Re-examining 'The Mean': Aristotle, Confucius, and the Nature of Virtue — Epoche B2
There is no Confucian doctrine of the mean for Aristotle's to be compared with A comparison built on two sentences Comparative ethics has a favourite pairing. Aristotle teaches that virtue lies in a mean between excess and deficiency [1] ; the Confucians teach zhongyong [2] , the doctrine of the mean; therefore two great traditions independently discovered the same thing, and the interesting work lies in describing how their versions differ. This research note argues that the pairing does not survive an inspection of the texts, and that the interesting work is elsewhere. The Analects contains two passages that could support the comparison [3] , and neither states a doctrine. The text that does develop zhong at length, the Zhongyong , is not Confucius's and does not mean by zhong an intermediate point between two extremes. What that text does contain, when read for what it says, turns out to parallel a different and more specific part of Aristotle's account than the one the comparison is usually made with. The conclusion is not that the traditions have nothing in common. It is that the received comparison identifies the wrong thing as the shared element, and gets the Confucian half of it from a sentence that means something else. Aristotle's doctrine, stated precisely The claim being compared has more structure than 'avoid extremes'. At Nicomachean Ethics II.6, 1107a1–2, Aristotle says that virtue is a state concerned with choice, lying in a mean relative to us, determined by reason and by the reason by which the person of practical wisdom would determine it. Four things are packed in. First, the mean is a property of a hexis , a state of character, not of an action or a feeling considered by itself. Second, every moral virtue is flanked by two named vices, one of excess and one of deficiency, and Aristotle supplies the pairs: courage between rashness and cowardice, generosity between prodigality and meanness, good temper between irascibility and inirascibility. Third, the mean is relative to us, so the intermediate amount varies with agent and circumstance and is not an arithmetical midpoint. Fourth, some things admit no mean at all — Aristotle names spite, shamelessness, envy, adultery, theft and murder at 1107a9–17, and says they are bad in themselves, not by excess or deficiency. The second of these is the load-bearing element and it is the one usually dropped in comparison. The doctrine is not advice to be moderate. It is a structural claim about every ethical virtue: that each occupies a position on a single scale flanked by two determinate failures, so that identifying a virtue involves identifying its two vices. What the Analects contains Now the Confucian side. The compound zhongyong occurs once in the Analects , at 6.29, where the Master says that zhongyong as a virtue is supreme, and that it has long been rare among the people. That is the whole of it. There is no analysis, no pair of flanking vices, and no indication of what the two characters mean when joined. The other passage always cited is 11.16. Zigong asks which of two disciples is better; the Master says one goes beyond and the other falls short; Zigong asks whether going beyond is therefore better, and the Master answers that going beyond is as bad as falling short. This is a remark about two people, made in reply to a question about ranking them. It is not offered as a general principle, no virtue is named, and nothing is said about a scale. Two passages, one of which is a single evaluative sentence and the other an answer about two named disciples, cannot bear a doctrine. Compare what the Aristotelian material consists of: a definition, a derivation from the theory of the soul, a table of virtues with their paired vices, an explanation of why the mean is relative to us, and an exception clause. The asymmetry is not a matter of degree. What zhong means in the Zhongyong The obvious reply is that the doctrine is in the Zhongyong , the chapter of the Liji that circulates under the title 'The Doctrine of the Mean'. Two things must be said about it. The first is authorship: the text is traditionally ascribed to Confucius's grandson Zisi and belongs to the Warring States period, not to Confucius. Attributing its doctrine to Confucius is the same kind of error as attributing the Enneads to Plato. The second is more important, because it concerns what the text says. Its opening chapter defines the term. Before pleasure, anger, sorrow and joy have arisen, that is called zhong ; when they arise and all hit the due measure, that is called he , harmony. So zhong names a state in which the feelings are not yet aroused at all. That is not an intermediate quantity of a feeling. It is the absence of arousal, understood as a centred condition from which a response can issue. A doctrine of the mean in Aristotle's sense would have to say that the right amount of anger lies between too much and too little; the Zhongyong says that before any anger there is zhong , and that anger arising at the due measure is he . The English title has done real damage here. 'The Doctrine of the Mean' was Legge's rendering, and it invites the reading it then appears to confirm. What actually parallels what Repairing the comparison is more interesting than abandoning it, because there is a genuine parallel once the terms are lined up correctly — and it is not with the doctrine of the mean. The Zhongyong 's phrase for feelings arising at the due measure, zhongjie , hitting the joint or the appropriate degree, is a claim about the fit between a response and its occasion. The second chapter sharpens it: the exemplary person's zhongyong consists in shizhong , hitting the centre according to the time. Timeliness is the operative idea. What is called for now is not what was called for yesterday, and getting it right is a matter of reading the occasion. That is a real counterpart to something in Aristotle — but to the clause 'relative to us', and to the role of p