6

Decision in the Eudemian Ethics1

Karen Margrethe Nielsen

Ever since renaissance scholars started including the Eudemian Ethics (EE) in editions of Aristotle’s Ethics in the 15th and 16th centuries, the treatise has been read through a Nicomachean prism. Though, questions about its authenticity have now been laid to rest,2 the treatise poses particular challenges for the reader, especially given the sorry state of the manuscript tradition. In his masterly study of the transmission history of the Eudemian Ethics, Dieter Harlfinger calls it “the most corrupt treatise” in the Corpus Aristotelicum,3 and Jonathan Barnes remarks that “the text of the EE is in a vile state – hideous corruption on every page.”4

The issue is unfortunately compounded by new errors in the apparatus criticus of the most recent edition, Walzer and Mingay’s Oxford Classical Text (OCT) (1991), which is the culmination of a scholarly relay run started by David Ross, continued by Walzer, and completed by Mingay. As Barnes notes in his review, the editors have perhaps been a little too ready to accept the emendations of modern scholars, and sometimes misattribute the emendations they accept, making it hard for the reader to trust both the accuracy of the apparatus and the judiciousness of their choices.5 Even modern translators who are aware of the issue, and prepared to reject proposals in the OCT where the manuscripts (MSS) can be made intelligible, find themselves hobbled by centuries of received opinion. The challenge, then, is to handle a manuscript tradition that is hideously corrupt, and at least five centuries of scholarly attempts to plaster over or repair the problems, a challenge that is made all the harder by our tendency to read the Eudemian Ethics as a treatise that either foreshadows or shadows the Nicomachean Ethics (EN) on most points of substance. Whether we follow the majority view, and read the EN rather than the EE as the ultimate expression of Aristotle’s thought, or follow Kenny, who argues for the reverse relationship, the temptation is to use assumptions about the concepts and arguments of the EN to illuminate obscure passages in the EE, without asking whether these assumptions are sound in their own right, and whether – and to what extent – they carry over from one treatise to the other.6 In the case I will consider in this chapter, restoring an MSS reading in the EE and reversing a Renaissance emendation make us confront what appears to be a conceptual divergence between the nominal definitions of prohairesis in the EE and EN. Taking the MSS at face value should make us reconsider standard analyses of prohairesis in Aristotle’s ethics.

Prohairesis is the keystone of Aristotle’s argument in both the Eudemian and Nicomachean Ethics.7 He defines virtue as a “state that decides” (hexis prohairetikê) (EN II 6, 1106b36–1107a1; EE II 10, 1227b5–11), and argues that decision distinguishes characters better than actions do (EN III 2, 111b7; EE II 11, 1228a3–4). We only use actions to judge character when the prohairesis of the agent is not easy to discern (EE II 11, 1228a14–16), for the decision reveals what someone has chosen to do and for what end. Prohairesis is further described as the efficient cause of action (praxis) – in this role, it is “hothen hê kinesis” (“that from which the movement arises”) rather than the telos (“that for the sake of which” we act) (EN VI 2/EE V 2, 1139a32–3). In the EN, and in the first common book, Aristotle maintains that virtuous agents decide on virtuous actions for their own sakes, as instances of eupraxia (EN I 8, 1098b22; II 4; VI 2, 1139a34; b3). While emotions may dispose non-virtuous agents to do right acts, they do not choose these acts as a result of a rational commitment (EN II 6, 1106a1–6; EE III 7, 1234a23–4).

The term “prohairesis” has no direct descendant in English, making it hard to capture its sense. In the Anglophone literature, it is rendered with terms ranging from “choice” to “commitment,” “rational choice,” “deliberate choice,” “purposive choice” (sometimes just “purpose”), “preferential choice,” “decision,” and “election,”8 the last following Aquinas’ Latin. Each has advantages and disadvantages. Which one we should prefer depends, in part, on how we understand Aristotle’s explication of the concept, and whether we wish to build Aristotle’s remarks about the nature of prohairesis into the very term, as “preferential choice,” “deliberate choice,” “purposive choice,” and “rational choice” all do.9 The term had not been regimented for philosophical use before Aristotle offered his definition: we need to understand what Aristotle means when he speaks of prohairesis to understand his analysis of virtue and human action.

6.1 Prohairesis in the Eudemian Ethics

In light of the many problems facing the reader of the EE, we should not be surprised if even the most able recent translators, benefiting from Harlfinger’s painstaking work on the MSS and stemma codicum, as well as a burgeoning critical literature, occasionally trip up and accept a reading that should have been rejected. In the case I’ll consider, the error of judgment is shared by all modern readers, and hard to avoid, because its source lies so far back in the reception history as to have been effectively obscured from view. However, once we trace the emendation back to its source and restore the original readings of the MSS, we will see that it distorts Aristotle’s argument.

The distortion I shall consider, at 1226b8, is not, remarkably, an unintended consequence of an attempt to correct a corruption in the MSS, but instead a conjecture that made it into Greek editions of the Eudemian Ethics starting in the 16th century, despite lacking manuscript basis, and despite being both unnecessary and philosophically ill-fitting in the context of the argument where it appears. The motive behind the emendation is conciliatory. The commentator, the German scholar Friedrich Sylburg (1536–1596), wished to smooth out what he (perhaps rightly) saw as a doctrinal discrepancy between the Eudemian and Nicomachean treatises, so that the Eudemian Ethics agreed with the theory with which he was familiar from the Nicomachean Ethics.10 He therefore made a proposal in the commentary included in his 1587 edition of the Eudemian Ethics that later editors have accepted without question or sufficient scrutiny. Based on his own commentary, where he lays out his case for overriding the MSS, we know that Sylburg was reading a perfectly wholesome line in the MSS, with no grammatical errors giving him reason to correct the transmitted text. Still, he thought something must be amiss, since the explanation of the term prohairesis offered by the Eudemian Ethics seemed to depart from what he took the parallel passage in the Nicomachean Ethics to say. As a result, he changed the Greek of the EE to make it conform to what he thought the EN stated. Sylburg’s emendation shows up in contemporary translations in the shape of almost comically contorted efforts to make sense of Aristotle’s line in the EE, and an oddly disjointed interpretation of the surrounding argument. The conjecture concerns Aristotle’s nominal definition of prohairesis, or decision, at EE II 10, 1226b8 – a definition that highlights the meaning of the term before Aristotle examines the nature of the mental act that the verb “prohaireisthai” denotes. Sylburg’s proposal is accepted in Susemihl’s Teubner edition (1884), as well as in Walzer and Mingay’s OCT, and presupposed in all modern English translations.11

Despite the universal acceptance it has enjoyed, the proposal obscures a distinctive feature of the Eudemian Ethics from view, a feature that, when properly understood, allows us to appreciate the Eudemian definition of decision on its own terms. Attempting to understand the Eudemian analysis of prohairesis without filtering it through preconceptions derived from the corresponding passages in the Nicomachean treatise, will, I wager, yield a better understanding of the progression of Aristotle’s argument in the Eudemian Ethics, and also potentially open a window into the development of Aristotle’s thought.

The disputed line appears in the context of Aristotle’s extended discussion of prohairesis in Eudemian Ethics II 10. Aristotle here seeks to determine its genus by considering puzzles that arise about prohairesis. The analysis parallels that of Nicomachean Ethics III 2–4, though in the EE, unlike the EN, Aristotle has already introduced the theme of decision in his preceding analysis of the voluntary.12 In both Ethics, Aristotle defines virtue of character as a hexis prohairetikê, a state that decides (EN II 6, 1106b36–1107a1; EE II 10, 1227b5–11).

In the EN, the full definition of virtue appears prior to the analysis of the voluntary and decision, and so motivates the ensuing inquiry into the nature of prohairesis. In the EE, Aristotle does not introduce the phrase hexis prohairetikê into his definition of character virtue until after he has concluded the topical analysis of prohairesis in II 10. Instead, he presents a provisional definition of virtue at EE II 5, 1222a6–12, which is fully articulated at the conclusion of book II 10, and repeated at the heading of book III (EE III 1, 1228a20–25), to specify that it is a state that decides. His approach in the EE is therefore inductive rather than deductive: Aristotle works towards the precise definition of virtue of character, starting from an analysis of decision, whereas his approach in the EN is deductive: Aristotle works from the precise definition of virtue as a state that decides, and then explicates the differentia that sets this state apart from others.

Whatever the reasons for this difference in approach, that decision is a central concept in Aristotle’s ethics cannot be doubted. In the EN, Aristotle says that “decision seems to be most proper to virtue, and to distinguish characters from one another better than actions do” (III 2, 1111b6–7), and in the EE, Aristotle remarks that virtues and vices all involve decision; therefore, states that are mere emotional means (cf. the doctrine of the mean) don’t qualify as character states (EE III 7, 1234a24–5). In both treatises, he defines decision as a “deliberate desire for actions that are up to us” (orexis bouleutikê tôn eph’hêmin [EN III 3, 1113a11]; orexis tôn eph’hautôi bouleutikê [EE II 10, 1226b17]), on the basis of its causal origin in deliberation. We start by having a wish (boulêsis) for an end. Once we have inquired about the means to that end through deliberation and have considered what we can do to promote it, we form a desire to do the act we judge we should do because it promotes our end best or well enough not to necessitate further inquiry. However, the underlying analyses of deliberation (bouleusis) in the EE and the EN seem, on the face of it, to highlight different aspects of practical reasoning, the EE emphasizing that deliberation seeks means to an end, and the EN taken to underscore the preferential nature of prohairesis as a choice of this course of action rather than that, and so the comparative aspect of the preceding deliberation. The nominal definitions of “prohairesis” that guide Aristotle in his inquiry also seem to diverge, at least if we don’t accept Sylburg’s conjecture and instead take the MSS of the Eudemian Ethics at face value, while preserving the standard analysis of the Nicomachean analysis.

The EE, like the EN, approaches the inquiry into the nature of prohairesis dialectically, by taking what “some say” as its starting point.13 In the EE, decision is said to be one of two things, either belief (doxa) or desire (orexis). In the EE, Aristotle thinks that this view has something in its favour, since decision seems to accompany both.14 However, Aristotle observes that it is neither of these singly. It cannot simply be defined as desire, for then it would have to be identical to wish, appetite, or spirit – the three types of desire identified by Plato. And it cannot be spirit or appetite, since these are shared with non-rational animals, and we can make decisions without feeling an appetite or emotion; these are always accompanied by pain, whereas a decision is not. However, it cannot simply be wish (boulêsis) either, since we wish for what we know to be impossible, but we do not decide to do what we know to be impossible – to be king of all mankind or immortal, for instance. Nor, in general, do we decide on things that are possible, but which we do not believe to be up to us to do or not to do. What we decide to do are those things that we believe to be in our own power. If decision is a kind of wish, it’s a specific kind of wish.

Does this entail, by elimination, that decision is a kind of belief? Decision is not belief simply speaking, for we have many beliefs about things that are not up to us, for instance that the diagonal is not commensurable with the sides. Further, belief is true or false, but decision is not true or false. It is not even any particular kind of thinking, for instance beliefs that something should or should not be done. What is missing is apparently some desiderative element making us commit to the act.

Aristotle now observes that belief and wish have something in common. Whereas we can have either a belief about an end or a wish for it, no one decides on the end, but rather on what promotes the end (outheis gar telos ouden prohaireitai, alla to pros to telos). This is the first appearance of the expression “to pros to telos” in the Eudemian Ethics. What we decide to do – to prohaireton – is what is “towards,” pros, the end, namely what promotes it, and what promotes the end are some of the actions that we believe are up to us to do or not to do. No one decides to be healthy, but rather to walk about or to sit down for the sake of his health (tou hugiainein heneken), and no one decides to be happy, but rather to go into commerce or to embark on a venture for the sake of being happy (tou eudaimonein heneka). “In general,” says Aristotle, “one who makes a decision simultaneously makes it clear what he is deciding to do and for the sake of what he decides to do it” (ti te kai tinos [heneka] prohairetai). The “for what” (tinos) is the thing for whose sake he makes the decision, but the “what” (to ti) is the thing he decides to do for the sake of the other, viz., for the sake of the end. This analysis is significant, since it already suggests that a prohairesis is a choice of one thing for the sake of another, an analysis that Aristotle endorses when he considers the meaning of the term “prohairesis” a few lines further on. The Eudemian Ethics, then, unlike the Nicomachean Ethics, explicitly connects the famous claim, familiar from the EN, about deciding on what is “pros to telos” to the meaning of prohairesis.

Aristotle identifies the end as an object of wish, since wish is primarily (malista) about ends; the end is also a subject matter of belief, since the agent believes that he should be healthy or that he should do well. Aristotle concludes that it is clear from these considerations that, though decision in some way involves wish and belief, it is not identical to either of these, for they are primarily (malista) concerned with ends, but decision is not. That wish is “malista” about the end, and decision “malista” about what promotes it, leaves open the possibility that a decision is a kind of wish, as Aristotle later seems to maintain. Without qualification, wish is for the end, just as without qualification, decision is for what promotes the end. With qualification, however, decision is for the end insofar as it is of “this for the sake of that,” an act insofar as it promises to promote our end. Conversely, with qualification, wish is for the means, since a decision is a deliberate desire for an act that is up to us, where the desire in question is a wish. By stating that wish is for the end, and decision for what promotes the end, Aristotle identifies their primary objects; he now needs to explain what distinguishes a wish simply speaking from a decision, and how a belief about what we should do differs from a decision.

At this point, Aristotle introduces deliberation (bouleusis) into his analysis. Its appearance is abrupt, but since we are looking for a differentia – whatever distinguishes a decision from a mere belief about what we should do or a mere wish – we should expect a new element to be introduced at this stage of the analysis.

Aristotle first considers what deliberation is about – what kinds of things are open to deliberation? There are some things that are capable of being and not being, and among these, some are up to us to do and not to do; these are the objects of deliberation. No one deliberates about the affairs of the Indians, or about squaring the circle, since the former isn’t up to us, and the latter can’t be done by anyone. We deliberate about what is up to us to do and not to do when we need to work out what promotes our end – while doctors can fail either in working out the right course of action or in paying attention to the right way to proceed, scribes can only make mistakes through inattention, since it is evident how to proceed in the application of their craft. Summing up the result of his analysis so far, Aristotle comes closer to pinning down its nature: since decision is neither belief nor wish taken separately (hôs hekateron), nor indeed both (amphô), since we can form a wish or a belief that we should do something suddenly (exaiphnês), while we cannot suddenly decide, it must be a product of both (ex amphoin), for both are present when a person makes a decision.

In the passage that is my main focus in this chapter, and to which I will return imminently, Aristotle now observes that the name itself provides a clue to the nature of decision as a product of wish and belief. A prohairesis is a kind of hairesis, he says, not simply, but – and here we reach the contested line that Sylburg would not accept – of one thing for another (“heterou pros heteron”) (1226b8). For this is not possible without examination and deliberation. So, he concludes, decision is the product of deliberative belief (it is ek doxês bouleutikês 1226b9).

This nominal definition provides Aristotle with the resources he needs to give a real definition of decision at the conclusion of the same paragraph (1226b9–20). The text of this paragraph is unfortunately a complete mess, caused mainly by the lexical similarity between boulêsis or bouleusis and their cognates, and the confusion to which this similarity gives rise. Although sorting through the apparatus to reconstruct a plausible reading of these lines is a worthy objective, let me simply state Aristotle’s conclusion: decision is a deliberate desire for what is up to us. In subsequent paragraphs, starting with “dio” at EE II 10, 1226b21, Aristotle defends the real definition he has presented by showing that it explains a number of beliefs that we hold, viz., beliefs about the voluntary. Having explained these beliefs in light of his definition of decision, he expands his analysis of deliberation, as being about what is “pros to telos” – we deliberate when we aim to trace our end back to an act that is up to us to do or not to do and that promises to promote that end. What, then, is the end? By nature, the good, but by corruption, and contrary to nature, something that is not good but only apparently good may be the end. This corruption is due to pleasure or pain. These observations allow Aristotle to define virtue at the conclusion of the chapter as a state of character that decides on what is mean relative to us in pleasures and pains (êthikê hexis prohairetikê mesotêtos tês pros hêmas en hêdesi kai lupêrois); viz. those pleasures and plains in respect of which a person’s enjoyment or distress indicates his character.

6.2 Prohairesis in the Nicomachean Ethics

Let us compare the Eudemian analysis of decision with its Nicomachean counterpart. The Nicomachean analysis of prohairesis follows a similar trajectory – arguing first that decision is not the same as any particular kind of desire, though it is close to wish, and second that it is not the same as belief, whether in general, or any specific kind of belief. It arrives at a nominal definition of decision via an observation about deliberation, but it offers an analysis of the meaning of the term that seemingly departs from that of the MSS of the EE:

Then what sort of thing, is decision, since it is none of the things mentioned? Well, apparently it is voluntary, but not everything voluntary is decided. Then perhaps what is decided is what has been deliberated (be)for(e) (to probebouleumenon). For decision involves reason and thought, and even the name itself would seem to indicate that what is decided [prohaireton] is chosen [haireton] (be)for(e) [pro] other things.

(EN III 2, 1112a13–17)

What appears to be missing from the EN is the EE’s detailed discussion, leading up to the nominal definition, of decision as being of one thing for the sake of another. Although the EN also underlines that we wish for the end, but deliberate and decide about what promotes it, this insight is not spelled out with the same kind of specificity that the EE offers. The discussion of deliberation and decision that follows the real definition of the EN likewise does not offer the explicitly teleological conception of deliberation and decision that we find in the EE, though the analysis of the EN is, arguably, compatible with the one we encounter in the EE. Most scholars have taken the “pro” in the EN’s nominal definition to have a preferential sense (e.g. Dirlmeier, Woods, Charles), whereas others (Aspasius, Joachim, Gauthier-Jolif, Irwin) take it to have a temporal sense. Those who favour the temporal reading typically claim that this makes better sense of the “probebouleumenon,” which they claim must mean “previously deliberated.” I shall later suggest that a third alternative is possible, and that this third interpretation of “pro” has gone unexplored, but I shall leave that to the side for now to focus on the nominal definition in the Eudemian Ethics.

Before we look at the textual issues, let me distinguish between three senses of “prohairesis” that we have encountered so far:

1. The preferential

2. The temporal

3. The teleological

Of these, the preferential and the temporal are standard interpretations of the “pro” in “prohairesis” in the EN. The teleological is, I have suggested, the only reading available of the “pros” in the MSS of the nominal definition in the EE, which cannot be taken in a temporal or a preferential sense. If we follow the existing interpretations of the EN, it would seem that the EE and the EN diverge over the meaning of prohairesis. Since this, in turn, suggests that the treatises operate with different conceptions of deliberation, either as something that precedes decision, or as the weighing of alternatives, or as inquiry into means to ends, these different nominal definitions also affect our understanding of a decision as an orexis bouleutikê. While the temporal reading appears to state the obvious – a decision requires prior deliberation – a point on which both the EE and EN converge, we might think that it fits in both contexts. However, this cannot be the sense of the EE nominal definition if we accept the MSS, and so not Aristotle’s reason for employing the compound noun prohairesis.

6.3 Textual Issues

It is time to look at the textual issues and Sylburg’s conjecture up close. The reading I just gave of the Eudemian nominal definition, “heterou pros heteron,” follows the MSS family descending from L, which Harlfinger dubs the “recensio Constantinopolitana.” On this reading, a decision is a kind of choice, not simply speaking, but rather a choice of one thing for the sake of another. The other main family of MSS, which Harlfinger calls the “recensio Messanensis,” instead offers the ungrammatical reading “heterou pros heterou” at 1226b8. This family has Codex Vaticanus 1342 (P in Walzer-Mingay; Pb in Susemihl) and Codex Cantabrigensis 1897 (C in Walzer-Mingay; Cc in Susemihl) as its primary extant members (P and C are “twins,” as Barnes notes, copies of the same archetype).

All MSS, including the corrupt ones, agree, however, that the preposition contained in Aristotle’s nominal explication of prohairesis in the Eudemian Ethics is “pros,” governing the accusative, rather than the preposition we find in the corresponding nominal definition in the Nicomachean Ethics, pro, governing the genitive.

The text actually printed in both the Teubner edition and the OCT is not, however, one that can be found in any MSS, but rather a result of combining parts of the reading in the recensio Messanensis with Sylburg’s proposal that “pros” should be “pro.” This yields a line reading “heterou pro heterou,” which does not appear in any of the 20 extant MSS of the EE.

Despite consulting MSS in both families, which as he acknowledges in his apparatus are unanimous in printing “pros” at 1226b8, Susemihl accepts Sylburg’s emendation of “pros” to “pro.” The MSS in the L-family, including Codex Marcianus, all have “heterou pros heteron” at 1226b8, which is unassailable on grammatical grounds. The MSS in the recensio Messanensis, on the other hand, all read “heterou pros heterou.” This makes no sense grammatically, and needs to be amended one way or another – either by replacing “pros” with a preposition governing the genitive (“pro” does the job), or by emending the last “heterou” to “heteron.” Assuming that the two heterous are correct, and preferring P to the grammatically correct Marc., Susemihl accepts Sylburg’s proposal. Familiarity with the parallel passage in the EN makes this solution seem more attractive than the alternative. However, the existence of the L-family makes the choice indefensible. A scholar considering only the L-family would have no reason to question the line. The existence of an alternate tradition offers only a weak desideratum, since L and its descendants are in much better shape than P and C and its descendants.15 Even someone working from both families of MSS would have reason to prefer L’s reading since it doesn’t cause a muddle. We can further explain how the second “heterou” made it into the MSS by duplication – the scribe’s eyes may have landed twice on the “heterou” preceding “pros,” leading to a classical, if minuscule, dittography. If this is right, the question we should be asking ourselves is not “utrum in alterum abiturum erat?”– which of the two readings would have been changed into the other – since the reading of L shows no signs of corruption. Rather, the question is, how did the conjectured archetype MSS of both families end up being corrupted in the recensio Messanensis? I have offered a conjecture of how that may have happened.

Against my explanation, one might object that “pro” is correct, and that the corruption to “pros” occurred at an earlier stage of transmission, either before the two families parted ways, or in the transition from this conjectured common ancestor to L. If the latter, whoever made L may himself have “corrupted” an archetype common to both main families into a grammatically permissible form. We cannot rule out that the MSS that he copied read “heterou pro heterou,” and that he added a sigma by mistake, and subsequently changed the last “heterou” to “heteron” in an effort to smooth over the wrinkle caused by the superfluous sigma. However, such a process of corruption seems harder to explain on the basis of familiar categories of scribal errors. The “s” would have had to appear out of thin air – why would he add it? We would furthermore have to believe that when he noticed that something was amiss with the grammar upon turning to the next word on the line, namely “heterou,” his impulse would not be to check the accuracy of the preposition that he had just written down, but rather to alter “heterou” to “heteron” to fit with his mistaken “pros.” This seems implausible.16

Alternatively, if we assume that the MSS were corrupted into reading “pros” even before the two families parted ways, we must assume that the two families dealt with the issue caused by the hypothesized corruption in different ways, either by smoothing over, as in the Byzantine tradition following L, or by changing the last heteron to “heterou,” or by simply letting it all hang out and printing an impossibility, as in the Codex Messanensis descending from P and C. My own view is that, on purely philological grounds, the hypothesis of dittography in the PC family makes more sense than the conjecture of an erroneous sigma creeping into the hyparchetype from which both main families derive or into the L-family. And as we shall see, there are also compelling philosophical reasons deriving from the surrounding argument for thinking that the L-family should be respected, and that Sylburg’s conjecture has caused a muddle in the way we interpret Aristotle’s theory of decision in the Eudemian – and possibly – Nicomachean Ethics. In other words, even if there had not been overwhelming evidence in the context of 1226b8 that “heterou pros heteron” is the correct reading, we should prefer it because it conforms to principles of sound textual criticism and principles of sound philosophical analysis.

The explanation that I have offered at first made me suspect that Sylburg was commenting on an edition of the EE which relied on the P and C family. However, this turned out not to be the case. When I consulted Sylburg’s commentary in the Balliol College special collections, it revealed that the Greek text on which Sylburg is commenting, included in his 1587 edition of the EE,17 in fact gives the L-family reading, “heterou pros heteron.” But in a section titled “Variae lectiones,” Sylburg expresses unhappiness with this:

p. 108, v.I, rectius forsan heterou pro heterou: ut significetur alterius prae altero electio: sicut scilicet in Nicomacheis, 40,13, prohaireton dictum tradit hôs on pro heterôn haireton. Aretius tamen vulgatam scripturam secutus vertit, alterius ad alterum.

(p. 271)

More correct, possibly, heterou pro heterou: so that it means alterius prae altero electio (the choice of one thing before another): just as in the Nicomachean Ethics 40,13, where the expression hôs on pro heterôn haireton renders prohaireton. Aretius, notwithstanding, having followed the commonly accepted manuscript, translates alterius ad alterum (of one thing for the sake of another).18

The entry is, as far as I can tell, not cited by any of the scholars who accept the emendation, though their rationale is likely similar to Sylburg’s. Without amending the text, the EE will diverge from the EN.

Now, Sylburg was no upstart, having studied with Henri Estienne (Stephanus) in Paris. His edition of Aristotle’s complete works, which includes an earlier edition of the EN and the Magna Moralia (MM), confirms that he was a dedicated and generally knowledgeable reader of Aristotle. Nevertheless, the principle on which he bases his conjecture seems weak: Sylburg assumes that the EE should conform to the sense he has extracted from the parallel passage in the EN, where he finds “pro,” and infers that the sense of “pro” must be preferential, so that Aristotle’s nominal definition states that a decision is a choice, not simply, but of one thing in preference to another. He rejects the MSS reading adopted by Leonardus Aretius (the renaissance scholar Leonardo Bruni, 1369–1444), who relied on what he calls “the commonly accepted manuscripts.” Bruni is one of the most important early Italian humanists; he invented the label studia humanitatis, from which “humanist” and “humaniora” derive, in addition to serving as chancellor of Florence. His Latin translation of the Nicomachean Ethics was widely disseminated; among other works in the corpus, he also edited the Eudemian Ethics. The MSS that Sylburg prints and comments on has the very same reading that Aretius relies on from “the commonly accepted manuscripts,” so it can’t be that Sylburg had access to scripts that supported his conjecture, in which case he would surely have cited them. Sylburg’s proposal is based, it seems, solely on his wish to make the EE conform to the EN, which he had edited a few years before the EE.

Once Sylburg’s conjecture is incorporated into editions of the EE, force of habit seems to have prevented scholars from examining his rationale critically. Consider first the three most recent editions of the text, Fritzsche’s Aristotelis Ethica Eudemia (Regensburg 1851); Susemihl’s Teubner (Leipzig 1884), and Walzer and Mingay’s OCT (Oxford 1991). Each prints “heterou pro heterou,” with exactly the same reference to Sylburg in the apparatus, and the same list of textual variants. Fritzsche, who adds his own Latin translation after the Greek text, renders the phrase “unius rei prae altera,” “one thing before another.”

Modern translators predictably follow suit, as a chronologically ordered selection reveals:

1. Solomon (1915, rev. 1984); transl. based on Susemihl:

“But we must ask – how compounded out of these? The very name is some indication. For choice is not simply picking but picking one thing before another; and this is impossible without consideration and deliberation; therefore choice arises out of deliberative opinion.”

2. Rackham’s Loeb (1936) prints heterou pro heterou without comment:

“But how purposive choice arises out of opinion and wish must be considered. And indeed in a manner the actual term “choice” makes this clear. “Choice” is “taking,” but not taking simply – it is taking one thing in preference to another; but this cannot be done without consideration and deliberation; hence purposive choice arises out of deliberative opinion.”

3. Dirlmeier (1962) translates:

“In gewisser Weise enthüllt dies allein schon der Name. Denn die Entscheidung ist eine Wahl, aber nicht in einfachem Sinn, sondern indem man dem einen vor dem anderen den Vorzug gibt.” He explains that “prohairesis ist wörtlich ‘die vorzuggebende Wahl’” (note ad 1226b6, p. 294), or as English translators sometimes say, “preferential choice.”

4. Woods’ (1992) translation again follows Sylburg, though his choice of terms is perhaps more careful than that of his Anglophone forebears. “To some extent the word ‘choice’ itself shows us: Choice (prohairesis) is a taking (hairesis), but not without qualification – a taking of one thing before (pro) another; that is not possible without examination and deliberation.” In his commentary ad loc, Woods offers the following justification for his translation:

“The Greek word here translated choice (prohairesis), and the corresponding verb, have the form of a compound of a word meaning “taking” with a preposition meaning “before”; the latter may have a purely temporal sense or it may have the force of “in preference to” (cf. English “I’d sooner do A than B”) (…) The same etymological observations occurs also at E.N. 1112a16–17, where Gauthier and Jolif, following Aspasius and Joachim, hold that the preposition has a temporal sense; i.e., the object of choice is that which comes first in the sequence leading to the end. But it seems clear that the prefix has the other sense in this passage, and the same is indisputably true at MM 1189a12–16 (which may be held to be decisive against Gauthier and Jolif’s reading of the E.N. passage: in MM there is reference to choosing the better instead of the worse) (…).”

Woods’ argument for thinking that the Eudemian and Nicomachean passages coincide is, alas, circular. Now that we know not just that the “pro” is Sylburg’s conjecture, but also that it rests solely on his assumption that the EE must mirror the EN, it is trivially true that “the same etymological observation also occurs in the EN.” Woods’ argument against the temporal reading of the “pro” further assumes that all three Aristotelian Ethics must agree on the meaning of “prohairesis,” and that the parallel passage in the MM offers a nominal definition that can only be taken in the preferential sense. This is just more of the same – unwarranted assumptions about the unanimity of Aristotle’s three ethical treatises. Even those of us inclined to hold that the MM in substance, if not in form, is a treatise by Aristotle, should resist the pernicious temptation to efface all doctrinal discrepancies so that they come out saying the same thing. Doctoring the text simply won’t help, if the surrounding argument pulls in different directions. Instead, we should at least allow the possibility that Aristotle’s thinking about prohairesis evolved, and that subtle differences between the analyses in the three Ethics need to be acknowledged, and, if possible explained, in a way that does not regiment all differences until all three treatises serve up the same muddled porridge. The possibility that Aristotle’s thought evolved should at least be kept a live option.

Two recent translations19 of all eight books of the EE unfortunately show little improvement on this score.

5. In Inwood and Wolf’s translation for Cambridge Texts in the History of Philosophy, the line is rendered as follows: “But in a way the term gives us an indication. Decision is choice, not unqualifiedly so, but of one thing in preference to another.” They then add a note, explaining that “Decision (prohairesis) is a compound of pro (in preference to, before) and hairesis (‘choice’)” (p. 35).

6. Anthony Kenny likewise offers a translation that is in equal parts lofty and odd: “Its very name already tells us something. Choice is a matter of picking – not random picking, but picking one thing in preference to another, and this is not possible without examination and deliberation.”

Notice the awkwardness with which Kenny renders “hairesis” “picking” and “prohairesis” “choosing.” The contrast between picking and choosing, first employed by Solomon, has no counterpart in Aristotle’s text. On the contrary, making a choice of lives – hairesis tou biou – is in the Eudemian Ethics described as the most important choice we can make, since it sets us on the right or wrong path, depending on the nature of the end that we pursue. That Heracles at the crossroads should be “picking” the life of virtue, and that a prohairesis by contrast is an existential, weighty choice, is pure invention – whatever it entails, Aristotle’s distinction between a hairesis and a prohairesis cannot be captured by this modern pair of contrasts. In his comments, Kenny thinks it is obvious that prohairesis

is a very solemn kind of choice, made after deliberation, and on the basis of a thought-out plan of life. Carrying out a monastic vow or a New Year’s resolution seems to be the closest thing in modern life to making an Aristotelian choice.

(p. 159)

Even if we agree that a prohairesis is a solemn kind of choice, it doesn’t follow that a hairesis is the opposite, especially since a prohairesis by Aristotle’s explanation in the EE is a kind of hairesis. The picking and choosing meme, which appears to originate with Solomon, has not improved with age. The translation is as dissonant now as it was when it was first voiced a century ago.

If instead of importing assumptions from the EN, we look at Aristotle’s definition in context, we will see that he anticipates this analysis of the relationship between hairesis and prohairesis already in book I. In the discussion of ultimate ends and the choice of lives in EE 1214a32; 1215b17; 21; 30; 35; 1216a13; 15; 21, Aristotle systematically relates one’s hairesis to the adoption of an end. The EN likewise opens with a discussion of what end is most choiceworthy (1094a15; also 1097a26 and following). In discussing the completeness of different ends, Aristotle systematically uses hairesis without pro. It seems, then, that whereas a hairesis is of the end, a prohairesis is of the means for the sake of this end. If this is right, we can make two related points. First, Aristotle thinks of a hairesis as a kind of boulêsis, but not just any boulêsis – it is rather a commitment to an end as good that lays it down as a target to be aimed at. The actions we do for the sake of this end are objects of prohairesis. Primarily, a prohairesis is of the means, but since we only act on decision when we commit to an end as good, our decision will also reflect the kind of life we take to be worth living. This is not simply a life devoted to the procurement of necessities – as the EE sternly observes, no one would choose to be alive rather than dead if all their activities were, in the words of the Politics, for the sake of living, rather than living well.

The Eudemian explication of the meaning of prohairesis highlights this relationship: a decision is a choice of one thing (an act) for the sake of another (an end), and this is what I have called the teleological conception of decision. The further discussion of deliberation that follows the real definition of prohairesis confirms that Aristotle thinks of deliberation as a search for a cause, namely the cause of the end that we want.

The deliberative part of the soul is the part that recognizes a particular kind of cause, for a purpose is one kind of cause, since a cause is the reason why something obtains (…) that is why those who have no fixed aim do not go in for deliberation.

(EE II 10, 1226b25–29)

The end that is kept fixed in deliberation functions like a hupothesis in the sciences – we reason from it, but not to it (EE II 10, 1227a8), since we need to deliberate with an end in mind: “A man who deliberates, once he has taken a survey from the point of view of the end, deliberates about what conduces to bring it within his reach, or what he himself can do towards the end” (1227a14–17). Aristotle’s claims about the fixity of the end, and his repeated insistence that we do not deliberate about the end, but rather about what promotes it, need to be handled with care, however. Although the end itself must be fixed like a hypothesis in order for deliberation to be possible, this is just the end “haplôs,” or without qualification, as Aristotle puts it in the EE – it does not preclude thinking about the end “kata meros.” The distinction between an end “haplôs” and an end “kata meros” corresponds to the differences between a general and a more specific conception of the end. Notwithstanding his view that the end for the purposes of any particular piece of deliberation is fixed, Aristotle thinks we make the end more specific by means of inquiry. He even calls this type of inquiry deliberation “kata meros”:

The end is by nature always a good, and one about which people deliberate in particular (kata meros), as a doctor may deliberate whether he is to give a drug, or a general where he is to pitch his camp. For them there is a good, an end, which is the thing that is best in the abstract (haplôs).

(EE II 10, 1227a18–21)

In the sphere of human activity, the “unhypothetical first principle” is living well, or eudaimonia, construed in the most formal way. But at any point beneath it, an end may be held fixed as a starting point for deliberation or may be considered a means to living well, by being part of it. In the latter role, it is subject to inquiry. That Aristotle allows for inquiry about ends is already evident from the discussion of the nature of happiness and its preconditions in EE I 2. He here takes a new tack in his inquiry:

Having acknowledged about these things that anyone capable of living in accordance with his own decision (zên kata tên hautou prohairesin) sets up20 some target for fine living (thesthai tina skopon tou kalôs zên), whether honor or reputation or wealth or culture, with a view to which he will carry out all his actions (as not having one’s life organized towards (pros) some end (ti telos) is a sign of great folly (aphrosunês pollês)), above all he must first determine in himself, neither impetuously nor hurriedly, in which of human affairs living well resides, and what are the necessary conditions for humans to possess it. For a healthy life is not the same as the necessary conditions for healthy living.

(EE I 2, 1214b6–15)

Everyone capable of living in accordance with his own decision does set up such a target. That is the import of saying that by nature, what is desired is the good. This good, which is always given some specification, whether hurriedly or through the careful consideration that Aristotle recommends, is the target at which people aim in all their actions (praxeis). If the target is wrong, the actions that we choose for its sake will also be wrong, and it is our task to ensure that we have specified the components of living well correctly, and that we do not confuse these components with merely necessary conditions for living well – things without which the best life won’t be possible, but the possession of which will not make us happy unless we use them well. It is in this light that we must understand Aristotle’s remarks in the coda to EE II 10. Having presented his definition of virtue as a state that decides, he now asks in EE II 11 whether virtue makes the target correct or the things that promote it. He responds that virtue makes the target correct, since this is not the result of deliberation or reasoning, and the target must be hypothesized as a starting point (1227b25). Just as a doctor does not inquire (skopei) about whether one ought to be healthy or not, but about whether one ought to take a walk or not, and a trainer does not inquire whether one ought to be fit or not, but whether one should take up wrestling or not (1227b25–28), no discipline in general inquires about its end (telos), but rather about what promotes it. The end is the starting point of thought, and the conclusion of thought is the starting point of actions. Virtue, then, is what ensures that we aim at the right end, but aiming at the right end requires careful thought, of the kind that Aristotle engages in when he examines the nature of living well.

6.4 Questions and Proposals

We can now revisit the question posed at the start about the relationship between the nominal definitions of prohairesis in the EE and the EN and what they tell us about Aristotle’s conceptions of deliberation and practical reasoning. Is the Nicomachean analysis of decision in tension with that of the EE, or can the two accounts, despite their differences, be interpreted in the same way? Restoring the MSS of the EE precludes the temporal and preferential readings of “prohairesis,” at least as far as the meaning of the term is concerned. If we deny that “pro” in the EN nominal definition can be read in the teleological sense, we should conclude that Aristotle changed his mind.

While Liddell and Scott’s dictionary does not highlight a teleological sense of “pro,” it is evident that Aristotle sometimes uses the preposition to mean “for the sake of.”

Consider Aristotle’s discussion of natural slaves as a type of servant in Politics I 4. Aristotle defines a slave as a sort of animate tool, and remarks that every servant is a “tool pro other tools” (kai hôsper organon pro organôn pas hupêretês) (Pol. I 4, 1253b32–33). Some translators take “pro” to mean “before” in some vague temporal sense. But the real meaning must be “for the sake of”: a servant is a “tool for the sake of tools” because the servant uses inanimate tools in his activities. Had tools been able to complete their work themselves, like the statues of Daedalus – ancient self-moving puppets – we should not have any use of servants. Alas, the shuttle does not move automatically, and a plectrum does not strike the chords of a lyre by itself. The master craftsman needs servants to carry out the work since only a servant can put inanimate tools to use. Jowett captures the sense well when he renders “pro” “for,” as does Reeve when he renders it “tools for using tools” (Aristotle: Pol. 6).21 Irwin and Fine translate “a tool prior to tools,” but then explain in a note that a slave is needed in order to use the other tools (my emphasis).22

This suggests that the prepositions “pros” and “pro” have overlapping senses: they can both mean “for the sake of,” and we need to expand the list of senses of “pro” distinguished in the literature. To the temporal and preferential senses, we should add a third, similar to that of “pro” in Latin expressions like “pro bono” and “pro forma.” This sense is omitted from Liddell and Scott’s dictionary, but included in other accounts of Greek prepositions, such as Gessner Harrison’s Victorian grammar A Treatise on the Greek Prepositions and on the Cases of Nouns with which These are Used (Philadelphia: J. B. Lippincott & Co, 1858).

From the entry on pro:

In some examples pro with the genitive case expresses the ground or the motive of an action. The object introduced by pro is that “in view of” which or “looking to” which, the action is performed, and so is regarded as its ground or motive. And, again, it is obvious that the notion of “looking to”, or “having in view”, is to be referred immediately to the primary sense of “before”, “in front of”, belonging to pro (…)

From the entry on pros:

Pros is only an augmented form of pro. Just as eis (ens) is a fuller form of en, and as ouk is of ou, and has the same root with the Latin pro and prae, with the English for and fore, seen in forehead, forecast, and fro, a transposed form of for, seen in froward. The radical fir in fir-st is only another from of for, as pri in pri-or and pri-mus is of pro and prae. Whether the s at the end of pros is inflectional, or merely formative, may admit of question.

That Aristotle uses “pro” and “pros” in the same sense is also evident at other points: he sometimes says that an inquiry is “pros to ergon” and sometimes that it is “pro ergou.”23 In light of this, three strategies for dealing with the apparent discrepancy between the EE and the EN nominal definitions present themselves. We can (i) emend the text of the EE so that it coheres with that of the EN – the strategy first proposed by Sylburg. This opens the EE nominal definition to temporal and preferential readings, of which the preferential seems to fit Aristotle’s claims better than the temporal in context. Alternatively, we can (ii) maintain that nominal definitions of prohairesis in the EE and the EN are indeed different – perhaps even incompatible – and seek to explain this fact, for instance as reflecting the development of Aristotle’s thought. Finally, we can (iii) accept that the nominal definitions differ lexically, but argue that they overlap in sense. They would overlap in sense if the “pros” in the EE is equivalent to the “pro” in the EN, insofar as both are teleological, or they could overlap because Aristotle does not privilege any of the three senses of “pro” in the EN, but instead wants us to hear all of them, including, but not restricted to, the teleological. If we prefer (iii), the seeming disparity in the nominal definitions comes down to a difference in emphasis rather than a substantial difference or change of mind.

The first and third approaches are conciliatory as far as doctrine is concerned, but only the second and third acknowledge what I take to be an indisputable fact: that Sylburg’s emendation corrupts the text of the EE.

In principle, we could, of course, also attempt to emend the MSS of the Nicomachean Ethics so that it corresponds to the EE; this would simply be the reverse strategy of Sylburg’s (i). This would be radical and inadvisable. Meddling with the manuscripts simply to preserve some preconceived assumption that the two treatises must cohere, not just in spirit, but also in the letter of their definitions, is unjustified and distorting. More fundamentally, the MSS of the Nicomachean Ethics are in much better shape than those of the EE, so even if the EE requires frequent editorial intervention, this “reverse” strategy of (i) is not a live option.

My own preference is for a version of (iii). Although I cannot defend my reasons for preferring (iii) in this chapter, the discrepancy between the two nominal definitions may in fact be less consequential than one might be inclined to think in light of existing interpretations of the Nicomachean theory of decision. For whereas modern scholars have asked us to choose between a preferential and a temporal reading of “pro,” I have argued that the preposition, just as its Latin counterpart, can have a teleological sense. Perhaps, then, the distance between the EE and the EN on this score is not unbridgeable, and we should use the EE MSS to shed light on the meaning of the nominal definition in the EN, travelling in the opposite direction of Sylburg. This does not require that we take the EN explication of the word to preclude other senses than the teleological. Aristotle may have wished to draw attention to several aspects of prohairesis at once in the EN, unlike in the EE, where the sense of the MSS must be teleological. The explication at EN III 2 lends itself to multiple readings, as my tentative translation reveals (“perhaps what is decided is what has been deliberated (be)for(e) (to probebouleumenon). For decision involves reason and thought, and even the name itself would seem to indicate that what is decided [prohaireton] is chosen [haireton] (be)for(e) [pro] other things” [EN III 2, 1112a13–17]). The surrounding argument does not tell decisively against any of them.

Whichever way we cut it, the questions raised by the MSS of the EE and Sylburg’s emendation show that we should pause before translating Aristotle’s notion “prohairesis” with terms that already decide the issue between the preferential, temporal, and teleological readings of “pro.” The proposals “purposive choice” and “preferential choice” already highlight one of the three senses while precluding other readings, which is unfortunate (especially if the EE and the EN diverge in their understanding of the term – the EE defending the teleological reading captured by “purposive choice” and the EN the reading captured by “preferential choice”). If we render “prohairesis” “deliberate choice” we will build into the very term some of the points Aristotle takes pains to state in the course of his inquiry, making them analytically rather than synthetically true. “Purposive choice” seems to capture what Aristotle takes prohairesis to be in the EE, but this reflects his real definition of decision as much as his nominal definition of the term, which simply motivates his analysis. My own sense is that the Nicomachean Ethics, too, albeit less explicitly, sees a decision as a choice of an act for the sake of an end, where the end is the object of wish and the act is discovered through deliberation with a view to the end. A decision seems to be primarily a choice of “this for the sake of that,” and a preferential choice only to the extent that we need to determine which of two courses of action promote the end best.24 To fully defend that contention, however, is a task for another day.

Notes

1. A version of this paper was presented at the meeting of the Southern Association of Ancient Philosophy in Cambridge in September 2016. My chair, Gabor Betegh, as well as Christopher Rowe, Catherine Rowett and David Sedley deserve special thanks for their challenging questions. In thinking about the Eudemian Ethics, I have benefited from many illuminating conversations with Terry Irwin, and especially during a graduate seminar on Aristotle’s Three Ethics that we gave in Oxford in Trinity Term of 2014. I presented an early version of this paper at a conference at the Humboldt Universität in Berlin in 2016, and a paper on a related topic at the conference organised by Giulio Di Basilio at Trinity College Dublin in 2018. I wish to thank him for taking the initiative to publish this volume, and the participants for helpful discussion. Lesley Brown and Gail Fine offered characteristically insightful comments as my argument evolved.

2. As late as 1884, Susemihl printed [Aristotelis Ethica Eudemia] “Eudemi Rhodii Ethica” on the cover of the Teubner edition, agreeing with Schleiermacher, 1835 and Spengel, 1841 that it was inauthentic, possibly written by Aristotle’s pupil Eudemus of Rhodes. (Schleiermacher idiosyncratically held that the Magna Moralia was the only authentic Aristotelian treatise on ethics, but this view never caught on). In the early 20th century, Jaeger, 1923, turned scholarly opinion around, and since then, the designation “Eudemus” has gone out of style.

3. Harlfinger, 1971, p. 3.

4. Barnes, 1992, p. 18.

5. See also A. Kenny’s cautious remarks in the preface to his new translation of the Eudemian Ethics, Kenny, 2011, p. xxix.

6. Kenny, 1978.

7. Translations of the EN follow Irwin, 1999. My translations from the EE are based on Kenny, 2011, with multiple changes that I won’t flag as such.

8. “Choice,” W. D. Ross (rev.) Urmson, Aristotle Nicomachean Ethics (in Barnes, 1984); “Commitment,” Chamberlain, 1984; “Rational choice,” Broadie, 1991, p. 78; “Deliberate choice,” Sorabji, 1980; “Purposive choice,” Kenny, 1979, p. 69; “Preferential choice,” David Charles in a series of books and papers; “Decision,” Terry Irwin, Aristotle Nicomachean Ethics 2nd ed. and 3rd ed. and a series of articles; “Election,” Irwin, 2007.

9. Explicating translations like “rational choice” and “deliberate choice” moot observations Aristotle makes about prohairesis in the course of his inquiry, making his claims analytically true, which they are not in the original Greek. While capturing the goal-directed nature of prohairesis, “purpose” does not work well when Aristotle refers to decision as the efficient cause of action. We need an English term that covers all usages equally well, and that does not pre-empt Aristotle’s analysis.

10. Sylburg studied under Henri Estienne (Stephanus) in Paris, and contributed to his Thesaurus Graecae Linguae (1572). As an editor for the publisher Johann Wechel in Frankfurt (1583–1591), Sylburg corrected many Greek manuscripts, including all of Aristotle (1584–1587). He later became librarian in Heidelberg.

11. As well as by Bekker, 1831 and Fritzsche, 1851, prior to Susemihl, and Rackham, 1936 after him.

12. In the Nicomachean Ethics, Aristotle makes multiple remarks about decision before the topical analysis of EN III 2–3, including in the discussion of the voluntary in EN III 1, where he defines vice as “ignorance in the decision,” namely ignorance about good and bad. He does not, however, seek to determine whether the voluntary could be the same as decision, as he does in the EE, but rather takes it for granted that they differ.

13. This dialectical approach reveals that prohairesis was discussed by philosophers prior to Aristotle; as EE II 11, 1227b14 also makes clear (see also EN III 2, 1111b12: “those who say”, 1112a1: “now presumably no one even claims that decision is the same as belief in general”). Although Aristotle’s treatment of decision in his ethics is the first systematic analysis that has survived, Xenophon uses the verb prohaireisthai in explicating Socrates’ position (and especially his denial of the possibility of akrasia) in the Memorabilia (e.g. at Mem. 2.7.10; 3.9.4; 4.5.7; and 4.8.11), though without regimenting it into a technical term. Plato uses the noun only once, at Parmenides 143c3; he occasionally uses the verb prohaireisthai, but only in a technical (and non-ethical) sense, to describe a stage in the method of division (diairesis), e.g. at Sophist 251e1; 255e1; Statesman 257c1; 279b5; Parmenides 136c3. I take Aristotle’s dialectical approach to refute Danielle Allen’s claim that Aristotle single-handedly introduced prohairesis into the philosophical lexicon. See Allen, 2006, p. 184. The many uses of the term in the Attic orators from the 350s and onwards was probably part of Aristotle’s inspiration, rather than an effect of Aristotle’s use of the term in his public lectures on rhetoric.

14. Contrast this with the more dismissive assessment at EN III 2, 1111b12–14: “Those who say decision is appetite or spirit or wish or some sort of belief would seem to be wrong.”

15. Harlfinger writes: “Fest steht, daß Susemihls Beurteilung von P und C als codices optimi nicht aufrecht zu erhalten ist; wenn wir schon mit solchen Werturteilen arbeiten wollen, dann würde wohl eher der Kodex L – obwohl anderthalb Jahrhunderte jünger – dieses Prädikat verdienen” (p. 28).

16. I am grateful to Christopher Shields and Terry Irwin for helpful discussion on these points. Neither one of them should be presumed to share my views.

17. [Aristotelous[Aristotelous Ēthikōn megalōn biblion 2 ... ] = Aristotelis Ethicorum magnoru[m] libri 2. Ethicoru[m] Eudemiorum l. 7. De virtutibus & vitijs l. 1. Theophrasti characteres ethici. Alexandri Aphrodis. quod virtus non sufficiat ad beatitudinem. Francofurti: Apud heredes Andreae Wecheli.

18. Sylburg’s references are to the text as printed in the edition in which his comments appear. It antedates Bekker’s edition by two and a half centuries, and so uses its own numbering (rather than the Bekker numbers familiar to us). Anyone who has tried to work with older editions will appreciate the benefits of the Bekker convention.

19. I have not consulted a third, Simpson, 2013. See C. Rowe’s review of all three, Rowe, 2015.

20. Walzer and Mingay print a comma after epistêsantas, and then obelize “thesthai,” even though it’s the unanimous reading in the MSS. “Dei thesthai” is a scholion from the margins of the Codex Vaticanus (P) added by someone needlessly worried about the grammar. See Barnes’ apt observations, Barnes, 1992, p. 29.

21. Joseph Karbowski ends up in confusion while commenting on the passage in Karbowski (2013). In footnote 33, Karbowski remarks: “Reeve’s translation ‘tools for using tools’ (Aristotle: Politics, 6) is not inaccurate, but it does not vividly convey the special status among tools slaves have suggested by the ‘pro’ language; cf. PA IV.10, 687a19. For this reason I prefer Brunt’s suggestion of ‘before’ for ‘pro’; see Brunt, Greek History, 387.” The reasoning here is not clear to me. Later, Karbowski takes the “pro” to merit talk of the slave as a “second-order tool.” To further complicate matters, Karbowski gives two divergent translations of the same line in short order on the same page. Just above the quoted passage he has offered: “So a piece of property is a tool for maintaining life; property in general is the sum of such tools; a slave is a kind of [ti] animate piece of property; and every assistant is like a tool beyond [pro] tools. (Pol. I.4, 1253b23–33).”

22. Fine, Irwin, 1995.

23. For instance, “pro ergou” (“contributing to the task”), at EE 1215a8; 1220a22; also EN 1113b28 (“ouden pro ergou” means “pointless”). He uses “pros to ergon” in the same sense at EE 1242a16 and EN 1098a31. In MSS of Plato’s Meno, scribes frequently diverge on whether pro and pros is correct, for instance:

• 74B3 probibasai W F (advance, go forward)/prosbibasai B T (liken, convince, “bring something/someone over to”).

• 75d6 proshomologêi BTW/proshômologei F/prohomologêi Gedike.

This shows that scribes are frequently confounded by these prepositions, and that in some cases, though not all, their choices affect the meaning of the passage. I am grateful to Lesley Brown for drawing this to my attention.

24. See my analysis in Nielsen, 2011, 2018.

References

1. Allen, D., ‘Talking about Revolution: A Political Change in Fourth-Century Athens and Historiographic Method’, in Goldhill, S. and Osborne, R. (eds.), Rethinking Revolution Through Ancient Greece. Cambridge University Press, 2006.

2. Barnes, J. (ed.), Aristotle: Complete Works. Princeton: Princeton University Press, 1984.

3. ———, ‘An OCT of the EE’, Classical Review, NS 42, 1992, pp. 27–31.

4. Bekker, I., Aristotelis Opera, 5 vol., Berlin: G. Reimer, 1831.

5. Broadie, S., Ethics with Aristotle. Oxford: Oxford University Press, 1991.

6. Bywater, I., Ethica Nicomachea. Oxford: Oxford Classical Texts, 1894 (first published in 1890).

7. Chamberlain, C., ‘The Meaning of Prohairesis in Aristotele’s Ethics’, Transactions of the American Philological Association 114, 1984, pp. 147–157.

8. Charles, D., Aristotle’s Philosophy of Action. London: Duckworth, 1984.

9. Dirlmeier, F. (ed.), Eudemische Ethik. Berlin: Akademie Verlag, 1962.

10. Fine, G. and Irwin, T. H., Aristotle: Selections, Indianapolis: Hackett, 1995.

11. Fritzsche, H, Eudemi Rhodii Ethica Eudemia, Ratisbonae: G. I. Manz, 1851.

12. Harlfinger, D., ‘Die Überlieferungsgeschichte der Eudemischen Ethik’, in Harlfinger, D. and Moraux, P. (eds.), Untersuchungen zur Eudemischen Ethik, Akten des 5. Symposium Aristotelicum. Berlin: de Gruyter, 1971, pp. 1–50.

13. Harrison, G., A Treatise on the Greek Prepositions. Philadelphia, PA: J. B. Lippincott & Co, 1858.

14. Inwood, B. and Woolf, R. (eds.), Aristotle: Eudemian Ethics. Cambridge: Cambridge University Press, 2013.

15. Irwin, T., Aristotle Nicomachean Ethics (2nd ed.). Hackett: Indianapolis, 1999.

16. ———, The Development of Ethics (vol. 1). Oxford: Oxford University Press, 2007.

17. Jaeger, W., Aristoteles, Berlin: Weidmann, 1923.

18. Karbowski, J., ‘Aristotle’s Scientific Inquiry into Natural Slavery’, Journal of the History of Philosophy 51(3), 2013, pp. 331–353.

19. Kenny, A., The Aristotelian Ethics. Oxford: Oxford University Press, 1978.

20. Kenny, A. (ed.), Eudemian Ethics. Oxford: Oxford University Press, 2011.

21. ———, Aristotle’s Theory of the Will. New Haven, CT: Yale University Press, 1979.

22. Nielsen, K. M., ‘Deliberation as Inquiry: Aristotle’s Alternative to the Presumption of Open Alternatives’, The Philosophical Review 120(3), 2011, pp. 383–421.

23. ———, ‘Deliberation and Decision in the Magna Moralia and Eudemian Ethics’, in Brink, D. Sauvé-Meyer, S., and Shields, C. (eds.), Virtue, Happiness and Knowledge: Essays for Gail Fine and Terence Irwin. Oxford: Oxford University Press, 2018, pp. 197–215.

24. Rackham, H. (ed.), The Eudemian Ethics, London: Loeb, 1935.

25. Reeve, C. D. C. (ed.), Aristotle’s Politics. Indianapolis, IN: Hackett, 1998.

26. Rowe, C., ‘Aristotle’s Other Ethics: Some recent Translations of the Eudemian Ethics’, Polis: The Journal for Ancient Greek Political Thought 1, 2015, pp. 213–234.

27. Schleiermacher, F., ‘Über die ethischen Werke des Aristoteles’, in Sämtliche Werke III 3, Berlin: G. Reimer, 1835, pp. 306-333.

28. Simpson, P. L. P., The Eudemian Ethics of Aristotle. New Brunswick: Transaction Publishers, 2013

29. Solomon, J. (ed.), Eudemian Ethics, in Barnes, 1984 (orig. 1915).

30. Sorabji, R., Necessity, Cause and Blame. Perspectives on Aristotle’s Theory. Ithaca, NY: Cornell University Press, 1980.

31. Spengel, L., ‘Über die unter dem Namen des Aristoteles erhaltenen ethischen Schriften’, Abh. der Bayer. Akademie III 2, 1841, pp. 437-496.

32. Sylburg, F., [Aristotelous Ēthikōn megalōn biblion 2 ... ] = Aristotelis Ethicorum magnoru[m] libri 2. Ethicoru[m] Eudemiorum l. 7. De virtutibus & vitijs l. 1. Theophrasti characteres ethici. Alexandri Aphrodis. quod virtus non sufficiat ad beatitudinem . Francofurti: Apud heredes Andreae Wecheli, 1587.

33. Walzer, R. R. and Mingay, J. M. (eds.), Aristotelis Ethica Eudemia. Oxford: Oxford University Press, 1991.

34. Woods, M., (ed.) Aristotle Eudemian Ethics. Books I, II, and VIII. Oxford: Oxford University Press, 1992.

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