McDowell (1980) The role of eudaimonia in Aristotle’s Ethics

McDowell, J. 1980. The role of eudaimonia in Aristotle’s Ethics. In Essays on Aristotle’s Ethics. Edited by A. O. Rorty, 359–376. Berkeley: Univ. of California Press. Argues that it is wrong to construe Aristotle’s account of happiness as an objectively rational account that is persuasive to any agent; rather, the article argues that Aristotle presupposes an audience whose proper upbringing mitigates the objectivity of any explanation.

 

1. In Book 1 of the Nicomachean Ethics, Aristotle evidently endorses the thesis that eudaimonia is the chief good, the end for all that we do. Following Anthony Kenny, we can distinguish at least two possible interpretations of that thesis: either as claiming that eudaimonia is that for whose sake all action is undertaken (an indicative thesis), or as claiming that eudaimonia is that for whose sake all action ought to be undertaken (a gerundive thesis). Kenny is reluctant to attribute any doctrine of the former kind to Aristotle.

But on the face of it an indicative thesis is what Aristotle appears to accept. At 1.12.8. 1102a2-3, he says: ". . it is for the sake of this [sc. eudaimonia] that we all do all that we do"; and there seems to be no prospect of taking this to express a gerundive thesis. 

And the general drift of Book 1 points in the same direction. At 1.2.1. 1094a 18-22, Aristotle says: "if, then, there is some end of the things we do, which we desire for its own sake (everything else being desired for the sake of this), ... clearly this must be the good and the chief good." Whether or not we suppose that the second "if" clause, which I have omitted, is meant as an argument for the truth of the first,3 what I have quoted appears to say that if the indicative thesis about a single end of action is true, then the single end whose existence it asserts is the chief good; and it is hard to resist the impression that eudaimonia figures in the later chapters of Book 1 as verifying the antecedent of that conditional: first at 1.4.2. 1095al7-20, on the strength of general consensus, and then at 1.7.3-8. 1097a25-b21, on the strength of its satisfying the two conditions, finality and self-sufficiency, which Aristotle argues that the chief good must satisfy.

 

2. Suppose someone says that everyone has a single end which he pursues in all his actions. We might ask: does he mean (i) that there is some end of action common to everyone? or (ii) that everyone has his own end, but one that may differ from his neighbor's?

1.4.2-3. 1095al7-28 indicates that Aristotle's answer would be "Both." Which answer is appropriate depends on the level of specificity with which ends are formulated. People have divergent views about what eudaimonia amounts to in substantive detail: if we formulate a person's end at a level of specificity at which such divergences appear, then ex hypothesi we cannot find that end shared by all (cf. (ii) above). But a thesis on the lines of (i) can be true nevertheless, in virtue of the availability of the term eudaimonia itself as a specification of the common end whose existence such a thesis asserts.

Aristotle himself has a specific view about what kind of life constitutes eudaimonia. He certainly does not hold that everyone aims to lead that kind of life. But this yields no argument against attributing to him a thesis like (i). It would be a mistake—a missing of the nonextensionality of specifications of aim or purpose—to think one could argue on these lines: eudaimonia is in fact such and such a kind of life; there are people who do not have that kind of life as their aim; therefore there are people who do not have eudaimonia as their aim.

If it is the availability of the specification eudaimonia which permits the unification of substantively divergent ends in life, the question arises whether the unification is merely verbal. I shall revert to that question in due course (§§ 7 ff.).

 

3. Kenny's reluctance to attribute an indicative thesis to Aristotle deserves sympathy. Even if we bracket the question whether any interesting unification of divergent ends in life is effected by the specification there is still room for suspicion of the claim that any one person has, in any interesting sense, a single end in all his actions. If "actions" means something like "voluntary or purposive doings," there is surely no plausible interpretation of the notion of eudaimonia which would make it true that all of anyone's actions are undertaken for the sake of what he conceives eudaimonia to be. Worse: that is conceded by Aristotle himself, when he recognizes the occurrence of incontinence. When someone acts incontinently in pursuit of a pleasure, he differs from an intemperate person—who would also pursue the pleasure—in that pursuit of the pleasure would conform to the intemperate person's conception of the sort of life a human being should lead (hence, his concep tion of eudaimonia); whereas for the incontinent person that is precisely not so. The incontinent person has a different conception of what it is to do well (i.e., of eudaimonia), but allows himself to pursue a goal whose pursuit in the circumstances he knows to be incompatible with what, in those circumstances, doing well would be.5 So his action, though volun tary, is not undertaken for the sake of (his conception of) eudaimonia.

 

4. But we can eliminate this counterexample, and so preserve the possibility of ascribing an indicative thesis to Aristotle, as Book 1 seems to re quire (§ 1 above), without accusing him of inconsistency. What is needed —and independently justifiable —is to equip Aristotle with a concept of action under which not just any voluntary or purposive doing falls.

The chief good is the end of the things we do (telos ton prakton: 1.2.1. 1094al8-19, cf. 1.7.1. 1097a22-23); and in the explicit statement of 1.12.8. 1102a2-3, quoted in § 1 above, the verb is prattein. Now we know in any case that prattein and its cognates have a quasi-technical restricted use at some points in Aristotle. At 6.2.2. 1139al9-20 and at Eudemian Ethics 2.6.2. 1222bl8-21, praxis (“action") is restricted to man and denied to other animals. Voluntary behavior, however, is allowed to other animals by 3.2.2. Illlb7-10. That passage suggests that we should connect the field of application of the restricted use of prattein and its cognates with the field of application of the notion of proairesis (stan dardly translated "choice"), since proairesis is similarly denied to non human animals (and also to children). As for proairesis, one might have thought, from 3.3.19. 1113a9-12, that just any deliberative desire to do something would count for Aristotle as a proairesis. But that does not square with the fact that, while denying that someone who acts incontinently acts on a proairesis (e.g., 3.2.4. Illlbl3-15), he recognizes that an incontinent act can issue from deliberation (6.9.4. 1142bl8-20). The best resolution is to suppose that a proairesis is a deliberative desire to do something with a view to doing well (eupraxia: see 6.2.4-5. 1139a31-b5).6 "Doing well" (eu prattein) is by common consent a synonym for "having eudaimonia" (1.4.2. 1095al9-20). So, given the conjecture that praxeis actions in the restricted sense—are doings that issue from proairesis, we have it guaranteed, by the implicit explanation of the restricted use, that all praxeis are undertaken for the sake of eudaimonia (i.e., eupraxia).

We might reach the same conclusion, without the detour through proairesis, from 6.5.4. 1140b6-7: . than itself, action ( praxis .while making has an end other cannot; for good action (eupraxia) itself is its end." This passage forces a further refinement into our picture. Aristotle here appeals to his distinction (cf. e.g., 1.1.2. 1094a3-5) between two sorts of application of the notion of an end, or of expressions like "for the sake of," according to whether or not that for whose sake something is done is distinct from that which is done for its sake. In the terminology which commentators have adopted from Greenwood, this is the distinc tion between productive means (where the end is distinct) and constitu ent means (where the end is not distinct).7 Now in order to respect the distinction between praxis and making, we have to recognize that, even if undertaken for the sake of eudaimonia, a bit of behavior need not there by be shown to be a praxis. To count as a praxis it must be undertaken as a constituent means to eudaimonia (that is, the agent's reason must be expressible on these lines: "Doing this is what, here and now, doing well is"), as opposed to a productive means (with the agent's reason expressible on these lines: "Doing well is doing such and such, and I cannot get into a position in which I can do such and such except by doing this").

 

5. Kenny does consider (p. 28) the possibility of getting round the problem posed by incontience (§ 3 above) in something like the way I have suggested: he contemplates the suggestion that since the incontinent person does not act on a proairesis ("choice"), one mist ascribe to Aris totle the thesis that whatever is chosen is chosen for the sake of eudai monia. Kenny rejects this suggestion on the basis of 1.7.5. 1097bl-5: "for this [sc. eudaimonia] we choose always for itself and never for the sake of something else, but honor, pleasure, reason, and every virtue we choose indeed for themselves (for if nothing resulted from them we should still choose each of them), but we choose them also for the sake of eudai monia, judging that by means of them we shall have eudaimonia." But this is inconclusive.

According to Kenny, it is clear that Aristotle "means not that on some particular occasion honor and pleasure are chosen both for their own sakes and for the sake of [eudaimonia], but that on some occasions they are chosen for their own sakes, and on other occasions for the sake of [eudaimonia]." This is open to dispute. Presumably Kenny's idea is this: the parenthesis shows that choosing those things for themselves is not choosing them as means to anything else; hence it can be true both that we choose them for themselves and that we choose them for the sake of (as means to) eudaimonia, only if the occasions of these choosings are different. However, the terminology of the parenthesis (note "resulted") suggests the possibility of a different construal, according to which what it shows is that choosing those things for themselves is not choosing them as productive means to anything else. With that construal of the paren thesis, the language of the passage is compatible with the idea that choos ing those things for themselves, so far from excluding their being chosen, on the same occasions, for the sake of eudaimonia, actually is choosing them as constituent means to eudaimonia.

However, although the language of the passage permits this interpreta tion, I am doubtful whether the substance does. Virtue and reason are surely not constituent means to eudaimonia (though they may be pro ductive means); nor is it obvious that that is the right view of the relation of pleasure and honor to eudaimonia. Such a view has its plausibility in the context of a conception of eudaimonia as an aggregate of indepen dently recognizable goods, and I shall be questioning (§§ 12-14 below) whether that conception is Aristotle's.

Suppose, then, that Kenny is right about the meaning of the passage: that, according to it, there are, or could be, choosings of, say, pleasure in the belief that the behavior motivated thereby will neither constitute nor produce eudaimonia—hence, choosings of pleasure other than for the 10 sake of eudaimonia. Even so, my suggestion is not refuted. The verb translated "choose" in this passage is not proaireisthai, which, with its cognate noun, has the quasi-technical use discussed in §4, but haireisthai, which can mean (what proaireisthai in Aristotle's quasi-technical use does not mean, and what he must sometimes have needed a word for) simply "prefer," or "choose" in an ordinary sense. In that case the conces sion that in the sense appropriate to this passage, there can be choosings of pleasure other than for the sake of eudaimonia need involve no more than the familiar point about incontinence (§ 3 above); or a similar point about pursuit of pleasure, not contrary to one's conception of eudai monia, as in incontinence, but engaged in by those (e.g., children or non human animals) who do not pursue eudaimonia at all. Such points pose no threat to the thesis that all behavior that issues from proairesis is undertaken for the sake of eudaimonia.

 

6. At Eudemian Ethics 1.2.1. 1214b6-12, Aristotle says: "...every body able to live according to his own proairesis should set before him some object for noble living to aim at —on which he will keep his eyes fixed in all his praxeis (since clearly it is a mark of much folly not to have one's life regulated with regard to some End). . . ."u Kenny remarks (p. 29): "The fact that this is made as a recommendation shows that what is recommended is not something that is already the case in the behaviour of all men."

Curiously enough, the "should" that occurs in the Loeb translation I have quoted12 corresponds to nothing in the text translated (although some manuscripts do have dein). One might argue that even if the text contains no "should," it needs in any case to be understood, because the parenthesis is evidently meant to back up a recommendation.13 But it is not obvious that the parenthesis cannot be understood differently, as a sort of gloss on the restriction "able to live according to his own proaire sis." In that case, with the Loeb text, the passage yields an indicative thesis about those to whom the restriction applies (sc. all but the very foolish).

In any case, once the character of the indicative thesis which I am ascribing to Aristotle is clear, it does not ultimately matter if this passage has to be read as making a recommendation. The recommendation is that those able to act on proairesis should do so, that is, should form a con ception of eudaimonia and act for its sake; that this is made as a recom mendation does not presuppose that a piece of behavior may both issue from proairesis and not be undertaken for the sake of eudaimonia.

 

7. Suppose Aristotle does wish to maintain that praxeis are (by definition) bits of behavior undertaken as constituent means to eudaimonia. What would be the point of such a thesis? If we can find something more than merely verbal unification of divergent ends in life effected by the specification "eudaimonia/' then "under taken as constituent means to eudaimonia" marks out, in spite of the divergences, a distinctive sort of reason an agent can have for behaving as he does.

In that case the point of the thesis can be to introduce us, by way of our grasp of that distinctive sort of reason, to a restricted class of bits of behavior which, because undertaken for that sort of reason, are of special interest in ethics. I suggest that we can indeed grasp such a distinc tive sort of reason: it is the sort of reason for which someone acts when he does what he does because that seems to him to be what a human being, circumstanced as he is, should do. The ethical interest of such behavior is that the behavior, with its reasons, is indicative of the agent's character.1

 

8. It is important not to be misled about the kind of classification of reasons I have in mind. One possible classification of reasons is by gen eral features of their content, into such categories as moral, aesthetic, or prudential. But that is not the kind of classification I have in mind.

To say that someone should do something is to say that he has reason to do it. Since reasons fall under categories of the sort I have just men tioned, it might seem to follow that uses of "should' fall under categories likewise. On this view, when "should" is used in characterizing the dis tinctive sort of reason that is involved in acting with a view to eudai monia, what is involved would have to be one such specific kind of "should," say a moral or prudential "should." But that is not how I intend the suggestion.

Consider a dispute on the following lines. One party (X) says that a human being should exercise certain virtues, including, say, justice and charity. The other party (Y) says: "Nonsense! That's a wishy-washy ideal, suitable only for contemptible weaklings. A real man looks out for himself; he certainly doesn't practice charity, or justice as you conceive Now when X applies his view to specific circumstances, he will pro duce reasons which, according to him, people so circumstanced have for acting as he says they should; and the reasons will belong to one of the categories into which reasons fall. As his position has been described, the reasons will, at least in some cases, be moral ones. Y's reasons will be of a different category: namely, reasons of selfish interest. If we can neverthe less understand the exchange as a genuine dispute, with the recognizable topic "How should a human being behave?" then we cannot take the "should" in the question to have a sense that permits it to be backed only by one of the favored categories of reason. And surely we can so under stand the exchange.

 

9. I have been using the word moral for a certain category of reasons to which a person may or may not think he should conform his life: a category of reasons on a level with, and distinguished by their content from, say, aesthetic reasons, so that if someone argues that human beings should not act in a certain way because it would be, say, inelegant, we might describe him as adducing not a moral but an aesthetic reason.15 Some philosophers may wanf to object, in the interest of a use of "moral" according to which the reasons to which someone thinks a human being should conform his life are, eo ipso, the reasons he counts as moral. Thus, in the case I have just mentioned, the person is described, accord ing to this view about the use of "moral," as thinking that the avoidance of inelegance is morally required; and similarly Y, in § 8, thinks looking out for oneself is morally called for. This is to insist that the "should" that fixes the topic of such disputes as that described in § 8 is a moral "should."

It is a terminological question whether we should use "moral" in this way. The terminological proposal does not conflict with the substance of my suggestion: namely, that we can make sense of a "should" (it does not matter whether we describe it as a moral "should") which, since it intel ligibly locates disputes of the sort described in § 8, is not proprietary to any one specific mode of appraisal—in the sense in which, on this termi nological proposal, moral appraisal is no longer a specific mode of appraisal.

Some will be tempted by a different way of insisting that the "should" in question does, contrary to my suggestion, belong to a specific cate gory: namely, the thesis that ultimately it stands revealed as a certain sort of prudential "should." This is not merely a terminological proposal. I shall postpone discussion of it until I have related the suggestion to Aristotle's text.

 

10. At 1.7.9-16. 1097b22-1098a20, Aristotle exploits the thesis that the ergon of man consists in rational activity, and the conceptual connec tions between the notions of ergon, excellence, and activity, in order to reach the conclusion that eudaimonia, the good for man, is rational activity in accordance with excellence. This passage is commonly taken as a (purported) argument for Aristotle's own substantive view about what eudaimonia is. But it can be read in such a way that the conclusion is (so far) neutral, as between Aristotle's own substantive view and, say, a view of eudaimonia corresponding to the position of Y in the dispute described in § 8. With such a reading, the point of the passage can be, not to justify Aristotle's own substantive view, but rather to help the reader to comprehend the distinctive kind of reason which, according to the suggestion of § 7, the concept of eudaimonia serves to delimit.

 

11. It will be protested that I have got this far only by ignoring that aspect of the sense of eudaimonia which makes the standard transla tion, "happiness," not completely inept. That aspect ensures that the term is correctly applied only to the life that is maximally attractive or desirable.

 

12. This protest begins with something indisputable: the concept of eudaimonia is in some sense a prudential concept. When Aristotle says that activity in accordance with excellence is eudaimonia, what he says can be paraphrased as the claim that two prima facie different interpreta tions of phrases like “doing well" coincide in their extension: doing well (sc. in accordance with excellence: living as a good man would) is doing well (sc. as one would wish: living in one's best interest). But we need to ask which way round this equation is to be understood.

If, as in the protest, the prudential nature of the concept of eudaimonia is taken to show that that concept yields something like a decision proce dure for disputes like the one described in § 8, then we have to suppose that we are meant to make our way into the equation at the right-hand side. The requisite idea of the most desirable life must involve canons of desirability acceptable to all parties in the disputes, and intelligible, in advance of adopting one of the disputed theses, to someone wondering what sort of life he should lead. Such prior and independent canons of desirability would presumably need to be constructed somehow out of the content of desires which any human being can be expected to have: thus, desires conceived as manifestations of a fairly stable and universal human nature, susceptible of investigation independently of adopting one of the disputed theses about eudaimonia.1

Such an explicit mention of human nature would be a sort of rhetorical flourish, added to a conclusion already complete without it. It is argu able, however, that human nature itself is more importantly involved in disputes like the one described in § 8. The suggestion would be that it is our common human nature that limits what we can find intelligible in the way of theses about how human beings should conduct their lives, and underlies such possibilities as there are of resolving such disputes, or at least of stably adopting one of the competing positions for oneself in a reflective way (aware that there are others). I do not intend to discuss these very difficult issues here. What I want to emphasize is that if, according to the position considered in this section, human nature is involved in this sort of way, then what it has is what David Wiggins calls "a causal and enabling role"; not the "unconvincing speaking part" which it would need to be credited with in the position considered in § 12.27