Bryn Mawr Classical Review 97.3.16


Stephen Engstrom and Jennifer Whiting (edd.), Aristotle, Kant and the Stoics: Rethinking Happiness and Duty. Cambridge: Cambridge University Press, 1996. Pp. ix + 310. $54.95. ISBN 0-521-55312-1.


Reviewed by Brad Inwood, Classics, University of Toronto, inwood@chass.utoronto.ca.

Aristotle, Kant and the Stoics is a collection of ten essays originally prepared for a conference at the University of Pittsburgh in 1994. This volume brings together papers by many of North America's leading exegetes of Aristotle, Kant, and ancient Stoicism in an effort to explore several key issues in ethics and the history of ethics. Anyone with an interest in ancient ethics should at least take note of this volume.

In their introduction, the editors question the common understanding of Aristotle and Kant as clearly marked opposites, "paradigmatic representatives of ancient and modern ethics". Their view of the traditional contrast is summarized as follows:

According to this line of thought, Aristotle is a teleologist who insists that all action be done for the sake of eudaimonia, while Kant is a deontologist who insists that moral action be done for the sake of duty alone; Aristotle's conception of practical reason is monistic insofar as he justifies all action by appeal to the single end of eudaimonia, while Kant's conception is dualistic insofar as his distinction between categorical and hypothetical imperatives reflects a sharp distinction between moral and nonmoral reasoning; and Aristotle sees right action as determined by the virtuous agent's perception of what is required by particular circumstances, while Kant sees duties as rules prescribed by reason alone. (2)
And they hold that this dubiously stark opposition "exercises a pervasive influence on contemporary ethics," an influence they hope to undermine by the publication of this collection.

Two lines of thought suggested this agenda. First, there is a movement in the study of Kant which aims to bring out Aristotelian aspects of his thought and a parallel movement in the study of Aristotle's ethics which aims to explore its Kantian features. And second, ancient Stoicism has long been thought to display sharply Kantian features within the framework of a typically ancient eudaimonism. Hence the Stoics might form a bridge between Aristotle and Kant, both in terms of their ideas and also historically, in view of their considerable impact on Enlightenment thought. As it turns out, though, the potential for exploring the intermediary role (historically and philosophically) of Stoic thought is somewhat underdeveloped.

The volume consists of five pairs of essays. John McDowell's Deliberation and Moral Development in Aristotle's Ethics' is matched with Barbara Herman's Making Room for Character'. McDowell emphasizes the role of character development in establishing the foundation of Aristotelian ethics, our conception of the goal of life, and Herman argues that Kant too has the resources to assign an important role to character in his theory. McDowell's argument is simultaneously less novel (no one denies that character formation is vital to any account of Aristotle's ethics) and more controversial, in that he presupposes a strong form of internalism in his reading of Aristotle. By contrast, if Herman is right about Kant (and I am in no position to judge) then she is opening up important new ground by showing that there is less difference than has been thought between virtue ethics and Kant. Her critique of the limitations of internalist tendencies in non-Kantian theory is a helpful corrective to the too ready assumption that all forms of externalism are a threat to virtue ethics. Like both Aristotle (in my view) and the Stoics, she is alive to the existence of naturalistic external constraints on human well being that must be taken into account in any successful moral theory. It is a pity that she does not pursue the similarities of her own version of Kantian theory (whether or not it was Kant's) to Stoicism; for as she frames the issues, her character-sensitive Kant might well have felt at home in a discussion with Stoics of the way rational and formal considerations interact with the foundations of moral theory.

Modus ponens is, free association is not, an authoritative form for discursive thought. The moral law offers a comparable compelling form to our practical maxims of action. Why we are responsive to either formal constraint is not a simple matter: in both cases the answer involves some kind of appeal to conditions for our inhabiting an ordered world. (44)
Clearly there is common ground with Stoicism here. The contrast comes in that Kant's story about those conditions invokes transcendental arguments while the Stoic theory appeals to cosmological theses (see also John Cooper's paper in this volume). Similarly, Herman refers to "the idea that, as rational beings, it is part of our nature that the evolution of our desires will be, to some high degree, reason-responsive" (53) -- a notion elaborated in different ways by Aristotle, the Stoics and even by Plato. If this Kantian theory of character is really Kant's then Herman has put us onto a trail which would facilitate an important trans-historical debate about the naturalistic foundations of ethics.

T.H. Irwin ('Kant's Criticisms of Eudaemonism') and Stephen Engstrom ('Happiness and the Highest Good in Aristotle and Kant') compare Aristotle and Kant on central ethical concepts. Irwin's thorough and acute analysis of Kant's assessement of what he understood to be ancient eudaimonism establishes the basis for an historical grasp of how Stoicism has helped to shape modern ethics -- a kind of philosophical Rezeptionsgeschichte. Irwin shows, with an historical acumen too rare in most discussions of such topics, that Kant (and Reid as well) significantly misunderstood ancient eudaimonism. Kant's central criticism of ancient ethics is that it tends to subordinate morality and the practical reason which accompanies it to non-moral considerations, such as 'inclination' or indeed any conception of the highest good which is not itself defined by a moral law. This criticism hits the Epicurean position hardest; but, as Irwin points out, Kant recognizes sporadically that Stoicism can escape it, since the Stoics do acknowledge the supremacy of morality even within the framework of eudaimonistic thinking. (Irwin claims that Aristotle and Plato would escape on the same grounds.) In so far as Kant held that the very structure of ancient ethics (with its focus on the summum bonum) entailed the erroneous subordination of morality, his critique was not consistent. In Irwin's account, Kant himself failed to recognize what Stoic and Aristotelian ethics had to offer; when we do so we can see the potential for a reconstruction of Kantian ethics which might embrace important elements of ancient ethical thought. The case supporting this claim is typical of Irwin's work in the history of ethics, in that it builds equally on detailed historical understanding and on a careful analysis of the argumentation underlying the doctrines.

Stephen Engstrom's exploration of Aristotle's and Kant's conceptions of happiness (Glückseligkeit, eudaimonia) and the highest good comes off less well. Like Irwin, Engstrom focusses on Kant's "criticism of the ancients for taking the highest good as their starting point in ethics" (107), and goes on to set this in the context of Kant's general criticism of the attempt to derive moral rules from some prior and non-moral conception of the good. His argument, in broadest outline, is that despite its prior articulation of the highest good, Aristotle's theory contains features which can save him from Kant's criticism, so that the Aristotelian highest good comes closer to Kant's theory than we might have thought (112-120). The grounds of similarity lie in what Aristotle says about eudaimonia as the divine activity of contemplative reason.

As an interpretation of Aristotle, Engstrom's case is far from compelling, and involves a certain amount of special pleading, in particular about the relationship of contemplative to practical reason. The problem of the relationship between the inclusive and the dominant conceptions of the good life in the NE ought to have been incorporated into the discussion, and more argument would be needed to support the assumption that human contemplative activity possesses the sort of autonomy which divine nous has in virtue of having itself as its own object. (There is something too simple in Engstrom's claim that "ethical life reaches its highest form through its recognition, from its contemplation of itself, of its own godlike character".) In the second half of the article Engstrom (like Herman in this respect) would have done better to bring in Stoicism more explicitly, since some of the alleged similarities between Kant and Aristotle would in fact be more plausible if asserted about the Stoics. The Stoics usually seem to emphasize more than Aristotle does the autonomy of rational activity of the soul vis-à-vis the external world, and Kant's notion that the life of virtue is one free of the vestiges of unruly desires is closer to idealistic Stoic conceptions of virtue than to Aristotle's realism (125). Engstrom's somewhat strained attempt to construct for Aristotle a conception of the highest good independent of external (and so un-Kantian) factors would also have worked much better for the Stoics than for Aristotle.

Allen Wood ('Self-Love, Self-Benevolence, and Self-Conceit') and Jennifer Whiting ('Self-Love and Authoritative Virtue: Prolegomenon to a Kantian Reading of Eudemian Ethics viii 3') raise the level of debate. Woods is frank about the nature of the task which the book has taken on:

Comparing Aristotelian and Kantian ethical theory is, by the nature of the case, less an exercise in the historiography of philosophy than a contribution to ongoing debates in ethical theory itself. Kant's knowledge of Aristotle's ethical writings may have been largely indirect and was in any case not deep. (141)
Despite that note of warning, Wood develops an elegant and well supported argument about the relationship of self to self and of self to other in both Kant and Aristotle. We seem to be free of the drift towards special pleading which threatens much trans-historical comparison in the history of philosophy. Despite the surprisingly extensive range of similarities between the two theories, Aristotle and Kant must differ (see also the papers by Cooper and Jerome Schneewind). Basic differences in philosophical anthropology, and the fact (often suppressed by modern commentators) that Kant is thinking in a (post-)Christian framework make disagreement inevitable. For Aristotle, the self-love of virtuous people and the friendship of the virtuous for each other is refreshingly, though naively straightforward. For Kant, even good people stand in need of an external standard and constraint. For Aristotle as for the Stoics (again, strangely neglected in any essay where they would be most welcome), human nature is at least in principle capable of achieving simple goodness. For Kant, as Woods reads him, human nature is so deeply resistant to reason, so fundamentally divided, that even the best of us has constant need of a moral rule as a counterbalance to our propensity to the dangerous kind of self-love. This is one point on which even proponents of ancient virtue ethics should be able to see that Kant's theory is better grounded and more realistic than Aristotle's.

In various ways, several of the papers in this volume attempt to find an interpretation of Aristotle which makes him less vulnerable to Kant's general critique of eudaimonism. Jennifer Whiting does so, by arguing that Aristotle's kaloskagathos has a kind of moral autonomy which far exceeds that of the merely agathos man and brings him closer to a godlike moral standing. Her detailed argumentative analysis of Eudemian Ethics viii 3 and related texts leads her to view the kaloskagathos as "maximally self-determined", as a man whose nous, "with which he identifies, serves as an internal standard -- a 'law within' -- regulating his conduct". Choosing contemplation and virtue for their own sakes and not for the sake of external rewards (as the typical Spartan might do) makes him "internally and not externally motivated" (192). This is indeed a Kantian reading of Aristotle. (Whiting explores, briefly but with discrimination, the issue of why there should be such a similarity between Kant and Aristotle. The influence of the Socratic conditionality thesis is a plausible candidate, and would indeed be a partial explanation for the similarity these view have to some facets of Stoic thought. Her tentative proposals on 164 for sorting out the complex lines of historical influence suggest a very promising line of enquiry; one wishes she had developed these speculations at greater length.)

Worries remain, however. Is the strain of thought Whiting nicely detects in EE viii 3 and argues for elsewhere typical of Aristotle's views? Or is it, as some have thought, an exceptional part of his theory, which may not cohere with other emphases elsewhere in the ethics? Further (and this is also a concern for Engstrom's paper), is the identification with nous in fact a sign of autonomy for Aristotle's ideal man? If the object of nous is somehow self-contained or self-reflective for man (as it seems to be for the unmoved mover in Metaphysics Lambda), then perhaps so. But if the objects of contemplation include universally framed truths about the natural world, then our thought (unlike god's) is far from self-sufficient, and unlike Kant, Aristotle would be recognizing the godlike quality of human reason in a very limited way, and only in the non-practical sphere. Finally, is even authoritative virtue (as Whiting translates kuria arete) self-determined as she claims it to be? Like McDowell's, her case depends on a strongly internalist interpretation of Aristotle's ethics which remains controversial.

Christine Korsgaard ('From Duty and for the Sake of the Noble: Kant and Aristotle on Morally Good Action') finds an important similarity between Aristotle and Kant in their views of the relationship between moral goodness and the exercise of practical reason in choice. Rather than considering their respective moral theories from a detached and schematic perspective, she attends to an aspect of ethics which often gets left to one side. Kant and Aristotle both acknowledge that rationality is the central and defining trait of human nature, and consequently that the character of rational choice is a source of the moral goodness which can be found in human action. The ways in which Kant and Aristotle work out this common idea are quite different, and Korsgaard clearly distinguishes their conceptions of natural sympathy and sense of humanity. In this respect her discussion resembles that of Allan Wood. And as in Wood's paper, an important opportunity is missed to bring in Stoic views on these issues, views which would have made a great contribution to the aim of the collection.

Julia Annas ('Aristotle and Kant on Morality and Practical Reasoning') is paired with Korsgaard. Building on the general characterization of ancient ethics she advanced in The Morality of Happiness (1993), her paper, more than any other in the volume, puts the distinctive contribution of Stoic ethical theory into play as a foil for the Kant-Aristotle contrast. She finds the Stoics taking clearer and sharper-edged positions than Aristotle in moral philosophy; in this they resemble Kant more than Aristotle, who is methodologically committed to respecting widely held moral beliefs. The Stoic tendency is rather to press theoretical distinctions until they conflict with such common beliefs. On several aspects of moral theory, but most interestingly on the characterization of practical reasoning and on axiology (where the Stoics strictly demarcated the conception of the good in terms of virtue), she finds Kant and the Stoics in comparable positions in contrast to Aristotle. And of course, just as contemporary moral philosophers track the debate between Kantian and broadly Aristotelian approaches, so in the ancient world the central debate in moral theory was between Stoics and Aristotelians. Much of what Annas says about ancient ethics is inevitably contentious (though less so here than in her 1993 book), but she succeeds in focussing the issues in moral theory which animate this volume more effectively than any other contributor.

I have complained about missed opportunities to integrate Stoicism with the Aristotle-Kant contrast, though the papers by Annas and Irwin certainly escape that criticism. In a way, the organizers can be held accountable, since there are two papers devoted to Stoicism, hived off into their own section (John Cooper, 'Eudaimonism, the Appeal to Nature, and "Moral Duty" in Stoicism' and Jerome Schneewind 'Kant and Stoic Ethics'). Perhaps it would have been better to encourage more authors to tackle the three philosophical positions advertised in the title. That said, the two papers dealing directly with Stoicism are excellent. Cooper, whose paper grew out of an extended critique of Annas' book, presents an authoritative and largely orthodox interpretation of Stoic ethics. It is bound to be critical of Annas, therefore, and this criticism perhaps occupies too much of the paper, which might better have engaged with other contributions to this volume. But what matters for the comparison to Kant is the selection of themes in Stoic ethics which Cooper emphasizes, and here his priorities cannot be bettered.

So on the Stoic theory virtuous persons act always, first, for the sake of their own happiness (referring all their actions to happiness as their single comprehensive 'end'); second, choosing virtuous acts for their own sake, as parts of their happiness; and third, choosing them as conforming to the universal law, that is to the will of Zeus. (262)
Cooper aims to "defend the philosophical respectability of each of these three elements in the Stoic theory of the virtuous life," and rightly emphasizes that it is the third element which offers the most enlightening parallel with Kant. For Kant, like the Stoics, connects morality with conforming to a rational law. The differences between their conceptions of rational law are far from trivial, but despite Aristotle's use of the idea of orthos logos they stand together against Aristotle (just as the Stoics stand with Aristotle as eudaimonists). In his final section Cooper makes a direct comparison between Kant and the Stoics. Despite all the similarities between them, there are more important differences. For the Stoics, morality and 'duty' cover a wider range of behaviours than in Kant; in ancient ethics generally one does not expect to find a distinct sphere of life distinguished as the moral sphere. Every aspect of our private and personal lives can be an expression of morality -- even the relatively pleasant aspects of life which involve acting out of normal inclinations; there is no special mark of morality in having to overcome a recalcitrant human nature. And the conception of universal reason in Stoicism is far more closely integrated with nature (indeed identified with it) in its cosmological role than it could possibly be in the eighteenth century, or after. As Cooper says,
Since Kant's conception of universal reason as the source of moral duties has its place in a political and religious world that made human freedom and equality in precisely this sense the central moral idea, one should not fail to see that the Stoic idea of universal reason, together with its consequences for ethical theory, constitute what amounts to a different moral universe. (278).
That conclusion sounds the right note for Schneewind's concluding paper, which somewhat subversively (given the theme of the collection) frankly emphasizes the importance of the discontinuities between Kant and the Stoics. For those readers whose interests and predilections are historical, Schneewind's paper will be pure delight. He argues trenchantly (against Paul Guyer and Barbara Herman) that by leaving deontology behind we are abandoning Kant himself and pursuing a generic and historically resituated Kantianism. And yet the elision of deontological features in Kant is at the heart of the case for reconciling him with the ancients. He then turns (like Cooper) to a critique of Annas' somewhat Kantian reading of Stoicism, emphasizing Kant's easily documentable dissatisfaction with the ancient school (a theme also developed by Irwin). Schneewind concludes with a brief and pointed appreciation of the differences between them, and with a proposal that most classicists and historians of thought will find appealing. We ought, he thinks, to focus our minds on why different issues in moral philosophy become salient in different periods -- for large cultural movements do have an impact on our philosophy, and it is false both historically and philosophically to try to tell the tale without including such changes. Much exciting comparative work in the history of ethics tends to flatten out such differences, resting on the implicit conviction that philosophical issues can be located against a single, stable ideological backdrop. Schneewind knows that they cannot. In his final section he suggests that:
moral philosophies can be viewed as sets of tools some cultures develop for coping with various problems in the ways we talk about or understand our common life. On that view, the changes will be as important as the continuities. (298)
This contingency in the history of ethics is important not just for those who simply like getting things right, historically (for that is not a universal or necessary preference); it is, as Schneewind also knows, philosophically important. Engagement with the differences and discontinuities "may also lead us to question our assumptions about the problems we put at the center of our own work" (299). And such questioning is essential to any philosophical endeavour.

There is an irony in this collection of papers. Philosophically and historically, the Stoics were the best bet for bridging the gap between Aristotelian and Kantian ethics. And yet those who work hardest to bring them together make least use of the Stoics, and those who engage most closely with Stoicism are least comfortable with effacing major differences between ancient and Kantian ethics; the major exception is Julia Annas, whose interpretation of Stoicism is important, though still controversial. Methodological inclinations clearly have enormous influence on each of the papers in this collection. Most readers of this review will probably lean to the historical side of the familiar polarity, but they should not therefore conclude that Kant has nothing to do with the Stoics and Aristotle, and that students of ancient ethics have nothing to learn from an understanding of Kant. For there are abundant starting points in this collection (especially in the papers by Irwin, Wood, Whiting, Annas, Cooper and even Schneewind) for a fully historicized understanding of the philosophical relationships between ancient and modern ethics. And such an understanding is vital to historians of ancient thought, if only because all of us have been educated in a world shaped more by Kant and his followers than we normally care to realize. This volume, for all its limitations, provides us with the best tools to date for tackling the task. It would be churlish indeed to complain that a great deal remains to be done.