4

Pleasure and Pain in the Eudemian and the Nicomachean Definitions of Moral Virtue

Marco Zingano

Both indisputably Aristotelian Ethics – the Eudemian (EE) and the Nicomachean Ethics (NE) – are set up as a conceptual analysis prompted by the definition of happiness. The definition of happiness is arrived at in the very first thematic section in both Ethics. Happiness is the chief and supreme good the obtainment of which all our actions are subordinated to. The definition of happiness is not only pivotal to moral philosophy as Aristotle conceives of it, but also provides the elements for the subsequent parts of the treatise, as the following sections build on and unfold what is mentioned in, or presupposed by it, both treatises being concluded with a discussion about first and second happiness, thus closing the loop in what looks like a ring construction.1

In the Eudemian Ethics, the definition of happiness comes by the end of the first section, which runs from EE I 1 to II, 1 1219b29. This first section discusses the aim of ethics and reports the commonly held three ways of life (EE I 1–5), delineates the method most suited to the investigation of moral topics (I 6), and then proceeds to obtain the definition of happiness (I 7 and II, 1 1218b31–1219b26), interpolating in between the criticism of Plato’s philosophical doctrine of the Form of the Good (I 8). This first section thus kicks off the philosophical analysis Aristotle is engaged in as he sets himself to examine “the philosophy of human matters” (NE X 9, 1181b15) from the vantage point of the actions and characters of the individuals. The definition of happiness comes by the end of this section and reads as follows:

ἐπεὶ δὲ ἦν ἡ εὐδαιμονία τέλεόν τι, καὶ ἔστι ζωὴ καὶ τελέα καὶ ἀτελής, καὶ ἀρετὴ ὡσαύτως (ἣ μὲν γὰρ ὅλη, ἣ δὲ μόριον), ἡ δὲ τῶν ἀτελῶν ἐνέργεια ἀτελής, εἴη ἂν ἡ εὐδαιμονία ζωῆς τελείας ἐνέργεια κατ’ ἀρετὴν τελείαν.

(EE II 1, 1219a35–39)

Solomon’s translation in the revised Oxford edition is as follows:

But since happiness was something complete, and living is either complete or incomplete and so also excellence – one excellence being a whole, the other a part – and the activity of what is incomplete is itself incomplete, therefore happiness would be the activity of a complete life in accordance with complete excellence.

From now on, I will speak of virtue instead of excellence whenever translating ἀρετή, but nothing hinges on this semantical choice for the argument I want to deploy. A more important point concerns how to translate τέλειον in its different uses here: for a life to be τελεία amounts presumably to having all or a good number of its parts in the sense of “being complete,” but it is not clear what it means for a virtue to be τέλειον. The passage states that τελεία virtue requires a whole, whereas an ἀτελής one is only a part of it, but the clarification is not very illuminating, for one can take the relation whole–part in a variety of ways. As a matter of fact, Aristotle himself acknowledges different senses of τέλειον, to wit, being complete and being perfect.2 These senses may diverge (a perfect Irish stew may not have all the possible ingredients and thus not be complete, and a complete one may not be perfect), but they may also converge; even so there is a difference to be made: something can be perfect because it is complete, or be complete because it is perfect. Which sense has τέλειον here? One might think that in so small a context, one and only one sense can be the case, but actually nothing prevents it from having two different senses in this passage. A τελεία life seems quite clearly a complete life, a life that has all, or most of, or the most important of its parts, as it stretches over time, but we are left clueless as to which sense τέλειον bears in the other case, when connected to virtue.

As we will soon see, the EE may be seen as the unfolding of this definition (and the NE alike, for that matter), as it unpacks what is only mentioned or presupposed in it. In so doing, light is shed on all parts of it, but, as we have just seen, some elements in it may resist elucidation – in our case, τέλειον remains undecided throughout the whole book. And this is not insignificant, for τέλειον is part and parcel of the phrase κατ’ ἀρετὴν τελείαν, and this phrase is crucial for understanding what Aristotle does mean by happiness. It is crucial because the κατά in κατ’ ἀρετὴν τελείαν is not only saying that happiness is brought about in accordance with a virtue of some sort (the τελεία one), but it is also saying, and more importantly, that happiness is obtained on the basis of it, thus vindicating a causal role for virtue in bringing about happiness, or, to say it more precisely, that virtue is the main cause of happiness. How virtue operates as the main cause of happiness is something the following books are bound to show, and will do it to a large extent, as they explore how virtue deploys its causal power in bringing about happiness, even if some issues remain vague, unexplored, or not fully charted – as we have already seen with τέλειον, in spite of τέλειον having an important task to discharge. Why τέλειον remains non-elucidated, or at least not totally elucidated, despite its significance for a correct grasp of the definition, is one of the cruces with which the interpreter must cope. This is not something peculiar to the EE approach, for in the NE as well, when Aristotle recalls his definition of happiness in NE I 13, just after having closed the first section devoted to producing its definition, he also says that happiness is brought about κατ’ ἀρετὴν τελείαν, and we are again left clueless as to how to construe τέλειον.

I want now to emphasize an important dissemblance between the Eudemian and the Nicomachean definition of happiness. At the surface level, a pretty similar definition seems to be found in the Nicomachean version, but some differences are worth noticing, and are actually pretty salient. To begin with, the Nicomachean definition of happiness takes place not at the end, but in the middle of the first thematic discussion, in which Aristotle, in a similar fashion to what happens in the EE, presents the aim of the treatise, gives some clues about its method, mentions common opinions about what is best, introduces us to the three ways of life, and comes out with his criticism of Plato’s philosophical stand on goodness. But this is a minor difference, and up to this point, the Nicomachean version is quite similar to the Eudemian one. But there are other differences, and quite significant ones. The Nicomachean definition runs as follows:

τὸ ἀνθρώπινον ἀγαθὸν ψυχῆς ἐνέργεια γίνεται κατ’ ἀρετήν, εἰ δὲ πλείους αἱ ἀρεταί, κατὰ τὴν ἀρίστην καὶ τελειοτάτην. ἔτι δ’ ἐν βίῳ τελείῳ

(NE I 7, 1098a16–18)

In Ross’ translation (revised by Urmson):

Human good turns out to be activity of soul in conformity with excellence, and if there are more than one excellence, in conformity with the best and most complete. But we must add in a complete life.

Ross sticks to the traditional rendering of κατά as “in accordance with,” which is not incorrect, but one would be better off translating it as “on the basis of” to stress the causal link between virtue and happiness. In the Nicomachean definition, the complete life seems to be envisaged as a clause or supplemental condition to the definition itself. It is no longer inside the definition, but is added to it as in an afterthought, signalled by ἔτι, “furthermore.” This may, however, be a product more of its linguistic clothing than of a different conception of its nature, for happiness continues to require the obtaining of two conditions to be the case: virtue being the case, and its occurring throughout the lifetime of the agent.3 A more salient difference lies in that the definition now gets involved with choosing one virtue among others: if there is more than one virtue, says the passage, happiness will be engendered on the basis of the best and most τελεία one. The τέλειον issue comes back again, under the description of κατὰ τὴν ἀρίστην καὶ τελειοτάτην, but the main issue now concerns more the fact that its role is directly linked to the discrimination of one and only one virtue among other virtues.

In his translation, Ross opens a new paragraph after “the best and most complete,” but this seems an overzealous rendering of the ἔτι clause, for it may still belong to the definitional context. But the rearrangement is noteworthy, as if it were driving a wedge between the two usages of τέλειον. For the complete life is added as a secondary clause; and the main point resides in what it means to be a τελεία virtue, referred to in the first, separate clause. Whatever τέλειον means in this clause, it should be taken in direct connection with its expression in the superlative case, or, as the text says, on the basis of the best and τελειοτάτη virtue, which seems to go in the direction of the most perfect virtue, and not necessarily the most complete. Much attention has been drawn, and rightly so, to the problem of determining the most τελεία virtue in contrast with the plurality of moral virtues. The contrast between one τελειοτάτη virtue and the many other moral virtues will become central in the second half of the last book of the NE, as an argument is there produced to identify which one is the τελειοτάτη virtue and, based on this, endeavours to ground a hierarchy between contemplative and political life.4 Surprisingly, however, it plays no relevant role from I 8 to X 5.5 The bulk of the Nicomachean treatise has no dealing with the superlative issue. A sign of this is the fact that Aristotle, when opening in the NE the second thematic section of his Ethics, the one devoted to defining moral virtue, in direct dependence on the definition of happiness, recalls the definition of happiness with no mention at all of this one–many virtue(s) issue, while accentuating the central (causal) place virtue has for there being happiness:

ἐπεὶ δ’ ἐστὶν ἡ εὐδαιμονία ψυχῆς ἐνέργειά τις κατ’ ἀρετὴν τελείαν, περὶ ἀρετῆς ἐπισκεπτέον ἂν εἴη

(NE I, 13 1102a5–6)

The issue of the meaning of τέλειον is again well in place, but the superlative sense and its corresponding choosing of one virtue among others are dropped from the text. No notice seems to be taken of it, at least not for the time being, and attention is again focused on the fact that virtue is qualified as τέλειον. A provisional translation, or rather a transliterating version that attempts to stay close to some of the points observed above, runs as follows:

Since happiness is an activity of soul on the basis of τελεία virtue, we must investigate the nature of virtue.

Such an investigation is made necessary indeed given the very definition of happiness, for the latter comprises the notion of virtue, but not only has no explanation been previously offered of what virtue means, but also, which is quite significant, it comes accompanied by an adjective whose meaning remains obscure. It is thus necessary to examine next the notion of virtue. And this is exactly what is done from NE I 13 to III 5, which can conveniently be labelled as the Aristotelian Treatise on Moral Virtue, for in it the definition of moral virtue is obtained, and its basic traits are also examined. The fact that it is the notion of moral virtue that is examined, and not the τελειοτάτη virtue, is surely a factor that produces some tension that persists from the beginning throughout the whole treatise, and will only be resolved in the second half of the tenth and last book. Tensions aside, because virtue is embedded in the definition of happiness, the next step is to define what (moral) virtue is; in so doing, we will be unpacking what comes along with the notion of happiness. After distinguishing moral virtue from intellectual virtue (NE I 13), making some comments on the way the former is acquired (II 1), and on the sort of accuracy one can expect to have in practical philosophy (II 2), as well as on its relation to pleasure (II 3), and after having dealt with a difficulty that might be raised concerning becoming virtuous by doing virtuous actions (II 4), Aristotle officially inaugurates his quest for a definition of moral virtue saying that “after these things, one must consider what virtue is” (NE II 5, 1105b19). The quest is framed as a “what-it-is question”: τί ἐστιν ἡ ἀρετή (1105b19): we now enter into the second definitional context, the one devoted to clarifying the nature of moral virtue, which comes as the logical sequence after happiness having been defined in the first section as a certain activity of soul on the basis of virtue. The outcome is this

ἔστιν ἄρα ἡ ἀρετὴ ἕξις προαιρετική, ἐν μεσότητι οὖσα τῇ πρὸς ἡμᾶς, ὡρισμένη λόγῳ καὶ ὡς ἂν ὁ φρόνιμος ὁρίσειεν.

(NE II 6, 1106b36–1107a2)

moral virtue is, therefore, a disposition issuing in decisions, lying in a mean relative to us, being determined by a piece of reasoning, that is, in the way in which the person of practical wisdom would determine it.

This is Bekker’s text, who follows the manuscripts; Bywater adopts a version based on Aspasius’ commentary, and Susemihl–Apelt mix both versions. According to the manuscripts, one has it that moral virtue is a disposition issuing in decisions by weighing the pros and cons of an action, the fulfilment of which consists in finding a mean relative to us determined by means of reasoning, or stating it otherwise, in the way in which the wise person would determine them. This sounds clear enough, but controversy lurks everywhere. One may wonder what the meaning of λόγῳ precisely is. Is it the reason by reference to which the intermediate term is determined6 in the sense of some reasoning based on which the agent decides for a certain option, or is it rather a sort of norm the agent is supposed to take into account to act well? On the former reading, moral particularism is in view; on the latter, morality can well consist of a set of fixed rules or generalizations moral agents are supposed to grasp and follow. One reckons that it is a mean term, and this already excludes both types of vice, excess and deficiency, but this does not go too far in determining what the agent ought to do. By making a reference to a λόγος, however, Aristotle gives a hint as to the sense to be given to what he has previously said about moral virtue, namely that moral virtue is στοχαστικὴ τοῦ μέσου (1106b15, 28): the sort of guessing by means of which one hits upon the mean term is not blind guesswork, but is steered by reasoning of some sort, although one can still wonder what sort of reasoning it is: how it operates, which rules it follows, and so forth.

Can we glean more information from this passage? Part of the difficulty this passage raises lies in the fact that it is often taken for granted that this definition puts together several pieces already discussed, and consequently should be clear as to their content and meaning. A good example of this expectation is provided by Aspasius’ commentary on this passage:

It is obvious that the definition has been rightly rendered. For in fact it has been shown to be a habitual state and to be choice-based, whether it is a choice or not; and it has also been shown to reside in a mean, not in accord with the thing but in the mean with respect to us.

(Konstan’s translation; my italics)7

Aspasius is right in saying that it has been shown that moral virtue lies in a mean, and that this mean is the one relative to us. But nothing has been shown in connection either to decision or to reason, let alone to the person with practical wisdom, except the allusion a bit earlier that moral virtues “are decisions of some kind, or require decision” (II 5, 1106a3–4). This remark cannot count as a clarification, for it only says that there should be some connection, making us expect to find it discussed at some later moment, but not yet showing which connection there is. In fact, the definition is rather forward-looking: one is set up to investigate what decision is, and what role the person with practical wisdom plays in determining it, if one wants to fully grasp what moral virtue really is. The definition itself is construed in this way, always engaging in a step forward: moral virtue is a disposition issuing in decisions, the decision-taking process consists in finding a mean relative to us, this mean relative to us is determined by a piece of reasoning, and reasoning occurs in the way in which the person of practical wisdom would determine it.8 Susemihl–Apelt’s text, as they mix the manuscripts version and Aspasius’, seems thus to be preferred, for it makes more apparent the forward-looking structure (a is b, b is tantamount to c, c is obtained through d, and d is explained by means of how e operates):

ἔστιν ἄρα ἡ ἀρετὴ ἕξις προαιρετική, ἐν μεσότητι οὖσα τῇ πρὸς ἡμᾶς, ὡρισμένῃ λόγῳ καὶ ὡς ἂν ὁ φρόνιμος ὁρίσειεν

(NE II 6, 1106b36–1107a2)9

Now, according to this definition, new terms are to be examined: decision is examined in III 1–5, and practical wisdom is examined in book VI; in between, from III 6 to V 11, the study of particular virtues allows us to have a better grasp of how moral virtue operates in each case as it strives for a medium point. We have already seen this strategy of introducing new elements not previously examined inside the definition so that we have to keep going and discriminate next what goes in the definiens: the definition of happiness brought in the not-yet elucidated notion of τελεία virtue, and the attempt to understand what this τελεία virtue is lead us to discuss disposition, decision-making process, and practical wisdom.

This forward-looking structure (pace Aspasius) gives us no pause in the continuous unfolding of the definition of happiness, as we still have to go forward and inquire into decision and practical wisdom. But there is something quite significant at this juncture, not in what is said in the definition, but in what the definition shrinks from saying. To see this, we have to turn to the Eudemian Ethics. In direct sequence of the definition of happiness reached at EE II 1, 1219b38–39 (quoted above), book II of the EE aims at reaching the definition of moral virtue. Similarly to the Nicomachean version, the EE overtly announces its goal in book II in terms of a quest for a definition:

After this, we must first investigate about moral virtue what it is and which are its parts (for it brings up to this point) and by which ways it is brought into existence

(EE II 1, 1220a13–15)10

Both Ethics are thus pursuing the same pattern of analysis as they carry on the analysis in terms of a search for the definition of moral virtue in a forward-looking procedure. The clause inserted in the Eudemian version, εἰς τοῦτο γὰρ ἀνῆκται, seems not to be clear at first sight and has consequently been variously translated: “for our inquiry has been forced back on this” (Solomon), “what amounts or comes to the same thing” (Inwood and Woolf; Kenny), “a questa, infatti, è giunta la discussione” (Donini), “car c’est à ce point qu’on est parvenu” (Décarie). What does it precisely mean? The verb ἀνάγω has different senses: (i) to bring from a lower to a higher point, bring up; (ii) bring someone before a jury; (iii) bring an offering, offer up; (iv) restore, bring back; (v) reduce, return; (vi) put out to sea, set sail; (vii) ascend to higher unity, only to register the most important ones. To complicate matters further, the subject of the clause is hidden. One may think that asking “what moral virtue is” comes down to asking “which are its parts” (as read in the translations by Inwood and Woolf and Kenny: sense v). It is true that inquiring into what moral virtue is requires determining which parts it has, but the former cannot be reduced to the latter, for the former is explanatory of the latter and not the other way round. Donini and Décarie seem to be on the right track. As I read it, Aristotle is warning us that a new step is reached: the definition of happiness leads up to the definition of moral virtue, which is our present task (sense i and vii). The general quest for a definition that steers the course in both Ethics, and which begins with the definition of happiness, does not come down to, nor amounts to defining moral virtue, but has this new inquiry as its second step. If I am correct, Aristotle takes time to announce it to his readers: this is our next task, to define moral virtue, after having defined happiness, the definition of which requires examining the nature of moral virtue, or saying it otherwise, it leads us up to that.

But there are significant differences in each version. Whereas in the NE the definition of moral virtue paves the way for the investigation of decision and practical wisdom, which come soon afterwards, as we have seen, in the EE the definition of moral virtue is reached after two preparatory steps, and its final version is produced not before, but after voluntariness and decision have been examined. The Eudemian first preparatory step is gained at EE II 3, 1220b34–35, a concluding remark saying that moral virtue has to do with intermediate states and is some sort of intermediacy:

Hence moral virtue must be related to certain means and be a certain mean point

(EE II 3, 1220b34–35)11

The second step is gained as soon as it becomes clear, granted that the emotions (τὰ πάθη) are defined by means of the pleasures and pains they are the expressions of (EE II 4, 1221b36–37: τὰ δὲ πάθη λύπῃ καὶ ἡδονῇ διώρισται), that dispositions are a certain way of reacting to the emotions, and that moral virtue is a character disposition:

It follows that moral virtue is the mean that is relative to each individual person or is concerned with certain means in pleasures and pains, and in the pleasant and the painful things as well

(EE II 5, 1222a10–12)12

The received text seems to call for emendation. Susemihl substituted καί for the ἤ, eliminating thus the disjunction on behalf of a conjunction (to be read in an epexegetical sense). Spengel expunged the αὑτόν, and is followed by Walzer and Mingay; Ross proposed to read τὴν ἠθικὴν ἀρετὴν καθ’ αὑτὴν ἕκαστον μεσότητα εἶναι instead of τὴν ἠθικὴν ἀρετὴν καθ’ αὑτὸν ἕκαστον μεσότητα εἶναι attributed to the manuscripts (no modern editor retains the second τήν in τὴν ἠθικὴν ἀρετὴν τὴν καθ’ αὑτὸν ἕκαστον μεσότητα εἶναι, found in both families according to Rowe).13 I think, however, we can keep the text of the manuscripts. It had just been said, in the preceding lines, that the mean term virtue consists in is the one relative to us, and this is why it is reasserted here that moral virtue is a mean in relation to each individual person. As to the disjunction, this is still a provisional step, to be further elaborated, and as things get clearer, we will see that the disjunction is in fact an alternation. But as things stands at this juncture, the disjunction opens different paths. Aristotle then goes on to state, in the third and final step, what the nature of moral virtue finally is. To do this, he goes through a rather convoluted discussion on voluntariness, what is up to us, and decision, which makes up the largest part of book II. Then, near the end of book II, he provides us with the much searched-for definition of moral virtue:

It follows then […] that moral virtue is a disposition that issues in decisions with regard to the mean relative to us in respect of those pleasant and painful things by which the person is said to have a certain sort of character, according as one is pleased or pained.

(EE II 10, 1227b5–10)14

That which is relative to each individual person or that which is concerned with certain means in pleasures and pains is now reframed in terms of taking decisions with regard to the mean relative to us in respect of pleasant and painful things. The definition is at last settled, and at the same time the most important difference between the Nicomachean and the Eudemian definition of moral virtue is brought to light. For in the Eudemian definition, pleasure and pain come to the fore and belong to the defining phrase, whereas in the Nicomachean version all mention of pleasure and pain is avoided and in its place one finds a reference to the reasoning the person with practical wisdom does when determining the mean term in an action. Is it only a literary device to express the same tenet, or is there something more substantial to their distinction? I think there is something philosophically significant going on here. Aristotle has definitely not two distinct theses, let alone two contrasting ones, but there is some different emphasizing that is worth noticing. The phrase κατὰ τὸν ὀρθὸν λόγον is not absent from Eudemian book II (see e.g. 5 1222a8), and reference is even expressly made at II 5, 1222b7 to the treatment this notion will receive later on in EE V = NE VI, the book devoted to examining practical wisdom. Decision and its role in determining the right mean is also present in the Eudemian discussion of voluntary acts and what is up to us. Still, stress is laid not on them, but upon being pleased or pained as one acts, and on acting well or badly in relation to pleasant or painful things: prominence is given to pleasure and pain within the definitional phrase of moral virtue. The connection between, on the one hand, acting well and being pleased, and, on the other, between acting in a mean way and being pained is reinforced in the EE by the idea that there are naturally pleasant things, which are at the same time good things, and unpleasant things by nature as well, which are at the same time morally inconvenient things, but which may on some occasions be pursued by people of perverted nature (see EE VII 2, 1237a4–5, 26–27).

In the NE, Aristotle is also eager to show the connection between moral virtue and pleasure. But this is done outside the definitional context of moral virtue. As a matter of fact, it is argued in NE II 3, by means of eight arguments, that there is a close connection between moral virtue and pleasure, but this is done before we begin the proper search for the definition of moral virtue (which starts at NE II 5). Moreover, NE II 3 examines their close connection expressly referring to pleasure or pain that supervenes on acts (NE II 3, 1104b4–5: τὴν ἐπιγινομένην ἡδονὴν ἢ λύπην τοῖς ἔργοις). The idea of pleasure as a supervenient end that attaches itself to an activity, and is morally good provided that the activity on which it supervenes is morally good, or is morally bad if that activity is morally bad, is present only in the Nicomachean treatise of pleasure. This may be controversial, but I am taking for granted that the common books partake of a Eudemian atmosphere, including the first treatise on pleasure transmitted to us at the end of NE book VII, whereas the second treatise on pleasure, found in the first chapters of NE book X, is rather to be taken as typically Nicomachean. Anyway, all this is in line with the attitude of moving pleasure and pain from inside to outside the definitional context of moral virtue. And this replacement is made transparent if we compare the two definitions we have of moral virtue, the one at EE II 10 and the one at NE II 6. To say it again from the vantage point of what is lacking in each definition: right reason and the way the person of practical wisdom determines the mean are not mentioned in the Eudemian version, while the connection with pleasure and pain is absent from the Nicomachean version, which stresses instead the role of reasoning in determining the right mean. It is surely a question of emphasis, but quite often philosophical issues hinge on giving or refraining from giving emphasis.

Another sign of a change of emphasis is this. In both Ethics, when discussing the connection between pleasure and virtue, Aristotle pauses to mention a pervasive attitude he finds in philosophers to define moral virtue as a sort of state of impassivity and rest. In the EE, he writes as follows:

It is on account of pleasures and pains that we call men vicious, for pursuing and avoiding in the wrong way or the wrong ones. That is why everyone tries προχείρως to define the virtues as impassivity or rest in the matter of pleasures and pains, and the vices as the opposite.

(EE II 4, 1222a1–5)15

Much hinges on the translation we give to προχείρως, which I left untranslated. It can mean readily or offhand, meaning a sort of spontaneous tendency people have when speaking about the nature of virtues. In this sense, it is rather neutral: this tendency may be wrong, or may not be wrong. But προχείρως can also mean hasty or hurriedly, in which case it carries a negative sense, indicating something which requires some correction, for it has been done hastily or carelessly. Nothing forces one to go either way, but if one takes into account the fact that the EE is more akin to dialectical contexts, and that dialectical arguing supposes that all people lean somehow the, or at least are able to recognize it as soon as they see it, it seems preferable to translate it in a more neutral way: this is why people naturally think of virtues as states of impassivity, for pursuing bad pleasures or pursuing them in the wrong way seems to be the cause of vice in them.

The NE strikes a pretty similar note, but nonetheless it sounds slightly different. After having said that people become bad by pursuing and avoiding pleasures and pains, either the pleasures and pains they ought not to or when they ought not to or as they ought not, Aristotle concludes as follows:

This is also why people define the virtues as certain states of impassivity and rest; not well, however, because they speak absolutely, and do not say ‘as one ought’ and ‘as one ought not’ and ‘when one ought or ought not’, and all the other things that may be added.

(NE II 3 1104b24–26)16

Maybe Aristotle is obliquely referring to Speusippus, or to Democritus, or to common theses in the Academy,17 or even to himself18 – we can no longer decide. He is, however, more cautious about its pervasiveness – he does not say “all people,” but only “(some) people.” And, more importantly, he is bluntly saying that they are wrong, and the reason they are wrong lies in the fact that they drop the clause “as one ought,” “when one ought” – those clauses right reasoning determines for each occasion. Maybe this is too subtle, but we see again pleasure and pain receding from the core of moral virtues on behalf of reasoning and the way in which the person with practical wisdom determines it, and this receding makes προχείρως now sound in a rather negative tone.

Unfortunately, the Magna Moralia (MM) is of little help in understanding how this change comes about. In MM I iv 10, Aristotle opens the section on moral virtue raising the question on its “what-it-is”:

After this, then, we must consider what virtue is

(MM I iv 10)19

In a first approach, moral virtue is seen as liable to be destroyed either by defect or by excess. Then another route is suggested, the one according to which one should define moral virtue in terms of pain and pleasure (I vi 1). As moral virtue is seen as a disposition to be correctly affected by emotions, and emotions “are either pains or pleasures or else do not come about without pain or pleasure” (I viiii 2: τὰ δὲ πάθη ἤτοι λῦπαί εἰσιν ἢ ἡδοναὶ ἢ οὐκ ἄνευ λύπης ἢ ἡδονῆς), we are to conclude that

It also clear from these considerations that, then, virtue is concerned with pains and pleasures

(MM I viii 2)20

This is all too close to the Eudemian approach, but does not shed any light on it. Moreover, the MM introduces the idea of a disposition issuing in decision only when discussing not moral virtue, but practical wisdom (see MM I xxxiv 11).

Much more rewarding is the way Aristotle examines the particular virtues, especially the first one, courage. In the Eudemian version, courage is connected to pleasure and pain in so far as it is relative to what is fearful: the courageous person faces what is fearful, while the coward is quickly and intensely frightened by any danger, and the reckless is overconfident in cases when one should be fearful. The definition of fearful things is

The fearful is, in general, what is productive of fear, and that in turn is whatever is manifestly capable of producing pain that is destructive

(EE III 1, 1229a33–35)21

Fearful things are restricted to those things that can bring about death; one might get a different type of painful feeling when, for instance, one experiences the pain of jealousy, but this is not what fear precisely means, nor the pain it brings about. The courageous person, then, has a certain attitude towards that kind of pain that appears to be imminent and which is capable of destroying life, that is, he is rightly disposed towards death and the pain of death as he faces it. Courage, then, is concerned with those fearful things that are capable of producing pain of a destructive sort, provided that they are imminent rather than distant. As such, death cannot be pleasant, and consequently no one who faces death because of the prospect of some pleasure, or even to avoid greater pains, could justly be said to be courageous:

Some even endure danger because of other pleasures. Spirit, after all, brings pleasure of a sort, since it comes with the hope of revenge. Still, no one who faces death because of this or some other pleasure, or to avoid greater pains, could justly be said to be courageous.

(Inwood, Woolf’s; EE III 1, 1229b30–34)22

There is no pleasure in acting courageously, for to act courageously is tantamount to facing what is deprived of any pleasure, to wit: death itself. Why should then the courageous person face death, if no pleasure is involved in an episode of courage? The courageous person endures the pain indeed, and they endure it because “reason bids them to choose what is fine” (EE III 1, 1229a2: ὁ δὲ λόγος τὸ καλὸν αἱρεῖσθαι κελεύει). Aristotle brings in the connection between being virtuous and doing what is fine via the notion of a rational decision in order to explain why the courageous person faces what is totally deprived of pleasure:

Now given that all virtue issues in decisions (we have said previously what we mean by this – virtue makes everyone choose for the sake of something, and this ‘for the sake of which’ is what is fine)23, it is clear that courage too, being a virtue, will make us endure what is fearful for the sake of something, and that will be due neither to ignorance (since virtue rather makes us judge correctly) nor to pleasure, but because doing so is fine, because, if it were not fine but foolish, one does not endure danger, since that would be shameful.

(EE III 1, 1230a26–33)24

Aristotle is referring back to EE II 7, 1227b35–28a7, where he says that decision is Janus-faced, for it has a two-side structure: one deliberates and decides about the means, but the means cannot but be means to an end, for the sake of which every deliberation and decision are taken. Decision is of something (a mean) for the sake of another thing (the end). However, he has not said in this passage that the end is the fine whenever the action is virtuous. This is something he says only when examining moral virtues one by one, beginning in III 1 with courage. And it seems that it is this new approach, in which doing something virtuously is tantamount to doing it for the sake of the fine, that comes to the fore and makes the presence of pleasure or pain recede as the essential feature when one acts virtuously, at the same time as reason takes a more central role in acting virtuously.

In the Nicomachean analysis of courage, the idea of acting courageously for the sake of the fine is so prominent that this idea manages to render the death itself the courageous person faces in battle the “noble death,” τὸν καλὸν θάνατον (NE III 6, 1115a33). The “for the sake of the fine,” τοῦ καλοῦ ἕνεκα, is ubiquitous in his analysis of courage: what is fearful is defined in terms of what is fine to endure (III 7, 1115b11–15); the courageous person endures danger for the sake of the fine (1115b23–24); he chooses the fine and endures danger (1116a11–12); he does this because it is fine to do it and shameful not to do it (8 1117a17). Aristotle generalizes the point a bit later, when examining liberality, and asserts that every moral virtue is directed towards what is fine as every moral action is done for the sake of the fine:

Now moral actions done on the basis of a virtue are fine and for the sake of the fine

(NE III 1, 1120a23–24)25

I think this theme, doing something virtuously insofar as one does it for the sake of the fine, becomes philosophically much more salient to Aristotle’s eyes than the idea that moral virtue is deeply connected to pleasure and pain. Moral action continues to have some connection with pleasure and pain, but this connection no longer appears inside the definitional phrase of moral virtue, it is no longer an essential feature of it. From now on, doing what is fine is the essential feature, and this is captured by the clause of “doing as one ought, when one ought, how one ought to do it,” or, to sum up, as reason bids us do, as reasoning strikes a balance between excess and deficiency. Acting for the sake of the fine and weighing the pros and cons of an action go hand in hand. One may wonder whether this change is brought about by a stronger realization that morality lies in rationality, and moreover that Aristotle chooses the fine instead of the good because the latter was too closely linked to personal advantage and usefulness, while the former was more apt to convey an idea of unselfishness. This all sounds quite plausible, and most likely is being broached here in a pregnant way for the first time. But I want to stress the more pedestrian idea that this change was occasioned by his growing acknowledgement that moral virtue is better understood when detached from the deep and constant company of pleasure people normally attached to it, and this was done in close connection with a better reassessing of the importance of acting as reason bids us to do, a better reassessing which was prompted by the study of the moral virtues one by one, especially because this study begins with courage. A sign of this is the fact that, in the NE, Aristotle justifies his close analysis of the particular virtues by saying

But we should not simply state this [scl.: what moral virtue is] in general terms; we should also show how it fits the particular cases. For with discussions that relate to actions, those of a general have a wider application, but those that deal with the subject bit by bit are closer to the truth.

(Rowe’s; NE II 7, 1107a28–31)26

They are closer to the truth indeed. We can see this movement of unyoking the pleasure from morality in other domains as well – the most significant one being the rejection of the life of pleasures as an acceptable way of life, performed already in NE book I, but cemented at NE X 1–5. The official sign of this disconnecting, a sort of weaning from pleasure, is the fact that pleasure and pain no longer figure in the definition of moral virtue. But the ties were strong between pleasure and morality, and Aristotle seems to be a bit regretful to have finally cut them. This may explain why, in two passages of the Nicomachean Ethics, he is keen to recall that there is also pleasure in the courageous action as one faces death (NE III 9, 1117a35–6, 17b15–16). I quote only the first, as the second one is a reminder of the first one:

Yet the end which courage sets before it would seem to be pleasant, but to be concealed by the attending circumstances, as happens also in athletic contests; for the end at which boxers aim is pleasant – the crown and the honours – but the blows they take are distressing to flesh and blood, and painful, and so is their whole exertion; and because the blows and the exertions are many the end, which is small, appears to have nothing pleasant in it (Solomon’s translation).

(NE III 9, 1117a35–b6)27

Pleasure is always involved in moral action, but its stronghold on well-doing is now weakened. It is no longer within the definition of moral virtue, but without. Still, it remains in the vicinity, as a vestige of its ancient definitional role, as we can glean from this passage. The disconnecting, however, of morality from pleasure is definitely sealed when pleasure and pain no longer figure in the very definition of moral virtue, as one can see from the contrast between the Nicomachean and the Eudemian versions of what moral virtue is. To compensate for the blank left in the definition by the unyoking of pleasure from moral virtue, Aristotle fills in the notion of acting for the sake of the noble when examining the virtues case by case. In these analyses, rationality – acting as reason bids us do – comes to the fore as that which the agents recognize as what they ought to do, however unpleasant the consequences of their action may prove to be – as is particularly the case of courageous actions of facing death on the battlefield.28

Notes

1. See Natali, 2017 for the unfolding of the Nicomachean argument in a sort of ring construction governed by the definition of happiness.

2. See Metaph. Δ 16. Aristotle lists three senses: (i) that outside of which no portion is to be found; (ii) that which cannot be surpassed relative to its genus; (iii) that which has reached its fulfilment. In the recapitulation of the senses listed in the chapter, Aristotle reduces them to two cases: (a) being complete and (b) being perfect. Sense (i) is the same as (a), and senses (ii) and (iii) are contained in (b), but in different ways: sense (ii) is linked to the relative superlative, and sense (iii) is connected to the absolute superlative.

3. In the Nicomachean version, it is quite clear that they are not two independent conditions, but the first implies the second, for whatever be the sense of τελεία in κατὰ τὴν ἀρίστην καὶ τελειοτάτην (either complete or perfect in both senses of perfection), it will be one such that its achievement requires time and completeness of life in the relevant sense, as no children can have “perfect” or “complete” virtue, but only natural virtue.

4. On the argument deployed at NE X 6–9, I refer to my treatment of this issue in Zingano, 2017, which improves upon my previous treatment in Zingano, 2014.

5. Except for some parts of book V 1, 1129b30–1130a1 and VI 12, 1144a3–6.

6. One might also wonder whether ὡρισμένη is to be taken in the sense of “being defined” or in the sense of “being determined.” I think ὡς ἂν ὁ φρόνιμος ὁρίσειεν in the next clause settles its meaning as “being determined”: the person with practical wisdom is not concerned as such with defining what moral virtue this (this is the philosopher’s task), but with determining what one ought to do in the current situation.

7. 48, 13–15: ὅτι δὲ ὁ ὁρισμὸς ὀρθῶς ἀποδέδοται, δῆλον. καὶ γὰρ ἕξις οὖσα δέδεικται καὶ προαιρετική, εἴτε προαίρεσίς ἐστιν εἴτε μή· δέδεικται δὲ καὶ ἐν μεσότητι οὖσα οὐ κατὰ τὸ πρᾶγμα ἀλλὰ τῇ πρὸς ἡμᾶς.

8. See Natali (op. cit.) on this way of construing the sentence.

9. Bekker’s text supposes as structure “a is b, b is tantamount to c, b is obtained through d, and d is explained by means of how e operates,” whereas Bywater gives “a is b, b is tantamount to c, c is obtained through d, and how e operates is explained by d”; in the former case, the third moment is backward-looking; in the latter, the fourth and last moment is backward-looking. In contrast, Susemihl–Apelt’s version is always forward-looking: “a is b, b is tantamount to c, c is obtained through d, and d is explained by means of how e operates.”

10. μετὰ ταῦτα σκεπτέον πρῶτον περὶ ἀρετῆς ἠθικῆς, τί ἐστι καὶ ποῖα μόρια αὐτῆς (εἰς τοῦτο γὰρ ἀνῆκται), καὶ γίνεται διὰ τίνων.

11. ὥστ’ ἀνάγκη τὴν ἠθικὴν ἀρετὴν περὶ μέσ’ ἄττα εἶναι καὶ μεσότητα τινά.

12. ἀναγκαῖον ἂν εἴη τὴν ἠθικὴν ἀρετὴν τὴν καθ’ αὑτὸν ἕκαστον μεσότητα εἶναι ἢ περὶ μέσ’ ἄττα ἐν ἡδοναῖς καὶ λύπαις καὶ ἡδέσι καὶ λυπηροῖς.

13. I printed the text of both families according to Rowe, who prefers to read ἀναγκαῖον ἂν εἴη τὴν ἠθικὴν ἀρετὴν τὴν καθ’ αὑτὸ ἑκάστην μεσότητα εἶναι καὶ περὶ μέσ’ ἄττα ἐν ἡδοναῖς καὶ λύπαις καὶ ἡδέσι καὶ λυπηροῖς, and explains in a note: “that each aretê is a μεσότης καθ’ αὑτὸ, i.e. not κατὰ συμβεβηκός (see, e.g. 1221b4), is rather more to the point than that it is a μεσότης in/in relation to each individual person, as the transmitted text would be saying” (C. Rowe, Eudemian Ethics, forthcoming).

14. ἀνάγκη τοίνυν […] τὴν ἀρετὴν εἶναι τὴν ἠθικὴν ἕξιν προαιρετικὴν μεσότητος τῆς πρὸς ἡμᾶς ἐν ἡδέσι καὶ λυπηροῖς καθ’ ὅσα ποῖός τις λέγεται τὸ ἦθος, ἢ χαίρων ἢ λυπούμενος. The clause “in respect of those pleasant and painful things by which the person is said to have a certain sort of character, according as one is pleased or pained” is not an explanatory, but a defining relative clause, for, as the text goes on to say, there are other things we are pleased or pained by without having thereby a special sort of moral character, such as liking what is sweet or what is bitter.

15. δι’ ἡδονὰς δὲ καὶ λύπας φαύλους εἶναι φαμέν, τῷ διώκειν καὶ φεύγειν ἢ ὡς μὴ δεῖ ἢ ἃς μὴ δεῖ. διὸ καὶ διορίζονται πάντες προχείρως ἀπάθειαν καὶ ἠρεμίαν περὶ ἡδονὰς καὶ λύπας εἶναι τὰς ἀρετάς, τὰς δὲ κακίας ἐκ τῶν ἐναντίων.

16. διὸ καὶ ὁρίζονται τὰς ἀρετὰς ἀπαθείας τινὰς καὶ ἠρεμίας· οὐκ εὖ δέ, ὅτι ἁπλῶς λέγουσιν, ἀλλ’ οὐχ ὡς δεῖ καὶ ὡς οὐ δεῖ καὶ ὅτε, καὶ ὅσα ἄλλα προστίθεται.

17. See for instance Topics IV 5, 125b22–23.

18. See Phys. VII 3, 246b19–20: ἡ μὲν γὰρ ἀρετὴ ποιεῖ ἢ ἀπαθὲς ἢ ὡδὶ παθητικόν, ἡ δὲ κακία παθητικὸν ἢ ἐναντίως ἀπαθές (“virtue disposes its possessor to be unaffected or to be affected thus and so, while vice disposes its possessor to be affected or to be unaffected in a contrary way”).

19. Μετὰ ταῦτα τοίνυν λεκτέον ἂν εἴη τί ἐστιν ἡ ἀρετή.

20. ἄρα ἀρετή ἐστιν περὶ λύπας καὶ ἡδονάς, καὶ ἐντεῦθέν ἐστι δῆλον.

21. ὅλως μὲν οὖν φοβερὰ λέγεται τὰ ποιητικὰ φόβου. τοιαῦτα δ’ ἐστὶν ὅσα φαίνεται ποιητικὰ λύπης φθαρτικῆς.

22. ἔνιοι δὲ καὶ δι’ ἄλλας ἡδονὰς ὑπομένουσιν. καὶ γὰρ ὁ θυμὸς ἡδονὴν ἔχει τινά· μετ’ ἐλπίδος γάρ ἐστι τιμωρίας. ἀλλ’ ὅμως οὔτ’ εἰ διὰ ταύτην οὔτ’ εἰ δι’ ἄλλην ἡδονὴν ὑπομένει τις τὸν θάνατον ἢ φυγὴν μειζόνων λυπῶν, οὐδεὶς δικαίως <ἂν> ἀνδρεῖος λέγοιτο τούτων.

23. One may translate καὶ τοῦτό ἐστι τὸ οὗ ἕνεκα τὸ καλόν slightly differently, depending on the way one parses it: “and this is the for the sake of which, the fine,” instead of “and this for the sake of which is what is fine.” I prefer the latter version because this is the first time the fine is explicitly introduced as a moral notion in the EE, and it would come as a surprise to find it simply in apposition to the for the sake of which.

24. ἀλλ’ ἐπειδὴ πᾶσα ἀρετὴ προαιρετική (τοῦτο δὲ πῶς λέγομεν, εἴρηται πρότερον, ὅτι ἕνεκά τινος πάντα αἱρεῖσθαι ποιεῖ, καὶ τοῦτό ἐστι τὸ οὗ ἕνεκα τὸ καλόν), δῆλον ὅτι καὶ ἡ ἀνδρεία ἀρετή τις οὖσα ἕνεκά τινος ποιήσει τὰ φοβερὰ ὑπομένειν, ὥστ’ οὔτε δι’ ἄγνοιαν (ὀρθῶς γὰρ μᾶλλον ποιεῖ κρίνειν) οὔτε δι’ ἡδονήν, ἀλλ’ ὅτι καλόν, ἐπεί, ἄν γε μὴ καλὸν ᾖ ἀλλὰ μανικόν, οὐχ ὑπομένει· αἰσχρὸν γάρ.

25. αἱ δὲ κατ’ ἀρετὴν πράξεις καλαὶ καὶ τοῦ καλοῦ ἕνεκα.

26. Δεῖ δὲ τοῦτο μὴ μόνον καθόλου λέγεσθαι, ἀλλὰ καὶ τοῖς καθ’ ἕκαστα ἐφαρμόττειν. ἐν γὰρ τοῖς περὶ τὰς πράξεις λόγοις οἱ μὲν καθόλου κοινότεροί εἰσιν, οἱ δ’ ἐπὶ μέρους ἀληθινώτεροι. The MSS oscillate in line 1107a30 between κοινότεροι (“a wider application”) and κενώτεροι (“are emptier,” the reading Susemihl–Apelt retain).

27. οὐ μὴν ἀλλὰ δόξειεν ἂν εἶναι τὸ κατὰ τὴν ἀνδρείαν τέλος ἡδύ, ὑπὸ τῶν κύκλῳ δ’ ἀφανίζεσθαι, οἷον κἀν τοῖς γυμνικοῖς ἀγῶσι γίνεται· τοῖς γὰρ πύκταις τὸ μὲν τέλος ἡδύ, οὗ ἕνεκα, ὁ στέφανος καὶ αἱ τιμαί, τὸ δὲ τύπτεσθαι ἀλγεινόν, εἴπερ σάρκινοι, καὶ λυπηρόν, καὶ πᾶς ὁ πόνος·διὰ δὲ τὸ πολλὰ ταῦτ’ εἶναι, μικρὸν ὂν τὸ οὗ ἕνεκα οὐδὲν ἡδὺ φαίνεται ἔχειν.

28. The topic of acting for the sake of the noble has been brought to the attention of contemporary commentators by Terence Irwin in his seminal paper, see Irwin, 1985; Irwin revisits the same topic in light of the discussions raised since the publication of his first paper in Irwin, 2011.

References

1. Irwin, T., ‘Aristotle’s Conception of Morality’, Boston Area Colloquium in Ancient Philosophy, 1, 1985, pp. 115–143.

2. ———, ‘Beauty and Morality in Aristotle’, in Miller, J. (ed.), Aristotle’s Nicomachean Ethics, Cambridge: Cambridge University Press, 2011, pp. 239–253.

3. Natali, C., Il Metodo e il Trattato, Roma: Edizioni di storia e letteratura, 2017.

4. Destrée, P. and Zingano, M. (eds.), Theoria – Studies on the Status and Meaning of Contemplation in Aristotle’s Ethics, Leuven: Peeters, 2014.

5. Zingano, M., ‘Eudaimonia, Razão e Contemplação na Ética Aristotélica’, Analytica 21, 2017, pp. 9–44.

If you find an error or have any questions, please email us at admin@erenow.org. Thank you!