Bryn Mawr Classical Review 94.03.23

Jacques Brunschwig and Martha C. Nussbaum (edd.), Passions and Perceptions: Studies in Hellenistic Philosophy of Mind. Cambridge: The University Press, 1993. Pp. xii + 364. $69.95. ISBN 0-521-40202-6.

Reviewed by Lee T. Pearcy, The Episcopal Academy.

Passions and Perceptions continues the alliterative series that began in 1980 with Doubt and Dogmatism.1 Readers familiar with the earlier volumes will know what to expect: essays by various hands exploring the contributions of the Hellenistic schools to some major philosophical topic. The papers in this volume have grown from presentations and discussions at the Fifth Symposium Hellenisticum held in August, 1989, and they reflect the state of play at the high end of the professional tour for students of ancient philosophy; from an amateur like me, readers of BMCR should expect not a continuation of the discussion, but simply an account of what I found.

In "Epicurean Hedonism," Gisela Striker sets out to rescue Epicurus' distinction between active and static pleasure (or, to use Epicurean terms, "kinetic" and "katastematic") from the criticism put forward by Cicero in De Finibus II.9-10, namely that it makes no sense to use the one term "pleasure" both for the kinds of activities that most of us would recognize as pleasure and for the bloodless Epicurean absence of pain a state which, as robustly hedonistic critics of Epicurus observed, resembled sleep or death more than it did pleasure. She proceeds by detaching the arguments of De Finibus II from the account of pleasure put forward by the Epicurean Torquatus in De Finibus I.29-42.

Striker succeeds in using Cicero against himself to show that he has misunderstood the nature of Epicurus' distinction between kinetic and katastematic pleasures. Epicurus did not, she argues, distinguish two different pleasures, but rather extended the meaning of pleasure to cover not only active movement away from pain and toward pleasure, but also states of body and mind characterized by aponia and atraxia. As he himself observes (Letter to Menoeceus 128 = Long and Sedley 21B,2), pleasure is what we want when we are in pain, and what we don't want (presumably because we have it) when we are not in pain. If, therefore, we are by nature creatures of appetite, as Epicurus believed, pleasure is both the arkhê (because all appetites can be referred to it) and the telos of human life. This move, Striker argues, "allowed Epicurus to identify the greatest pleasure with the good life" (p. 17). Striker's persuasive argument has the additional merit of illuminating Cicero's philosophical and doxographical technique and rehabilitating De Finibus as a source for Epicurean ethics.

"Annicéris et les plaisirs psychiques. Quelques préables doxographiques," by André Laks, also addresses a doxographical issue, not incidentally but head-on. Diogenes Laertius II.86b-91a records the doctrines regarding pleasure of "those who remained in the school of Aristippus and were known as Cyrenaics." Klaus Döring has argued that it is possible to sift from this account at least two layers of hedonist argument: an original Cyrenaic doctrine attributable to the elder Aristippus, and an anti-Epicurean polemic originating with Aristippus' successor Anniceris.2 Laks re-sifts the evidence and reaches a different conclusion: (1) Anniceris' doctrine on pleasure was a natural outgrowth of the Cyrenaic doctrine held by the others who "remained in the school of Aristippus," not a radical departure in response to Epicurus' distinction between kinetic and katastematic pleasures; (2) Diogenes' references to Epicurus are "l'effet d'une construction doxographique, beaucoup plus que le reflet du débat"; (3) it was not Epicurus, but instead his fellow-Cyrenaic Hegesias whose doctrine forced Anniceris to reformulate traditional Aristippean hedonism.

Hegesias accepted the school's dissociation of happiness (eudaimonia) from pleasure and its recognition of pleasure as the aim (telos) of life. The rejection of eudaimonism entailed at least a demotion of happiness to a second-order good; Hegesias went a step, or several steps, further and asserted that happiness was impossible (TH\N EU)DAIMONI/AN O(/LWS A)DU/NATON EI)=NAI, D.L. II.94a) and that the altruistic virtues like gratitude, friendship, and kindness did not exist. It was Hegesias' radical pessimism, Laks maintains, not Epicurus' distinction of katastematic from kinetic pleasures, that led Anniceris to re-discover within traditional Cyrenaic doctrine a place for happiness and altruism. Laks' analysis has the merit of redirecting attention to the debate between Hegesias and Anniceris and of situating our evidence for their doctrines in the context of the Cyrenaic school's rejection of eudaimonism. If Striker's article showed that analysis of an ancient philosopher's thought could clarify doxography, Laks shows doxography clarifying ancient doctrinal disputes.

Julia Annas in "Epicurus on Agency" and David Furley in "Democritus and Epicurus on sensible qualities" explore ways in which Epicurus distanced himself from some of the counterintuitive propositions that had been put forward as consequences of atomism. In each case we see Epicurus attempting to save the phenomena, whether of moral intuitions or of sense-perception, and so to narrow the distance between the world of atoms and void and the world as it seems to us.

Annas takes up the question of what exactly Epicurus says about agency and responsibility in the fragments of On Nature preserved on a papyrus from Herculaneum edited most recently by David Sedley.3 Sedley has argued that Epicurus, unhappy with the strict determinism entailed by a reductive interpretation of atomist materialism, moved toward something like emergent dualism by distinguishing the qualities arising out of sufficiently complex states of matter like the soul from that matter itself, and giving those qualities a role in agency.

Annas, on the other hand, reads the fragments of On Nature to show that Epicurus need not have held a dualist view of any kind. In her view, his seemingly dualist distinction of "we (ourselves)," our "constitution," and its "development" represents nothing more than different ways of talking or thinking about the same thing. We can think of ourselves as our atomic constitution or as the development of that constitution. Rationality, and hence agency, is in fact the capacity for development inherent in some atomic constitutions: those of human beings and, interestingly enough, of some animals.

"Epicurus' account of agency," Annas concludes, "while incomplete and jargon-ridden, turns out to be relatively simple and attractive" (71). So it does, on her account of it. It is not clear to me that she has proven that Epicurus did in fact hold this simple and attractive notion of agency, but she has certainly shown that he might have held it. In the present state of our evidence, that may be all that can be said.

Furley explores a contradiction between the views on the value of sense-perception reported to have been held by Democritus and Epicurus. Both started from the same theory of atoms and the void, and both explained the mechanics of sense-perception in essentially the same way. Yet for Democritus the evidence of our senses was untrustworthy; for Epicurus, PA/NTA TA\ AI)SQHTA\ EI)=NAI A)LHQH= (Sextus Empiricus Adv. Math. VIII.63). The simplest solution to this contradiction is to suppose that Epicurus used alethe to mean "real" rather than "true." The contradiction thus dissolves. Democritus is asserting that our senses do not give us reliable reports about the true state of objects in the world. Epicurus agrees, but adds that the sense impressions themselves are not imaginary; they are events in the real world. This was the solution adopted by Furley himself twenty years ago.4

Now, however, Furley wants to argue a stronger interpretation: "that Epicurus asserted both that sense impressions are true, and that they directly convey the truth about external objects themselves, but that the content of sense impressions needs to be specified very exactly" (92). To support his argument he privileges two testimonies: the surviving part of Theophrastus On the Senses, and two anti-Epicurean treatises of Plutarch, Against Colotes and On the Impossibility of a Pleasant Epicurean Life. These show, he argues, that the two atomists' apparent agreement on atomism and the mechanics of perception masks an important difference over the ontology of perceptible objects. Democritus held that ta aistheta were not among the properties of the external world. Epicurus maintained that they were real properties both of the images (eidola) streaming from objects in the world and of the objects themselves. For Epicurus, a red ball really is in some sense red; for Democritus, its redness is an illusion.

Democritus, Furley suggests, was influenced by the Eleatic denial of our senses' report that objects in the world are subject to change. Epicurus, however, had the advantage of Plato's discussion of being and not-being in the Sophist and Aristotle's analysis of the problem, in particular his distinction of substratum from quality in Physics I. True enough; and there is also the rather likable Epicurean desire to account for all the variety and attractiveness of the way the world seems to us. Epicurus wants to save the phenomena because he admires and takes pleasure in them; Democritus, one feels, wishes that the phenomena would go away. The essays by Furley and Annas have the merit of showing us the philosophical rigor with which Epicurus pursued his goal.

The four essays by Striker, Laks, Annas, and Furley all either deal directly with Epicureanism or take it as a point of departure for consideration of related doctrines. The remaining seven papers have much the same relationship to Stoicism. Three of these, Martha Nussbaum's "Poetry and the passions: two Stoic views," Brad Inwood's "Seneca and psychological dualism," and James Hankinson's "Actions and passions: affection, emotion, and moral self-management in Galen's philosophical psychology," stand out not only for length (they take up nearly one-third of the text in this volume), but also for importance and for their relevance to areas of classical studies other than philosophy.

The Stoics are notorious for paradox. In "Poetry and the passions," Nussbaum sets out to discover their strategies for dealing with one implicit in their own doctrine and practice: on the one hand, the Stoics are unrelentingly hostile toward the passions, "holding that they should be not just moderated, but completely extirpated from human life.... On the other hand, no other ancient school is more sympathetic to the poets, those notorious feeders of passion" (99). What then was the use of poetry to the Stoics?

Nussbaum is able to distinguish two Stoic views on poetry and its effects, grounded in two different views on the nature of the soul. According to what she calls the "cognitive view," emotions are evaluative judgements taking place in a one-part soul; poetry educates by forming or changing these judgements. In the "non-cognitive view," emotions are movements in the non-rational parts of a tri-partite, essentially Platonic soul; poetry educates through rhythm, harmony, and melody by imposing order on these non-rational movements.

Nussbaum traces the non-cognitive view through Posidonius and Diogenes of Babylon, who is for her an important, neglected figure in the history of this question. For Posidonius' doctrine she takes Galen (On the Opinions of Plato and Hippocrates) at his word: Posidonius held a dualist version of psychology, distinguishing the rational part of the soul from two other, non-rational elements. Nussbaum does not mention an alternative, less trusting interpretation of Galen, according to which Posidonius' dualism becomes an artifact of Galen's polemic.5 That remains, however, a minority view on Posidonius, and discussion of it might have weakened the thrust of her argument.

Nussbaum, who is no dispassionate analyst, declares for the cognitive view. The non-cognitive position is "weak"; it "betrays a serious misunderstanding of the passions" and "yields an impoverished view of education" (121). Critics naturally like the cognitive view; it offers more purchase to analysis and commentary, and in a time when theory dominates the discourse of literature it must appear the more attractive and closer to the way we read. Poets, I suspect, will feel instinctively that there is still something to be said for the non-cognitive view, and that the cognitive view, with its emphasis on narrative content, character, and other matters independent of poetic form fails to account for the force of poetry that is not dramatic or narrative. Nussbaum has not convinced me that the distinction between non-cognitive mousike and cognitive poietike (102) is solely a matter of jargon -- perhaps because I do not believe that, as she puts it in a characteristically vivid example, "If I am angry because someone has murdered my child, whom I deeply love, you could play Mozart until the year 2000 without altering my state" (113).

The cognitive view, however, has the advantage that it allows us to talk about poetry, which is, as the cognitive view sees it, a matter for logos. Nussbaum concludes by exploring four tools with which Stoic partisans of the cognitive view resolved the paradox of poetry's dangerous attractiveness. Their tool kit consisted of censorship, writing new poetry, allegorical interpretation, and what Nussbaum calls "the art of critical spectatorship." There is, she rightly observes, little evidence for Stoic censorship of poetry, and what evidence there is for Stoic writing or rewriting of poetry does not produce a desire for more. Allegorical interpretation, with its rich history, can be seen as the Stoics' principal theoretical tool for domesticating the dangerous wildness of poetry, and Nussbaum has some interesting things to say about it. In her reconstruction of Stoic literary theory, however, allegory is only part of a larger project: to redefine the relationship between poetry and its audience by teaching what amounts to a new way of reading. The ideal Stoic reader is not moved by poetry or carried away by it. He is actively critical, detached, and self-aware. His response is in fact hermeneutic, not affective, and becomes part of the Stoic project to abolish the passions.

Brad Inwood's "Seneca and psychological dualism" fittingly occupies the center of Passions and Perceptions. It looks forward to essays like David Sedley's and back to the papers by Annas and Nussbaum, and its clarity and rigor focus attention on a central problem in one Stoic discussion of the passions. Inwood's discussion of how Seneca uses metaphor to advance philosophical argument, and how philosophers have been misled by his practice, will interest literary critics and others interested in the interactions between philosophy and other forms of text.

In the beginning, however, Inwood draws a red herring across the reader's path by opening the question of whether Seneca can accurately be described as "eclectic." That term cannot, as Inwood knows, account for Seneca's problematic relation to orthodox (that is, Chrysippean) Stoicism, and in fact the question of Seneca's eclecticism vanishes almost as soon as it is raised. The more interesting question, as Inwood's title suggests, is whether Seneca holds any version of a dualist psychology, and in particular whether De Ira II.1-4 and related passages can be interpreted as dualist. In answering this question, Inwood inevitably deals with an important doxographical issue. De Ira II, it has been suggested,6 shows the influence of Posidonius. It is thus different from, and perhaps incompatible with, the Chrysippean Book I.

Inwood's reply, grounded in a profound scepticism about the possibility of applying traditional Quellenforschung to an author like Seneca, is that Seneca cannot be described as a dualist, Posidonian or otherwise. Seneca is in fact an innovator within the tradition of Stoic orthodoxy who develops a theory of exactly what kinds of psychological events can and cannot be described as rational, in the sense that they are the product of assent in a mature rational animal. Seneca, according to Inwood, uses Chrysippus' idea of psychological inertia (the idea that, like an overbalanced runner, we often find ourselves carried headlong by an impulse to passion) to expand the Stoic concept of propatheiai. These primi motus include many psychological phenomena that seem to be passions but to Seneca are not.

James Hankinson has been among the most vigorous proponents of Galen as an original philosopher. In "Actions and passions: affection, emotion, and moral self-management in Galen's philosophical psychology" he continues his examination of Galen's understanding of the connection between mind and body, focusing on Galen's views on the nature of pathe and on his account of causation and responsibility.

Galen is never easy to read, but Hankinson has developed a way of getting at his philosophical intention which must, I think, be correct. It depends on two methodological principles: first, that when Galen tells us that terminology is important, he is alerting us to a distinction in his own usage; and second, that in Galen, philosophical arguments can be recovered from therapeutic recommendations. Hankinson sifts Galen's distinction of action (enargeia) from passion (pathos), and of passions from errors (hamartemata) to show that Galen carefully separated two senses of pathos and enargeia: passion and action in accordance with nature, and passion and action contrary to nature. Thus a single event may happen to be called both a pathos and an enargeia; an enargeia contrary to nature (a weak pulse, for example) may be a pathos in accordance with nature.

Hankinson then applies this distinction to Galen's therapeutic recommendations for dealing with the passions. These reveal that Galen had a consistent position on the will, freedom, and responsibility. Passions are, in Galen's view, responsive to the will, and the will in turn is subject to the passions. Self-control consists of regulating this feedback loop so as to bring the passions into a state of natural function, making them pathe in accordance with nature. The passions may be in some sense beyond an individual's control, but an individual is nonetheless accountable for regulating them, and for their outcomes.

Although Galen argued that the best physician was also a philosopher, and indeed made that proposition the title of one of his propaedeutic works, he remained first and foremost a physician. Many in the modern Galenic revival read him as a philosopher,7 although it remains an open question whether he ought to be understood as an original, if unusually garrulous and polemical, philosopher or as a derivative philosophical dilettante. It also remains, at least to my mind, an open question whether Hankinson has shown us what kind of philosopher Galen is, or what kind of philosopher Galen would be if he was a philosopher. The answer to that question depends on whether we agree with Hankinson that, for example, when Galen asks rhetorically "How could anything disobey itself or reject itself or fail to follow itself?" he is "asserting that certain two-place relations are self-evidently irreflexive" (190). In this and similar cases, the difference between a proposition stated and a proposition implied seems to matter.

The remaining four essays examine specific points of Stoic doctrine. In treating them briefly, I do not mean to suggest that they are trivial or lacking in wider interest. In particular Jean-Louis Labarrière, "De la 'nature phantastique' des animaux chez les Stoïciens" will prove suggestive to anyone who hopes to rescue the animal rights movement from thinking that is often as fuzzy as the creatures it defends. In "Le concept de doxa des Stoïciens à Philon d'Alexandrie: essai d'étude diachronique," Carlos Levy locates Philo's account of Joseph's dream of the fat and lean cattle (Genesis 41; Philo, De Somniis II.15) in the context of Stoic and Academic debates over opinion and knowledge. Along the way he makes a case, rather as Hankinson has done for Galen, for paying serious attention to Philo as a figure in the history of philosophy. Phillip Mitsis, in "Seneca on reason, rules and moral development," suggests that the Stoic distinction between praecepta and decreta forms part of a plausible account of an intellectualist, rule-based moral theory. In "Chrysippus on psychophysical causality," David Sedley looks closely at De Fato 7-9 to see what Cicero's exposition reveals about Chrysippus' ideas on the way in which our minds, which to the Stoics were corporeal entities, affect and are affected by the rest of the physical universe.

Sometimes a collection of essays arising from a learned conference will profess a theme which, upon examination, turns out to be less a common focus than a shared excuse for publication. Passions and Perceptions avoids this trap. The contributors bring different ways of doing philosophy and different rhetorical styles to bear on a common set of problems: not quite "Hellenistic philosophy of mind," but more nearly "Hellenistic, and in particular Stoic, readings of the Platonic tripartition of the soul." Subsidiary themes also emerge, in particular (and welcome it is) the revaluation of Galen, Philo, Plutarch, and other authors who have been seen as marginal to the central canon of ancient philosophy. The Fifth Symposium Hellenisticum must have been a stimulating conference, and I hope that the supply of alliterative titles has not run out.


  • [1] Doubt and Dogmatism: Studies in Hellenistic Epistemology, edd. M. Schofield, M. Burnyeat, and J. Barnes (Oxford, 1980); Science and Speculation: Studies in Hellenistic Theory and Practice, edd. J. Barnes, J. Brunschwig, M. Burnyeat, and M. Schofield (Cambridge, 1982); The Norms of Nature: Studies in Hellenistic Ethics, edd. M. Schofield and G. Striker (Cambridge, 1986); Matter and Metaphysics, edd. J. Barnes and M. Mignucci (Naples 1988).
  • [2] K. Döring, Der Sokratesschüler Aristipp und die Kyrenaiker (Wiesbaden-Stuttgart 1988).
  • [3] "Epicurus' refutation of determinism," in SUZHTHSIS: Studi sull'epicureismo greco e latino offerti a Marcello Gigante (Naples 1983) 11-51.
  • [4] "Knowledge of atoms and void in Epicureanism," in Essays in Ancient Greek Philosophy, ed. J. P. Anton and G. L. Kustas (New York 1971) = Cosmic Problems (Cambridge 1989) 161-171.
  • [5] Janine Fillion-Lahille, Le De ira de Séneque et la philosophie stoïcienne des passions (Paris 1984), discussed by Inwood in the following essay.
  • [6] By E. Holler, Seneca und die Seelenteilungslehre und Affektpsychologie der Mittelstoa (Kallmünz 1934) and, with a different view of Posidonius, by Fillion-Lahille, Le De ira de Séneque.
  • [7] See John Scarborough, "Galen Redivivus: An Essay Review," Journal of the History of Medicine and Allied Sciences 43 (1988) 313-321.