Jump to ContentJump to Main Navigation
Moral Psychology and Human Action in Aristotle$

Michael Pakaluk and Giles Pearson

Print publication date: 2011

Print ISBN-13: 9780199546541

Published to Oxford Scholarship Online: September 2011

DOI: 10.1093/acprof:oso/9780199546541.001.0001

Deliberation and Choice in Aristotle

(p.159) 7 Deliberation and Choice in Aristotle
Moral Psychology and Human Action in Aristotle

Heda Segvic

Oxford University Press

Abstract and Keywords

How can every virtuous action be chosen, on Aristotle's view, if choice requires deliberation (since surely some situations do not permit time for deliberation)? How can Aristotle claim, and what does he mean in claiming, that deliberation is not of the end? This chapter offers distinctive interpretations of key notions in Aristotle's moral psychology in answering these questions — including choice (prohairesis), wish (or ‘rational wanting’, boulēsis), practical wisdom (phronēsis) and happiness (eudaimonia). Aristotle's notion of ethical deliberation reduces neither to instrumental reasoning, nor even to explicit reasoning that can be captured in practical arguments.

Keywords:   practical reason, choice, deliberation, phronimos, wish, practical wisdom, happiness, practical arguments

There was no obvious questioning, nor figurings, nor diagrams. There was, apparently, no considered loopholes. It appeared that the swift wings of their desires would have shattered against the iron gates of the impossible (Crane 1967: 222–3).

1. What is deliberation?

Our understanding of what deliberation (bouleusis) is for Aristotle should start from his remarks about euboulia, the virtue of deliberating well. For that is deliberation at its best.

At NE 6.7, 1141b12–14 he characterizes the unconditionally good deliberator, ho haplōs euboulos, as ‘the person who is capable of aiming (stochastikos) in accordance with calculation (kata ton logismon) at what is the best for a human being of things pursued in action’.1 This characterization very nearly sums up Aristotle's understanding of practical wisdom, phronēsis, the overarching virtue of practical intellect. He makes the point himself, by way of preface to the characterization of the good deliberator just quoted. He says: ‘we say that deliberating well (to eu bouleuesthai) is above all the ergon of the practically wise person’ (1141b9–10). An account of some things is to be sought in the work they do. Aristotle uses the term ergon to capture the kind of activity that defines a thing of this sort. (Thus he attempts to specify what makes human beings the beings they are in his ergon argument at NE 1.7.) The ergon of a thing is best characterized as what the thing does that makes it what it is. Accordingly, to specify the practically wise person's ergon is to spell out precisely what makes (p.160) him practically wise. To say that deliberating well is above all the ergon of the practically wise person is to say that what makes a person practically wise is, beyond all else, the person's euboulia, deliberative excellence.

That deliberative excellence is a defining feature of practical wisdom is a point Aristotle makes more than once. At NE 6.5, 1140a25–8, he notes: ‘It is thought to be a mark of a person of practical wisdom to be able to deliberate well about what is good and expedient for himself, not in some particular respect, for example, about what sorts of thing conduce to health or to strength, but about what sorts of thing conduce to the good life in general.’ And a little bit further on, at 1140a30–1: ‘It follows that in general the person who is good at deliberating (ho bouleutikos) is practically wise.’2

A virtue for Aristotle is always something idion, specific or proper to the activity of which it is a virtue. Euboulia is what makes one good at deliberation—it is that activity's ‘proper’ virtue. If bouleusis, deliberation, were thought of as a process of reasoning in the narrow sense—going through the steps of an argument, whether deductive or inductive, or working out by which means to attain a goal—then euboulia, as its proper virtue, could not consist in aiming at the right goal, or a tendency to aim at the right goal,3 in accordance with reasoning or calculation. Rather, it would consist in being good at reasoning or calculating in a narrower sense. A good deliberator would be someone who produces good inferences, which can serve him well in the pursuit of whatever goal he has.

The characterization of the good deliberator given at NE 6.7 is reinforced by the account of deliberative excellence at NE 6.9, the chapter that specifically deals with this virtue. At 6.9, 1142b21–2 euboulia is described, and in fact defined, as correctness of deliberation (boulē) which achieves a good (hē agathou teuktikē), or which tends to achieve a good (‘tending to achieve’ is a possible, and, I think, the intended meaning of teuktikē). A reliable tendency to achieve the end requires aiming in the right direction, and more. One's conception of the end has to be correct, and the steps one takes in an attempt to attain the end must be the right steps. The characterization of euboulia in terms of a tendency to achieve the end is therefore more demanding than the characterization given at 1141b12–14, where this virtue is taken to be responsible only for the correct aiming.

Given the account of deliberative excellence in NE 6.9, it is not surprising that this excellence is regarded as a definitive mark of practical wisdom. Someone whose use of his reason reliably enables him to aim at, and attain, the best life for a human being is someone who uses his reason as well as reason can be used for practical purposes. The two characterizations together make it clear that (p.161) correctness of deliberation demands directedness at the right goal, not just formally correct reasoning in pursuit of an independently given goal.4

The characterization of euboulia as a tendency to achieve the right goal appears in the context of a very careful examination of what sort of correctness is involved in deliberative excellence (1142b8ff.). A good deliberator could not be someone who tends to achieve the goal if he were not good at reasoning, especially of the means–end sort. At 1142b22–6 Aristotle mentions the need to reach the right goal through a correct inference, noting that reaching the right end through a wrong inference (pseudē syllogismō(i)) is ‘not yet’ euboulia. The remark shows that Aristotle, even as he makes the point that producing correct inferences is necessary for deliberative excellence, is not tempted to identify this virtue with an inferential skill.

The good deliberator, as someone who aims at, and tends to attain, the right goal by making correct choices, is someone who is good at effectively directing his desiring, and more specifically his aiming in action, toward goals that he chooses as ways of living a good life. What he is good at—deliberation—is not simply reasoning in the service of a practical goal, but an effective determination of desire toward the right goal by means of reasoning.

A puzzle concerning the relationship between deliberation and virtue of character points toward the same understanding of deliberation. The puzzle is the following. All virtuous actions are in Aristotle's view chosen, and therefore result from deliberation. (He defines choice as what is determined by deliberation at NE 3.2, 1112a15.) Yet if by ‘deliberation’ we understand a process of (p.162) explicit reasoning or calculation, of the means–end or some other sort, it is simply not reasonable to claim that all virtuous actions are done after deliberation. This is so even if we allow for the possibility that the relevant piece of reasoning was carried out quite some time before the action.5 Many virtuous actions are not deliberated in the usual sense of this word. Courageous actions come to mind, but virtue more generally seems not to require, or sometimes even permit, calculation. Aristotle would have recognized this more than anyone else. His insistence that we become, say, just by doing just actions indicates that it is not an excogitated action that counts as virtuous, but one that stems from one's properly formed emotional and desiderative propensities. In some cases immediacy of response may be a distinguishing mark of a courageous, or otherwise virtuous, person. To assume in such cases that deliberation must have been carried through some time before the action seems like an attempt to save Aristotle's view that virtuous actions are chosen at any cost.

The view that virtuous actions involve choice, and therefore deliberation, is not detachable from Aristotle's understanding of virtue. He builds it into the definition of virtue. On one reading, the famous definition of virtue of character at NE 2.6, 1106b36–1107a2 states: ‘Virtue, then, is a state having to do with choice (hexis prohairetikē), lying in the mean which is relative to us and which is determined by reason, such reason by which the practically wise person would determine it.’6 This definition has been variously interpreted. The precise sense in which the ethical hexis is related to choice has been a matter of debate. Virtue could be a state concerned with choice, a state which chooses, a state formed as a result of choice, and so on. Virtue of character is in fact connected with choice in more than one way, and the difficulty lies in determining which connection Aristotle has in mind when he defines virtue of character. It seems most likely that virtue is defined as a prohairetic state in the sense that it is a disposition to make correct choices. These choices are determined by reason, and the determination is of the sort the person of practical wisdom would make.

The practically wise person, defined as he is above all by his deliberative use of reason, figures in the very definition of virtue of character.7 The reason for this is that ethical virtues—courage, moderation, or gentleness, for instance—are all without exception deliberative virtues. Each involves a capacity to use reason to reach the right choice in a certain well‐defined range of circumstances—too well defined, perhaps, for Aristotle seems anxious to combat the Socratic tendency to construe particular virtues in the broadest terms possible. Socrates wanted each virtue to expand to fill the whole field of virtuous action, since he (p.163) was convinced that there only was one thing properly called virtue.8 For Aristotle, the well‐delimited circumstances pertaining to each particular virtue naturally provoke a definite range of desiderative or emotional responses: fear, appetitive desire, or anger, in the case of the three virtues mentioned. An agent's response to such specific circumstances is virtuous only if the resultant action is motivated by desires and passions that are not only moderate, but whose moderateness has been shaped by their reasoned outlook on life. The training of emotions that qualifies one as a brave, temperate, or gentle person essentially involves a training of the intellect. The standard of what counts as moderate is set by reason. Aristotle was therefore firmly convinced that all ethical virtue involves rational judgement. Virtue of character not only requires, but, being a disposition to choose correctly, directly embodies the excellence in deliberation that is the defining characteristic of the practically wise person.

The admitted difficulties, however, in reconciling Aristotle's view that all virtuous action is chosen, and therefore deliberated, with the fact that many virtuous actions are not excogitated have prompted the suggestion that deliberation in some cases be regarded hypothetically—not as an actual process of arriving at a choice, but rather as a theoretical device deployed for the purpose of explaining or justifying actions. Even when a moral choice has not been reached by explicit calculation—as so often seems to happen—we can, it is suggested, regard it as deliberated, since the person's reasons for acting, will, when produced, constitute a deliberative argument in favour of the choice made. Deliberation could be said to lie behind every virtuous action in a hypothetical guise. If the agent acted on reasons, it is as if he had deliberated in accordance with a practical argument that spells out those reasons (Cooper 1975: 9–10).

As a view about application of practical arguments to actions, the hypothetical approach is a highly reasonable one. Yet it cannot be the correct view of Aristotelian deliberation. The remarks that follow are meant to show that we need to draw a clear line between practical arguments and deliberation.

Actions resulting from deliberation are not the only ones that can be represented by practical arguments. A person motivated by a non‐rational desire, either appetitive or spirited,9 acts for reasons. These reasons can be set out in a form of a practical argument. Aristotle does so himself. In his account of weakness of will in NE 7.3, he mentions two practical arguments, which result in a weak‐willed action of eating something sweet. The premises of one of these arguments are: ‘Everything sweet is pleasant’ and ‘This is sweet’ (1147a32–3). The other argument's universal premise, which goes unheeded, ‘forbids tasting’ (p.164) (1147a32). Aristotle does not want us to imagine that the weak‐willed agent is assailed by the thought expressed in his conveniently provided universal premise, ‘Everything sweet is pleasant’. The akratic need not have reasoned from this thought, combining it with the observation that the piece of cake in front of him is sweet to reach the conclusion that this piece of cake is pleasant, and should therefore be tasted (see the other helpfully provided premise ‘Everything sweet should be tasted’, at 1147a29). The premises ‘Everything sweet is pleasant’ and ‘This is sweet’ (as well as ‘Everything sweet should be tasted’) are simply a way of expressing in words the standpoint of the appetitive desire that here motivates the weak‐willed action. The awkwardness of the verbal expression of the appetitive impulse comes as no surprise.

Since on Aristotle's view a non‐rational desire carries with it an appearance of goodness—namely, a valuation of the object of desire as good—the premise of a practical argument representing the desired object as good (or more specifically, pleasant, in the case of an appetitive desire) can be used with full propriety to represent the non‐rational desiderative motivation that prompted the action. Nonetheless, the akratic actions so represented by practical arguments typically do not stem from deliberation. Even someone who did not make the assumption that desire involves valuation could quite legitimately use practical arguments to represent in quasi‐deliberative form the agent's reasons for action.

If one therefore permits deliberation in a hypothetical guise in order to secure the link Aristotle envisages between deliberation and virtue, one will have permitted too much. Weak‐willed actions, which stem from a breakdown of practical rationality, will count as deliberated. The point of linking virtue with deliberation and choice will have been lost.

A practical argument is a theoretical device for explaining or justifying actions; deliberation is not. Aristotle defines deliberation as a zētēsis, a search or an enquiry (with considerable insistence: NE 6.9, 1142a31–2, a34, b2, b14, b15), or a skepsis,10 an investigation, of some sort, which, when successfully completed, results in a choice (prohairesis). He takes this characterization of deliberation seriously. The difficult task of course (valiantly tackled in 6.9) is to explain what kind of search or enquiry deliberation is. But whatever kind of enquiry or search it should turn out to be, it is an actual process of arriving at a choice.11 It is, moreover, part of Aristotle's account of choice that it is a desire of some sort (p.165) which is causally explained by deliberation.12 This could not be so if deliberation were not something actual—an actual process, or activity, of some sort. Deliberation must therefore be understood straightforwardly as the process of arriving at a choice, and not be regarded in a hypothetical guise.

In the light of this, we must distinguish between practical arguments and deliberation. When an action is deliberated, and a practical argument is used to explain it, the two will of course be closely linked. The premises of the argument should in some form capture the content of the reasoning, whether implicit or explicit, that motivates the agent to act as he does. This should be so at any rate if the practical argument is accurately to reflect the considerations that move the agent to act as he does. However, practical arguments extend further than deliberation and choice. Actions that are not chosen—in the sense of stemming from prohairesis—are no less susceptible than chosen actions to representation by a practical argument.

Scholarly discussion about practical reason in Aristotle was for a long time focused on the so‐called ‘practical syllogism.’ Even a cursory reading of Aristotle's ethical treatises, however, strongly indicates that his discussion of practical reason centres on deliberation rather than practical argument.13 The expression ‘practical syllogisms’ occurs at most once. As used by Aristotle, the term syllogismos does not refer to a syllogism, but more broadly to a deductive inference. In the practical context, syllogismos may well be used in a non‐technical sense, as a counterpart to the verb syllogizesthai, which can refer broadly to putting things together, or reaching a conclusion, by means of reason, but not necessarily by deductive inference. Moreover, in the passage that allegedly uses the term ‘practical syllogisms’, the words corresponding to ‘practical’ probably do not modify the term translated as ‘syllogisms’.

The sentence reads: hoi gar syllogismoi tōn praktōn archēn echontes eisin, [epeidē toionde to telos kai to ariston, hotidēpote on—estō gar logou charin to tychon], NE 6.12, 1144a31–3. On one reading tōn praktōn goes with oi syllogismoi: ‘For the inferences which deal with acts to be done are things which involve a starting point [viz. ‘since the end, i.e. what is best, is of such and such a nature,’ whatever it may be—let it for the sake of the argument be what we please]’ (in Ross's translation, substituting ‘inferences’ for ‘syllogisms’). On an alternative, and far more plausible reading, tōn praktōn goes with archēn: ‘For the inferences which contain the starting point of things done in action run as follows: [‘since the end, i.e. what is best, is of such and such a nature,’ etc.].’ That ta prakta, things done in action, have a starting point, archē, which is the goal of action, is something Aristotle points out repeatedly. The phrase itself, archē tōn praktōn, is an established part of his ethical discourse. See, for instance, his explanation at (p.166) NE 6.5, 1140b16–17: hai men gar archai tōn praktōn to hou heneka ta prakta, ‘the starting points of things done in action are that for the sake of which things are done in action.’ By contrast, tōn praktōn as modifying syllogismoi is odd and unprecedented. Moreover, what in this case remains of the clause, archēn echontes eisin, is syntactically extremely awkward.14 The alternative reading is much to be preferred.

Aristotle's interpreters have tried to impose a common form on the notoriously varied kinds of argument‐type structures he supplies when describing actions. However, even if there is something he would call a ‘practical syllogism’, there is reason to doubt that it would be an argument of a special sort, governed by distinctive rules of inference in the way deductive arguments are governed by the rules of deductive inference. Some of the arguments Aristotle produces are deductive in nature, others not; some are not even valid. It would be misguided to expect every argument he uses for illustrative purposes to exhibit a single canonical form.

This does not mean that practical arguments are not important theoretical tools. When our concern is to explain or justify an action, as it often might be, it is well worth our while to try to cast the considerations that motivated the agent in a rationally assessable form. Since the activity of figuring out what to do sometimes does take the form of explicit calculation, a practical argument will not explain the action in an adequate way unless it captures the central elements of the reasoning that led the agent to his action. Some approximation to the agent's reasoning must therefore be sought if the explanation is to be a good one. Yet most explanations do their job quite well without being true to life in capturing the deliberative process. Moreover, a single action, even one that is motivationally not especially complex, can be adequately represented by ex post facto arguments of different logical forms.

Explicit calculation is only the limiting instance of a range of shapes that the process of figuring out what to do might take. But Aristotle's determination to regard all virtuous action as deliberated is not sheer obstinacy. It is an expression of his considered view of the extent to which reason is engaged in moral action. It is also based on a recognition of what the process of arriving at a choice—which it is natural to call ‘deliberation’—is actually like. There is no reason to expect the process of arriving at a choice to much resemble going through steps of a rational argument, even if the considerations that guided the agent can ex post facto be captured in the form of an argument. The deliberative process is not necessarily explicitly calculative; where it does involve explicit calculation, this hardly resembles going through steps of ‘practical arguments’ as they occur in philosophy textbooks.

(p.167) Aristotle's construal of deliberation, allowing as it does ways of arriving at a choice that are not marked by explicit reasoning, is broader than what is termed ‘deliberation’ in English, and probably similarly broader than bouleusis in its ordinary uses. Yet it is a natural enough extension of the term. Aristotle may sometimes have spoken of ‘deliberation’ in a way that comes closer to the narrow, more calculative, construal. He helps himself to a more conversational usage, for instance, when he distinguishes between two types of akrasia, impetuosity (propeteia) and weakness (astheneia). The weak person, Aristotle says, deliberates but does not abide by the result of his deliberation ‘because of his affect’ (pathos), whereas the impetuous person is driven by his affect ‘because he has not deliberated’ (NE 7.7, 1150b19–22). In particular, Aristotle observes, quick‐tempered people (oxeis) and ardent people (melancholikoi) tend to be impetuous akratics; they ‘do not wait for reason, since they tend to follow appearance’—the quick‐tempered because of the urgency (tachutēs) of their impulses, and the ardent because of the intensity of their impulses (NE 7.7, 1150b25–8).

In saying that the quick‐tempered and the ardent fail to ‘wait for reason’, Aristotle probably wants to say that they do not pause to reflect on their action, or to calculate its consequences. However, the quick‐tempered and the ardent who are described as driven by their affect ‘because they have not deliberated’ also fail to deliberate in the sense that they do not, on the occasion on which they yield to impetuousness, act on their choice. Both the impetuous and the weak akratics have exercised their deliberative abilities well enough to be credited with having the right choice; when weak‐willed, they do not act on the deliverances of their deliberative effort.

Seen as a process of arriving at a choice, deliberation is an activity that modifies an initial set of desires and concerns which the agent brings to bear to his action into another set of desires which are wants or willings. This activity is deliberative only if the modification of desires can be regarded as a result of an exercise of reason.15 Deliberation, as Aristotle understands it, is not simply reasoning that serves a practical purpose, but an effective direction, or redirection, of one's will by means of reasoning.

(p.168) 2. Deliberation ‘is not of the end’: the practical starting points

Aristotle repeatedly states that deliberation is not of the end, to telos, but of ta pros to telos—things that are toward the end, or, as this phrase used to be translated, the means toward the end16 (NE 3.3, 1112b11–12 and 1112b33–4; EE 2.10, 1226a7–17 and 1226b9–12; EE 2.11, 1227b25ff; see also Rhet. 1. 6, 1362a18–19). Thus, for instance, Aristotle says, a doctor does not deliberate whether his patient should be healthy or not, but whether, say, he should take walks or not (EE 2.11, 1227b25–6; NE 3.3, 1112b12–13).

The claim that deliberation is not about the end has suggested to some that ethical deliberation is only about means, and not about ends. However, this view clashes with the picture that emerges with clarity elsewhere in the Ethics, in particular, in NE Books 6 and 7. There Aristotle claims, among other things, that choice, which is a product of deliberation, reveals character better than actions do. Instrumental reasoning is hardly indicative of character. Moreover, the overall picture of ethical deliberation and choice provided in Books 6 and 7 does not support the view that ethical deliberation is of the means only. To rely only on the few passages already quoted, if ethical deliberation were only of the means, its excellence could not consist in aiming at, and tending to achieve, the right goal.

The debate whether Aristotle in NE Book 3 argues that ethical deliberation is always about instrumental means and never about ends appears, however, to be settled. There seems to be an agreement among the more recent interpreters of Aristotle that as a result of ethical deliberation we choose not only the means to happiness, but also the ends, in particular, the constitutive parts of happiness. What settled the debate was in part the recognition that the phrase ta pros to telos does not refer just to instrumental means, but also to the constitutive parts of an end.17

However, even if this is granted, problems remain. If on Aristotle's view we deliberate about the constituents of happiness, then we do, in an important sense, deliberate about happiness as well; hence one needs to explain what he has in mind when he insists that deliberation and choice are not ‘of the end’ or ‘about the end’. Furthermore, since the examples Aristotle gives at EE 2.10–11 (p.169) and NE 3.3, are restricted to deliberation in the technical domains, a doubt may still linger in some minds that Aristotle models ethical deliberation, at least in NE 3 and EE 2, after technical deliberation.

A closer look at several passages from the Nicomachean and the Eudemian Ethics shows, I believe, what point Aristotle is trying to make by saying that deliberation is not of the end. Eudemian Ethics Book 2 is especially useful here:

We deliberate not about the ends, but about what is toward the ends (tōn pros ta telē). For the doctor does not deliberate whether he shall heal, nor the orator whether he shall persuade, not the statesman whether he shall produce law and order (eunomia), nor does any other practitioner of art deliberate about the end. Rather they lay down the end (themenoi to telos), and then examine how and by which means to bring it about […] For a deliberator seems to search and analyse in the way described, as though analysing a diagram. (NE 3.3, 1112b11–21)

Since the deliberator always deliberates for the sake of something, and there is always some goal (skopos) with reference to which he investigates what is useful, nobody deliberates about the end, this [sc. the end] being a starting point and a hypothesis (archē kai hypothesis), like hypotheses in the theoretical sciences […], but everyone's investigation is about what contributes to the end (tōn pros to telos pherontōn) […]. (EE 2.10, 1227a5–12)

For the doctor does not look into whether his patient should be healthy or not, but rather whether he should walk about or not; nor does the physical trainer look into whether one should be fit or not, but whether one should wrestle. In the same way, no other [science or art] is about the end. Just as in the theoretical sciences, the hypotheses are starting points, so in the productive [sciences or arts] the end is the starting point and hypothesis (archē kai hypothesis). Given that this thing needs to be healthy, if that is to come about, such‐and‐such must be the case, as, in the other area, if a triangle contains two right angles, such‐and‐such must be the case. (EE 2.11, 1227b25–32)

Nobody deliberates about the end, but that is presupposed (keitai) by everybody […]. (EE 2.10, 1226b9–10 (Walzer and Mingay 1991; 10–11 in Susemihl 1884))

Those who do not lay down some end (hois mētheis keitai skopos) are not deliberators. (EE 2.10, 1226b29–30)

In the Nicomachean passage deliberation, as a search (zētēsis) or investigation (skepsis), is compared with geometrical analysis, in particular, with the attempt to solve a geometrical problem by recognizing that the figure in question can be analysed as containing, for instance, a certain kind of triangle. Now, if by pointing out that deliberation is not about the ends Aristotle were at pains to stress that deliberation is always concerned with instrumental means, it would be rather peculiar for him to invoke the example of geometrical analysis, which has nothing to do with means‐end reasoning. But it would be equally wrong to assume that the reference to geometrical analysis indicates that Aristotle intends to present theoretical reasoning, or more specifically the sort of reasoning involved in solving geometrical problems, as a paradigm for practical reasoning. The point Aristotle wants to make in the passages quoted is quite different. The Eudemian Ethics goes some ways toward spelling it out.

(p.170) The reason why deliberation is not about the end, according to the first two Eudemian passages quoted, is that the end plays a different role. It is a hypothesis—something laid down or posited. The end is what in the practical sphere corresponds to a hypothesis proper, that is, a hypothesis in theoretical sciences. The Greek term hypothesis refers to something that is laid down, posited, or presupposed—provisionally perhaps, but not necessarily in a tentative way. When Aristotle says that the end is a hypothesis, or that we lay down or posit the end (in the remaining three passages), he invites us to think of the end in deliberation as something taken as a fixed point. In order to deliberate, the deliberator has to hold something constant; there can be no reasoning or deliberation without presuppositions. In the third Eudemian passage quoted, EE 2.10, 1226b9–10, Aristotle directly links the claim we are interested in here; that we do not deliberate about the end, with this point about presupposition: ‘Nobody deliberates about the end, but that is presupposed (keitai) by everybody’ (or, ‘laid down by everyone’—not, as Woods translates it: ‘that is there for everyone.’)18 The line of thought begun at 1226b9 ends at 1226b29–30 with the claim (quoted last): ‘Those who do not lay down some end are not deliberators.’

The reasoning involved in deliberation is not in fact hypothetical in our sense of the word. Deliberation does not consist in figuring out what would have to be done if the deliberator had such and such an aim in view. A mere opinion that something ought or ought not to be done is put aside at 1226b23–4, shortly before the last of the passages quoted, as a potential candidate for the starting point of deliberation. To say that deliberation is not of the end, but that the end is laid down in deliberation, is to affirm that deliberation starts from an actual practical commitment to some end. Without such a commitment there is no deliberation—those who do not lay down some end are not deliberators.

Deliberation starts with a state of aiming at a particular telos, and this involves having a desire for the object at which one aims. As the practical counterpart to the theoretical hypothesis—what we might call the ‘practical hypothesis’—the end which the deliberator is committed to, the desire for which guides his deliberation, is an integral part of the deliberation. If the deliberation is spelled out as a practical argument, the desire for the end may be formulated as the first, universal, premise.19

What is laid down by the deliberator, however, is not simply an end, but a certain conception of the end. One cannot deliberate how to bring about health without having some conception as to what health involves and what sorts of (p.171) things might bring it about (see Meta. 7.7, 1032b15–1033a1). A person is not a technical deliberator unless he is both committed to the end that defines his technē, craft, and guided in his deliberation by his expert conception of that end. Likewise, a person is not an ethical deliberator without having eudaimonia as the end in his deliberation and without having some (non‐expert) conception of eudaimonia, which is presupposed in deliberation.

In Book 3 of the Nicomachean Ethics, as well as Book 2 of the Eudemian, Aristotle focuses mostly on the likeness between technical and ethical sort of deliberation, but the likeness he wants to draw our attention to is not the concern with instrumental means. Rather, he is trying to explain what makes the reasoning involved in deliberation practical. What makes both technical and ethical deliberation practical is (in part) that deliberation presupposes aiming at a particular goal, and starts with a certain conception of the goal.

The differences between ethical and technical deliberation emerge when some issues are discussed that are not broached in the NE 3 or EE 2. The most striking differences are due to the role the conception of the end plays in each type of deliberation, and especially the sort of guidance that each conception provides. Once Aristotle has explored these differences, ethical deliberation will be seen to lie far apart from technical deliberation. (I will mention some of the differences in section 5 below.) The fact that Aristotle in NE 3.3, and EE 2.10–11 discusses ethical and technical deliberation in one breath is due to the evidently introductory purpose of these chapters.20 He is working his way toward an account of the use of reason in ethical deliberation by comparing and contrasting this use of reason with other uses of reason: solving mathematical problems, figuring out what cure to apply to make a man healthy. These comparisons do not lead to the view either that theoretical problem‐solving or that technical deliberations are paradigms for the practical use of reason.

3. Ethical deliberation and reflection

When discussing courage, Aristotle makes the following observation:

Hence also it is thought the mark of a more courageous man to be fearless and undisturbed in sudden alarms (en tois aiphnidiois phobois) than to be so in those that are foreseen; for it must have proceeded more from the state of character (apo hexeōs), because less from preparation (ek paraskeuēs); foreseen acts may be chosen by calculation and reason (ek logismou kai logou), but sudden acts (ta exaiphnēs) in accordance with one's state of character (kata tēn hexin). (NE 3.8, 1117a17–22)

(p.172) On the other hand, we have seen that, according to Aristotle, virtuous action is chosen, and therefore deliberated. Moreover, ethical deliberation, in Aristotle's view, always involves a conception of the final end, the good life. At NE 6.12, 1144a31–3, he says that ‘the inferences which contain the starting point of things done in action run as follows: “since the end, i.e. what is best (to telos kai to ariston), is of such and such a nature”’. A virtuous person's deliberation, one might think, also has to start from such a premise.

The thought that the virtuous person's action ensues on deliberation the first premise of which states a conception of the good life may well seem to conflict with the quoted description of an unpremeditated courageous action which Aristotle presents as a sign of greater courage than a premeditated and calculated one. Some have sought a way out of the conflict by assuming that to telos kai to ariston, the end that figures as the premise of deliberation, is a more specific end than the good life. Hardie, for instance, claims that the end referred to by the phrase to telos kai to ariston is not the ultimate good, ‘but the kind of end a soldier has in view when he sets himself to behave well in a battle’ (Hardie 1968: 252).

However, this cannot be right. to telos kai to ariston is Aristotle's standard expression for the ultimate end (kai is clearly epexegetical here). And in any case, Aristotle believes that if ethical deliberation, whether narrowly or broadly construed, is to be carried out properly, it is of importance that the agent's overall conception of eudaimonia be available at every point during the deliberative process. It is not his view that one explicitly consults one's whole conception of the good life. Not all of it is explicit or available for scrutiny. Aristotle does not seem to impute a fully explicit conception of eudaimonia even to the practically wise and virtuous person. His thought rather is that one's overall evaluative outlook on the conduct of life should be on call in every situation of action.

Coming to a deliberative situation equipped with a conception of eudaimonia is not to be understood as starting with a large premise containing one's whole conception of the good life, ‘Eudaimonia is…’ and deducing from that premise what one is to do here and now. If deliberation generally speaking does not consist in deductive argument, there is no reason to assume that ethical deliberation starts from a grand premise specifying the person's general ethical outlook. The way in which deliberation involves the deliberator's whole conception of eudaimonia is best described by starting from an incisive remark by David Wiggins: ‘few situations come already inscribed with the names of all the concerns which they touch or impinge upon’ (2002b: 231).21

Human life involves facing situations in which it is not immediately, or even upon reflection, clear what is at stake, from the point of view of the person (p.173) involved. A given situation will typically not activate a virtuous person's whole conception of the good life; it will activate only a part of it. However, which part is activated depends on what the whole conception is. A virtuous person's overall evaluative outlook is responsible both for bringing forth those features of the situation that he perceives as practically relevant in his initial scanning of the situation, and also for bringing about his specific response to the situation thus perceived.

If this is a fair interpretation of Aristotle—as I believe it is—there is no reason to worry whether he represents ethical deliberation as starting from a ‘grand premise’ stating the agent's overall conception of the good life. He is not arguing that in acting virtuously, for instance, we reason from such a premise. If we are in the business of explaining an action by citing a practical argument, we may start with the grand premise—or not, depending on what our explanatory purpose is. In most cases, we are interested in what made the agent respond to this situation in the way he did; the story of how he came to analyse the situation as involving this or that salient concern is not to the point.

If one looks at a situation which has already aroused some deliberative response in the agent, then one can say, with Hardie, that ‘the end and the best’ is something less general than the agent's whole conception of eudaimonia. However, if one thinks of an agent who has just faced, or is about to face, a new situation, ‘the end and the best’ must be taken to refer to the ultimate good. But even in the first case, on Aristotle's view, the whole conception of eudaimonia is involved among what he calls the deliberative starting points.

A conception of the good life, which is the starting point of ethical deliberation, is a set of evaluative attitudes—from simple desires to more complex evaluative attitudes which involve a desiderative component, such as choices, practical concerns, commitments, and so on22—which, if their content were fully spelled out, would jointly amount to some specific picture of how one should live one's life. An ordinary person's conception of eudaimonia is to a large degree implicit; it is also usually vague and full of gaps in parts, not well integrated, and, more often than not, inconsistent. Nonetheless, it is Aristotle's view that most human adults have evaluative attitudes which involve such substantive valuations and which jointly amount to an evaluative outlook on the manner in which they should conduct their life.

Deliberation is ethical if the agent's overall moral outlook guides his perception of the situation in which he acts, activating this or that part of it. The concerns or interests that are activated by the situation are always specific. Yet, Aristotle, following in the steps of Socrates, has a holistic conception of the exercise of reason that backs virtuous action. Moderation, justice and courage (p.174) are to some extent independent motivational propensities. Yet no single action of any kind is fully virtuous unless it is backed by an integral view of how one ought to go about things in life.

The person of practical wisdom is in many ways not unlike the ordinary moral agent. He would not of course be practically wise if his conception of the good was inconsistent, or his conception misguided. But he is like the rest of us in that his conception of the good life is largely embodied in his dispositions to view particular situations, and respond to them, in certain ways. He does not reach his decisions from a carefully worked out scheme of values, which he simply applies to particular situations; his insight, Aristotle insists, is bound up with recognition of particulars. In some cases, a virtuous response demands an action that is unpremeditated and uncalculated, like the courageous action Aristotle speaks of at NE 3.8, 1117a17–22.

Yet for Aristotle the practically wise person is a far more reflective character than some recent interpretations have made him out to be. Aristotle attributes the correctness of the virtuous person's choice to reason, not merely to the appropriate direction of desire and emotion, if this is seen as not mediated by reason. Virtuous action requires a width of perspective, and articulateness, that is easily lost sight of when we get too focused on the picture of the virtuous person responding to particular features of situations as they present themselves to him in his own life. The sort of perceptiveness Aristotle has in mind when he speaks of practical wisdom as ‘this eye of the soul’ (NE 6.12, 1144a30) can only exist in someone whose dispositions to act are shaped by a use of reason that is reflective and critical. Aristotle describes such a use of reason in a way that lays stress on its social element. His discussion of two special virtues of the intellect brings this out.

The first virtue is sunesis or eusunesia, comprehension or good comprehension (NE 6.10). A person who has sunesis understands and assesses well what other people have to say about their deliberation and reflection on practical matters. Unlike practical wisdom, sunesis is not prescriptive (epitaktikē), but rather critical, or concerned with judgement (kritikē) (1143a8–10). In exercising comprehension, one develops one's interpretive skills and skills of critical assessment that are necessary for conducting one's own deliberation well. Through exercising comprehension, one also gains access to other people's conceptions of the ends they pursue. These provide material, and sometimes the motivation, for one's own deliberation. Aristotle believes that especially in the beginning of our ethical development a reliance on other people's judgement is indispensable in order to get us going at all; it is indispensable, in other words, if we are eventually to come up with a conception of eudaimonia that is our own. Sunesis as a virtue, however, is not represented as relevant only to moral development. Someone who is capable of attaining what is best for the human being of things pursued in action has the ability to judge well the things others say about ethical conduct.

(p.175) Aristotle's emphasis on the ethical relevance of sunesis throws some light on an issue to which he gives little direct discussion. This is the issue of the role ethical reflection plays both in the acquisition of one's conception of the good life and in deliberation. Deliberation is an exercise of reason which is practical strictly speaking; it is a process, or rather an activity, which directly affects desires and volitions. But one does not modify desires for the sake of modifying desires. The impetus for changing desires comes from a person's evaluative outlook. This outlook is not acquired only through deliberation. People think about practical issues in a manner that is not strictly practical in Aristotle's sense of the word; they reflect on their past actions, the content of the ends they tend to pursue, the relationship among the particular ends they go for. The direction of their deliberation is often affected by such reflection. It seems then that Aristotle needs to introduce here a distinction between the exercises of reason that are practical strictly speaking, and the exercises of reason the point of which is action as well, ultimately—but not necessarily directly or immediately. Although sunesis itself is restricted to thinking of, and judging about, what other people have to say about ethical issues, in attaching importance to this virtue, Aristotle implicitly recognizes the practical significance of ethical reflection. For, that exercise of intellect which is concerned with the conduct of life, but is not practical in the strict sense—being, as he puts it, merely critical—is or involves what we usually call practical reflection.

Aristotle appears to address the issue of ethical reflection in one passage in the Eudemian Ethics. At EE 2.10, 1226b25–6, he ascribes to the deliberative part of the soul an ability to contemplate the ends of human action. He says: ‘that part of the soul is deliberative which contemplates (to theōretikon) the cause of a certain sort; for that‐for‐the‐sake‐of‐which is one of the causes’. Apart from this passage, we know that he regards his own ethical theory, as developed in the Ethics, as practical. This seems to indicate, first, that ethical inquiry on his view includes both deliberation, which is directly or properly speaking practical, and ethical reflection, which is practical in an indirect sense.

The virtue of sunesis or eusunesia conspicuously displays a social element. It shows Aristotle concerned, as he is throughout the Ethics, with other people's appearances of goodness. The desirability of some virtuous actions is often revealed only to those whose vision has been shaped by what others have to say on the matters of human life. The kind of practical insight that Aristotle thinks characteristic of practical wisdom is also shaped by an understanding of others, as another virtue of practical reason—consideration or suggnōmē (NE 6.11)—reveals.

What is peculiar from our own point of view is that the virtue of being considerate and forgiving is classified by Aristotle as an intellectual virtue. There is no talk here about the emotion of compassion, but rather about a correct discrimination (krisis orthē) of what is decent or equitable (tou epieikous), 1143a20. Understanding the other person and the willingness to extend (p.176) forgiveness is ascribed to a discriminatory insight of some sort (1143a19–20). This does not mean that the insight is based on some abstract principle of fairness; rather decency in dealing with other people, which is a characteristic of all good people, seems to be a matter of insight into particulars (1143a31–2). It is the sharpening of one's insight into the circumstances of the other people's actions, and into what moves them to act as they do, that leads to considerateness which tends to forgive. Yet the forgiveness is not unbounded: it is guided by a proper grasp of the other person, and a sense of fairness (which is another way to translate to epieikon, what is decent or equitable).

Sunesis and suggnōmē are attitudes of mind distinct from practical wisdom. They each display a certain other‐oriented understanding, and are not immediately bent on action. Yet even if different in account, comprehension and forgiveness are found in the practically wise (1143a25–31). When we think of the practically wise person as someone who makes right choices, we have also to think of him as someone whose choices result directly from the kind of reflective and discriminatory insight that forms the core of sunesis and suggnōmē.

4. Choice

Deliberation is a process which, if successfully carried through, leads to a prohairesis, a choice or decision. Aristotle's notion of choice is the most original part of his ethics.23 Choice, on this view, is a psychological attitude that involves both a belief and a desiderative state.

At EE 2.10, 1227a3–5, Aristotle says: ‘It is clear that choice is neither simply wish (boulēsis) nor belief (doxa), but belief and desire (orexis) together when as a result of deliberation they are brought to a conclusion.’ The claim made here is apparently only that choice involves desire and belief, not that it is an attitude of some third, distinct, kind.24 However, at NE 6.2, 1139b4–5, choice is characterized as ‘desiderative understanding’ (orektikos nous) or ‘rational desire’ (orexis dianoetikē), which comes closer to suggesting that choice is a state of the soul that is sui generis, and not merely a combination of a desire and a belief.

The following, preliminary, characterization of choice, or rather of a chosen action, is suggested by the passages I cite below. A chosen action, on Aristotle's view, is caused by a certain kind of wish (boulēsis), which we may call a (rational) (p.177) wanting or willing.25 My wish to ϱ is a wanting or willing to ϱ just in case the wish to ϱ involves a belief that my wishing to ϱ will bring it about that I ϱ. My action (of ϱ‐ing) is chosen just in case: (a) it is caused by my willing to ϱ, and (b) this willing to ϱ is caused by deliberation.

Let me first provide the grounds on which I attribute this preliminary characterization of a chosen action to Aristotle. With regard to (b), see EE 2.10, 1226b16–20: ‘choice is deliberative (bouleutikē) desire (orexis) for those things which are up to us…By deliberative desire, I mean desire whose origin (archē) and cause (aitia) is deliberation (bouleusis)’. For (a), see NE 6.2, 1139a31–3: ‘Now the origin of action is choice—[the origin in the sense of] the source of motion, not that for the sake of which—while the origin of choice is desire (orexis) together with reason which is for the sake of something [i.e. which aims at some end].’ The origin in the sense of ‘the source of motion’ is what later came to be called an efficient cause, or simply, cause.

Conditions (a) and (b) are well attested; what of the claim that a wish involved in a choice is a willing or wanting, as characterized above? For this, we should look at the discussion of choice at EE 2.10, 1226a17ff. and NE 3.2, 1111b19ff, where Aristotle argues that choice is a wish, but not just any kind of wish. Choice is said to be a wish for what is possible (not, for instance, a wish to square a circle, or a wish for immortality); for what can be otherwise; for what is in human power to do or not to do; for what is in the power of the agent to do or not to do (thus not, for instance, a wish concerning affairs in India, on the assumption that one has no connection whatsoever with India). Aristotle appears to be saying that a choice is a wish for what the agent believes is possible; what he believes is in human power, etc. See NE 3.2, 1111b23–6: ‘Further, wish is also for things that are not achievable through one's own agency, for instance, a victory for some actor or athlete; but nobody chooses anything of that sort, but what one thinks (oietai) can come about through one's own agency.’ The relevant kind of wish to ϕ involves a belief on the part of the deliberator that what he wants to attain by ϕ is the sort of thing he can achieve by his own agency. Why is a wish that is involved in a choice restricted in this way? We should first recall here Aristotle's concept of telos: I can wish for a victory for a particular athlete, but I cannot aim at it. In addition, Aristotle believes that chosen actions are caused by their choices. A part of his concept of human agency is that we in some sense know, or believe, that we are, through our choices, causes of our actions. Through a choice or decision we knowingly make ourselves into causes of our actions. Thus, when I choose to take a walk, my choice involves an awareness that my determination to take a walk will bring it about that I take a walk.

(p.178) The preliminary characterization of a chosen action, as it stands, does not specify a sufficient condition for an action to be chosen. Suppose that I want to speed through an intersection against the red light, as a result of deliberation. Suppose further that, being suddenly seized with fear, I change my mind about this, still in time safely to stop before the intersection. However, in panic, I press the accelerator, and indeed race through the intersection against the red light. How would Aristotle deal with this sort of case? Observe that choice on his view is not just any combination of belief and desire (or perhaps, is not a combination of just any sort of belief and desire), but a combination that involves aiming at some goal. As Aristotle puts it at NE 6.2, 1139a32–3, the origin of choice is desire together with ‘reason which is for the sake of something’ (logos ho heneka tinos). On his view, both the aiming to ϱ and the action of ϱ‐ing are caused by deliberation, and if the action itself is chosen, the same causal process must be involved both in the aiming to φ and the action of φ.

A more precise Aristotelian characterization of a chosen action might therefore run somewhat like this: An agent's action of φ is chosen if: (i) it is caused by the agent's willing to φ; (ii) it involves the same state of aiming as that which is involved in the agent's willing to φ, and (iii) the agent's willing to φ is caused by deliberation. The fact that Aristotle's account of a chosen action involves a two‐step causal story is not, I think, problematic. A willing caused by deliberation—a choice—need not result in an action. If the choice does lead to an action, he takes it that the deliberation which causes the choice also causes the action. The state of aiming involved in the agent's willing to ϱ and that involved in his ϱ are assigned to the same deliberative process. The causal story behind an action is unitary. The conception of the end that is the starting point of deliberation and leads to a chosen action is the same as, or is a direct causal ancestor of, the conception that is involved in the willing to ϱ that causes the action.

I have used the term ‘wanting’ or ‘willing’ to bring out an aspect of choice which is separate from the causal origin of choice in deliberation. The willing is a desiderative state of some sort; unlike mere desire, however, it involves a preparedness to bring about a definite state of affairs in the world through one's own effort. Aristotle's restriction of choice to such changes as can be brought about through human action, and through one's own action, has to be seen in this light. One cannot be bent on bringing about a state of affairs through action without implicitly believing that the state of affairs in question is of the sort that can be brought about through that very action (and therefore, through one's own action, and through human action). The question of the causal origin of this sort of desiderative state, which corresponds fairly closely to what we might call willing, is logically separate from the account of the state in question.

Why then does Aristotle not separate the notion of willing, which is implicit in his concept of choice, from the causal origin of choice in deliberation? The reason for this is probably that he regards the two as more than accidentally (p.179) conjoined. We should recall here that deliberation for him is not necessarily a process of explicit figuring out what to do in a particular situation. Deliberation is a process that leads through an exercise of reason to a preparedness to act. The fact that Aristotle never considers the possibility that there could be a willing, or a choice, without deliberation, seems to indicate that he thought of the state I have called willing as essentially having its source in reason. If he did think of deliberation as a process of deliberative reasoning in the more ordinary sense, he would have been drawn to accept the instances of non‐deliberated volition. However, working out what to do in an explicit way is merely one way in which reason can be exercised in determining what to do in a particular situation. Aristotelian deliberation covers this broader range of cases.

Aristotle does not ascribe what I have called willing or wanting to a separate faculty, the will. He does not speak of anything that could naturally be translated as ‘the will.’26 He does take it that reason in its practical employment gives rise to something we have good reason to call willings or volitions. That reason can do this is what makes it in Aristotle's view practical. If we do not think of the will as a separate faculty, but as an ability of reason to effectively determine action, we can say that the will in this sense is implicit in Aristotle's account of deliberation and choice. To this extent Aristotle is indeed a precursor of modern conceptions of the will.

5. The person of practical wisdom as the standard of ethical judgement

The claim that the phronimos, the practically wise person, is the standard of ethical correctness, is one of the central claims of Aristotle's ethics. I shall conclude my discussion of his conception of practical reason by looking into some of the reasons he has for making this claim.

Phronēsis, practical wisdom, can be described as a virtue of practical judgement. Whereas reasoning, in the strict sense of this word, does not aim at truth itself, but at the preservation of truth from premises to conclusion, judgement aims at truth itself. When Aristotle describes the euboulos, good deliberator, as (p.180) ‘the person who is capable of aiming in accordance with calculation (logismos) at what is best for a human being of things pursued in action’ (NE 6.7, 1141b12–14), he characterizes the euboulos as someone who, in addition to being able to reason well in the formal sense, has the virtue of good judgement about practical affairs.27

Yet practical wisdom is not just the virtue of having good judgement about practical affairs; the excellent use of reason that characterizes the practically wise person is itself practical in nature. An employment of reason is practical only if it directly affects a desire, or a complex psychological attitude which involves a desire, and thus has an impact on the person's motivation to act. The phronimos is someone who is as good as one can be at using his reason in modifying his desires, and in arriving at choices to act, with a view to living the best possible life.

To say that a person who has the virtue of practical judgement is the standard of ethical correctness may appear to be trivial. The phronimos is understood to begin with as someone who makes the right choices as the result of his excellent use of practical reason. But by making this claim, Aristotle himself clearly intends to make a non‐trivial point. He defines virtue of character as a mean of some sort which is determined by reason, namely, ‘the reason by which the practically wise person would determine it’ (NE 2.6, 1107a1–2). His response to Protagorean relativism is formulated in the claim that things which appear good to the phronimos are, as a matter of fact, good (NE 3. 4 1113a25–1113b2). Aristotle's response to Protagoras indicates that he wanted to make his understanding of ethical objectivity depend on his account of practical reason. However exactly he wanted to do this, he could not defeat Protagoras by the mere notion that the person of practical wisdom is by definition an embodiment of ethical correctness.

Here are some considerations that motivate Aristotle's characterization of the practically wise person as the standard of ethical correctness.

(1) There is an element of irreducible subjectivity in Aristotle's account of what the good life for a human being consists in. This might seem at first surprising. One might think that Aristotle, as a staunch ethical objectivist, might have been tempted by the following two (broadly Aristotelian) lines of thought. First, the extent to which a person does well in life depends on how many of his worthwhile ends he realizes, and perhaps also on which combination of these ends he realizes. According to the second Aristotelian strand of thought, a human being, as a natural creature of some sort, endowed with the (p.181) capacities characteristic of his species, does well in life to the extent that he develops and exercises such capacities in a way typical of the well‐functioning mature specimens of his species.

These two lines of thought both draw on Aristotle's reflections about eudaimonia. However, on his view the good life for a human being is not a simple function of either sort. What is missing in these characterizations of the good human life is Aristotle's thought that a human being does not do well in life unless he lives in accordance with his own conception, of what doing well in life consists in. This conception has to be one's own, in the sense of being to a large degree a result of one's own deliberation. It has also to be one's own in the sense that it is not a picture of the good life one merely speculates or fantasizes about, but is rather a conception that is operative in what one aims at in life. If one is to give an ideal representation of the human good, as Aristotle understands it, it will not do to provide an impersonal list of worthwhile human ends (perhaps along with some description of the deliberative priorities that attach to these ends in specific contexts). Nor will a list of the specifically human capacities do (accompanied perhaps with a description of the typical ways these capacities are exercised by mature human beings). What Aristotle needs is an ideal representation that embodies the agent's aiming at eudaimonia, and doing so under a conception which is a cumulative result of the agent's own previous deliberation. An appropriate device of representation for an ideal human life is someone who does these things well, and that is the Aristotelian phronimos.

(2) Aristotle's characterization of the phronimos as the standard of ethical correctness has to do with the sort of thing a conception of eudaimonia is, and also with the sort of guidance such a conception provides in ethical deliberation. Two features essentially belong to ethical deliberation. Firstly, ethical deliberation is a procedure which is not guided by genuinely independent standards, and secondly, the conception of eudaimonia any agent—including the practically wise one—is guided by is incomplete.

Whereas the final end of various technical competences plays a constructive role in technical deliberation, such a constructive role seems to be lacking in ethical deliberation. We have some independent standards for judging whether the final end which defines a given craft (e.g. health or safety in navigation, seen as defining medicine and the art of navigation) is reached in particular cases. These standards are independent in the sense that they do not depend on technical conceptions of what health or safety in navigation involve, and how they are best produced or secured. We can judge whether people are healthy or ill to some degree independently of the conception of health which is involved in medicine as a technical expertise. Hence, in arts or crafts we can adjust our activities to the effects these activities tend to produce. These effects can serve as our reference point, since we have some independent ways of judging whether they are produced or not. By contrast, practical wisdom in some sense ‘refers’ everything to the ultimate good, but there are no fully independent (p.182) ways of judging the sort of success in the conduct of life which eudaimonia amounts to.28

We do derive some independent evidence for the soundness of our valuations concerning the conduct of life by observing, for instance, what sorts of actions tend to wreck other people's lives, or make their lives successful. But, we take such an evidence to be limited in significance because it involves an external point of view, in other words, precisely because it is independent. Although Aristotelian ‘eudaimonia’ is considerably more objective a term than ‘happiness’ is, eudaimonia shares with happiness the fact that it cannot be adequately judged independently of the agent's own perspective.

In addition, as ethical agents, we do not have anything resembling an effective procedure for deciding how to act in a satisfactory way in light of the variety of valuations that make for our conception of the good life.29 In ethical deliberation, we have to work from both ends. Ethical deliberation is guided by the deliberator's general ethical outlook; this outlook may be regarded as that which is laid down or posited in a particular deliberative situation. However, the general ethical outlook may in a given situation point in different directions, or in no particular direction. In such cases some work on the valuations that go into one's ethical outlook may be necessary.

The agent's conception of eudaimonia contains valuations of unequal weight. Some parts of his ethical outlook are more important to him than others. Thus, deliberation may lead him to abandon a part of his ethical outlook in favor of another part, one which he is more committed to and less willing to give up. Conflicts in the agent's conception of eudaimonia are typically revealed and resolved in particular deliberative situations. Solutions to such conflicts can sometimes be reached without articulating the conflicting concerns and explicitly working on a solution to the problem. But some practical difficulties are sufficiently tough to require such an articulation, and may not be resolved even when the person's most carefully thought out evaluative judgements and his best deliberative skills are employed in an attempt to solve them.

In some cases, the agent's general ethical outlook will not point in any particular direction. The cases I have in mind are not of the sort in which the agent does not have any clue as to what is going on in the situation from the ethical point of view. I am assuming that the agent can recognize some of (p.183) the ethical considerations that are of relevance to the case at hand, yet these considerations are not sufficient either for him to adopt a conclusive interpretation of the situation, or to move him to act in any particular way. On occasion, one has to make a stab in the dark in the hope that the implications of one's action will be apparent at a later point. Sometimes this is the only way in which one can acquire the valuations one needs in order to deal with the new circumstances.

Yet another situation arises when the agent finds that the concerns he took himself to be firmly committed to lead unambiguously to a solution to the practical difficulty that in the given circumstances strikes him as a wrong thing to do. Pondering over the grounds for this disparity may lead to a revision of the original commitment.

These remarks are meant to explicate the kind of indeterminacy which Aristotle believes is inherent in ethical deliberation. This indeterminacy is not something that only the agents who are not fully virtuous and practically wise have to face. The practically wise person is not someone who does not have to make a stab in the dark, or to revise his commitments; rather, he is someone who is good at judging when it is appropriate, or permissible, to make a stab in the dark, or to make an appropriate revision.

This brings me to the incompleteness of the conception of eudaimonia. The practically wise person does not have at his disposal all the valuations he would need in order to reach a correct choice as to how to act in all the situations he might face. We should not be misled here, again, by the fact that the phronimos represents a certain ideal. The phronimos can only be in possession of whatever his specific excellences—above all, euboulia, the general deliberative excellence—involve, since he is nothing but an embodiment of these excellences. Euboulia does not consist in one's being equipped with a full body of practical knowledge which would enable one to arrive at a correct choice in any given situation. Aristotle's conception of the good deliberator is primarily a conception of someone who, in addition to having some body of practical knowledge, has the ability to acquire the knowledge he needs in order to deal with the new circumstances he finds himself in. Thus, Aristotle thinks of practical wisdom as, among other things, an excellence of learning.

This leads us to another reason why Aristotle chooses to describe the phronimos as the standard of the correctness of ethical judgement. If he thought that possessing a definitive body of practical knowledge were involved in being practically wise, he could refer to that knowledge as the standard of ethical correctness. However, he believes that the virtue of good practical judgement essentially involves the ability to revise and expand the necessarily incomplete conception of eudaimonia which the agent has acquired to a large degree as the result of his deliberations in the very specific circumstances of his own life. Thus, it makes better sense for Aristotle to refer to the person of practical (p.184) wisdom himself as the model of ethical correctness rather than the necessarily incomplete knowledge this person has at any point in time.

Aristotle thinks of his conception of the good practical judgement as standing in sharp contrast to the Platonic position on the excellent use of reason in practical affairs. Aristotle complained that the Platonic form of the good is not something that can guide human action (see NE 1. 6 and EE 1. 8). But Aristotle must have been aware of the fact that Plato intended the form of goodness to be something that is prakton in the sense of providing guidance for action. Plato took the form of goodness to be a paradeigma of some sort, a pattern or model, not only in the ontological sense, but also in the sense of being some sort of guide or reference point for our actions. He seemed to believe that if we knew the form of goodness, or the good‐itself, we would be able to recognize what is good in all its guises, and thus would be able to attain fully reliable practical guidance.30 His view might even have been that knowing the good‐itself, that which he describes as truly good, cannot be understood independently of the ability to recognize what is good or bad in the particular circumstances of one's life.

However closely Plato might have connected knowledge of the good itself with choosing the right courses of action, Aristotle would regard the very notion of such a kind of knowledge as misguided. Complete and exact knowledge concerning practical affairs is not, in the language of NE 1., ktēton for human beings: it cannot be ‘had’ or possessed by us. What we can have, ideally, is a procedure for dealing with practical problems which rests on some knowledge, and which helps expand that knowledge to cope with the new circumstances we encounter.

Ethical deliberation, on Aristotle's view, is distinct both from the theoretical and from the technical employment of reason. Whereas the theoretical use of reason does not involve a modification of one's desires, the technical use of reason, as Aristotle thinks of it, does. However, a technical expert has the benefit of having some standards of successful practice in his area that are independent of his own expert conception of successful practice.31 If this is so, then there is on Aristotle's view no technē, art or craft, of living well. (p.185) Nonetheless, Aristotle believes that with regard to practical affairs, there is something he would describe as a sound procedure, methodos.32 On his view, there is a way of going about things in life which, although it does not amount to an exact or complete procedure, does reliably lead to a considerable measure of success in the conduct of life. (p.186)


(1) Logismos, calculation, as Aristotle uses it here, looks to be a near synonym for bouleusis, deliberation. See NE 6.1, 1139a12–13: ‘Deliberation (to bouleuesthai) and calculation (logizesthai) are the same.’

(2) See also 6.9, 1142b31–2: ‘since, then, deliberating well is a mark of people of practical wisdom […]’.

(3) Stochastikos can have either meaning.

(4) A widely used translation of the Nicomachean Ethics renders the first two passages quoted above on deliberative excellence in a way that entirely obscures the point I have been attempting to bring out. Irwin (1985a) at 1141b9–10 reads: ‘deliberating well is the function of the intelligent [i.e. practically wise] person more than anyone else’. However, malista, ‘above all,’ undoubtedly modifies ergon, ‘the function’. The meaning is that deliberating well characterizes the practically wise person more than anything else does (characterize him). It would be extremely awkward for Aristotle to point out that deliberating well is a function of the practically wise person more than of anyone else. (‘Anyone else’ is not in the Greek; it is added by Irwin.) The practically wise person is an embodiment of the ideal practical use of reason, so it goes without saying that he is a better deliberator than anyone else. What does not go without saying is that excellence in deliberation is more than anything else a definitive mark of a practically wise person, namely, that it defines the excellent use of practical reason which makes for practical wisdom. As I already pointed out, this reading is supported elsewhere, for instance, by 1140a25–8. The characterization of the unconditional euboulos at 1141b12–14 reads in Irwin's rendition as follows: ‘the unconditionally good deliberator is the one whose aim expresses rational calculation in pursuit of the best good for a human being that is achievable in action’. The central claim that the good deliberator is ho stochastikos—the one who is capable of aiming, or the one who aims at the best goal for human beings—has been lost here. Irwin's translation steers us toward the view that the excellent use of reason consists in calculation. This fits Irwin's rather narrow understanding of deliberation in Aristotle (see Irwin 1988 and 1989). Both translations, however, are far from the Greek. (The translation of 1141b12–14 is otherwise not felicitous: it is not a prerogative of a good deliberator that his aim expresses rational calculation in pursuit of eudaimonia. The bad deliberator's aim might be expressive of his (bad) rational calculation in pursuit of eudaimonia just as well.)

(5) As Cooper (1975: 7–8) has plausibly argued.

(6) Reading Bywater's text: ὡρισμένῃ λόγῳ καὶ ᾡ̑ ἂν ὀ φρόνιμος ὁρίσειεν at 1107a1–2. ᾡ̑ is Aspasius' reading; the manuscripts have ὡς.

(7) Some of Aristotle's thinking behind the view that the practically wise person is the standard of correctness of ethical judgement is discussed in section 5 below.

(8) See Plato's Laches and Protagoras for a Socratic analysis of courage, and Charmides for an analysis of moderation, and the Protagoras for the view that virtue is one and indivisible.

(9) Appetite (epithumia) and spirit (thumos) are for Aristotle two types of non‐rational desire. In recognizing these desires, he follows Plato, and especially the Republic.

(10) Skepsis, which means primarily looking into some matter, is used in this context especially in the Eudemian Ethics. For skepsis, see EE 2.10, 1226b8, 1227a12. For the relevant usages of the related verb skopein, see EE 2.10, 1226b1 and 2.11, 1227b26; also NE 3.3, 1112b16.

(11) Cooper is well aware that deliberation is described by Aristotle as an actual process of arriving at a choice, and he often treats it as such. In taking it sometimes as a hypothetical device, he offers a reasonable solution to the conflict created by Aristotle's insistence that virtuous action is chosen and his considered views about virtuous action as not requiring explicit calculation. However, deliberation is never regarded by Aristotle hypothetically. Whereas Cooper's solution works for practical arguments, it does not help us to understand bouleusis.

(12) See EE 2.10, 1226b19–20, and the discussion of choice in section 4 below.

(13) The centrality of deliberation in Aristotle's ethics has been fully recognized by Wiggins (1975–6: 29–51; 1978–9).

(14) Stewart (1892) who wonders if ἔχοντές εἰσιν is sound, proposes that it be taken as equivalent to ἔχουσιν.

(15) A transition from one set of desires to another can in some sense be a result of an exercise of reason, yet not be deliberation. An activity of designing computer programs might lead to a person's desire to give up his present occupation and do computer programming instead. Yet this desire need not be a result of deliberation. If the desire to do computer programming is a result of deliberation in Aristotle's sense, the relevant exercise of reason is not just that involved in designing computer programs. The point of calling the relevant employment of reason ‘practical’ is to call attention to the fact that reason in this employment specifically focuses on desires and their content, and does something with them or about them.

(16) Ta pros to telos means roughly ‘[things] with reference to the end’.

(17) Cooper (1975: 19ff); Hardie (1980: 256); compare also Greenwood (1909: 46–7). Note that this wider interpretation should not be restricted to the phrase ta pros to telos. This phrase is in any case equivalent to ta tou telous heneka or ta tou telous charin, ‘things which are for the sake of the end’—see, for instance, EE 2.10, 1226a7–11. That the expressions heneka or charin (‘for the sake of’) covers constituents as well as instrumental means is suggested already by the following passage: ‘we choose honour, pleasure, intelligence, and every sort of virtue both because of themselves […], but we also choose them for the sake of happiness (tēs eudaimonias charin), thinking that through them we shall be happy’ (NE 1.7, 1097b2–5). Honour, pleasure, and intelligence are not represented here as instrumental means to happiness. Virtue, in fact, cannot on Aristotle's view of happiness (NE 1. 7) be such a means.

(18) See Woods (1992). Keitai does the job here of the perfect passive of tithenai, meaning to ‘lay down’ or ‘posit’—the very verb from which the word hypothesis is derived. This is a standard usage of the verb keisthai. Note that keitai is used in the same way in the passage cited just below: dio hois mētheis keitai skopos, ou bouleutikoi (1226b29–30).

(19) This does not mean that the deliberator reasons from that premise. The practical argument merely spells out in some form the content of the considerations that bring about the action.

(20) This is independent of the question of the order of composition of, say, Book 3 of the Nicomachean Ethics vis‐à‐vis Books 6 and 7.

(21) The lines of thought I briefly mention in this paragraph are discussed insightfully, and at length, by Wiggins (1975–6: 43–4) and by McDowell (1980).

(22) No claim is made here that these more complex evaluative attitudes cannot be analysed into beliefs, on the one hand, and desires, on the other. That level of analysis is in any case typically too fine‐grained for ethical purposes.

(23) The term prohairesis is used only once in Plato, at Parm. 143 c. Deliberation might well have been of interest to the Sophists (in Plato's Protagoras, 318e5–319a2, Protagoras claims to teach his pupils euboulia), but we have no evidence that they developed anything like the conceptual analysis of deliberation and choice that Aristotle offers.

(24) A similar claim is made at De Motu 6, 700b23: ‘choice shares both in intellect (dianoia) and in desire (orexis)’. See also De An. 3. 10, 433a22–30.

(25) I take it that ‘wish’ is an appropriate translation for boulēsis—as, for instance, when we say that we wish to be immortal, or to rule over mankind—but the most interesting kind of wish, that involved in choosing, is in fact more appropriately called ‘willing’ or ‘wanting’.

(26) Kenny (1979) translates boulēsis as ‘will.’ However, boulēsis is simply a wish. One may wish for the impossible. Aristotle's own examples are a wish for immortality, and a wish to have a kingly power over the whole of mankind (see EE 2.10, 1225b32–4 and NE 3. 2, 1111b22–3). A different kind of example are wishes for what is possible but not (normally) attainable by one's own agency, such as a wish that one's favourite team win a competition (cf. NE 3. 2, 1111b23–4). These desires are not practical in any sense, and certainly do not amount to wanting or willing of any kind. Eventually, the word boulēsis will be used in the sense that warrants its translation as ‘will’, but it is not so used in Aristotle. As I argued (Segvic 2000), the term boulēsis was used by Plato's Socrates in the Gorgias in a way that picks out certain strands of the concept of the will. Aristotle, however, contrasts boulēsis with choice, prohairesis. It is in his account of choice that an implicit notion of willing is to be found.

(27) The term logismos is intended here widely. A logismos in the narrow sense is calculation, especially of the arithmetical sort, but more broadly it is any activity of figuring things out by reason, logos, thus also an activity of finding out the truth in practical affairs insofar as this activity is a reasoned process. In particular, logismos should not be understood as restricted to means–end reasoning.

(28) For the Aristotelian idea that the ultimate good is a reference point of some sort, see NE 1.12, and especially the following: ‘By not praising pleasure although it is a good we indicate, he [Eudoxus] thought, that pleasure is better than the things which are praised; and the god and the good are of this [better] sort, since the other things are referred to them (pros tanta gar kai t'alla anapheresthai)’ (1101b28–31) (cf. EE 2.1, 1219b11–13). Although Aristotle does not say so, the thought that the good is a reference point of some sort is also Platonic: see Phaedo 76de, Republic 484c, and Segvic (2004).

(29) When Eudoxus speaks of ‘referring’ things to the ultimate good, pleasure, he might have in mind such an effective procedure. On Aristotle's view, such a procedure is not available to us.

(30) This view is suggested by Plato's Republic. Socrates in the early Platonic dialogues makes a similar assumption. When he proposes the view that virtue is knowledge, he seems to have in mind an all‐comprehensive body of knowledge which provides an effective practical guidance in all circumstances of action. However, Socrates does not make it clear whether he thinks that such a knowledge could be attained. He certainly denies that he himself has the knowledge in question.

(31) For an insightful examination of the similarities and differences between the practical and the technical use of reason, see Broadie (1991: 190–8, 202–12). This examination reveals, among other things, that our grasp of what is involved in a technical expertise is in many ways not adequate enough to provide a firm basis for the comparison between technical expertise and practical wisdom.

(32) Compare the opening sentence of NE: ‘Every craft (technē) and every methodos, and likewise every action (praxis) and choice (prohairesis) is thought to aim at some good […].’ A methodos is usually understood by Aristotle's interpreters as an approach followed in a science, epistēmē, or a craft, technē, or sometimes even as the science or the craft itself. See for instance, Burnet (1900) and Stewart (1892). However, Aspasius (1889) already made an attempt to construe the term broadly. A methodos, I believe, is no more than a procedure, or a way of going about things, that is sound and to some degree systematic. This is precisely what, on Aristotle's view, the practically wise person has, although the procedure he follows is neither scientific nor technical.