7
Mi-Kyoung Lee
7.1 Introduction
Aristotle’s two treatises on ethics, Nicomachean (EN) and Eudemian Ethics (EE), have three books in common (EN books V–VII = EE books IV–VI), the so-called “common books,” one of which is a book on justice (EN book V = EE book IV).1 Aristotle’s treatment of justice is the most detailed treatment of any of the virtues in either the EN or the EE. At the same time, it is less polished than the treatment of the other virtues in the undisputed books of the Nicomachean Ethics and in the Eudemian Ethics and shows signs of editorial compilation.2 It is also the only treatment of a moral virtue common to the EN and the EE; for the others, the EN and the EE have independent treatments. The book on justice is commonly read as part of the Nicomachean Ethics, and only rarely considered in the context of the Eudemian Ethics. In this chapter, I will consider the book on justice in the context of both Ethics, and will consider what the Eudemian Ethics says about justice if we read it with the common book on justice.
The question of which treatise the common books belong to is difficult. The traditional view is that they belong to the Nicomachean Ethics, and that the Nicomachean Ethics is the later, and more mature work.3 But in 1978, Anthony Kenny argued, on the basis of stylometric analysis, that (i) the common books have more stylistic affinities with the Eudemian Ethics, and furthermore, that (ii) the Eudemian Ethics is the later and more mature work.4 Kenny’s evidence for Claim (i) has attracted some consensus that the common books originated in the Eudemian Ethics, whereas Claim (ii) has not. For it is compatible with Claim (i) that the Nicomachean Ethics is the later, more mature work; the common books may have been works in progress first written at the same time as the EE but revised for inclusion in the Nicomachean Ethics. Recently, Oliver Primavesi has offered evidence that runs counter to Claim (i); he argues that the common books were originally part of the manuscript of the Nicomachean Ethics, and that they were only later added by editors to the undisputed books of the Eudemian Ethics (I–III, VII–VIII) to fill a gap. Primavesi’s evidence is based on the book numbers in the manuscripts and catalogues for the Nicomachean and Eudemian Ethics; they present strong though indirect evidence that the common books were in the manuscripts of the Nicomachean Ethics, as they were assembled after Aristotle died, not in the manuscripts of the Eudemian Ethics when that work was catalogued by editors in the Hellenistic period.5 As Dorothea Frede rightly points out, this does not settle any questions about the order in which Aristotle wrote his ethical treatises – and thus any questions about which treatise the common books belong to must depend not only on stylometric and historical evidence, but also on focused studies concerning the similarities and divergences in doctrine between the common books and the undisputed EE and EN. Frede has accordingly put forward considerations in support of the thesis that the common books (CB V–VII) fit doctrinally better into the Nicomachean Ethics, and are inconsistent on important philosophical points with the Eudemian Ethics. In this chapter, I will offer additional support to Frede’s line of argument concerning the common book on justice by considering in more detail and with additional evidence what the Eudemian Ethics together with the common books have to say about justice. For reasons of space, I cannot do a complete three-way comparison of CB V, the EE, and the EN on justice, though I will regularly bring in the EN when it helps to illuminate the way CB V fits or fails to fit with the EE.
As I will argue, the undisputed books of the EE are ignorant of many important ideas in the common book on justice. In the EE, Aristotle never alludes to his distinction between two kinds of justice, and it is often unclear whether he has in mind general or particular justice. He does not call justice “complete virtue” as he does in the common book on justice, and bestows that epithet upon kalokagathia “nobility-and-goodness” (EE VIII 3). In EE VII, he seems to be unaware of his solution in the common book on justice to the problem of whether injustice to the self is possible. He is also inclined to analyze justice as a mean between taking too much and taking too little; he has not yet come to recognize the objection that taking too little is not a sign of a defective disposition of which greed is the excess. However, the treatment in CB V 8 of voluntariness in just and unjust actions is on several points closer to the EE than to the EN.
7.2 Five Theses about Justice in CB V
I will proceed by listing (without much argument)6 five major arguments that Aristotle puts forward about justice in CB V, and I then consider whether the undisputed books of the EE are consistent with or whether they diverge from those points.
7.2.1 Thesis 1: The virtue dikaiosunê “justice” is a disposition to do what is just (to dikaion) and to do just actions (dikaiopragein)
In the opening lines of the common book on justice, Aristotle announces that
We see that all men mean by justice [δικαιοσύνην] that kind of state of character which makes people disposed to do what is just [ἀφ’ ἧς πρακτικοὶ τῶν δικαίων εἰσὶ] and makes them act justly [ἀφ’ ἧς δικαιοπραγοῦσι] and wish for what is just [βούλονται τὰ δίκαια].
(CB V 1, 1129a6–9)7
The point of this is sometimes obscured in translation, because of the ambiguity of the English term “justice,” which can refer to the personal virtue (“he is a just and righteous man”) or the abstract quality of justice (e.g. John Rawls’ two principles of distributive justice).8 However, the difference is clearer in the Greek terms: the abstract nouns to dikaion “the just” and to adikon “the unjust” (which are formed from the adjective dikaios, -a, -on “just” “right” and its opposite adikos, adikon “unjust” “wrong”) are distinct from the virtue terms dikaiosunê and adikia.9 To dikaion and to adikon refer to objects of just or unjust action: states of affair, distributions, allocations, or exchanges which are “just” or “unjust,” “fair” or “unfair.” By contrast, dikaiosunê and adikia and the personal adjectives dikaios and adikos are virtue terms belonging to persons (and by extension to a polis), characterizing them as being “just” or “unjust.” When, for example, Rawls defines justice he is not talking about the personal virtue, but justice “as the first virtue of institutions, as truth is of systems of thought” (Rawls, 1999, p. 3). Thus, far from being trivial or circular, the claim that dikaiosunê is the disposition to “do what is just” (and the corresponding claim that adikia is the disposition to “do what is unjust”) connects two concepts, that of the virtue term dikaiosunê and the moral term “to dikaion,” and shows that Aristotle recognizes that the notion of a fair or just outcome is prior to that of a fair or just person. This sets the terms of his project: to explain what to dikaion and to adikon are (to set out his “principles of justice,” so to speak), and then to define the virtue terms dikaiosunê and adikia (the “good of persons” as Rawls puts it)10 in terms of those concepts. This is exactly what he goes on to do: in CB V 1–5 he offers definitions of to dikaion, and then defines (in CB V 5) the concepts of just action and just character in terms of it.
In CB V 6 and 8, Aristotle makes the thesis more precise by distinguishing the following:
1. (i) Doing what is just (to dikaion)
2. (ii) Just action (dikaiopragein)11
3. (iii) Just action that speaks of a just character (dikaiosunê)12
Category (i) consists of performing some action type, such as obeying a law or correctly making a just distribution. But merely conforming to the law or bringing about a just distribution is not sufficient for acting justly (category [ii]), since one can obviously obey or disobey a law kata sumbebêkos, i.e. by accident, or unknowingly produce an unjust distribution. Just actions in category (ii) must be done knowingly and voluntarily. (I will return to this in Section 7.3.) To do a just act voluntarily in sense (ii) however is not the same as (iii) just action that speaks of a just character; for that, Aristotle says that one must act from prohairesis or “choice.” These distinctions capture the idea that merely complying with the law is easy but acting in a way that shows good character is more difficult.
This internal structure is unique to justice; for no other virtue does Aristotle make a threefold distinction corresponding to his distinction between justice as a state of affairs (to dikaion), justice in actions (dikaiopragein), and justice in character (dikaiosunê). Aristotle does not attempt to define courage (for example) in terms of doing what is courageous (ta andreia) or courageous action (andreiopragia?), or define sôphrosynê in terms of doing what is temperate (ta sôphrona).13 The reasons for this include, first, the fact that to dikaion and ta dikaia were common in everyday language, while the corresponding adjectives were not so used in the case of the other virtues; second, justice alone cannot be defined purely in terms of a state of the soul (e.g. passions, appetites, emotions), but has to be defined in terms of an agent’s intended object, and whether the agent correctly aims at what is just. One cannot know what the virtue dikaiosunê is without knowing what it is for an action or state of affairs to be dikaion; doing what is just and avoiding what is unjust are requirements for being just and unjust in character, respectively.14 This, in turn, implies that the just person acts on reasons and that those reasons involve a correct evaluation of what things are worth, what people deserve, and what they are owed. By saying that the just person is one who does just actions, or brings about just states of affair, Aristotle makes it essential to being a just person that one possess the intention to do what is just – the just person doesn’t do just actions unknowingly or by accident. Justice is not a matter of feelings or (mere) desires – one must have an awareness of something as being the just thing to do, and hence one must have a rational intention to do it, in order to qualify as being a just person. This is, of course, a consequence of the central role that practical reason and deliberation play in Aristotle’s theory of virtue; justice, more than any other moral virtue, requires practical reason and deliberative awareness of what is just and morally valuable.
In the undisputed books of the EE, however, Aristotle nowhere acknowledges this distinction between the virtue of justice which is a disposition, and just actions and states of affairs.15 Indeed, the EE says very little about dikaiosunê as a virtue. The term appears 15 times in the undisputed books, mostly to serve as an example of a virtue, not as an object of analysis. Notably, in his list of virtues in EN II 7, Aristotle lists the virtue dikaiosunê, whereas in the Eudemian Ethics he lists the term dikaion (EE II 3, 1221a4). This is the only virtue in the EE list for which he lists the abstract noun (or neuter adjective), and not the virtue term – without any explanation or apology. One might argue that his reason for listing the abstract noun dikaion instead of the virtue term dikaiosunê is because the former, not the latter, is a mean, and that he is here, already, registering his awareness that justice is anomalous. But then he does not clearly say so – and later in the chapter he explains the triad kerdos–zêmia–dikaion by naming the two associated vices as being profit-seeking (κερδαλέος) and loss-seeking (ζημιώδης) (EE II 3, 1221a23) – and not, as he holds in the common book on justice, the one opposite for to dikaion, namely, to adikon. In this respect, the EN list looks more thought through than the EE list. I will mention other problems with this triad below.16
7.2.2 Thesis 2: The justice terms are not univocal; there are two ways that something can be said to be just (dikaion) and correspondingly two kinds of justice (dikaiosunê)
The first type of justice has to do with what is required by the law: what is just is what is legal, and what is unjust is what is contrary to the law (CB V 1, 1129a32–1129b1, 1129b11–1130a13). Unjust actions are those which violate laws in our community that prevent us from doing harm to each other, such as theft, murder, fraud, rape, violent attack; these action types are “unjust” in the sense of being criminal and illegal. To be a just person in this sense is to be law-abiding and conform to moral and legal norms; to be unjust is to violate those norms. The second type of justice has to do with equality: what is just is fair or equal, and what is unjust is unfair or unequal (CB V 1, 1129b1–11, V 2, 1130a14–1130b5). The paradigm example of this type of injustice is an unfair or unequal division of some shared good. Someone who is just in this sense is “fair-minded,” aiming at, or respecting, fair distributions, allocations, or determinations between people, such as a fair and equitable judge. The two types of justice are usually labelled “general” and “particular,” because Aristotle describes them in terms of whole and part, so that general justice is the genus of “virtue” and particular justice is one species of virtue among many.17
The EE does not acknowledge the distinction between two kinds of justice; it is often unclear whether Aristotle is referring to justice in the general or particular sense. Sometimes it is clear that justice in the general sense is intended, e.g. at EE II 7, 1223b10–12, “the continent man will act justly (dikaiopragêsei), for continence is more of a virtue than incontinence is, and virtue makes people more just” – since Aristotle has in mind continence with respect to actions done from appetitive desire (epithumia), such as adultery or excessive drinking, these actions would be “unjust” in the general sense of breaking the law, since such actions do not fall under the remit of particular justice (CB V 2, 1130a24–32; for another example, see EE II 7, 1223a36). But in EE II 3, when Aristotle lists dikaion in his table of virtues, he fails to acknowledge Thesis 2, the distinction between two kinds of justice, and lists the triad kerdos–zêmia–dikaion, which evidently corresponds to justice in the particular sense. By contrast, in the EN II 7 list of virtues, he explicitly acknowledges that there are two kinds of justice: “As for justice, since it is a term used in more than one way, we shall distinguish its two varieties after discussing the other virtues and say how each variety is a mean” (tr. Crisp) (EN II 7, 1108b7–9).18 One might argue that we cannot put much weight on the fact that the list of virtues in the Eudemian Ethics lacks general justice, since EE II 3 is notoriously rough and has numerous difficulties.19 But EN II 7 does alert us to the distinction between two kinds of justice and is clearly looking forward to CB V 1–2 – which suggests that CB V 1–2 with its distinction between two types of justice was probably written after the EE, and was written for the EN. If we can speculate, Aristotle inherited the notion of particular justice from Plato, especially in the Republic, where Plato thinks of justice as a particular virtue and not as equivalent to virtue as a whole; furthermore, Platonic justice incorporates the notion of geometrical proportionality, of equals to equals, but this only applies to what Aristotle would call distributive justice.20 Thus, it would make sense that Aristotle should begin with an early precursor to his notion of particular justice in the EE; then, when he wrote the common book on justice for the EN, he realized that there is not one but two kinds of justice, which is the position reflected in EN II 7.21
7.2.3 Thesis 3: General justice is “complete” virtue, and is a kind of super-virtue, which encompasses all the others
In CB V 1–2, Aristotle says that general justice is “complete virtue, although not without qualification, but in relation to another [ἀρετὴ μέν ἐστι τελεία, ἀλλ’ οὐχ ἁπλῶς ἀλλὰ πρὸς ἕτερον]” – and then offers quotations to support the thought that it is universally thought to be “the greatest of the virtues” (CB V 1, 1129b25–1130a1).22
Aristotle concludes:
Justice in this sense, then, is not part of virtue but the whole of virtue, nor is the contrary injustice a part of vice but the whole of vice
(CB V 1, 1130a8–10)
Justice is therefore teleios “complete” or “perfect” in two senses: it is “complete” in the sense of encompassing or entailing all the other virtues, and it is also “perfect” in the sense that it is the fullest exercise and expression of the virtues, because the just person exercises them even with respect to others and not just himself.23 Whereas the first point makes general justice extensionally equivalent to possessing all the virtues, the second point implies that to be just in the general sense is an even greater achievement than it is to merely possess the other virtues.
Although there are a few passages in the EE where Aristotle may be referring to general justice in the sense of being generically equivalent to virtue (cited above), he does not recognize in the EE the privileged position that he gives to general justice in the passage just quoted from CB V 1–2 on account of its being “complete or perfect virtue.” For example, at EE II 1, 1219a35–9, Aristotle defines happiness as “activity of a complete life in accordance with complete virtue [ζωῆς τελείας ἐνέργεια κατ’ ἀρετὴν τελείαν],” with no mention of general justice here. Arguably the phrase – which also appears at EN I 7, 1098a17–18, I 9, 1100a4–5, I 13, 1102a6 – is used in different senses in different contexts, with no inconsistency.24 Furthermore, EE II 1 and EN I are preliminary, and we can’t yet expect answers to the question of what “perfect virtue” might be or entail. However, when in the final chapter of the EE, Aristotle names καλοκἀγαθία “nobility-and-goodness” as a super-virtue which he says is “complete virtue [ἀρετὴ τέλειος]” (EE VIII 3, 1249a16–17), thereby using the same expression that he uses to describe general justice in CB V 1, we might have expected some awareness that his candidate earlier in CB V 1 for “complete virtue” was dikaiosunê. If CB V 1 was originally EE book IV, then wouldn’t Aristotle have remembered in the conclusion to the book, EE VIII 3, that he had earlier given that crown to dikaiosunê and not to kalokagathia? Perhaps kalokagathia is simply the same as what Aristotle calls “general justice” in CB V 1. But he certainly never says so; in EE VIII 3 he is happy to regard justice and temperance as individual virtues that are parts of kalokagathia. Of course, EE VIII 3 is itself not well integrated into the rest of the EE – among other reasons because it introduces the topic of kalokagathia as though it has been discussed before (VIII 3, 1248b10–11), though this is the first mention of it in the EE.25 It’s possible that EE VIII 3 was not written as the conclusion for the EE, and that it, like other parts of Aristotle’s books, may have originally been an independent essay that was later attached to the EE.
7.2.4 Thesis 4: What is just (to dikaion) in the particular sense is what is equal
Aristotle subdivides particular justice into three kinds: distributive justice (CB V 3), rectificatory justice (CB V 4), and reciprocal justice (CB V 5).26 For all three, his main insight is that justice in the sense of fairness has to do with the correct apprehension of the values of goods and of the desert and merit in competing claims to those goods. Aristotle’s aim is to show that there is an objective basis to judgments about what is just and unjust, namely, that they all have to do with a kind of equality: distributive justice and reciprocal justice aim at proportional equality, rectificatory justice aims at arithmetical equality.
In order to do this, Aristotle must explain the sense in which what is just is a mean or intermediate. Accordingly, he argues in CB V 3–5 that for all three types of particular justice, the just is a mean because it is a kind of equality (ison ti).
1. The mean in distributive justice (to dianemêtikon dikaion) is proportional equality. For each division or distribution between two parties, the shares that each party receive should be proportional to the merit or desert of each of the parties, and the share of the thing divided which each receives. Aristotle concludes, “This, then, is what the just is – the proportional (τὸ ἀνάλογον); the unjust is what violates the proportion. Hence one term becomes too great, the other too small” (CB V 3, 1131b16–20). When we talk about fairness in distributions, we are aiming at equality of the proportional kind.
2. The mean in rectificatory justice (to diorthôtikon dikaion) is the arithmetical equal. Each rectification aims at the equal; thus, for example, if party A suffered an involuntary and undeserved loss at the hands of B, the rectification aims at “the equal,” namely, the restoration to A of what they lost from B. He concludes, that “the just is intermediate between a sort of gain and a sort of loss [ὥστε κέρδους τινὸς καὶ ζημίας μέσον], namely, those which are involuntary; it consists in having an equal amount before and after the transaction” (CB V 4, 1132b18–20). When we talk about justice or fairness in rectification, we have in mind that the compensation should be “arithmetically” i.e. exactly equal to the loss suffered by the victim.
3. The mean in reciprocal justice (to antipeponthos) is proportional equality (CB V 5, 1133a5–14): in trade exchanges or business transactions, in which one party A has rendered a benefit to another party B, B is obliged to return not a precisely equal return, but a return that is equally beneficial for A. When, therefore, we talk about justice or fairness in reciprocal exchanges, such as when two people are bartering, this aims at an equality in the value for each, as determined by need or demand.
Aristotle’s goal is to explain how judgments about fairness and justice are normative: they aim to correctly hit on the relevant kinds of equality for each case. Knowing how to fairly divide shares, or to fairly compensate someone for a loss, or to make a fair exchange – all of these concern a type of equality that the just person gets right and correctly aims at, and the unjust person gets wrong and misses.
The undisputed books of the EE do not show any awareness of these results in CB V. One can admittedly find, in the undisputed books of the EE, both the claims that
1. (i) Justice (to dikaion) is a mean (EE II 3, 1221a4)
and that
1. (ii) Justice (to dikaion) and friendship are a kind of equality (EE VII 9, 1241b11–13).
But Aristotle does not put the two together in the way that he does in the common book on justice: in the undisputed books of the EE he doesn’t recognize that justice is a mean because it is a kind of equality. This omission is particularly striking in EE VII 9, where one might have expected him to give a nod to the earlier treatment of justice as a kind of equality in the common book on justice. Its omission suggests (to me) that these represent two independent strands of thought, which he only attempted to put together in the common book on justice.
Furthermore, Aristotle’s understanding of Claim (i) in the EE, that justice is a mean, is simplistic compared with the way he understands it in the common book on justice. There, he analyzes to dikaion as a mean between profit (κέρδος) and loss (ζημία) (EE II 3, 1221a4), in a table listing the virtues together with the extremes between which each virtue is a mean. This triad is quite puzzling.27 For it suggests that the just outcome or state of affairs is always an intermediate between profit and loss. But why should it not be just (or at least not unjust) to sometimes earn a profit, and sometimes sustain a loss? In a fair transaction, is it not possible for both parties to gain from the exchange? Maybe he means by “profit” “more than one deserves” and by “loss” “less than one deserves.” What is just (to dikaion) would then be a mean between undeserved profit and undeserved loss. But then one wants to complain that “profit” does not mean “more than one deserves” (it means an advantage or gain relative to what one had before), and also that the notion of desert cries out for explanation. Such an explanation is forthcoming in CB V 4, 1132b11–20 but is not hinted at here in EE II 3.
Finally, the “profit–loss–justice” triad from EE II 3 is insufficiently general, since it doesn’t apply to general justice (general justice does not have a mean); even for particular justice, it only fits rectificatory justice (with some problems, as I just noted), not the other two forms, distributive and reciprocal justice. In the fuller argument in CB V 3–5, Aristotle uses the language of “profit–loss–mean” only for rectificatory justice. And he explains that rectificatory justice has to do with compensation for someone who has sustained an involuntary loss or injury from another party, one in which that second party is culpable: “the just is intermediate between a sort of gain and a sort of loss [κέρδους τινὸς καὶ ζημίας], namely, those which are involuntary; it consists in having an equal amount before and after the transaction” (1132b18–20). The judge’s task is to equalize the victim’s position, to make it what it was before the injury. Hence, the victim would wrongly “profit,” so to speak, if she were compensated too much by the injuring party and would suffer a “loss” if she were compensated inadequately. This is odd for a number of reasons (as Frede, 2019, p. 97 points out) – first because he assimilates all cases of injury to the language of monetary loss (odd to say that a murderer makes a profit, the murdered a loss!) – which is presumably why Aristotle remarks on the fact that the terminology comes from trade (rectificatory justice involves “a sort of gain and loss” [κέρδους τινὸς καὶ ζημίας], CB V 4, 1132b18). Again, it is odd to describe just compensation for one party (suppose Alex was assaulted by Ben) as being an intermediate between Alex’s profit (when Ben compensates too much) and Alex’s loss (when Ben fails to compensate or does not compensate sufficiently). In the common book on justice, Aristotle modifies the triad to fit the case of rectificatory justice, and drops it entirely from his account of distributive and reciprocal justice. I conclude, then, that the triad at EE II 3 very much looks like a “first stab” by Aristotle at characterizing justice before he worked out the details in the common book on justice.
7.2.5 Thesis 5: Just action is intermediate between doing injustice and having injustice done to one. But justice (the virtue) is not a mean between two vices, because having injustice done to one is not a vice. (Corollary: It is not possible to do an unjust action to oneself, and it is not possible to suffer injustice voluntarily.)
As I noted under Thesis 1, Aristotle distinguishes between (i) just outcomes or states of affair, (ii) just actions that aim to bring about such outcomes, and (iii) the virtue of justice which consists (among other things) of a disposition to act justly. In his treatment of particular justice, he analyzes the first item “what is just” (to dikaion and to adikon, CB V 3–5), and concludes by saying, “We have now defined the unjust and the just [Τί μὲν οὖν τὸ ἄδικον καὶ τί τὸ δίκαιόν ἐστιν, εἴρηται]” (CB V 5, 1133b29). Next, he defines (ii) just action (dikaiopragia) as follows:
These having been marked off from each other, it is plain that just action is intermediate between acting unjustly and being unjustly treated; for the one is to have too much and the other to have too little [διωρισμένων δὲ τούτων δῆλον ὅτι ἡ δικαιοπραγία μέσον ἐστὶ τοῦ ἀδικεῖν καὶ ἀδικεῖσθαι·τὸ μὲν γὰρ πλέον ἔχειν τὸ δ’ ἔλαττόν ἐστιν].
(CB V 5, 1133b30–2)
This is puzzling. First, it is not clear whether this is about particular justice, or whether this also applies to general justice. If the latter, then it is not clear whether and how the same action, described as being just (in the general sense) will have a mean that is compatible with its mean when it is described as (say) a courageous action. If the former, then the types of just actions Aristotle has been discussing in the common book on justice up to this point have to do with a judge’s determinations of what would be fair in distribution, rectification, etc. But when a judge makes an incorrect rectification, or a distributor makes an incorrect distribution of goods, surely the judge’s mistake is not to fail to hit the intermediate between acting unjustly and being unjustly treated, but to fail to hit the intermediate between the victim being compensated too much and too little. The judge being unjustly treated is not one of the extremes. Similarly, someone who makes an unfair distribution – of political authority, say – has not failed to hit the intermediate between acting unjustly and being unjustly treated. Rather, his mistake is to fail to hit what is just (to dikaion), which in this case is a proportional equality between divided shares. Why then does Aristotle conclude that just actions are intermediates between acting unjustly and being unjustly treated?
Either we have to suppose that Aristotle is still in the grip of the Eudemian Ethics’ triad – which holds that being just is an intermediate between gain and loss – even though Aristotle has substantially modified that claim in the common book on justice. Or, as I believe, we should simply abandon the assumption that Aristotle is trying to fit just action into the doctrine of the mean – or that he’s talking about the just actions of a judge – instead, he is evidently talking more generally about right action and wrongdoing. Recall Glaucon’s remarks about the origins of justice in Plato’s Republic:
[Justice] is intermediate [μεταξὺ] between the best and the worst. The best is to do injustice without paying the penalty; the worst is to suffer it without being able to take revenge. Justice is a mean between [ἐν μέσῳ ὂν] these two extremes.
(Rep. II, 359a5–8)
The sense in which justice is a “mean” in the Republic has nothing to do with Aristotle’s doctrine of the mean, where a mean is intermediate between two (vicious) extremes; instead, Glaucon is saying that it’s intermediate on a scale from best to worst. The very best option is the ability to do injustice with impunity; the very worst is to suffer injustice with no way to protect oneself and no possibility of revenge. Being just by adhering to the laws of civil society is a way of staving off the worst option, while forswearing the best option. In my view, Aristotle is making a similar, and utterly familiar, point, echoing Glaucon (with approval): a just action is in between doing injustice to others (by getting more than one should) and having an injustice done to one (by getting less than one should). But he is not attempting to say (as the doctrine of the mean would imply) that every just action is an intermediate between two extremes, both of which are unjust actions by that agent. For example, if Charles and Dana have made a contract, and Charles abides by that contract, his just action is intermediate between cheating her which might be better for him, and being cheated by her which would be worse for him. If Charles is cheated by Dana, he has what is unjust done to him – but this is not an unjust action that he has done. Hence the trio (i) Charles cheating Dana, (ii) Charles abiding by his contract with Dana, and (iii) Charles being cheated by Dana is not parallel to the more standard example of the doctrine of the mean, where (i) Charles acts rashly, (ii) Charles acts courageously, and (iii) Charles acts cowardly. For Charles being cheated by Dana is (obviously) not a vicious action by Charles, and is not a defective state of Charles.
This is confirmed when Aristotle goes on to explain how justice the virtue is a mean at CB V 5, 1133b29–1134a16. For, he says, justice the virtue (i.e. particular justice) is not like the other virtues because it is not an intermediate between two (extreme) vices; for justice has only one opposite, injustice. Hence, it is anomalous with respect to the doctrine of the mean, which holds that each virtue is intermediate between two vices. Dikaiosunê is an intermediate only in the sense that it aims at an intermediate, namely, to dikaion which is a mean between two extremes. Thus, justice the virtue relates to the intermediate by producing or aiming at it (to dikaion), and injustice relates to the extremes by producing or aiming at too much or too little relative to the mean.
This allows us to address a famous complaint against Aristotle that goes back to Hugo Grotius, namely, that Aristotle’s doctrine of the mean seems to imply that if someone gets less than his share voluntarily, he thereby willingly suffers injustice, or does an injustice to himself, which indicates that he has a defective disposition and is unjust.28 In CB V 9–11 – chapters of the common book on justice that are seldom read – Aristotle actually asks whether it’s possible to do an injustice to oneself, or to suffer injustice voluntarily. He considers the case of Glaucus and Diomede in Iliad VI 232–6, saying
one who gives what is his own, as Homer says Glaucus gave Diomede “Armour of gold for brazen, the price of a hundred oxen for nine” is not unjustly treated; for though to give is in his power, to be unjustly treated is not, but there must be someone to treat him unjustly. It is plain, then, that being unjustly treated is not voluntary.
(CB V 9, 1136b9–14)
His point is that if you voluntarily make a bad bargain, as Glaucus does when he exchanges his own golden armour for his guest-friend Diomedes’ bronze armour, there is no injustice, and he does not do an injustice to himself or suffer an injustice, since he voluntarily and knowingly agrees to the exchange. Someone else must do the injustice to him.29 Later, Aristotle makes this point more explicitly: one cannot do an injustice to oneself, whether by getting a bad bargain in a transaction, or even in suicide when one harms oneself intentionally (CB V 11, 1138a4–28) – for one thing, no one has a boulêsis “rational desire” for being unjustly treated, which is a requirement for actions done from choice (CB V 9, 1136b3–9, b23–5). To choose to do an injustice to oneself requires that one have a boulêsis for suffering injustice, and Aristotle thinks that no one has such a desire – not even someone who is willing to do harm to himself (e.g. suicide), or someone who is willing to endure an injustice in order to avoid a worse evil (e.g. protecting a friend from a tyrant even at the cost of one’s life). Furthermore, epieikeia (often translated “equity,” though that is inadequate) is the virtue of one who takes less than his share, or less than he is owed, this being the mark of a “decent” or epieikês person whose actions exemplify the virtue of supererogation (CB V 10).
In sum, Thesis 5 holds that the virtue of justice is not an intermediate between two vices, but has only one opposite, namely, injustice. When in CB V 5 Aristotle explains how the virtue of justice and the vice of injustice are intermediates, he correctly sees that the intermediacy should be located in the object of justice, not in justice itself. What is just is an intermediate in the sense spelled out in Thesis 4. Aristotle’s decision to treat dikaiosunê and adikia differently from the other virtues is raised as a problem for the theory by Grotius, since it implies that the doctrine of the mean is not perfectly general. And it must have been embarrassing for Aristotle to discover that in the case of the central virtue (justice!) the triad of excess–mean–deficiency does not apply. If the doctrine of the mean is meant to explain all the virtues, we would expect that general and particular justice as dispositions must be “intermediates.” But it is to his credit that he comes to realize this over the course of the common book on justice.
The undisputed books of the EE do not acknowledge Thesis 5 – and indeed at one point Aristotle says something inconsistent with it. For in EE II 3, Aristotle explains what it is to be a just person by reference to this analysis of the just as intermediate between profit and loss. For he goes on to explain that “The profiteer is the one who tries to get more from every situation, the loss-maker is the one who never does so other than occasionally [κερδαλέος δὲ ὁ πανταχόθεν πλεονεκτικός, ζημιώδης δὲ ὁ μηδαμόθεν, ἀλλ’ ὀλιγαχόθεν]” (EE II 3, 1221a23–4); the profiteer is “pleonectic,” wanting to get the larger share of any good, in cases where he is not entitled to get it. Thus, presumably, the just person gets profits whenever it is appropriate, neither too much nor too little. (Or so the reader must guess; Aristotle does not spell out the conclusion regarding the just person [ho dikaios], since in EE II 3 he does not discuss any of the virtues, only pairs of vicious extremes.) But this represents the opposite of the profiteer as the zêmiôdês or “loser,” someone who habitually sustains a loss; the “loser” is by implication vicious, and has a defective disposition correlating with the vice of being kerdaleos “the profiteer.” But why should we suppose that the person who loses or sustains a loss is necessarily vicious? EE II 3 seems to assume that justice is a mean between two vices, and has not yet come to see the point that the loser is not doing an injustice and is not an unjust person. In the EN, Aristotle drops any reference to “the loser” and it does not show up again.30
Furthermore, at EE VII 6, 1240a14–20, Aristotle asks whether someone can be a friend to oneself, and notes that this question is like the question, whether anyone can commit an injustice against himself. (No such remark about whether injustice to oneself is possible is raised in the undisputed books of the EN.) Aristotle says that such questions can be settled “in the same way”: namely, by noting that insofar as a human soul is bipartite, one can make sense of “wronging oneself,” e.g. by the appetite “wronging” the rational part of the soul, but otherwise, one cannot. Fine – but this is not the solution offered in the common book on justice, which I described above.31 Again, note that Aristotle does not in EE VII 6 simply refer back to the discussion in CB V 9–11 (which would be EE book IV) or summarize the explanation given there, which suggests that the common book was never a part of the EE at all, or at least that the EE VII book on friendship was written independently of, and without any awareness of, the common book on justice.
7.3 Voluntariness in CB V 8
So far, most of the evidence we have considered suggests that the common book on justice fits better with the Nicomachean Ethics. This is not universally true, however, since his treatment of voluntary action in CB V 8 fits better with the EE than with the EN. In my view, CB V 8 resembles the EE account of voluntary action on a number of points, in ways that suggest that it is both an application of and a refinement of the EE account of voluntariness. These are refinements insofar as Aristotle is applying his account of voluntary action to the legal context in CB V 8, thus paving the way for the EN III 1–5 account of voluntariness.
A complete discussion of this topic would take us from justice to the complex question of Aristotle’s evolving views about voluntariness.32 There are two other passages in CB V 8 which bear on the question of how that book is related to the EE, but for reasons of space I will simply focus on what seems to me the most significant and telling point.33 The common book on justice treats cases of ignorance and cases of coercion as being involuntary (with the EE), and not as voluntary (with the EN). In CB V 8, Aristotle holds that among involuntary actions, some are pardonable, when they are done in and from ignorance (ἀγνοοῦντες … δι’ ἄγνοιαν ἁμαρτάνουσι), whereas they are involuntary but not pardonable when they are done in but not from ignorance (ὅσα μὴ δι’ ἄγνοιαν, ἀλλ’ ἄγνοοῦντες διὰ πάθος), “owing to a passion which is neither natural nor such as man is liable to” (CB V 8, 1136a5–9). That is, he holds that actions done in but not from ignorance – as for example when one acts in the grip of a passion such as anger or fear – are involuntary, but still inexcusable. Similarly, at EE II 9, 1225b8–10, Aristotle holds that
whatever a man does that is in his power not to do, and does it not in error and of his own volition, he must needs do voluntarily; this is what voluntariness is. What he does in error, and because of the error, he does involuntarily.
Therefore, acts done in ignorance (or “in error” in Kenny’s tr.) brought on by passion are involuntary. But Aristotle changes his mind about this in EN III 1, where he holds that actions done in though not from ignorance are voluntary (1110b24–27). Furthermore, the striking phrase in the CB passage “owing to a passion which is neither natural nor such as man is liable to [διὰ πάθος δὲ μήτε φυσικὸν μήτ’ ἀνθρώπινον]” (CB V 8, 1136a8–9) clearly echoes the EE account, in which Aristotle develops an account of the voluntary in terms of our natural and intrinsic impulses, and the involuntary as what is contrary to those natural and intrinsic impulses, and as originating from an external source. This idea is dropped in the EN.
Finally, at CB V 8, Aristotle holds that a person who repays a debt through fear does so involuntarily (1135b4) – which hearkens back to the EE account of involuntary actions as those which do not stem from one’s own nature or which one cannot forbear acting on. In the EE, coerced actions are not voluntary (EE II 8, 1225a2–19), whereas the EN holds that such coerced acts are voluntary, “though in the abstract perhaps involuntary; for no one would choose any such act in itself” (EN III 1, 1110a18–19). On these points, CB V 8 seems to be closer to the EE than to the EN, which holds that actions done from passion and in ignorance, and coerced actions are “voluntary,” and hence offers a wider definition of voluntary action than the one offered in the EE or CB V 8.
At the same time, CB V 8 makes significant additions to and advances beyond what EE II says about voluntary action – in ways that foreshadow the EN account of voluntary action. David Charles has explored this in detail and proposed “an intelligible route from EE II to NE III 1–5 via CB V 8” (Charles, 2012, pp. 24–26). On Charles’ account, the common book on justice introduces a legal context for questions of voluntariness, which the EE acknowledges with its mention of the aims of legislators (EE II 10, 1226b38) but does not go into; from there, Aristotle begins to rethink the prominence of the notion of nature in the EE, and replaces it with a more narrow and strict notion of choice (hairesis) as the relevant starting point for voluntary actions and states. Charles argues that EE II distinguishes between voluntary and involuntary actions by stressing that the latter are based on desires that are not natural (i.e. common to everyone) or ones that our nature is not able to bear. In the EE, Aristotle is not so much concerned with the question of how to assign responsibility especially in cases of ignorance and coercion. By contrast, Aristotle in CB V 8 is very much interested in the question of how to deal with ignorance and coercion – and that is a point on which CB V 8 paves the way for the EN.
7.4 Conclusion
In sum, certain parts of CB V fit better with the Nicomachean Ethics, others with the Eudemian Ethics. In my view, the Eudemian Ethics represents an early phase of Aristotle’s thinking about justice. The Eudemian Ethics seems unaware of the distinction between two kinds of justice. Aristotle accepts a simplified formula for understanding justice as a mean between profit and loss. His commitment to the idea that justice is intermediate between profit and loss may be a sign of the lingering influence of Plato’s way of thinking about justice in the Republic, where Plato has Glaucon put forward the thought – evidently spelling out what the many think about justice – that it is an “intermediate” between acting unjustly and being unjustly treated; Plato also seems to have in mind what Aristotle calls particular justice, though presumably Aristotle’s point is that Plato did not distinguish clearly between the two senses. Finally, Aristotle seems to think in the Eudemian Ethics that justice and friendship are both virtues of social life: many friendships revolve around just exchanges of goods and benefits between friends, and are dissolved over quarrels and disputes concerning what is just and fair.34
At some point, Aristotle got to work on his treatise on justice, and there, in the common book on justice, he began to develop new ideas and concepts of which the Nicomachean Ethics, but not the Eudemian Ethics, seems to be aware. These include Aristotle’s distinction between general and particular justice; his new conclusions about how justice fits into the doctrine of the mean; and his treatment of justice as a form of equality. The common book on justice therefore “belongs” to the Nicomachean Ethics in the sense that it is better integrated, doctrinally, with important points concerning justice than the Eudemian Ethics is.
These conclusions must be carefully qualified. Kenny’s stylometric data gives some evidence that on points of style, and vocabulary, the common books are more like the Eudemian Ethics; though I’m sceptical about some of Kenny’s data, the general impression is certainly that parts of the common book of justice may very well be coeval with the Eudemian Ethics. The main reason why the common book on justice cannot be connected tightly with either book remains the fact that it is itself loosely organized and shows signs of editorial compilation; that helps to explain why CB V 1–5 go better with the EN and CB V 8 with the EE without any inconsistency. It is probably hopeless to try to pinpoint exactly when Aristotle wrote the Eudemian and Nicomachean Ethics relative to each other, and relative to the common books – especially if the common book on justice was a work in progress, somewhere between lecture notes and a polished treatise. Even if one accepts my conclusions concerning justice, one will therefore want to be cautious about drawing conclusions about what it even means for the common books to “belong” to one treatise rather than another. Similar caveats are required for questions about developmentalism in Aristotle, and the extent to which Aristotle’s views about justice have undergone demonstrable change. Even so, one can say with some confidence that Aristotle’s theory of justice grew and developed, and shows signs of his willingness to rethink “received wisdom” and to add and revise in interesting and important ways to the base that he started with in the EE.35
Notes
1. In what follows I shall refer to the common books as CB and use their Nicomachean book number: CB V = Nicomachean Ethics book V = Eudemian Ethics book IV.
2. In CB V, chapters 1–2 have significant stretches that are repetitive, as do chapters 3–5; there are “doublets” repeating almost verbatim the same thoughts in these chapters. Chapters 6–8 are an assemblage of topics with some overlap; chapters 9–11 are loosely connected by a common thread of argument. Overall, CB V is not as polished as the introductory books for the EN and the EE, but even in the doublets, it is possible to discern important new points being made in the “double,” much as a professor might add new points to the same set of lectures. Jackson, 1879 reorganizes chapters 5–11 as do Gauthier and Jolif, 1970, but see Rowe, 1971, pp. 104–105 for a judicious analysis of the given text with reasons against any such reorganization.
3. In the 19th century, the Eudemian Ethics was held to be inauthentic by Schleiermacher and Spengel (thus, Susemihl’s 1884 Teubner edition is titled “the Ethics of Eudemus of Rhodes”). In the 20th century however, von der Mühll, 1909 and Kapp, 1912 established the authenticity of the Eudemian Ethics; Jaeger championed this and argued that it provides an important clue to Aristotle’s development since it is earlier and closer in philosophical doctrine to the Protrepticus than the Nicomachean Ethics (Jaeger, 1923 [Engl. tr. 1934/1948]). Some of Jaeger’s views were anticipated by Thomas Case, 1910 (reprinted 1996). For critical assessment of Jaeger’s developmentalist hypothesis, see Rowe, 1971, pp. 73–76; Bobonich, 2006, pp. 16–18.
4. Kenny adduces evidence based on the relative frequency of particles and other common terms in the EE, EN, and the common books. Many scholars accept Claim (i) (for discussion and references, see the Introduction in this volume by Di Basilio, n. 16, p. 13), but fewer accept Claim (ii) (for reservations see Irwin, 1980). For considerations about whether CB V fits better with the Eudemian or Nicomachean Ethics, see Kenny, 2016, pp. 6, 51, 55, 60–69, 276; Frede, 2019, pp. 95–99. Kenny identifies certain concepts used in the common book on justice, such as Aristotle’s use of concepts of arithmetical vs. proportional equality (Kenny, 2016, p. 276), his use of the part/whole distinction, and argues that these concepts and other concepts are used more frequently in the EE than in the EN (Kenny, 2016, pp. 60–69); to me, these seem inconclusive for establishing that the common book on justice “belongs” to the EE.
5. According to Primavesi, 2007, there were two sets of Aristotle’s works – and two different book catalogues for Aristotle – one for the Library in Alexandria (which used the later Hellenistic numbering system), and one for the older collection which was supposedly preserved in Scepsis (and whose dramatic rediscovery in the 1st century bc Cicero reports). Primavesi’s key finding is that the books of the EN follow the older system of numbering, whereas the manuscripts of EE’s books are numbered by the Hellenistic alphabetic numerals. Primavesi can therefore confirm Harlfinger’s careful preference for the EN as the original manuscript home of the disputed books, citing evidence from book marking (2007, pp. 70–73). (Harlfinger’s 1971 study of the manuscripts is described by Kenny as saying that the disputed books were as much a part of the manuscript tradition of the EE as they were of the EN. But this is not what Harlfinger concludes [as Frede, 2018 notes; see also Huby’s 1973 review of Moraux and Harlfinger, 1971].) The Hellenistic catalogue of Aristotle’s works only has a five-book Ethics, which must be (as Jaeger thought) the Eudemian Ethics minus the three disputed books (I–III, VII–VIII). After the rediscovery of the Nicomachean Ethics, its books V–VII were added to the five-book EE to become EE IV–VI. This, of course, does not imply anything about the order in which Aristotle wrote the EE, the EN, and the common books; it is possible that the EE was written after the EN, that the common books originally belonged to the EE, and that editors then took the common books and added them to the EN. But it would then be puzzling why the common books were missing from the oldest manuscripts of the EE. For a recent discussion of the manuscripts making up the Aristotelian corpus, see Hatzimichali, 2016.
6. I argue for this in more detail in my Justice in Aristotle’s Moral and Political Philosophy, under contract with Oxford University Press.
7. Translations of the EN are from Ross/Brown, 2009; translations of the EE are from Kenny, 2011, unless otherwise noted.
8. The same ambiguity can be found in German (Gerechtigkeit) and in French (la justice). The question of whether the Greeks have the notion of the abstract quality of justice (and whether dikê can ever have this meaning) is a classic debate going back to Havelock, 1969, Lloyd-Jones, 1971, Gagarin, 1974.
9. These, in turn, derive from an older word hê dikê which, by Aristotle’s time, is mostly used to mean “lawsuit.” In the EE and EN, Aristotle uses this term only five times, mostly to mean “lawsuit” or “trial” (EE VII 10, 1243a9, CB V 6, 1134a31, EN VIII 13, 1162b30, EN IX 1, 1164b13), and only once in the older sense of “justice” in a bit of poetry he quotes to illustrate an older notion of reciprocity as justice (CB V 5, 1132b27).
10. In Part III of Theory of Justice, Rawls defines justice as a good of persons (Rawls, 1999, pp. 380–386).
11. Aristotle’s terms for “just action [dikaiopragia]” (CB V 5, 1133b30) and “acting justly [dikaiopragein]” (twice in the EE, five times in the EN, nine times in CB V, once in the Rhetoric, and once in the Topics) are his own coinage, invented in order to have something corresponding to the term adikein “act wrongly” (e.g. Rhet. 1373b22 adikein kai dikaiopragein).
12. Cf. CB V 6, 1134a17–23, V 7, 1135a8–15, V 8, 1135a15–17, 1135b19–25.
13. Aristotle does use the expression ta andreia for “courageous actions” (MM 1.34.23.5) and to andreion in the sense of “manliness” or “courage” at Rhet. 1364b37 (which reflects its use elsewhere at Th. 2.39, E. Supp. 510, Andr. 683); he uses ta sôphrona for “temperate actions” (EN II 4, 1105a18 ff.), though he does not use to sôphron in the sense of sôphrosunê which reflects its use elsewhere at e.g. S. Fragm. 786, 683; E. Hipp. 431, Th. 1.37, 3.82. But none of these are the counterpart to Category (i) above.
14. Aristotle would still qualify as an “evaluational internalist” (cf. Slote, 2001) and not an “evaluational externalist” (cf. Driver, 2001), because he thinks that what is required for justice is to intend to aim at just outcomes, and not simply the external fact of one’s having brought them about.
15. With the possible exception of EE VIII 3, 1248b21: “justice, both itself and the actions based on justice” [δικαιοσύνη καὶ αὐτὴ καὶ αἱ πράξεις], though he does not use the vocabulary or conceptual distinctions that he draws so carefully in CB V 6 and 8.
16. It is well known that there are problems with the list of virtues at EE II 3 – e.g. its inclusion of phronêsis which is not a virtue of character. D. J. Allan proposes that the third column listing the virtues was an interpolation, and hence that phronêsis and dikaion were not originally on the list; in support of this is the fact that when he goes on to review the table, he only focuses on the vices, and not the virtues (Allan, 1966, p. 148). Di Basilio argues (in an unpublished paper “The Table at Eudemian Ethics II 3 1220b38”) that the list in EE II 3 may not have been a list of virtues at all, but a list of pathê, which would also make sense, especially given his introductory remarks at EE II 3, 1221a13–15 and a possible back reference to the list at EE III 7, 1234a23–28.
17. Aristotle argues that there is a particular virtue of justice distinct from general justice at CB V 2, 1130a14–1130b5. Although he does not use the terms “general” and “particular,” he does use the “whole”/“part” language characteristic of the genus/species relation at CB V 1, 1130a8–10, V 2, 1130a14–16, 1130b6–32 passim.
18. Cf. Frede, 2019, pp. 95–96.
19. Many thanks to Daniel Wolt for helpful discussion. Another explanation (suggested to me by Giulio Di Basilio) would be that the list at EE II 3 is meant to be a list of examples.
20. Many thanks to Dorothea Frede for help with these points.
21. Arguably, Aristotle is doing something similar when he carves out a notion of orexis as distinct from epithumia, and when he carves out a notion of hypolêpsis as distinct from doxa (cf. Moss and Schwab, 2019, p. 11).
22. Cf. Lee, 2014.
23. Kenny, 2016, pp. 67–68 argues that Aristotle’s phrases – “perfect virtue,” “part”/“whole” of virtue – are Eudemian and not Nicomachean, but his statistics are contestable. For example, “perfect virtue” occurs only 3 times in the EE (II 1, 1219a37, 39, and VIII 3, 1249a16), and 4 times in the EN (I.7, I.9, I.10, I.13); likewise, meros appears 14 times in the EE, and 12 times in the EN.
24. Cf. Woods, 1982, p. 90. For example, Aristotle appears to think that complete virtue includes both the practical and theoretical virtues in EE II 1, 1220a2–6.
25. is found at Kalokagathia Politics I 13, 1259b34, EN IV 3, 1124a4, X 9, 1179b10; it (and its cognates, e.g. kalokagathos) is not found in the EE outside this chapter. EE book VIII 1–3 are three independent essays, and were probably added by the editor to the EE by way of conclusion. It is unlikely that EE VIII 3 was originally intended by Aristotle to serve as the conclusion for the treatise; see Wolt, 2022 who argues that the reason why the EE discusses , whereas the EN does not, is that the role played by as the unified expression of all the virtues, is taken over by phronêsis in the EN, and that EE VIII 3 was originally intended to supply a similar “capstone” argument that CB VI 12–13 plays in the EN. For more on in the EE, see Bonasio in this volume, though she does not comment on this question about teleios aretê in CB V 1 vs. EE VIII 3.
26. There is some uncertainty about whether he regards reciprocal justice as one of the subspecies, for it is not listed in the introduction to particular justice in CB V 2. But CB V 5 gives an analysis that narrows down the traditional notion of justice as reciprocity (“quid pro quo”) to the justice of fair business transactions (as when two people trade goods, or when a society uses currency as a medium of exchange).
27. Kenny, 2016, p. 214 notes the inconsistency, as well as the possibility of excising this triad as being a later addition to the text. But as Frede, 2019, p. 96 points out, if we excise it, then justice will not be on the list of virtues at all, which would be peculiar. As noted in n. 15 above, Allan, 1966 proposes that the third column listing the virtues could be a later addition, because the subsequent discussion does not mention any of the virtues, only the vices from the first two columns. If so, then the virtue for kerdos-zêmia is not justice, and “dikaion” was an editor’s (possibly mistaken) addition. It’s hard to imagine what other virtue Aristotle could have in mind though, unless it is, as Di Basilio has suggested, not a list of virtues in the first place. (Thanks to Daniel Wolt and Giulio Di Basilio for discussion.)
28. Grotius critiques Aristotle’s virtue-oriented approach to justice in the Prolegomena to his On the Law of War and Peace §44 in 1625 (Grotius, 2005, vol. 3, pp. 1757–1758); cf. Schneewind, 1990, pp. 46–48.
29. Aristotle is tempted by the thought that perhaps he gets something else out of the exchange besides worse armour – gratitude and the benefits of guest-friendship (CB V 9, 1137a1).
30. The terms kerdaleos and zêmiôdês do not occur in the EN, and only occur in the EE in this passage.
31. Admittedly, Aristotle does mention at CB V 11, 1138b5–12 that “metaphorically and in virtue of a certain resemblance there is a justice, not indeed between a man and himself, but between certain parts of him,” but this is not the main solution offered in V 9 and 11.
32. See Charles, 2012 for a detailed examination of CB V 8, and whether it fits better with the EE or the EN accounts of voluntariness.
33. There is a forward-looking reference in the Eudemian Ethics (EE II 10, 1227a2) to CB V 8, 1135b9–11, but no such reference in the EN III 1. However, there is no substantial difference since Aristotle makes the essential point at EN III 2, 1112a14–16. Also, Aristotle’s use of mê kata sumbebêkos at 1135a23–26 is more reminiscent of EE II 9, 1225b1 6 (though I find the latter very obscure), whereas in EN III 1 Aristotle drops the idea that actions done in ignorance are “incidental.”
34. Here again I cannot go into the details. Outside the common book on justice, Aristotle mentions justice most frequently in his books on friendship in the EE and EN; there, he maintains a close connection between friendship and justice (to dikaion), and this connection helps to explain some aspects of his treatment of justice – in particular, its orientation towards the common good, and towards the types of rules and principles people must live by in communities and groups that are held together by such a commitment to the common good. He thinks that justice and friendship are the twin goals of any community (koinônia); they are both virtues of people who live together in societies and communities, bound together by law aiming at the common good. Justice is a necessary condition of koinônia “community” as well as friendship (EE VII 9; EN VIII 9). In both the EE and the EN, he connects these in turn with equality; Aristotle says that justice (to dikaion) is a kind of equality (ison ti), and that philia too lies in equality (isotês) (EE VII 9, 1241b12–14).
35. I am very grateful to Giulio Di Basilio, Dorothea Frede, Peter Hunt, Gagan Sapkota, Gisela Striker, and Daniel Wolt for helpful comments and discussion. I also thank Paula Gottlieb for sharing with me some of her forthcoming book on the Eudemian Ethics.
References
1. Allan, D. J., ‘Review of Eudemische Ethik by Franz Dirlmeier’, Gnomon 38, 1966, pp. 138–149.
2. Bobonich, Christopher, ‘Aristotle’s Ethical Treatises’, in Kraut, R. (ed.), The Blackwell Guide to Aristotle’s Nicomachean Ethics. Malden, MA; Oxford: Wiley-Blackwell, 2006, pp. 12–36.
3. Case, Thomas, ‘Aristotle’, in Encyclopaedia Britannica (1910); reprinted in Wians, W. (ed.), Aristotle’s Philosophical Development. Lanham, MD: Rowman and Littlefield, 1996, 1–40.
4. Charles, David, ‘The Eudemian Ethics on the “Voluntary”’, in Leigh, F. (ed.), The Eudemian Ethics on the Voluntary, Friendship, and Luck: The Sixth S.V. Keeling Colloquium in Ancient Philosophy. Leiden; Boston: Brill, 2012, pp. 1–28.
5. Driver, Julia, Uneasy Virtue. Cambridge, UK; New York: Cambridge University Press, 2001.
6. Frede, Dorothea, ‘On the So-Called Common Books of the Eudemian and the Nicomachean Ethics’, Phronesis 64(1), 2019, pp. 84–116.
7. Gagarin, Michael, ‘Dike in Archaic Greek Thought’, Classical Philology 69(3), 1974, pp. 186–197.
8. Gauthier, René Antoine and Jolif, Jean Yves (eds.), Aristotle. L’éthique à Nicomaque (2e éd). Louvain, Paris: Publications universitaires; Béatrice-Nauwelaerts, 1970.
9. Gottlieb, Paula, The Nicomachean and Eudemian Ethics. Cambridge: Cambridge University Press, Forthcoming.
10. Grotius, Hugo, The Rights of War and Peace, trans. Richard Tuck, 3 vols. Indianapolis, IN: Liberty Fund, 2005 (originally published in 1625).
11. Harlfinger, Dieter, ‘Die Überlieferungsgeschichte Der Eudemischen Ethik’, in Moraux, P. and Harlfinger, D. (eds.), Untersuchungen Zur Eudemischen Ethik (Oosterbeek, Niederlande, 21–29 August 1969). Berlin: Walter de Gruyter, 1971, pp. 1–50.
12. Hatzimichali, Myrto, ‘Andronicus of Rhodes and the Construction of the Aristotelian Corpus’, in Falcon, A. (ed.), Brill’s Companion to the Reception of Aristotle in Antiquity. Leiden: Brill, 2016, pp. 81–100.
13. Havelock, E. A., ‘“Dikaiosune”. An Essay in Greek Intellectual History. (In Tribute to George Grube, the Distinguished Author of “Plato’s Thought”)’, Phoenix 23(1), 1969, pp. 49–70.
14. Huby, Pamela M., ‘The Eudemian Ethics’. Review of Paul Moraux and Dieter Harlfinger (eds.), Untersuchungen zur Eudemischen Ethik, The Classical Review 23(2), 1973, pp. 149–151.
15. Irwin, T. H., ‘The Aristotelian Ethics and Aristotle’s Theory of the Will by Anthony Kenny’, The Journal of Philosophy 77(6), 1980, pp. 338–354.
16. Jackson, H., The Fifth Book of the Nicomachean Ethics of Aristotle. Cambridge: Cambridge University Press, 1879.
17. Jaeger, Werner, Aristoteles: Grundlegung einer Geschichte seiner Entwicklung. Berlin: Weidmannsche Buchhandlung, 1923.
18. ———, Aristotle: Fundamentals of the History of His Development, trans. Richard Robinson (2nd ed.). Oxford: The Clarendon Press, 1948.
19. Kapp, Ernst, Das Verhältnis der eudemischen zur nikomachischen Ethik. Dissertation, Freiburg Br., 1912.
20. Kenny, Anthony (tr.), Aristotle : The Eudemian Ethics. Oxford; New York: Clarendon Press; Oxford University Press, 2011.
21. ———, The Aristotelian Ethics: A Study of the Relationship between the Eudemian and Nicomachean Ethics of Aristotle (2nd ed., 1st ed. 1978). Oxford: Clarendon Press, 2016.
22. Lee, Mi-Kyoung, ‘Justice and the Laws in Aristotle’s Ethics’, in Lee, M-K. (ed.), Strategies of Argument: Essays in Ancient Ethics, Epistemology, and Logic. New York: Oxford University Press, 2014, pp. 104–123.
23. ———, ms. Justice in Aristotle’s Moral and Political Philosophy.
24. Lloyd-Jones, Hugh, The Justice of Zeus, Sather Classical Lectures, v. 41. Berkeley: University of California Press, 1971.
25. Moss, Jessica and Schwab, Whitney, ‘The Birth of Belief’, Journal of the History of Philosophy 57(1), 2019, pp. 1–32.
26. Primavesi, Oliver, ‘Ein Blick in den Stollen von Skepsis: Vier Kapitel zur Frühen Uberlieferung des Corpus Aristotelicum’, Philologus 151(1), 2007, pp. 51–77.
27. Rawls, John, A Theory of Justice, revised ed. (1st ed. 1971). Cambridge, MA: Harvard University Press, 1999.
28. Ross, W. D. and Brown, Lesley, trans., Aristotle: The Nicomachean Ethics. New York: Oxford University Press, 2009.
29. Rowe, C., The Eudemian and Nicomachean Ethics: A Study in the Development of Aristotle’s Thought. Cambridge: Proceedings of the Cambridge Philological Society, 1971, 3(3), Supplement.
30. Schneewind, J. B., ‘The Misfortunes of Virtue’, Ethics 101(1), 1990, pp. 42–63.
31. Slote, Michael, Morals from Motives. Oxford; New York: Oxford University Press, 2001.
32. Susemihl, Franz, Eudemi Rhodii Ethica. Leipzig: In aedibus B.G. Teubneri, 1884.
33. von der Mühll, Peter, De Aristotelis Ethicorum Eudemiorum Auctoritate. Göttingen, 1909.
34. Wolt, Daniel, ‘Phronêsis and Kalokagathia in Eudemian Ethics VIII.1’, Journal of the History of Philosophy 60, 2022, pp.1–23.
35. Woods, Michael D., Aristotle’s Eudemian Ethics: Books I, II, and VIII. Oxford: Clarendon Press, 1982.