A puzzle about knowledge and virtue
What role, according to Aristotle, does knowledge play in the exercise of the virtues of character? This is a simple question to ask, but a difficult one to answer simply. In order to answer this question, we have to tackle an elliptical passage in Nicomachean Ethics II.4, which is part of Aristotle’s solution to a different puzzle about virtuous action.
That puzzle is (very roughly) this: if we have to do the virtuous thing in order to become virtuous, but we have to be virtuous in order to do the virtuous thing, how do we ever get started? Aristotle notes that we talk about actions and corresponding states of soul in two different ways, (let’s label them) logically – the actions accord with a certain state – and causally – the actions are the product of that state. So, too, with virtue: there are actions that are courageous because they are what a courageous person would do and actions that are courageous because they flow from the courageous person’s courage. Puzzle solved: when we’re on the way to virtue, we have to do the first kind of action so that when we acquire virtue, we are capable of the second.
Where the terrain gets treacherous is when Aristotle decides to try to characterize the state of the soul that differentiates the latter from the former case:
[Actions] that come about in accordance with the virtues are not done justly or temperately just because they are a certain way, but only provided that the agent is also in a certain state when he acts: first, provided he has knowledge (eidōs), second provided he exercises choice and chooses things because of what they are, and third, provided he acts being himself stable and unchanging. These [conditions] are not taken into account, by contrast, with regard to possessing any of the crafts, except for knowing (to eidenai). But in connection with [possessing] the virtues, knowledge is of no or little significance, while the other [conditions], which arise from doing what is just and temperate many times, have no small influence but rather are all-important (1105a28-34).
Here’s how I’d frame the question I have about this passage given my topic: what kind of knowledge is important enough that it merits being one of the three conditions that differentiates the state of those performing actions virtuously from those merely doing what is virtuous but is not that important compared to the other two conditions?
The standard answer is: knowledge of what you’re doing, i.e., the kind of knowledge you need for an action to be voluntary (cf. EN III.1, 1111a2-6). But I don’t think this answer can work.
For one, voluntariness just isn’t in play here. The question is not what makes something a virtuous action as opposed to any old action, but rather what characterizes someone doing something virtuously as opposed to someone doing something that happens to be virtuous on their way to possessing virtue. And it’s hard to see why the latter wouldn’t always be voluntary, too. Why then would Aristotle mention voluntariness here?
Second, voluntariness seems to belong just as much to craft action as to virtuous action, and to make each no more and no less what it is than the other. Aristotle’s remark that knowledge is relevant for craft, but relatively unimportant for virtuous action demands that we understand something other than knowledge of what one is doing.
What, then, does he mean to pick out with ‘knowledge’?
 I say standard because this is the view of, at minimum, Sarah Broadie (‘[T]he sheer ability to know what one is doing’ – 2002, p. 301), John Burnet (‘It is essential that he should know what he is doing’, citing EN III.1 – 1900, ad loc.), C.C.W. Taylor (‘It is clearly a necessary condition of an action’s instantiating a virtue that the agent should know what he or she is doing in performing that action’ – 2006, p. 84), and St. Thomas Aquinas (ille qui facit opus virtutis non operetur ex ignorantia vel a casu sed sciat quid faciat – ad loc.), in their commentaries.