Bryn Mawr Classical Review 04.03.24


C.D.C. Reeve, Practices of Reason: Aristotle's Nicomachean Ethics. New York: Oxford University Press, 1992. $49.95. ISBN 0-19-823984-X.


Reviewed by J. Bussanich, University of New Mexico.

Aristotelian FRO/NHSIS, practical wisdom, comprises the practical intellectual skills for living a moral and happy life. But how practical is it? Recall that in the Politics only the ruler is said to have perfect moral virtue (i 1260a17) and practical wisdom (iii 1277a25). The rulers, the practically wise, are few, a point echoed in the Nicomachean Ethics, which is addressed to the morally mature (EN 1095a) -- a case of the enlightened (or: the properly trained!) conversing with their familiars. Plato too insists that political wisdom, i.e., the science of guardianship (Rep. iv 428d6), is the virtue of the smallest class (428e7-9). Aristotelian FRO/NHSIS, like its Platonic counterpart, is an architectonic or supervisory type of knowledge and so not primarily practical in the common meaning of the term. In his fine new book David Reeve argues, against scholarly orthodoxy, that FRO/NHSIS is a kind of scientific knowledge, despite its concern with the particularities of life.

Reeve states the puzzle he wishes to explore in these terms. If scientific, demonstrative knowledge is of what holds always and by necessity and if ethical knowledge concerns things that vary and are contingent and about which exactness is unattainable, how can this practical type of knowledge be described as scientific? A comprehensive answer to this difficult question is framed in four chapters. Chapter 1 focuses on dialectic and scientific knowledge of ethical universals. Chapters 2 and 3 articulate, in turn, the particular and universal functions of FRO/NHSIS and its relation to NOU=S and EU)DAIMONI/A. In the concluding fourth chapter Reeve traces the implications of his argument for current debates about the questions whether EU)DAIMONI/A is an inclusive or dominant end, what is the proper role of external goods in the happy life, and on the relation between EN x and Met. vii.

The analysis conducted in chapter 1 is crucial for the overall argument of the book. Reeve makes a distinction between unconditional scientific knowledge and what he calls 'plain scientific-knowledge'. The former concerns necessary universal truths ('things that cannot be otherwise'), whereas the latter 'is of what holds always or for the most part' (Met. 1027a21). (Reeve [14-15] is rather more sanguine than other commentators that Aristotle has a worked-out logic of 'for the most part' propositions. For some healthy skepticism on this point see J. Barnes, Aristotle's Posterior Analytics, Oxford 1975, pp. 184, 229. In any case Reeve's argument is too compressed to be fully convincing.) Since, in Reeve's view, ethical principles (as enmattered universals) only 'hold always or for the most part' unconditional knowledge of them may not be possible. However, he argues, this judgment should not lead us to conclude that moral knowledge is limited to perception of particulars or to decisions about individual actions. Knowledge of ethical universals also, like the de re necessities that constitute natural laws in the biological works and the Physics must, he thinks, be scientific in some way (12-15). Aristotle's notion of natural necessity is looser than conceptual or analytic necessity, but comparing the Ethics to the scientific works does not ease the difficulties. Explanations of nature often take the form of contingent non-universal truths (e.g., 'most tigers have stripes'), so Aristotle must still confront his famous requirement that knowledge is only of universals or necessary truths. The problem remains whether Aristotle can claim scientific knowledge of either nature or human life given their fundamentally contingent features.

One way to loosen the stringent requirements for knowledge is to reformulate the criterion of exactness. Since 'exactness in a science is a function of its level of abstraction from matter' (21) precise scientific knowledge will hold of immaterial objects, e.g., those treated by mathematics and theology. Reeve bravely faces the difficulty that ethical universals (like their counterparts in the natural sciences) involve matter and that Aristotle is explicit that ethical truth can only be described 'in outline': PA=S O( PERI\ TW=N PRAKTW=N LO/GOS TU/PW| KAI\ OU)K A)KRIBW=S O)FEI/LEI LE/GESQAI (EN ii.2 1104a1-2). Reeve attempts to escape these constraints by specifying things that are 'unconditionally fine' about which the excellent person has 'unconditional knowledge'. This (almost Platonic) claim rests on the accuracy of the virtuous person's moral judgment. He cites EN iii.4 1113a30-33: 'For each state of character has its own individual view of what is fine and pleasant and the excellent person differs most from the others because he sees the truth in each, being a sort of standard and measure of what is fine and pleasant'; cp. Rep. ix 581-591. Now the attainment of truth is the function of all intellectual faculties (1139a27ff), so while we can agree with Reeve that the excellent person's knowledge is objectively true, the difficulty remains that there are both practical (correct decision about an appropriate action) and theoretical truths (the sum of the angles in all triangles is 180°). Reeve should clarify why he thinks (since he seems to imply that) this grasp of practical moral truth is 'unconditional scientific knowledge'.

Reeve works towards an affirmative answer to this question by making the following points (25-26): (1) only matter poses an obstacle to unconditional scientific knowledge; (2) NOU=S and QEWRI/A are immaterial; (3) as human beings we are primarily NOU=S and we have NOU=S of ethical first principles; (4) the ultimate ethical first principle, EU)DAIMONI/A, is QEWRI/A, the activity of NOU=S. (Thus, 'knowledge of the first principle of ethics is the most exact form of unconditional scientific knowledge.') Reeve concludes that, (5) because FRO/NHSIS is for the sake of NOU=S and SOFI/A, (6) the FRO/NIMOS must possess wisdom and (7) hence FRO/NHSIS must be unconditional scientific knowledge (26). First, (1)-(4) are non-controversial; Reeve concludes that they support 'demonstration of unconditionally necessary ethical propositions' (28), though he admits that the extent of such knowledge is rather limited. (He might have noted that there are few if any actual cases of demonstrative arguments in the entire Aristotelian corpus. See Barnes 'Aristotle's Theory of Demonstration' Phronesis 14 (1969): 125-52.) Concretely, perhaps only EN x.7-9 exemplifies his description, with the rest of the treatise devoted to more contingent matters (more on this later). (5) is unsupported in the immediate context (though discussed in chapter 4); more importantly, (6) flies in the face of EN vi.6, which distinguishes sharply between FRO/NHSIS and SOFI/A; and (7) would seem to be at odds with vi.5, which explicitly declares that FRO/NHSIS is not E)PISTH/MH. In short, Reeve here seems to blur the distinction between the practical and theoretical intellects, i.e., FRO/NHSIS is assimilated to SOFI/A, though he states it clearly enough in later sections of the book (cf. 73-77, 94-97). Another point that left me puzzled is the assertion (25-26) that we have unconditional scientific knowledge of the ultimate ethical first principle, EU)DAIMONI/A, a claim that would seem opposed to Aristotle's view that knowledge of ultimate first principles is not scientific or demonstrative but NOU=S. By comparison, his account of plain scientific knowledge of ethical truths is clear (if not fully convincing). It concerns derived principles like the fine, the just, and the right, about which we can provide demonstrative causal explanations derived from our intuitive grasp (NOU=S) of the unconditional first principle EU)DAIMONI/A.

One reason the claims proffered in the important opening sections of the book may strike the reader as overstated is that Reeve's more detailed analyses of NOU=S and QEWRI/A and related matters, on which the conclusions of chapter 1 might be thought to depend, are to be found in chapter 4. However, I proceed on the assumption that an author's argument should be followed as presented.

Having characterized knowledge of ethical first principles, in the remainder of chapter 1 (31-62) Reeve addresses the important methodological topic of dialectic. He lucidly expounds these familiar points. Arguments to first principles are dialectical, while arguments from first principles are scientific (i.e., demonstrative); and EN is a compendium of the former. The starting point of dialectical arguments is endoxa, common or reputable beliefs, whose inconsistencies present puzzles to be solved through critical inquiry. Noting the limitation that dialectic is unable to prove or justify the truth of first principles (39), Reeve nevertheless believes that it 'clarifies first principles by going through the aporiai', both empirical and conceptual (40). The dialectical procedure is illustrated by a detailed account of Aristotle's famous analysis of A)KRASI/A in EN vii.1-3. Reeve adds some valuable remarks about Aristotle's reliance on experience to acquire accurate information about the world and how this empirical method supports a naturalized epistemology. Particularly interesting here is his exploration of the determinant effects of a person's 'desire structure' on her beliefs; see also the later discussion of emotions as 'modes of practical perception' (71-72). (This feature of Reeve's discussion will be familiar to readers of his Philosopher-Kings, Princeton, 1988.) It is perhaps worth noting that this non-cognitive, affective 'method' for attaining the good -- a person is habituated by training to choose or delight in the good: we 'reach our complete perfection through habit' (EN ii 1103a26, a quasi-Humean account) -- is often distinguished from the dialectical method, whereas Reeve, plausibly enough, sees them as complementary.

The concluding sections of chapter 1 (§§9-10) relate dialectic to NOU=S via discussion of the crucial passage at EN vi.11 1143a35-b5. Here NOU=S is employed to define both (a) the theoretical grasp of first principles (the culmination of the inductive process that extends from perception of particulars to conceptual grasp of universals) and (b) the perception of particulars. One problem is that neither first principles nor particulars are demonstrable, i.e., neither are objects of E)PISTH/MH, a difficulty that Reeve might have highlighted a bit more. Instead he stresses the continuity between the empirical and the theoretical, based on texts that apply the axiomatic model to practical thinking. This approach renders Aristotle more consistent than he probably is by weaving together seamlessly the dialectical and inductive methods. In Reeve's terms, as noted above, 'clarifying first principles' comes very close to 'grasping first principles'. It would have been helpful if he had attempted to show how the coherentist justification typical of dialectic fits the intuitive grasp (NOU=S) of the principles.

Chapter 2 provides a succinct though comprehensive account of the role of FRO/NHSIS in Aristotle's ethical theory, including informative outlines of deliberation, decision, and the nature of ethical virtue. FRO/NHSIS is concerned with both universals and particulars; in the pursuit of virtuous action FRO/NHSIS discerns that a particular action is an instance of the relevant ethical universal. In this context Reeve properly stresses the focus of FRO/NHSIS on practical perception and, at the same time, its dependence on a more abstract knowledge of universals (71-73). Though Reeve's analysis is lucid and informative it does not alleviate my doubts that Aristotle has not removed the inconsistencies between a method relying on the practical perception of contingent particulars and the formal demonstrative model of scientific knowledge of invariant truths.

Returning again to the relation between FRO/NHSIS and wisdom he argues that not only is FRO/NHSIS for the sake of the latter but that it 'brings it into being', all this based on cursory analysis of a brief passage in 1145a (96). The point is elaborated further in a comparison of divine and human NOU=S. NOU=S influences the good person as final cause, FRO/NHSIS exercises control as an efficient cause (135). More significantly, he employs the distinction to support these conclusions: 'study expressing wisdom is primary eudaimonia; practical activity expressing phronesis is secondary eudaimonia; and the latter is for the sake of the former' (97). Despite some differences this thesis is fairly close to Richard Kraut's attractive interpretation, which is developed at considerably greater length in his recent Aristotle on the Human Good (Princeton 1989), so further discussion is unnecessary.

Chapter 3 is devoted to almost introductory tours of landmarks in that extremely well explored territory EN i -- completeness, self-sufficiency, and choiceworthiness of ends, the function argument. While I agree with Reeve that the function argument especially supports the view that Aristotle's ethical theory rests on metaphysical foundations, his account offers little that carries the argument forward beyond the important detailed studies in its favor by Irwin, Sorabji, J. Lear, Bernard Williams, and Alasdair MacIntyre; nor does it offer much in the way of rebutting the arguments of those who insist that Aristotle's methods of ethical investigation need not rely on his metaphysical principles, e.g. Tim Roche 'On the Alleged Metaphysical Foundation of Aristotle's Ethics' Ancient Philosophy 8 (1988): 49-62.

In chapter 4 Reeve discusses divine and human NOU=S, QEWRI/A, and the degrees of happiness, with consideration of the theme of external goods. He is to be commended for bringing Aristotle's theology to bear on the concepts of EU)DAIMONI/A, NOU=S, and QEWRI/A, unlike Irwin, Nussbaum and others in the Ango-American philosophical tradition who find it irrelevant or embarrassing (Kraut is a notable exception). Reeve's account of human and divine NOU=S is excellent, though I have one objection. Following Irwin's translation of EN (Indianapolis 1985) he renders QEWRI/A as 'study', but Reeve offers no reasons why we should prefer this English term to the more traditional 'contemplation'. Now 'study' is not perhaps inaccurate when applied to the pure intellectual activity of the human NOU=S, but it does not work so well, it seems to me, when applied to Aristotle's god. Moreover, the summary of Met. xii 7 & 9, that god 'is immutably and eternally studying' (134), supports Reeve's conclusion that 'god, in a way, just is theology, a demonstrative structure of essentially matterless universals' (141); in fact 'god studies only essentially matterless universals' (144n9). In his characterization of both human and divine QEWRI/A Reeve appears to draw upon Irwin's definition of human QEWRI/A as study: 'Aristotle is not thinking of the inquiry needed to find answers I do not already have; he probably thinks of surveying the deductive structure of a demonstrative science, seeing how each proposition is justified by its place in the whole structure' (Irwin 1985, 427). But, in the at least prima facie different context of Met. vii, I cannot see why we should think that this mental inspection of a scientific taxonomy is the meaning of divine QEWRI/A. Aristotle's god, we should remember, contemplates only himself. But for Reeve this divinity is a matterless universal, or a demonstrative structure. On the first alternative, what are his particular instances? Or should we conceive of Aristotle's god as a set of necessary propositions? Reeve's all too brief foray into theology leaves too many questions unstated let alone unanswered. Perhaps one should note Reeve's gratuitous concluding remarks -- that science has 'decisively killed off' Aristotle's conclusions about happiness, god, and contemplation, and that they have 'become incredible' (196). It is a bit disconcerting to be advised throughout the book that Aristotle's arguments are good and then to have the rug snatched from underfoot.

This book is excellent in many respects: Reeve is an exceptionally clear and concise writer and a formidable dialectician who marshalls his arguments very well indeed. Many of his positions are right on the mark, in my view, e.g., on the relevance of Aristotle's metaphysical principles to his ethical investigations; on the degrees of happiness; on the role of external goods in Aristotle's conception of happiness; and on the importance of non-cognitive factors in the pursuit of moral ends. At the same time, it seems to me that many of Reeve's discussions, while presented lucidly and crisply, are to a great extent condensed versions of other scholars' work: Irwin on first principles, Irwin and Barnes on endoxa, Irwin and Kraut on the nature of ends, Cooper on A)KRASI/A, to mention a few examples. In one respect this feature of the book is an asset: it would serve well as a comprehensive introduction, for the experienced student, to Aristotle's thought via the Ethics -- arguably the best way to bring students to this most difficult of ancient philosophers. Supporting evidence for this suggestion: throughout Reeve quotes very extensively from across the entire Aristotelian corpus (I estimate that primary texts comprise more than one fourth of the book); he covers a bewildering variety of topics, too many for a pathbreaking scholarly contribution (for that one does need the 600 pages Irwin fills up in his Aristotle's First Principles, Oxford 1988), but attractive in an introductory book; and, most important, Reeve is a vigorous, exciting writer who make one want to know more about Aristotle. So the book is not just for students, I recommend it to all Aristotle scholars.