. 337d–e; Criti. 106b. Cf. A. Bloom, The Republic of Plato (New York, 1991), p. 444. 13 Tht. 158d, trans. M.J. Levett, rev. M. Burnyeat, in Complete Works, ed. J. Cooper (Indianapolis, 1997). Cf. Tht. 172d; D. Allen, ‘A Schedule of Boundaries’, Greece & Rome, 43 (1996), pp. 157–68. 14 Cra. 437d, trans. C.D.C. Reeve, in Complete Works, ed. Cooper. Cf. E.S. Staveley, Greek and Roman Voting and Elections (Ithaca, 1972), pp. 73–100. 15 Grg. 452e, 454b, 454e, 455a, 466a, 471e–72c, 480b–481a, 486, 508b, 508d, 511b–d, 515e, 516d, 519b, 521e, 522a–27e. Cf. Hipp. Maj. 304a, 304d; Euthyd. 290a; Phdr. 261b; Tht. 174c. 16 Clit. 408b; Laws 766d, 768d. 17 Arist. Rhet. 1358a. 18 See e.g. Hom. Il. 18.497–508; Hes. WD 225–9; Aesch. Eum. 433–5; Hdt. 1.96–7; Arist. Pol. 1253b25; Stob. Flor. 2.7.iii. 19 See D. Cammack, ‘The Democratic Significance of the Athenian Courts’, forthcoming. 20 Arist. Pol. 1274a. C op yr ig ht ( c) Im pr in t A ca de m ic 2 01 3 F or p er so na l u se o nl y -n ot fo r re pr od uc tio n of the Aristotelian agreed. Thucydides and Aristophanes each depicted the Athenians as obsessed with judging, the ‘Old Oligarch’ cast the courts as a sine qua non of the political system, and numerous orators described them as the ‘bulwark’ or ‘highest organ’ of the democracy. In large part, this prominence stemmed from the courts’ role in disciplining politicians. Athens had its share of powerful political leaders, yet those leaders were prevented from ruling outright by their subjection to pop

In the study of Plato, two points seem so obvious as hardly to need restating. One is the special place of justice in his writings. As Eric Havelock observed, though Plato devoted several dialogues to single virtues, only justice received the honour of a treatise in ten books: the Republic, or 'On Justice' as its first editors subtitled it. 3 Yet justice is also prominent elsewhere, as in the Euthyphro, where it eventually supersedes holiness as the principle regulating man's relations with the gods, or the Theaetetus, an inquiry into knowledge trained specifically on the question of what is just. Indeed, as Jay Kennedy has recently shown, justice was often literally central: the cluster 'philosophy, justice and god' recurs at the exact centre of many Platonic texts. 4 If justice was in some sense Plato's lodestar, then Athenian democracy was the port from which he set sail, and this is the other obvious feature of his work. Athens is the target of explicit criticism in the Protagoras and the Laws, but all Plato's writings are shot through with scepticism about the kind of democratic norms most Athenians took for granted. 5 The trial of Sokrates in 399 provides the obvious occasion for articulating this scepticism in the Euthyphro, Apology, Crito and Phaedo, but the perils of majority rule are also ties and verdicts feature throughout Plato's work, whether the topic is ostensibly judicial or not. 12 Some of these references are easy to miss. 'Are we to fix the limits of truth by the clock?' asks Sokrates in the Theaetetus, which readers may not recognize as a nod to judicial practice unless they know that speakers in Athens' courts -and only in its courts -were subject to strict time limits. 13 Similarly, the line 'Are we to count names like votes and determine their correctness that way?' in the Cratylus may be read as voicing scepticism about majoritarianism in general, unless it is known that counting votes (as opposed to estimating a majority from raised hands) was a distinctively judicial practice. 14 Other items are less obscure. The Gorgias initially associates rhetoric with 'the courts, council, assembly and other places', then with 'the courts and other places', and finally focuses solely on the courts. 15 In the Cleitophon, Sokrates identifies politics fully with judging, and in the Laws, a polis is said to be no polis at all without law-courts properly established, while those who are excluded from judging are said to be justified in feeling they have no share in the polis at all. 16 In the ancient Greek context, this emphasis is less strange than it may seem. Aristotle divided politics into two parts, deciding what is advantageous (to sympheron, a task for assemblies) and deciding what is just (to dikaion, a task for courts); 17 and while today politics is primarily associated with the former, that is, prospective policy decisions, the ancient Greeks seem to have associated it far more with the latter, that is, retrospective dispute-resolution. 18 Moreover, judicial activity was especially significant in democratic Athens. 19 Aristotle believed it was specifically through its power in the courts that the Athenian dÂmos had gained control of the entire political system. 20 The author of the Aristotelian AthÂnaiÜn Politeia agreed. 21 Thucydides and Aristophanes each depicted the Athenians as obsessed with judging, the 'Old Oligarch' cast the courts as a sine qua non of the political system, and numerous orators described them as the 'bulwark' or 'highest organ' of the democracy. 22 In large part, this prominence stemmed from the courts' role in disciplining politicians. Athens had its share of powerful political leaders, yet those leaders were prevented from ruling outright by their subjection to popular rule through the courts. A wide variety of political charges, including treason, accepting bribes, deceiving the dÂmos, making illegal proposals and proposing disadvantageous laws, meant that Athenian politicians spent a good deal of time defending themselves in court and often received stiff penalties. 23 As Sokrates notes in the Gorgias, the careers of Kimon, Themistokles and Miltiades all came to an end through judicial action; 24 those of many others, from Perikles to Demosthenes, were temporarily stayed. 25 In short, the courts were a crucial vehicle of democracy in Athens, and Plato surely recognized this.
This article suggests that Plato did more than recognize it, however. Against those who might interpret Plato's philosophy as being relatively insulated from his political context, it reads his engagement with justice as a direct confrontation with the democratic and highly politically charged mode of justice pursued in Athens' popular courts. 26 The result of that confrontation was a novel, even revolutionary account of justice that radically challenged the foundations of Athenian democracy and has not yet ceased to inform Western political thinking. Nonetheless, this is not to suggest that Plato's account had no Athenian antecedents. In fact, his approach recalls another strand of Athenian judicial practice, namely homicide trials. This was the only area of Athenian adjudication in which a right answer was assumed to exist independently of the views of the judges, an idea Plato seems to have taken up with enthusiasm. Indeed, Plato's unique contribution to political thought may be characterized as having treated a relatively minor element of Athenian judicial practice as paradigmatic of justice in general, with remarkably enduring effects. The political dimension of this venture ought to be better known, and not only for the sake of improving our understanding of Plato's philosophy. If

Copyright (c) Imprint Academic 2013
For personal use only --not for reproduction thinking about justice remains indebted to Plato's and what the political implications may be.

I From to dikaion to dikaiosynÂ
One indication of Plato's confrontation with Athenian judicial norms appears in one of the most familiar scenes in Western philosophy, the clash between Thrasymachos and Sokrates in Book I of the Republic. 27 Though the claims that arise in this encounter are complex, the topic under discussion seems clear: it is the nature of justice. Yet two Greek terms, both regularly translated 'justice', appear in this section. 28 One is to dikaion, what is right or just. This is the subject of Simonides' definition, discussed earlier in the dialogue, 29 what Thrasymachos, in his initial interjection, demands Sokrates define, 30 what Thrasymachos himself defines as 'nothing other than the advantage of the stronger' (later 'the good of another, the advantage of the stronger and the ruler'), 31 and the substance of the first round of Sokrates' interrogation of Thrasymachos. 32 The other is dikaiosynÂ, the main subject of the earlier conversation between Sokrates and Polemarchos, 33 the search for which Sokrates likens to the search for gold, 34 and the substance of the second round of Sokrates' interrogation. 35 Most importantly, it is dikaiosynÂ, not to dikaion, that forms the focus of the rest of the work. 36 This raises two important questions.
(1) Are to dikaion and dikaiosynÂ true synonyms? (2) If not, why might Plato have wished to turn the conversation from to dikaion to dikaiosynÂ in this way?
To begin with the first question, there are good reasons to suppose that to dikaion and dikaiosynÂ were conventionally distinct and that the difference between them would have been plain to Plato's early readers. One is that Aristotle, perhaps only a couple of decades later, included them separately in the list of human goods presented in the Rhetoric. 37 Another is that Aristotle apparently wrote separate treatises on each, one entitled Peri tÜn dikaiÜn, 'On Just Things' (the plural of to dikaion), and the other Peri dikaiosynÂs. 38 A survey of their usage prior to Plato confirms this distinctness. To dikaion is as old as extant Greek literature. It appears twice in the Odyssey, counterposed first with violence and then with the mistreatment of guests. 39 Another early example (c.500-25) is epigraphical. A bronze plaque laying out rights of pasturage in a newly settled Lokrian community states that in the absence of family members closer than brothers who can inherit the right, men may pasture 'according to what is just' (ka to dikaion). 40 Evidently, to dikaion could denote either an action or an outcome, 41 and may thus be translated either 'the just thing' or 'what is just', although, as Enrico Pattaro argues, the best rendering may be 'what is right' or 'what is as it ought to be', in line with the root dikÂ, which he translates 'right'. 42 Significantly, to dikaion cannot denote a person. A person can be dikaios (or dikaia if female), but the neuter adjective dikaion must refer to a thing, and the article to suggests 'thing in general' -hence 'what is right' in an impersonal sense. 43 DikaiosynÂ, on the other hand, is a younger term and used differently. 44 We find it first in Herodotos' History, written in the last half of the fifth century, where it would seem that like other Greek words ending -osynÂ (such as sÜphrosynÂ, 'moderation', or polypragmosynÂ, 'busybodiness'), it denotes a personal quality -a virtue or vice, in this case a virtue. Often the relevant agent is a king: dikaiosynÂ is the attribute that enables kings to judge (and therefore to rule) soundly. 45 But the term is also used in relation to fairmindedness and trustworthiness in non-ruling individuals. 46 In either case, 'righteousness' is a good translation, since it signifies a personal quality rather than an action or state of affairs. 47 Havelock, moreover, suggests that, although we cannot know for sure, the term was probably coined not long before we first see it. While other -osynÂ words appear regularly in this period, dikaiosynÂ appears just nine times before the end of the fifth century and only becomes common currency by the end of the second decade of the fourth century. 48 As a result it seems possible that it originated no earlier than 450, 'to express a notion that had not hitherto demanded it'. 49 Accordingly, although the English word 'justice' may indicate either 'what is right' or 'righteousness', and 'justice' is frequently used to render both to dikaion and dikaiosynÂ, to dikaion and dikaiosynÂ are not synonyms. 50 This matters a great deal in the Republic, because it suggests that Thrasymachos and Sokrates are interested in two distinct questions. Thrasymachos wants to define what is right, which calls for a description of something impersonal, some class of actions or states of affairs. Like to dikaion, both his suggested definitions, 'the advantage (to sympheron) of the stronger' and 'the good (to agathon) of another', feature the substantive use of the neuter adjective and thus make sense linguistically as equivalents, whatever their possible philosophical demerits. But to sympheron, 'the advantageous', and to agathon, Such an account is exactly what we get in the rest of the Republic, as Sokrates investigates the soul of the righteous agent. Of course, since the state of someone's soul is difficult to discern from the outside, Sokrates' account of dikaiosynÂ includes a description of the activity by which a righteous soul may be known: the definition of dikaiosynÂ offered at the end of Book IV, 'doing one's own business', is of this form. 52 But though dikaiosynÂ can be manifested in action, Sokrates' righteous agent is righteous whether he acts or not. 'Doing one's own business' refers primarily to the activity of the three parts of the soul, not to the actions of the agent whose soul it is. 53 Being righteous, on this account, is thus a permanent psychic state as opposed to a practice (as Aristotle would later characterize it). 54 To dikaion, by contrast, necessarily presupposes action, either directly (by itself denoting an act) or indirectly (as the origin of an outcome that is dikaion).
The difference between Thrasymachos' and Sokrates' accounts has been described as that between an act-centred and an agent-centred conception of justice. 55 As the foregoing suggests, this gets at something valuable, but it is nonetheless arguably better avoided, since it is not clear that there exists at this point in the Republic a single concept 'justice' that the objects of Thrasymachos' and Sokrates' interest can be said to be different conceptions of. At this stage, all we have is one term denoting 'what is right' and another denoting 'righteousness', and as yet no very clear understanding of their relationship. This brings us to the second question asked above. Plato evidently wishes to turn from discussing what is right to discussing righteousness. Why?
Two clues in the text suggest an answer to this question. One is that, as Glaukon's contributions at the beginning of Book II (a retelling of the myth of Gyges and a sketch of the lives of a righteous man deemed unrighteous and an unrighteous man deemed righteous) seem designed to suggest, he who possesses Sokratic dikaiosynÂ will possess it whether or not anyone else sees this is the case. 56 In this respect, Plato employs dikaiosynÂ almost as a nonevaluative term: that is, it does not express a judgment on the part of any particular agent. It simply denotes an enduring personal attribute, like having blue eyes. To dikaion, however, both within and beyond Plato's works, cannot help having evaluative force. Its counterposition to violence, maltreatment of guests and wrongful pasturage expresses a judgment on these activities -and this presupposes the existence of an agent doing the judging. indication of this appears in Thrasymachos' definitions. What is right, on his account, is right in relation to a particular agent: to the stronger, to another, or to a ruling party, i.e. a tyrant, an oligarchical elite, or a dÂmos. What is important here is not who the agent is but simply the fact that to dikaion presumes the existence of some agent in relation to whom a given action or state of affairs can be said to be dikaion. 57 In the examples from the Odyssey, the relevant agent is Penelope and her community. In the Lokrian decree, it is the Lokrians. To dikaion by itself does not suggest anything about the agent involved, but neither does it simply denote a given action or state of affairs: it is itself a judgment on them, which implies the existence of a judge.
The other clue appears at the beginning of Book I, when the theme of right action is first canvassed. The conversation begins with the elderly Kephalos worrying about being asked to 'pay the penalty' (didonai dikÂn) in the world below for misdeeds committed in this one. 58 This leads him to rattle off a list of possible wrongs: cheating another man 'even unintentionally', playing someone false, or remaining in debt to a god for a sacrifice or to a man for money. 59 Kephalos is not interested in whether these actions are right or wrong, or what the basis for making a judgment of that sort might be: that is, he is not interested in investigating the concept of 'right' itself. He cares only about whether these actions have been performed: that is, effectively, whether or not he possesses dikaiosynÂ. Sokrates confirms that this is the object of Kephalos' anxiety with his first comment. 'But this very thing, dikaiosynÂ, is it really truth-telling without qualification and giving back whatever one has received?' 60 This is the first time the word dikaiosynÂ itself appears in the text. But Sokrates seems correct to identify righteousness, not what is right, as the thing on Kephalos' mind.
As the discussion moves away from Kephalos' personal predicament, it is easy to lose sight of its origins in this passage. But these origins are revealing. As we have seen, Sokrates, as Plato depicts him, is not interested in classifying different kinds of action; he wants to consider agents in themselves. Specifically, the Kephalos episode suggests that his interest lies in judging these agents. Kephalos expects to be held accountable after death for misdeeds he may not even know he has committed, and it is that perspectivethat of the omniscient, immortal judge -that is developed in the rest of the work. Contrast this with Thrasymachos' approach to to dikaion, which has the perspective of a human agent at its foundation. Thrasymachos is not terribly 57 The equivalence is not exact, since as emerges in response to Cleitophon, Thrasymachos ultimately resists the notion that the ruling agent is himself the final judge of what is right (Rep. 340b-341a). Nonetheless, Thrasymachos' rulers do function as standards or measures of to dikaion. 58 Rep. 330d-e. 59 Ibid., 331b. 60 Ibid., 331c.
interested in the identity of that agent: he seeks only to give a general account of to dikaion in relation to it. Certainly no judgment is offered as to the suitability of that agent to function as a standard of right in the first place. Accordingly, if Thrasymachos represents a possible approach to the question 'what is right?', we can say that Plato turns the tables on it. Rather than take the relevant human agent for granted and seek to describe the kind of things that can be said to be dikaion in relation to it, Plato, in effect, turns to the judging agent and asks: 'What is this agent's relationship to what is right?' This not only marks a shift from an act-centred to an agent-centred approach: it also opens up a space from which the judging agent itself, whose character is not scrutinized under Thrasymachos' approach, can be judged and found wanting. This is arguably a political move because 'what is dikaion?' is the question that ordinary Athenians asked themselves every day in the popular courts when judging the disputes that came before them, while 'what is dikaiosynÂ?' is a lever that can be used to question their right to sit in the seat of judgment in the first place.

II Deciding to dikaion in the Popular Courts
In Book I, Chapter 3, of the Rhetoric, Aristotle divides politics into two parts: deciding what is advantageous (to sympheron) and deciding what is just (to dikaion). 61 On Aristotle's account these are not merely two distinct questions. They also fall to different institutions to decide. Deciding what is sympheron is a task for the assembly, deciding what is dikaion a task for the courts. 62 In classical Athens, what was dikaion was in almost all cases decided democratically, by taking the majority view of hundreds of randomly selected ordinary citizens as the verdict of the polis, following a public trial in which speeches were heard from both sides. 63 Athens was not alone in deciding disputes this way, and the mode of reasoning about to dikaion featured in our Athenian sources was surely not unique to Athens. 64 Where Athens stands out is in the quality of evidence we have relating to judicial decision-making by its citizens, and -especially in relation to Plato -in the significance of this activity for the history of political thought.
The essential question decided by judges in the Athenian courts was whether the defendant deserved punishment for a specified act. This question 620 D. CAMMACK 61 Arist. Rhet. 1358a-59a. 62 Aristotle does not deny that questions of right will sometimes be discussed in the assembly, or of advantage in court. But he does argue that these considerations are beside the point. Rhet. 1358b. 63 Hansen, Athenian Democracy, pp. 178-224; Lanni, Law and Justice, pp. 31-40. Homicide trials were decided differently, as is discussed in the next section. was broken down into two parts: first, whether or not some penalty was deserved, and second, what it should be. In some cases, penalties were fixed by law: for example, the third conviction for making an illegal proposal in the assembly or council led automatically to loss of citizenship. 65 Normally, however, once conviction was announced, the prosecutor and defendant each proposed what they considered an appropriate punishment, and the judges decided between them (they were not allowed to split the difference). 66 Dispute-resolution in the Athenian popular courts thus involved two distinct judgments: whether the defendant's actions had been dikaia or adika, right or wrong, and if wrong, how best to re-establish to dikaion in the eyes of the community. 67 In making these judgments, judges were subject to two forms of guidance. The first was the dikastic oath, which was taken by every citizen listed on the judicial roll, six thousand citizens annually selected at random from those who volunteered. The exact content of the oath is disputed, but at a minimum it seems to have included the following four pledges: to vote in accordance with the laws and decrees of the Athenians; 68 to vote only about matters pertaining to the charge; 69 to listen to both sides impartially; 70 and to judge (dikazein) with one's most righteous judgment (dikaiotatÂ gnÜmÂ). 71 The oath-taker then called on Zeus, Apollo and Demeter and invoked a curse on himself and his household should he break his word. 72 The second form of guidance was the speeches given in court. Since judges were not expected to have prior knowledge of cases or of potentially relevant laws or decrees, all the material on which their decisions were based had to be provided by litigants themselves during the trial. 73 The content of the litigants' speeches depended entirely on what they believed would sway the judges; they could use the time allotted to them in any way they chose.  Inevitably, this meant that speakers presented what Adriaan Lanni has called a 'wide-angle' view of cases. 75 This might include the background to the dispute, including any previous legal actions, 76 extenuating circumstances or aggravating factors; 77 attacks on the character of the opposing party, along with that of his ancestors, family, friends and associates; 78 discourses on one's own (and one's family's) excellent reputation and history of service to the polis; 79 and emotional appeals to the judges to consider the effects of their verdict, including bringing weeping children up onto the platform. 80 Speakers also commonly quoted laws and decrees they considered relevant 81 (although not necessarily the law under which the charge had been brought), 82 discussed decisions given in similar cases, 83 and quoted lines of poetry. 84 All these elements were viewed in the same light as 'evidence'. 85 But what the judges took to be relevant to their decision was left entirely up to them. Once both sides had been heard and had been given a chance to respond to each other, the judges simply lined up to cast their ballots, in secret, for either prosecutor or defendant. They neither discussed the case among themselves nor left any record, apart from the vote itself, of what they had found to be persuasive.
This manner of adjudication has been viewed with extreme scepticism. The evaluation of Henry Maine is typical: 'The Greek intellect, with all its mobility and elasticity, was quite unable to confine itself within the strait waistcoat of a legal formula . . . questions of pure law were constantly argued on every consideration which could possibly influence the mind of the judges. No durable system of jurisprudence could be produced in this way.' 86 Lofberg, in 1917, agreed: 'a startling amount of all kinds of irrelevant matter was brought into nearly every case'. 87 According to B.B. Rogers, translator of Aristophanes' Wasps, 'it would be difficult to devise a judicial system less adapted to the due administration of justice'. In his view, a 'large assembly' could 'rarely if ever form a fit tribunal for ascertaining facts or deciding questions of law'. Its members 'lose their sense of individual responsibility to a great extent, and it is apt to degenerate into a mere mob, open to all the influences and liable to be swayed by all the passions which stir and agitate popular meetings'. 88 Grote, by contrast, gave a relatively sympathetic account. 89 More recent studies, such as those by Allen and Lanni, have also succeeded in showing that Athenian adjudication at least made sense in its own terms: it was not an anomaly in an otherwise essentially modern legal system. 90 Other scholars, such as Edward Harris, have sought to show that Athenian judges took the constraints of the rule of law more seriously than has been supposed. 91 Yet however we evaluate the system, one point is clear: Athenian dikastai had absolute discretion over their verdicts, both as to what they should be (what we may call a first-order decision) and how they should be reached (a secondorder decision).
To be sure, the dikastic oath required judges to judge in accordance with the laws and decrees of the polis. But since judges were free to decide for themselves how, if at all, the laws and decrees brought to their attention applied to any particular case, the interpretative options available to them were not necessarily reduced. 92 Similarly, though precedents were sometimes cited by the litigants, there was no requirement that they be followed by the judges, nor could any such requirement have been enforced. Prosecuting Aristokrates on a charge of illegal proposal in 352, Demosthenes readily admitted that similar proposals had previously been allowed. But the key question, he argued, was not whether such things had happened, but whether they ought to have happened. 'Do not let them tell you that those old decrees were upheld by other juries', he urged: 'ask them to satisfy you that their plea for this decree is fairer than ours'. Failing that, he argued, 'I do not think that you ought to give greater weight to delusions of others than to your own judgment'. 93 Crucially, moreover, the judges' power to give greatest weight to their own judgment was strongly protected. In a political system notorious for holding responsible persons to account, only judges and assembly-goers were deliberately left unaccountable for their decisions. 94 Judges were also considerably better protected than assemblygoers from pressure to justify themselves informally. The lack of discussion among judges, the secret ballot, and the fact that decisions could not be appealed 95 meant that whatever considerations each judge took to be dispositive was an entirely private matter.
Such judicial discretion may provoke anxiety. What if judges came up with the wrong verdict? What if innocent individuals were punished for crimes they had not committed, purely on account of the ignorance or prejudice of the judges? In the context of classical Athens, this anxiety is often expressed in the form of a specific question. What about Sokrates? Even those sympathetic to the Athenian mode of adjudication readily describe his execution as an 'outrage'. 96 Was not his death a direct result of the Athenians' unlearned, populist and discretionary approach to judicial decision-making?
An important clarification is necessary here. Athenian judges were surprisingly seldom asked to decide the facts of a case. Those were usually agreed by both sides. Their task was primarily interpretative. They had to decide if the actions of the defendant had been dikaia or adika, just or unjust, and this complicates the notion of a 'wrong' verdict considerably. 97 For example, when Leokrates was accused of treason by Lykourgos in 330, the relevant action -Leokrates' departure from Athens after the battle of Chaironeia eight years earlier, when the citizenry had been asked to stay in the polis to defend it against further attack -was not disputed. It was admitted that Leokrates had gone abroad with his family and that he had spent the intervening years working overseas as a corn merchant. 98 What the judges had to decide was whether or not this constituted treason, which was a question of interpretation rather than fact. Or take cases of illegal proposal. The fact that a given defendant had made a given proposal was not at issue: that was a matter of public record. What the judges had to decide was whether the proposal in question had been paranomÜn, 'beyond the laws', which was, again, an interpretative task.
Something similar can be said about the charges against Sokrates. Sokrates was accused of impiety (asÂbeia) for not believing in the gods worshipped by the rest of the polis and for corrupting the youth. 99 But not even Plato suggested that Sokrates did not hold heterodox religious beliefs. Indeed, his mention of Sokrates' daimon in the Apology and the fact that he left open the possibility that Sokrates was a monotheist rather confirms the case. 100 Equally, it was common knowledge that Sokrates was happy to impart his views to whichever young men cared to listen, and that he did this for free. 101 The question for the judges was whether or not this behaviour was impious, and a narrow majority decided it was. 102 Not all cases in Athens followed this pattern. Sometimes the facts were in dispute, and then the judges had a trickier task: they had to decide who was lying. 103 But most of our surviving cases do conform to this model, and in large part this is because of another well-known feature of Athenian laws, their vagueness. In a modern legal setting, when faced with a question such as 'was this act treason?', 'is this proposal illegal?' or 'is this teaching impious?', the natural first step would be to determine the relevant definitions of 'treason', 'illegality' and 'impiety' provided in the laws and consider whether the defendant's actions accorded with them. If they clearly did (or did not), one might be tempted to argue that the verdict was -or should be -a foregone conclusion. But this step was not available in Athens. 104 Athenian laws were famously unspecific: as Robin Osborne put it, they had an 'open texture'. 105 Generally, as Lanni explains, the laws simply stated 'the name of the offence, the procedure for bringing the suit under the law, and in some cases the prescribed penalty'; they did not 'define the crime or describe the essential characteristics of behavior governed by the law'. 106  104 It is not really clear that a legal decision can be a foregone conclusion in any legal system, strictly speaking: that is to say, the connection between law and act is always a matter of interpretation (a point to which I return in the conclusion). Nonetheless, the necessity of interpretation is especially obvious in the Athenian case. 105  of Athens, then that man, while chained up, is to be tried before the dÂmos, and if he is found guilty, he is to be killed by being thrown into a pit and his money confiscated and a tithe given to the goddess.' 107 Exactly what 'wronging the dÂmos' meant in this context was left to the judges to decide; and the same was true of almost every other wrong prohibited in Athenian law. 108 What 'treason', 'illegality' or 'impiety' meant was a question for judges, not legislators, to decide. 109 On one view, this vagueness was a major weakness in the Athenian legal system. In the words of Moses Finley, the judges 'had too much latitude, in the sense that they could not only decide on a man's guilt but could also define the crime he had committed'. Finley added: 'When impiety -and this is only an example -is a catch basin, no man is safe.' 110 It may be natural to worry that virtually any act might be defined as criminal on this basis. But the claim that a term may function as a 'catch basin' directs us to think further about the way that words acquire meaning -and this in turn raises the possibility that the standard Athenian approach to adjudication may not have been the recipe for injustice that many have feared.
Arguably, there is a deep analogy between the conventional Athenian conception of 'what is right' (to dikaion), as revealed in the activity of the popular courts, and an account of language that is widely accepted today. The Athenians seem to have assumed that what is dikaion is intersubjectively constituted in the same way as is the meaning of any term. That is, they seem to have imagined that there is no 'external' or 'objective' right answer to the question of whether any given act is dikaion or adikon beyond the answer given by the community itself. The only available measure or standard for such an answer was other members of the same community, and for this reason the decision of a sufficiently representative panel of Athenian citizens could not be 'wrong'.
To clarify this argument in the linguistic context, consider the word 'treason'. It would make no sense to claim that 'treason' has an objective meaning independent of the way it is used by the community of English speakers. If that were the case, languages could not evolve, though of course they do. Equally, however, it would make no sense to claim that the meaning of 'treason' is wholly subjective, that is, that it can mean one thing for me -'betraying one's country', say -and something else for you -'cheating on one's spouse' -with no way to decide which definition is better. If that were the case, all communication would be impossible. 111 A more plausible account starts and finishes at the level of the linguistic community. The range of meanings of any given term is intersubjectively constructed through its use by members of that group over time: it is produced, defined, sustained and changed exclusively in relation to the group itself. 112 Accordingly, if one wants to determine the meaning, or meanings, of the term 'treason', there is nothing for it but to explore how it is used by members of the linguistic community in question, in as many contexts as seems necessary. Conceivably, there is no other standard by which to ascertain the meaning of any given word.
The Athenians seem to have approached 'what is right' in their popular courts in a similar way. The question of whether a defendant deserved punishment for a given act was not, for them, one that had an 'objectively' correct answer, to be deduced either from the laws themselves or in some other way. But equally, they did not think that 'what is right' in any particular case was fully subjective, that is, one thing for one citizen and another for another, with no way of choosing sensibly between them. Rather, both the actions and the ideas of the ancient Athenians suggest that, in the normal run of things, they took 'what is right' to be intersubjectively constituted: both diachronically, in the sense that the idea of to dikaion held by each citizen was tested and refined by his interactions with the rest of the community over time, and synchronically, in the sense that every particular verdict was a snapshot of the views of a random sample of the community at a particular moment.
Importantly, moreover, it was taken for granted that there would often be disagreement over what was dikaion in any particular case. The absence of disagreement could be treated as a sign of a frivolous lawsuit: in public cases, prosecutors who failed to win at least a fifth of the judges' votes were heavily fined. 113 But this merely suggests the significance, on this conception of to dikaion, of not taking the view of any single citizen or small group as decisive, but rather of canvassing the views of a large sample, and the larger the better in important or controversial cases. This is exactly what the Athenians did. For cases involving large sums of money, or that were especially politically significant, the Athenians doubled or tripled the number of judges required to hear the case, from 500 to 1000 or 1500. 114  minimize the chances of getting a freak result. The Athenians seem to have aimed to represent as closely as they could the view of the whole community on the matter, congruent with the limitations of space and expense (since judges were paid for their time).
There is a clear affinity between the conception of 'what is right' sketched here and the philosophy of Protagoras, at least as suggested by the line 'man (ho anthrÜpos) is the measure (metron) of all things: of those which are, that they are, and of those which are not, that they are not'. 115 Moreover, although Plato repeatedly construed this claim as though it referred to a single man, 116 thus making Protagoras open to charges of both subjectivism ('all appearances exist') and relativism ('what appears to you is true for you') 117 -and although various modern scholars have followed Plato on the subject referred to in this line, though not necessarily on the inferences 118 -it is certainly possible that Protagoras' 'man' was meant to denote the species or the community rather than a single individual. 119 If that is the case, then Protagoras' teaching would seem to have been very close to Athenian practice as I have described it.
But the idea that human beings can function as measures of 'what is' and 'what is not', particularly in respect of to dikaion, was not unique to Protagoras. It also appears in Aristotle. Aristotle's objection to the emotional manipulation of judges by speech-makers rested on precisely this foundation: one ought not 'warp' the 'rule' (kanÜn) that one was going to use. 120 Aristotle also believed it was impossible to lay down laws about things that were subjects for deliberation (peri hÜn bouleuontai), including the relationship of law to 'particular matters' (hekasta). On such questions, he said, 'men do not deny that it must be for a human being to judge', they merely dispute how many men ought to perform that task; 121 and Aristotle himself took it for granted that provided certain conditions were met, it was better for a large group of ordinary citizens to produce these judgments than a single outstanding man, or even a few outstanding ones. 122 Protagoras seems to have suggested the same thing, albeit with fewer conditions. 123 The Athenian approach to adjudication suggests that most Athenians held similar views. Interestingly, however, the Athenians did not reason this way in all cases. Homicide trials operated very differently, revealing an alternative conception of to dikaion at work -one echoed, in striking ways, by that advanced by Plato.

III Deciding to dikaion in the Homicide Courts
Though most disputes in Athens were heard in the popular courts in the manner described above, cases of homicide were treated differently. 124 To begin with, they were normally judged by a restricted group of people. Altogether there were five distinct homicide courts, although four of these seem to have been subsets of the fifth, the Areopagos council, which also gave its name to the homicide courts as a whole. 125 In an earlier era the Areopagos had been the aristocratic governing body of the city, but in the classical period it consisted solely of people who had been selected by lot to be one of the nine chief archons, and its function was almost exclusively to judge cases of homicide. Provided they passed their audit (euthyna), all archons joined the Areopagos on a permanent basis. Hansen estimates it probably included around one hundred and fifty people at any one time, around two-fifths of whom will have been over sixty. 126 The entire body heard cases of intentional killing, including wounding, arson and poisoning resulting in death, 127 while panels of fifty-one judged cases in the other courts. The Palladion heard cases of unintentional homicide and the killings of slaves, metics and foreigners; the Delphinion, cases in which the defendant admitted having killed but argued that he had acted lawfully and therefore did not deserve punishment; the Prytaneion, cases in which an animal, inanimate object or unknown agent had caused death; and the court at Phreatto, cases of homicide against citizens who were already in exile for another offence, meaning that they had to argue their case from a boat anchored offshore. 128 As well as being judged by a distinct set of citizens, homicide trials operated significantly differently from those held in the popular courts. Three 124 Commercial maritime suits were also treated differently in that greater reliance was placed on written contracts (Lanni, Law and Justice, pp. 167-71). However, in other respects the process was the same as that described in the previous section. 125 Our sources refer to judges in the smaller courts as ephetai, and some have argued that they were not Areopagites but randomly selected citizens. See  pre-trials were held; 129 defendants were banned from public and sacred spaces before the trial; 130 trials were held in the open air; 131 the most solemn sacrifices were made; 132 and an elaborate series of oaths was taken. At the start of trials, the prosecutor swore that the defendant had committed the offence, and the defendant swore he had not. During trials, witnesses swore both as to the truth of their testimony and as to whether the defendant had committed the crime. Finally, after the verdict was announced, the winner swore that he had told the truth and that the judges had decided correctly. 133 The mode of argumentation in homicide suits was also strikingly different. For once, a relevancy rule applied: speakers were not allowed to go 'outside the issue' (exÜ tou pragmatos), 134 which seems to have been interpreted as prohibiting discussions of character and possibly emotional appeals to the judges. 135 There was another important difference. In the popular courts, as we have seen, the judges were asked to decide what was right or just, that is, what was dikaion. But another pair of concepts regularly appears in connection with homicide. That is truth (alÂtheia) and what is true (to alÂthes). 136 In homicide trials, judges had not only to decide what was right; they also had to decide what was true. Antiphon emphasized the distinction this suggested. 'The laws, the oaths, the sacrifices, the public announcements and all the other things that happen in a homicide suit are very different from other procedures because the facts themselves (auta ta pragmata), concerning which the stakes are greatest, must be known correctly (orthÜs gignÜskesthai)'. 137 Evidently, something was deemed to be at stake in homicide trials that was not at stake in others.
These differences demand an explanation. A significant factor may be the age of the procedures, as Lanni has suggested. Homicide procedures were the oldest in use in Athens and no doubt represented an earlier way of doing things. 138 Yet the allusions to truth that feature in these cases and the additional note of reverence for the Areopagos that appears throughout our liter- ary sources suggests something more. 139 The severity of the penalties might seem relevant: intentional homicide led automatically to execution, burial outside Attika and confiscation of property, while unintentional homicide resulted in exile. 140 But death and exile were also regularly used as penalties in other cases, so that suggestion will not suffice. 141 A better explanation is arguably found in the religious significance of homicide cases. An unnatural or improper death was said to leave a miasma or polluting stain on the polis. 142 Part of the purpose of homicide trials was thus to attribute responsibility to the right person (or animal or object) in the right way, so the stain could be removed. 143 Several ancient authors cited pollution as the explanation for holding the trial out of doors. 144 Religious significance also helps to explain why these cases remained in the hands of the Areopagites, since the chief archons traditionally had significant religious duties. 145 It also explains the elaborate oath-taking, particularly the final oath sworn by the victorious party: the idea was apparently to transfer any miasma resulting from a wrong verdict from the judges to the victorious party himself. 146 Most important, the religious dimension of homicide may explain the discursive difference between speeches in homicide trials and those in all other kinds. Arguably, the stakes seemed greatest in homicide suits because not only other citizens, but also the gods, were deemed to be an interested party. The gods' concern that their shrines and temples not be polluted may have suggested to the Athenians that an 'external' or 'objective' right answer existed to the question of what was dikaion in cases of homicide in a way that was not true in other cases. In a case of treason or illegal proposal, the only interested parties, conceivably, were other citizens: there could thus seem to be no 'right' or 'wrong' verdict on these questions beyond what the citizen community itself took to be right. In cases of homicide, however, the supposition that the gods also had a view may have altered the nature of the reasoning involved. It meant that in homicide cases -and only homicide cases -Athenian judges could be thought of as struggling to reach a decision which in some sense existed independently of their own judgment, i.e. that reflected a truth beyond their own norms and evaluations.
It is worth clarifying that, just as in other types of suit, the question before the judges in homicide trials was often one of interpretation rather than fact. Again, both sides might agree on the events leading up to the death; what was at issue was who or what had ultimately been responsible for it (and hence for the subsequent pollution). 147 This was a much murkier issue and raised puzzling issues of causation that the Athenians seem to have found fascinating. A famous, though probably fictional, example involved the death of a boy from a javelin thrown by a classmate as he ran across a gymnasium. Was the thrower responsible, or the victim, or even the javelin itself? 148 A modern coroner would have the option of recording a verdict of 'accidental death' in these circumstances, but that was not available in Athens, possibly since the problem of miasma would remain. Cases of this sort suggest why the court at the Prytaneion was deemed necessary. Even if death had been caused by an inanimate object, it was still important to interpret the chain of events correctly, in order to ascertain the innocence of any associated human agent and to cast the offending object beyond the boundary of the polis. 149 Again, as in the popular courts, not all homicide suits followed this pattern. Sometimes the events leading up to the death were uncertain, and then the relevant question was simply 'did the defendant do it?', a factual question familiar from any whodunnit. 150 In such cases, the truth of the matter clearly rested on relevant facts, and these clearly existed independently of the views of the judges. In cases such as that involving the javelin, however, the truth of the matter was essentially a question of interpretation. It raised the same issues as deciding to dikaion in the popular courts, except that in cases of homicide, to dikaion could be supposed to have a reality beyond the views of the judges, because the gods could be presumed to care about the result. From the perspective of the judges, what was dikaion in a case of homicide could thus be conceived of as a question with an objectively correct answer rather than an intersubjectively constituted one. In effect, the interest of the gods converted a question of 'right' into one of 'truth' or 'fact'. What was right became something that could be known, or not known, in a way that to dikaion in other cases could not be; and this, I suggest, is where Plato comes in.

Plato's Intervention
Plato's signal contribution to political thought may be characterized as taking the idea that to dikaion exists independently of the norms of the political community -a claim that, on the standard Athenian view, applied only to a limited subset of religiously significant cases -and applying it to all human experience. There is always an externally or objectively correct answer to questions of right and wrong on the Platonic view: it is the answer accepted by the gods (or, as often in Plato's telling, god), who are supremely righteous, and it applies in all cases because the gods have an interest in everything we do. Like the truth in homicide cases, moreover, it can only be known or discerned; it is not created by any agent, be it a single man, a group, or even the gods themselves; 151 and it can be discerned only by a select few: those who possess dikaiosynÂ, which, midway through the Republic, is identified as the state of having gazed on and internalized the true 'form' of to dikaion, 'the right' or 'the just'. On Plato's account, this state is accessible to philosophers alone. Yet, arguably, the doctrine of the forms recalls the kind of knowledge the Athenians imagined themselves to be groping after in homicide trials, but that they do not seem to have imagined existed in any other context, such as cases of treason or proposing disadvantageous laws.
The contours, foundations and implications of this set of claims are aired and explored across Plato's works, in ways both large and small. We may note, for example, the transformation, in the Euthyphro, of Meletos' claim that Sokrates had corrupted young men into the claim that Meletos 'knows how the youth are corrupted and who are those who corrupt them': a turn from an interpretive question, the answer to which would depend on an intersubjectively constituted understanding of what it meant to corrupt, to a question of knowledge or fact. 152 Or the distinction drawn between true and false judges in the Apology, where Sokrates explicitly strips the title 'judge' (dikastÂs) from those who have failed to find him innocent: being a true judge evidently meant coming up with the right answer (or at least, we can say, what Sokrates considered to be the right answer). 153 Or Sokrates' celebrated acceptance of the laws of Athens in the Crito: this is often regarded as a contradiction of his position in the Apology, when he defies the Athenian public with the words 'Men of Athens, I am grateful and I am your friend, but I will obey the god rather than you'. 154 What is less often recognized is that when, in the Crito, Sokrates blames his conviction on the interpretation of the laws offered by the men of Athens rather than on the laws themselves, he thereby rejects the authority of the citizen body to decide for itself what is dikaion, which was the foundation of Athenian law. 155 Finally, we should note the connection between the effort to establish objective meanings for words in the Cratylus and Plato's awareness, revealed in two dialogues, that there was an analogy between the construction of language and that of to dikaion on the conventional conception of that term. Alkibiades, in the first dialogue of that name, states explicitly that the people who had taught him ta dikaia were the same as those who had taught him Greek. 156 The same idea appears in the Protagoras. 157 This suggests that Plato understood that the intersubjective construction of language itself had to be denied if his more significant attack on the intersubjective construction of justice was to succeed. 158 A full analysis of Plato's works in this light would be of great interest, but we may focus here on two points. The first is Plato's enthusiastic (and seemingly innovative) development of the idea that the gods care about every human wrong, not merely some discrete subset of wrongs. This thought appears in a variety of places, including the Republic: the divine judges (dikastas) who reward and punish human beings after death attend to 'all the wrongs they had ever done', not merely to certain categories of wrong. 159 The same perspective features in the Gorgias, in the activity of Minos, Rhadamanthys and Aiakos, the supremely just judges of the afterlife, 160 and in the Phaedo. 161 But it appears most fully in Book X of the Laws. First, the argument 'that the gods exist, and that they are good and honor justice (to dikaion) more than do men' is identified by Klinias as 'the best defence of all our laws'. 162 The Athenian Stranger agrees: the non-existence of the gods must be disproved as a necessary prelude to obeying the laws. But two other 'false notions' about the gods must also be removed: that they 'exist, but pay no heed to human affairs', and that they 'do pay heed, but are easily won over by prayers and offerings'. 163 All three propositions -the non-existence of the gods, their lack of interest in the doings of men, and their willingness to transgress justice (to dikaion) when paid off -are then comprehensively attacked. 164 One point is especially noteworthy: the argument that 'God also cares for the World-All' is wrapped into the argument for his existence in a way that his unwillingness to be 'seduced away from justice with gifts' appears not to be. 165 Care for all things would, on this account, appear to be presupposed by God's very existence. Accordingly there would seem to be no sphere in which human beings may decide what is right for themselves. There is always an independently existing answer, which they may discern successfully, or not.
The second important point is that with which this article began: the transition from to dikaion to dikaiosynÂ as the object of attention in the Republic and its implications for both Plato's philosophy and for later understandings of justice. From the beginning of Book II dikaiosynÂ is established as the focus of the work, 166 and from Book IV it is formally defined as the state in which the elements of the soul 'do their own business'. 167 The question we may now explore is what this allows Plato to do.
As we have seen, the question 'what is to dikaion', as raised by Thrasymachos and pondered daily by ordinary Athenian citizens in the popular courts, functions in effect as an invitation to all comers to exercise their judgment. The question 'what is dikaiosynÂ', by contrast, directs attention to the nature of the judging agent. Presumably, this turn would seem desirable if one wished to argue that only someone with dikaiosynÂ can judge correctly what is dikaion, although Plato nowhere articulates this explicitly. What is clear, though, is that directing attention to a judge's personal qualities raises the possibility that not everyone will possess the necessary credentials to judge well -that is, not everyone will be found to possess dikaiosynÂ. This is not a necessary inference: Protagoras, for example, is depicted as suggesting that a sense of dikÂ, right, was common to all human beings. 168 But asking the question 'what is dikaiosynÂ?' does open up a space from which would-be judges can themselves be judged and found wanting. If they are discovered to lack this virtue, their responses to the question 'what is to dikaion?' might be ruled out -and rightly so, on this approach.
The crucial question then is: who is to be disbarred from judging on this basis? Or, put another way: who possesses dikaiosynÂ and how do they acquire it? In Book IV, we are told that dikaiosynÂ means the possession of a rightly-ordered soul (a condition very close to sÜphrosynÂ, a significantly 164 Ibid., 888c-907d. 165 Ibid., 902-3, 907a. 166 Rep. 357d. Cf. e.g. 363a, 368b-c, 392c-d, 427d, 432b-c, 443d, 517d, 545a, 612b. 167 Ibid., 443d. 168 Prt. 323a-e. older term). 169 But Plato goes further. In Books V and VI, he posits a new notion of to dikaion that goes before and anchors all particular manifestations of that concept, and he does this explicitly in order to provide an intellectual basis for dikaiosynÂ. The new conception of to dikaion he advances is an eternal 'idea' or 'form' gazed on by a select few, and this act, or state, of gazing is then established as what defines dikaiosynÂ as a human quality. 170 It is important to notice that the existence of to dikaion as an ideal form is not intrinsically presupposed by the concept dikaiosynÂ, which as we have seen predated Plato. Rather, the form to dikaion is brought in by Plato as a way of demarcating clearly the difference between those who possess dikaiosynÂ and those who do not. 171 Equally, however, it is important to notice that Plato arrives at the existence of the form to dikaion via his investigation of the concept of dikaiosynÂ rather than as a direct response to Thrasymachos' original question, 'what is to dikaion?'. This is presumably because without being channelled through specific human agents, the existence of to dikaion as an eternal form might be nothing to us. Just as the existence of the gods may mean nothing to us without their intervention in human affairs, the existence of an eternal form could be irrelevant to human society without some mediating channel. If no human being can get to the forms, they might as well not exist, or, at least, we would have little option but to act as though they do not. Thus the manifestation of the form to dikaion in a human being is a critical step in Plato's account. This is where dikaiosynÂ comes in. He who has dikaiosynÂ is the philosopher, who gazes on eternal realities, including to dikaion in its true, timeless form. 172 This seems a critical move in Plato's argument for two reasons. The first is political. As we saw above, judicial activity was of great political significance in the ancient Greek poleis, and it formed a major theme in Plato's works. But it is especially important in the Republic. This is presaged in an illuminating way during the search for dikaiosynÂ in Book IV. Sokrates has just 'caught sight of something', though Glaukon has yet to see it. 173 To help him, Sokrates returns to the theme of 'everyone doing one's own work', and poses the question: 'Look at it this way if you want to be convinced. Won't you order your rulers to act as judges in the city's courts (tas dikas . . . dikazein)?' 'Of 170 That access to the forms is limited to only a select few is generally asserted in Plato's works, rather than argued for. Rep. 493e-494a, 503b; cf. Pol. 297c. 171 As the analogies of the divided line and the cave show, Plato envisaged a series of steps towards the forms. 'Right opinion' was a possible state; true knowledge was not necessary in order to act well. Nonetheless, the boundary between right opinion (orthÂ doxa) and knowledge (epistÂmÂ) was clearly marked.
course', Glaukon replies. 174 Sokrates goes on to establish that the judges will aim for no citizen to have what belongs to another or to be deprived of what is his own, which leads directly to the 'discovery' of dikaiosynÂ. But what is arguably most significant about this exchange is the apparent naturalness of the assumption that the rulers of a polis and its judges will be identical. Later, this equation is confirmed in a considerably showier way with the establishment of the philosopher-rulers, but its seeds are already present. Judging has already been identified as a constitutive function of a ruler. Hence, in the Socratic polis, the rulers must be those who know what is truly just, which is to say the philosophers. From the democratic Athenian perspective, what is significant about this result is that it mirrors the way the Athenian dÂmos ruled: by virtue of its control of the courts. It thus makes transparent the connection between Plato's preoccupation with the theme of justice and his hostility to Athenian democracy with which this article began. If there is a right answer to questions of justice that only the philosopher can 'see', there is no way anyone else can legitimately act as a judge, and hence, on the standard Greek conception, rule.
The Republic embraces this result directly. 175 In the Laws, the position softens. Taking into consideration the significance of participating in judging for acquiring the feeling that one is a real citizen of a polis, which he admits to be beneficial, the Athenian Stranger allows some space for democratic judicial activity of the ordinary Athenian kind. 176 But he radically alters the political consequence of this activity by stripping the democratic judges of Magnesia of final judicial authority. All cases will be open to appeal to a higher court, composed of men elected for the task, and it is they who will hold supreme authority. 177 As in the Crito, Plato resists the right of ordinary citizens to interpret their laws for themselves. Yet this was not only the foundation of the administration of justice in Athens; it was arguably the foundation of the rule of the dÂmos overall. 178 Plato's philosophical intervention thus posed a radical and pointed challenge to the strength and extent of dÂmokratia in Athens.
We may thus offer another answer to the question recently posed by Danielle Allen: 'What did Plato do?' 179 What Plato arguably did was to attack Athenian democracy by formulating a new conception of justice and its administration that was not open to being controlled by ordinary citizens, hence challenging their grip on rule over the polis itself.
The second reason that the manifestation of to dikaion in a human being proves a critical move in Plato's argument is conceptual. I suggested above 174 Ibid., 433e. 175  that the dikaiosynÂ of the philosopher functions as a channel through which to dikaion, the form, can be made active in human society. It would be equally accurate, on the account given in the Republic, to say that the philosopher embodies to dikaion. The philosopher not only gazes at the eternal realities, he will also 'endeavor to imitate them' and 'as far as may be, to fashion himself in their likeness and assimilate himself to them'. 180 Indeed, not only will the philosopher attempt to fashion himself in their likeness, he may be required to reproduce their likeness in others. He is imagined 'stamping on the plastic matter of human nature in public and private the patterns (paradeigmata) that he visions there'. 181 The philosophers will 'glance at' to dikaion, to kalon and to sÜphron 'in the nature of things', and 'alternately at that which they are trying to reproduce in humankind', and there will be no better 'craftsman' of sÜphrosynÂ or dikaiosynÂ or any other virtue. 182 This process of assimilation and reproduction may sound straightforward, yet it has a profoundly important result. In the soul of Plato's philosopher, the conventional distinction between dikaiosynÂ and to dikaion, the virtue within and the state of affairs without, itself dissolves. As in the case of the judgerulers, this dissolution is presaged in an illuminating way earlier in the Republic, in the analogy posited between the city and the soul. When both dikaiosynÂ and to dikaion are translated 'justice', the significance of this analogy and of the coming merger of the two concepts into one in the person of the philosopher is lost. But this significance is profound. We ought to feel it is as strange to seek for dikaiosynÂ in a politeia, a political system or structure, as it would be to seek for courage or moderation in that structure. That is, the search for virtue among the members of the relevant body would not be strange; what is strange is to search for it among the relations of the members of the body to each another. Similarly, the suggestion that to dikaion can be found among members of a political body was entirely familiar. To dikaion, 'what is right', was standardly used as a way of marking out the relations between citizens. 183 But Plato does not pursue that idea. Instead, he explicitly sets out to discover dikaiosynÂ, righteousness, in the relations between individuals in a politeia, rather than in individuals themselves.
The enduring significance of this move is that, conjoined in the person of Plato's philosopher, the concepts to dikaion and dikaiosynÂ normally used to denote 'what is right' and 'righteousness' respectively -become effectively interchangeable. 184 This interchangeability is definitively established in Book VII of the Republic, in the course of the analogy of the Cave.
The philosopher leaves the cave to gaze at the forms, including to dikaion; when he returns, however, the object of his attention is identified not as to dikaion, but as dikaiosynÂ. 185 It is also presaged at the end of Book I. 186 When, following the failure of his discussion with Thrasymachos, Sokrates muses on his continued ignorance of what dikaiosynÂ is, Plato has him say something that, to a Greek, must have sounded very strange indeed. 'Now I know nothing,' Sokrates says, 'for if I don't know what to dikaion is, I shall hardly know if it is a virtue (aretÂ) or not'. 187 Most Greeks would surely have found it very difficult to understand how to dikaion, 'what is right' in the sense of an outcome, could possibly be construed as a virtue, as if it were equivalent to dikaiosynÂ, 'doing what is right'. Yet Plato, in the Republic, aims to provide the conceptual apparatus necessary to equate the two.
The dramatic and profoundly influential result of that move was that a concept of 'justice' that could encompass both 'righteousness' and 'what is right' became thinkable, apparently for the first time. Yet this new concept, connoted by Plato as dikaiosynÂ / to dikaion used interchangeably (and in English by the capacious term 'justice'), was not equally constituted by its antecedents. Rather, righteousness as a form of human activity became the handmaiden of a strictly intellectual conception of what is right: right as a form of knowledge independent of human agents.
Strikingly, the epistemological groundwork of this thought may precede Plato. It is arguably indicated in Sokrates' reported dictum, 'aretÂ is epistÂmÂ', virtue is knowledge. 188 If so, then the distinction between human activity and outcomes or states of affairs may already have seemed dissolved. Moreover, some such dissolution may indeed be necessary if what is right is to be conceived as having objective reality, rather than being intersubjectively constituted as the Athenians seem to have believed. It was left to Plato, however, rather than Sokrates, to show in writing how this might be done; and to the extent that we remain persuaded by a conceptualization of justice as a form of knowledge, accessed through experts, that ought to govern human activityas opposed to a form of activity, developed in community, that itself helps to constitute what is known -Plato seems partly responsible for getting us there.

V Conclusion
While none of Plato's institutional proposals -'philosopher-kings', 'nocturnal councils' and so on -were to gain long-term traction, his influence on the 185 Rep. 517d. 186 See also ibid., 336a and 343c. 187  conceptualization of justice has been profound. To be sure, the immediate context was unpropitious. The main foothold of an objective conception of to dikaion in Athens, i.e. the distinctive treatment of homicide, began to be eroded in the late fifth century. Rather than being heard by the Areopagos in the traditional fashion, an alternative procedure known as apagÜgÂ, previously used for offences such as theft, highway robbery and seizure of persons, began to be used to bring homicide before the popular courts, where it was judged in the ordinary way. 189 The significance of the Areopagos as a vehicle for ascertaining both 'what is right' and 'what is true' may thus have been fading. On the other hand, from the 340s the Areopagos began to gain power in another way, through another procedure, apophasis. 190 This procedure cast the Areopagos in the role of an apparently neutral fact-finding institution: it required the Areopagites to produce a preliminary report, most commonly in relation to charges of treason, corruption and official misconduct, to establish the facts of a case before it went to the popular courts. 191 Yet if this new role could be construed as political, it scarcely helped Plato's cause. It certainly confirmed the special relationship of the Areopagos to knowledge; 192 yet the fact that the popular courts were ready to acquit defendants whom the Areopagos found to have committed the acts of which they were accused only confirms the final authority of ordinary citizens as judges. 193 In the longer term, however, Plato's innovations, in particular the focus on dikaiosynÂ, could be seen to be gaining ground. A useful witness here is Diogenes Laertios. Prior to Plato, works whose titles were later translated 'On justice' bear the title Peri tou dikaiou, while later ones bear the title Peri tou dikaiosynÂ. 194 The only exception here is Aristotle, who wrote separate works on each. 195 It is tempting to speculate on the contents of Aristotle's texts, especially given the possible influence of Protagoras on his thought. 196 Apparently he at least resisted the dissolution of the distinction between to dikaion and dikaiosynÂ that Plato had advanced. Judging from his extant works he will have exhibited a less intellectualist position as well. But as Diogenes also relates, Aristotle's school was short-lived. His immediate lineage died with his student Theophrastos. 197 Not until the middle ages was he fully revived, and the Aristotle of the Thomists was a very different character.
The reception of Plato has been different. This is not the place to trace the path of his works to the present day, but something significant may be inferred from the fact that, alone among classical authors, we appear to have all, or nearly all, his writings. Platonism was evidently sufficiently congruent with Christianity for his works to be continually recopied (in the Greek east if not, until the fifteenth century, the Latin west). 198 The importance of this fact can hardly be overstated. It means Plato has been able to exert a more abiding influence on Western political thought than any other ancient thinker.
This influence remains visible today. For though the standard Athenian approach to adjudication is both compatible with democratic politics and philosophically defensible, the dominant modern conception of justice and its execution is arguably much closer to Plato's. Of course, few would accept Plato's argument about the forms as it stands. But an alternative intellectualist approach to justice is available, according to which justice is knowledge because the nature of transgressions is defined in the laws. Hence judicial decision-making requires expertise in law. Since ordinary people lack this expertise, they cannot act as judges (though they may participate in judicial activity in a more attenuated way, as jurors).
This view may seem more attractive than Plato's, but it has the same shape and shares some of the same assumptions; and it is not unproblematic. One reason for doubt goes back to the point about language made above. The core supposition of the alternative intellectualist approach is that the whole body of law considered together is, or could be, effectively self-interpreting. If one had access to all the relevant information given in the laws, one could make a perfect judgment. So Montesquieu: judgments ought to be 'fixed' by being 'ever conformable to the letter of the law', and judges 'no more than the mouth that pronounces the words of the law'. 199 These hopes are echoed in the work of Ronald Dworkin, for example, whose superhero-judge Hercules assumes that the laws with which he deals are structured by a coherent set of principles, to be applied in the same way to each case that comes before him. 200 Both accounts reveal a commitment to the idea that there is a right answer to every legal problem available in the laws themselves, which can be ascertained if only the judges in each case are sufficiently disinterested to act as a proper vehicle.
An Athenian democrat, however, could he be brought to understand this view, would probably respond that laws are not self-interpreting, in the same way and for the same reason that any linguistic item is not self-interpreting. The need for human beings to judge does not arise merely because the laws themselves cannot speak. Human agents -specifically, members of the linguistic community in question -are necessary because only they can decide on the meanings of the terms laid down in laws and how every particular case relates to them. No amount of technical training will help with that, nor will the lack of it make one a less capable judge.
As Aristotle argued, 'men do not deny that it must be for a human being to judge' such matters; they merely dispute how many men ought to perform that task. 201 Plato's preference was for a very small number, on the basis of an account of justice that rested on the internalization of outside knowledge. Athenian democrats opted for many, on the basis of their membership in a community whose judgments were taken to be the final measure of what was right for those within it. Not the least remarkable feature of democracy today is that the conception of justice held by many self-described democrats seems closer to Plato's than to that of his fellow citizens.

Daniela Cammack
HARVARD SOCIETY OF FELLOWS