I've been pondering this apparent disparity:
(1) "Nowhere in the Eudemian discussion of voluntariness ... does Aristotle show any concern to establish the voluntariness of states of character" (Meyer, 1994, p. 43).
(2) "It is clear, therefore, that also virtue and vice would be in the class of voluntary things" (EE 1223a19-20).In her book Meyer notes that (2) is perhaps an "exception" and leaves it at that; in her article, the line is simply left out. In her book Meyer says, implausibly, that in (3) 'vice' and 'virtue' appear to refer to actions, not states of character; in her article, the difficulty receives no mention. [N.B. See note (*) below.]
(3) "A necessary consequence is that vice is voluntary, as also virtue" (EE 1228a7-8).
Then there are texts in NE. Someone might urge that all of III.5 counts against her view, but especially the following lines:
Well, if each of us is himself somehow responsible for causing his disposition in himself, he will also be somehow responsible for the appearance in question. If he is not somehow responsible for his disposition, no one is responsible for its being the the case that he himself does bad things; he does these things because of ignorance of the end, thinking that by means of them he will get what is best. In this case, aiming at the end will not be self-chosen (Rowe and Broadie trans., 1114b3-5).
ei) me\n ou)=n e(/kastoj e(autw=| th=j e(/cew/j e)sti/ pwj ai)/tioj, kai\ th=j fantasi/aj e)/stai pwj au)to\j ai)/tioj: ei) de\ mh/, ou)qei\j au(tw=| ai)/tioj tou= kakopoiei=n, a)lla\ di' a)/gnoian tou= te/louj tau=ta pra/ttei, dia\ tou/twn oi)o/menoj au(tw=| to\ a)/riston e)/sesqai, h( de\ tou= te/louj e)/fesij ou)k au)qai/retoj,
Bywater's apparatus indicates: 1114b3. mh/, ou)qeij] mhdei\j Lb Mb G
On Bywater's text, the passage contradicts Meyer, because Aristotle would in essence be affirming the conditional, "if character is not voluntary, our actions are not voluntary", and by Modus tollens one gets the offending claim.
By the way, the reading that Bywater opts for--which is the reading of the best single ms., Kb-- is accepted also by Bekker, Ramsauer, Stewart, Burnet, and others (including, evidently, Chris Rowe).
Meyer deals with this in her article as follows:
Modern readers often assume that both Aristotle and the objector [sc. in NE III.5] agree on the principle that if a person acts as he is disposed to, then his action is not up to him (or does not originate in him) unless it can be shown that the disposition itself is up to him or originates in him (for example, Hardie 1980: 175). In a nutshell, the principle is that responsibility for action requires responsibility for character. This is a very common modern assumption about responsibility, and it seems to be what motivates the hypothesis we rejected at the beginning of this chapter: that establishing responsibility for character is the main goal of the account of voluntariness. But is there any evidence that Aristotle endorses such a principle?So here Meyer presents the reading which conflicts with her thesis as if it were an innovation or conjecture of John Burnet! And the reading which some scholars have favored (including Rassow and Susemihl), and which certainly is defensible, is referred to not as "a possible reading" but rather as, unambiguously, "the better reading". It's also not clear that the problem goes away on the alternative reading: Aristotle's argument in substance would appear to be the same.
Two passages in NE III.5 may give the misleading impression of articulating or implying the principle: (a) 1113b17-21 and (b) 1114b3-4. However, (a) concerns the asymmetry thesis about action, not the thesis of responsibility for character. On Burnet's (1990) [sic] reading of of 1114b3 (ei de me outheis), (b) does appear to articulate the principle. But the better reading is "if no one -- ei de medeis -- is responsible for his wrongdoing ... ." So nowhere in NE III.5 does Aristotle even articulate, let alone endorse, the principle.
In a footnote in her book Meyer at least gives a reason in favor of the reading that does not explicitly conflict with her view. Maybe I'll discuss that later.
I may also draw attention to another chapter in NE (can you guess which?)--not considered by Meyer or listed in her Index locurum--that also can present the very misleading impression that her thesis is false.
(*) After writing this post, I discovered that, in parenthetical remarks in her discussion of NE later in the article, Meyer twice cites these texts (along with several others). But she gives conflicting explanations. In one parenthetical remark, she cites the EE passages as one among several passages that really are talking about actions, even though the language is such that it "it might equally well refer to states of character" (pp. 151-2). In the other, she cites those passages as ones in which Aristotle is not asserting in that we are responsible for being virtuous or responsible for being vicious, but rather merely putting forward a biconditional claim, that we are responsible for being virtuous if and only if we are responsible for being vicious (p. 154).