Plato against the vote: the Republic's case for rule by those who know

The Republic makes a claim no politician alive could survive making out loud. Political authority belongs to those who understand what is good for a city; most people do not understand it and cannot be brought to. Socrates says so at 473c–d, in Benjamin Jowett's rendering: "Until philosophers are kings ... and political greatness and wisdom meet in one ... cities will never have rest from their evils."

Plato is not grumbling about a bad election result. He is saying that competence to govern is a rare achievement, distributed roughly as a talent for higher mathematics is, and that organising a state on any other principle is a category mistake. Every modern instinct about equal citizenship is offended by this. The offence is not a misunderstanding: Plato meant it.

His reasons were not abstract. Born around 429 BC into a family close to the centre of Athenian public life, he watched his city lose the Peloponnesian War, watched the Thirty take power in 404 with two of his own relatives among them, and watched the restored democracy execute his teacher in 399. The Seventh Letter is bitter about both: the Thirty, he writes, "made the former government seem by comparison something precious as gold", and then the democracy that replaced them killed the one man who had refused to collaborate with them. We have covered the allegory of the cave elsewhere. What follows is the political argument it was written to serve.

The argument opens with a ship whose crew cannot navigate#

The image in Book VI, at 488a–489a, is the sharpest thing Plato wrote about democratic politics. Picture a ship whose owner is "taller and stronger than any of the crew, but he is a little deaf and has a similar infirmity in sight, and his knowledge of navigation is not much better". The sailors quarrel over the helm; "every one is of opinion that he has a right to steer, though he has never learned the art of navigation". They flatter the owner, drug him, take the ship, and call anyone who actually studies the winds and the stars "a prater, a star-gazer, a good-for-nothing". The philosopher's uselessness, Plato is saying, is a fact about the ship.

Underneath the image lies an older argument, from Book I, where Socrates presses Thrasymachus on what a craft is: the physician in the strict sense pursues the patient's advantage, not his own (341c–343a). Assembled, the case is strong. There are better and worse answers to political questions; some people are better placed to find them; wherever a practice has a truth to hit, we hand the decision to whoever can hit it. Nobody proposes a referendum on the management of sepsis. The Athenians had already made one exception of exactly this shape: most magistracies were filled by lot from 487–486 BC, but the ten stratēgoi, the generals, were elected, on the ground that command in war is a skill and a bad general kills you. Plato notices the exception and asks why it stops at the army.

The analogy fails at the joint where interests are decided#

It fails in two places. In medicine the patient sets the goal and the doctor supplies the means. A seventy-year-old with a slow cancer may reasonably prefer six comfortable months to eighteen unpleasant ones, and no oncological expertise settles that for her. Politics is largely goal-setting. There is no technical answer to whether a percentage point of growth is worth trading for cleaner air, because the question is not about facts alone.

Plato's reply is the load-bearing part of his metaphysics. He denies the disanalogy, because he thinks the good is one determinate thing about which there is knowledge, and that the many are wrong about their own interests as a sick man is wrong about what he wants to eat. If that holds, politics really is a craft. If it does not, the ship image comes apart at the first joint and every later move in the Republic is unsupported.

The second failure is administrative. Medicine has a public, contestable apparatus for certifying expertise — examinations, licensing bodies, journals, malpractice courts. In politics the credential is itself in dispute, and Plato's answer, a thirty-year training programme with repeated selection, produces a body that certifies its own successors.

Estlund's reply: being right and being in charge are two different things#

Modern political philosophy has a name for this: epistocracy, rule by the knowledgeable, a term coined by the Brown philosopher David Estlund in an essay called "Why Not Epistocracy?" (2003). Estlund is the useful opponent because he concedes so much. Even granting that some people "know better what should be done than others", he writes, "it simply doesn't follow from their expertise that they have authority over us." He calls the slide the expert/boss fallacy.

What blocks the inference is a condition on legitimacy Estlund takes from the Rawlsian tradition — the qualified acceptability requirement, summarised by David Copp in Ethics in 2011 as the claim that "a state is legitimate only if its coercive enforcement of law can be justified on a basis that is acceptable to all possible qualified points of view". Expertise fails that test not because experts are ignorant but because their standing as experts is itself reasonably disputable. Estlund adds a blunter worry, the demographic objection: any workable criterion of political expertise will select a narrow group whose biases would damage the outcomes epistocracy exists to secure.

None of this refutes Plato's epistemology. It changes the subject from who is right to who may coerce — a legitimate move, and a change of subject all the same. Estlund's own view, epistemic proceduralism, concedes half of Plato's point: democratic authority depends partly on democracy's tendency to produce good decisions. Jason Brennan pushes the other way in Against Democracy (Princeton, 2016).

The Republic is a book about schooling with a constitution attached#

Strip out the curriculum and the Republic would be a pamphlet. Books II and III are given over to what children may be told: gods must not be shown lying, quarrelling or grieving, Homer is cut, musical modes are ranked by the characters they form. Book X generalises the policy, invoking "an ancient quarrel between philosophy and poetry" (607b) and expelling imitative poetry because it "awakens and nourishes and strengthens the feelings and impairs the reason" — though Socrates offers, almost sportingly, to readmit her if a defence can be made.

Book VII supplies the higher programme: arithmetic, plane geometry, solid geometry (which Socrates concedes is an underdeveloped subject), astronomy and harmonics, ten years of it from twenty to thirty; five years of dialectic; fifteen years in subordinate office; then rule, at fifty (537b–540a). Nobody in Kallipolis governs before their sixth decade, and everyone who governs has spent fifteen years taking orders. Whatever else the philosopher-king is, he is no licence for clever young men in a hurry.

The organising claim is that a city is the soul written in larger letters (368e–369a, 434d–435a) — doubted since Bernard Williams's 1973 paper on the analogy as either a covert metaphysical thesis or a heuristic doing more work than it can bear. The political upshot survives the doubt. A polity is produced by what its citizens are trained to enjoy; get the loves wrong and no arrangement of assemblies and courts will save you. Hence the legislation about lullabies, and hence the fact that the question of who decides what may be taught keeps returning, as in the dispute now before the courts in Texas over whether a university may stop a professor teaching the Symposium.

The quarrel with the sophists was about whether speech answers to anything#

Protagoras of Abdera, who lived from roughly 490 to 420 BC, left a sentence that Plato quotes at Theaetetus 152a and never stopped arguing with: "Of all things the measure is man, of the things that are, that they are, and of things that are not, that they are not." What he meant by it is still disputed. Plato heard it as making appearance sovereign: if how things seem to me settles how they are for me, the craft of making things seem is the only craft there is. Gorgias of Leontini supplied the technique. In the Encomium of Helen he calls speech "a powerful lord, which with the smallest and most invisible body accomplishes the most divine works", and likens its action on the soul to that of drugs on the body — persuasion indifferent to truth by design, and rather proud of it.

The charge Plato lays in the Gorgias is precise. At 464b–465c Socrates distinguishes two genuine arts of the body, gymnastic and medicine, and two of the soul, legislation and justice; each has a counterfeit aimed at gratification instead of good. The proportion at 465c: as self-adornment is to gymnastic, so is sophistry to legislation, and as cookery is to medicine, so is rhetoric to justice. The pastry-cook knows what will please, the physician what will help — and "if the physician and the cook had to enter into a competition in which children were the judges ... the physician would be starved to death".

That is a polemic, and we know the sophists largely through the man attacking them; they taught a practical skill to citizens who had to argue their own cases in court, and a market for that is not sinister. What Plato fastened on was narrower and harder to dismiss: the skill can be sold with no requirement that the seller know whether what he says is so. Whether public speech is organised around hitting the truth or around producing an effect is unsettled in every generation, and that is where the Gorgias reaches the present without being dragged there.

The noble lie is where Plato's own argument turns on him#

At 414b–415d Socrates proposes what Jowett translates as "one royal lie which may deceive the rulers, if that be possible, and at any rate the rest of the city" — gennaion pseudos in the Greek, softened by F. M. Cornford and rendered flatly as "lie" by Allan Bloom. Citizens are to be told that their upbringing was a dream and that they were in fact formed beneath the earth, which is their mother; and that the god who made them mixed gold into those fit to rule, silver into the auxiliaries, iron and bronze into farmers and craftsmen, with the rider that children may move between classes when their metal differs from their parents'. Socrates proposes to try it on the rulers first, and is visibly uncomfortable throughout.

Reported accurately it is less lurid than its reputation: no racial doctrine, no permanent caste, and a mobility clause an Athenian aristocrat would have found radical. Reported accurately the objection is still fatal. Plato's whole case is that authority should follow knowledge, and the noble lie rests the cohesion of the best city on a proposition the rulers know to be false and the ruled must not examine. A regime that legislates a founding myth before it can begin has conceded that its citizens cannot be governed by reason alone — that the philosophers' title to rule cannot be made good to the people it governs. Estlund's acceptability requirement is the refusal to concede precisely that. Plato concedes it in Book III and never takes it back.

Popper's charge is too blunt, and the replies are not as clean as they sound#

Karl Popper wrote The Open Society and Its Enemies in New Zealand during the Second World War; Routledge published it in London in 1945, the first of its two volumes titled The Spell of Plato. The charge has three strands: that Plato is a historicist for whom change is decay; that he wants a closed society, a tribal collectivism set against the individualism of democratic Athens; and that Kallipolis is totalitarian in structure, with fixed classes, controlled breeding, censorship and an unaccountable ruling caste.

Three replies are standard. The first is that the ideal city is a paradigm rather than a programme, on the strength of 592a–b, where Socrates says that "in heaven ... there is laid up a pattern of it ... which he who desires may behold, and beholding, may set his own house in order". The difficulty is that he also insists the city could come into being (450c–d, 473c, 499b–d): if Plato did not mean it as a goal worth striving for, something other than Socrates' explicit statements has to show it. The second reply is anachronism, pressed by Ronald B. Levinson in In Defense of Plato (Harvard, 1953), a 674-page rejoinder arguing that Popper read fourth-century Athens through the categories of 1943.

The third and most popular is that the late dialogues show a different Plato. The Statesman holds that a genuine expert would rule better without law, since law resembles a self-willed man who admits no questioning and cannot prescribe for every case (294a–c), then concedes from 297e onwards that no such expert exists, so the rule of law is the best available second. The Laws, unfinished at Plato's death around 347 BC, designs Magnesia: 5,040 households, private property, a constitution mixing monarchic and democratic elements, elected officials audited by scrutineers, and magistrates described at 715d as servants of the law.

That story is tidier than the evidence. Magnesia also has a Nocturnal Council meeting before dawn to study theology and supplement the laws, a body of philosopher-rulers under another name; and Myles Burnyeat argued in 1992 that the Laws is in some respects a less practicable ideal than the Republic, since it demands virtue of every citizen. Popper's critics have had to concede that the censorship, the state myth and the controlled breeding really are in the text. What stays in dispute is the inference from them to the totalitarian state.

Where this leaves a reader who has to vote next year#

The honest answer to Plato is not that experts are sometimes wrong; everybody knows that, Plato included. The better answer is the one Elizabeth Anderson set out in Episteme in 2006, in a paper testing three versions of the epistemic case for democracy — the Condorcet jury theorem, the diversity-trumps-ability result, and John Dewey's experimentalism, which she prefers. Democratic institutions, she argues, pool information held by people who are experts in nothing. Her example is community forestry in South Asia, where councils that excluded women turned out to be worse at forestry, because the women gathered the fuel and knew the trees. Set that against the ship image and it does damage: the crew are not merely quarrelsome, they know things the pilot cannot see from the bridge.

Where the technical content is high the difficulty is sharper; there is now a literature on making scientific advice accountable without making it popular, covered here in our report on Zeynep Pamuk's proposals. The risk on the other side needs naming without euphemism. Regimes founded on the principle that those who know should rule have a poor record, and not because the principle is silly. The credential can always be claimed, and no mechanism has been found for checking it that does not require the consent of people the claimant has just declared unqualified. None of the twentieth century's vanguards was run by philosophers. Every one of them said it was run by people who understood.

What Plato saw, and what most modern defences of democracy concede, is that a polity is made by what its citizens are taught, and that a public formed by flattery will choose flatterers. Where he gave up — on the capacity of ordinary people to be educated into judgement — is where he should be resisted, on the evidence rather than on principle. A democracy that took the Republic seriously would spend its money on schools and on the conditions of honest public argument. A duller conclusion than rule by the wise, and harder to arrange.

---

Plato was born around 429 BC and died around 347 BC. The cover image is a Roman copy of a fourth-century BC portrait attributed to Silanion, in the Musei Capitolini, Rome. Translations from Plato are Benjamin Jowett's unless otherwise stated.