Confucius: how a good person is formed and a good society built

Almost everyone has met the name. Few people who have not studied China could say what the man thought, and the few sentences that circulate under his name, about silence and golden ages and a thousand-mile journey, are mostly things he never said. The real figure is both less quotable and more interesting. Confucius was a teacher in a small, quarrelsome state in eastern China two and a half thousand years ago, who asked one question above all others and spent his life turning it over: how does a person become good, and can a decent society be built out of anything other than good people?

That the question sounds unremarkable is part of why he is underrated in the West. The habit here is to measure a philosopher by the systems he builds, and Confucius built none. He left no treatise, proved no theorems, defined almost none of his central terms. What survives is a book of fragments, remembered conversations between a teacher and his students, assembled after his death by people who had not all known him. To read him as the author of a doctrine is to miss him; he is closer to a portrait than to an argument, a record of how one person met particular situations and of the character that showed through.

It is worth saying what this piece will not do. It will not present Confucius as a sage with the answer to modern life, nor reduce him to the caricature of obedience and hierarchy his name often carries. Both the halo and the sneer get in the way. The aim is narrower: to set out what he argued, to separate it from what his successors built on it, and to say where it still presses on questions we have not solved.

The historical man and the legend#

The traditional dates are 551 to 479 BC, in the Spring and Autumn period, when the Zhou kingdom had fractured into rival states and the old ritual order was visibly failing. He was born Kong Qiu in the state of Lu, in what is now Shandong. His own tradition called him Kongzi or Kong Fuzi, 'Master Kong'. The familiar Latin form, Confucius, is the work of Jesuit missionaries who reached China in the sixteenth and seventeenth centuries and rendered Kong Fuzi into Latin for European readers; the man never heard the word.

The biography most people half-remember, the wandering official rebuffed by princes, the editor of the classics, comes largely from the Records of the Grand Historian compiled by Sima Qian around 100 BC, four centuries after Confucius died. The Stanford Encyclopedia of Philosophy is firm that these later sources are unreliable, shaped by legend and by the needs of the writers who produced them. The sober conclusion is that we know rather little about the events of his life with any confidence. We are on firmer ground with the shape of his thought, provided we read the right text in the right way.

The Analects as a made thing#

That text is the Lunyu, known in English as the Analects. It is not a book Confucius wrote. It is a compilation of sayings and short exchanges put together by his disciples and, over several generations, by their disciples in turn, reaching something like its present form only in the Han dynasty, well after his death. In the 1990s the scholars E. Bruce Brooks and A. Taeko Brooks argued that its twenty books are layered, composed by different hands at different times, so that the work records an evolving school rather than a single voice. The lesson holds even if one doubts the details: the Analects is an edited object with a history, to be read through that history rather than past it.

This changes how the book should be used. It is not a scripture to be quoted flatly but a collection in which a teacher gives different answers to the same question depending on who is asking, corrects himself, and sometimes declines to speak at all. The inconsistency is not a flaw to be ironed out. It is the method. Confucius seems to have believed that the right response depends on the person and the moment, and the text preserves that conviction in its very untidiness. Good translations exist in English, by James Legge in the nineteenth century and by D. C. Lau, Roger Ames and Henry Rosemont, and others since; the quotations below follow Legge.

Ren: humanity is something you do with others#

The central term is ren, written with the characters for 'person' and 'two'. It is translated as humaneness, benevolence, goodness, and no single English word holds it. The composition of the character is a clue worth more than a definition: to be fully human takes at least two people. Ren is what a person becomes by developing himself, morally, in and through his dealings with others.

Confucius never pins it down. Asked what ren is, he answers one student that it is to love others, another that it is to master oneself and return to ritual, another with a rule that has echoed ever since. In Legge's rendering: 'What you do not want done to yourself, do not do to others.' The refusal to give a single definition is itself the teaching. Ren is not a concept you grasp and then apply; it is shown in a life, not stated in a formula, and the Analects would rather display it in a hundred small exchanges than trap it in one sentence.

Here the distance from a certain modern individualism is real. The question 'who am I?' is not, for Confucius, answered by looking inward alone. A person is constituted in his relationships, as child, parent, friend, teacher, student, subject and ruler, and these are not roles the self puts on over a prior, naked identity. They are what the self is made of. Some contemporary interpreters, Ames and Rosemont among them, press this so far as to call Confucian ethics a 'role ethics', holding that the Western 'individual' is, in Confucius's world, an abstraction and the real units are relationships. That is a strong reading, and not every scholar accepts it, but it marks where Confucius and the modern West least easily understand each other.

The junzi: nobility you earn#

The word junzi once meant something like 'the ruler's son', a term of birth: a nobleman. In Confucius's hands it is quietly revolutionised into a term of character. The junzi, the 'gentleman' or 'exemplary person', is noble in conduct rather than by blood. This is among the most consequential moves he made, and Europe would not make its own version until the Renaissance humanists, and later the Enlightenment, argued that true nobility is virtue.

The practical edge of the idea is education. Moral excellence, for Confucius, is not an inheritance but an achievement, reached through study, self-discipline and practice, and therefore in principle open to anyone. 'In teaching there should be no distinction of classes,' he is reported to have said, and his own students included poor men. Learning, xue, is the first word of the Analects, and the opening line sets its tone. Legge gives it as: 'Is it not pleasant to learn with a constant perseverance and application?' He also warns against learning that never becomes thought, and thought that never submits to learning: 'Learning without thought is labour lost; thought without learning is perilous.' Education here is not the acquisition of a trade. It is the building of a self.

Li: ritual as the grammar of a shared life#

To a modern ear 'ritual' suggests empty form, the hollow performance of gestures. For Confucius, li is closer to the invisible grammar of social life. Just as speech is impossible without the rules of a language, a common life cannot be carried on without shared forms of conduct, the ways people greet, mourn, eat together, mark an occasion. These are not decoration laid over human relations; they are the medium in which relations are conducted, the daily language by which people tell one another what they are to each other. The American philosopher Herbert Fingarette drew out this idea for Western readers in 1972, comparing ritual acts to what J. L. Austin called 'performative' utterances: a handshake does not represent an agreement, it is one.

But Confucius adds a warning that cuts against any merely ceremonial reading. Form without feeling is worthless. Asked about the root of the rites, he answered, in Legge's version, that 'in festive ceremonies, it is better to be sparing than extravagant', and that 'in the ceremonies of mourning, it is better that there be deep sorrow than a minute attention to observances'. The point of mourning is grief, not a flawless performance. Ritual matters because it carries real regard between people; drained of that regard it is noise. A society in which people no longer know how to address or disagree with one another may be suffering less from a failure of law than from the breakdown of this shared grammar.

Government by virtue, not only by rules#

Confucius's political thought is strange to anyone raised on the modern theory of the state, which tends to locate good government in good institutions. For him, good rules are not enough; the character of the ruler, his de or moral power, is what finally governs. The most famous image is of the ruler who leads by virtue as 'the north polar star, which keeps its place and all the stars turn towards it'. Where a ruler is upright, Confucius thought, people follow without being forced.

He draws the contrast sharply between two kinds of order. 'If the people be led by laws, and uniformity sought to be given them by punishments, they will try to avoid the punishment, but have no sense of shame. If they be led by virtue, and uniformity sought to be given them by the rules of propriety, they will have the sense of shame, and moreover will become good.' Translate this into a present-day argument and it becomes a challenge to liberalism. Should a legal order merely set the outer limits of conduct and stay neutral among ways of life, as modern liberal theory tends to hold, or is the cultivation of good citizens part of the point of politics? Confucius takes the stronger line: politics cannot be separated from character, and even just laws, administered by people without virtue, will produce injustice. Whether that line can survive the diversity of a large modern society, without licensing the state to dictate how its citizens should live, is exactly the question his critics press, and it is a fair one.

Zhengming: when words come loose from things#

One of his most suggestive political ideas is zhengming, usually translated 'the rectification of names'. Asked what he would do first if given charge of government, Confucius answered that he would rectify names. The point is not pedantry about vocabulary. When a ruler is called a ruler but does not act as one, when a father does not father or a minister does not serve, the bond between the name and the reality snaps, and public life unravels from there. Legge's translation of the consequences is bleak and precise: 'If names be not correct, language is not in accordance with the truth of things. If language be not in accordance with the truth of things, affairs cannot be carried on to success', and so on down to a final collapse in which 'the people do not know how to move hand or foot'. An allied saying puts the ideal plainly: good government is present 'when the prince is prince, and the minister is minister; when the father is father, and the son is son'.

This is where a modern reader should hesitate before nodding too quickly. The doctrine can be read two ways. As a demand that those in authority actually do what their titles claim, it is a sharp tool against cant, and George Orwell's analysis of political language is its distant descendant. As a demand that everyone keep to the conduct proper to their fixed station, it can serve as a charter for hierarchy, a way of telling the son and the minister to stay in their place. Which reading Confucius intended is not fully decidable from the text, and the ambiguity is honest to record rather than resolve.

What Confucius did not say#

Much of what passes for 'Confucianism' is the work of successors, and the distinction matters. The systematic doctrine of the Five Relationships, the fixed hierarchy of ruler and subject, father and son, husband and wife, elder and younger, friend and friend, is set out not in the Analects but by Mencius in the fourth century BC, and it is Mencius, not Confucius, who argued at length that human nature is good. Xunzi argued the opposite, that nature is bad and must be reshaped by ritual. The cosmic correlations that later tied Confucian ethics to the movements of heaven belong to the Han thinker Dong Zhongshu, who helped make Confucian learning state orthodoxy. The Analects itself was not even the centre of the curriculum until the Song dynasty, when Zhu Xi placed it among the 'Four Books' that every examination candidate would thereafter memorise. The imperial examination system, the ancestor cult as state machinery, the rigid family patriarchy: these are the achievements, and the distortions, of two thousand years of followers. Confucius is responsible for the seed, not for every tree grown from it, and a reader who blames him for the whole of later Chinese history is making the same mistake as one who credits him with it.

Why he is read again#

Three days ago we marked his traditional birthday; the better memorial is to take the arguments seriously. Where an earlier West waved Confucius away as the prophet of a stagnant order, parts of the present West are tempted to adopt him as a ready-made corrective to individualism, a comfortable Eastern wisdom to set against Western excess. That is its own kind of condescension; he is more useful as an irritant than as a balm.

Set beside the Greeks, the contrast is instructive. Socrates told us to know ourselves, questioning citizens one at a time in the public square, a scene the West still retells when it argues about Plato's cave. Confucius worked in a different key, teaching not by interrogation in the agora but in conversations on the road, at table, in the course of a ritual, and his instruction was less 'know yourself' than 'cultivate yourself, and do it in company'. His sharpest challenge to the present is the claim that character is prior to rules, in ethics and in politics alike. It is a claim worth resisting as much as accepting, and the resisting is the point.

It even reaches the newest arguments. As AI systems take over more of what used to be human judgement, and as philosophers find themselves hired by the laboratories building them, Confucius's question refuses to go away. He would probably not have asked whether a machine can think. He would have asked what using it does to the person who uses it, and whether a society that outsources its judgements has not quietly lost the habit of forming them. His lasting bequest is not a set of answers but the stubborn idea that becoming human is work, that the work is done among other people, and that no institution, and no instrument, can do it for us.