Bryn Mawr Classical Review

Bryn Mawr Classical Review 2003.05.27

W.W. Fortenbaugh, Aristotle on Emotion. A Contribution to Philosophical Psychology, Rhetoric, Poetics, Politics and Ethics. First published in 1975. Second Edition with a new epilogue.   London:  Duckworth, 2003.  Pp. 142.  ISBN 0-7156-3167-5.  £12.99 (pb).  

Reviewed by Peter Aronoff, Marymount School (

Duckworth provides a welcome reprint and slight update of William Fortenbaugh's classic work on the emotions in Aristotle. As F. notes in the epilogue, when he first wrote this book, he was very much a pioneer. Now, however, there is a sizeable industry devoted to Aristotle on the emotions and to the philosophy of emotions more generally. In large part F. inspired the work on Aristotle that followed, and so it is very good to have the book back in print. For reasons of space I will not, however, try to show where F.'s thoughts fit into the current landscape of Aristotelian scholarship or current philosophical thinking about the emotions so much as to describe and evaluate this book on its own terms.1 First I will describe the main outlines of F.'s arguments, and then I will present two criticisms. I should stress right from the start that the book is extremely solid, and that it will be essential reading for anyone concerned with Aristotle's rhetoric, poetics, psychology, ethics, or politics.

The book is highly compressed but abundantly clear. The length of the work (92 pages for the original, plus a new 34 page epilogue) belies its weight, so to speak. F. presents a very broad range of material in a tight space, and readers must pay attention to every detail. Still, F. writes quite clearly and he often discusses key points in more than one place. In fact, although F. includes a great deal of detail, the book's central thesis is overwhelmingly clear: as a development of debate and inquiry in Plato's Academy, Aristotle presents a cognitive view of emotion. According to F., this cognitive analysis of emotion allows Aristotle to rehabilitate rhetoric and poetry and to make significant advances in psychological, ethical, and political theory. F. organizes his material into four chapters: Aristotle's analysis of emotional response; a new political-ethical psychology; consequences for political theory; and consequences for ethical theory. The epilogue takes up a number of more specialized topics and allows F. to consider other interpretations of Aristotle on the emotions.

F. argues that Aristotle's analysis of the emotions is cognitive. A cognitive view of the emotions emphasizes the role of thought in emotion. So, for example, in the case of fear a cognitive analysis will emphasize that we fear what we think will harm us. Fear, on such an account, is not merely a blind or irrational feeling; instead the emotion is a thoughtful response to the world around us. What a person fears (or hates, pities, feels anger at, etc.) follows from how that person views the world. The importance of this insight cannot be emphasized strongly enough, since it goes counter both to a very large tradition of philosophy and psychology and to much of ordinary talk about emotion. According to these rival views, emotions are essentially brute, irrational, thoughtless feelings. Against such views, F. presents an Aristotle who puts thought rather than feeling at the heart of his analysis of emotion.2 This is not to say that feeling has no importance, and F. makes very clear that Aristotle does not study emotion purely from a cognitive viewpoint. That said, F. certainly thinks that cognition is central to Aristotle's understanding of the emotions, and he feels that such a cognitive approach to emotion allows Aristotle to make significant breakthroughs in other fields where emotion is important.

According to F., Aristotle builds on Plato as he develops his cognitive view of emotions and Aristotle's improvements allow him to answer Platonic attacks on rhetoric and poetry. In the Philebus Plato shows awareness that emotions like fear involve thoughts, but he does not make clear exactly what role thought has in emotion. Aristotle, on F.'s account, specifies a causal role for thought. While Plato in the Philebus simply says that fear occurs 'with' (meta) a thought, Aristotle explicitly replaces the vague 'with' and prefers the more explicit 'because of' (dia). This advance in analysis allows Aristotle to give emotion a more respectable place in human life and to answer some of Plato's attacks on rhetoric and poetry. If rhetoric and poetry move us partly by altering or stimulating our emotions, this is no longer necessarily a disparagement of them. Emotions rely on thoughts, and therefore Aristotle shows us the possibility of reasoned and reasonable emotional appeals from orators or poets.

Moreover, F. argues that Aristotle's theory of the emotions allows Aristotle to develop a more plausible moral psychology than Plato. F. claims that Aristotle replaces Plato's tri-partite analysis of human motivation and moral psychology with a bi-partite analysis. In the bi-partite analysis, human beings have a rational and an arational side to them. The emotions belong to the arational side, and in an ideal case a person's emotions are obedient to the person's reason.3 F. believes that bi-partition is a significant advance on tri-partition. In the tri-partite soul, each part has its own reason, emotion, and desire, but in the bi-partite soul, emotion and desire are grouped together and clearly separated from reason.

Once Aristotle advances such a bi-partite psychology, he puts it to use in his politics and ethics. Since animals entirely lack reason, Aristotle gains a principled way to show how ethics and politics are distinctly human achievements. Aristotle uses the distinction, on F.'s account, to make important points about children, slaves, and women. The arational side of a person develops first, and so this explains how to approach the education of children and how to explain their proper role. We must habituate their arational side and give them reasons; we should not expect children to formulate their own rules of conduct. On the other hand, since reason does develop as people age, educators and leaders should attend to the proper education of children's reason very diligently. More sinisterly, bi-partition also plays a role in explaining slavery and the place of women in society. According to F., Aristotle makes a consistent and valid argument for natural slavery, even if we may reject his factual premises.4 The argument runs roughly as follows. A natural slave lacks a developed logical side; he or she cannot deliberate, reason, or reflect on choices. Such a person, however, does possess an arational side. Thus, he or she can obey reason. Since such a person lacks an internal reason, it is appropriate that an external reason should rule his or her behavior. As a child should take rules from a parent, a natural slave should take guidance from someone else -- someone who does possess reason. The case for women is parallel. For Aristotle women possess reason, but in a very weak or attenuated sense.5 According to F. this is how and why Aristotle justifies his attitude towards women and their place in society. F. also explains how Aristotle's conception of emotion and bi-partite psychology allows Aristotle to explain virtue and virtuous action.

My first criticism concerns the status of the Aristotelian analysis of emotions which F. describes. In a nutshell, should we believe that Aristotle had a proper theory of the emotions and that this theory appears, in a consistent form, in his Rhetoric, De Anima, and ethical works? F. seems to believe that the right answer is yes to both questions, but I remain unconvinced. Notoriously, we do not find Aristotle's examination of emotion where we would have expected it, namely in his psychological or ethical works. Instead, Aristotle discusses the emotions primarily in the Rhetoric. His stated concern there is not the emotions per se, but rather the aspiring orator's ability to rouse or alter the emotions of his audience, so that they will go along smoothly with his overall argument. For example, a defense attorney may try to inspire pity for his client or anger at the prosecution in order to achieve an acquittal. This is a very limited aim, and it is far from clear that it requires Aristotle to offer a comprehensive theory of the emotions. Moreover, in the De Anima Aristotle suggests that an adequate explanation of emotion would have to offer and integrate material and dialectical explanation (I 1). What Aristotle calls 'dialectical' here corresponds to his own definitions in the Rhetoric, but he refers to such definitions very diffidently: the [dialectician] would define [anger] as a desire of returning pain for pain, or some such thing (403a30-31). The dialectician's definition, as many have noted, looks like what Aristotle says himself in his Rhetoric. The problem is that 1) 'or some such thing' appears to heavily qualify the value of this definition and 2) Aristotle goes on to say that any such definition is incomplete, since it lacks sufficient attention to the material side of emotion. As Gisela Striker points out, however, Aristotle nowhere offers the unified theory that he demands in the De Anima.6 In the face of this it seems safer to describe Aristotle's comments in the Rhetoric as more of a sketch than a theory and to be more open to disagreements between that sketch and what Aristotle says about emotions in other places.7

My second criticism concerns F.'s handling of two very controversial topics, namely natural slavery and women, in Aristotle. To be blunt, what troubles me is that F. does not do more to distance himself from Aristotle's position here. Toward the end of his consideration of Aristotle on women, F. writes, "Aristotle's view of women may be false" (61), and he adds in a footnote a bit later, "This is not to say to say that Aristotle's view of women is altogether free of prejudice" (61, n.2). First, this disclaimer is far too weak given that Aristotle argues that women lack the ability to use reason to govern emotion; secondly, even this weak disclaimer only highlights that F. offers no disclaimer of any kind when he discusses Aristotle's views on (so-called) natural slavery. At times, F.'s language even suggests that Aristotle was right:

When Aristotle says that slaves do not possess the deliberative capacity, he is not drawing a conclusion based on the menial role of slaves. Rather he is indicating why slaves have the role they do. They lack the capacity to deliberate, that is to say the ability to act with forethought (1252a31-2). When this deficiency is combined with bodily strength suitable for necessary tasks (1254b28-9), then the role assigned to slaves by society seems to be a natural role. Similarly in the case of women a reference to their psychological make-up combined with their bodily condition explains their role within the household and therefore ultimately their peculiar kind of virtue (59).

F. may take it for granted here that his readers will automatically assume that he is merely explicating Aristotle's argument, but I wish that he had been more explicit. We have very strong reasons to deny the factual premises of Aristotle's argument (women are thus and so; there are natural slaves), and F. has, I think, gone too far in his defense of Aristotle.


1.   For the help of more specialist readers, I will include some information on these topics in footnotes.
2.   Many if not most contemporary philosophers take a similarly cognitive approach to the emotions, and for such philosophers Aristotle is a crucial trailblazer. See for example, William Lyons Emotion and Michael Stocker and Elizabeth Hegeman Valuing Emotions.
3.   It may seem surprising that F. would label the emotions 'arational' after putting so much emphasis on cognition within emotion. There is no contradiction, however, and in fact, F. highlights an important detail in Aristotle. As F. points out the contrast here is not between 1) thoughts and 2) drives but rather between 1') reasoned reflection and 2') emotional response (24-25). In the case of the second contrast, both reasoned reflection and emotional response are cognitive in a broad sense. For Aristotle, 'arational' may have two senses (see Nicomachean Ethics I 13). In one sense something is arational insofar as it does not possess reason at all. Aristotle gives the example of digestion and growth; these processes occur in humans entirely without thought or reason. On the other hand, something can be arational insofar as it can hear or obey reason but not reason or calculate for itself. Aristotle refers to a self-controlled person. Such a person obeys reason, but there is a part of him that resists. The person wants to do things that reason rejects, and even if the person obeys reason and remains self-controlled, Aristotle believes that the refractory desires show that there must be another part of the person. This part possesses reason in a sense since it can obey and follow reason, but it is arational insofar as it cannot, on its own, formulate reasons. Aristotle's final analogy is to a parent and child. A child may possess and follow a rule which a parent gives him, even if the child himself cannot (yet) arrive at a rule of his own.
4.   This exact way of putting the matter is mine not F.'s. I find this section of F.'s book disturbing, and I will have more to say about this below. For the moment, however, let me stress that I do not want to say that F. himself argues for natural slavery or the inferiority of women.
5.   In the Politics Aristotle notoriously writes that a slave entirely lacks deliberation, but that a woman possess deliberation but it is akuron and a child possesses deliberation but it is incomplete (1260a12-14). akuron here appears to mean something like 'unauthoritative'. The point seems to be that women can deliberate but their deliberation is (often?) unable to govern their behavior authoritatively. Again, see below for more on this.
6.   Striker makes this point in "Emotions in Context: Aristotle's Treatment of the Passions in the Rhetoric and His Moral Psychology", which appears in Essays on Aristotle's Rhetoric edited by Amélie Rorty (at pages 286-88). My thoughts on this issue are heavily dependent on this article, although I may very well have used its ideas in ways Striker would disavow. For other suggestions that Aristotle offers less than a robust theory of the emotions, see John Cooper "An Aristotelian Theory of the Emotions" in the Rorty volume and Jonathan Barnes in The Cambridge Companion to Aristotle (page 267).
7.   A secondary and more specialized problem with F.'s position is that he not only describes Aristotle as having a consistent theory, but he argues that the theory meets the standards which Aristotle sets in the Posterior Analytics for a demonstrative science (12-16). There is good reason to doubt, however, that anything meets those standards. See, for example, Jonathan Barnes 'Aristotle's Theory of Demonstration' in Articles on Aristotle: 1. Science.

Read Latest
About BMCR
Review for BMCR
Support BMCR

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010