Eudemian Ethics II 6 is meant to introduce Aristotle's discussion of voluntary action in II 7–9. The majority of II 6, however, consists of a somewhat obscure discussion of the ways in which humans, alone among animals, are origins of action (ἀρχαὶ πράξεων). It is not at all clear how that topic is meant to relate to the topic of voluntary action until the following passage, towards the end of the chapter, in which Aristotle relates being the cause (αἴτιος) and origin of action to praiseworthiness and blameworthiness and to the voluntary (1223a9–16):

My focus here is on lines 13–16, the last clause in the first sentence (δῆλον … πράξεων) and the sentence that comes after it (ληπτέον … πράξεων). Ιn both places the inclusion of πράξεων seems odd, at least given how these sentences are usually interpreted. According to the standard interpretation, what Aristotle is saying in the first part is that virtue and vice concern those actions of which one is oneself the origin and cause. If this is right, Aristotle is expressing himself in a rather roundabout way. Far more natural would be something like: δῆλον ὅτι καὶ ἡ ἀρετὴ καὶ ἡ κακία περὶ πράξεων ὧν αὐτὸς αἴτιος καὶ ἀρχή. The oddity of the phrase that Aristotle actually uses is reflected in the way in which translators have rendered the sentence in English. For example, KennyFootnote ³ translates the sentence as follows: ‘It is clear that virtue and vice have to do with matters where the man himself is the responsible source of his actions’ (RackhamFootnote ⁴ and SolomonFootnote ⁵ translate almost identically), and WoodsFootnote ⁶ translates as follows: ‘It is evident that virtue and vice have to do with those things of which a man himself is the cause, a starting point of actions.’

The wording of the second sentence is no less strange. Again, according to the standard interpretation Aristotle is saying there that it must be ascertained what actions one is the cause and origin of. If this was what Aristotle wanted to say, though, it is unclear why he should include πράξεων in the place that he does. Presumably this is why Dodds, according to the apparatus criticus of the Oxford Classical Texts edition, secludes the whole phrase καὶ ἀρχὴ πράξεων and Woods, in his translation, simply leaves out πράξεων (translating ‘So we must determine of which a man himself is the cause, and a starting point’).Footnote ⁷ Other translators simply take ποίων with πράξεων, despite the placement of the latter term (for example, Rackham, Solomon and Kenny).

Of the major English translations, Inwood and Woolf'sFootnote ⁸ recent translation is the most faithful to the text: ‘So it is clear that both virtue and vice are concerned with what one is oneself a cause of, as starting point of action. One must then grasp the sorts of things that one is a cause of as starting point of action.’ The reason that Inwood and Woolf are alone in translating the passage in this way, however, is that this is a very odd way for Aristotle to express himself, at least if the standard interpretation is right. What is it to be the cause of something as a starting point of action? Is it just to be the starting point (and hence the cause) of an action, as the standard interpretation would seem to require? If so, why didn't Aristotle just put it in that, simpler, way?

Now, I do not think that these considerations rule out the standard interpretation, nor do I think that proponents of the standard interpretation must depart from the manuscript text. I do think, however, that they give us at least some reason to prefer an alternative. My aim here is to provide such an interpretation.

The key assumption that I wish to call into question has to do with the role that πράξεων plays in relation to ἀρχή. On the standard interpretation πράξεων serves to delimit the range of items that the ἀρχή at issue is the ἀρχή of. Hence to say that x is an ἀρχὴ πράξεων is to say that x is an ἀρχή and that the things that x is an ἀρχή of are actions. On my alternative reading, to say that x is an ἀρχὴ πράξεων is not to say what things x is an ἀρχή of but rather to say the manner in which x came to be an ἀρχή of those things. One may be an ἀρχὴ πράξεων of something that is not an action—for example, a state of character or an event or even a quality or an artefact perhaps—so long as one came to be the origin of that thing by virtue of exercising agency, the distinctively human capacity for action. On this reading, one may be an ἀρχὴ πράξεων of a state of character by virtue of producing that state of character through one's own actions. Similarly, if someone throws a rock and shatters a window, that window's shattering is not itself an action, but it is caused by the action the agent performed. Hence the agent is an ἀρχὴ πράξεων of that event. To capture this quasi-technical sense that ἀρχὴ πράξεων has here I suggest we translate it as ‘action-origin’Footnote ⁹ and otherwise follow Inwood and Woolf's translation.

This interpretation coheres quite well with the rest of E th. E ud. II 6. Consider, for example, the lines with which Aristotle opens the chapter and introduces the concept of an ἀρχὴ πράξεων (1222b15–20):

The contrast here between actions and mere animal behaviour is introduced by a more general contrast between those ways of being origins that are distinctively human and those that are not. Something serves as the origin of other things like itself, for example, merely in so far as it is a living thing (and as such has the capacity for reproduction). So, although one way in which humans function as origins is by reproducing, this is not a way of being an origin that they have qua human but rather qua living thing. In this respect, the kind of origins that human beings are in so far as they have the capacity to reproduce differs from the kind of origins that they are in so far as they have the capacity to produce actions. And it is because actions reflect a distinctively human capacity in this way that Aristotle wishes to call special attention to them. Philosophers today might speak of this capacity by speaking about ‘agency’, but, since Aristotle has no such term in his philosophical lexicon, he innovates and instead speaks about humans as ἀρχαὶ πράξεων.

If this interpretation is right, it has important implications for how we understand the project of the chapters that follow, II 7–9, and their relationship to the discussion of action in II 6. Whereas on the standard reading the investigation of the voluntary is meant only to shed light on what actions we are responsible for and are origins of, on my reading the goal of those chapters is rather different. What Aristotle is interested in is what things reflect an exercise of the distinctively human capacity to produce actions, which things, that is, are attributable to an exercise of our own agency. Hence it makes sense that in the eventual answer he gives, he would appeal to a distinctively human cognitive capacity, defining the voluntary as what is ‘on the basis of thought’ (κατὰ τὴν διάνοιαν, 1225b1) in II 9. What exactly he has in mind in defining the voluntary in this way is a bigger question, one that I will pursue in detail in a future project.