How can virtue be in a mean, if as far as the best and the good are concerned it is an extremity? (2017 Q3)

Related: notes · essay plans · Write a critical commentary on Aristotle’s Doctrine of the Mean. (essay)

Virtue can be in a mean even though it is an extremity as far as the good is concerned, because the axes along which it is the mean are non-evaluative. The virtuous state is a stable disposition which generates the appropriate actions and feelings in each situation, where those actions and feelings are intermediate between less-appropriate responses in non-evaluative respects. In this essay, I will set out Aristotle’s Doctrine of the Mean, and clarify what is meant by a mean. Then, I’ll demonstrate that Aristotle’s theory is not committed to any contradiction in saying that virtue is at once a mean and an extremity of the good. Finally, I’ll argue that the titular question, although missing the mark in its precise objection, does point towards a more substantial worry for Aristotle’s doctrine – that it is either trivial or incoherent.

Aristotle introduces his Doctrine of the Mean (DoM) in Book II of the Nicomachean Ethics, and he uses the concept of the mean throughout his discussion of virtues of character in Books III-V, as the theoretical machinery behind what it is to be virtuous. Given that the central object of the Ethics is to provide an account of the good life, and Aristotle establishes in I.7 that this is the life of virtuous activity, it becomes clear when Aristotle declares that virtue is a “state (hexis) which decides, consisting in a mean (meson)” that we will need to properly understand the mean before we can fully grasp Aristotle’s ethical theory.

So, what does Aristotle have to say about the mean? He tells us that virtue – and by extension the mean – is located between two vices (“one of excess and one of deficiency”), and also that it concerns both feelings and actions. Following Young (2006), we can thus decompose the DoM into two claims:

(1) [Location thesis]: The mean for a particular character trait is a virtuous state located between two vices for that same trait, one of excess and one of deficiency.

(2) [Intermediacy thesis]: The mean for a particular character trait is a disposition towards responding with actions and feelings that are intermediate between those generated by the corresponding vices for that trait.

Crucially, in either case the mean is restricted to being that which is intermediate in a state of character, action, or feeling. The mean need not be intermediate with respect to every conceivable dimension; that would be absurd. Indeed, for some dimensions the “mean” would clearly be at an extreme – for example of torturing – and Aristotle pre-empts this particular objection by noting that some traits and actions do not admit of a mean as “their name automatically entails baseness”.

The poser of the original objection might be unsatisfied by this defence, and argue that, unlike torture, goodness is a perfectly reasonable dimension along which to evaluate actions. But this misses the point about what the DoM is meant to capture. Aristotle’s goal with the DoM is to demonstrate that the evaluative, moral features of an action (or character trait, or feeling) can be analysed in terms of how that action relates to the mean, an object defined along descriptive axes like “enjoyment of bodily pleasures”. There is no contradiction in saying that virtue consists in a mean (i.e., involves intermediate actions and feelings), while also maintaining that the outcomes it realises are evaluatively optimal. The question’s complaint is a bit like saying: “How can Goldilocks’s porridge be right in the middle, if as far as her enjoyment is concerned it is an extremity?” The answer, of course, is that it was in the middle with respect to temperature and sweetness, and this is an entirely natural way to understand intermediacy. Hursthouse’s characterisation of reaching the mean as “hitting the bullseye” helps us see this more vividly: the virtuous agent is skilled at performing the difficult task of getting an action right in all the ways it must be, in order for it to go well, just like the darts player manages to throw their dart with just the right force, and in the right direction (and so on), in order to attain the highest possible score.

However, although this particular objection to the DoM is mistaken, it does point towards a more substantial worry for Aristotle’s theory – that it is either trivial or incoherent. We argued above that the mean in which virtue consists is an intermediate state with respect to actions, feelings, and character traits. So far, we have said nothing of where this mean lies – only that it exists somewhere. The trouble is that for many apparent virtues, there don’t really seem to be the adjacent vices that our theory demands. Aristotle’s discussion of justice in Book V, for example, struggles to identify the vices which are meant to flank the virtue of special justice (not of suffering and doing injustice, for the former is involuntary; not of taking too much and taking too little, for the latter is admirable and what the decent agent does). Similarly, Aristotle concedes that the man who viciously enjoys bodily pleasures too little “seems hardly human and has no name” – yet Aristotle still asserts that there is a virtue about temperance. So the theory appears incoherent, and fails to capture the range of virtues that Aristotle thinks it ought to.

One might try to salvage Aristotle’s theory by weakening the claim made about the nature of virtue, but this descends into triviality and runs into difficulties close to those raised by the titular question. In particular, we could interpret the DoM as merely claiming that virtue consists in a mean insofar as it is neither too much nor too little. But we are now smuggling in evaluative content in such a way that renders the doctrine meaningless and uninteresting: virtue is, as the question states, the optimal state with regard to goodness – so of course it is neither too much nor too little; that is precisely what it means to be the optimum.

In this essay, I discussed the importance of the DoM and distinguished between two interpretations of it. I then argued that on each interpretation virtue can be a mean despite being the extreme of goodness, in that the mean is about actions, feelings, and states of character. Finally, I showed that even if this objection to DoM can be dissolved, the theory still suffers from a dilemma between triviality or incoherence.

I applaud your desire to explore less well-known parts of Aristotle’s theory, and this is an excellent learning tool. I encourage you to keep revising this way.

With respect to the content: I agree, broadly speaking, with your answer, but I found the second half of the essay unconvincing and hard to follow. I suspect you’re doing too much. Try to focus solely on developing your defense of Aristotle’s position.

Examiners’ report:

Many candidates were able to produce reasonable discussions of Aristotle’s account of the mean. For instance, many candidates were able to explain that being in a mean doesn’t imply always choosing the option that is mid-way between two extremes. Disappointingly few candidates really focussed on the question asked, which was about how these remarks about virtue’s being in a mean can be compatible with the sense in which virtue is an extremity. Some of the better candidates connected the idea of virtue’s being in an extremity with the difficulty of being virtuous, and hence with the idea that being virtuous is a great achievement.

Post-Claude & examiner report notes