I

Aristotle claims in the Metaphysics that in order to be resourceful ([phrase omitted]) in first philosophic inquiry it is useful to go through perplexity well ([phrase omitted]) (1) In the following essay, my focus will be on the role that going through perplexity well plays in practical inquiry. The specific claim I seek to defend, or at least make initially plausible, is that the right sort of perplexity is necessary not only for the dialectical inquiry concerning practical wisdom Aristotle undertakes in the Nicomachean Ethics, but also for the practical deliberative inquiry he explores and describes in that work. Ethical deliberation, on Aristotle's account, is a kind of inquiry, and the intellectual virtue of good deliberation ([phrase omitted]) a particular kind of rightness ([phrase omitted]) in inquiry. What Aristotle's discussion of good deliberation in Nicomachean Ethics 6.9 makes clear, however, is that while the kind of rightness exemplified in good deliberation is a kind of rightness in thinking, it cannot be equated either with rightness of opinion or with rightness of scientific knowledge ([phrase omitted]) Genuine deliberative thought ([phrase omitted]) as opposed to both opinion and scientific knowledge, is "not yet an assertion" ([phrase omitted]) so the rightness involved cannot be identified with arriving at the right account ([phrase omitted]) of what one is to do. It is not to be found in the answer to the question about which one deliberates. The rightness involved, I will claim, is rather to be found in an achieved clarity about the fundamental ethical question about which one deliberates. Moreover, I will claim that it is precisely in and through going through perplexity well that one achieves clarity about the question.

More generally stated, I maintain that going through perplexity about fundamental ethical questions is an essential condition of virtue in the "authoritative" sense ([phrase omitted]) An investigation of the precise character and extent of deliberative perplexity, however, stands outside the scope of the present essay. My goal, instead, will be to show the role marked out for deliberative perplexity in the Nicomachean Ethics. I will do so by offering an interpretation of the relation between Aristotle's account of akratic ignorance in Nicomachean Ethics 7 and the emphasis at the beginning of book 7 on the necessity of going through perplexity ([phrase omitted]) when inquiring into akrasia. Along the way, I hope to shed some additional light on Aristotle's conception of endoxa, his account of the so-called practical syllogism, and the distinction between ethical virtue simply and "authoritative" virtue. But my intention throughout will be to examine the role that perplexity about the phenomena of ethical life plays in the kind of thoughtfulness required for excellence of character.

II

At the beginning of Nicomachean Ethics 7, Aristotle describes the way one must proceed when inquiring into the problem of akrasia. The broadest question that orients his discussion is this: In what way are human actions that are motivated by the desire for pleasure and which contravene our prior ethical commitments morally blameworthy? In order to investigate this question well, Aristotle claims, one must make evident or clarify what he calls endoxa (usually translated as "common" or "reputable opinions") about self-restraint and lack of self-restraint, either all of these endoxa or at least the most authoritative ([phrase omitted]) of them. One does this by "setting out the phenomena" and "first puzzling through" them.

   A. Just as in the other cases, by setting out the phenomena [phrase
   omitted] and first going through perplexity ([phrase omitted]) one
   must in this way bring to light ([phrase omitted]) in the best case
   all of the endoxa about these experiences, or if not, the majority
   and most authoritative of them: for if we can unravel [phrase
   omitted] the difficulties ([phrase omitted]) and hand down the
   endoxa, we will have shown enough. (2)

The subsequent arguments in book 7 suggest that the phenomena that Aristotle seeks to set out are complexly interrelated with the "reputable opinions" about pleasure and pain, knowledge and ignorance, which his discussion brings to light. Indeed, it has become common, in the half century since G. E. L. Owen's influential 1961 article, "Tithenai ta phainomena," for commentators (1) to identify phainomena and endoxa in dialectical contexts, and (2) to understand both the relevant phainomena and the endoxa primarily as isolable, truth-evaluable propositions that have been or would be assented to by the many, the wise, or "the most well-known and reputable ([phrase omitted]) among the wise. (3) While there has been a good deal of dispute in recent commentary concerning the precise scope and aims of the endoxic method referred to in this well-known passage, there is broad agreement about the propositional character of the endoxa themselves. (4)

The dominant approach to endoxa, however, runs the risk of missing something fundamental about Aristotle's setting out of the phainomena concerning human beings' relation to pleasure. As indicated by both extremes, as it were, of his treatment of common opinions about pleasure, Aristotle's account of the relevant phenomena is not restricted to the words people use. Instead, he directs our attention to the lives people lead when we are seeking to ascertain their suppositions about happiness, (5) and points us toward the harmony or disharmony between an individual's deeds and his words as the credible perceptible basis for discerning that about which someone has conviction. (6) These passages indicate that the modes of opinion that Aristotle brings to light by puzzling through the relevant human phenomena can be in tension with the words people use. Indeed, as Aristotle claims near the very end of the work, "practical truth is judged ([phrase omitted]) from deeds and lives, since these are what are authoritative." Truth in matters relating to action is discovered, for Aristotle, by examining things that have been said, not in the abstract, but with reference to deeds and lives. "If they harmonize with deeds," Aristotle claims, "we should accept them, but if they are discordant, we should suppose them to be just words." (7)

A narrowly propositional construal of endoxa, moreover, is difficult to reconcile with the distinction Aristotle draws between real and merely apparent endoxa in his account of eristic deduction. (8) According to that account, a genuine endoxon is distinguished from an apparent endoxon in that the phantasia, the imaginative presentation in virtue of which one might be moved to action or assent, is not superficially manifest ([phrase omitted]) in a genuine endoxon. As Kurt Pritzl puts it, "genuine endoxa carry an import or meaning, a 'why and wherefore', that is not given in their own presentation. The why and how can be sought. The merely apparent endoxon, however, is not a real endoxon because it lacks this depth." (9) This feature of a genuine endoxon, what we might call its zetetic depth, obviously cannot be identified with the truth or falsity of a given propositional assertion, or with how broadly shared is the disposition to assent to that assertion, or with the sincerity with which that assertion would be assented to. Trivial truths, for example, would not qualify as genuine endoxa despite the fact that the many, the wise and the most endoxos among the wise would all assent to their propositional expression. (10)

What Aristotle means by "superficiality" and the significance of its contrary, zetetic depth, is further illuminated by the discussion in Rhetoric 3.10 of the contrast between "well-regarded" ([phrase omitted]) and superficial enthymemes. Aristotle indicates that the key difference is that the former but not the latter kind of enthymeme requires, and hence provides the occasion for, thought.

   B. Thus neither are superficial ([phrase omitted]) enthymemes well
   regarded ([phrase omitted])--for by 'superficial' we mean those that
   are altogether clear, and about which there is no need to inquire
   ([phrase omitted])--nor are those [well-regarded] we fail to
   understand when said, but those which when they are said at once
   ([phrase omitted]) understanding ([phrase omitted]) comes to be, or
   if it was not already existing ([phrase omitted]) thinking [phrase
   omitted] comes soon after. (11)

The point is elaborated by further comments in Rhetoric 3.10 and 3.11 about avoiding superficiality in metaphor. Aristotle claims that "well-regarded enthymenes" should be expressed neither with a metaphor that is wholly alien, which is difficult to comprehend ([phrase omitted]) nor with a superficial metaphor, which produces no experience ([phrase omitted]) in the listener. (12) A good metaphor brings a situation to life by making it active or actual ([phrase omitted]) and "bringing (it) before the eyes" ([phrase omitted]) and it does so by revealing the unexpected in the apparently familiar. This is best achieved, Aristotle claims, when the hearer is initially misled, because what one has learned is clearer when it goes contrary to what one has held to be the case. When this occurs, "it is like the soul saying 'How true; then I was wrong.'" (13)

Note that in passage [B], while Aristotle rejects superficial enthymemes on the ground that "there is no need to inquire into them," the conception of inquiry he is invoking here cannot be restricted to cases where the hearer explicitly articulates questions about an enthymeme or a metaphor. For he claims that in cases where knowledge was in some sense already present, understanding ([phrase omitted]) can occur as soon as ([phrase omitted]) the enthymeme is expressed. What is essential to inquiry in these latter contexts is, rather, the need for active thinking to disclose the significance of something that one could miss or mistake. This active thinking involves the possibility of experiencing, of "bringing before the eyes," some thought content that we are already in some sense in possession of but nonetheless could fail to appreciate or apprehend. And, it seems, were this not possible, neither would it be possible for us to inquire about it in the requisite sense. "Well-regarded" enthymemes and apt metaphors, like good jokes, are things we have to get, and they could not be that sort of thing were they not the sort of thing we could also miss. And this will be true even if, as it were, we already got the joke before we were told it. I maintain that this interdependence between, on the one hand, the possibility of success or failure in apprehending the significance of some thought content and, on the other, the possibility of active thinking of that content also characterizes genuine endoxa. In the context of this interdependence, I believe, we are in a position to understand Aristotle's conception of the complex hermeneutic work that must be done to bring the endoxa to light. (14)

The endoxa, I maintain, should not be understood as isolated practical or theoretical positions or propositions that one could endorse or affirm (or as dispositions on the part of the agent to endorse or affirm such propositions) but rather as phenomenally concrete modes of human orientation to a shared practical world. (15) More specifically, the endoxa should be understood as lived responses to certain fundamental philosophic and ethical questions. Aristotle's dialectical method, I maintain, centrally involves making the endoxa cognitively and experientially salient as a means of disclosing those fundamental philosophic and ethical questions. (16) These questions come to light as dilemmas, genuinely contradictory pairs of assertions, about which one can be perplexed. (17) The ultimate goal of dialectical inquiry is disclosing the relevant pragmata to the thinker who has gone through these perplexities. Alternatively expressed, in dialectical inquiry, one seeks to grasp intellectively the phenomena of human experience by seeing how that experience can genuinely be put into question. One does that by reanimating ways of questioning that experience, by seeing the endoxa in the light of and as shedding light on those ways of questioning. For this reason Aristotle claims that dialectic is an art of examination that is the way to the principles (or starting points) of all ways of inquiring. (18)

It is, I have claimed, an error to take endoxa as propositional assertions isolated from the lived context of questioning to which the endoxa are responses. To do so is, in a sense, like considering a punch line in isolation from the rest of the joke. This error is not, however, merely imposed upon the phenomena from without by ancient or modern scholarship. It is, rather, a danger inherent in political life as such. Indeed, Aristotle makes this clear through a striking simile. In his treatment of akrasia "of the spirit" ([phrase omitted]) he writes:

   C. For it is as if spiritedness seems to hear logos in some way,
   but to mishear it, like swift servants who run off before they hear
   what is said in its entirety and then err in carrying out the
   command. (19)

On the interpretation I offer here, Aristotle's dialectical analysis of the ethical failing of akrasia is in many ways about this danger. The akratic agent, according to Aristotle, is one who has acquired through habituation an orthos logos, a "correct account" of or a "right reason" for, what he ought to do or refrain from doing. In the moment of akratic weakness, these two sides, as it were, of the orthos logos come apart in the experience of the akratic agent. He comes to experience the account he has of the ethically virtuous human life merely as an authoritative pronouncement or command to which he has assented, but he ceases to experience it as a way of understanding his concrete practical situation. In contemporary philosophic terminology, the orthos logos goes from being an "internal" reason to being an "external reason" for him to act or refrain from acting in the prescribed or proscribed way. (20)

Aristotle's account of the possibility of akratic weakness is, I believe, of interest as an analysis of one of the central phenomena of ethical life. My focus here, however, is primarily on the light it sheds on the possibility of the contrary phenomenon of authoritative virtue. By exploring what happens, on Aristotle's analysis, when an agent loses his grasp on his habitually acquired account of a good human life, I hope to understand better Aristotle's account of the possibility of grasping it, and making it one's own, in the first place. That is, I want to understand Aristotle's account of how, when someone "acquires intellect" ([phrase omitted]) a state that was merely habitual or natural virtue can become authoritative ([phrase omitted]) virtue. (21) 1 want to show, moreover, the role that perplexity plays in this possibility. To do so, we will first need to understand how Aristotle's account of prior perplexity in Metaphysics 3 can apply to practical as well as theoretical inquiry. To this account we now turn.

III

At the beginning of Metaphysics 3, Aristotle makes a series of striking claims about the necessary role that aporia plays in resourceful inquiry. The passage is complex, and as that complexity will be significant for what follows, I quote it at length.

   D. It is necessary in relation to the sought after knowledge for us
   to first confront those things about which it is first necessary to
   be perplexed.... For those who wish to be resourceful [phrase
   omitted] [in inquiry] it is useful to go through perplexity well
   ([phrase omitted]) for the later resourcefulness is the release
   ([phrase omitted]) from that about which one is formerly perplexed
   ([phrase omitted]) and it is not possible to undo ([phrase omitted])
   the bond of which one is ignorant. But the perplexity of thought
   makes this clear ([phrase omitted]) about the area of concern [phrase
   omitted] for by means of that by which one is perplexed one
   experiences something similar to those who are in bonds; for in
   both cases it is impossible to go forward. This is why one must
   have considered all the difficulties ([phrase omitted]) beforehand,
   both for these reasons and also because those who inquire without
   going through perplexity first are like those who are ignorant of
   where they need to walk, and in addition to that, because he does
   not even know whether he has discovered what he is looking for or
   not known it. For the end is not clear ([phrase omitted]) to such a
   person, but it is clear to him who has been perplexed [phrase
   omitted] (22)

The first thing we should notice about this well-known account of perplexity is that Aristotle does not merely assert that insofar as one happens to be perplexed about an area of inquiry, one must go through or examine that perplexity if one wants to make progress. Rather, his argument is that if one wants to be resourceful in an inquiry, one must have first been perplexed. (23) Now, this claim is difficult to understand, since one might think that perplexity is precisely what we hope to avoid in inquiry. Indeed, one might reasonably think that perplexity is a sign of at least a local failure in our inquiry. It is, in fact, easy to read the above passage as identifying aporia with such a local failure, for Aristotle might seem to be saying that being in a state of perplexity is like being tied up in knots. While it is easy to see why it would be useful to have a look at a knot if we were tied up, we might think it would be better to avoid getting tied up in the first place. Why, therefore, does Aristotle claim that one must first have been perplexed if one wants to be resourceful in inquiry?

Our first step in answering this question is recognizing that Aristotle does not in fact identify the perplexity of thinking ([phrase omitted]) with that which ties or binds the inquirer in this passage, nor does he identify being perplexed with being tied or bound. What binds the inquirer, rather, is a problem or difficulty ([phrase omitted]) in some area of concern ([phrase omitted]) Aristotle claims that this difficulty or area of concern is something about which one is, or can be, perplexed ([phrase omitted]) but that is not the same as saying that the perplexity is the difficulty. Aristotle draws precisely this sort of distinction between perplexity and that about which one is perplexed in Topics 6.6, where he offers an objection to those thinkers who define aporiai as "equality in opposite reasonings." Such a definition of aporia, Aristotle claims, fails to distinguish between that which produces an aporia and the aporia produced. In this passage, Aristotle indicates that an aporia should be understood as the recognition on the part of an inquirer of an apparent equality of reasons on either side of a genuine philosophic problem, "for it is when we reason on both sides and it appears to us that everything can come about either way, that we are in a state of aporia about which of the two ways to take up." (24)

Indeed, in both this passage [D] from the Metaphysics and the passage [A] from Nicomachean Ethics 7 with which we began, rather than being another name for the difficulty itself, going through perplexity well is presented as the achieving of a kind of clarity with regard to an otherwise obscure area of concern. In Nicomachean Ethics 7.1, going through perplexity is how one makes manifest the endoxa about the experience of akrasia; in Metaphysics 3.1, perplexity makes clear the bond ([phrase omitted]) that ties one down. Moreover, Aristotle also claims in the latter passage that one cannot untie a knot of which one is ignorant. This suggests that, far from being identified with being bound, prior perplexity is a necessary condition of freedom from bondage.

Aristotle claims that it is only insofar as one has already been perplexed that one is in a position to see the telos, the end one seeks. Moreover, despite the fact that this claim about inquiry arises in the particular context of clarifying the role of aporiai in "the knowledge sought after" in the Metaphysics (presumably knowledge of being), Aristotle does not obviously restrict his claim about inquiry to that context. It is not merely misguided inquirers about being qua being, but those who inquire without going through perplexity per se, of whom it is said that the end is unclear to them.

Now, not every question or series of questions constitutes an inquiry ([phrase omitted]) for Aristotle. In inquiry, as Aristotle indicates in Posterior Analytics 2, we seek the middle term ([phrase omitted]) that is the explanatory ground ([phrase omitted]) of something about which we can have knowledge. For the middle term to be explanatory, it must allow us to see the connection between the fact that something is the case ([phrase omitted]) and why it is as it is ([phrase omitted]) Asking someone for directions to Larissa, for example, or looking up the word "virtue" in the dictionary is not an inquiry in Aristotle's sense, because such a query seeks only the "that" and does not seek the "why." Inquiry is impossible, moreover, in cases where the relation between the "that" and the "why" is already evident. If we wanted to know why we were unable to make progress, and saw that we were chained to the ground, there would be nothing further to inquire about--at least, that is, insofar as we saw our being chained to the ground in the right way. If we saw our inability to make progress as a case of being bound, in seeing it as a case of being bound we would know why we were unable to make progress. When we inquire, therefore, we seek for a not yet evident middle term ([phrase omitted]) that makes manifest the relation between the "that" and the "why."

The foregoing account of inquiry, however, is subject to one very significant complication. While Aristotle describes all inquiry as seeking a middle term, he also makes clear that the explanatory ground ([phrase omitted]) we seek can either be the same as what it explains or something different from it ([phrase omitted]). (25) In the latter case, we take the subject of our inquiry as a complex particular and inquire "in respect to the part" ([phrase omitted]) In this case, the object of our search is why that complex particular is as it is. In the former case, we take the subject of our inquiry as some unitary being and undertake our inquiry simpliciter ([phrase omitted]) In this case, the object of our search is to know what it is ([phrase omitted]) to be that thing, and if we grasp what it is, we will also understand "the why." Insofar as we seek to know what something is simpliciter, moreover, our inquiry is ultimately directed toward grasping the essence or "what it was to be" of that about which we inquire. (26)

With this we come to our crucial qualification of Aristotle's claim that every inquiry seeks the middle term. For, on Aristotle's account, the "what it was to be" of an entity is that very entity understood as an immediate and indemonstrable explanatory ground and principle. (27) Thus, for some inquiries at least, those directed toward understanding entities in terms of their principles, our search for a middle term is directed not toward something separate from and alongside that entity, but rather toward seeing that very entity as an immediate explanatory ground. This explanatory ground is immediate ([phrase omitted]) in two senses. First, it is immediate in the sense that it cannot be articulated into distinct terms or parts requiring further mediation. Second, it is immediate in the sense that it brings together into one thought the aspects of our inquiry that required mediation, the "that" and the "why." The essence of an entity, then, is in one sense a middle term that is another sense immediate; in our intellectual apprehension of the "what it was to be," we see what something is as why it is as it is.

An inquiry, therefore, seeks an explanatory middle term, which can be either demonstrable or indemonstrable, and either distinct from that which it explains or another way of conceiving that very thing.

IV

Even with an appropriately delimited conception of inquiry, it is not easy to understand the claim that prior perplexity is a necessary condition of resourceful inquiry in all cases of inquiry, nor will I seek to defend that claim here. Instead, I will try to show why prior perplexity should be understood as a necessary condition for excellence in ethical deliberation as Aristotle describes it in Nicomachean Ethics 6. (28) Deliberation, Aristotle claims, is a kind of inquiry. More specifically, Aristotle says, in Nicomachean Ethics 3.3, that deliberation is an inquiry and calculation not about ends but about things in their relation to ends ([phrase omitted]) and, in Nicomachean Ethics 6.9, that the virtue of good deliberation ([phrase omitted]) is "rightness, in accord with what is advantageous, toward the end ([phrase omitted]) about which end practical wisdom is a true conviction." (29) The change in number from ends ([phrase omitted]) to end ([phrase omitted]) reflects the fact that in the former context Aristotle is speaking about ends internal to particular practices such as medicine, oratory, and political rule, (30) while in the latter about the unqualified end ([phrase omitted]) of acting well ([phrase omitted]) as such. (31) The person of practical wisdom ([phrase omitted]) is one who is able to deliberate and calculate finely about things in relation to an ethically serious end ([phrase omitted]) and in relation to the good life as a whole ([phrase omitted]). (32) 1 will argue that, on Aristotle's account, prior perplexity is necessary for this kind of excellent ethical deliberation just as it is for genuine first philosophic inquiry and for precisely the same reason. The end in relation to which one has undertaken the inquiry is unclear to the person who has not been perplexed, but it is clear to him who has gone through perplexity well.

One might object, at this point, that in bringing together what Aristotle says about the telos of an inquiry in the Metaphysics with what he says about the telos of good human action in the Nicomachean Ethics I am guilty of a simple conflation of two distinct senses of telos in Aristotle's work. That is, so the objection would go, I have conflated the notion of the terminus of a theoretical inquiry with the goal or purpose of ethical action. Why this is not the case will become clearer as we proceed, but in advance of that argument this much can be said. On Aristotle's account, the terminus of excellent deliberative inquiry is an action that is an instance of acting well ([phrase omitted]). (33) Acting well is "an end, and one's longing is for this end." (34) Acting well is, moreover, a final end, in the sense that it has no further end beyond itself. (35) Thus, when one exemplifies the intellectual virtue of good deliberation as Aristotle describes it in Nicomachean Ethics 6, the action that one seeks in one's inquiry is an end in both of the above senses. To be clear about the end of the inquiry is to be clear about that for the sake of which one acts.

In order to understand Aristotle's account of good deliberation, we must see that the crucial question is not whether the good deliberator knows what he should do, but rather in what sense he knows it. Good deliberation is, on Aristotle's account, the inquiry whereby the good deliberator comes to know in one sense that which he already knows in another. Insofar as he has been brought up in fine habits, and has come to be pleased and pained at what he ought, in the ways that he ought, the good deliberator knows and possesses in one sense what it is to act well. What he achieves through excellent deliberative inquiry is the active contemplation of what this means for him as an agent in the concrete ethical situation he finds himself. Since, however, practical wisdom is knowledge of action in the concrete ethical situation, or as Aristotle puts it, the ultimate particular thing, (36) the good deliberator comes to know what he ought to do by seeing this action as what he ought to do. Alternatively expressed, he comes to understand why one ought to act in such and such a way through a certain kind of perception, (37) by seeing this as what he ought to do. (38) Deliberative ethical inquiry is, therefore, that inquiry through which the good deliberator comes to see his own ethical agency--in Aristotle's terms, his choice ([phrase omitted])--as the explanatory middle term between the right account he possesses through habituation and his perception of his concrete ethical situation. Deliberation is the inquiry that brings together under one perception the general premise "such and such a [person] ought to do such and such a thing" with the particular premise "here and now is such a thing and I am such a [person]," and it does so by bringing to light the sort of person the deliberator has chosen to be. 39 What defines the good deliberator is not whether or not he actively considers concrete alternatives to a specific course of action. (40) It is not, for most of us or for Aristotle, a mark of good ethical thinking that one weigh all the options before helping a friend or resisting injustice. (41) What defines the good deliberator is not only that he does what he ought to do, but also that his obligation is something to which he has chosen to be ethically responsive. Alternatively stated, while good deliberation does not require in every case actively taking up whether or not this is the right thing to do, it does require that the good deliberator see why it is the right thing to do, and that requires having actively taken up the question what the right thing to do is. But he can have taken this question up only insofar as this question is, or has been, a genuine question for him.

Now, for many readers familiar with the Nicomachean Ethics, the claim that the virtue of good deliberation requires seeing why some course of action exemplifies acting well might seem directly contradicted by Aristotle's explicit claims about the appropriate audience for works like the Nicomachean Ethics. Does not Aristotle claim that the well-educated members of his audience, those brought up in fine habits, are in possession of the "that" and hence have no need of the "why"? Not exactly. What Aristotle writes is this:

   E. Perhaps it is necessary for us, at least, to begin from the
   things known to us. Hence he who will listen adequately to the
   noble things and the just things, and to the political things
   generally, must be brought up nobly by means of habituation. For
   the "that" is a principle ([phrase omitted]) and if this were
   sufficiently apparent ([phrase omitted]) there will be no need of the
   "why" in addition, and a person of the sort indicated has or would
   easily get hold of principles. (42)

Aristotle makes two claims here. First, he claims that a person who has received the right sort of habituation either has or can easily get the principles of good action, the "that." Second, he claims that if this were sufficiently manifest, then there will be no need of the "why." The key to understanding Aristotle's claim here, as I will demonstrate below, is recognizing that one can have acquired through habitual education a principle of good action that is nonetheless not "sufficiently apparent." The principle can become sufficiently apparent, moreover, only to the deliberative inquirer who has gone through perplexity. To see why this is so, however, we need to look at Aristotle's account of good deliberation in Nicomachean Ethics 6, and his account of a particular kind of failed deliberation, a deliberative inquiry that he describes as beginning from the right hypotheses and arriving at the right conclusion, but going through a false middle term.

V

As we have seen, Aristotle claims that we do not deliberate about ends but rather about what is toward the end ([phrase omitted]) and while most commentators and translators treat this phrase as equivalent to "the means toward the end," there is a broad consensus that this account does not commit him to a merely instrumental view of deliberation. (43) It is common in the literature to defend this interpretation by invoking a distinction between constitutive and instrumental means. The point of this distinction is to claim that Aristotle's account of deliberation can allow for noninstrumental deliberation about the constituent parts of the final end of human happiness or, as David Wiggins presents it, noninstrumental deliberation about the correct specification of that final end. If, to take one of Wiggins's examples, our deliberation concerns as a posited end an interesting holiday, noninstrumental deliberation about what is toward (pros) that end would be concerned with "what really qualifies as an adequate and practically realizable specification of what would satisfy this want." (44) While it is certainly correct to argue that Aristotle's ton pros to telos need not, indeed should not, be interpreted to refer to merely instrumental means, it is far from clear how invoking the notions of constituent means or correct specification clarifies the relevant opposition between instrumental and noninstrumental reasoning. (45) More significant, however, is the fact that this notion of deliberative inquiry as concerned with constituent means is very hard to reconcile with Aristotle's description of the specific kind of "rightness" ([phrase omitted]) involved in right deliberation in Nicomachean Ethics 6.9.

   F. Now, since rightness is manifold, it is clear that rightness of
   deliberation is not every kind of rightness; (i) for the person
   lacking self-restraint or the base person will happen on, as a
   result of calculation, whatever he sets before himself to know
   ([phrase omitted]) with the result that he will have deliberated
   rightly but nonetheless have gotten hold of something very bad. But
   to have deliberated well seems to be something good, for such
   rightness of deliberation is good deliberation, which is apt to
   happen on what is good, (ii) But it is possible to happen on this
   also as the result of a false syllogism; one can happen on what one
   ought to do but not through that which one ought, the middle term
   ([phrase omitted]) being false. So that this would not yet be good
   deliberation, according to which one happens on what one ought to
   do but not indeed through that which ([phrase omitted]) one ought.
   (46)

In this passage, Aristotle articulates two distinct ways in which one could fail to deliberate well. The first [F.i] is fairly straightforward. Here Aristotle distinguishes the sort of person who sets before himself the wrong conception of how he should act from the good deliberator, who sets before himself the right conception. In this case, Aristotle reasonably concludes that any deliberative process that reliably achieves something bad for the deliberator cannot be an example of the virtue of good deliberation. The second way that one could fail to deliberate well [F.ii] is more complex. Here Aristotle is not concerned with someone who has set before himself the wrong conception of what he ought to do. Rather, his concern is with the way in which someone might fail to deliberate well even if he "happens upon" what he ought to do. The failure diagnosed is one of arriving at what one ought to do through a false deliberative syllogism, specifically, through a false middle term. Aristotle concludes that one cannot deliberate well if one happens on what one ought to do through ([phrase omitted]) that which one ought not.

It is common, among those commentators who discuss the false syllogism, to take the phrase "through which" ([phrase omitted]) here, like ta pros elsewhere in the Nicomachean Ethics, to mean "the means" and to understand the false syllogism as one that advocates the right action but chooses the wrong means. But this, it seems, cannot be the right way to understand the false middle term. The supposition that one could do the right action by the wrong means is plainly inconsistent with Aristotle's conception of eupraxia. When one acts well, one acts virtuously, with all the attendant specifications of time, place, means and manner that Aristotle uses to discriminate virtuous action. If one chooses the wrong means to some otherwise choiceworthy result, one has, ex hypothesis not happened upon what one ought to do. (47) This will be the case whether we conceive of the means in question as instrumental means or constitutive means. It will be a fortiori the case if we conceive of the false syllogism as an incorrect specification of the constituent end. Moreover, the very characterization of the second type of failed deliberation as involving a false syllogism with a false middle term should indicate that Aristotle is not thinking about deliberation as involving constitutive means. Aristotle's description of the false deliberative syllogism in Nicomachean Ethics 6.9 exactly parallels his claim in Prior Analytics 57a29-30 that in theoretical syllogisms concerning particulars one can come to true conclusions through falsehoods ([phrase omitted]) This should strongly suggest that, whatever the differences between practical and theoretical inference, in the Nicomachean Ethics as in the Prior Analytics, the problem diagnosed concerns the character of the inference and not the correct or incorrect specification of the constituent end.

If, however, the falseness of the false syllogism cannot be due to an incorrect specification of constituent means, given the assumption that the deliberator hits upon what he ought to do, how are we to understand it? Ex hypothesi, the deliberator who hits upon what he ought to do through the false middle term is neither vicious nor akratic. If he hits upon what he ought to do through the practical syllogism, he must do what he ought to do, for the conclusion of a practical syllogism is an action. And if he does what he ought to do, he must not only perform the right deed, he must do so at the right time, in the right way, with the right means. He must, moreover, experience the appropriate pleasure and pain in relation to the deed. (48) He must, as Aristotle says of the courageous person, be pleased and pained at what he ought to be, in the way he ought to be, and to the extent that he ought to be. (49) He must, that is, in a certain way possess ethical virtue. What, then, is the difference between the excellent deliberator and the deliberator who proceeds through the false middle term?

This is the very question Aristotle raises in Nicomachean Ethics 6.12, a chapter devoted to the perplexity one might go through ([phrase omitted]) concerning how wisdom and practical wisdom could be of any use. With regard to practical wisdom in particular, the question takes the following form. If practical wisdom concerns things that are just, noble and good for a human being to do, and the ethical virtues are states or conditions ([phrase omitted]) to reliably do what we ought, how will we be better as agents (literally, how are we "more practical," ([phrase omitted]) by knowing these things? For if the virtues are states in the same way health is, such things are said to exist not because we do something, but by possessing the relevant hexis. (50)

Aristotle begins his response to this question by turning to the common endoxon that there are people who do the just things but are not yet just.

   G. For just as we say that some people who do just things [phrase
   omitted] are not yet just, the sort of people who do the things
   ordained under the laws ([phrase omitted]) but do so either
   involuntarily, through ignorance, or through something other and
   not through these things themselves ([phrase omitted]) (and yet they
   do what they ought to do, even so many things as it is necessary
   for the morally serious human being to do). (51)

Such people, or some of them, seem to correspond to the deliberator who falls short of deliberative excellence by hitting upon what he ought to do through the false middle term. They are neither akratic nor vicious; they do as much as ([phrase omitted]) the morally serious person does. But, to fully correspond to the type of failed deliberator we are concerned with, this person cannot be someone who acts involuntarily (for if he were, that would mean that he would not be acting virtuously in any sense), nor can he act through ignorance (for he knows, in some sense, what he ought to do). There is, however, another possibility. He could be doing just deeds ([phrase omitted]) not through or because of ignorance ([phrase omitted]) but in ignorance ([phrase omitted]) nonetheless.

Aristotle makes the distinction between doing something because of ignorance and acting in ignorance in his discussion of voluntary action in Nicomachean Ethics 3.1. Aristotle's examples of a person doing something because of ignorance range from a slip of the tongue to someone who acts under the supposition that "the pointed spear had been blunted." His privileged examples of acting in ignorance are of two kinds: the actions of a human being with a corrupt character and actions done in anger or drunkenness. When one does something because of ignorance, one does not act voluntarily and is not subject to praise or blame. When one acts in ignorance, to the contrary, one is held accountable for what one has done because one is held to be responsible for the state one is in. Let us bracket for now Aristotle's difficult account of the way in which one is responsible (or "coresponsible") for having acquired a corrupt character and focus only on the account of drunkenness. (52) In what sense, we can ask, is drunkenness a state of ignorance? The drunken man knows, we might think, what he ought to do and what he ought not to do. He hasn't, strictly speaking, forgotten about some prohibition. Instead, we might say that his inhibitions were lowered. He could still tell you what it is he's not supposed to do. He might even tell you that he knows he is not supposed to do it. It is, rather, that whatever internalized prohibition he has that would normally inhibit his action doesn't appear immediately relevant. He can't really see, in the moment, how the prohibition applies to him.

As we will see below, this is how we should understand Aristotle's account of the akratic agent, whose knowledge and ignorance are explicitly likened to those of a drunken man. Before turning to that discussion, however, we should note the following. Aristotle in his account of voluntary and involuntary action focuses on people who in ignorance, but not because of ignorance, do what they shouldn't do. He does not discuss but obviously leaves open the possibility of people who, in ignorance, do what they should. These ignorant right-doers would presumably not be drunkards, but might be those who are in another way "ignorant through carelessness ([phrase omitted])." (53) It is precisely such an individual who could approximate in his deliberation the kind of false syllogism described in [F] above. The ignorant right-doer would hit upon in his deliberation what he ought to do, indeed all the things he ought to do. He would do the very things the morally serious ([phrase omitted]) human being does. He would do "the things ordained under the laws" ([phrase omitted]) which means, as Aristotle indicates with his only other use of this phrase in the corpus, he would do "the just things ordained under the law with reference to the whole of virtue." (54) But he would do so in ignorance, despite possessing the right account of what to do.

What would such a human being look like? Well, in one sense, he would look just like the virtuous individual. But, unlike the person who has virtue in the authoritative sense, his virtues would be merely habitual. Now, it is important not to misunderstand this claim, and this claim is liable to be misunderstood in two distinct ways.

First, to say that someone's virtues were merely habitual is not to say that he acts mindlessly. That is, it does not mean that he would be acting robotically; he would feel pleased and pained at the right things, at the right time, and to the right extent. Nor would his emotional reactions, as John McDowell emphasizes, be the result of brute tendencies, "like a trained animal's behavioral dispositions." (55) They would, rather, be more like responses to internalized paternal warnings, censures, and exhortations. (56) But, Aristotle indicates, one can listen to such admonitions more or less rationally. In Nicomachean Ethics 1.4, Aristotle draws an opposition between a person who can "listen adequately ([phrase omitted]) to the noble things and the just things" and another person who is incapable of such listening, but can nonetheless listen to Hesiod's exhortation to "obey one who speaks well." (57) This opposition between two modes of listening is then carried over into Aristotle's initial division of the soul into rational and nonrational parts. In 1.13 he claims that the nonrational part of the soul is twofold ([phrase omitted]) with the desiring part of it somehow participating in reason because it can listen to and heed the command of reason. (58) He apparently supports this claim by noting that this is the way we say ([phrase omitted]) that the person who listens to and heeds the commands of his father and friends possesses reason ([phrase omitted]). (57) It would seem, however, that possessing reason simply by obeying one's father's commands is to be reasonable in a fairly attenuated sense. Aristotle further argues that if the appetitive part possesses reason, then "that which possesses reason will be twofold as well: what possesses it in the authoritative ([phrase omitted]) sense and in itself ([phrase omitted]) on the one hand, and, on the other, what has it in the sense of being apt to listen as one does to one's father."'" Significantly, Aristotle's explication of these two twofold divisions in the soul leaves it unclear how we are to think of a desiring part of the soul understood as belonging to putatively distinct rational and nonrational parts of the soul. (61) But whether we think of the desiring part of the soul as one thing, two things, or somehow indeterminately double, (62) Aristotle's account raises the possibility of a kind of listening to that which has been ordained by law or custom that is in a certain sense unthinking or, as he suggests, careless. The person who was unthinkingly obedient would be someone who, as stated in [G], does the just things not because of the just things themselves ([phrase omitted]) but because of something else ([phrase omitted]) That something else or "other" would be the authority of the laws or rules to which one was obedient. In such a case, it would be the authority of the laws, rather than their justice, that one was heeding. And while such obedience is not simply mindless, neither, we could say, is it mindful. We can call it unthinking obedience.

The second way that the distinction between merely habitual and genuine or authoritative virtue could be misunderstood is if one thought of the distinction between mindfulness, on the one hand, and unthinking obedience, on the other, in terms of the presence in one case and the absence in the other of an explicit, discursive, or thematic account about what one is or is not to do. But this would be to mistake the crucial issue. The question, again, is not whether the person who unthinkingly obeys the dictates of reason, law, or custom possesses the right account of good action. He does so by hypothesis. The question is in what sense he knows it. As we will see by turning to Aristotle's account of akrasia, whatever explicit account one is or can be in possession of is not sufficient for practical wisdom and virtue in the authoritative sense.

VI

Aristotle distinguishes two possible ways that an akratic agent could act in contravention of the account he possesses of right action. He could fail to attain the final premise of the practical syllogism; that is, he could fail to achieve an adequate sense of what that right account calls for him to do in the moment. Alternatively, he could attain the final premise, but attain it "in such a way that his having it does not amount to his knowing it." (63) As David Charles has argued, the latter sort of akratic is able to rehearse all the arguments against doing what he does. (64) Such an akratic agent, we can suppose, insofar as he attains the final premise, reflectively endorses the prohibition his action contravenes. Yet, he performs the action nonetheless. It is this type of akratic agent, a type we can call the reflective akratic, that will interest us here.

In the initial chapters of Nicomachean Ethics 7, Aristotle goes through a series of aporiai concerning whether the akratic agent knows what he ought to do or is ignorant of it. This question becomes reformulated as a question concerning the sense in which the akratic agent could possess the knowledge of what he should do but nonetheless not use it. In his answer to this question, Aristotle makes a distinction between two different senses of "to have." He writes:

   H. For in the case of having but not using science, we see [phrase
   omitted] that the 'having' ([phrase omitted]) is different, such that
   a person both has it ([phrase omitted]) in a way and does not have it
   ([phrase omitted]) just as someone who is asleep, mad, or drunk. (65)

The reflective akratic possesses, in one sense, the right account of acting well, the final end of human action. When he acts akratically, however, he does so without the active contemplation of this end. He acts, we might say, not mindlessly but thoughtlessly. This thoughtlessness is not mere inattention. It is not that, as it were, the right account merely slipped his mind. Nor is it the case that he comes to reject the final premise as false. But there is something about him that thwarts his ability to see clearly, in the moment, the way that this act is not good for him. He possesses the right account, Aristotle suggests, in something like the way that a man might drunkenly recall some mnemonic device from childhood without being able to recall what it was supposed to help him remember. It is his inability to actively contemplate the knowledge or opinion of the end he possesses through ethical habituation that renders the akratic agent susceptible to doing that which he, in some sense, "knows" he should not do.

The character of the akratic's thoughtlessness is brought out by a nice ambiguity in Aristotle's first statement of the Socratic paradox denying that akrasia properly speaking exists.

   I. This account, then, contends against the phenomena that come
   plainly to sight ([phrase omitted]) and one must investigate ([phrase
   omitted]) if in fact this experience ([phrase omitted]) occurs
   through ignorance, what the character of the ignorance is. For it is
   manifest that before he comes to be in the experience [phrase
   omitted] a person who lacks self-restraint does not think [phrase
   omitted] at least, [that he ought to act as he then proceeds to
   act]. (65) 66 65

The bracketed phrase in the translation supplies in this context the most reasonable complement, lacking in the Greek, for the verb oiomai--"to think," "to believe," or "to consider." The akratic agent does not believe that he is doing what he believes he should be doing. But when he acts akratically, he does not in some sense really believe, but merely recites, the principle his action contravenes. He is, at that moment at least, like a dissolute actor playing the role of a stoic sage. This fact, however, escapes him.

We see this gesture repeated at 1146b23-4 and again at 1152a5-6. In both passages, Aristotle contrasts the akratic agent with the licentious person, who takes it as a maxim that he should always pursue present pleasure.

   J. For the vicious person is led on by what he chooses, maintaining
   that he ought always to pursue the present pleasure, whereas the
   person lacking self-restraint does not think [that he ought to],
   but pursues [the pleasure anyway].

The italicized portion of [J] translates [phrase omitted] which in context is most naturally understood to mean that the akratic does not think that he should always pursue present pleasure, but he pursues it nonetheless. But all that Aristotle explicitly says here is that the akratic "does not think, rather he pursues." Aristotle's indication of the thoughtlessness of the akratic is clearer still at 1152a4-6.

   K. Those lacking self-restraint and the licentious are similar as
   well, though they are in fact different: both pursue the bodily
   pleasures, but the latter does so while considering he ought to,
   the former while not considering [[phrase omitted]].

The significance of Aristotle's repeated indirect implication that akrasia involves a failure "to believe" or "think" ([phrase omitted]) is seen when we recognize that Aristotle's primary use of the verb oiomai in the Nicomachean Ethics is in reports of endoxa, both those apparently relied upon in the account and those explicitly brought into question. When Aristotle responds to the Platonists who consider the good to be something separate, (67) or claims that "on the basis of the lives they lead, the many and crudest seem to believe, not unreasonably, that the good and happiness are pleasure," (68) or puzzles over what Socrates "used to think" about akrasia (69) or Eudoxus "used to think" about pleasure, (70) he uses a form of oiomai.

I have suggested, following Pritzl, that what distinguishes a genuine endoxon from a merely apparent endoxon is a kind of zetetic depth. In Pritzl's words, cited above, "genuine endoxa carry an import or meaning ... that is not given in their own presentation." It is precisely the lack of such depth that characterizes the akratic's possession of the right account of human action at the moment he acts akratically. As we have seen, the akratic we are considering reflectively endorses (in some sense) the prohibition his action contravenes. But his belief in the rightness of the prohibition, while in some sense sincere, is also in some sense abstract. We can provisionally express the abstract character of his belief as follows. He assents to a proposition or body of propositions the lived import of which he fails to grasp. When he says to himself that he knows he really shouldn't do what he proceeds to do, "he merely speaks, as a drunk man states the sayings of Empedocles." (71)

The claim that the akratic differs from the licentious person in that the latter thinks or believes he should pursue present pleasure while the former does not think is mirrored by corresponding claims about choice (7100alocols). Throughout Nicomachean Ethics 7, Aristotle claims that the licentious person chooses to pursue present pleasures and the akratic agent does not choose. (72) In fact, Aristotle never says the akratic agent chooses; he never uses the verb "to choose" ([phrase omitted]) positively when describing the akratic. Indeed, at 1150a25, akratic agents are referred to as "those who do not choose" ([phrase omitted]) Instead, akratic agents act contrary to choice ([phrase omitted]) or contrary to the right account in them ([phrase omitted]) This suggests a particular way of understanding the akratic's susceptibility to temptation. The akratic has acquired through habituation a conception of the final end of human action but has not made that conception his own. The choice he contravenes is not a choice he has made but a conception of the choiceworthy ([phrase omitted]) life he has received through his education. His choice has been chosen for him.

Aristotle's argument also suggests a crucial reason for the akratic's failure to choose. Akratic agents have acquired true opinion regarding the final end but consider ([phrase omitted]) themselves to have precise knowledge ([phrase omitted]) of that end. (73) In contrast to his possession of the right logos of what he ought to do, which he is not said to think, he does think that he knows what he ought to do. In contemporary terms, we could say that the akratic's thinking involves a kind of second-order belief falsely ascribing to himself first-order knowledge about the final end. The akratic fails to see that which, according to passage [H], we come to see by going through the puzzles raised in book 7. He fails to see "that the hexis is different." What that means in the broader context of the argument of book 7 is that he fails to see that what he takes to be precise knowledge of what he ought to do is only, at best, true opinion concerning that end, a true opinion that he has acquired through habituation but has not made his own. (74)

Alternatively expressed, the reflective akratic has been persuaded of the correctness of his habituation toward virtue. Aristotle indicates as much when he contrasts the akratic's commitment to the correct account in him with his pursuit of pleasure in contravention to that account: "though he has been persuaded of one thing, he does other things nonetheless." (75) As opposed to the licentious individual, who is persuaded to pursue pleasure, the akratic pursues pleasant things without having been persuaded to do so. Moreover, as we will see in a moment, the conflict in the akratic is experienced in precisely these terms, as occurring between an account of which he has been persuaded and his perception of something as an immediate source of pleasure. It is this that ultimately distinguishes the reflective akratic from the excellent deliberator. The akratic's experience of the correct account of human action, unlike that of the good deliberator, is essentially mediated by an authority other than the most authoritative part of himself. (76)

VII

In section IV above, I claimed that we should understand ethical deliberation as that inquiry through which the good deliberator comes to see his choice as the explanatory middle term between the right account he possesses through habituation and his perception of his concrete ethical situation. I also claimed that the middle term sought, the ethical choice of the excellent deliberator, is immediate in the sense that the good deliberator's choice is not merely a link in an inferential chain that connects a universal conception of the good human life with perceived aspects of the broader ethical context. It is, rather, an intellective perception (nous) (77) whereby the deliberator understands the universal, "such a [person] ought to do such a thing," precisely insofar as he perceives that "here and now is that sort of thing and I am that sort of [person]."

In both On the Soul 3.11, and On the Movement of Animals 7, Aristotle presents deliberative inference as a thinking together of a universal imperative concerning an agent's action--in On the Movement of Animals, that a human being must walk ([phrase omitted]) (78)--with a particular self-recognition on the part of the deliberating agent--in On the Movement of Animals, that oneself is a human being. But when Aristotle presents deliberative inference in Nicomachean Ethics 7.3, in the context of the question of the akratic's ignorance, the form is notably different. There he offers an oddly abbreviated practical syllogism.

   L. (N)othing prevents someone who holds both from acting contrary
   to scientific knowledge, making use of the universal premise but
   not the particular one ([phrase omitted]) for actions are
   particulars. The universal also differs, for it concerns himself
   ([phrase omitted]) and it concerns the fact ([phrase omitted]) for
   example, that dry things are advantageous for every human being and
   that he himself is a human being, or that this is diy. (79)

In contrast to his treatment of practical inference in De Anima or Metaphysics, here the syllogism contains neither any imperatival statement nor indeed any obvious reference to action. This odd facet of the passage is emphasized by the fact that the term I have translated as "fact," pragmatos, is primarily used in the Nicomachean Ethics to refer to things done. (80) But here it apparently refers to dry things ([phrase omitted]) usually understood to be dry foods. Instead of a universal imperative saying a human being should or must do something to or with dry things, however, we have the universal statement that dry things benefit every human being. This too is odd because Aristotle makes clear elsewhere that this claim, if expressed in this unqualified way, is simply false.

In De Sensu 444a16-17, Aristotle claims that food, whether dry or moist, is often unhealthy. Moreover, at Nicomachean Ethics 8.8 he indicates that neither the wet nor the dry is unqualifiedly good; rather "what is good for the dry is not to become wet but to come to the middle condition, and similarly in the case of heat and the rest." (81) Thus it would seem that "dry things are good for every human being" is at best an elliptical expression of the right logos, which if given fuller expression would be something more like "dry things are good for every human being, as long as that human being is not already dry." The question is why the deliberative inference here makes use of only half a premise, a question that becomes more pressing once we realize that the adjective "dry" ([phrase omitted]) was used to refer not only to dry foods but also to fasting or abstaining from all food. It was used to describe severe or frugal habits, and when so used it was opposed to the pleasant ([phrase omitted]) Moreover, Aristotle tells us in On the Soul that what is tasteable is wet, and that a tongue that is thoroughly dry ([phrase omitted]) will perceive nothing. (82) If this is the case, then the half-premise "dry things are good for every human being" is in great danger of becoming a counsel of physical and psychic deprivation.

Aristotle's description of spiritedness ([phrase omitted]) in his account of the metaphorical akrasia related to spirit is illuminating here. He says that spiritedness seems to hear reason in some way, "but to mishear it ([phrase omitted]), like swift servants who run off before they hear all of what is said and then go wrong in doing what is commanded." (83) This suggests the interpretive possibility that the elliptical formulation of the universal premise is Aristotle's representation, not of the right logos simply, but of the right logos as misheard by the akratic or the potential akratic. Moreover, if we return to Aristotle's treatment of the wet and the dry at Nicomachean Ethics 8.8, we receive further support for this interpretive possibility. There Aristotle suggests that an extreme condition (such as the wet) does not in itself ([phrase omitted]) aim at its opposite extreme (such as the dry) but does so only accidentally ([phrase omitted]) From such an extreme state, however, it may appear that the opposite extreme is what is desired, even though in truth "the desire ([phrase omitted]) is for the middle." It is with this possibility in mind, the possibility that under extreme conditions one can misconstrue the natural or necessary object of one's desire, that we should approach Aristotle's account of the explanatory ground of simple or unqualified ([phrase omitted]) akrasia considered with respect to nature ([phrase omitted]) in Nicomachean Ethics 7.

VIII

Simple akrasia, or akrasia in an unqualified sense, Aristotle claims, is a lack of self-control with respect to the pleasures related to bodily necessities ([phrase omitted]) that is, "the sorts of pleasures as those bound up with nourishment ([phrase omitted]) and the sexual need ([phrase omitted])." (84) These pleasures are the focus of his account of akratic conflict considered according to nature, an account that begins as follows.

   M. For example, i. if all the sweet must ([phrase omitted]) be
   tasted, and this thing here is sweet (it being some one particular
   thing), it is necessary that one capable and not prevented from
   doing so must do this at once ([phrase omitted]) Whenever, then, ii.
   the universal is present that hinders us from tasting [phrase
   omitted] and iii. another universal is also present, to the effect
   that all the sweet is pleasant, and this thing here is sweet and
   this premise is active ([phrase omitted]) and the desire happens to
   be present, the one premise says to avoid this ([phrase omitted]) but
   desire leads the way ([phrase omitted]) for it is able to set in
   motion each of the parts. (85)

The most obvious interpretive difficulty presented by this passage is that it is not easy to see how it is meant to be illustrative of akratic conflict. The first universal premise [M.i] that appears in the passage, and the only one that is expressed in terms of an obligation ([phrase omitted]) is the premise "all the sweet must be tasted." It is not clear how we are to understand this hypothetical universal obligation to taste the sweet, nor is it clear in what sense the obligation so expressed is meant to be intemperate. As A. W. Price notes, citing De Sensu 442al-2, Aristotle elsewhere identifies the sweet with that which nourishes an organism, and hence an apparent necessary condition for health. (86) The second universal that appears [M.ii], to the contrary, does not look nearly so reasonable. It is presented as a brute obstacle to tasting. This is strange since Aristotle makes clear earlier in the Nicomachean Ethics that taste plays little or no role in licentiousness and lack of self-restraint. (87) More significantly, later in the Nicomachean Ethics, Aristotle will claim that it is those who have not tasted ([phrase omitted]) of unmixed and liberal pleasure who take refuge in pleasures of the body. (88) He claims, moreover, that those who have not tasted ([phrase omitted]) of the noble and truly pleasant will have no idea ([phrase omitted]) of these pleasures. Such people, he says, do not naturally abstain from base pleasures but do so only out of fear of retribution. (89) Yet, the present universal hindering one from tasting ([phrase omitted]) is is referred to without any specification of who is not to taste, what they are not to taste, or why.

This unbounded and unspecified universal apparently comes into conflict in the akratic with another present universal [M.iii] that says that all the sweet is pleasant, and with the judgment this is sweet. It is unclear, however, how a universal preventing one from tasting can come into conflict with a universal affirmative proposition, much less with the particular perceptual judgment "this is sweet." No mention is made in [M.ii] or [M.iii] of what one ought or ought not do, of the agent or his action. The human being confronted with a choice seems to have been forgotten. Nor is it clear how either of the present universals relate to the universal premise [M.i] with which Aristotle begins. What are we to make of these strange aspects of Aristotle's account?

We are better positioned to understand what Aristotle is doing here once we recall that the perceptual judgment of something as sweet is one of Aristotle's chief examples in the Nicomachean Ethics of a judgment that is relative to the good or bad state of the perceiver. (90) The puzzling features of Aristotle's account make sense once we realize that Aristotle is not merely analyzing akratic conflict in this passage. He is depicting that conflict as it is experienced by the akratic in the moment of akratic irresolution. It is for the akratic that, in the conflict between a desire for a present pleasure and that in him that inhibits fulfilling his desire, the former appears as something actual ([phrase omitted]) while the latter appears as a blank prohibition telling him not to taste. It is the akratic who forgets himself, his action, and his choice; he does not make use of the universal concerning himself ([phrase omitted]) He is in a certain sense a spectator to the conflict within him. It is experienced as occurring between, on the one hand, a prohibition that speaks to the desiring part of his soul with the voice of a father saying "flee from this" ([phrase omitted]) and, on the other hand, a present desire that leads ([phrase omitted]) him.

Aristotle claims to have shown by his argument that one comes to be akratic by means of an account ([phrase omitted]) in a way ([phrase omitted]) and by means of an opinion that is not in itself ([phrase omitted]) but only accidentally ([phrase omitted]) opposed to right logos. The latter point is emphasized by repetition. While the desire is opposed to the right account, the opinion is not, (91) yet it is the opinion rather than the desire that Aristotle cites as the cause of one coming to be akratic. The difficulty here is understanding how an opinion that is not in itself contrary to right reason can be made accidentally opposed to right reason by desire. Yet that very difficulty indicates the defining characteristic of opinion in akratic mental conflict. Opinion in the reflective akratic is, or has become, separated from any animating desire. The prohibition the akratic's action contravenes appears to him in the moment of akratic conflict as an unexplained imperative, a proverbial dead letter. The dictates of right logos sound in the akratic's head like mere words, and hence are no longer action guiding. Right logos becomes mere opinion, a thinking that moves nothing. (92)

Insofar as the correct account of right action comes to be experienced by the akratic as merely an authoritative prohibition, desire comes to be experienced as opposed to deliberation rather than informed by it. For this reason the desire, even if it is directed toward a necessary and natural object, cannot be the correct longing that Aristotle claims is necessary for good practical thinking. (93) This is the sense that the desire for the sweet in akratic conflict is opposed to correct reason. It is not because it is incorrect to desire the truly sweet.

Aristotle presents the akratic as experiencing a conflict between two universals [M.ii and M.iii] that are not formally contradictory. They can become contradictory, however, if the universal statement [M.iii] "all the sweet is pleasant" takes on the form of an imperative commanding one to taste. Given the resources of the passage, it seems that it can take that form only if we understand it as an exegesis of the first universal premise [M.i] and a partial explanation of its obligatory character. "All the sweet must be tasted" is then seen as a specific mode of presentation of the premise "one ought to pursue the pleasant."

Aristotle maintains that whenever we have a desire for something, we perceive it as pleasant. (94) In On the Soul he claims that whenever we perceive something as pleasant we pursue it, and that pursuit is a sort of affirmation. (95) Moreover, in Nicomachean Ethics 7.13 he claims that the pursuit of pleasure by all living things is a sign ([phrase omitted]) that it is somehow ([phrase omitted]) best. (96) He first avers that "since neither the same nature nor the same hexis is or is opined to be the best, all do not pursue the same pleasure, though all do pursue pleasure." But he immediately qualifies this with the suggestion that "perhaps they pursue not the pleasure they suppose ([phrase omitted]) or would affirm ([phrase omitted]), but in fact the same pleasure, for all things by nature possess something divine." (97) The Nicomachean Ethics as a whole will arrive at the position that pleasure is so intimately bound up with a choiceworthy life that we are left with the question, posed but significantly left unanswered in the work, of whether we choose pleasure for the sake of living or choose life for the sake of pleasure. (98)

The akratic's failing, therefore, is not that he finds sweet things pleasant, nor is it that he pursues pleasure in contravention of an empty demand not to taste. His failing is, rather, that he experiences the conflict in himself in precisely these terms, as a struggle between an active perception of something as pleasant, on the one hand, and a brute imperative on the other. Whatever complex normative and affective orientation to a life well-lived he has acquired through his ethical habituation is reduced in the moment of akratic conflict to a bare command, a command he formally accepts as authoritative, but the content of which he cannot experience as motivating. Rather than reflecting his habitually acquired understanding of moderate action and desire, (99) the akratic experiences the right account he possesses in terms of decontextualised universal propositions like "Dry things (or 'austere practices,' ([phrase omitted]) benefit every human being." (100)

As we have seen, such bare propositional assertions are more plausibly understood as expressions of the conflict within the akratic agent than as expressions of the correct account of human action. But this does not mean that the akratic simply creates such ascetic dicta out of whole cloth. Instead, as Aristotle makes quite clear, the opinion that pleasure is "altogether base" ([phrase omitted]) and to be avoided as such is very much part of the fabric of contemporary ethical discourse. But he also indicates that this opinion is, as it were, merely an opinion and not a genuine endoxon.

   N. For some say that pleasure is the good, others that pleasure is,
   to the contrary, an altogether base thing--some of these latter
   perhaps are those that have been persuaded ([phrase omitted]) that
   pleasure is such in fact, the others are those who suppose [phrase
   omitted] it to be better with a view to our life to declare that
   pleasure is among the base things, even if it is not. For the many,
   they suppose, tend toward it and are in fact enslaved to pleasure.
   Hence one ought to lead them toward its contrary, since in this way
   they might arrive at the middle. (101)

Here Aristotle divides those who claim that pleasure is worthless into two groups: (1) those who think it better to say that pleasure is base, even though they don't actually believe it, and (2) those who are perhaps persuaded by such claims. No one, it seems, genuinely thinks or supposes that pleasure is base. For this reason, the claim that pleasure is altogether base has, on Aristotle's account, the opposite effect from that intended by its moralizing proponents. He writes: "Whenever such arguments are discordant with what is perceived, they are treated with contempt and undermine the truth," a statement that would seem to apply equally well to the role of such claims in akratic conflict as to their fate in public disputation. (102)

Aristotle's own account in the Nicomachean Ethics apparently begins by occupying a moralizing antihedonist position similar in its extremity to the one he denounces as truth undermining in book 10. In Nicomachean Ethics 1.5, he speaks of the life of enjoyment chosen by the majority of people as slavish and bovine. (103) While this does not directly contradict his later valorization of the contemplative life on the basis of the purity and stability of the pleasure inherent in it, (104) it is hard to reconcile with his claim that life and pleasure are so closely yoked together ([phrase omitted]) that it does not appear possible to separate our choosing to be active and alive from our choice of pleasure. (105) Moreover, a careful reading of the book 1 criticism of the life of enjoyment in the context of the work as a whole indicates that what is at issue is the way such a life is taken up rather than the centrality of pleasure in it.

Aristotle suggests that most people's devotion to that life is at least in part due to their being influenced by the example set by those in positions of authority ([phrase omitted]) Later we learn that most people wish to be honored by those with authority in the hope that they will be benefitted by them. (106) They seem in this way, Aristotle says, to choose honor not through itself ([phrase omitted]) but incidentally ([phrase omitted]) Thus it turns out that the many, who choose the life of pleasure, and the refined and active ([phrase omitted]) who choose honor, are more alike then they differ; they both choose honor in a sense. But, Aristotle avers, honor is more superficial ([phrase omitted]) than what we are seeking ([phrase omitted]) (107) a claim that we are now better able to understand. Neither the refined nor the majority of people seek their respective ends of activity and pleasure directly; instead, their inquiry is essentially mediated by the authority of others who, as it were, do the work of inquiring for them. Their experience of the essential interrelation between activity and pleasure is obscured by their appropriation of one-sided pronouncements for or against the life of pleasure. They thus neglect that which Aristotle claims is to be neglected least of all ([phrase omitted]) the dispute over whether and in what sense pleasure is to be pursued. (108)

IX

Aristotle introduces the account of akratic conflict examined in the previous section with the following claim.

   O. For one opinion is a universal, and the other is about
   particulars, over which perception is already authoritative. And
   whenever one thing ([phrase omitted]) comes to be from these, it is
   necessary that, on the one hand, the conclusion is thereupon
   ([phrase omitted]) affirmed by the soul and, on the other hand, in
   the productive case ([phrase omitted]) immediately ([phrase omitted])
   done. (109)

On the interpretation I have offered, the "one" that comes to be from the universal opinion and the other opinion over which perception is authoritative is not a distinct propositional content over and above the major and minor premises in the practical syllogism. It is, rather, the unified understanding that brings these premises together into one thought. (110) More specifically, it is the intellective perception of one's ethical agency in its concrete specificity as exemplifying an ethical universal. To this intellective perception there corresponds a conclusion ([phrase omitted]) that the soul affirms, a conception of what is to be done that Aristotle calls the "final premise" ([phrase omitted]). (111) It is this final premise, Aristotle claims, that the akratic agent either doesn't possess, or does possess but not in the way that his having is a knowing. (112)

David Charles has similarly argued that ([phrase omitted]) should be identified with the conception of what one ought to do that the akratic contravenes in his action. For that very reason, however, he translates ([phrase omitted]) as "proposition" rather than "premise" in this context. (113) My reason for retaining "premise" over "proposition" is that I believe our translation should allow us to understand protasis here as both a demonstrative premise and a dialectical premise ([phrase omitted]). (114)

Aristotle defines a dialectical premise as a contradiction in the form of a question ([phrase omitted]) in a dialectical inquiry. (115) More precisely, it is a question that takes up one side of the contradictory pair of assertions that make up a dialectical problem. A dialectical problem, Aristotle says in Topics, is an object of contemplation ([phrase omitted]) that is focused either on choice and avoidance or on truth and knowledge. (116) He says, moreover, that "some dialectical problems are useful to know in relation to choosing or avoiding, for example, whether pleasure is to be chosen or not." (117) Note that Aristotle does not claim here that it is useful to know the answer to a dialectical problem, but rather that it is useful to know the problem itself. Note, moreover, that the dialectical problem Aristotle singles out as significant in relation to choice is the problem of whether or not pleasure is to be chosen. (118)

I have argued that the dialectical examination of akrasia in Nicomachean Ethics 7 should be understood as making manifest the problem of akratic conflict as experienced by the akratic would-be agent. I have also tried to show how Aristotle's depiction of akratic conflict suggests a particular etiology for at least some cases of akratic conflict. In akratic conflict the question of whether and in what sense pleasure is to be pursued is experienced as closed to dialectical inquiry. The akratic possesses, as it were, an authoritative answer to that question without an adequate experiential grasp of the question itself. He has the "that" without the "why" and, moreover, has the "that" in a way that the end is not sufficiently clear to him.

Aristotle's account of akrasia is followed by the first of the two dialectical examinations of pleasure which bracket the books on friendship. Immediately after the second account of pleasure, Aristotle returns to the question of happiness that governs the Nicomachean Ethics as a whole, but he does so as if he were posing the question for the first time. (119) In the context of this new beginning, Aristotle offers his most decisive arguments for the superiority of the life of theoretical contemplation. An interpretation of these two perplexing accounts of pleasure or the vexed problem of the precise relation between them stands outside the scope of this essay, but I will conclude with the following brief suggestion. Aristotle's dual examinations of the problem of pleasure in Nicomachean Ethics 7.11-14 and 10.1-5 emphasize the essential inseparability of pleasure and activity, two aspects of human happiness that are held apart in the conceptions of happiness held by the proponents of the life of enjoyment and the life of honor discussed in Nicomachean Ethics 1.3. This inseparability is best exemplified and most clearly achieved in the life of contemplation. This is a life of inquiry into the very questions to which the endoxa are lived responses. Aristotle prepares the way for his readers to take up these questions by showing the way that our political experience can obscure our access to these questions as questions. One reason Aristotle shows or makes manifest the lived character of akratic conflict rather than simply telling his readers how they should understand that conflict is that he is trying to make manifest the difference between a mere account and the active contemplation of that account, the difference between the hexis as a potential comprehension of ethical agency and the actual comprehension of ethical agency. He brings the endoxa to light by bringing the problem of akratic conflict before the eyes, as it were, and allowing his readers to puzzle through the phenomena. This is how his account helps to unravel the difficulties and hand down the endoxa. If he can do this, he says, he will have shown enough.

[Please note: Some non-Latin characters were omitted from this article]

Deep Springs College

Correspondence to: [email protected].

(1) Metaphysics 2.1.995a27-8.

(2) NE 7.1.1145b2-7. Translations from the Nicomachean Ethics in this essay follow Robert C. Bartlett and Susan D. Collins, Aristotle's Nicomachean Ethics (Chicago: The University of Chicago Press, 2011). I have, however, freely altered those translations when necessary. Remaining translations are my own unless otherwise indicated.

(3) Topics 1.1.100b21-23. Among the most influential responses to Owen's article ("Tithenai ta phainomena," in Aristote et les problemes de methode, ed. S. Mansion (Louvain: Presses Universitaires de Louvain, 1961) are Martha Nussbaum, The Fragility of Goodness (Cambridge: Cambridge University Press, 1986), 240-56; Terence Irwin, Aristotle's First Principles (Oxford: Clarendon Press, 1988); C. D. C. Reeve, Practices of Reason: Aristotle's the Nicomachean Ethics (New York: Clarendon Press, 1992), 34-66; David W. Hamlyn, "Aristotle on Dialectic," Philosophy (1990): 465-76; and Robin Smith, "Dialectic and Method in Aristotle" in From Puzzles to Principles? Essays on Aristotle's Dialectic, ed. May Sim (Lanham, Md.: Lexington Books, 1999). Recent synoptic treatments that defend the centrality of Aristotle's dialectical or endoxic method include Richard Kraut, "How to Justify Ethical Propositions: Aristotle's Method," in The Blackwell Guide to Aristotle's Nicomachean Ethics (Malden, Mass.: Blackwell Publishing, 2006), and especially C. D. C. Reeve, "Aristotle's Philosophic Method," in The Oxford Handbook of Aristotle, ed. C. Shields (Oxford: Oxford University Press, 2012), 150-70. For recent arguments against see Gregory Salmieri, "Aristotle's Non-'Dialectical' Methodology in the 'The Nicomachean Ethics'," Ancient Philosophy 29, no. 2 (2009): 311-35, and especially Dorothea Frede, "The Endoxon Mystique: What Endoxa Are And What They Are Not," in Oxford Studies in Ancient Philosophy, vol. 43, ed. Brad Inwood (Oxford: Oxford University Press, 2012), 184-215. John Cooper's treatment of the passage in Aristotle: The Nicomachean Ethics, Book VII Symposium Aristotelicum, ed. Carlo Natali (Oxford: Oxford University Press, 2009), rightly emphasizes the question of the specific relation between the "methodology" outlined in Nicomachean Ethics 1145b2-7 and the specific arguments of Nicomachean Ethics 7. Unlike Cooper, however, I do not think that taking this question seriously argues against the broader relevance of the passage in Aristotle's work.

(4) A significant exception is Kurt Pritzl's treatment of endoxa in "Opinions as Appearances: 'Endoxa' in Aristotle," Ancient Philosophy 14, no. 1 (1994): 41-50, an article to which I will return below. More recently, Christopher P. Long has argued for a more dialogical and "ecological" understanding of the logoi in question. While I am in broad agreement with the dialogical character of Long's approach, I believe that he fails to recognize key limitations to the identification of ta legomena with ta phainomena in Aristotle's work. See Christopher Long, Aristotle on the Nature of Truth (Cambridge: Cambridge University Press, 2011), 49-71.

(5) NE 1.5.1095b 14-16.

(6) NE 10.1.1172a26-b7.

(7) NE 10.8.1179a17-22.

(8) Topics 1.1.100b24-101al.

(9) Pritzl, "Opinions as Appearances," 45.

(10) While he does not argue against a propositional construal of endoxa, in Aristotle: Topics: Books I and VIII (New York: Clarendon Press, 1997), Robin Smith emphasizes the difficulty of discriminating between genuine and apparent endoxa, and hence between dialectical and eristic arguments, simply on the basis of the opinion expressed (48).

(11) Rhetoric 3.10.1410b21-26.

(12) Rhetoric 3.10.1410b31-33.

(13) ([phrase omitted]) Rhetoric 3.11.1412a21-22.

(14) As Pritzl writes, "(t)his is what 'proving' consists in for endoxa--a 'showing' of the truth given partially or obscurely in them (compare EE 1216b32; Phys 211a6-110). This sense of 'showing' rather than 'establishing' or 'certifying' is fully within the range of semantic possibilities for deiknunai." Pritzl, "Opinions as Appearances," 45.

(15) Compare NE 4.6.1126bll-4, 9.10.1170b10-14.

(16) This way of conceiving Aristotle's dialectical method is similar in crucial respects to, for example, Stephen Salkever's account of "Aristotle's philosophical pedagogy" in "Teaching the Questions: Aristotle's Philosophical Pedagogy in the Nicomachean Ethics and the Politics," The Review of Politics 69, no. 2 (2007): 192-214, and more generally to other recent accounts of the Nicomachean Ethics as a protreptic work (see his 193 n. 1 for references). Those accounts, however, for the most part share in the predominant conception of endoxa as propositional. Moreover, unlike those accounts, I am concerned with articulating a particular philosophic conception of the role of perplexity in practical inquiry.

(17) On the role of contradictory assertions in Aristotle's conception of dialectic, see C. W. A. Whitaker, Aristotle's De Interpretation (Oxford: Oxford University Press, 2002).

(18) Topics 1.2.101b3-5.

(19) NE 7.6.1149a25-28.

(20) Compare Posterior Analytics 1.10.76b23-26.

(21) NE 6.13.1144b12-14

(22) Metaphysics 3.1.995a27-b2.

(23) For an excellent account of the role of prior perplexity in this passage, differing in some detail from the one I present here, see Vasilis Politis, Routledge Philosophy Guidebook to Aristotle and the Metaphysics (London: Routledge, 2005), 64-74.

(24) Topics 6.6.145M6-20. Compare Vasilis Politis, "The Place of 'Aporia' in Plato's 'Charmides'," Phronesis 53, no. 1 (2008): 4.

(25) Posterior Analytics 2.3.90a34-35.

(26) Posterior Analytics 2.8.93a19-22.

(27) Metaphysics 7.17.1041a9.

(28) The significance of prior perplexity for practical inquiry is suggested in the passage from Metaphysics 3.1 by Aristotle's invocation of the similarity between those who have not gone through perplexity and those who don't know where they must walk, a favoured metaphor for practical reasoning in Aristotle's works. It is further underlined by the implicit references in this passage to Plato's Meno, where the aporiai arise in an inquiry about the eidos of virtue. The most obvious allusion in this passage to the dialogue is the reference in 995b 1 to the eponymous paradox about the possibility of inquiry. But Aristotle's claim at 995a28 that is it useful or serviceable ([phrase omitted]) to one who wants be resourceful in inquiry to go through perplexity well echoes Socrates' defense of the usefulness ([phrase omitted]) of perplexity as a means to discovery ([phrase omitted]) at Meno 84b9-10. Moreover, Meno himself seems to be an excellent example of an inquirer who has not already been perplexed and to whom the end does not appear.

(29) NE 6.9.1142b32.

(30) NE 3.3.1112b12-5.

(31) See Nicomachean Ethics 3.3.1112b12-5, 6.2.1139b1-4, and 6.5.1140b7. The doctor in Nicomachean Ethics 3.3 is deliberating, as A. W. Price puts it, "qua doctor" and not "as a man." This does not preclude him from deliberating about whether or not this whole doctoring thing is really for him, after all. Most treatments of the issue, Price's included, do not emphasize adequately the significance of the relation indicated here between deliberation in the context of a particular technical or semitechnical practice and the deliberator's identification with and aspiration toward the particular social role of authoritative practitioner. To deliberate medically is conceived here in terms of an inquiry where the deliberator seeks to ascertain "the things the art of medicine commands and as he who possesses that art commands" (1138b32-3). As the phronimos is to ethical virtue, so the hiatros is to the art of medicine. See A. W. Price, Virtue and Reason in Plato and Aristotle (Oxford: Oxford University Press, 2011), 218.

(32) NE 6.5.1140a28.

(33) On the Movement of Animals 7.701a12-13. See Price, Virtue and Reason, 289-91.

(34) NE 6.2.1139b4.

(35) NE 6.5.1140b6-7.

(36) NE 6.8.1142a24-26.

(37) NE 6.8.1142a26-30.

(38) As John McDowell puts it, "having the right goal ... is inseparable from the ability to know what is to be done occasion by occasion," and this is "what practical wisdom is." See McDowell, "Some Issues in Aristotle's Moral Psychology," in Companions to Ancient Thought: 4: Ethics, ed. Stephen Everson (Cambridge: Cambridge University Press, 1998), 114.

(39) On the Soul 3.11.434a16-22.

(40) On this point, see Karen Margrethe Nielsen, "Deliberation as Inquiry: Aristotle's Alternative to the Presumption of Open Alternatives," Philosophical Review 120, no. 3 (2011): 383-421.

(41) In choosing these two examples, I am deliberately avoiding the question of whether the individual virtues Aristotle examines in books 3 and 4 are, as virtues, potentially in conflict with one another. See Rebecca Stangl, "A Dilemma for Particularist Virtue Ethics," Philosophical Quarterly 58, no. 233 (2008): 665-78.

(42) NE 1.5.1095b3-8.

(43) See Nielsen, "Deliberation as Inquiry," 404-08, and Heda Segvic, "Deliberation and Choice in Aristotle," in From Protagoras to Aristotle: Essays in Ancient Moral Philosophy, ed. Myles Burnyeat and Charles Brittain (Princeton, N.J.: Princeton University Press, 2009), 168-71, with references.

(44) David Wiggins, "Deliberation and Practical Reason," Proceedings of the Aristotelian Society, the Nicomachean Ethics Series, vol. 76 (1975-76): 38.

(45) Part of the difficulty involves unavoidable scope ambiguities involved in any end sufficiently circumscribed to differentiate it from eudaimonia, the final end of human beings as such. For any end such as an interesting holiday, any boundary that we draw between facilitating conditions and constituent means is likely to seem arbitrary. We can ask, for example, whether a pleasant flight is merely an instrumental means to the end of a pleasant holiday, or whether it is a constituent means and part of the proper specification of that holiday. To answer that question, it seems, we need to ask some such question as "What do I want out of a pleasant holiday, anyway?" This latter question is, in turn, hard to distinguish in principle from the question "Why should I go on a holiday?" and, ultimately, "Why should I seek the pleasant?" That is, the very attempt to distinguish instrumental means from proper specification of the end leads to questions about the end as such. Compare McDowell, "Some Issues in Aristotle's Moral Psychology," 115 n. 19. Similar issues arise in debates in deontic logic between wide and narrow scope readings of "ought" in hypothetical imperatives.

(46) NE 6.9.1142b16-26. My translation is based on Bartlett and Collins with substantial alterations. Worthy of note is the fact that I follow Irwin in accepting ([phrase omitted]) in 1142b19, commonly bracketed by editors but attested by most manuscripts.

(47) This error, as Paula Gottlieb shows in The Virtue in Aristotle's Ethics (Cambridge: Cambridge University Press, 2009), has a long history. Her response to Aquinas, who suggests choosing to steal as a means of helping the poor as an example of a false middle term, is apposite. "According to Aristotle, the correct action would be to help the poor at the right time, in the right way, from the right sources, and so on, and so the false syllogism would not have the right conclusion, since it is never the right time to help the poor from the wrong sources. We need to find an alternative interpretation where the middle term is incorrect but the conclusion is still true." Her own interpretation takes a number of steps in the right direction. In particular, her work is exceptional in its focus on the identification of the middle term in the practical syllogism with the agent's self-conception. But she does not avail herself of the resources opened up by this insight, and it is hard to see how her criticism of Aquinas does not apply to her own analysis of the false syllogism, which also involves an agent who helps the poor by stealing (168).

(48) NE 2.3.1104b2-5.

(49) NE 3.7.1115b 18-20.

(50) NE 6.12.1143b 18-28.

(51) NE 6.12.1144a13-17.

(52) See NE 3.5.1141b23.

(53) NE 3.5.1141al-5.

(54) NE 5.11.1138a6; compare NE 5.2.1130M8-19.

(55) McDowell, "Some Issues is Aristotle's Moral Psychology," 120.

(56) NE 1.13.1102b34.

(57) NE 1.4.1095b4-14.

(58) NE 1.13.1102b28-31.

(59) NE 1.13.1102b31-33.

(60) NE 1.13. 1103al-3.

(61) NE 1.13.1102a30. See Michael Davis, The Soul of the Greeks: An Inquiry (Chicago: The University of Chicago Press, 2011), 55-60.

(62) NE 1.7.1098a5-6.

(63) NE 7.3.1147b9-13.

(64) See David Charles, "NE VII.3: Varieties of Acrasia," in Aristotle: The Nicomachean Ethics, Book VII Symposium Aristotelicum, 9-39. Charles argues that ([phrase omitted]) in ([phrase omitted]) should be translated as "proposition" rather than "premise." He offers a further defence of this translation of protasis with reference to the use of the term in the Analytics in David Charles and Paolo Crivelli, "'Protasis' in Aristotle's 'Prior Analytics'," Phronesis 56, no. 3 (2011): 193-203. Compare David Charles, Aristotle's Philosophy of Action (Ithaca, N.Y.: Cornell University Press, 1984), 109-60. My reasons for retaining "premise" rather than "proposition" are given in section IX below.

(65) NE 7.3.1147al 1-14.

(66) NE 7.2.1145b27-33.

(67) NE 1.4.1095a26-31.

(68) NE 1.5.1095b 14-17.

(69) NE 6.13.1144b 19, b29; 7.2.1145b23.

(70) NE 10.2.1172b9, b18.

(71) NE 7.3.1147b 13-14.

(72) NE 7.2.1146a32; 7.3.1146b22; 7.4.1148a4-10, 17; compare NE 5.8.1135b9-10.

(73) This claim may seem contradicted by Aristotle's response to the question he poses at 1146b24-27 about "whether it is true opinion but not knowledge that those who act akratically contravene." His response is that "there is no difference in relation to the account ([phrase omitted]) for some people, when they are opining are not uncertain but think they know things precisely ([phrase omitted]) At first glance, the most natural reading of the passage is to take it as arguing that the distinction between opinion and knowledge makes no difference for our understanding of akrasia, on the grounds that some people can have the same degree of conviction regarding opinion as others have regarding what they know. But, if that were Aristotle's argument, he would be guilty of very weak reasoning. In particular, he would be guilty of conflating two quite distinct questions: (1) the question of whether the difference between opining and knowing is significant for an adequate explanation of the possibility of akrasia, and (2) the question of whether the relevant explanatory difference between those who opine and those who know is to be found in the comparative strength of their convictions. But it is clearly possible to answer the second question negatively, as Aristotle does, while answering the first question affirmatively. More generally, one could think that the difference between opining and knowing plays necessary role in explaining the possibility of akrasia, without thinking that it is sufficient to explain its occurrence.

What, then, me we to make of Aristotle's claim in this passage that the difference between opinion and knowledge "makes no difference to the account"? The "account" referred to is, I believe, not Aristotle's account of the possibility akrasia, but rather the correct account of good human action that is shared both by akratic and nonakratic agents. The difference is not in that account but in the sense in which the agent knows or possesses that account. That is, the significant difference is in the hexis rather than in the logos.

(74) David Bostock objects that Aristotle's prima facie claim that the akratic agent lacks knowledge when he undergoes the experience of akrasia is "wholly unrealistic" and offers as evidence in support of his claim the fact that "even as I bite into the eclair I may well be thinking to myself, "I should not be doing this." And there is no reason at all to say that I fail to understand these words that I think to myself; I do know perfectly well that I should not be doing what I am doing." This objection, however, fails to pay sufficient attention to the distinction between the kind of knowledge required to be a competent speaker of English uttering the sentence "this is bad for me" and the kind of knowledge required to "know perfectly well that I should not be doing what I am doing." To know, rather than merely accept on trust, that eating light meats is healthy or eclairs unhealthy for human beings, one would need to possess at least as much knowledge of human physiology as the medical doctor, and this is knowledge that the akratic agent in general presumably lacks (1143b25-28). Moreover, one must know, rather than opine, that whatever the contribution made by not eating this eclair to one's health is, all things considered, better for me than the pleasure I take in eating it. How to correctly perform such a calculation in each case between the healthy and the pleasurable, however, is far from obvious. As Aristotle points out, even contemplation can sometimes harm your health (1153a20). See David Bostock, Aristotle's Ethics (Oxford: Oxford University Press, 2000), 133.

(75) NE 7.2.1146a35-b1.

(76) NE 9.8.1168b29-34.

(77) NE 6.11.1143b6.

(78) On the Movement of Animals 7.701a13.

(79) NE 7.3.1147al-7.

(80) See NE 2.6.1106a28; 2.8.1109a6, 12; 4.6.1126M2; and esp. 5.10.1137b18.

(81) NE 8.8.1159b21-23.

(82) On the Soul 2.10.422a34-b5.

(83) NE 7.6.1149a26-28.

(84) NE 7.4.1147b23-27.

(85) NE 7.3.1147a29-35.

(86) Price, Virtue and Reason, 244. See also GA 4.8.776a25-30, where it is said that in all things the part that provides nutriment is the sweetest.

(87) NE 3.10.1118a24-b9.

(88) NE 10.6.1176b19-21.

(89) NE 10.9.1179bl 1-16.

(90) NE 3.4.1113a25-31,10.6.1176a12-16.

(91) NE 7.3.1147a35-b3.

(92) Compare NE 6.2.1139a37.

(93) NE 6.2.1139a31.

(94) NE 1.8.1099a7-11, 3.10.1117 b28-31.

(95) On the Soul 3.7.431al0.

(96) NE 7.13.1153b26.

(97) NE 7.13.1153b29-32.

(98) NE 10.4.1175a18-19.

(99) NE 7.8.1151a15-21.

(100) NE 7.3.1147a5-6.

(101) NE 10.1.1172a27-33.

(102) NE 10.1.1172a36-b1.

(103) NE 1.5.1095b19-20.

(104) NE 10.7.1177a22-27.

(105) NE 10.4.1175a18-21.

(106) NE 8.8.1159a19.

(107) NE 1.5.1095b23-4.

(108) NE 10.1.1172a25-26.

(109) NE 7.3.1147a25-28.

(110) Compare Posterior Analytics 1.14-15.

(111) NE 7.3.1147b9.

(112) NE 7.3.1147b 11-12.

(113) Most commentators take "the final premise" as the minor premise, in contradistinction to the major premise. As Ronna Burger notes, however, "the final premise" cannot refer to, for example, the perception "this is sweet" in the example that follows because this perception is said to be in energeia. See Ronna Burger, Aristotle's Dialogue with Socrates: On the Nicomachean Ethics (Chicago: The University of Chicago Press, 2008), 143. Compare Charles, "Varieties of Acrasia," 42 nn. 2-4 for Charles's account of his place in the scholarly debate.

(114) This is not to say, of course, that adverting to the difference between "premise" and "proposition" in English is sufficient to clarify the issue. "Dialectical premise" is, however, the standard translation of ([phrase omitted]) in the literature. See, for example, Reeve, "Aristotle's Philosophic Method."

(115) Prior Analytics 1.1.24a25, Topics 1.10.104a8.

(116) Topics 1.11.104b 1-5.

(117) ([phrase omitted]) Topics 1.11.104b5-7.

(118) See Topics 2.1.108b34-109al.

(119) NE 10.6.1176a30-32.