Week 10

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC*

By Gerasimos Santas

I. Introduction

Plato’s theory of social justice in the Republic is antidemocratic, by his own lights as well as by historical and contemporary consensus. His reasons for criticizing democratic theory are not as clearly agreed upon, except in very broad outline: he thinks that democracy prizes freedom far too much and knowledge far too little. Even if his reasons are better understood than I am implying, they certainly have not been found very convincing; a sign of this is that many scholars ignore or lay aside his political theory of social justice, as if it were not there at all, and concen- trate on his ethical theory of justice in the soul and of the good person, which looks like a virtue ethics — something that is getting attention in contemporary discussions of ethics.

Part of the reason for this state of affairs may be that we have witnessed the horrid excesses of some antidemocratic regimes, by comparison to which modern democracies can shine. Another reason, I think, is the fault of Plato’s own lack of surface system: his informal dialogue style some- times masks some fundamental ideas and their influence on his thinking, while allowing him to tread lightly when he pleases and even to substi- tute character sketch, irony, and sarcasm when argument is needed. The lighter passages provide much needed (sometimes comic) relief and make the Republic very readable and popular even when the reader hates the ideas. But they make the book harder, not easier, to understand.

A case in point is Plato’s discussion of the (presumably Athenian) democratic constitution. He correctly identifies its main principles as free- dom and equality. When it comes to criticism, however, he seems to shower it with irony and sarcasm: twice he calls it “most beautiful” (Rep. VIII.557c);1 and after he notes that it disdains his fundamental principle of social justice (as if that by itself were a criticism), he says that “it is a delightful form of government, anarchic and motley, assigning a kind of equality to equals and unequals alike” (Rep. VIII.558c4–6). Well, democ-

* I wish to thank the other contributors to this volume, and its editors, for many helpful comments.

1 All references inside parentheses are to John M. Cooper, ed., Plato: Complete Works (Indianapolis, IN: Hackett Publishing Company, 1997). Translations are by Paul Shorey, Plato: The Republic (Cambridge, MA: Harvard University Press, 1956), unless otherwise indicated.

70 © 2007 Social Philosophy & Policy Foundation. Printed in the USA.

racies are not anarchic — somebody is always ruling. And why not treat equals and unequals alike? My two children may be very unequal, but I treat them alike and that is thought to be fair.

Plato writes far too briefly here. Except for the theory of Thrasymachus, he proceeds indirectly with other oppositions; he does not subject Glaucon’s probably democratic theory of justice to the Socratic elenchus (cross- examination), nor the various constitutions he considers unjust in Rep. VIII. After the first book, Plato seems rather to put up his view as an alternative to these other theories and invite us, the readers, to choose. To appreciate his reasons against democracy we need to look at some of his reasons for his own view. Thus, his reasons against treating equals and unequals alike are likely to be found in his reasons for his own principle of social justice.

In any case, his criticisms of democracy could not be misunderstand- ings; he was well acquainted with the democracies of his day and lived most of his life in arguably the best of them. His thought in ethics and politics has great originality, breadth, and depth, and is still worth study- ing. Moreover, much as we may prefer democracy to other practically available forms of government, we are hard put to deny that democracies, ancient and modern, have displayed faults and imperfections of many kinds. Can we then learn from Plato’s criticisms at least how democracies might become better, where by “better” we mean at least more just and/or more conducive to the good of all their citizens? This is the main question I wish to pursue in this essay.

II. Abstract or Formal Theory and Empirical Assumptions

I propose to proceed by examining briefly Plato’s handling of three subjects relevant to my question: private property and wealth; knowl- edge; and freedom. I want to start, however, by observing a distinction I will be relying on, a distinction between (1) Plato’s more or less formal or abstract theories and (2) his application of such theories, by means of assumptions or empirical premises, to reach substantive political or eth- ical theses about property, knowledge, and freedom. I hope that I can use a noncontroversial version of this distinction. The main use I will make of it will be to try to locate where in his theories Plato may have made empirical or other mistakes and where his theories can be revised accord- ingly while remaining essentially Platonic; and also where his assump- tions have held up and democracies can learn from them.

First, let me illustrate the distinction in an important case. In Rep. I, Plato has Socrates define function, then characterize being a good thing in terms of it, and finally characterize virtue also in terms of it: (1) The function, or ergon, of each thing that has a function is (a) what it alone can do or (b) what it can do better than anything else (352e2–353a11). (2) A

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 71

thing that has a function is good if it performs its function well (353e). (3) The virtue of anything that has a function is that by which it performs its function well (353b14–c7). I put these three definitions or assumptions on the side of abstract or formal theory. In contrast, the substantive thesis that the function of the eyes is seeing I take to be reached on the basis of, or explained by, the empirical proposition that we can see only with the eyes (together with the definition of function). The case is similar when Socrates himself applies the abstract theory to convince Thrasymachus of his ethical thesis that justice makes the just man happy (353d–354b): Socrates uses all three abstract definitions, along with the metaphysical or empirical proposition that we can live, deliberate, and decide only with the soul, and the ethical thesis that justice is the virtue of the soul. I think it is clear that Socrates could be correct in his abstract theory and incorrect in one or more of the substantive premises he uses to reach his thesis.2

(Aristotle provides another example of the distinction: he had an abstract theory of function as activity unique and essential to a kind of thing; and he applied the theory to try to discover, say, the function of the leaves of plants; but he was mistaken about what activity he thought was unique and essential to leaves; the function of leaves — photosynthesis — was not in fact discovered until the middle of the twentieth century — by a bio- chemist. We have a similar story with the function of the heart — it took William Harvey’s work and some assistance from Italian physicians to discover the function of the heart. We have a deceptive familiarity with some organs, and we come to think it is easy to tell what the functions of things are.)

I think this kind of thing happens again and again in the Republic, and it is not always noted by critics. Plato is sometimes criticized without his critics’ adequately locating what is supposed to be his mistake: is it the abstract theory, or a metaphysical (or empirical) premise used in the application of the theory, that is at fault? If we can locate the mistakes, in the abstract theory or in the empirical or metaphysical substantive assump- tions, we can determine where the theory needs revision, and in what science we need to search in order to find truths to replace the mistakes.

I begin with Plato’s treatment of property and wealth, since it is clear that he proceeds in the way I have indicated; and both his proposals here and current democratic practices can be critiqued and can be brought closer together.

III. Property and Wealth in Plato’s Completely Good City

I take up Socrates’ proposals about private property and about extremes of poverty and wealth at the end of Rep. III and the beginning of Rep. IV,

2 For another illustration of the division of labor between abstract principles and empir- ical premises at the heart of Plato’s theory of justice, see Gerasimos Santas, Goodness and Justice (Oxford: Blackwell Publishing, 2001), 112–13.

72 GERASIMOS SANTAS

rather than the more elaborate proposals for common property and fam- ily in Rep. V. The latter are made on the basis of (a disputed notion of ) unity being the greatest good and civil war the greatest evil. In the earlier books, however, the concern is about the effects of private property and wealth on guarding and ruling the city, and about the effects of extremes of wealth and poverty on the craftsmen and the providers of the city.

In Rep. III, Plato is primarily concerned with the native qualities that make people best suited to defend and rule the city and with the early education of such persons with a view to the best performance of these functions. At 412c, Socrates remarks that as the best farmers are those who are “the most at farming,” so the best guardians are those who are “the most at guarding” the city. He continues: “They must then to begin with be intelligent in these matters and capable, and furthermore careful of the interests of the city. That is so. But one would be most likely to be careful of that which he loved. Necessarily. And again, one would be most likely to love that whose interest he thought to coincide with his own, and thought that when it [the city’s interest] prospered he too would prosper, and if not, the contrary” (Rep. III.412c–d7). Next, Socrates proposes a series of tests and trials for the rulers and the soldiers, trials of pleasure and pain, desire and fear, to insure that they would in all circumstances preserve the conviction that they must do what is best for the city, and that what is best for the city is best for them and not otherwise. He ends up with the myth of the metals rounding out their education. To sum up, Socrates has outlined the rulers’ and soldiers’ native qualities, the engen- dering of beliefs and desires about what is best for the city, and tests and trials to preserve these beliefs so that, in all circumstances, the rulers and soldiers act in accordance with their desire for the good of the city.

But then (Rep. III.415e–417b) we learn that the correct native qualities and the right education are not enough. There is still a danger that the guardian dogs will turn into wolves. “Must we not guard then by every means in our power against our helpers treating the citizens in such a way and, because they are stronger, converting themselves from benign assistants into savage masters?” (416b1–3). The guardians’ education, even if it is really a good one, is not enough of a safeguard. “In addition to such an education a thoughtful man would affirm that their houses and the possessions provided for them ought to be such as not to interfere with the best performance of their own work as guardians and not to incite them to wrong the other citizens” (416c5–d1). If they had private prop- erty, private houses, land of their own, and money, they would be “house- holders and farmers instead of guardians”; they would become not helpers but “enemies and masters” of their fellow citizens, and end up causing a shipwreck for themselves and the city (417a–b). Therefore, they must not have private property, private houses, land, or money.

This argument assumes the abstract functional theory, the previous assignment of the three main functions in the city (ruling, defending, and

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 73

providing), the determining of the innate qualities suited best for each of these functions, and the early public education of the citizens. It then proceeds by supposing that the rulers and the soldiers would be dis- tracted from performing their functions or would not perform them well if they were allowed private property, private houses, land of their own, and money. Whether this supposition is true is clearly an empirical issue. (It could well be that all the preceding issues, except the abstract theory itself, are empirical, but I lay this aside here.) Is the supposition true?

In the absence of experiments (as far as I know), we can only go by historical and contemporary experience. To begin with, Plato’s proposals here are clearly a recognizable version of separating institutionally polit- ical and military power from property and wealth. These are reforms (relative to the Athenian practices of his day) for the economic and polit- ical institutions of the city. They are immediately comparable to attempts in modern democracies to institute some such separation between wealth and political power, by means of blind trusts, financial disclosure require- ments, and campaign rules for raising and spending money.

We can also see that Plato’s proposals here are extreme: his guardians’ manner of life is more Spartan than their counterparts in Sparta — and Plato knows it. He has Glaucon assent to the proposals emphatically, but his Adeimantus immediately makes a serious objection: to deprive the guardians — rulers and soldiers — of all these things would make them unhappy; for these are the very goods the possession of which is thought to make men happy (Rep. IV.419e–420a). Indeed, these are the goods that Thrasymachus’s heroes and men with the ring of Gyges go after. The objection is fundamental, since Plato agrees that everyone pursues his or her own happiness and Plato’s guardians would have more power than anyone else to pursue these goods. But his Socrates does not retreat. After observing that he and his interlocutors set out not to make any one class in the city especially happy (as Thrasymachus’s cities do), but to make the whole city as happy as possible — for they thought that only in such a city would they find justice — Socrates challenges the conception of happiness used in the objection. He would not be surprised, he says, if his guardians were “most happy,” and suggests that their happiness will be found not in the possession of fine houses and lands, but in being “the best crafts- men in their own work.” “[A]nd so, as the entire city develops and is ordered well, we must leave it to nature to provide each group with its share of happiness” (Rep. IV.421c3–6, trans. Grube/Reeve). The rulers will have a ruler’s happiness, to be found primarily in the best performance of their work as rulers, and/or in the satisfaction of being the best crafts- men in their own work; the soldiers similarly will have a soldier’s hap- piness, and the craftsmen and traders will have the kind of happiness appropriate to their nature and work.

It is doubtful, to say the least, that in the real world people can be found who are fit to be Plato’s rulers (even by his own lights, since his Socrates

74 GERASIMOS SANTAS

does not make it, though for different, epistemic reasons), or who would be “most happy” without nearly any of the goods that most people enjoy and that Socrates excludes. No rulers that we know of, as far as I can tell, have been satisfied to live in the conditions of deprivation that Socrates describes. Nor have any known states required their rulers to live so. Even in states where the means of production are publicly owned, per- sonal property is allowed; and even where it is not, rulers live in palatial public houses (e.g., the former Soviet Union). We know of actual democ- racies that have been compatible with public ownership of the means of production, such as socialist England after the Second World War, but it allowed personal private property for all, including rulers, and even allowed considerable wealth. John Rawls, our most fundamental recent defender of democratic theory, considers his democratic principles of justice to be compatible with such public ownership, though not with the abolition of personal private property.3

In sum, we have no actual experience, so far as I know, that would show that Platonic rulers or soldiers, living under such conditions as Plato specifies, can rule well or be happy or both. Plato’s reforms are most extreme and there is no assurance that they can be realized.

Nonetheless, he has an important point. Advocates of democracy acknowledge that a close link between ruling (or soldiering) and large properties or great wealth pose a danger to ruling well (or defending the country well), since, by the advocates’ own lights, ruling well must involve (as far as possible) political equality in ruling and equality of maximum civil liberties. Wealthy men are not excluded from office, but on attaining office they have to disclose their wealth, their wealth may have to go into a blind trust, and they have to rule behind a veil of ignorance — ignorance of how their public decisions will affect their wealth. The unending attempts and fights for campaign reform in the United States acknowledge the same danger.

But these democratic measures to separate wealth from ruling are fee- ble compared to Plato’s, and have been feeble and largely ineffective as means to their declared end (equality in ruling and in civil liberties). For example, about a third of the members of the U.S. House of Representa- tives are millionaires in annual income, not just assets, and the same is true of the Senate,4 while U.S. presidents are millionaires when they run for office or can become so the moment they leave office (by selling their memoirs); but only one percent of the population they represent are millionaires. Even when they are not millionaires, people running for

3 John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1999), 53, 57. 4 Reported by Agence France Presse, June 30, 2004, on the basis of annual financial disclo-

sure statements by U.S. congressmen; the figures are for the year 2003. These figures vary from year to year, but it is clear enough that a far greater proportion of U.S. congressmen are wealthy than the general population.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 75

office need millions of dollars for their campaigns, and most of them most of the time have to put themselves in debt to millionaires.5

There must be some ground in between these extremes, which would not make Plato’s city less Platonically just or a democratic city less dem- ocratic. John Rawls recognizes the ineffectiveness and resulting unfair- ness of the current relevant democratic practices, and his proposals are somewhere in between those practices and Plato’s proposals on property and wealth.6 Given the distinction I am using, I think Plato can accept the idea that his guardians could rule well even if they had some modest property or lived in better public quarters, and still hold on to the essence of his theory of social justice; likewise, democracies can move closer to Plato on the link between ruling and wealth and still remain democratic.

The other proposal that Socrates brings up next is to set limits to wealth and poverty for the rest of the population, the craftsmen and traders, who are allowed private property in the completely good city (Rep. IV.421d– 422b). Once more, the argument proceeds on the basis of the abstract functional theory and empirical assumptions about what is needed for the artisans to perform their function well. A potter who grew rich, Soc- rates claims, would no longer mind his craft; he would become idle, negligent, and a worse potter. But a potter who was poor would be unable to provide himself with tools and other things he needs, and once more would not be able to do his work well. “From both causes, then, poverty and wealth, the products of the arts deteriorate and so do the artisans. . . . Here then is a second group of things [i.e., wealth and pov- erty], it seems, that our guardians must guard against and do all in their power to keep from slipping into the city without their knowledge . . .” (421e4–422a1). Generally, Socrates claims, wealth saps motivation in crafts- manship; a tanner does his job well, becomes rich, gets bored with tan- ning, and runs for office. In contrast, Socrates may sculpt well, but has no money to buy marble.

Once more, modern democracies acknowledge the need to set wealth ceilings and poverty floors. Such measures as inheritance taxes, now under debate in the United States, presumably were designed to prevent extreme concentrations of wealth over generations, while such things as “the negative income tax” and Rawls’s difference principle are clearly designed to eliminate extremes of poverty.7 Democracies’ reasons may be different from Plato’s: extremes of wealth and poverty produce inequal-

5 In England in the nineteenth century, John Stuart Mill was worried about a similar problem when he spoke of “the anomaly of a democratic constitution in a predominantly plutocratically constituted society.” See J. H. Burns, “J. S. Mill and Democracy, 1829–61,” in J. B. Schneewind, ed., Mill (Garden City, NY: Doubleday, 1968), 286.

6 Rawls, A Theory of Justice, 242–51. 7 The difference principle states that “the higher expectations of those better situated are

just if and only if they work as part of a scheme which improves the expectations of the least advantaged members of society.” Ibid., 65. For Rawls’s extended discussion of this principle, see ibid., 65–73.

76 GERASIMOS SANTAS

ities in ruling and in the enjoyment of civil liberties. In the United States, for example, some 36 million citizens are below the official poverty line, and most likely their participation in elections and governing is very unequal to the participation of the wealthy. Rawls draws a distinction between liberty and the worth of liberty;8 for example, all citizens have an equal right to free speech, but the wealthy have greater access to the means of exercising this right, such as newspapers, radio, and television. Or consider the right to travel: all citizens can have it equally, but its exercise will vary widely with the variation of means — the poor go nowhere. The democratic ideals of freedom and equality cannot be effec- tively realized with extremes of wealth and poverty. Rawls proposes constitutional measures to compensate for the unequal worth of liberty caused by inequalities in income and wealth.9

Actually, Plato’s reforms on extremes of wealth and poverty seem too Spartan (see Rep. IV.422 on wealth and fighting wars). They are presum- ably guided not by considerations of equality as in democracies, but by the aim of ensuring that the city is ruled justly (by Plato’s standard of justice for the city). But how much wealth or poverty are too much? Given the enormous differences in size, numbers, and technologies between ancient Greek city-states and modern nation-states, Plato’s view on how much wealth is necessary or best to do one’s job well, and how much is too much, might need considerable revision.

At the same time, the modern democracies’ “war on poverty” seems like another lost war, and there seems to be no substantial agreement on the limits of wealth and poverty. Rawls’s difference principle ties the fortunes of the less fortunate (in talents and social and economic resources at birth) to the fortunes of the more fortunate in all these respects. The principle still allows great inequalities in property and wealth, to allow for incentives and productivity and for the more talented and hard- working to be compensated for costs of their additional education and training. But the upward movement of the more fortunate is allowed only so long as it brings with it an improvement for the less fortunate.10 Sad to say, the difference principle is the most controversial of all Rawls’s prin- ciples (and the objections do not come from the poor), even though argu- ably the principle seems fair, and its concrete implementation seems far more fair than the present state of affairs (great wealth seemingly at the expense of great poverty).

In any case, both Plato’s ideal city and modern democracies can agree that property and wealth, and extremes of wealth and poverty, can sig- nificantly affect ruling and important social and economic functions — and thus require regulation.

8 Ibid., 179 –80. 9 Ibid., 196–200.

10 Ibid., 65–73.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 77

In sum, I suggest that while Plato seems correct in supposing that doing well the social task that a citizen can do best is one important human good (and one important component of the citizen’s happiness), he may be considerably mistaken in the empirical assumptions he makes about how much of the usual material goods are necessary or effective for doing one’s job well, whether one is a ruler, a soldier, or an artisan. Plato’s completely good city and its citizens might require more of these goods and still remain Platonically just. But he is right to point to the dangers of wealth and poverty, and democracies are subject to these dangers as much as other regimes, and need to lean more in Plato’s direction.

In the Republic, Plato integrates his politics and economics in a different way from modern democracies; he proposes far greater economic equal- ities than are dreamed of by any democracy, though he proposes them in an almost paradoxical mixture with political inequalities. The deeper bases for these economic equalities are Plato’s functional theory, the pri- macy of doing well what one does best as part of the human good, and the parallels between ruling and the other arts and sciences. Modern democracies, too, try to moderate economic inequalities; but their deeper basis is clearly political equality; economic inequalities are based on eco- nomic realities (e.g., motivational incentives and costs of training) and are to be mitigated mainly for the sake of maintaining political equalities and the meaningful exercise of liberties. If we take Rawls’s well-ordered soci- ety as a modern democratic ideal, a democracy highly approximating it would eliminate the worst existing economic inequalities through the institutional requirements of the difference principle (such as a negative income tax) and of the principle of fair equality of opportunity.11 Such a society would lean in Plato’s direction to some degree, though for differ- ent reasons.

IV. Platonic Knowledge and Democratic Rule

Plato’s ideal city is usually thought to be antidemocratic and elitist with respect to knowledge. But this clearly does not include all knowledge: not knowledge in the crafts, or in the arts and sciences generally, but knowl- edge required for ruling the city well, wisdom of what is good for the city as a whole in its internal and external relations (Rep. IV.428b–429a). This is the knowledge (not the other kinds) by reason of which a city is called wise; it is the possession of this virtue that entitles some of the citizens to rule over others, and not the consent of the ruled — though the virtue of (civic) temperance implies consent by all citizens that the wise should rule, and presumably brings harmony between rulers and ruled and makes the ideal city stable.

11 Ibid., 57–81, 86–93.

78 GERASIMOS SANTAS

This much is clear enough in Rep. IV, and by itself it is not necessarily elitist or even antidemocratic. Maybe all the citizens can acquire such wisdom with appropriate educations; and maybe the best way to find out if the ruled consent (and the city is temperate) is by elections of some kind. It depends on whether some of Plato’s empirical assumptions are true; for example, whether he is correct about the natural lottery, sym- bolized by the myth of metals (Rep. III.415), the idea that nature distrib- utes talents and abilities unequally at birth. It may be that, contrary to what he thought, there is enough gold in everyone’s soul, though not necessarily in the same proportions to the other metals, so that everyone can acquire the wisdom in question.

At the same time, modern democracies can lean in Plato’s direction to some extent and still be democracies: literacy tests for voting would not be objectionable if there were no citizens so poor they could not afford an education (and assuming the tests were applied impartially); indeed, absent the poor, the requirements for voting might include a high school edu- cation. Why would it be antidemocratic to require informed voting under such conditions? And absent the poor, the requirements for running for office might go beyond residency and age, to a certain level of education. We have such requirements for other professions; why not for ruling? Even if it is the consent of the ruled that entitles some to rule over others, surely the wisdom of Rep. IV would be one of the rulers’ virtues?

As we all know, however, Rep. V, VI, and VII wipe out these possible overtures of moderation and even conciliation. The apparent innocence of Rep. IV about wisdom turns out to be an illusion. Socrates draws a very strong distinction between knowledge and opinion, requiring the meta- physics of the Forms; the wisdom of Book IV is now said to require knowledge of the Form of the Good; and such knowledge, in turn, is said to require such a demanding higher education in the sciences that it is highly doubtful that there is enough gold to go around so that more than a small minority can attain such knowledge. As if all this were not enough, Plato’s metaphysics of the Form of the Good is so obscure that after some twenty-four centuries we have no interpretation of it which is intelligible enough and will bear the enormous political burden Plato puts on knowl- edge of his good. Somebody has to know the Form of the Good in order to rule a city well (that is the political burden), but nobody does! Plato sets the bar so high that according to him even his Socrates of the Republic — the very character who proposes all the important ideas of this great work — has no knowledge of the Form of the Good (Rep. VI.506c2–3).

I think we should ask whether Plato’s theory of justice can be isolated in thought from his metaphysics and epistemology and still remain a significant theory of justice, including some of its antidemocratic ele- ments. (Here my distinction between abstract or formal theory and empir- ical assumptions takes a slightly different shape: it is the distinction between the abstract theory of justice and the metaphysics and epistemology, which

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 79

can also be abstract.) Of course, the Republic would not be the great and comprehensive work it is without its middle books (V, VI, and VII). Some interpreters think these books are the very essence of the work. And some think that the middle books show that Plato thinks we cannot do ethics and political philosophy well without some assumptions or theories in metaphysics and epistemology; and this is certainly one tradition in moral philosophy.

But I do not want to dispute these points; indeed, I think they are largely correct. I am not suggesting that we rewrite the Republic, or read it and try to understand it without the middle books; nor am I attacking the integrity and unity of the work. What I wish to suggest is that Plato’s theory of justice, both of the just city and of the just person (though I am concerned here only with social justice), can have a significant and instruc- tive interpretation without his special metaphysics and epistemology; certainly, there can be theories of justice and the human good arguably without epistemology or metaphysics.12 Or we can have Plato’s theory of justice and the human good with a weaker metaphysics and epistemol- ogy: without Forms, or with Forms as properties without self-predication, and with knowledge as justified true belief, perhaps highly probable belief rather than certainty (at least in the nonformal sciences).

The way the Republic is written seems to allow, to some degree, for such an imaginary isolation. The theory of what justice is and the defense of justice are put forward by Socrates in Rep. II, III, IV, the first half of V, and the later books VIII and IX. Socrates develops the theory of what justice is in the first group without any reference, not to speak of essential ref- erence, to the theory of Forms or the conception of knowledge that goes with it. Even after the Forms and knowledge of them has been elaborated in the middle books, these theories are not used explicitly in the classi- fication and rankings of unjust cities and persons in Rep. VIII. To be sure, though not explicitly used, the metaphysics and the epistemology may be necessary for the main arguments in the defense of justice in Rep. IX.13

Still, the only later passage in which the theory of Forms is explicitly used is a small stretch of Rep. X, where the theory is used to explain the nature and value of works of art. It would seem, then, that Plato himself allows that his theory of what justice in the city and in a person is can be understood, or at least stated, without the theory of Forms — and similarly

12 See John Rawls, “Justice as Fairness: Political, not Metaphysical,” Philosophy and Public Affairs 14 (1985). For a similar attempt to distinguish and examine “the rule of reason” in Plato’s Republic apart from his controversial metaphysics and epistemology, see Fred D. Miller, Jr., “Plato on the Rule of Reason,” The Southern Journal of Philosophy 43 (2005): 50–83.

13 David Keyt has pointed out to me that the third argument that justice pays in Book IX, the argument about real and illusory pleasures, especially at 585d11–e4, may contain a reference to Forms. I am inclined to agree, at least in the sense that the argument may be best understood by reference to knowledge of the Platonic Forms. Also, Plato’s defense of the philosopher as the most just of men and the best qualified to rule would indeed be difficult without his metaphysics and epistemology.

80 GERASIMOS SANTAS

for at least some of his defenses of his theory of justice, against the rival theories of Glaucon and Thrasymachus, and against the timocratic, oli- garchic, and democratic theories of justice.

It might be argued that Plato’s justice is more deeply understood and better defended if we bring in and integrate with it his metaphysics and epistemology (and perhaps the grand metaphysical teleology of the Timaeus). The “longer road,” deferred in Rep. IV and taken up in Rep. VI, points to such integration.14 And this is certainly an interpretation of the Republic that maintains its unity and integrity. But still this longer road is not adequately explained because of the obscurity of the Good and because no one ever traveled such a road. Plato’s theory of justice may be better understood with his metaphysics and his epistemology, and perhaps bet- ter defended; though given how prolific the ontology seems to be and how demanding Plato’s theory of knowledge, there can be serious doubts about the practicability of any theory of justice that depends on them. In any case, I do not propose that we regularly separate the theory of justice from the epistemology and metaphysics and read and try to understand the Republic without the middle books; but only that we look for a moment at the theory of justice in isolation from the epistemology and metaphys- ics and see whether Plato’s antidemocratic proposals and his criticisms of democracy are still significant and instructive. This we can certainly do.

The question remains, though, how significant the remainder from this isolation experiment is. Some might think that Plato’s justice without his metaphysics and epistemology is not Plato’s theory anymore. Or, even if it is, it is now trivial, no longer interesting, no longer a significant theory of justice.

I do not think this is correct. One sign that it is not correct is that the two principal questions of the Republic, what justice is and whether we are better off being just rather than unjust, can be and are answered mostly in passages that make no explicit reference to or use of the theory of Forms or the epistemology that goes with it. The first question is answered on the basis of the functional theory and empirical assumptions, such as the functions of the city (and of the soul), the natural lottery, and what talents and education are necessary or best for performing well various social (and psychic) functions. The first defense of justice is founded in the health-justice analogy, which relies on the conceptions of organisms and functions of organs in the medicine and biology of the day, not on Pla- tonic metaphysics or epistemology; and the attack on Platonic injustices, and the ranking of constitutions and persons (Rep. VIII), do not rely on the metaphysics and the epistemology.

Another sign that we still have a significant theory of justice is that the contrast and opposition to the theories of justice of Thrasymachus and

14 See Terry Penner, “The Forms in the Republic,” in Gerasimos Santas, ed., The Blackwell Guide to the Republic (Oxford: Blackwell, 2005), 239 –40.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 81

Glaucon remain intact, absent the metaphysics and epistemology. Con- trary to Thrasymachus, Platonic justice is to be found in a city so struc- tured as to promote the happiness of the city as a whole (Rep. IV.420b7–8), not the greatest happiness of the rulers; and the happiness of the city is to be found in doing well what one does best, not in power, property, and wealth. And contrary to Glaucon, in Plato’s theory of justice without the metaphysics and the epistemology, the freedom to do as one pleases and get away with it is not the greatest good (indeed, perhaps not a good at all); the functions of the city-state go far beyond the minimal Glauconian function of protecting citizens from harming each other; and the nature of justice is not to be found in the circumstances of conflict and competition for scarce material goods and in a contract made because of them.

Yet another sign that we still have a significant theory of justice is that all the oppositions between Platonic justice and timocracy, oligarchy, democracy, and tyranny (Rep. VIII) also remain intact. Absent the meta- physics and epistemology, Plato’s justice is still not honor based, not wealth based, not freedom and equality based, and not power based. The theory of justice that remains, absent the metaphysics and epistemology, is still meritocratic: the ideal city is a meritocracy — merit being talent and education for the main social functions — and it is just for the city to enforce this, so that there is no free choice of occupation. With respect to the function of ruling the city, the ideal city is an epistocracy (to use David Estlund’s term)15 — a special case of meritocracy, though with a reduced but still significant conception of knowledge.

Such meritocracy is still antidemocratic. Rawls takes pains to show that his theory of justice is not meritocratic, especially his principle of fair equality of opportunity: the principle that those with similar inborn tal- ents and abilities and similar willingness to develop and use them should have similar prospects of success.16 This looks meritocratic. It is not a case of justice as absolute or arithmetical equality, but of proportional equality: equality in life-prospects based on equality of merit (talent or ability, and motivation). And since it is a principle of justice, it is enforceable (by such institutions as free public education), though it is enforceable at the level of institutions and not individual actions, and as a case of pure proce- dural justice, not a justice of outcomes. It is difficult to believe that Plato would object to this principle by itself (though he would have a different conception of “prospects of success”). But of course the whole of Rawls’s theory is not meritocratic; his well-ordered society allows for maximum equal liberties, including free choice of profession (formal equality of opportunity).17 And that society would not be an epistocracy: consent of

15 David Estlund, “Why Not Epistocracy?” in Naomi Reshotko, ed., Desire, Identity, and Existence (Kelowna, Canada: Academic Printing and Publishing, 2003), 53. Epistocracy is, literally, the rule of knowledge, as democracy is the rule of the people.

16 Rawls, A Theory of Justice, 63, 73, 86–93. 17 Ibid., 62, 73.

82 GERASIMOS SANTAS

the ruled, not knowledge of the rulers, is the basis for the authority of ruling.

But if we abstract from the issue of the basis for ruling, and if we moderate Plato’s wisdom to the claim that ruling well requires some attainable and publicly understood knowledge (or at least education), as other professions do, rather than Plato’s stronger claim that it requires knowledge of the Form of the Good, then we can have something that seems more reasonable and acceptable, and democracies can lean in a significant way in Plato’s direction. As I mentioned earlier, given the absence of poverty and given universal education, the qualifications for various offices can be raised so that at least a minimum of relevant knowl- edge (or education) is required, as the qualifications for physicians or attorneys have been raised over the years. Democracy allows this now for civil service positions. In the absence of poverty and with free public education, ruling with higher qualifications for office can still be based on the consent of the governed.

Another constitutional device leaning in the direction of epistocracy would be plural voting. John Stuart Mill and John Rawls both believed that such a practice was compatible with democratic justice: instead of one citizen, one vote, we have one citizen at least one vote; some citizens have more than one vote in proportion to their education (rather than their knowledge).18 There are several reasons for such a proposal. To begin with perhaps the least important, the ideal of equal political liberty, with the precept of “one elector, one vote” and all that this implies about representation, is extremely difficult to attain, even to highly approxi- mate: representation in the United States Senate, for example, completely disregards the implication of this precept — that members of legislatures, with one vote each, represent the same number of electors — and even the U.S. House of Representatives, designed to be truly representative, only approximates it.

The main reason for scholocracy (the rule of scholars) is the idea that the better educated would rule better or more wisely. Thus, where there are populations in which some are better educated than others (a univer- sal condition, it seems), the more educated would have more votes so that their opinions would have greater influence. Mill perhaps was influenced by the Platonic analogies between ruling and other arts; he thought that in a common enterprise in which everyone has an interest, such as ruling, and such other arts as medicine and navigation, everyone should have a say — but not necessarily an equal say, since some can have more relevant knowledge than others and they can act better for everyone’s benefit,

18 Ibid., 203–6. Rawls includes a critical discussion of plural voting proposals by Mill. See also Burns, “J. S. Mill and Democracy, 1829–61,” 318–28; and Estlund, “Why Not Epistocracy?” 61. Estlund calls a democracy with more than one vote for the educated a “scholocracy” — the rule of scholars.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 83

including the benefit of those with a lesser say.19 Rawls, with his priority of liberty, which prohibits trade-offs between liberties and other primary goods, would restrict the range of benefits that unequal political liberties might provide to an increase (or at least not a decrease) of the other liberties. And to Mill’s condition that plural voting must be for the benefit of all, including those with the lesser votes, Rawls adds the condition that it must be acceptable to those with the lesser votes. His statement on this is worth quoting:

Government is assumed to aim at the common good. . . . To the extent that this presumption holds, and some men can be identified as having superior wisdom and judgment, others are willing to trust them and to concede to their opinion greater weight. The passengers of a ship are willing to let the captain steer the course, since they believe that he is more knowledgeable and wishes to arrive safely as much as they do. There is both an identity of interests and a notice- ably greater skill and judgment in realizing it. Now the ship of state is in some ways analogous to a ship at sea; and to the extent that this is so, the political liberties are indeed subordinate to the other free- doms that, so to say, define the intrinsic good of the passengers. Admitting these assumptions, plural voting may be perfectly just.20

David Estlund disagrees. He argues that even if we concede the value of educated plural voting and educated government (for some level of education), and even if we concede good will on the part of the educated classes (they act consciously for the benefit of all the citizens), there is still a reasonable objection that the educated classes carry with them dispro- portionately certain privileges of social or economic class, race, or gender, which countervail the epistemic advantages of their education (and might influence their decisions). Estlund argues that even if we correct for any known biases in the educated and governing class (e.g., demographic disproportions in race, gender, religion, etc.), there may still be unknown “latent” or “conjectural features” (unknown, presumably, even to the governing class itself — since good will is conceded) of such dispropor- tions that presumably could influence unfairly the decisions of ruling. Estlund’s essential point is that if someone made this objection, it would not be unreasonable, and so a constitutional or legislative enactment of plural voting would not be acceptable to him or her; it would violate the

19 John Stuart Mill, Essays on Politics and Society, ed. J. M. Robson (Toronto: University of Toronto Press, 1977), 466–81.

20 Rawls, A Theory of Justice, 205. The “ship of state” analogy, of course, was made famous by Plato in the Republic, VI.488–89. For a recent complete analysis of the analogy, see David Keyt, “Plato and the Ship of State,” in Santas, ed., The Blackwell Guide to Plato’s Republic, 188–213.

84 GERASIMOS SANTAS

(presumably essentially democratic) idea that a person “has a right to be ruled on grounds that are acceptable to her.” 21

I am unsure what to say about this objection. It would seem that any governing or ruling group, no matter how corrected for all known unrep- resentative features, may have unknown latent or conjectural features that influence its decisions unfairly. Thus, Estlund’s objection seems to be an equally reasonable objection to all forms of governance, if it is reason- able to begin with. (The case seems to me similar to the construction of samples from which we extrapolate to whole populations; all we can do is exercise due diligence to make sure that known features of the sample are random or representative of the whole population. But no matter how much correcting we do, we can never be sure that there is not some hidden feature that makes the sample unrepresentative; the objection of unknown latent or conjectural unrepresentative features seems to apply to all samples.) In any case, there are other devices which we use that may counter such hidden unfairness, such as veils of ignorance of various kinds (blind trusts, blind refereeing, blind clinical trials), financial disclo- sure statements, and conflict-of-interest laws. So I am inclined to lay Estlund’s objection aside.

A second objection to plural voting points out that political decisions often involve war and peace and have human as well as other costs. Why should a mother of two sons of military age, one may fairly ask, have fewer votes than a single man with a Ph.D.? 22 Perhaps she should not, and others besides the better educated should have more than one vote. Plural voting itself does not exclude unequal weighing of votes for other groups besides the educated. It is worth noting, though, that the weighing of votes in favor of the educated is not based on representation of their interests, but on the idea that it promotes better governing for all — a common interest.

In any case, we can perhaps agree that to accept scholocracy is to lean in Plato’s direction in a significant way; and that scholocracy is still dem- ocratic if every citizen has at least one vote, and if it is acceptable to those with the lesser votes.

V. Plato’s Criticisms of Democratic Freedoms

Plato’s criticisms of democratic freedoms are not easy to find or under- stand well. As far as I can tell, there are at least three main such criticisms in the Republic: Plato’s criticism of the psychic freedom of the democratic man;23 his criticism of free choice of profession; and his advocacy of the

21 Estlund, “Why Not Epistocracy?” 60–61. 22 This objection was brought to my attention by David Keyt. 23 For Plato’s criticism of the psychic freedom of the democratic man, see Santas, Goodness

and Justice, 138–49.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 85

censorship of poetry.24 I confine myself here to his criticism of free choice of occupation, especially free choice of going into politics. The criticism is indirect, because it is clear that there is no free choice of occupation in the completely good city, but we have to hunt for Plato’s reasons.

That there is no such free choice of occupation in the completely good city becomes clear when Socrates proposes that a principle he and his interlocutors used all along is what (social) justice is: “that each one man must perform one social service in the city for which his nature was best adapted” (Rep. IV.433a5–6). But Socrates thinks he has to give reasons for this identification — that this is social justice and not some other thing. His last argument for this is the most relevant here. If a cobbler undertook the work of a carpenter or a carpenter the work of a cobbler, or one of them did both (the two chief violations of the principle), then this would not injure the city a great deal. “But when I fancy [that] one who is by nature an artisan or some kind of money maker, tempted and incited by wealth or command of votes or bodily strength or some similar advantage, tries to enter into the class of the soldiers, or one of the soldiers into the class of the counselors and guardians, for which he is not fitted, . . . or the same man undertakes all these functions at once . . . then this kind of substi- tution or multitasking is the ruin of the city. . . . [T]his is the greatest harm to the city and the thing that most works it harm . . . and what works the greatest harm to the city is injustice” (Rep. IV.434a9 –c4).

This makes it clear that doing what one wants to do (as one’s occupa- tion) is not permitted if one wants to do anything for which one is not best suited by nature and education — and the same holds true if one wants to engage in several occupations at once, or wants to move from an occu- pation one is best suited for to an occupation for which one is less suited. All these things are prohibited by the principle of social justice, which remarkably in our passages is applied to “child, woman, slave, free, arti- san, ruler and ruled” (Rep. IV.433d2–4).

But why do such violations bring the greatest harm to the city? No reason is given in the present passages. We have to go all the way back to the passages where Socrates first proposed the principle of division of labor by talent and education, to see his reasons for it: division of labor (rather than each one doing all necessary things to satisfy all of his or her needs) makes production of food, shelter, and clothing easier; people are born with different natures suited for different occupations; people are better at doing one thing (occupation) well rather than several at once; and doing well at an occupation requires education and time free from other concerns. As Socrates sums it up: “The result, then, is that more things are produced, and better and more easily when one man performs one task according to his nature, at the right moment and at leisure from

24 There may be other restrictions of freedoms in the Republic — for example, the deceptive lottery by which men and women are mated for eugenic purposes.

86 GERASIMOS SANTAS

other occupations” (Rep. II.370c3–5). Notice that Socrates is not saying quite generally that one cannot do several “things” at once or succes- sively; he is talking about the division of labor along occupational lines; and occupations, unlike smaller tasks, require education and considerable time. He is saying that one cannot perform several occupations as well as one; and he is also saying that, given the “natural lottery” assumption (that nature distributes talents and abilities unequally), one cannot per- form as well in an occupation for which one is not suited by nature as in an occupation for which one is. When the model is expanded beyond provisioning to defending and ruling the city, the same reasons work as well. This is why violations of this principle are harmful to the city.

It is difficult to dispute Socrates’ reasons in a major way. The division of labor enhances productivity enormously.25 When the division is along occupational lines (rather than the fine division of labor in an assembly line of a modern factory), occupations require considerable education, training, and a considerable part of one’s lifetime. One can dispute the “natural lottery” assumption to some extent, but when it is applied only to the three major social functions (providing, defending, and ruling), it may be mostly true. Can we claim that all human beings can be equally good at provisioning their city, defending it, and ruling it, if they are given similar educations? Perhaps only an extreme behaviorist would make that claim.

And why do interchanges of the three functions or multitasking cause the greatest harm to the city? Socrates does not say explicitly. The most likely reason is that the principle of organization (division of social labors by talent and education) is foundational for the other virtues — or is at least a necessary condition for them. If, for example, those best suited to defending the city were ruling the city, they would not be doing it as well, and it would be difficult if not impossible for them to acquire the virtue of wisdom — they would have no gold or not enough gold in their souls. Considerable mismatching of talents, educations, and occupations might well result in no one doing his job well — perhaps in gross incompetence.

Socrates’ arguments, however, overlook an important possibility. If a society has free choice of occupation, and citizens have the freedom to choose whatever occupation they want, the good results that Socrates claims for his principle of justice might still obtain by and large and in the long run, if information and appropriate educations are available, and incentives are provided for choosing occupations according to one’s abil- ities and education. But to allow for this possibility would be to give up the principle of division of labor by talent as a principle of justice, for this principle requires that occupations are matched to talents as a matter of justice. Democracies can allow for the value of the results of matching occupations to talents, but they do not require such matching as a matter

25 See Adam Smith, The Wealth of Nations (New York: The Modern Library, 1937), 3–22.

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 87

of justice; if some of their citizens pursue a career other than what they are best at, they might be regarded as foolish, irrational, or inefficient, but not as unjust. Plato could not allow for this possibility — free choice of occu- pation even with good results — without giving up his theory of social justice.

But how can he argue against this possibility? And why do advocates of democracy insist on free choice of occupation?

Plato might argue that if his principle of justice is adopted for the city, those officials charged with assigning occupations to citizens would be more competent to make such choices than the citizens themselves; and if this is so, then in his city the good results are more likely to obtain (and obtain more uniformly) than in a city with free choice of profession. Provisioning the city, defending it, and ruling it would be better done in Plato’s city than in a society with free choice of these professions. But this argument is rather opaque, because in fact Plato does not say who would be making such choices, what their education would be, and why they are more likely to make correct choices than educated citizens themselves. Moreover, if we look forward to modern nation-states, the enormity of populations makes this argument dubi- ous: in effect, Plato would have a command system for choice of pro- fession, and in economics at least — in provisioning the city — such systems fare poorly. An alternative system of free choice of profession, with information, appropriate educations, and incentives for choosing the career one is best suited for, might do as well.

This might seem like a dispute about the instrumental value of a com- mand system of choice of profession versus a social structure that allows free choice of profession by individuals — the value depending on how well each promotes the same good end of matching talents to careers. But unlike the previous cases we considered — where Plato could accept revi- sions in his empirical or metaphysical or epistemological assumptions and still maintain the essence of his theory of justice — here the stakes are too high for both Plato and the advocates of democracy. We have seen that free choice of profession is incompatible with Plato’s principle of social justice, and this is too essential a change for him to accept. At the same time, the advocates of democracy want to argue that free choice has value on its own and that this tips the scales decisively.26 Indeed, Rawls is explicitly not content to leave the fate of liberty to standard empirical assumptions as utilitarian or teleological theories of justice do; rather, he embeds maximum equal liberty and the priority of liberty into his first principles. Here the distinction between justice and empirical assump- tions finds its limits; the dispute cannot be resolved by modifying — hopefully in the direction of truth — empirical assumptions used in the construction of the theory of justice or in its applications.

26 Rawls makes this argument in A Theory of Justice, 73, 184–85, 205–6.

88 GERASIMOS SANTAS

VI. Conclusion

We have discussed Plato’s antidemocratic theory of social justice by distinguishing between the abstract and essential parts of his theory and the empirical or other assumptions he uses in applying that theory. We have argued that his application of the abstract theories may have con- tained empirical mistakes, and it may have been burdened too much with a prolific metaphysics and a demanding epistemology. Accordingly, we attempted to look at Plato’s theory of social justice in imaginary isolation from any discernible empirical mistakes and from his metaphysics and epistemology. When we did that, we found that some of his proposals and criticisms of democracy are well worth our attention, especially in the case of governing. His attempt to separate ruling and wealth and to establish economic floors and ceilings for his ideal city seems especially instructive in view of problems in these areas that ancient and modern democracies have experienced. Isolating and looking at the Republic’s theory of social justice apart from Plato’s epistemology and metaphysics may be more problematic, though our efforts in this direction were not intended as a rewriting of the Republic but only as a thought experiment. Here we found Plato’s insistence that superior wisdom and education is the central virtue of rulers instructive; and in this respect some modern defenders of democratic justice, such as John Stuart Mill and John Rawls, have leaned somewhat in Plato’s direction. Finally, we argued that Plato’s criticism of democratic free choice of occupation is less persuasive than some of his other criticisms of democracy.

Philosophy, University of California, Irvine

PLATO’S CRITICISMS OF DEMOCRACY IN THE REPUBLIC 89

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.