<i>Fractured Goodness: Aristotle’s Response to Plato’s Form of the Good</i>

In the first book of his Nicomachean Ethics (EN), after his remarks on different ways of life in 1.5, Aristotle interrupts the discussion of happiness, the highest end, and turns to ‘universal’ goodness (1.6), by which he means to the Platonic Form of the Good. Against the existence of a Form of the Good, he sketches a series of technical arguments to the effect that the good is not, in fact, something universal. Most of these arguments point to a relevant diversity among goods—a diversity that is, in Aristotelian terms, incompatible with the good’s being ‘universal’, ‘common’, and ‘one’ (1096a28). Perhaps the most important diversity is that the good is not univocal: it is said in as many ways as ‘being’ (1096a23–24). Aristotle adds that, even if there were a Form of the Good, it would be of no use for ethics. Then he returns, without much ado, to the zêtoumenon agathon, the highest end he has been discussing in 1.1–5.Aristotle’s readers are usually at a loss with this chapter, which looks like a foreign body in EN 1; and while some tend to regard it as a failure, others downplay its relevance for Aristotle’s ethical project. Against both of these tendencies, Christopher Shields’s Fractured Goodness (FG) offers a welcome antidote. First, the book defends the relevance of EN 1.6 by arguing that the attack against Plato reveals a challenge that Aristotle has to answer himself: If goods are so diverse that the good is not univocal, then how can they be compared qua goods? And how can there be anything like a science of goodness? As Shields argues, the answer to these questions lies in the concept of core dependent homonymy (chap. 9.2), which makes possible both comparisons among goods, based on a causal perspective (chap. 9.3), and a science of goodness qua goodness, modeled after the science of being qua being (chap. 10). Readers familiar with Shields’s work will notice the close connection to his Order in Multiplicity (1999). In a way, FG is a sequel to this influential study. Second, FG insists that the arguments against Plato are not in fact a failure. Instead of being ‘partisan’ (chap. 2.3) toward Aristotle, however, Shields aims at a fair assessment that does not ignore the problems those arguments raise but nonetheless identifies their valid points. This is done in a careful and detailed study of EN 1.6. This meticulousness, the persistence that Shields both shows and expects from his readers, is one of the main virtues of his book.As this is not the place to follow Shields into the details, I will confine myself to some observations that might help to characterize his project. I will also try to raise a worry that I think is likely to arise for readers of FG.The core of FG is a close reading of EN 1.6, devoting one chapter to each of Aristotle’s arguments against the ‘Platonist’ (chaps. 2–8) (see chap. 1.1.7 on the identity of Aristotle’s target). The strategy of those core chapters might be put as follows. After presenting the argument at stake in the sketchy form of EN 1.6, Shields explains why that argument is not convincing, as it stands. As it turns out, the different objections of EN 1.6 raise different problems. Some seem to rest on equivocations, to be unspecific, or to beg the question; others do not specify what the charge exactly is, or they fracture goodness to an extent that Aristotle could hardly find attractive. To pick out just one example, the argument from the diversity of the sciences (1096a29–35) seems to rest on observations concerning actual sciences, which are, as Shields asserts, insufficient to establish the impossibility of a unified science of goodness (chap. 5.1).In the next step, Shields suggests a ‘development’ of the argument as presented in EN 1.6, drawing on other parts of Aristotle’s writings. Again, this development takes different routes in different cases; and by following these routes, Shields turns his interpretation of EN 1.6 into a study of some fundamental aspects of Aristotle’s philosophy. Thus in the case of the argument from the sciences he considers Aristotle’s conception of the ‘requisites’ of a science from the Posterior Analytics (chap. 5.2), and develops, in a veritable tour de force, what he calls the “predicational purview” of that science (135).Finally, Shields concludes his reading in each case with a reassessment, pointing out that, in its developed form, the argument makes a valid point against the Platonist (which does not mean that the latter might not have an answer to it).There is a lot to learn from, and a lot to ask about, these chapters. What stands out is how consistently they correlate the argument of EN 1.6 with the topic of nonunivocity, sometimes to an extent that might appear strained, given the highly compressed nature of Aristotle’s objections (the argument from the sciences is a good example). This brings me to an observation that I take to be crucial for a proper assessment of Shields’s perspective. The chapters of FG just mentioned are, as it were, engaged in a reconstructed debate, for they confront a developed version of the Aristotelian arguments with a comprehensive account of the Form of the Good. Now, it surely is an important lesson that such a developed version can make a point that should make the Platonist think. One might worry, however, that by leading that reconstructed debate, FG tends to lose sight of what Aristotle is actually doing in EN 1.6, what he thinks he has achieved, and how that affects the course of EN 1. In contrast to Shields, Aristotle draws on a biased and selective view of Plato’s theory. He apparently wants to make a specific point against Plato and ignores many aspects that the latter’s partisans would find indispensable. If this selectiveness is no accident, then it seems natural to ask, Why does Aristotle turn to the Form of the Good at this point of EN 1, and why does he treat it in that peculiar way? On these questions, FG is obviously less informative.A similar observation can be made concerning Shields’s account of the relevance of EN 1.6. As already mentioned, he argues that the variety of goodness spells a problem for the comparison of goods that Aristotle himself is engaging in, for this comparison requires a common standard.1 While a Platonist approach finds that standard in the “propinquity metric” (195), Aristotle’s solution lies, as Shields holds, in the tool of core dependent homonymy. The problem is of course that ‘core dependence’ (pros hen) is only one of the tools mentioned in EN 1.6 and that Aristotle does not develop the respective theory. The ambitious agenda of FG, chapters 9–10, is hence (1) to show that core dependence, not analogy, is in fact Aristotle’s answer to the question how the different goods are related, (2) to spell out this core dependence, and (3) to outline how it makes possible a science of goodness qua goodness and a comparison among goods.I cannot do justice to this extremely subtle discussion, so let me just note that it again deals with a reconstruction, for it asks, What would a science of goodness qua goodness, which Aristotle nowhere develops, look like? Drawing on Aristotle’s remark that going into details would be the matter of a “different philosophy” (1096b29–30), one might characterize FG as being about that different philosophy rather than about the importance of (the proposed reading of) EN 1.6 for the aims of EN 1. To be sure, this is a perfectly legitimate and important project. But it is obviously less interested in what we can learn about Aristotle’s engagement with the Form of the Good from the very context in which it appears. I believe, however, that this context throws a different light on some of Shields’s assumptions.Throughout EN 1.1–5, Aristotle conceives of goods as ends and of the highest good as a highest end. This teleological outlook grounds his general remarks on the ethical project as well as his discussion of ‘opinions’, because all candidates for happiness treated in EN 1.1–5 are (highest) ends. When he turns to the Form of the Good, Aristotle does not just introduce a “contender” (176) for the highest good but a different way to conceive of it: not as a highest end but as something that possesses goodness to the highest possible degree. A question lying behind this “interlude” (17) is thus whether it is right to think of goods as ends. Aristotle does not raise this question as explicitly as in EE 1.8, but since the overall dialectic of EN 1 is still shaped by it, one might want to describe that dialectic differently than Shields: not in terms of ‘normative’ versus ‘axiological’ (chap. 1.3), nor ‘circumscribed’ versus ‘absolute’ (chap. 1.2), but in terms of a ‘teleological’ versus a ‘paradeigmatist’ view of the best good.A related difference can be observed with regard to the comparison of goods. Shields is right in claiming that EN 1.6 offers a challenge to Aristotle’s own project of comparing goods. But this challenge is more specific than Shields assumes, because Aristotle has already spent five chapters on that project. He has explained how goods can be compared if they are conceived as ends. So the specific question seems to be whether this teleological account of the highest good is in any way affected by the results of the anti-Platonic argument. And although it is impossible to be sure on that matter, one might find this specific question present in the ‘Platonic Rejoinder’ that Shields discusses in chapter 7, for the intrinsic goods are here characterized as being pursued for their own sakes (1096b10–11). If they, too, are different qua good, one might worry that a teleological account of the good leads into an irreducible plurality of (highest) goods. However that may be, given the context of EN 1, Shields would have to explain why Aristotle should find problematic something that he has been doing all along: comparing goods.Relating EN 1.6 to the context of EN 1 might finally affect our position toward the alternative between analogy and core dependent homonymy. Putting aside the question whether analogy really reintroduces “Platonic univocity, simply at a higher level” (212) (I think it does not), it can hardly be denied that the argument of EN 1.1–5 is shaped by the language of analogy—that is, by the remark that the good-as-end is different in different cases (1095a23) (see Brüllmann 2011). It is this remark that reappears in 1.6 when Aristotle sketches what an analogy is (1096b29) and in 1.7 when he returns to the teleological perspective (1097a20). If we take this observation seriously, then the option ‘analogy’ (or a combination of analogy and core dependence: 210n32) should not be dismissed so easily. In this case, we might also take a different stance on the alternative ‘absolute’ versus ‘circumscribed’ goodness. For whatever Aristotle’s position on the existence of absolute goodness might be, his very choice of a teleological approach—at least as laid out in EN 1.1–5—seems to imply a circumscribed perspective.These observations do not diminish the achievements of this absorbing book, which covers new ground in scholarship and combines passion for details with an impressively wide scope. They rather show how stimulating it is.