5
Giulio Di Basilio
5.1 Introduction
Both Aristotle’s Eudemian (EE) and Nicomachean Ethics (EN) devote significant attention to the topic of the voluntary (τὸ ἐκούσιον; cf. EN III, chs. 1 and 5; EE II, chs. 6–9).1 Although these accounts share important core ideas, a number of divergences between them, of varying degrees of philosophical significance, have been pinpointed by scholars. In what follows, I want to focus on one particular divergence, namely the fact that the EN does, while the EE does not, tackle the question of whether states of character (ἕξεις) are voluntary. I will start from the EN and show how the positive answer to that question is reached in the context of the Nicomachean account of voluntariness. I will then turn to the EE to show that, despite appearances, none of the relevant texts argues for a similar conclusion.
5.2 Voluntariness of Character States in the EN
I start on a linguistic note. Aristotle is wont to refer to the phenomenon at stake as the voluntary (τὸ ἑκούσιον). This nominalizes the slightly more common Greek adverb (ἐκών, ἑκουσίως). Standardly in Greek, ἑκών, and its opposite ἄκων, qualify actions (πράξεις); it is common in Greek to wonder, for instance, whether someone did something ἑκών or ἄκων. Aristotle’s practice of preferring the noun to the adverb no doubt serves the purpose of presenting the voluntary as an independent object of enquiry, as it were. But it also leaves the door open to an extension of the notion of voluntariness beyond its standard domain, i.e. human action. This is what is achieved in EN III 5, where the voluntary embraces character states (ἕξεις). This is the central claim that I will focus on in this section.
To begin with, it bears pointing out that there are a few neighbouring claims defended in EN III 5, and it will be useful to determine how they stand to one another. On the one hand, Aristotle’s conclusion goes as follows: If “the virtues are voluntary” (1114b22 ἑκούσιοί εἰσιν αἱ ἀρεταί), or similarly “up to us and voluntary” (b29 ἐφ’ ἡμῖν καὶ ἑκούσιοι), then “the vices will be voluntary too; for [they are] equally [voluntary]” (b24–25). This formulation is connected to the ostensible polemical target of this chapter, namely some kind of asymmetry thesis, which denies that virtue and vice are equally voluntary.2 There is, however, a more general conclusion secured at the end of this chapter, namely that states of character are voluntary. Indeed, Aristotle concludes also that “actions and states [of character] are not voluntary in the same way” (1114b30–31), which makes clear that he takes himself to have shown not only that virtue and vice, but also that character states more generally are voluntary. Of course, the two conclusions are not unrelated, for it is qua states of character that virtue and vice are voluntary. It is then worth bearing in mind that Aristotle’s conclusion will extend beyond virtue and vice, for instance to lack of self-control and other intermediate character states too. This broader conclusion and the way Aristotle formulates it alert us to an important qualification appended to it, namely that actions and character states are not voluntary in the same way, which I will return to and discuss in detail later on. Thus, we can state the results of Aristotle’s argument in EN III 5 as follows: actions and states of character are voluntary, and although voluntariness does not apply equally to both, qua states of character virtue and vice are equally voluntary.3
I now move on to examine the way in which the EN establishes the claim that character states are voluntary. Once we get clearer on how the EN treats this claim, we will be in a better position to consider whether the same conclusion is ever established in the EE. Let us start by recalling the account of voluntariness defended in EN III 1 for actions. There, Aristotle starts off by considering two involuntary-making factors, namely (i) force and (ii) ignorance (1109b35–110a1). Either of these factors is sufficient for a subsequent action to be involuntary, as in the case of being carried off by the wind, or in the case of someone firing a gun mistakenly believing it to be unloaded. What follows in EN III 1 is in large part an attempt to refine (i) and (ii) and show when they are legitimately given application. Important qualifications are added, notably that the force at stake has to be external to the agent who contributes nothing to the ensuing outcome, and that the ignorance which exculpates has to be non-culpable unawareness of the circumstantial features of one’s action; but (i) and (ii) are, by and large, vindicated in the rest of the chapter.4 A positive account of the voluntary is arrived at as that of which the principle is internal to the agent who is aware of the particulars in which the action takes place (EN III 1, 1111a22–4). On this view, for an action to be voluntary two requirements have to be met, which Aristotle derives from the involuntary-making factors (i) and (ii), namely (i*) the internality of the principle and (ii*) the knowledge condition. It would seem that by the end of EN III 1, Aristotle has left us with an ingenious account of what it is for an action to be voluntary, but has not so far broached the question of what it is for a character state to be voluntary.
This is not entirely correct, however. In EN III 1, in an attempt to pre-empt an objection to the idea that forced actions originate in what is external to the agent, the qualification appended to (i*), Aristotle argues as follows: if someone were to claim that pleasant and fine things force one to act upon them, then all actions would turn out to be forced, which is clearly absurd. On the contrary, one should not blame anything external for the attraction that thing exercises on us but rather oneself for being easy prey to it (1110b9–17). This remark brings in for the first time in Aristotle’s examination the importance of one’s character. For, to be easy prey to something (εὐθήρατος) is to be such that one cannot help acting on it when the opportunity presents itself. It is in virtue of one’s character that certain things strike us as pleasant and fine (1113a31). Already in EN III 1, Aristotle clearly implies that people are somehow open to praise and blame for their character in addition to their action, which anticipates the later examination of EN III 5.5
Having defended the above account in EN III 1 and having looked in the intervening sections at the notions of decision (ch. 2), deliberation (ch. 3), and wish (ch. 4), in EN III 5 Aristotle turns to virtue and vice. A large part of what Aristotle does in EN III 5, I submit, is to show that the account of the voluntary advocated in EN III 1 can be extended beyond actions. First, though, he shows that the account of the voluntary advocated earlier applies equally to virtue and vice (1113b3–1114a21).6 This is indicative of his resistance to asymmetry claims about virtue and vice, which, as we have seen, will be emphasized in the conclusions to the chapter. Next, he brings in ordinary practices of punishment and reward, both on the level of individual citizens and of legislators, to support his point that virtue and vice are equally voluntary (1113b21–26). Afterwards, he argues that one can be responsible for one’s own ignorance which results in wrongdoing, for instance if one does something prohibited by the laws in ignorance of the prohibition. In this way, it turns out that a condition like ignorance can be up to one and something for which people can be responsible.7 This amounts to a first extension of the account of voluntariness advocated for actions in chapter 1. That this is the argumentative strategy of the text can be gleaned from the way Aristotle goes about showing this, for he points out that one’s condition of ignorance can be voluntary precisely by showing that (i*), the requirement that the principle be internal to the agent, is fulfilled in this case. (Presumably, he does not think that [ii*], the knowledge requirement, can be fulfilled in the present case.) At this point, there is something of a turn in the chapter and the remainder is taken up by two objections as well as Aristotle’s replies to them (first objection at ll. 1114a3–30; second ll. 1114a30–b25). From now on, every claim can be interpreted one way or the other as relevant to the thesis of voluntariness of character states. Since it is the first objection which brings in the question of the voluntariness of character states, I will focus on it in greater detail.8 In view of Aristotle’s account of voluntariness in EN III 1, according to which what is voluntary is what fulfils (i*) and (ii*), the question whether character traits too are voluntary comes down to whether (i*) and (ii*) hold of one’s character too.
Aristotle is sanguine that the principle of acquiring a given character trait lies within the agent, and hence that (i*) holds of character states. The origin of a given character trait lies in the agent’s past actions. If someone were to say she cannot help but act the way her character prompts her to do, the retort would then be that she’s to blame for acquiring her character in the first place. One cannot deflect responsibility by saying, for instance, that one was born with their character, for character is not something one is born with, but is rather the product of a process in which the agent has the leading role.9 The argument used to support this (1114a3–31) relies heavily on Aristotle’s account of habituation defended in EN II 1–4. For someone to be of a certain type (τοιῦτός ἐστιν, 1114a3) comes down to having certain character states (ἕξεις), and character states are acquired by previously engaging in corresponding activities (ἐνέργειαι). Earlier in EN II 2, actions were said to be in control (κύριαι, 1103b30–31) of engendering corresponding character states through habituation. What Aristotle is doing in this later chapter is looking at the process from the opposite vantage point, so to speak: earlier, we were told that actions have the capacity to engender corresponding character traits; now that those character traits come about as a result of one’s previous actions. There are good textual indications that, in reply to the first objection of EN III 5, Aristotle is deliberately showing how (i*) applies to the case at hand by resorting to the mechanism of habituation: most clearly, he concludes that “the principle is in oneself” (ἡ γὰρ ἀρχὴ ἐν αὐτῷ, 1114a19) when it comes to forming character states.
Now, notoriously, Aristotle does not pause to ask questions which are likely to occur to us, in particular whether it is true that people are in control of the activities formative of corresponding character traits early on when the influence of one’s social environment might resemble external force. He presumably believes that in all such cases the principle is still internal. Any kind of influence exercised on the habituand will have to follow an internal route, and so long as the route is internal, then (i*), as defended earlier in chapter 1, will be fulfilled. We are likely to feel some discomfort at his solution on this point, but since this has been commented upon enough by others,10 I will refrain from taking this further. Instead, I will move on to the question as to how, if at all, the second requirement for voluntariness applies to character states, namely (ii*), the knowledge requirement.
The question to be asked is: what must one have knowledge of in order for the resulting character state to be voluntary? An answer to this question will amount to defining (ii*), the knowledge condition for voluntariness, for the case of character states. In the case of action, the requisite knowledge has been spelt out at length in EN III 1 as knowledge of the circumstances in which the action occurs (1110b18–1111a21). By contrast, Aristotle is somewhat elusive on the knowledge requisite for a given character trait to be voluntary. He never appears to tackle this issue as clearly as he does for (i*), when he shows by appeal to his account of habituation that the origin of character states lies within the agent. In principle, having argued that (i*) is fulfilled for character traits, Aristotle might presume that it is enough for all the actions leading up to the acquisition of a given character trait be voluntary in order for the ensuing character trait to be voluntary too. We should, however, review attentively any text from EN III 5 that might seem to bear on the question of the knowledge involved in the formation of character states. One such text is
“Not knowing that [character] states [derive] from being active [in a certain way] in each domain is a sign of great insensibility”11
(1114a9–10)12
In the text leading up to this, Aristotle has argued that the principle that character traits derive from previous activity is evident in the case of athletes training ahead of competitions. Here, we find a pre-emptive response to a possible objection: if someone were to disavow knowledge that character traits are formed by acting in a certain way in the past, the reply would be that the objection is hardly believable. There are two implications of this claim, both of which are prima facie puzzling: first, it seems to suggest that knowledge of this principle is necessary in order for the resulting character trait to be voluntary. Second, it implies that Aristotle considers the core idea of the account of habituation previously defended in book II as a piece of knowledge to which most people ordinarily somehow have access.
The thought is that nobody can reasonably disavow knowledge of the principle whereby one’s character is the product of, and results from, one’s past actions. The obvious retort is that such knowledge can hardly be expected of anyone, of children, for instance. Since earlier in the ethics Aristotle has emphasized the importance of early habituation, he appears to be particularly vulnerable to this criticism. A possible solution to this complication would be to suppose that he is here addressing an audience of mature agents.13 Such a restriction may be less abrupt than one might at first think, if we suppose, as some critics have suggested, that he has been addressing mature agents since the beginning of EN III 5 by focusing on actions out of decision.14 Even so, the challenge can be reiterated by asking why one should suppose that every adult knows that actions are formative of character. It seems plausible to think that some people might not know this, or that, even if they do, they fail to take it to heart and cannot as a result be expected to act accordingly.
It also seems puzzling to claim that everyone knows that character is formed by engaging in activities, a principle sometimes referred to as “like state from like activity.”15 For it has taken Aristotle careful argument to defend his own view, and nowhere in his account of habituation does he seem to take it for granted as obvious. To be sure, he does draw support from ordinary practices of rewards and punishments as well as from Plato as trailblazer for the importance of habituation; and he has resorted to the etymology of character virtue (ἠθικὴ ἀρετή) from habit (ἔθος) as evidence that the former derives from the latter (EN II 1, 1103a17–18); but he has by no means given the impression to be appealing to ordinarily accessible knowledge in defence of his account in EN II 1–4.
Perhaps Aristotle’s point is that this sort of knowledge is available under normal circumstances to most people via the admonitions of one’s educators and the like.16 If so, the idea is not that everyone possesses factual knowledge of the principle that character is formed by past actions, but rather that everyone has plenty of opportunities to learn this under normal circumstances. If Aristotle is right and opportunities to learn this are aplenty, then one will be culpably ignorant in failing to cop on to this principle. Understood in this sense, this type of ignorance would be analogous to the case of ignorance of what is prohibited by the laws considered earlier in EN III 5, in so far as it would concern what one can be reasonably expected to know under normal circumstances. I think there is a fitting analogy, employed in this chapter and elsewhere, which supports this reading, namely the analogy of bad and wretched people being like those who become sick by disregarding their doctor’s advice (1114a16). The very same image had been used at the end of EN II 1–4 to rebuke those who do not bother to engage in virtuous activity to acquire virtue but prefer to take refuge in discourses: there too Aristotle had said that their case is analogous to refusing to take one’s doctor’s advice and hoping to stay in good health regardless (1105b12–16). If this is the thought behind the puzzling claim we have been focusing on, perhaps Aristotle may further take himself to be contributing to this protreptic endeavour by reminding people of the formative power of engaging in the right activities and abstaining from bad ones.17
We are still faced with the question about the kind of knowledge necessary for someone to be said to have acquired one’s character voluntarily. I suggest we look at one last text from EN III 5 which seems to deal with this question, if only indirectly.
As we have mentioned earlier, Aristotle’s conclusion has it that actions and character traits are not voluntary in the same way (1114b30, οὐχ ὁμοίως, etc.). In a bid to expand on the requisite qualification, Aristotle goes into some more detail about what exactly, in the case of character traits, we do not have knowledge of. It is now worth quoting the whole passage.
Actions and states, however, are not voluntary in the same way: for we are in control of actions from beginning to end, so long as we know the particular circumstances, but [we are in control] of the beginning of states, while the progressive advancement [of the state] in the particular circumstances is not known to us [καθ’ ἕκαστα δὲ ἡ πρόσθεσις οὐ γνώριμος], as in the case of diseases; but since it was up to us to use [the particular circumstances] this way or that way, for this reason [states] are voluntary.
(EN III 5, 1114b30–1115a2)18
It is clear that Aristotle is qualifying the second requirement of voluntariness, namely (ii*) the knowledge condition. The qualification at stake is asserted elsewhere in this chapter in a number of slightly different ways, which are presented as equivalent:
1. while people are in control of their actions from beginning to end, they are only in control of the beginning of their character traits (1114b31–1115a1);
2. everyone is responsible for their own character state in a way (1114b1–2, ἕκαστος ἑαυτῷ τῆς ἕξεώς ἐστί πως αἴτιος);
3. we are jointly responsible in a way of our character traits (1114b22-23, τῶν ἕξεων συναίτιοί πως αὐτοί ἐσμεν).
There is a close link, in other words, between the fact that we are only in control of the onset of character formation, the way voluntariness applies to character states, and our qualified responsibility for them. Scholars have discussed at length the nature of the qualification at stake.19 A common explanation has it that Aristotle is making room for other partners responsible for our character states, presumably nature and upbringing.20 In my view, this is only partly correct: it is true that there is a qualification of the voluntariness of character states at stake here; but this has nothing to do with the involvement of other causal factors like nature and upbringing in addition to the agent. If that were the case, we would expect Aristotle to qualify (i*) the internality of the principle requirement, whereas it is (ii*), the knowledge requirement, which is at stake in the above passage. More precisely, the qualification is introduced in view of there being some kind of ignorance involved in character formation.
Before developing this, let us take our bearings by focusing on the general sense of the qualification. This is best understood as a restatement of a vivid claim made earlier in EN III 5 about character traits: while we are not in control of retrieving a stone once it has been thrown, it was clearly up to us not to throw it in the first place (1114a16–19). The thought seems to be that there is a certain momentum in the way in which character traits develop to the effect that our control over them gradually diminishes as they become ingrained. The explanation for this is presumably to be found in the idea that it is in virtue of one’s character traits that things appear to us in a certain way, and, most importantly that some of them strike us as pleasant and fine (1113a31). Once established, then, character traits will incline us to pursue the things that strike us as pleasant and fine. Since one cannot act upon the way things appear to us to the same extent as one can take action to pursue the things that strike us as pleasant and fine, our control over the former is bound to be inferior to our control over the latter.
Now, presumably with his stone example, Aristotle has exaggerated the extent to which character cannot be changed once established. For one thing, earlier Aristotle contrasted the mechanism of acquisition of character virtue through habituation with the case of natural movements by arguing precisely that a stone cannot be habituated to fall upwards (1103a18–23). This implies that forming one’s character is not like the throwing of a stone. The image of the stone being irretrievable once thrown is probably best understood inter alia as a warning to make sure we throw the right stones, as it were; that is, that we develop the right habits.21 Nor does the image rule out the possibility of character change: elsewhere, Aristotle points out that “it is not impossible, or not easy, to alter by argument what has long been absorbed as a result of one’s habits” (EN X 9, 1179b15–20). The parallel with diseases in EN III 5 makes clear that Aristotle is especially concerned with cases of people having developed bad habits and not being able to act otherwise once their character has become ingrained. In other words, since it will take an awful lot of effort to break a bad habit once it is formed, it is better not to acquire it in the first place; once again, a protreptic thread runs through this chapter.
There is another idea in the last text quoted which still deserves comment, namely the fact that our diminished control over character states is, on the face of it, explained by a certain ignorance. For there, Aristotle compares and contrasts the knowledge condition, i.e. (ii*), in the case of actions and character states. In EN III 1, the relevant type of knowledge requisite for voluntariness of actions was elaborated as knowledge of the particulars in which the action occurs (1111a22–4). If someone has this type of knowledge whilst acting, their control over the ensuing action is complete. In the case of character states, by contrast, the progressive advancement of the state is not known to us. The idea seems to be that people are in no position to know the exact contribution of individual actions to the incipient character state. There is reason to think that this idea concerns especially bad actions and their contribution to the development of vice and other undesirable character traits. The parallel with diseases is instructive here: in this case too it is hard to know in advance what kind of behaviour will be sufficient to trigger the onset of a disease. Someone may do action x1 at time t1 and not be aware that x1 contributes to developing disease y, and then do action x2 at time t2, action x3 at time t3, and so on, all the while contributing to disease y; yet there is no way to know in advance which x will cause the onset of disease y to the point where one is no longer capable of preventing it.22
If this is along the right lines, two remarks are in order. First, it suggests that this qualification is Aristotle’s last minute concession to the idea that vice is involuntary and not to blame. For, as we have seen, Aristotle is prepared to concede that there is an element of ignorance connected to the impossibility to know in advance the contribution of individual actions to the formation of character states, and this type of ignorance is of the kind which exculpates. However, the concession does not undermine the voluntariness of character. Second, with his examination Aristotle most likely intends to exhort his audience to pay close attention to the habits one sets out to develop. Since we are not able to know in advance what kind of effect activities, in particular bad ones, will have on us, the rational reaction to this is an overall policy of caution and the adoption of a reflective attitude towards the character traits one is in the progress of developing.
5.3 Voluntariness of Character States in the EE
Let us briefly take stock. In the previous section, I hope to have shown that the question addressed in EN III 5, namely whether character states are voluntary, is not simply tacked onto Aristotle’s previously reached conclusions about voluntariness: on the one hand, that question has been alluded to previously in Aristotle’s EN; on the other, the affirmative answer reached to that effect in large part amounts to an extension to the character states of Aristotle’s account of voluntariness for actions. In view of all of this, it will be of interest to consider whether the EE takes any stance on the question of character states, whether they are voluntary or otherwise.
Some critics have read into a couple of texts of the EE arguments for the conclusion that character states are voluntary.23 More generally, there are a few texts of the EE that might seem to deal with voluntariness of character states, and it will be worth reviewing them all in what follows. The first relevant text occurs at the end of EE II 6. This chapter prefaces Aristotle’s enquiry into voluntariness, which is to follow in the next chapter, with a series of remarks on what it means to be causally responsibly for, and the originating principle of, actions.24 Towards the end of the chapter, Aristotle goes back to his enquiry into virtue of character, which is the principal focus of EE book II as a whole.
Since virtue and vice and the deeds [that originate] from them are objects of praise and blame; for, one does not bestow praise and blame on what happens either because of necessity, chance, or nature, but on those things for which we are causally responsible. Indeed, of those things for which someone else is causally responsible it is that person who receives praise or blame. It is clear that both virtue and vice concern those things that one is causally responsible for and the originating principle of.
(EE II 6, 1223a9–1223a15)25
Aristotle reasons as follows: (a) virtue and vice receive praise and blame, respectively; (b) one praises and blames not what happens by necessity, chance, or nature, but rather that for which one is causally responsible; hence, (c) virtue and vice concern those things for which one is causally responsible. Now, (a) does not make it clear whether what is meant is virtue and vice as states, or only the actions which derive from them, or both. Nor does the reference to praise and blame help us settle this, for praise and blame are reactions appropriate in principle both to actions and character states.26 Since explicit mention is made of the deeds which flow from virtues and vices (cf. τὰ ἀπ’ αὐτῶν ἔργα),27 it seems that actions are clearly within the scope of the argument; hence the two options left would seem to be either that the argument is about both character states and action, or about actions exclusively.28 Nevertheless, in (c) the conclusion drawn concerns only the actions of virtue and vice (the ταῦτα in [c] presumably picking up on τὰ ἀπ’ αὐτῶν ἔργα of [a]); this implies that the focus of the argument is on the actions characteristic of virtues, things like returning a debt or abstaining from overeating as manifesting justice and moderation, respectively.
Aristotle could have argued as follows: (a) virtue and vice receive praise and blame; (b) one praises and blames not what happens by necessity, chance, or nature, but rather that for which one is causally responsible; hence, (c*) virtue and vice are things for which one is causally responsible. It is significant that he stops short of drawing this conclusion. I suggest that the weaker conclusion, namely (c) above, is more in line with Aristotle’s purposes in EE II 6. This chapter paves the way for the examination of the voluntary by situating human action in the natural world as Aristotle conceives of it. Aristotle argues that although human beings are unique in nature in being originating principles of action, still action does not lie outside of nature, so to speak; this is the reason why action is characterized as (a type of) movement (1222b28–29). Aristotle sets up to show that virtue and vice are up to us in the sense that it is open to us to engage in the actions characteristic of them in view of the domain in which they occur.29 This is the domain of virtue; and of vice as well. This text shares this much with EN III 5, namely the refusal to leave room for asymmetry claims about virtue and vice. Indeed, human beings are causally responsible for, and originating principles of, opposite outcomes (1222b41–42), hence what applies to virtue applies to vice just as well. Nevertheless, unlike EN III 5, the argument here is centred on action. Once we bring into clearer focus the objectives of this section, it at once becomes clear that there is no intention on Aristotle’s part to drive home the stronger point that virtue and vice, understood as character states, are things for which we are responsible.30
The last text quoted does not use the notion of voluntariness. What follows immediately in EE II 6 brings in the voluntary by establishing an equivalence class between that for which one is responsible, previously used in the text above, and the voluntary.
It is hence necessary to grasp which actions one is causally responsible for and originating principle of. Everyone agrees that one is causally responsible for voluntary things and things in accord with one’s decision, whereas one is not causally responsible for involuntary things. But everything which has been decided on is also clearly done voluntarily. Hence, it is clear that both virtue and vice will be of voluntary things
(EE II 6, 1223a15–20)31
Here, the same conclusion reached earlier is re-affirmed by using the newly introduced notion of voluntariness. Earlier, the conclusion had been that virtue and vice concern those things that one is responsible for and principle of, whereas now virtue and vice are within the domain of voluntary things. There is no denying that the expression used in the conclusion is unhelpfully ambiguous, the genitive “of voluntary things” (a20, τῶν ἑκουσίων) potentially meaning things as different as “virtue and vice concern voluntary things” and “virtue and vice are voluntary.” However, in these lines Aristotle intends the new conclusion to follow from the previous one on the strength of the invoked consensus about the connection between the notion of voluntariness and responsibility. (While introduced for the first time here, the notion of decision is left in abeyance during the examination of the voluntary in EE II 7–9, but will reappear in the next and last text to look at.) Since, as I said earlier, this chapter is devoted to assigning a somewhat clearer ontological domain to human action within the natural world, I suggest that our reading of these lines should be in keeping with the general understanding of this chapter. Thus, Aristotle is best understood as claiming that virtue and vice are within the domain of voluntary things in the sense that the actions characteristic of them are voluntary.32
So much for EE II 6. The last relevant text from the EE occurs in the final chapter of book II, i.e. EE II 11. This chapter is, in Woods’ words, “something of an appendix to the discussion of virtue of character and decision, to which as a whole nothing corresponds in the EN.”33 It will be helpful to remind us briefly of the question addressed there. The topic of the chapter is whether (i) virtue prevents mistakes with regard to one’s decision and makes the goal correct, or whether (ii) virtue makes one’s reasoning correct (1227b12–15). Aristotle comes down on the former side of the question and accuses those who come down on the latter side of mistaking virtue for self-control (ἐγκράτεια). What emerges fairly clearly is Aristotle’s firm commitment to the idea that virtue is responsible in a special way for giving its possessor the right goal at which to aim. Despite being on a par with virtue as a praiseworthy state, self-control falls short of this. Indeed, only the virtuous person aims at the mean (τὸ μέσον), which is an indication of the fact that virtue is responsible for making one’s goal right (1227b37–38). What is characteristic of virtue is that its possessor takes action with the right end in view. Looking at someone’s decision (προαίρεσις), which has been previously examined in EE II 10, is of help here as it enables us to assess someone’s goal in acting. Aristotle has suggested that decision has a standard structure, namely that it is always of something for the sake of something (1226a11–13;1227b36–37), where the first variable refers to an action and the second to the goal the agent has in view in acting as she does (call this “decision schema”). This nexus of ideas is put to use in what follows in the text, which reads thus:
And it is because of this [sc. the decision schema] that we judge what one is like from their decision: this is what one acts for the sake of, not [simply] what one does. Similarly, vice too makes one’s decision be for the sake of contrary things. If, therefore, someone, when it is up to him to do fine things and to refrain from doing shameful ones, does the opposite, it is clear that such a human being is not good. It follows that it is necessary that vice and virtue are voluntary: for, there is no necessity to do wretched things. For these reasons it is the case both that vice is blamed and virtue is praised: for, neither does one blame shameful and bad things done involuntarily, nor does one praise good things [done involuntarily], but rather voluntary ones.
(EE II 11, 1228a2–11)34
The argument is reminiscent of the passages from EE II 6 already looked at, not least for its conclusion that virtue and vice are voluntary. If, at this point, Aristotle can go into some more detail it is because he has defined in the meantime the notion of decision, which is key to his argument here.35 In particular, here decision serves the purpose of showing that that which is true of virtue is, by parity of reasoning, true of vice as well. For someone’s character is manifested most clearly in their acting in a certain way for the sake of a given end, namely in their decision. The best condition of one’s character, i.e. virtue, is responsible for decision’s aiming at the right goal, namely the mean; by parity of reasoning, vice will be responsible for decision’s aiming at the opposite goal. (Presumably by using the plural “opposite goals” Aristotle refers to the two opposite directions in which it is possible to depart from the mean either by excess or deficiency.) Whenever someone does a bad act, it being up to them to refrain from it, one is voluntarily bad and reveals one’s character accordingly; they have chosen a bad end as worth pursuing and are prepared to take action accordingly.
On the face of it, here there might seem to be more to Aristotle’s talk of virtue and vice than simply their respective characteristic actions. For one thing, in the preceding chapter, namely EE II 10, virtue has been defined as a character state (1227b5–11, esp. ἕξις at l. 8). Hence, one might think that from there on the default understanding of virtue will be according to its definition. Similarly, Aristotle’s controversial claim that virtue provides the right goal is more naturally construed as a claim about virtue as character state than virtue as pattern of actions. On the other hand, in the last text quoted, the emphasis of Aristotle’s argument is placed on action, as the use of the verb πράττειν and cognate throughout makes clear. I suggest, therefore, that, not unlike EE II 6, here the focus is on the voluntariness of the actions which manifest virtue and vice and that Aristotle is content with showing that the actions of virtue and vice are equally up to us. To be sure, there is no reason to deny that Aristotle may well believe virtue and vice to be voluntary in a sense which goes beyond the voluntariness of the actions they give rise to; but, contrary to what we have seen in EN III 5, he is not in the business of giving an argument to that effect.
5.4 Conclusion
If my argument above is correct, EN III 5 is the only place where Aristotle tackles the question of the voluntariness of states of character by giving an argument in its defence. I hope to have shown that that conclusion is admirably consonant with the account of the voluntary previously established for action in EN III 1. When we turn to the EE, despite appearances we do not find any argument to that effect. As often in similar cases, it seems easier to think that the EN was written with the benefit of the EE, otherwise it would be unclear why Aristotle would have left out the question of the voluntariness of character states entirely in the latter work; at the same time, we have found evidence that the two works pursue significantly different agendas in a way which resists simplistic assessments. Both of them deserve assiduous attention.
Notes
1. I use “voluntary” for τὸ ἐκούσιον and “involuntary” for τὸ ἀκούσιον; “cause” or “causally responsibly” for αἴτιος. None of these translations is without its difficulties, but they are preferable to their competitors.
2. That is, the Platonic Socrates, according to Sauvé Meyer, 2011, ch. 5; the Plato of the Timaeus and the Laws, according to Kamtekar, 2019, p. 60. Despite there being important differences between their interpretations, they are agreed that some kind of asymmetry thesis is Aristotle’s polemical target.
3. See Lienemann, 2018, p. 452, 512; contrast Sauvé Meyer, 2011, p. 128.
4. I disregard another puzzling requirement for involuntariness, namely that the agent has to regret the action done; for one thing, this is not mentioned in the final account of involuntariness (1111a22-24); for the other, it can be thought of as a symptom of, and not a condition for, involuntariness.
5. Along similar lines, see Furley, 1977, pp. 49–50.
6. These lines have been much discussed with regard to the determinism/compatibilism debate, see Destrée, 2011; Bobzien, 2013.
7. Sauvé Meyer, 2011, p. 139 argues that the reference is to Socrates and his conviction that vice lies in ignorance of the good. See also Lienemann, 2018, pp. 511–512.
8. Careful analyses of the second objection are Grgić, 2017; Kamtekar, 2019, pp. 80–82 helpfully suggests that the second objection re-articulates the challenge posed by the first by means of some ideas accepted by Aristotle himself.
9. Broadie, 1991, p. 165 has emphasized how the argument of EN III 5 is largely aimed at refuting the idea that character is innate. Although her remarks are very helpful, she perhaps overly restricts Aristotle’s arguments to this particular target for criticism.
10. Lienemann, 2018, pp. 479–488; Sauvé Meyer, 2011, pp. 123–124; Broadie, 1991, p. 167; Sorabji, 1980, pp. 266–269; Furley, 1977, p. 53.
11. On the meaning of the expression κομιδῆ ἀναισθήτου, see Bondeson, 1974, pp. 61–62. I disagree with Bondeson that people “who are ignorant of this are beyond the bound of sense and are probably so deficient that they cannot be counted as moral agents” (p. 64).
12. τὸ μὲν οὖν ἀγνοεῖν ὅτι ἐκ τοῦ ἐνεργεῖν περὶ ἕκαστα αἱ ἕξεις γίνονται, κομιδῇ ἀναισθήτου.
13. Irwin, 1980, p. 140; Cooper, 2013, pp. 308–309.
14. The suggestion is Donini’s, who however does not accept it in the end, see Donini, 2003, p. 11.
15. Lawrence, 2011, p. 246.
16. This is the suggestion made in Donini, 2003, p. 19; see also Lienemann, 2018, pp. 492–493.
17. For the idea that Aristotle’s ethics, and the EN in particular, belongs to the genre of protreptic discourses, see Hutchinson, Johnson, 2014. According to them, in the EN Aristotle’s protreptic efforts are aimed at exhorting his audience to philosophy; but they do make room for the idea that there might be instances of apotreptic discourses against vice and protreptic discourses in favour of particular virtues, see esp. pp. 396–397.
18. οὐχ ὁμοίως δὲ αἱ πράξεις ἑκούσιοί εἰσι καὶ αἱ ἕξεις· τῶν μὲν γὰρ πράξεων ἀπ’ ἀρχῆς μέχρι τοῦ τέλους κύ-ριοί ἐσμεν, εἰδότες τὰ καθ’ ἕκαστα, τῶν ἕξεων δὲ τῆς ἀρχῆς, καθ’ ἕκαστα δὲ ἡ πρόσθεσις οὐ γνώριμος, ὥσ-περ ἐπὶ τῶν ἀρρωστιῶν· ἀλλ’ ὅτι ἐφ’ ἡμῖν ἦν οὕ-τως ἢ μὴ οὕτω χρήσασθαι, διὰ τοῦτο ἑκούσιοι. Scholars have doubted both whether this passage did not originate as a marginal note and, more plausibly, whether its position in the MSS is correct, for more details see Stewart, 1892, p. 281; Frede, 2020, pp. 47–48 suggests, following Rassow, moving it to 114b25; Gauthier, 1970, ad loc., suggests moving it earlier to 1114a31.
19. Sauvé Meyer, 2011, p. 127 argues that Aristotle intends to establish only what she calls qualified, as opposed to full, responsibility for one’s character states; see also Lienemann, 2018, pp. 508–509. Other interpreters, e.g. Ott, 2006; Grgić, 2017, argue against this by maintaining that Aristotle is not qualifying one’s responsibility for character states at all. Ott, 2006, p. 74 argues that if Aristotle were to qualify one’s responsibility for character, then he would have to qualify one’s responsibility for actions too; but this flies in the face of (i)–(iii), which clearly state that there is a difference between the voluntariness of actions and character states. Grgić, 2017, pp. 118–119 argues that the qualification has to do with there being other explanatory factors which account for people’s character states, notably goals in addition to efficient causes (i.e. us); this is, of course, true, but it would fail to apply to character states exclusively, for that is true of actions as well.
20. Irwin, 1999, p. 211; Destrée, 2011, p. 310. Modern commentators follow Aspasius, who argues along similar lines, but mentions nature and chance, see 70.33–80.3.
21. Di Muzio, 2000 is an informative discussion of the issues involved here. He argues that Aristotle does not rule out character change and moral reform. In particular, the stone example, on his view, is only meant to show that one cannot change one’s character simply by wishing to do so. Cf. also EN IX 3, 1165b13–22; Cat. 10, 13a23–31; Pol. 1332b4–11.
22. Aspasius’ reading seems to be along similar lines, see 80.21–29; see also Donini, 2003, p. 12, n. 26.
23. See Wolt, 2019; Broadie, 1991, p. 162.
24. I use “causally responsible” for αἴτιος and “originating principle” for ἀρχή. Although cumbersome, these seem close to how Aristotle introduces them in EE II 6. I sometimes omit the qualifications and use “responsible” and “principle” for brevity’s sake.
25. ἐπεὶ δ’ ἥ τε ἀρετὴ καὶ ἡ κακία καὶ τὰ ἀπ’ αὐτῶν ἔργα τὰ μὲν ἐπαινετὰ τὰ δὲ ψεκτά· ψέγεται γὰρ καὶ ἐπαινεῖται οὐ διὰ τὰ ἐξ ἀνάγκης ἢ τύχης ἢ ὑπάρχοντα ἀλλ’ ὅσων αὐτοὶ αἴτιοι ἐσμέν, ὅσων γὰρ ἄλλος αἴτιος, ἐκεῖνος καὶ τὸν ψόγον καὶ τὸν ἔπαινον ἔχει· δῆλον ὅτι καὶ ἡ ἀρετὴ καὶ ἡ κακία περὶ ταῦτ’ ἐστιν ὧν αὐτὸς αἴτιος καὶ ἀρχὴ πράξεων.
26. Contrast Simpson, 2013, pp. 260–261.
27. Cf. 1220a19–21.
28. The issue could be settled by taking the καί at l. 9 as epexegetic. I have preferred not to iron out the ambiguity of the argument. I have no objection in principle, however, to adopting this construal as a result of my argument. I thank Dan Ferguson for pressing me on this issue.
29. On the connection between this chapter and the project of the EE, see the helpful suggestions in Müller, 2015, pp. 215–216; Wolt, 2019.
30. On this passage, I agree with Sauvé Meyer, 2011, p. 43; however, she goes on to mention the conclusion drawn at EE II 6, 1223a15–20, which is the next passage I will look at, as the only place where possibly the voluntariness of character traits might be affirmed.
31. ληπτέον ἄρα ποίων αὐτὸς αἴτιος καὶ ἀρχὴ πράξεων. πάντες μὲν δὴ ὁμολογοῦμεν, ὅσα μὲν ἑκούσια καὶ κατὰ προαίρεσιν τὴν ἑκάστου, ἐκεῖνον αἴτιον εἶναι, ὅσα δ’ ἀκούσια, οὐκ αὐτὸν αἴτιον. πάντα δ’ ὅσα προελόμενος, καὶ ἑκὼν δῆλον ὅτι. δῆλον τοίνυν ὅτι καὶ ἡ ἀρετὴ καὶ ἡ κακία τῶν ἑκουσίων ἂν εἴησαν. Wolt, 2019, stresses the importance of the optative mood used in the conclusion which, on his view, would signal that he hopes to demonstrate this claim in what follows. This is plausible, and one might further suggest that the conclusion in question is established in EE II 11, the last passage I will consider below.
32. Similar considerations apply to the version of this argument given in MM I 9, 1187a19–23.
33. Woods, 1992, pp. 151–152. I have replaced Woods’ “choice” with “decision”’ for the sake of consistency with the rest. A very helpful discussion of this part of EE II 11 is Lienemann, 2018, pp. 345–352.
34. καὶ διὰ τοῦτο ἐκ τῆς προαιρέσεως κρίνομεν ποῖός τις· τοῦτο δ’ ἐστὶ τὸ τίνος ἕνεκα πράττει, ἀλλ’ οὐ τί πράττ-ει. ὁμοίως δὲ καὶ ἡ κακία τῶν ἐναντίων ἕνεκα ποιεῖ τὴν προαίρεσιν. εἰ δή τις, ἐφ’ αὑτῷ ὂν πράττειν μὲν τὰ καλὰ ἀπρακτεῖν δὲ τὰ αἰσχρά, τοὐναντίον ποιεῖ, δῆλον ὅτι οὐ σπουδαῖός ἐστιν οὗτος ὁ ἄνθρωπος. ὥστ’ ἀνάγκη τήν τε κακίαν ἑκούσιον εἶναι καὶ τὴν ἀρετήν· οὐδεμία γὰρ ἀνάγκη τὰ μοχθηρὰ πράττειν. διὰ ταῦτα καὶ ψεκτὸν ἡ κακία καὶ ἡ ἀρετὴ ἐπαινετόν· τὰ γὰρ ἀκούσια αἰσχρὰ καὶ κακὰ οὐ ψέγεται οὐδὲ τὰ ἀγαθὰ ἐπαινεῖται, ἀλλὰ τὰ ἑκούσια.
35. Müller, 2015, p. 249, n. 71 agrees that the EE does not tackle the question of the voluntariness of character. However, he argues that someone acting out of decision is responsible not only for the action done but also for “being the kind of person who does the action” (p. 250). Strictly speaking, though, all that Aristotle is entitled to conclude is that the agent is responsible for the kind of person she wants to be (see p. 248, where, as I understand it, Müller argues along these lines).
References
1. Aspasius, In Ethica Nicomachea Quae Supersunt Commentaria, ed. G. Heylbut, CAG XIX.1. Berlin: Reimer, 1889.
2. Bobzien, S., ‘Found in Translation: Aristotle’s Nicomachean Ethics 3.5, 1113b7-8 and its Reception’, Oxford Studies in Ancient Philosophy, 45, 2013, pp. 103–148.
3. ———, ‘Choice and Moral Responsibility in Nicomachean Ethics III 1–5’, in Polansky, R. (ed.), The Cambridge Companion to the Nicomachean Ethics. Cambridge: Cambridge University Press, 2014, pp. 81–109.
4. Bondeson W., ‘Aristotle on Responsibility for One’s Character and the Possibility of Character Change’, Phronesis 19, 1974, pp. 59–65.
5. Broadie, S., Ethics with Aristotle. Oxford: Oxford University Press, 1991.
6. Charles, D., ‘The Eudemian Ethics on the “Voluntary”’, in Leigh, F. (ed.), The Eudemian Ethics on the Voluntary, Friendship, and Luck. Leiden: Brill, 2012, pp. 1–27.
7. Cooper, J., ‘Aristotelian Responsibility’, Oxford Studies in Ancient Philosophy, 45, 2013, pp. 265–312.
8. Destrée, P., ‘Aristotle on Responsibility for One’s Character’, in Pakaluk, M. and Pearson, G. (eds.), Moral Psychology and Human Action in Aristotle. Oxford: Oxford University Press, 2011, pp. 285–319.
9. Di Basilio, G., ‘Habituation in Aristotle’s Ethics. The Eudemian Ethics, the Common Books, and the Nicomachean Ethics’, The Journal of the History of Philosophy 59, 2021, pp. 531-557.
10. Di Muzio, G., ‘Aristotle on Improving One’s Character’, Phronesis 45, 2000, pp. 204–219.
11. Donini, P., ‘Volontarietà di vizio e virtù. (Aristot. Ethi. Nic. III 1–7)’, in Berti, E. (ed.), Napolitano Valditara, L., Etica, Politica, Retorica. Studi su Aristotele e la sua presenza nell’età moderna. L’Aquila: Japadre, 1989, pp. 3–21 (repr. in Donini, 2010).
12. ———, Aristotle and Determinism. Leuven: Peeters, 2010.
13. ———, Abitudine e saggezza. Torino: Edizioni dell’Orso, 2014.
14. Echeñique, J., Aristotle’s Ethics and Moral Responsibility. Cambridge: Cambridge University Press, 2012.
15. Frede, D. (ed.), Nikomachische Ethik. Berlin: Akademie Verlag, 2020.
16. Furley, D. J., ‘Aristotle on the Voluntary’, in Barnes, J., Schofield, M., and Sorabji, R. (eds.), Articles on Aristotle: 2. Ethics & Politics. London: Duckworth, 1977, pp. 47–60.
17. Gauthier, R. A. and Jolif, J. Y., Aristote: L’éthique à Nicomaque. Louvain: Publications Universitaires de Louvain, 1970.
18. Grgić, F., ‘Aristotle on Co-Causes of One’s Dispositions’, Elenchos, 38, 2017, pp. 107–126.
19. Hutchinson, D. S. and Johnson, M. R., ‘Protreptic Aspects of Aristotle’s Nicomachean Ethics’, in Polansky, R. (ed.), The Cambridge Companion to the Nicomachean Ethics. Cambridge: Cambridge University Press, 2014, pp. 383–409.
20. Irwin, T. H., ‘Reason and Responsibility in Aristotle’, in Rorty, A. (ed.), Essays on Aristotle’s Ethics. Berkley: University of California Press, 1980, pp. 117–155.
21. ———, Aristotle’s Nicomachean Ethics (2nd ed.). Indianapolis: Hackett, 1999.
22. Kamtekar, R., ‘Aristotle contra Plato on the Voluntariness of Vice: The Arguments of Nicomachean Ethics 3.5’, Phronesis 64, 2019, pp. 57–83.
23. Kenny, A., Aristotle’s Theory of the Will. London: Duckworth, 1979.
24. Lawrence, G., ‘Acquiring Character: Becoming Grown-Up’, in Pakaluk, M. and Pearson, G. (eds.), Moral Psychology and Human Action in Aristotle. Oxford: Oxford University Press, 2011, pp. 233–283.
25. Lienemann, B., Aristoteles’ Konzeption der Zurechung. Berlin: De Gruyter, 2018.
26. Müller, J., ‘Agency and Responsibility in Aristotle’s Eudemian Ethics’, Phronesis, 60, 2015, pp. 206–251.
27. Ott, W., ‘Aristotle and Plato on Character’, Ancient Philosophy 26, 2006, pp. 65–79.
28. Sauvé Meyer, S., Aristotle on Moral Responsibility (1st ed. 1993). Oxford: Oxford University Press, 2011.
29. Simpson, P., The Eudemian Ethics of Aristotle. New Brunswick, NJ: Transaction Publishers, 2013.
30. Sorabji, R., Necessity, Cause, and Blame. London: Duckworth, 1980.
31. Stewart, J. A., Notes on the Nicomachean Ethics of Aristotle. Oxford: Clarendon Press, 1892.
32. Wolt, D., ‘The Aim of Eudemian Ethics ii 6–9’, Ancient Philosophy 39, 2019, pp. 137–149.
33. Woods, M., Aristotle. Eudemian Ethics Books I, II, and VIII. Oxford: Clarendon Press, 1992.